Journal für Reproduktionsmedizin und EndokrinologieLaboratory testing: General screening for thrombophilia prior to the prescription of oral contraceptives (OC) is not recommended

Offizielles Organ: AGRBM, BRZ, DVR, DGA, DGGEF, DGRM, D·I·R, EFA, OEGRM, SRBM/DGE

Krause & Pachernegg GmbH, Verlag für Medizin und Wirtschaft, A-3003 Gablitz

Journal für

Reproduktionsmedizin und Endokrinologie– Journal of Reproductive Medicine and Endocrinology –

Andrologie • Embryologie & Biologie • Endokrinologie • Ethik & Recht • Genetik Gynäkologie • Kontrazeption • Psychosomatik • Reproduktionsmedizin • Urologie

Indexed in EMBASE/Excerpta Medica/Scopus

www.kup.at/repromedizinOnline-Datenbank mit Autoren- und Stichwortsuche

Contraception and Thrombophilia - A statement from the

German Society of Gynecological Endocrinology and

Reproductive Medicine (DGGEF e. V.) and the

Professional Association of the German Gynaecologists

(BVF e. V.)

Rabe T, Luxembourg B, Ludwig M, Dinger JC

Bauersachs R, Rott H, Mueck AO, Albring C

J. Reproduktionsmed. Endokrinol 2011; 8 (Sonderheft

1), 178-218

https://www.kup.at/repromedizin

178 J Reproduktionsmed Endokrinol 2011; 8 (Special Issue 1)

Contraception and Thrombophilia

Contraception and Thrombophilia –A statement from the German Society for Gynecological Endo-

crinology and Reproductive Medicine (DGGEF e.V.) and theProfessional Association of German Gynaecologists (BVF e.V.)

T. Rabe1, B. Luxembourg2, M. Ludwig3, J. Dinger4, R. Bauersachs5, H. Rott6, A. O. Mueck7, C. Albring8

Venous thromboembolism (VTE) is responsible for more than half a million deaths annually in the European Union, most in older people following surgery,but some in women of reproductive age using various hormonal contraceptives. In some parts of the population inherited defects of the blood coagulationsystem (factor V Leiden, prothrombin G20210A, protein C, protein S and antithrombin deficiency) are responsible for an increased risk of VTE, which is alsoinfluenced by concomitant factors: e.g. long-distance travel, immobilisation, advanced age, cigarette smoking, high BMI, surgery, malignancy, fluid loss,pregnancy, oral contraceptive use and hormone replacement therapy (HRT).Laboratory testing: General screening for thrombophilia prior to the prescription of oral contraceptives (OC) is not recommended. Laboratory testing forthrombophilia should be limited to women with a positive family and/or personal history of VTE or vascular occlusion.– Factor V Leiden is by far the most common congenital thrombophilia. Heterozygous factor V Leiden (5-fold increased VTE risk) is present in 3–13%,

homozygous factor V Leiden (10-fold increased VTE risk) in up to 0.2–1% of people of European origin.– Prothrombin mutation G20210A: Autosomal dominant mutation inheritance (2% of people of European origin) leads to a 3-fold increase in VTE risk is

substantially increased if one or more additional risk factors are present such as factor V Leiden or protein C, S, or antithrombin deficiency.– Protein C and protein S: VTE risk increases with protein C or S deficiency (odds-ratio 3–15 and 5–11, respectively).– Antithrombin deficiency leads to a 4 to 50-fold increase in VTE risk depending on the type of deficiency.Female hormonal contraceptives containing progestogens with or without combination with a synthetic estrogens (mainly ethinylestradiol [EE]) or anatural estrogen (e.g. estradiol or its derivative estradiol valerate) affect the incidence of VTE in healthy women without known risk factors as follows (VTEcases per 10,000 woman-years):– No method-related increased risk (3–4): Non-hormonal contraceptives (e.g. tubal sterilisation, condoms, spermicides, behavioral methods, copper IUDs)– No or only slightly increased risk (3–4): Levonorgestrel IUS, progestogen-only pill, estrogen-free oral contraceptives– Moderately increased risk (3–10): Combined OCs (COCs) with < 50 µg EE containing norethisterone, norethisterone acetate, levonorgestrel, norgestimate,

chlormadinone acetate, dienogest; COCs with estradiol valerate and dienogest; vaginal combined estrogen/progestogen ring, depot injectables– Moderately increased risk (6–14): COCs with < 50 µg EE containing desogestrel, gestodene, cyproterone acetate or drospirenone; combined estro-

gen/ progestogen contraceptive patchDetection of women at risk for VTE via family and personal history is absolutely required before any hormonal therapy (e.g. contraception, hormonalreplacement). General screening for thrombophilia is not recommended. Additional individual risk factors must be considered. Each patient should beadvised about early symptoms of vascular occlusion. For patients with an increased risk of VTE a risk-benefit analysis must be done regarding non-hormonal choices and non-contraceptive benefits of individual hormonal treatment (e.g. for COCs: regular menstrual cycles, less dysmenorrhoea, improve-ment of acne vulgaris). Shared decision-making and informed consent are strongly recommended. J Reproduktionsmed Endokrinol 2011; 8 (SpecialIssue 1): 178–218.

Key words: thrombophilia, factor V Leiden, prothrombin 20210, protein C, protein S, antithrombin, venous thromboembolism, screening,hormonal contraceptives, risk groups, patient counseling, personal history, family history

Received and accepted: May 31, 2011From the 1Universitäts-Frauenklinik Heidelberg, the 2Institut für Transfusionsmedizin und Immunhämatologie, Abteilung Molekulare Hämostaseologie, DRK BlutspendedienstBaden-Württemberg-Hessen und J. W.-Goethe-Universität Frankfurt a.M., the 3amedes, Zentrum für Endokrinologie–Kinderwunsch–Pränatale Medizin im Barkhof, Hamburg,the 4Center for Epidemiology and Health Research (ZEG) Berlin, the 5Medizinische Klinik IV, Angiologie am Klinikum Darmstadt, the 6Gerinnungszentrum Rhein-Ruhr, the7Universitätsklinikum, Institut für Frauengesundheit Baden-Württemberg, Tübingen and the 8Berufsverband der Frauenärzte e.V., München, GermanyCorrespondence: Thomas Rabe, MD, PhD, MD (hons), Professor Obstetrics and Gynecology, Department of Gynecological Endocrinology and Reproductive Medicine, Univer-sity Women’s Hospital, Medical School Heidelberg, D-69115 Heidelberg, Voßstraße 9; e-mail: [email protected]

Preliminary Remarks

This statement addresses venous throm-boembolic complications in women,with and without the use of various typesof contraception. Because epidemiologi-cal studies have also associated com-bined oral contraceptives (COCs) withan increased risk of arterial thromboem-bolism (myocardial infarction, transientischemic attacks, ischemic strokes), sec-ondary attention is devoted to arterialthromboembolic events.

This statement focuses on the risk asso-ciated with thrombophilia – other poten-tial risk constellations such as obesity,heavy smoking, PCO syndrome, diabe-tes mellitus, insulin resistance etc. haveto be considered on an individual basis –including the resulting diagnostic andtreatment consequences. These recom-mendations do not release physiciansfrom their professional duty to attend toeach individual case, including the pro-vision of extensive information to thepatient about treatment options and theireffects and/or side effects.

Disclaimer

Medical knowledge is constantly chang-ing. Standard safety precautions must befollowed, but as new research and clini-cal experience broaden our knowledge,changes in treatment and drug therapymay become necessary or appropriate.Readers are advised to check the mostcurrent product information provided bythe manufacturer of each drug to be ad-ministered to verify the recommendeddose, the method and duration of admin-istration, and contraindications. It is the

For personal use only. Not to be reproduced without permission of Krause & Pachernegg GmbH.


J Reproduktionsmed Endokrinol 2011; 8 (Special Issue 1) 179

responsibility of the practitioner, relyingon experience and knowledge of the pa-tient, to determine dosages and the besttreatment for each individual patient.Neither the publishers nor the authorsassume any liability for any injury and/or damage to persons or property arisingfrom this publication. For any legal mat-ters the court of jurisdiction is Heidel-berg, Germany.

1. What is Hemostasis?

Hemostasis is a crucial physiological re-action which ensures that bleeding stopsand blood vessels close following an in-jury. In addition to the blood vessel’s en-dothelium, platelets and plasmatic co-agulation factors play a major role in he-mostasis.

A number of reactions are triggeredwhen a blood vessel is damaged:– the vessel constricts, which reduces

blood flow– platelets are activated to adhere and

aggregate, forming a platelet throm-bus

– plasmatic coagulation is activated,forming a fibrin mesh that reinforcesthe initial thrombus.

When a blood vessel is damaged, suben-dothelial structures are exposed, of whichcollagen and tissue factor (thromboplas-tin) play an especially strong pro-coagulatory role. Platelets bind to ex-posed collagen within seconds. In theprocess, the von Willebrand factor formsa “bridge” between the collagen andplatelets (Fig. 1a). Tissue factor (TF), anintegral-membrane protein which is ex-pressed from e.g. fibroblasts and smoothmuscle cells, initiates plasmatic coagula-tion. The TF/factor VIIa complex acti-vates factor X (FXa), which together withits co-factor Va converts prothrombin intothrombin (Fig. 1a). Thrombin catalyzesthe conversion of soluble fibrinogen intoinsoluble fibrin. Fibrin polymerizes into afibrin mesh, which is mechanically stabi-lized via cross-linking by factor XIIIa. Invivo coagulation takes place on cell sur-faces, such as TF-expressing cells and ac-tivated platelets (Fig. 1b).

For decades a model was taught accord-ing to which plasmatic coagulation isinitiated by 2 different systems (extrinsicand intrinsic coagulation systems). It isnow clear that these two systems are in-separable. For one thing, the TF/FVIIacomplex also activates factor IX. For an-

other thing, polyphosphates are releasedwhen platelets are activated, which binddirectly to factor XII and activate it.Moreover, it has also been shown that ri-bonucleic acid (RNA), which is releasedfrom damaged cells, also induces activa-tion of the classical intrinsic coagulationsystem.

In physiological terms, the coagulationprocess is limited by coagulation inhibi-tors at the site of the vessel lesion. These“naturally produced” anti-coagulants in-clude:– “tissue factor pathway inhibitor”

(TFPI), which inhibits the TF/FVIIa/FXa complex

– antithrombin, which inhibits espe-cially thrombin and factor Xa

– protein C and protein S.

The vessel’s endothelium assumes animportant role in these anti-coagulatoryprocesses. The effect of antithrombin isstrengthened by heparan sulfate on thevessel’s surface. The protein C system isactivated when thrombin binds to its in-tegral-endothelium receptor thrombo-modulin. Via the complex of thrombo-modulin and thrombin, protein C is con-verted into its active form, namely acti-

Figure 1a. Schematic representation of primary and secondary hemostasis. Mod. from [1].



vated protein C (APC). Together with itsco-factor protein S, APC inhibits factorsVa and VIIIa (Fig. 1a).

At the end of the wound-healing pro-cess, the fibrinolytic system ensures thatthe vessel reopens. The main enzyme infibrinolysis is plasmin (Fig. 1a). Plasmindissolves the fibrin clot, producing fibrindegradation products such as D-dimers(Fig. 1a).

Deficient regulation of hemostasis,whether due to an excess of pro-coagu-latory factors or to a decline or defectivefunctioning of anticoagulatory mecha-nisms, induces a tendency to developthromboses (thrombophilia).

2. Thromboembolism –

Etiology, Clinical Rel-

evance and Diagnosis

2.1. Prevalence of Thromboem-bolic DiseaseApproximately 1.1 million cases ofvenous thromboembolism (VTE) are di-agnosed in the European Union everyyear, including deep venous thrombosis(DVT) and pulmonary embolism, ofwhich 150,000 cases end in death [2].Also of note is the fact that most throm-boembolism cases are asymptomatic andare therefore not diagnosed. Cohen et al.(2007) estimate that around 220,000deaths across Europe are due to undiag-nosed pulmonary embolism. VTE is

therefore a serious health problem thatclaims more victims per year in the EUthan do breast cancer, HIV/AIDS andtraffic accidents. The incidence in bothsexes rises exponentially with age [3–5],with VTE occurring very rarely inyoung, healthy women. According toHeit et al. 60% of all VTE could beattributed to hospitalization or nursinghome residence [6, 7]. These figuresclearly indicate that VTE represents anenormous risk for certain populationgroups, whereas the vast majority of theyounger population faces only a slightrisk.

Approximately one out of every tendeaths in hospitals (one percent of allpatients admitted) is due to pulmonaryembolism [8].

Venous thromboses and venous throm-boembolism (VTE) occur primarily inthe lower extremities and pulmonaryvessels. They occur less frequently in theupper extremities, and rarely in otherblood vessels (e.g. liver, mesentery, kid-ney, brain or retinal vessels).

A distinction is made between VTE in-duced by reversible risk factors (second-ary VTE) and that which is not (idio-pathic VTE).

Reversible (strong) risk factors include:surgery, hospitalization, immobilizationin plaster casts or other fixed bandagesin the month before diagnosis, and ma-lignancies. Weaker factors include estro-gen treatment, pregnancy, long-distancejourneys (e.g. > 8h) and the above-men-tioned strong risk factors within a periodfrom 3 months to 1 month prior to diag-nosis.

Common to all definitions of non-idio-pathic VTE is the identification of acutereasons (e.g. surgical procedures,trauma, immobilization). This distinc-tion is of limited practical relevance,however, because: 1. the proportion ofwhat are termed idiopathic VTE is de-clining as scientific knowledge ad-vances, and 2. bias presumably plays arole in determining the incidence of idio-pathic VTE in connection with COCs,because mention of the COC “risk fac-tor” in clinical practice often suffices toterminate the search for further VTE riskfactors.

Figure 1b. Coagulation cascade in vivo. (Graphic kindly made available by Novartis Behring, Marburg).



Of special note here is that venousthromboses, and also pulmonary embo-lism, often remain unrecognized. Theyfrequently cause non-specific, minorsymptoms, which are often not properlyunderstood by patients. This means thatdiagnoses are only made following atargeted search, and this search in turnis frequently triggered by the mentionof risk factors. Overall, thromboembo-lism represents an under-diagnosedcondition with a high number of unre-ported cases.

2.1.1. Incidence of Venous Thrombosis(Fig. 2)The incidence, or number of new cases,in Germany is 1–1.8 per 1,000 residentsper year (higher rate in women thanmen). The incidence has increased overthe past few decades. Both a rise in riskfactors (e.g. increased weight) and ad-vances in diagnostics play a role here.Incidence also increases with age(Fig. 3) (see also [5]).

The incidence of arterial occlusion isalso low for women of fertile age. A

large-scale study of oral contraceptiveusers showed the incidence of stroke forwomen under 50 years of age to be 20out of 100,000 (EURAS, Dinger et al.2007 [9]).

The risk of venous thrombosis and em-bolism as well as arterial occlusion de-pends on sex and age. Venous throm-boses and thromboembolism are rare inyoung women who do not show risk fac-tors.

The incidence of serious complications(e.g. pulmonary embolism) is lower thanthe incidence of acute DVT in the leg byapproximately a factor of 10, and deathsdue to thromboembolic complications inCOC users are extremely rare – they areobserved without other identifiablecauses in approximately 1–4 per millionwomen using the Pill. The risk of mortal-ity is due essentially to failure to identifythe underlying condition (venous throm-bosis or pulmonary embolism).

For VTE only, the following figures ap-ply: incidence ~0.0008, lethality ~0.005,deaths ~4 in one million woman-years[9].

Use of COCs increases the risk by a fac-tor of 2–6 [10].

2.2. Etiology of Thromboembo-lismThe presence of a thrombophilic geneticmutation (e.g. factor V Leiden, pro-thrombin G20210A, hereditary deficien-cies of antithrombin, protein C, proteinS, etc.) increases the underlying risk ofthrombosis, which is further increasedby the use of COCs; see Table 1 [11, 12].

Thromboembolism is a multi-factor con-dition, whose risk can increase on a mul-tiplicative basis with the number of riskfactors.

2.2.1. Additional Dispositional RiskFactorsIn addition to COC type and thrombo-philic aspects, various other factors in-crease the risk of venous thrombosis orarterial occlusion.

In more than half of individuals with he-reditary anomalies, venous thrombosisdoes not occur spontaneously. Variousother risk factors function as triggers(Tab. 2), such as:

Figure 2. Genesis of venous thrombosis (with kind permission of www.internisten-im-netz.de).

Figure 3. Risk of venous thrombosis by age (per 10,000 women/year) for COC users. Source: LASS study interimreport: http://clinicaltrials.gov/ct2/show/NCT00676065; Dinger, 2010 personal communication.



– Age: The risk of a thromboembolicevent increases exponentially withage. Below the age of 40, the risk ofsuch an event is approximately 1 in10,000 (0.01%), at age 60 it is ap-proximately 1 in 1,000 (0.1%), andabove 80 years it is approximately 1in 100 (1%) per year [13–16].The risk of thrombosis increases withage, lack of movement, ageing ofthe vascular system and other factors.If hereditary susceptibility factors(thrombophilia) are present, throm-boses occur earlier, often before theage of 45.

– Use of oral hormonal contracep-tives (OC)

– Hormone replacement therapy– Cigarette smoking: Not all studies,

however, confirm an increased risk ofVTE for smoking. The EURAS study,for example, did not when adjustmentwas done for other risk factors [17]

– Obesity– General lack of movement, long pe-

riods of sitting with bent legs (air andcar travel, computer work)

– Immobilization: Illnesses requiringlong periods of bed rest, injuries fromaccidents, bone fractures, surgery,plaster casts

– Other illnesses: Malignancies andmyeloproliferative diseases, cardiacinsufficiency, infections, nephroticsyndrome

– Central venous catheters– Pregnancy, puerperium

The risk of arterial thromboembolicevents or cerebrovascular insults in-creases with:– Age– Cigarette smoking– Positive family history, i.e. occur-

rence of arterial thromboembolicevents in a sibling or parent < 50 yearsof age. If hereditary predisposition issuspected, a medical specialist shouldbe consulted before a decision to usea COC is made.

– Obesity (BMI > 30 kg/m2)– Dyslipoproteinemia– Arterial hypertension– Migraines

– Valvular heart disease, atrial fibrilla-tion, cardiac insufficiency

– Postpartum– Diabetes mellitus– Other diseases: Malignancies and

myeloproliferative diseases, vasculi-tis, chronic inflammatory diseasessuch as rheumatoid arthritis

Note: The presence of a major risk fac-tor or multiple risk factors for venous orarterial disorders can also be a contrain-dication for COC prescriptions.

2.3. Clinical Diagnosis ofThromboembolism

2.3.1. SymptomsTypical symptoms of deep vein throm-bosis in the leg (Fig. 4):– Swelling– Spontaneous, strain-dependent pain

alleviated by elevation– Tenderness from pressure on inner foot

and along vein with the thrombosis– Pain in the calf on flexing the foot– Increased prominence of visible veins

Table 1. Risk of venous thrombosis with thrombophilia, with and without oral contraception. Because some results are lim-ited, the data for with/without OC use come from different studies. Risk with versus without OC use is therefore not directlycomparable; the columns must be considered separately (e.g. for heterozygous prothrombin G20210A mutation, one shouldnot conclude that the risk with OC use doubles from 3–6).

Thrombophilia DVT risk, OR DVT risk with OC, OR

Factor V Leiden mutation, heterozygous 5 16(Data from a meta-analysis of heterozygous and a fewhomozygous cases. The VTE risk for homozygote

Factor V Leiden mutation, homozygous 10 carriers with OC use has thus far not been sufficientlystudied, and could lie considerably higher)

Prothrombin G20210A mutation, heterozygous 3 6(Data from a meta-analysis of heterozygous and a fewhomozygous cases. The VTE risk for homozygote

Prothrombin G20210A mutation, homozygous due to rarity, no data carriers with OC use has thus far not been sufficientlystudied, and could lie considerably higher)

Prothrombin G20210A mutation heterozygous + 4–15 8–17factor V Leiden mutation heterozygous

Congenital protein S deficiency 5–11 5

Congenital protein C deficiency 3–15 6–24

Congenital antithrombin deficiency type I/II 4–50 depending on type of 13AT deficiency 28% of OC users suffer thrombosis

Factor VIII elevation 5–8 9–13

Antiphospholipid antibodies (lupus anticoagulants, 2–16 depending on antibody insufficient study resultsanti-cardiolipin antibodies, anti-β2-glycoprotein I or combination thereofantibodies)

Hyperhomocysteinemia risk rises by 1.3 for each insufficient study resultsincrease of 5 µmol

Lipoprotein (a) > 300 mg/l 1.8 no data

MTHFR C677T polymorphism not elevated not elevated



Typical symptoms of pulmonary em-bolism:– Sudden or gradual dyspnea, during

exertion or at rest depending on thestage

– Respiration-related thoracic pain– Therapy-resistant pneumonia of inde-

terminate origin– Coughing, blood traces in sputum– Tachycardia– Syncope

Note: The symptoms are extremelyvariable. All symptoms can occur either

individually or in combination. Deepthromboses and pulmonary embolismcan also occur without symptoms.

Possible symptoms of a venous (sinus)or arterial thrombosis (insult) in thecentral nerve system:– Unusual, strong and/or persistent

headache– Impaired vision: sudden partial or

complete loss of sight, double vision– CNS symptoms, slurred speech or

aphasia, vertigo, sudden weakness orpronounced numbness on one side or

in one part of the body, impaired co-ordination

– Collapse with or without focal sei-zures

Thromboses can occur less frequently inother locations, such as venous throm-boses in the arm including swelling withor without pain, or in the mesentery(possibly acute abdomen), or myocar-dial infarction.– COC users should be strongly urged

to consult a physician if they showsigns of thrombosis.

2.3.2. Recurrent Venous Thrombo-embolismAround 30% of patients with VTE intheir histories show a recurrence within10 years, with the highest risk in the firstyear following the initial diagnosis [18,19].

2.3.3. Summary– Identification of venous thrombosis

and resulting pulmonary embolism iscrucial for prompt treatment. Unrec-ognized DVT carries a high risk ofpulmonary embolism, and unrecog-nized pulmonary embolism is linkedwith high mortality.

– Typical symptoms of DVT such aspain, swelling and/or tautness in theleg should be reported as promptly aspossible to a physician in order to ini-tiate diagnostic procedures. The Wells

Table 2. Risk factors for venous thromboembolism. Mod. from: [ScottishIntercollegiate Guidelines section 10]

Age

Exponential increase in risk with age. In the general population:< 40 years: annual risk 1/10,00060–69 years: annual risk 1/1,000> 80 years: annual risk 1/100(May reflect immobility and coagulation activation)

Weight

3-fold risk if obese (body mass index > 30 kg/m2)(May reflect immobility and coagulation activation)

Varicose veins

1.5-fold risk after major general/orthopaedic surgery, but low risk after varicose vein surgery

Previous VTE

Recurrence rate 5%/year, increased by surgery

Thrombophilia

Low coagulation inhibitors (antithrombin, protein C or S)Activated protein C resistance (e.g. factor V Leiden)High coagulation factors (I, II, VIII, IX, XI), prothrombin G20210AAntiphospholipid syndromeHigh homocysteine

Other risks for thrombotic states

Malignancy: 7-fold increased risk compared with the general populationHeart failureRecent myocardial infarction/strokeSevere infectionInflammatory bowel disease, nephrotic syndromePolycythaemia, paraproteinaemiaBeheçt’s disease, paroxysmal nocturnal haemoglobinuria

Hormone therapy

Oral combined contraceptives, HRT, raloxifene, tamoxifen (3-fold risk)High-dose progestogens (6-fold risk)

Pregnancy, puerperium

10-fold risk*

Immobility

Bed rest > 3 days, plaster cast, paralysis (10-fold risk)Risk increases with duration

Prolonged travel see text

Hospitalisation

Acute trauma, acute illness, surgery (10-fold risk)

Anaesthesia

2-fold greater risk for general (versus spinal/epidural)

* Note: Puerperium risk > pregnancy

Figure 4. Venous thrombosis in the leg. Source: R.Bauersachs.



Score can be used to estimate theclinical probability of venous throm-bosis in the leg (Tab. 3) or pulmonaryembolism (Tab. 4). It combines ex-amination results with VTE risk fac-tors [21]. However, because the WellsScore cannot reliably diagnose or ex-clude a thrombosis, it may only beused in conjunction with other diag-nostic parameters (see the diagnosticalgorithm for venous thrombosis inthe leg and pulmonary embolism inFigure 5).

– Patients should also be acquaintedwith the “ACHES” checklist for earlywarning signs of venous and arterialocclusion (Tab. 5).

2.4. Clinical Factors for Assess-ing the Risk of Coronary HeartDisease and VTE

2.4.1. Family History

Value of family history of venousthrombosis as a predictive factor forindividual risk, also with respect tothrombophilic factors:

A family history of venous thrombosiscan indicate the presence of genetic riskfactors. Carriers of genetic factors have ahigher risk of first-time venous throm-bosis, and a higher risk still if environ-mental factors are also present. For ex-ample, factor V Leiden mutation syner-gistically increases the risk of venousthrombosis for women who take oralcontraceptives [22]. Because generallaboratory screening for thrombophilicfactors is not cost-effective [11, 23], re-search is focusing on identifying criteriathat increase the probability of finding

Table 4. Wells Score for determiningclinical probability of pulmonary em-bolism (following German S2 guidelineon diagnosing and treating venousthrombosis and pulmonary embolism,2010) [21]

Clinical features Score

Previous venous thrombosis or +1.5pulmonary embolismRecent surgery or immobilization +1.5Cancer +1Hemoptysis +1Heart rate > 100 beats/minute +1.5Clinical symptoms of venous +3thrombosisAlternative diagnosis less likely +3than pulmonary embolism

Score 0–4: Pulmonary embolism unlikely;score > 4: Pulmonary embolism likely

Table 3. Wells Score for determiningclinical probability of venous thrombo-sis in the leg (following German S2guideline on diagnosing and treatingvenous thrombosis and pulmonaryembolism, 2010) [20].

Clinical features Score

Active cancer 1.0Paralysis, paresis, recent plaster 1.0immobilization of lower limbBed rest (> 3 days); major surgery 1.0(< 12 weeks)Pain/stiffness along deep venous 1.0systemEntire leg swollen 1.0Calf swelling > 3 cm compared to 1.0asymptomatic legPitting edema 1.0Collateral superficial veins 1.0Previous documented DVT 1.0Alternative diagnosis at least as –2.0likely as DVT

Score ≥ 2.0: high probability of venousthrombosis in the leg; score < 2.0: prob-ability of venous thrombosis in the legnot high

Table 5. Checklist for typical symptoms of blood clots

ACHES checklist for signs of arterial or venous thrombosisA = Abdominal painC = Chest pain: sudden appearance and spread into left arm; sudden strong coughing

without apparent causeSudden shortness of breath

H = Headache: New occurrence, long duration, one-sided, worsening of a migraine, cre-scendo character, scotoma, impaired speech

E = Eye problems: Impaired vision, partial or complete blindness or double visionS = Swelling of the leg: strong pain and/or swelling of one legAdditional symptoms: Weakness, numbness in one part of the body, dizziness or faintness

Figure 5. Diagnostic algorithm for venous thrombosis in the leg or pulmonary embolism for patients with stablehemodynamics. Mod. from [German S2 guideline on diagnosing and treating venous thrombosis and pulmonaryembolism, 2010]. (CUS = compression sonography)



genetic risk factors in laboratory tests.Family history is one of these criteria.Various studies have examined the valueof family histories as surrogate param-eters for identifying known genetic riskfactors for venous thrombosis [24–28].These studies suggest that family histo-ries are not very suitable for identifyingknown genetic risk factors. Some stud-ies, however, have shown a link betweenfamily history and the occurrence ofvenous thrombosis [29, 30]. This alsoapplies to OC users. The LASS study1

showed that COC users with a positivefamily history for VTE showed a three-fold higher VTE risk than COC userswith a negative family history [17]. Thequestion also arises of whether familyhistory is of additional value in predict-ing individual risk of venous thrombosiswhen genetic risk factors have alreadybeen identified. The case control studyby Bezemer et al. (2009) [31] addressesthis issue.

Case-control study by Bezemer et al.(2009) [31]:– Study objective: The case-control

study by Bezemer et al. (2009) [31]examined the value of family historyfor determining the risk of venousthrombosis in connection with knownrisk factors.

– Study population: A multivariantanalysis of environmental and geneticrisk factors for venous thrombosiswas performed as part of a popula-tion-based case-control study thatused blood samples and informationabout family and environmental fac-tors from 1,605 patients with first-time venous thromboses and from2,150 control persons.

– Definition of family history: Pa-tients were asked whether their par-ents, brothers or sisters had had avenous thrombosis, and if so at whatage. Because the patients’ partnersserved as the control persons, chil-dren were not included in these histo-ries. A family history was consideredpositive if at least one of these first-degree relatives had had a venousthrombosis.

– Results (see Table 6): A total of 505patients (31.5%) and 373 control per-sons (17.3%) reported a venousthrombosis in one or more first-de-gree relative. A positive family his-

tory increased the risk of venousthrombosis by a factor of more than 2(odds ratio 2.2, 95% confidence inter-val 1.9–2.6), and a positive familyhistory with more than one relativeincreased the risk by a factor of upto 4 (3.9, 95% CI: 2.7–5.7). Familyhistory correlated only poorly withknown genetic risk factors. Familyhistory correlated with the occurrenceof venous thrombosis in patients bothwith and without genetic or environ-mental risk factors. The risk ofvenous thrombosis increased with thenumber of demonstrated risk factors.For persons with genetic and environ-mental risk factors and a positive fam-ily history, the risk was up to 64 timeshigher than for those who had a nega-tive family history and no known riskfactors.

– Conclusions: Family history is a riskindicator for first-time venous throm-bosis, regardless of whether other riskfactors are identified. In clinical prac-tice, family history could be moreuseful than laboratory testing forthrombophilia in assessing the risk ofvenous thrombosis.

Summary:– Family history of deep venous throm-

bosis and pulmonary embolism,which is reported by approximately3% of women of fertile age, is astrong predictor for the risk of VTE.

– Family history of coronary heart dis-ease (CHD): Occurrence in parentsbefore the age of 45 years (somesources use 50): Myocardial infarc-tion in the mother; stroke, thrombo-sis, thromboembolism in either par-ent.Diseases/conditions in the patients’grandparents and in the siblings oftheir parents can be added to the as-sessment.For CHD risk above and beyond VTErisk, metabolic conditions includinglipid metabolic disorders, diabetes mel-litus, hypertension etc. also play a role.

– Family history of fatal myocardial in-farction/stroke before the age of 50,which is reported by approximately2% of women in fertile age, is astrong predictor of cardiovascularrisk [9].

– If family history is positive for cardio-vascular disease, laboratory testingmay be needed for further clarifica-tion (e.g. thrombophilia parameters

for VTE, lipid status for arterialthromboembolism), possibly alsofamily testing.

– Family history of cardiovascular dis-ease is an accurate predictive param-eter for assessing probability of samein the patient and other family mem-bers.

2.4.2. Risk Factor: TravelThe following analysis is based on a2010 Internet publication from the Cen-ters of Disease Control in Atlanta, USA(Barbeau: Deep Vein Thrombosis andPulmonary Embolism 2010)2 that takesinto account surveys and meta-analysesby Anderson et al. (2003) [32],Goodacre et al. (2005) [33], Kuipers etal. (2007) [34, 35], and Geerts et al.(2008) [36].

It examined known risk factors and dif-ferent types of travel. A population-based case-control study of adults whowere treated for a (first-time) VTEshowed that long periods of travel (≥ 4hours) double the risk of VTE. The riskincreased most in the first week aftertravel, but remained elevated for twomonths. Air travel did not show a differ-ent effect from bus, rail or car travel,which suggests that the increased riskfrom air travel is due primarily to thelength of inactivity. Additional risk fac-tors include factor V Leiden mutation,oral contraceptives for women, BMI> 30 kg/m2, and height > 190 cm. Someof these effects were most prevalent forair travel. In addition, persons under160 cm in height only showed a greaterVTE risk after longer periods of airtravel. These results suggest that addi-tional factors combine with air travel toplay a role in elevated VTE risk.

Clinical StudiesTwo subsequent retrospective cohortstudies examined VTE frequency and airtravel.

The first is a cohort study of 2,630healthy Dutch commercial pilots [37].The incidence of VTE in this group was0.3 per 1,000 person-years. When thedata were adjusted for age and sex, therate did not differ from that for the gen-eral Dutch population. There was no

1 http://clinicaltrials.gov/ct2/show/NCT00676065

2 http://wwwnc.cdc.gov/travel/yellowbook/2010/chapter2/deep-vein-thrombosis-pulmonaryembolism.aspx



association between VTE incidence inthe pilots and the number of hours theyflew.

The second study examined 8,755 em-ployees of several international organi-zations [34]. VTE frequency followingflights of over 4 hours was 1.4 per 1,000person-years. The absolute risk for VTEwas given as 1 per 4,656 flights. TheVTE rate for women was higher, espe-cially for those taking oral hormonalcontraceptives. The incidence was alsohigher for persons with BMI > 25 kg/m2

and height < 1.65 m or > 1.85 m. VTErisk increased with flight duration andthe number of flights during an 8-weekperiod, with a 3-fold risk for persons

who took five or more long-distanceflights (≥ 4 hours). Each additionalflight increased the risk of VTE by a fac-tor of 1.4. The risk was highest in thefirst two weeks following a long-dis-tance flight, and returned to baselineafter 8 weeks.

Both studies examined populationgroups that were younger (average age35–40 years) and healthier than the gen-eral population, so the results are nottransferrable to a population group withheightened risk.

Preventive Measures for TravellersSeveral randomized, controlled studieshave assessed the effect of preventive

measures on the risk of VTE followingair travel [38]3.

All the studies examined the risk ofasymptomatic DVT in travelers forflights of ≥ 7 hours. All travelers wereencouraged to exercise at regular inter-vals during the flight and to drink onlynon-alcoholic beverages. Ultrasoundtests were done between 90 minutes and48 hours post-flight to determine thepresence of DVT in the leg. The effect of

Table 6. Family history and VTE. Mod. from [31], with permission.

Family Historya No. (%) Odds Ratio (95% CI)

Patients with Control Per stratum of type Relative to the groupvenous thrombosis subjects of risk identified with no known riskfactors

and fegative family history

No Known Risk Factors

All n = 389 n = 1538 … …Negative 261 (67.1) 1286 (83.6) 1 (Reference) 1 (Reference)Positive

Any relative 128 (32.9) 252 (16.4) 2.5 (1.9–3.2) 2.5 (1.9–3.2)Relative < 50 y 53 (13.6) 98 (6.4) 2.7 (1.9–3.8) 2.7 (1.9–3.8)> 1 Relative 23 (5.9) 27 (1.8) 4.2 (2.4–7.4) 4.2 (2.4–7.4)

Environmental Risk Factors Onlyb

All n = 823 n = 378 … …Negative 596 (72.4) 310 (82.0) 1 (Reference) 9.5 (7.8–11.5)Positive


Genetic Factors Onlyc



Environmental and Genetic Factors



CI: confidence interval; a: History of venous thrombosis among parents, brothers, and sisters; b: Surgery, injury, immobilization, and pregnancyor puerperium within 3 months before the index date, use of oral contraceptives or hormone therapy at the index date, and diagnosis ofmalignancy within 5 years before or within 6 months after the index date; c: Low levels of antithrombin, protein C, or protein S; factor V Leidenmutation; or prothrombin 20210 mutation

3 http://wwwnc.cdc.gov/travel/yellowbook/2010/chapter2/deepvein-thrombosis-pulmonary-embolism.aspx; http://www.who.int/cardiovascu-lar_diseases/wright_project/phase1_report/WRIGHT%20REPORT.pdf



compression stockings, aspirin, low mo-lecular-weight heparin and various natu-ral extracts with anti-coagulatory prop-erties were examined. None of the phar-macological interventions showed a sig-nificant effect. Compression stockings(10–20 mmHg and 20–30 mmHg) re-duced the risk of asymptomatic DVT.Four travelers in one study, however,developed superficial thrombophlebitisafter wearing compression stockings.None of the travelers participating in thestudies showed symptomatic DVT orpulmonary embolism.

All travelers are encouraged to ensuresufficient hydration, wear loose-fittingclothing and flex their calves at regularintervals on extended trips. Compressionstockings show a favorable effect if otherVTE risk factors are present. Currentlythere are no convincing data showingthat pharmacological interventions sig-nificantly reduce the risk of VTE fromtraveling.

Summary of recommendations to pre-vent VTE from long-distance travel [36,39]:– The following general measures are

recommended for travelers spending> 8 hours on an airplane: avoid tightclothing on the lower extremities andaround the waist, ensure adequatefluid intake, and exercise (flex) calfmuscles on a frequent basis (grade 1C).

– These same general measures are rec-ommended for long-distance travel-ers with additional VTE risk factors.If active thrombosis prevention is un-der consideration on account of el-evated VTE risk, properly fitted knee-length graduated compression stock-ings (GCS) that provide pressure of15–30-mmHg (grade 2C) for theankle area can be recommended, or asingle prophylactic pre-flight shot oflow molecular-weight heparin(LMWH) (grade 2C).

– Aspirin is not recommended to long-distance travelers as a preventive mea-sure for venous thrombosis (grade 1B).

Risk groups for travel-related throm-boembolism, following the Inter-national Consensus Statement bySchobersberger et al. (2008) [40]:

Group 1: Low RiskLong-distance travelers without risk fac-tors listed for Groups 2 and/or 3

Group 2: Medium RiskPresence of two or more of the followingfactors– Oral contraception– Hormone replacement therapy– Pregnancy or puerperium– Family history of venous thrombosis– Documented thrombophilia– Marked varicose veins, chronic

venous insufficiency– Obesity (BMI > 30)– Age > 60 years

Group 3: High Risk– Previous VTE– Manifest malignancy or other serious

disease– Immobilization, e.g. plaster cast– Major recent surgery

2.4.3. Risk Factor: SurgeryPerioperative Use of Hormonal Contra-ceptivesThe American College of Chest Physi-cians [36] assigns surgical operations to3 different categories for thrombosis risk:Low risk: Minor operations on other-wise healthy, active patients.Medium risk: Most general surgical,open gynecological and urological op-erations.High risk: Hip and knee joint endo-prostheses, hip fractures and spinal cordinjuries. The thrombosis risks are shownin Table 7 [42].

Every type of combined hormonal con-traception increases the risk of thrombo-sis. However, the current German throm-bosis prevention guideline no longerrecommends discontinuing COC use be-fore surgery. This is due to the extendedperiod of residual hypercoagulation ofapproximately 6 weeks following cessa-tion of use and the risk of unplannedpregnancy. Instead, patients should beprovided peri-operative with sufficientthromboprophylaxis (in accordancewith the current thrombosis preventionguideline).

Citation from the German AWMF (Ar-beitsgemeinschaft der Wissenschaftli-chen Medizinischen Fachgesellschaften)guideline on preventing venous throm-boembolism (www.awmf.org/leitlinien/aktuelle-leitlinien/ll-liste/deutsche-gesellschaft-fuer-chirurgie.html): “Aspecial evaluation of the LASS-Study onrequest of the FDA showed in the firstthree months after major orthopedic sur-gery a 7-fold higher risk for VTE whencompared to OC use independent of anyoperation. Compared to non-users of OCthe risk had been increased 2-fold[Dinger 2011, personal communica-tion]. Despite a large study population ofmore than 17,000 women, this is not sta-tistically significant.” And “the risk ofunplanned pregnancy if OC use is dis-continued before surgery should be

Table 7. Risk of venous thromboembolism in surgical patients without prophylaxis(According to Geerts et al. (2001) [41] and Geerts et al. (2004) [42].

Risk category Deep vein thrombosis (%) Pulmonary embolism (%)

Calf Proximal Clinical Fatal

Low risk – minor surgical 2.0 % 0.4 % 0.2 % < 0.01%operations, age < 40 years,no additional risk factors*)

Moderate risk – minor surgical 10–20% 2–4% 1–2% 0.1–0.4%operations with additionalrisk factors*)

or surgical operations in patientsaged 40–60 without additionalrisk factors

High risk – surgical operations 20–40% 4–8% 2–4% 0.4–1.0%in patients > 60 yearsor surgical operations in patientsaged 40-60 years withadditional risk factors*)

Highest risk – surgical operations 40–80% 10–20% 4–10% 0.2–5%in patients > 40 yearswith multiple risk factorsor hip or knee arthroplastyor major trauma or spinal cord injury

*Additional risk factors include one or more of the following: advanced age, cancer, priorvenous thromboembolism, obesity, heart failure, paralysis, or presence of a molecularhypercoagulable state (eg, protein C deficiency, factor V Leiden).



weighed in relation to the reduced risk ofthrombosis. Discontinuing OC use is notrecommended. Users of hormonal con-traceptives should receive physical andmedication-based thromboprophylaxisbefore more extensive surgery”.

The LASS study (http://clinical-trials.gov/ct2/show/NCT00676065)showed a 7-fold higher VTE risk for OCusers in the first three months followingmajor surgery [Dinger, personal com-munication]. The authors therefore rec-ommend for the (infrequent) cases thatfertile women make long-term plans formajor surgery (e.g. hip or knee replace-ment) that they discontinue OC use atleast 6 weeks before surgery. In suchcases, a 3-month period should alsoelapse before resuming COC use.

For minor surgery, hormonal contracep-tion can be resumed or started for thefirst time 14 days after an ambulatoryprocedure or hospital discharge.

2.4.4. Summary– Some coagulation factors return to

normal only 2–3 months followingdiscontinuation of oral hormonal con-traceptives.

– Before every operation, surgeonsshould ask patients about possible useof combined oral contraceptives, vagi-nal ring, contraceptive patch or otherforms of hormonal contraception.

– Women with oral contraceptivesshould receive peri-operative throm-boprophylaxis according to the cur-rent thrombosis prevention guideline.

2.5. Diagnosing Venous Throm-bosis in the Leg and Pulmo-nary EmbolismCorrectly diagnosing venous thrombosisin the leg is essential for appropriate treat-ment and for preventing subsequent pul-monary embolism. To determine or ex-clude venous thrombosis in the leg andpulmonary embolism, medical associa-tions recommend diagnostic algorithms.If used consistently, VTE lethality can bereduced (interdisciplinary German S2guideline from the AWMF on diagnosingand treating venous thromboembolismand pulmonary embolism, 2010: Diag-nostik und Therapie der Venenthrom-bose und der Lungenembolie). Thesealgorithms cover the following:– Clinical diagnostic measures: The

clinical probability of venous throm-

bosis in the leg or of pulmonary em-bolism can be estimated using theWells Score (Tab. 3, 4) and should bedocumented. The Wells Score coverstypical clinical symptoms of VTE,and includes examining and palpatingthe affected part of the body as well asdetermining the presence of VTE riskfactors. Because a clinical examina-tion alone is not sufficient to deter-mine or exclude VTE, further diag-nostic measures should be undertaken(D-dimer test for lower clinical prob-abilities and imaging procedures forhigher probabilities).

– Laboratory tests: A blood test to de-termine the D-dimer concentration ishelpful in excluding VTE. This testshould only be done after determiningthe clinical probability of VTE. D-dimers are degradation products of theproteolysis of cross-linked fibrin. Theyindicate increased fibrin formationwith secondary fibrinolysis, as occurswith VTE. If the test is negative, it ex-cludes VTE with a high probability.But the sensitivity of individual D-dimer tests varies, and is usually not100%. It therefore has to be combinedwith a score (Wells) for the clinicalprobability of VTE. If the Wells Scoreshows a low clinical probability andthe D-dimer concentration lies in thenormal range, VTE can be excluded. Ifthe clinical probability is low but theD-dimer concentration is elevated, fur-ther tests must be done. The same istrue for high clinical probability ofVTE with a normal D-dimer concen-tration. If the clinical probability ishigh, therefore, a D-dimer test can beomitted in favor of proceeding directlyto imaging tests.Elevated D-dimer values, however,do not always indicate venous throm-bosis. These values can be high forother reasons, such as following sur-gery or injury, during an infection, orin conjunction with a tumor. If the D-dimer test is positive, further diagnos-tic means must be undertaken to de-termine or exclude VTE.

– Imaging procedures: Actual diagno-sis of venous thrombosis in the legand pulmonary embolism is done byimaging procedures.The gold standard for diagnosing legDVT in routine practice is non-inva-sive imaging by ultrasound, i.e. com-pression sonography (Fig. 6). It isconsiderably less stressful for pa-

tients than x-rays with contrast agents(phlebography).

– For diagnosing pulmonary embolism,multi-slice spiral CT pulmonary an-giography is generally used for pa-tients with stable hemodynamics. Im-aging via ventilation/perfusion scin-tigraphy is also possible. Scintigra-phy, however, yields a high propor-tion of non-usable diagnostic results.Pulmonary angiography is rarely in-dicated these days. Echocardiographyis used in determining or excludingright ventricular dysfunction.

2.5.1. Summary– Diagnoses of venous thrombosis in the

leg and pulmonary embolism shouldbe based on algorithms that encom-pass the clinical probability of VTE,laboratory testing of D-dimer concen-trations, and imaging procedures.

– Deep venous thrombosis of the leg:The gold standard for DVT diagnos-tics in daily practice is non-invasiveultrasound imaging as compressionsonography.

– Pulmonary embolism: Multi-slicespiral CT angiography or ventilation/perfusion scintigraphy are non-inva-sive imaging tests for pulmonary em-bolism. Echocardiography is used todetermine right ventricular dysfunc-tion.

2.6. SummaryPrevalence of VTE: An estimated 1.1million cases of venous thromboembo-

Figure 6. Venous thrombosis in the leg shown by com-pression sonography (thrombosis in right femoral vein– VFC). The vein is distended, with non-compressiblediameter (AFC = common femoral artery). Source: B.Luxembourg.



lism (including DVT and pulmonaryembolism) occur every year in the Euro-pean Union, and are associated withmore than 150,000 deaths [2]. VTE istherefore a serious health problem thatclaims more victims annually in the EUthan breast cancer, HIV/AIDS and trafficaccidents. However, the risk is associ-ated to a very high degree with age andhospitalization [3–5, 7], which meansthat VTE represents an enormous riskfor certain population groups whereasthe majority of the younger populationfaces only a very slight risk.

Mortality: Pulmonary embolism is thecause of death for approximately one outof every ten patients who die in hospital(1 percent of all patients admitted) [8].

Age dependency in women (per 10,000women/year): < 20 years: 4.3; 20–29years: 8; 30–39 years: 13; 40–49 years:23.9; ≥ 50 years: 50.1) [9].

Symptoms of DVT and pulmonaryembolism:– Venous thromboembolism is

underdiagnosed because its symp-toms can often be non-specific oreven absent at first.

– Typical symptoms of venous throm-bosis in the leg are pain, sensitivity topressure, edema, swelling, cyanoticskin coloring, and/or dilated superfi-cial veins in the affected leg.

– DVT is often asymptomatic for bed-ridden patients on account of reducedhydrostatic pressure.

– The greatest risk associated with DVTis the possibility that a pulmonaryembolism may develop. Typicalsymptoms of pulmonary embolisminclude thoracic pain, dyspnea, cough-ing, hemoptysis, treatment-resistantpneumonia, tachycardia and/or syn-cope. Pulmonary embolism is oftenthe result of a number of different de-velopments at different points intime, and its symptoms are often ini-tially indistinct or even absent. Deathcan occur suddenly and unexpectedly.

Risk factors:– Numerous hereditary and acquired

risk factors can contribute to VTE.Typically more than one factor con-tributes to VTE pathogenesis, i.e.VTE is a multifactorial disease.

– Each patient’s risk profile should bedetermined, and thromboprophylaxis

measures should be considered fortypical risk situations such as surgeryor immobilization.

– Numerous thrombophilic factors el-evate individual risk.

– Oral hormonal contraceptives andhormone replacement therapy also in-crease the risk by a factor of 2–6 forhealthy women, and have a multipli-cative effect on patients with knownthrombophilic factors.

Early determination of risk:– Family history: Family histories of

cardiovascular conditions includingVTE are an important instrument fordetermining risk.

– Travel-related thrombosis: The riskof VTE is increased by trips of 4 ormore hours, regardless of whetherthey are by plane, car or bus. For tripsof more than 8 hours, general mea-sures are always recommended (e.g.exercise, sufficient hydration). Throm-boprophylaxis consisting of suitablecompression stockings or applicationof heparin (LMWH) are recom-mended only for high-risk patients;aspirin is not recommended. Patientstaking oral contraceptives but withoutfurther DVT risk factors generallyonly have a low to medium risk ofthrombosis from long-distance travel.Before departure, presence of any ad-ditional risk factors should be deter-mined which could change the riskcategory assignment.

Diagnosis:– Diagnoses of venous thrombosis in

the leg and pulmonary embolismshould be based on algorithms thatencompass the clinical probability ofVTE, D-dimer laboratory tests, andimaging procedures.

– Deep venous thrombosis of the leg:The gold standard for DVT diagnosisin regular practice is non-invasive ul-trasound imaging as compressionsonography.

– Pulmonary embolism: Multi-slicespiral CT angiography or ventila-tions/perfusion scintigraphy are non-invasive imaging tests for pulmonaryembolism. Echocardiography is usedin determining right ventricular dys-function.

Thrombosis prevention guidelines:The consensus guidelines from theAmerican College of Chest Physicians

(ACCP) [36] are revised every 2–3years, and are considered the interna-tional standard. In Germany, the nationalguidelines published by the AWMF(Arbeitsgemeinschaft der Wissenschaft-lichen Medizinischen Fachgesellschaf-ten) need to be given preference. TheGerman S3 thrombosis preventionguideline was issued in 2009 [43]. TheGerman S2 guideline for diagnosing andtreating venous thrombosis and pulmo-nary embolism was published by theAWMF in 2010

2.7. Important Websites:English Associations

GuidelinesManagement of venous thromboem-bolism:A Clinical Practice Guideline from theAmerican College of Physicians and theAmerican Academy of Family Physi-cians (2007) [44].

EAST Practice Parameter Workgroupfor DVT Prophylaxis (2011) [45]

Prevention of venous thromboembo-lism:Guideline from the National Institute forHealth and Clinical Excellence (2010):Venous thromboembolism (surgical)[46].

American College of Chest PhysiciansEvidence-Based Clinical Practice Guide-lines (8th Edition) (2008): Prevention ofvenous thromboembolism [36].

Oral contraceptives and the risk ofvenous thromboembolism:Guideline from the Royal College ofObstetricians and Gynaecologists(2010): Venous Thromboembolism andHormonal Contraception [47].

SOGC clinical practice guidelines: Oralcontraceptives and the risk of venousthromboembolism: an update (2011)[48, 49].

3. Laboratory Tests and

Patient Information/

Counseling

3.1. General PreliminaryRemarksVenous thromboembolism (VTE) is amulti-factor condition with roles playedby environmental factors, acquired risk



factors such as age, excess weight andoral contraceptives, and hereditary fac-tors. Numerous studies have investigatedgenetic factors in VTE, ranging fromcandidate gene studies to genome-wideassociation studies. They show that ge-netic variants lead either to an excess ofprothrombotic factors or to a deficiencyin anti-thrombotic factors.

Factor V Leiden mutation is the most of-ten and best studied genetic predisposi-tion factor for VTE, followed by the pro-thrombin G20210A mutation and defi-ciencies in protein S, protein C and anti-thrombin [50].

A large number of additional laboratoryparameters and genetic variants associ-ated with VTE risk have been studied.These have not been included in routinethrombophilia screening thus far be-cause it is unclear how significant theyare for treatment decisions in generalpractice. For example, there is a knownassociation between blood group andVTE risk, with O and A2 showing alower risk than other blood groups [51].This association is based among otherthings on higher Willebrand factor andfactor VIII levels – two known risk fac-tors for VTE – in individuals who do nothave blood groups O or A2. How andwhether VTE risk changes here with theuse of oral contraceptives is unclear.Blood group therefore does not play arole in prescribing oral contraceptives.

Various types of genetic polymorphismare often associated with only a slightlyhigher risk of thrombosis. For example,an association between VTE risk andpolymorphism in the CYP4V2 regionwas recently described, although the riskwas only slightly elevated (odds ratio:1.14–1.39) [52]. Moreover, this VTErisk was further weakened on adjustmentfor other VTE risk factors [52].

These examples show that a range of ge-netic determinants are involved in therisk of VTE which however play no rolein clinical practice because it is entirelyunclear what influence they have on VTErisk with the use of oral contraceptives.

Various individual factors increase therisk of VTE in connection with geneticthrombophilic factors. For example, car-riers of factor V Leiden or prothrombinG20210A mutation show a multiple in-

creased risk of VTE if they use oral hor-monal contraceptives [53, 54], undergohormone replacement therapy [55], orsmoke cigarettes [56]. This suggests thatfor many individuals, genetic thrombo-philic factors alone are not sufficient totrigger VTE. The accumulation of riskfactors, which can also occur on a tran-sient basis, can however trigger VTE. Itis therefore more important in clinicalpractice to identify and avoid acquiredrisk factors than to determine a numberof different types of polymorphism.

3.2. Indications for Thrombo-philia TestingThrombophilia testing should only beperformed if the results are of clinical sig-nificance regarding the patient’s family,life situation, age, desire for children, etc.

From the perspective of gynecology,thrombophilia testing should be done ifone of the symptoms or conditions de-scribed in Table 8 is present.

3.3. Laboratory Testing forThrombophiliaLaboratory testing for thrombophiliashould cover the following parameters inTable 9 a, b.

Although not advisable for the generalpopulation, thrombophilia screeningmakes sense for risk groups, e.g. womenwith a positive family history (multipleoccurrence of thromboembolism in first-degree relatives or thromboembolism infirst-degree relatives at a young age) be-fore prescribing oral contraceptives.

Accumulated risk factors can also be acontraindication for prescribing COCs.Special mention should be given here tocardiovascular risk factors linked to ahigher risk of arterial thromboembolism(e.g. age > 35 years plus strong nicotineuse or presence of multiple cardiovascu-lar risk factors such as obesity, arterialhypertension, or known hyperlipidemia).Additional tests are recommended inindividual cases, e.g. blood glucose,HbA1c, lipid status, lipoprotein (a), thy-roid hormones, homocysteine, CRP,blood counts, creatinine.

If homocysteine levels are high, it maybe necessary to clarify the cause; in theevent of low to medium hyperhomo-cysteinemia, benefits of folic acid, vita-min B6 and B12 regarding vascular oc-clusion are not confirmed.

3.4. Preliminary Considerationsfor Thrombophilia TestingWhat should be considered when takingblood samples and interpreting the re-sults?

Numerous conditions and medicationsinfluence some thrombophilia param-eters. These especially include preg-nancy and puerperium, ovulation inhibi-tors, anti-coagulants, and acute-phasereactions. Potential influences are listedin Table 10.

The following parameters are not influ-enced:– Molecular genetics (factor V Leiden

mutation, prothrombin G20210A mu-tation)

– antiphospholipid antibodies (anti-cardiolipin antibodies, β2 glycopro-tein I antibodies) except lupus antico-agulants

Table 8. Indications for thrombophiliatesting

Thromboembolism at a young ageRecurrent thromboembolism of unclearoriginThrombosis in atypical location (sinusveins, mesentery veins, etc.)Suspicion of antiphospholipid antibodies(e.g. patient has systemic lupus erythe-matosus) or antiphospholipid syndromeThree or more spontaneous miscarriages(possibly two depending on individual de-mand and distress)StillbirthConsideration of oral contraception pre-scription with family history of thrombo-sis (first-degree or possibly second-de-gree relatives with thromboembolismbefore the age of 50)Pregnancy or planned pregnancy withown history of thromboembolism

Table 9a. Clinically relevant thrombo-philia markers

APC resistance test for factor V Leidenmutation, or genetic testing right awayfor factor V Leiden mutation

Prothrombin G20210A mutation

Antithrombin

Protein C

Protein S

Factor VIII

Antiphospholipid antibodies (lupus anti-coagulants, anti-cardiolipin antibodies,anti-β2-glycoprotein I antibodies)



Table 9b. Diagnostic tests for thrombophilia. Mod. from [Seligsohn U, Lubetsky A, Genetic Susceptibility to Venous Thrombo-sis. N Engl J Med 2001; 344: 1222–31].

Test Genetic basis for test Conditions or states that Factors that can distort determi-

can influence test results nation of thrombophilia para-

meters

High significance

APC resistance Factor V Leiden mutation (other Lupus anticoagulants, antibodies Lupus anticoagulants, thrombinpolymorphisms/mutations not against protein C. Only for certain inhibitors, direct factor Xa inhibitorspart of routine testing) test procedures: (pregnancy, oral vitamin K antagonists, heparin in

contraceptives, increased fac- high concentrations, coagulationtor VIII level, protein S deficiency) factor deficiencies (in part only for

certain test procedures)

Factor V Leiden mutation G1691A in exon 10 of factor V gene Genetic testing not suitable follow-(heterozygous/homozygous) ing liver or allogeneic stem cell

transplants

Prothrombin mutation G20210A in a non-coding area Genetic testing not suitable follow-of prothrombin gene ing liver or allogeneic stem cell

transplants

Elevated factor VIII level Physical and mental stress, Lupus anticoagulants, unfractionatedpregnancy, oral contraceptives, heparin in therapeutical doses,increased age, acute-phase thrombin inhibitors and direct fac-response, liver disease, corticoid tor Xa inhibitors can distort testtreatment, Cushing syndrome, procedures for factor VIII activityhyperthyroidism

Lupus anticoagulants Infectious diseases High heparin concentrations, vita-min K antagonists, thrombin inhibi-tors, direct factor Xa inhibitors

Anticardiolipin and β2-glyco- Infectious diseasesprotein I antibodies

Intermediate significance

Protein C deficiency >250 different mutations Acute thrombosis, treatment withvitamin K antagonists, vitamin Kdeficiency, liver disease, sepsis,disseminated intravascular coagu-lation, antibodies against protein C

Protein S deficiency >200 different mutations Acute thrombosis, treatment with Lupus anticoagulants, thrombin in-vitamin K antagonists, vitamin K de- hibitors, heparin and factor V Leidenficiency, pregnancy, oral contracep- mutation can distort test procedurestives, liver disease, malignancies, for protein S activitytreatment with asparaginase, sepsis,disseminated intravascular coagula-tion, chronic inflammatory intestinaldisease, HIV, nephrotic syndrome,antibodies against protein S(e.g. with lupus erythematosus)

Antithrombin deficiency >200 different mutations Acute thrombosis, heparin treat- Thrombin inhibitors and direct fac-ment, preclampsia, liver disease, tor Xa inhibitors can distort testsepsis, disseminated intravascular procedures for AT activitycoagulation, nephrotic syndrome,treatment with asparaginase, exu-dative enteropathy, major surgery

Low significance

Elevated homocysteine level Mutations in genes encoding Deficiencies of folic acid, vitamin B6methyltetrahydrofolate reductase or vitamin B12; increased age,(MTHFR) or cystathionine renal disease, smoking, hypothyroid-β-synthase ism, malignancies, medication

(e.g. MTX, phenytoin)

Dysfibrinogenemia >200 different mutations Liver disease, disseminated intra- Thrombin inhibitorsvascular coagulation



Depending on the test procedure, APCresistance can be influenced by preg-nancy and oral contraceptives as well asby high doses of anticoagulants. Othertest procedures are available, however,that are not subject to these influences(information can be provided by clinicallaboratories).

Care should therefore be taken to ex-clude the above-mentioned influences

on thrombophilia testing. If this is notpossible, always make sure to advisethe laboratory of the clinical situation(e.g. pregnancy week, current medica-tion) and have the results evaluated byan experienced hemostaseologist.

3.4.1. Diagnostic SamplesSee the notes in Table 11 on taking andtransporting blood samples for thrombo-philia testing.

3.5. Counseling Patients aboutRisk Factors3.5.1. GeneralClear information is key for counselingpatients and deciding on the right treat-ment and/or contraception. In Germany,legislation went into effect on Febru-ary 1st 2010 that stipulates extensivecounseling with patients before and aftergenetic screening. Counseling must in-clude discussion of the relevant findings

Table 10. Influence of different 2nd and 3rd-generation progestogens, progestogen-only preparations, pregnancy, vitamin Kantagonists, and acute phase reaction during thrombosis on different clotting parameters

Protein S Protein C Antithrombin D-dimer Lupus anti- Factor VIIIcoagulants

2nd-generation OVH decrease slight increase – slight increase – slight increasepossible possible possible possible

3rd-generation OVH decrease slight increase slight increase slight increase – increasepossible possible possible

Progestogen- increase slight decrease – – – –only Pill possible possible

Pregnancy and strong decrease – – marked increase – marked increasepuerperium (up to possible (slight increase with duration of6 weeks postpartum) possible toward pregnancy

end of pregnancy)

Vitamin K antagonists marked decrease marked decrease – increase possible false positive –(phenprocoumon, when drug use results possible,warfarin) stopped if diagnostic guide-

lines not followed(confirmation and

mixing tests)

Heparin treatment – – decrease possible increase false positive Test usuallywhen drug use results possible, aPTT-dependent,

stopped if diagnostic guide- therefore influ-lines not followed ence possible(confirmation and with unfrac-

mixing tests) tioned heparinsif no heparin

neutralizer used

Acute phase, consumption- consumption- slight decrease elevated – often stronglyacute thrombosis dependent slight dependent slight possible elevated

decrease possible decrease possible

OVH = Ovulation inhibitor; – = no influence

Table 11. Notes for taking and sending blood samples for thrombophilia tests

Test Material Note

Molecular genetic (factor V Leiden mutation, 1–5 ml EDTA blood sample is transportable, Do not centrifuge or freeze EDTA bloodprothrombin G20210A mutation) and can be sent by post sample!possibly other PCR analyses

Clotting tests (protein C, protein S, antithrom- 1–3 ml frozen citrate plasma, strongly Alternatively the citrate blood can be senttbin, lupus anticoagulants, APC resistence, centrifuge citrate tube 2× and pipette super- as whole blood to the lab via courier withinfactor VIII) natant without cells into neutral tube. 4 hours. Citrate whole blood should never

Store and transport at –20 °C. be frozen or refrigerated!

Homocysteine 1–3 ml EDTA blood. Immediately centrifuge Fasting blood samples should be taken.blood tube and store plasma separately becausehomocysteine can otherwise enter plasma fromerythrocytes and lead to false high results!If immediate centrifuging is not possible, storesample on ice. Use of special tubes (acid citrate,fluoride) can increase sample stability.



in Table 9a as well as of the followingclinical parameters. For acne patients,every elevated risk factor requires con-sideration of non-hormonal dermato-logical treatment to minimize individualrisk as much as possible.

Clinical parameters that must be con-sidered when counseling patientsabout hormonal contraception:

Age: The risk of suffering a thromboem-bolic event increases exponentially withage. The risk is approximately 1 in10,000 (0.01%) per year in those under40 years of age, approximately 1 in1,000 (0.1%) at the age of 60, and ap-proximately 1 in 100 above the age of80 (1%) [13–16, 57].

However, the VTE incidence rate forwomen of fertile age who do not takehormonal contraception has been cor-rected upwards over recent years, to ap-proximately 4 VTE per 10,000 woman-years [9].

Family history: Risk increases withpositive history of cardiovascular dis-ease in the parents below the age of 45.

Patient history: History of venous orarterial thromboembolism, localizationand degree of seriousness of throm-boembolism, causal connection withexogenous events, evaluation in contextof family history, bodily status, addi-tional risk factors, and laboratory con-firmed thrombophilia.

Contraception duration: VTE developsmost frequently during first year of OCintake [9, 57].

The following individual risk factors inpatient history (Caution: For long-termOC use, the potential for these symptoms/conditions to appear/reappear must beregularly determined, and indications forcontinued use reevaluated).

Cardiac disease: coronary heart dis-ease, cardiac insufficiency, valvularheart disease, atrial fibrillation (excludehyperthyroidism).

Thyroid: thyroid dysfunction (hyper/hypothyroidism).

Cigarette smoking: Although severalstudies have described cigarette smok-

ing as increasing the risk of VTE [4, 58–66] it probably cannot be viewed as a rel-evant risk overall [67–69]. A greater riskof arterial cardiovascular disease, how-ever, must be considered.

Obesity: Increased risk of VTE and arte-rial cardiovascular disease.

Also, as body weight increases, contra-ceptive effectiveness decreases for bothImplanon® (should be removed or re-placed before three years in overweightwomen, see summary of product charac-teristics – SPC) and Evra® (≥ 90 kg, seeSPC). This has also been discussed fororal contraceptives: An association wasnot found in the EURAS study, but theINAS study in the USA, with a high per-centage of overweight participants, hasshown a decrease in effectiveness forwomen of BMI > 35 that is statisticallysignificant but of minor clinical relevance[70, 71].

Immobilization: Increased risk withimmobilization (e.g. following accidentsor surgery with long periods of bed rest),plaster casts, lack of activity due to acuteinfections or inflammatory diseases.Hormonal contraception need not be in-terrupted before surgery if the surgeon isnotified and appropriate pre- and post-operative heparin treatment is adminis-tered (see also section 2.4.3).

Lipid metabolic disorder: Increasedrisk of arterial cardiovascular disease.

Diabetes mellitus: Increased risk of ar-terial cardiovascular disease. See alsocontraception recommendations for pa-tients with diabetes mellitus.

Arterial hypertension: Increased riskof arterial cardiovascular disease.

Malignancies, myeloproliferative dis-eases: Increased VTE risk, in part alsogreater risk of arterial thromboembolism.

Nephrotic syndrome: Increased risk ofvenous and arterial thromboembolism.

Migraines: Increased risk with first oc-currence or worsening of migraines,marked hemiplegic and/or crescendocharacter, scotoma.

Lupus erythematosus: Increased riskwith inflammatory reactions, which can

affect all parts of the body includingskin, joints and organs. Increased risk ofvenous and arterial thromboembolismespecially if antiphospholipid antibodiesare present.

Postpartum: Increased VTE riskshortly after giving birth! Risk: 51 per10,000 births in the first three monthspostpartum [72].

See also recommendations for womenwho are nursing!

3.5.2. Patient Counseling about GeneticScreeningAs of February 1, 2010, the GeneticDiagnosis Act (Gendiagnostikgesetz, orGenDG) in Germany stipulates that pa-tients must receive appropriate counsel-ing and provide written consent beforegenetic testing is done. Once the resultsare obtained, patients must receive ap-propriate counseling again, which mustbe documented, from a specially quali-fied physician (e.g. human geneticist orother specialist with relevant additionalqualification).

Key information from the GenDG4 is asfollows (see footnote for German/English versions5).

The German Genetic Diagnosis Act(Gendiagnostikgesetz, GenDG) wentinto effect on February 1, 2010. Theaim of this legislation is “to determinethe requirements for genetic examina-tions … and to prevent any discrimina-tion and disadvantage based on geneticcharacteristics, especially in regard tothe duty of the state to protect humandignity and to ensure the individual rightto self-determination via sufficient in-formation” (§1 GenDG). The Act hasspecial consequences for all physicianswho perform or initiate genetic analyses.Genetic testing and counseling must beboth initiated and performed by physi-cians. A distinction must be drawn be-tween diagnostic and predictive testing.Predictive analyses require consultationwith a physician who specializes in hu-man genetics.

4 www.gesetze-im-internet.de/bundesrecht/gendg/gesamt.pdf5 https://www.eshg.org/fileadmin/www.eshg.org/d o c u m e n t s / E u r o p e / L eg a l W S / G e r m a n y _GenDG_ Law_German_English.pdf.



Duty to inform (§9): Before every ge-netic test, the physician in charge mustinform the patient about the purpose,type, scope and significance of the test.The GenDG stipulates that all remainingsample material be destroyed immedi-ately after the test is concluded, and thedocumentation must be destroyed after10 years. Patients must be informed thatthey can choose to have the documenta-tion kept for longer periods of time. Pa-tients must also be informed about risksassociated with the testing, about theirright not to be informed, and about theright to revoke their consent. This infor-mation can be provided via suitablereading material, or in person. The infor-mational content must be documented inwriting.

Consent (§8): Patients must sign thatthey have received adequate informationand that they agree to have the plannedgenetic analysis performed. Consentmust also clarify whether the results willbe provided to further individuals be-sides the physician. Here too the patientshave the right to revoke their consent.

Counseling (§10): When the results areavailable, the physician is to provide thepatient with counseling based on spe-cialist genetic knowledge. If the resultsyield signs of genetically conditionedillnesses other than the original indica-tion for genetic testing, consultationwith a physician specializing in humangenetics is to be offered.

Consultation with a specialist in humangenetics or medical doctor with certifi-cation in genetic examinations is re-quired both before and after a predictivegenetic test. This consultation shouldalso and especially cover possible medi-cal, psychological and social issues con-nected with the test and the results. Thepatient should also be informed of sup-port measures for psychological andphysical difficulties. The physician mustdocument the content of the consulta-tion.

3.5.3. Interpreting the Laboratory ResultsThis section includes basic informationabout thrombophilic parameters, as wellas prevalence, associated VTE risk, andchanges thereto with use of hormonalcontraception, and also indications forthrombosis prevention. In addition to themedication described here for VTE risk

situations (e.g. surgery, inactivity duringacute illness, infections, immobilizationof extremities via e.g. plaster casts),physical measures should also always betaken (early exercise and/or preventionof inactivity during illness whereverpossible, as well as prophylactic stock-ings).

Hormonal contraceptives increase therisk of VTE, especially for women withthrombophilia. Because side effects andcost make it inadvisable to take throm-bosis prevention medication on a con-tinuous basis together with contracep-tives in order to reduce the risk of VTE,the choice of contraception is especiallyimportant for women with thrombo-philia.

3.5.3.1. Factor V Leiden MutationFactor Va is normally inactivated by acti-vated protein C (APC). The factor VLeiden mutation is marked by guanineinstead of adenine in nucleotide position1691 in the factor V gene. This in turndestroys a cleavage site for APC in thefactor V molecule (FVR506Q). Thechanged structure in the Leiden variantof factor V impairs the degradation offactor Va by APC (factor V becomes “re-sistant” to inactivation by APC) and fac-tor Va retains its coagulation-promotingeffect. This leads to an imbalance be-tween coagulation-inhibiting and -pro-moting influences, which in turn in-creases the tendency for thromboses todevelop (thrombophilia).

The APC-resistant phenotype can be de-termined in plasma samples. Tests forthe APC-resistant phenotype show a sen-sitivity and specificity of 98–100% forfactor V Leiden mutation. The genotype(factor V Leiden mutation, hetero- orhomozygous) is determined by molecu-lar genetic testing. It is an autosomaldominant hereditary condition.

Heterozygous carriers are found in thegeneral European population with a fre-quency of 3–13%, homozygous carrierswith a frequency of 0.2–1% [73]. InAsians and Africans, by contrast, themutation occurs rarely (< 1%, [73]).Factor V Leiden mutation is commonlyfound in European VTE patients (10–50%).

Heterozygous carriers have an approxi-mately 5-fold increased risk of VTE

(95% confidence interval 4.4-5.5) [12].The risk of thrombosis with homozy-gous factor V Leiden mutation was longoverestimated. One study calculated an80-fold increased VTE risk for homo-zygous carriers [74]. However, severalstudies and a meta-analysis by Gohil etal. have shown that the relative VTE riskin homozygous carriers is “only” ap-proximately 10 times increased (95%confidence interval 6.7–13.3) [12, 75].A meta-analysis by Segal et al. [76] de-termined a relative risk of 17.8 (7.98–39.89) for VTE occurrence in homozy-gous family members of patients withknown factor V Leiden mutation.

A meta-analysis by Wu et al. [77] calcu-lated the VTE risk (odds ratio) forwomen with factor V Leiden mutationtaking oral contraception. Due to thesmall number of homozygous carriers,this meta-analysis unfortunately couldnot determine separate risk values forhetero- and homozygous carriers. Thepooled analysis of hetero- and homozy-gous carriers (with a very small numberof homozygous cases) taking oral con-traception showed a 15.6-fold increasedVTE risk (95% confidence interval 8.7–28.2) [77]. The VTE risk for homozy-gous carriers taking oral contraceptionhas thus far not been sufficiently studied,but is presumably considerably higherthan shown by the pooled analyses.

Continuous thrombosis prophylaxis isnot necessary for patients who have nothad VTE. Thrombosis prophylaxis inVTE risk situations consists of low mo-lecular-weight heparin or fondaparinux.New oral anti-coagulants such as Riva-roxaban or Dabigatran can be used, butthus far are only authorized for VTE pro-phylactic purposes for major orthopedicsurgery. Dosing is done in accordancewith the German S3 guideline on pre-venting venous thrombosis (www.awmf.org). The significance of factor V Leidenmutation for the VTE recurrence riskand the duration of anticoagulation treat-ment post-VTE is discussed below.

3.5.3.2. Prothrombin G20210A Muta-tionProthrombin is the proenzyme of theserine protease thrombin, which con-verts fibrinogen to fibrin.

Substitution of adenine for guanine inposition 20210 of the prothrombin gene



leads to higher plasmatic prothrombinlevels, and is associated with an approxi-mately three-fold higher risk of throm-bosis (95% confidence interval 2.2–3.5)for heterozygous carriers [12]. This mu-tation is determined exclusively by mo-lecular biological methods.

Heterozygous carriers are found in 1.7–3.0% of the general European popula-tion. Homozygous carriers are very rare(< 0.1 %, Rosendaal et al. [74]). On ac-count of this low prevalence, the data isthus far not sufficient for estimating theVTE risk for homozygous carriers. Aheterozygous prothrombin G20210Amutation is found in 7–16% of patientswith VTE.

A meta-analysis by Emmerich et al. [54]showed a relative VTE risk of 7.14 (95%confidence interval 3.4–15.0) for womentaking oral contraception; a meta-analy-sis by Wu et al. [77] showed a relativerisk of 6.1 (95% confidence interval 0.8–45.6). These meta-analyses includeddata from both hetero- and homozygouscarriers although with only a small num-ber of homozygous participants. TheVTE risk for homozygous carriers tak-ing oral contraceptives is presumablyconsiderably higher, but has thus farbeen little studied.

The same thrombosis prophylactic mea-sures should be taken here as for factor VLeiden mutation.

3.5.3.3. Compound Heterozygotes forFactor V Leiden and ProthrombinG20210A MutationsIf both prothrombin G20210A and fac-tor V Leiden mutations are present, thereis a 4 to 15-fold increased relative risk ofVTE (Wu et al. [77]: OR = 4.0 with a95% confidence interval of 1.0–16.0,Emmerich et al. 2001 [56]: OR 14.7 witha 95% CI of 3.5-62.0; these analyses in-cluded hetero- and homozygous carriersbut with a small overall number of ho-mozygous cases).

The odds ratio for VTE with oral contra-ception for individuals carrying both theprothrombin G20210A and factor VLeiden mutations is 8-17 (Wu et al. 2005[77]: OR = 7.9 with a 95% confidenceinterval of 1.7–37.4; Emmerich et al.[56]: OR = 17.0 with a 95% CI of 3.6–72.8; these analyses included hetero-and homozygous carriers but with a

small overall number of homozygouscases).

The same thromboprophylaxis measuresshould be taken here as for factor VLeiden mutation.

3.5.3.4. Antithrombin DeficiencyAntithrombin (AT) is the major antago-nist of thrombin, although it also inhibitsother coagulation factors such as IXa,Xa and XIa. The effect of AT is acceler-ated multiple times by heparin.

Hereditary AT deficiency can result fromreduced AT production. Blood levelsshow a parallel reduction of AT antigenand AT activity (type I AT deficiency). Intype II AT deficiency, AT moleculesform that show limited heparin- orthrombin-binding capacity; this type ischaracterized by lower AT activity whilethe antigen concentration is largely nor-mal. With rare exception the patients areheterozygous carriers. Thus far only afew homozygous carriers with type II ATdeficiency have been described [78].This is due to embryogenic lethality ofserious congenital AT deficiency.

Congenital AT deficiency is found in ap-proximately 0.2% of the general popula-tion and in approximately 1–3% of pa-tients with VTE.

AT deficiency is diagnosed by repeatedtesting of AT activity. This procedureidentifies not only type II but also type IAT deficiency.

It may also be necessary to determine theAT antigen concentration and do mo-lecular biological testing to find the typeof AT deficiency.

Family screening can be helpful in deter-mining hereditary AT deficiency. Ge-netic testing is the only way to conclu-sively demonstrate homozygosity, be-cause homozygous carriers can show ATactivity comparable to heterozygous car-riers. Thus far more than 270 differentmutations of the AT gene (SERPINC1)are known that can lead to AT deficiency[78]. Inheritance is generally autosomaldominant.

Before undertaking time-intensive andhigh-cost diagnostic procedures, how-ever, acquired AT deficiency should beexcluded. Antithrombin levels are often

reduced with acute thromboembolism.The same is true for heparin treatment,impaired liver synthesis, heightened ATconsumption due to surgery or trauma,and protein loss via the kidneys (neph-rotic syndrome) or intestines.

VTE risk depends on the type of AT defi-ciency. Patients with type II HBS (hep-arin-binding defect) AT deficiency havea lower thrombosis risk than patientswith type I or other forms of AT defi-ciency. The relative risk of thrombosislies between 4 and 50.

The VTE risk for hereditary AT defi-ciency and use of oral contraception hasthus far been little studied. A meta-analysis by Wu et al. [77], which coveredonly two studies and did not differentiateamong AT deficiencies, yielded an oddsratio of 12.6 (95% CI 1.4–115.8) forwomen with AT deficiency taking oralcontraception.

Continuous thromboprophylaxis is gen-erally not necessary for patients whohave not suffered a thromboembolicevent. For thrombotic risk situations,however, care must be taken to ensuresufficient prophylactic measures. It isimportant to note that heparin-basedthromboprophylaxis is only of limitedeffectiveness for AT deficiency, becauseheparin needs AT to work. It is recom-mended to determine the type, durationand dose of thromboprophylaxis for e.g.surgery in consultation with an experi-enced specialist in hemostaseology.

3.5.3.5. Protein C DeficiencyAlong with thrombin, protein C binds tothe endothelial receptor thrombomodulinand thus becomes activated protein C(APC). Its anti-thrombotic effect derivesfrom cleaving factors Va and VIIIa as wellas from activating fibrinolysis. Protein Calso inhibits inflammation and apoptosis.Hereditary protein C deficiency is foundin 0.2–0.4% of the general populationand in 2–5% of VTE patients.

Hereditary protein C deficiency is foundby repeated determination of protein Cactivity combined with the exclusion ofacquired causes of protein C deficiency.Acquired protein C deficiency is ob-served most often in conjunction withacute thromboembolism, impaired liversynthesis, or treatment with vitamin Kantagonists. Sepsis, especially meningo-



coccal sepsis, can also lead to severe ac-quired protein C deficiency.

Molecular biological testing is only sel-dom needed to show hereditary protein Cdeficiency. Thus far more than 250 differ-ent mutations are known in the protein Cgene (PROC) that can lead to deficiency.Most individuals with protein C defi-ciency are heterozygotes and often sufferthromboembolism already as youngadults (autosomal dominant inheritance).Homozygous carriers show severe pro-tein C deficiency (protein C activity often< 1%) and usually already develop pur-pura fulminans, disseminated intravascu-lar coagulation and venous thromboem-bolism in the neonatal period.

VTE risk estimates for protein C defi-ciency differ strongly. Odds ratios of3–15 are given. The thrombosis risk forindividuals in affected families (proteinC deficiency plus thromboembolism inat least one family member) is gener-ally higher than in unselected patientgroups.

VTE risk with oral contraception for he-reditary protein C deficiency has thus farbeen little studied. A meta-analysiswhich covered only two studies calcu-lated an odds ratio of 6.3 (95% CI 1.7–23.9) [77]. This analysis, however, alsoplaces the risk of thromboembolism forprotein C deficiency without oral contra-ception, based on data from a singlestudy, at only 2.5 (95% CI 1.2–5.1). Astudy of family members of patientswith protein C deficiency found use ofhormonal contraceptives to be associ-ated with a relative thrombosis risk of23.6 (3.7–535.6) [79].

VTE prophylaxis for protein C-deficientpatients in thrombosis risk situationsgenerally consists of low molecular-weight heparin or fondaparinux.

If a vitamin K antagonist should beused to treat VTE, it must be noted thatrapid decline in the already low proteinC levels can lead to coumarin necrosis.To prevent this, coumarin must bedosed very low at the beginning andheparin must be administered untilthe treatment-appropriate INR range isreached.

Protein C concentrate (Ceprotin®) is in-dicated in the case of severe protein C

deficiency with purpura fulminans orcoumarin necrosis, as well as short-termprophylaxis for surgery or at the start ofcoumarin treatment.

3.5.3.6. Protein S DeficiencyProtein S is also a coagulation inhibitor.Protein S is a co-factor in the inactiva-tion of factor Va and VIIIa by activatedprotein C.

Protein S deficiency occurs considerablymore often in acquired than hereditaryform. It occurs under oral contraception,in pregnancy, with acute thromboembo-lism, impaired liver synthesis, treatmentwith vitamin K antagonists, inflamma-tory bowel disease and HIV.

Hereditary protein S deficiency is foundby repeated determination of free pro-tein S antigen levels or protein S activityin plasma, combined with exclusion ofacquired protein S deficiency. Molecularbiological testing of the protein S gene(PROS1) is only rarely necessary, but itcan help differentiate acquired from he-reditary protein S deficiency and alsoshow homozygous inheritance. A prob-lematic aspect is that currently availablemethods only enable a mutation to be de-termined in approximately 50% of cases.Family testing can therefore be helpfulin determining hereditary protein S defi-ciency. Thus far more than 200 differentmutations in the PROS1 gene are knownthat can lead to protein S deficiency. In-heritance is generally autosomal domi-nant, and affected individuals are usu-ally heterozygous mutation carriers. Ho-mozygous or compound heterozygouscarriers are very rare and often alreadysuffer purpura fulminans and recurrentVTE in the neonatal period.

Hereditary protein S deficiency is foundin 0.2–2% of the general population andin 1–7% of VTE patients. VTE risk esti-mates for protein S deficiency differstrongly. Odds ratios of 5–11 are given.The thrombosis risk for individuals inaffected families (protein S deficiencyplus thromboembolism in a family mem-ber) is generally higher than for unse-lected patient groups.

VTE risk with hereditary protein S defi-ciency with oral contraception has thusfar been little studied. A meta-analysisby Wu et al. [77] that covered only twostudies calculated a VTE odds ratio for

protein S deficiency with oral contracep-tion of 4.9 (95% CI 1.4–17.1).

Continuous thromboprophylaxis is notnecessary for patients who have not hadVTE. Thromboprophylaxis in VTE risksituations consists of low molecular-weight heparin or fondaparinux. Dosingis done in accordance with the GermanS3 guideline on preventing venousthrombosis (www.awmf.org).

Protein S deficiency can also lead tocoumarin necrosis at the start of cou-marin treatment, so as in the case of pro-tein C deficiency, care must be taken toensure gradual coumarin dosing whentreatment starts.

3.5.3.7. High Factor VIII LevelsFactor VIIIa is a co-factor in the activa-tion of factor X by factor IXa. As such, ithas a pro-coagulatory effect.

High factor VIII levels are found by re-peated determination of factor VIII ac-tivity in plasma.

A large number of factors can lead to atemporary rise in factor VIII levels, suchas acute phase reactions, especiallyacute and chronic infections as well asauto-immune diseases, acute throm-boembolism, pregnancy, malignancies,liver diseases and medication. FactorVIII is also influenced by blood type,age and weight (increases with age andBMI: [80]). Persistent high levels of fac-tor VIII are associated with a higher riskof thrombosis.

Increased factor VIII activity is found inapproximately 5–10% of the generalpopulation and in approximately 10–30% of VTE patients. Patients withheightened factor VIII have an approxi-mately 5 to 8-fold greater risk of VTE[77, 81].

The relative VTE risk with oral contra-ception for higher factor VIII levels is8.8 (4.1–18.8) to 13.0 (4.9–34.3) [77,81, 82].

VTE prophylaxis for individuals withhigher factor VIII levels in VTE risk situ-ations generally consists of low molecu-lar-weight heparin or fondaparinux.Dosing is done in accordance with theGerman S3 guideline on preventingvenous thrombosis (www.awmf.org).



3.5.3.8. Antiphospholipid SyndromeAntiphospholipid antibodies are a hetero-geneous group of antibodies againstphospholipid protein complexes. Basedon the current state of research, the rel-evant antiphospholipid antibodies are lu-pus anticoagulants, anticardiolipin anti-bodies and β2-glycoprotein I antibodies.They are associated with antiphospho-lipid syndrome, or APS for short.

APS is defined by persistent evidence ofantiphospholipid antibodies in patientswith venous or arterial thromboembo-lism or pregnancy complications (≥ 3otherwise inexplicable miscarriages be-fore the 10th week of pregnancy, ≥ 1 mis-carriage or stillbirth with no unusualmorphological features in the ≥ 10th

week of pregnancy, or premature birthbefore the 34th week due to placental in-sufficiency or [pre]clampsia).

Lupus anticoagulants were first de-scribed in patients with systemic lupuserythematodes (SLE). The term is mis-leading, because the antibodies occurnot only in connection with SLE, and es-pecially because the tendency is not to-ward bleeding but rather toward throm-bosis. There is a propensity for venousand arterial thromboembolism.

Eighty percent of APS patients arewomen. In the general population, lupusanticoagulants are found at a rate of0–1.7% [83], anticardiolipin antibodiesat 2.7–23.5% [83] and β2-glycoprotein Iantibodies at approximately 3% [84].Antiphospholipid antibodies are foundin 2–10% of patients with VTE.

Lupus anticoagulants are found by per-forming two screening tests followed bytwo confirmation tests. Anticardiolipinand β2-glycoprotein I antibodies aredetermined with the help of ELISA testprocedures. Anticardiolipin antibodiesare only viewed as a criterion for APS ifanticardiolipin IgG or IgM antibodiesshow at least a medium to high titer(> 40 GPL or MPL or titer > 99th percen-tile).

Antiphospholipid antibodies are presentin approximately 50% of cases associ-ated with other diseases (autoimmunediseases, especially systemic lupus ery-thematodes, malignancies, infections,drug-associated). They can occur on atransient basis in the course of infec-

tions. Excluding transient antibodies isthe reason for repeating the antibody di-agnostic procedure after 12 weeks, asprescribed by international guidelines.

Caution: Lupus anticoagulants extendthe aPTT. This is an in-vitro phenom-enon, which is typically not associatedwith a propensity for bleeding. Despitethe extended aPTT, there is a propensityfor thrombosis!

For patients without an underlying au-toimmune condition, the relative risksare as follows:– for VTE with:

● lupus anticoagulants: 4.1–16.2 [85]● anticardiolipin antibodies (medium

to high titer): 0–2.5 [85]● β2-glycoprotein I antibodies: 2–4

[84, 86]– for arterial thrombosis:

● lupus anticoagulants: 8.7–10.8 [85]● anticardiolipin antibodies (medium

to high titer): 0–18.0 [85]● β2-glycoprotein I antibodies: 0–8

[85].

Patients with positive antiphospholipidantibodies in multiple tests (lupus anti-coagulants + anticardiolipin antibodies+ β2-glycoprotein I antibodies) have thehighest risk of thrombosis [87].

The risk of venous and arterial throm-boses from oral contraceptives with thepresence of antiphospholipid antibodieshas not yet been sufficiently studied. Thealready existing risk of cerebral is-chemia has been found to increase ap-proximately 5-fold, and the risk of myo-cardial infarction approximately 4-fold[88].

Prevention:Acetyl salicylic acid should be consid-ered as thromboprophylaxis for patientswith systemic lupus erythematodes andpersistent demonstrated antiphospho-lipid antibodies [89].

Modification of classic reversible car-diovascular risk factors such as arterialhypertension and hypercholesterolemiawhere applicable, in order also to reducethe risk of arterial thrombosis [89].

Because antiphospholipid antibodies oc-cur on a secondary basis in 50% of cases,it can be necessary to clarify their gen-esis.

Thromboprophylaxis in VTE risk situa-tions for persons who have not hadthromboembolism consists of low mo-lecular-weight heparin or fondaparinux.Dosing is done in accordance with theGerman S3 guideline on preventingvenous thrombosis (www.awmf.org).

VTE with antiphospholipid syndrome isan indication for long-term anticoagula-tion. When treating VTE, it must be con-sidered that lupus anticoagulants usuallyextend the aPTT and that aPTT is notsuitable for monitoring purposes whenunfractionated heparin is administered.

3.5.3.9. Mild HyperhomocysteinemiaHomocysteine is an intermediate prod-uct of amino acid metabolism, formedby demethylation of the amino acidmethionine. The amino acid cysteine canbe formed from homocysteine via theenzyme cystathione β-synthase (CBS).Methionine can be synthesized from ho-mocysteine via the enzyme methyl-tetrahydrofolate reductase (MTHFR).Folic acid and vitamins B12 and B6 areco-factors in homocysteine metabolism.

Hyperhomocysteinemia is associatedwith VTE (VTE in the leg, pulmonaryembolism) and cardiovascular eventssuch as myocardial infarction and stroke.

Possible causes of hyperhomocystein-emia are: folic acid and/or vitamin B12or B6 deficiencies, polymorphism/muta-tion in the genes encoding MTHFR orCBS, renal insufficiency, nicotine abuse,high coffee consumption, medication(e.g. methotrexate, theophylline, anti-convulsives), hypothyroidism, and oth-ers.

Mild hyperhomocysteinemia is found inapproximately 11% of women in Europeaged 20–40 years [90]. Mild hyper-homocysteinemia is found in 6–30% ofpatients with VTE.

An increase of 5 µmol/l in the homocys-teine level is linked with a venous throm-bosis risk of 1.3 (95% CI 1.0–1.6) [91].The odds ratio is 1.2 (95% CI 1.1–1.3)for coronary heart disease, and 1.8 (95%CI 1.6–2.0) for ischemic stroke [92].

The thromboembolism risk for individu-als with hyperhomocysteinemia who aretaking oral contraception has not beensufficiently studied. Due to the only



slightly higher risk of venous thrombo-sis in connection with mild hyperhomo-cysteinemia and the insufficient data inconnection with hormonal contracep-tion, it is generally not recommended todetermine homocysteine levels in con-nection with use of the Pill. One study,however, has shown an increased risk ofcerebral ischemia for hyperhomocys-teinemia and oral contraception (OR 6.2;95% CI 1.7–22.0 [93].

Preventive measures in the case ofknown mild to medium hyperhomocys-teinemia:

Homocysteine levels can be lowered bysubstituting folate, vitamin B12 andvitamin B6, but not the occurrence ofVTE or cardiovascular events (myo-cardial infarction, stroke). The benefitof vitamin intake has not been demon-strated [94–97].

Clarification of hyperhomocysteinemiagenesis where applicable to excludetreatable causes. Modification of classiccardiovascular risk factors where appli-cable. On account of their homocys-teine-elevating effects, no consumptionof nicotine, no or only low consumptionof coffee.

3.5.3.10. Homozygous Methyltetra-hydrofolate Reductase (MTHFR) C677TPolymorphismC677T polymorphism leads to reducedMTHFR enzyme activity. Compared toCC genotype carriers, TT carriers haveapproximately 25% higher homocys-teine concentrations (approx. 2.5 µmol/l) in plasma. The effect on homocysteinelevel depends on folate intake.

Approximately 10% of Europeans areMTHFR 677 TT carriers [98].

VTE risk is not significantly influencedeven by MTHFR677 polymorphism inhomozygous form (odds ratio for Euro-peans: 1.1 with 95% CI 0.97–1.2, notsignificant [12]; the relative risk of anischemic stroke is 1.4 [95% CI 1.1–1.8],TT versus CT and CC genotype carriersamong Europeans; [99]). The risk of myo-cardial infarction is not increased [100].

The influence of MTHFR C677T poly-morphism on the risk of thrombosis withoral contraception has not been suffi-ciently studied. Due to insufficient data,

molecular genetic testing for MTHFRC677T polymorphism is not currentlyapplicable to Pill use in routine cases.Further studies are needed to clarifywhether MTHFR C677T polymorphismis relevant to prescription decisions forpatients with a higher risk of cerebral is-chemia. One study calculated a relativerisk of ischemic cerebral insult of 5.4(95% CI 2.4–12.0) for homozygous car-riers of MTHFR 677 polymorphism withoral contraception [101]; another studyshowed an odds ratio of 8.9 (95% CI3.7–21.1 [102].

3.5.3.11. Significance of ThrombophiliaParameters Post-VTEAn increased risk of VTE recurrencecould not be clearly demonstrated for theprothrombin G20210A mutation [76,103]. Carriers of homo- or heterozygousfactor V Leiden have a slightly higherrisk of VTE recurrence (odds ratio 1.56,95% confidence interval 1.14–2.12 and2.65, 95% CI 1.2–6.0) [76]. Individualswith antithrombin, protein C or protein Sdeficiency show a higher VTE recur-rence risk [104]. Thus far, however,there are no studies that show benefits ofextended anticoagulation for patientswith these thrombophilia parameters.National and international guidelinesthat specify post-VTE anticoagulationduration therefore currently do not takethrombophilia diagnostics into account.By contrast, long-term anticoagulationis recommended for patients with anti-phospholipid syndrome (www.awmf.org/leitlinien/aktuelle-leitlinien/ll-liste/deutsche-gesellschaft-fuer-angiologie-gesellschaft-fuer-gefaessmedizin.html;[105].

3.6. General Recommendationsfor Apparently Healthy WomenIn individual cases, thrombosis can alsobe triggered in apparently healthy non-risk women (i.e. negative own and fam-ily histories, negative lab tests) by theuse of hormonal contraceptives, whichcan lead to pulmonary embolism and inrare cases to death. As with the at-riskpopulation, additional factors play arole, such as exsiccosis due to diarrheaand severe vomiting, immobilization,limited movement, and low-pressureconditions on long flights. Therefore allwomen (as well as men who face simi-lar risk constellations although not tak-ing the Pill!) should adhere to the fol-lowing.

3.7. General Recommenda-tions

3.7.1. Immobilization/InactivityFor surgery and plaster casts, as well asimmobilization/inactivity with “inter-nal” conditions such as infections: Pro-phylactic anticoagulants in accordancewith the German S3 guideline on pre-venting venous thromboembolism(www.awmf.org/leitlinien/aktuelle-leitlinien/ll-liste/deutsche-gesellschaft-fuer-chirurgie.html)!

Air or car travel > 4 hours: Thrombo-prophylaxis with low molecular-weightheparin only for at-risk individuals!Compression stockings depending onrisk constellation. For this see section2.4.2. Car travel should include regularstops with exercise.

3.7.2. Air TravelClinics at the Munich and Frankfurt air-ports can provide counseling and treat-ment. For long-distance flights, check inand boarding should be done earlyenough to place carry-on luggage in theoverhead compartment instead of underthe seat so as to ensure leg room. An ex-ercise program to promote circulation isrecommended, including regular legmovement and occasional walking ifpossible. Seats at aisles or emergencyexits are preferable due to greater legroom. Avoid consumption of alcohol orsleeping tablets on long trips.

In general, sufficient hydration shouldbe ensured during extended periods ofconfinement (e.g. on trips or duringother periods of inactivity/immobiliza-tion). See next section. At least 0.25 literof water should be consumed every 2hours during air travel [40].

3.7.3. HydrationThe human body consists of up to 70%water. This percentage is slightly lowerfor women than for men on account ofthe relatively higher fat content in tissue.To prevent dehydration, at least 1–3 li-ters (at least 3% of body weight) of waterneed to be consumed per day. Fluid lossof 2% already leads to a decline in physi-cal performance, concentration andshort-term memory.

An adult body loses 200–400 ml of wa-ter daily via the skin, 400–600 ml via thelungs, 1500 ml via the urinary system



and 100 ml via the intestinal system. Incertain circumstances the body can loseconsiderably more fluid via the skin(sweat) such as during strenuous exer-cise or fever.

Daily fluid requirements for adults withnormal hydration status is 30–40 ml perkilo of body weight per day. For a bodyweight of 70 kg, that would be approxi-mately 2.5 liters/day. In the event offever, daily adult fluid needs rise byapproximately 10 ml per kilo of bodyweight per 1 °C temperature above 37 °C(S3 guideline from the Deutsche Gesell-schaft für Ernährungsmedizin: www.awmf.org). Diarrhea and vomiting can alsocause major fluid loss, which must berapidly restored.

If intake is lower than output, there is adanger of exsiccosis: sensation of thirst,dry skin and mucous membranes, dimin-ished skin turgor, decreased urine pro-duction, elevated serum and urine mo-larity, headache, nausea, paresthesia,muscle spasms, tachycardia, hyperten-sion, increased temperature, weight loss,followed by agitation up to and includ-ing delirium. The clinical symptoms in-crease with the extent of fluid loss.

A useful rule of thumb for preventing de-hydration in hot environments or duringstrenuous physical activity is to monitorthe frequency and type of urination. Emp-tying a full bladder containing colorlessor lightly colored urine at least every 3–5hours shows the absence of dehydration.

If input exceeds output volume, there isa danger of edema (legs, lungs), ascites,combined with weight gain and possi-bly arterial hypertension. It is oftennecessary to limit the amount of fluidintake for conditions such as cardiac,liver and renal insufficiency, as well asedema.

3.7.4. Risk FactorsAvoid additional risk factors such assmoking, excess weight and malnutri-tion. Regular exercise is extremely im-portant. If other clotting disorders areidentified in addition to factor V Leiden,individual consultation with a hemo-staseology specialist is recommended.

3.7.5. ContraceptionUse of oral hormonal contraception inparticular, but also vaginal rings (Nuva

Ring®), hormonal patches (Evra®) orother steroid hormones increases the riskof thrombosis. The appropriate contra-ception depends on the individual riskprofile, and should be determined inconsultation with the gynecologist andpossibly also with a hemostaseologist.

3.7.6. PregnancyWomen with a positive own or familyhistory of venous or arterial thrombo-embolism should consult with their phy-sicians when planning for pregnancyin order to assess the risk of thrombosis,clarify the existence of additional riskfactors and discuss prophylactic mea-sures and possible treatment duringpregnancy and puerperium. Women whotake vitamin K antagonists should be in-formed of the teratogenic properties ofthese preparations.

3.7.7. Patient InformationCoagulation defects such as heterozy-gous factor V Leiden mutation are com-mon, with the latter occurring in 5% ofthe population. The majority of factor VLeiden mutation carriers do not knowthat they have a higher individual risk.Multiple factors are generally involvedin VTE, which means that it is very im-portant for preventive purposes to iden-tify additional acquired risk factors. Es-pecially women taking hormonal contra-ception should be informed of additionalthromboembolism risk factors so thatappropriate prophylactic measures canbe taken in situations with a higher riskof thrombosis.

3.8. Summary

3.8.1. Laboratory Screening– Thrombophilia screening of all

women taking hormonal contracep-tion is not justified.

– Indications for thrombophilia test-ing: Thromboembolism at a youngage, the need to clarify otherwise un-explained recurrent thromboses/em-bolism, thromboses in unusual loca-tions, consideration of oral contra-ception prescription with family his-tory of thrombosis, suspicion of anti-phospholipid antibodies, post-throm-boembolic patients with planned orcurrent pregnancy, ≥ 3 miscarriages,stillbirth (Tab. 8).

– Thrombophilia testing plays only asecondary role in deciding the dura-tion of anticoagulation post-VTE.

– Genetic Diagnosis Act (GenDG):Genetic testing for thrombophiliamust comply with the GenDG patientinformation requirements beforehandas well as counseling and documenta-tion regarding the results.

– Costs: Covered by country-specifichealth insurance policies or on a pri-vate basis.

3.8.2 Patient Information/Counseling– General recommendations for pre-

venting thromboses in healthy indi-viduals have been described in detail.

– VTE is generally a product of mul-tiple factors. Ongoing thrombopro-phylaxis is not necessary for asymp-tomatic individuals with thrombo-philia. Additional acquired risk fac-tors can trigger VTE. Informationshould therefore be provided aboutVTE risk factors and care should betaken to ensure thromboprophylaxisin risk situations.

– Contraception, in particular oral con-traceptives, increase the risk of VTE.Because continuous thrombopro-phylaxis is not an option for individu-als taking hormonal contraception, itis crucial to assess individual risk ofthromboembolism in deciding for oragainst the use of hormonal contra-ception.

– As a basis for counseling patients, thispaper compiles VTE risk informationfor clinically relevant thrombophiliaparameters with and without use ofcontraception.

– Due to the low frequency of somethrombophilia parameters, e.g. homo-zygous factor V Leiden mutation orhomozygous prothrombin G20210Amutation, in some cases sufficientdata is not available to precisely as-sess VTE risk with oral contraception.

– Over and above this, it is crucial toselect a contraceptive with the lowestpossible cardiovascular risk (see Sec-tion 4).

– In all cases, the question of whetheror not to take an oral contraceptive isan individual decision for individualswith thrombophilia, and the diagnos-tic and treatment guidelines presentedhere are solely for informational pur-poses.

– Carrying a thrombophilia alert (in-cluding family history, patient his-tory, lab test results, risk assessment[like allergy, cortisone, etc. alerts]) ishelpful for correctly assessing throm-



bosis risk in situations such as emer-gency surgery.

4. Contraceptive Selection

for Women with Thrombo-

philia and/or Previous

Thromboembolism

4.1. Preliminary RemarksThis statement focuses on venous throm-boembolitic complications in women,with and without the use of various typesof contraception. Because epidemiologi-cal studies have also associated the useof combined oral contraceptives (COCs)with an increased risk of arterial throm-boembolism (myocardial infarction,transient ischemic attacks, ischemicstrokes), secondary attention is devotedto arterial thromboembolic events.

This statement concentrates on the riskassociated with thrombophilia – otherpotential risk constellations such as obe-sity, heavy smoking, PCO syndrome,diabetes mellitus, insulin resistance etc.have to be considered on an individualbasis – including the resulting diagnosticand treatment consequences. These rec-ommendations do not release physiciansfrom their professional duty to care foreach individual case, including provid-ing extensive information to the patientabout treatment options and their effectsand/or side effects. Neither the authorsnor the publishers assume any form ofliability for or related to the informationcontained herein (see disclaimer).

In assessing the VTE risk of differentcontraceptives it is important to knowVTE incidence rates for different agegroups in the healthy population as wellas for groups with additional risk factorssuch as obesity and positive family his-tory. The incidence rate for women whodo not use hormonal contraceptives hasbeen corrected upwards in recent years[9, 57, 106]. Comparative VTE risks forpregnancy and puerperium have also hadto be revised, because incidence rates(20–30 VTE/10,000 woman-years) areevidently higher than previously as-sumed [72, 107–109].

In its 2004 “Medical Eligibility Criteriafor Contraceptive Use” and its 2008,2009 and 2010 updates, the WHO hasanalyzed different patient health situa-tions and provided recommendations for

Table 12. Medical Eligibility Criteria for Contraceptive use. WHO 2004, 2008, 2009,2010.

Category Clarification

1 No restriction for the use of the contraceptive method Always usable2 Advantages of using the method generally outweigh the

theoretical or proven risks Broadly usably3 Theoretical or proven risks usually outweigh the advantages

of using the method Caution/Counseling4 Unacceptable health risk if the contraceptive method

is used Do not use

selecting the appropriate contraceptivemethods. These recommendations fallinto categories 1 through 4 (Tab. 12).

The final section of this paper will ad-dress the WHO recommendations ondeep venous thrombosis, pulmonary em-bolism, known thrombogenic mutations,superficial venous thromboses, ischemiccardiac diseases, stroke, hyperlipidemiaand systemic lupus erythematosus, in-cluding antiphospholipid antibodies(Tab. 13).

4.2. Thrombophilia Risk ofDifferent Contraceptives

4.2.1. Combined Oral Hormonal Contra-ceptives (COC)Cigarette smoking: VTE risk with theuse of oral hormonal contraceptivesshows a continuous increase with thenumber of cigarettes smoked per day[111] (Fig. 7). The odds ratio for > 20cigarettes is 1.9. However, not all studieshave shown similar results, and it isquestionable whether smoking in factrepresents a risk for VTE (see Section3.5.1).

Age and BMI: The EURAS studyshowed VTE incidence rates (VTE/10,000 woman-years) for non-over-weight women of 1.7 for women under25 years of age, 4.9 for women aged25–39, and 19.9 for women > 40, allwith a BMI < 25 (Fig. 8). For all three ofthese age groups, risk increases addi-tionally for women with higher BMI val-ues. For the age group under 25, forexample, risk increases from 1.7 VTE/10,000 woman-years for BMI < 25–7.7for BMI 25–30, and to 14.9 for BMI> 30 [Dinger 2008, personal communi-cation]. The risk from COC also dependson the following factors of influence.

Duration of use: The maximum riskfrom COC exists within the first 3

months following the start of use [11,113–116]. In the first half year there is a6 to 8-fold increase in thrombosis riskover baseline compared to that for asame-age comparative group [117]. It isimportant to note that after a break in use(due to planned pregnancy, break-up ofpartnership, lack of prescription), thereis once again an increased risk. Thisholds regardless of whether the same ora different preparation is taken [114].Short-term (e.g. 1–2 cycle) breaks in useshould therefore be viewed critically. Bycontrast, changing the preparation with-out interrupting use is not associatedwith increased risk [114].

Hemostatic changes caused by COC usereverse approximately 6 weeks after useis stopped [116]. This is confirmed bythe LASS study [118], which found noincreased risk following two months ofnon-use.

VTE risk is highest at the start of OC use,and decreases over time. If use is inter-rupted for 4 weeks, VTE risk is just ashigh as at the start of the first Pill phaseand then again shows a continuous de-crease (Fig. 9, 10) [114].

Estrogen dose: Risk depends on thedose of ethinyl estradiol (EE) [111, 119].– Dose reduction from 50 to 30–40 µg

EE: A lower risk been shown for areduction in dose from 50 to 30–40 µg ethinyl estradiol, although theresults are partially conflicting [111,118]. Epidemiological studies haveshown that VTE incidence in womenwithout other VTE risk factors whouse COCs with low estrogen content(< 50 µg ethinyl estradiol) is 20 to 40cases per 100,000 woman-years. Bycontrast, incidence in the typicalpopulation of OC users is approxi-mately 90 cases per 100,000 woman-years [9].



Table 13. Summary of classifications for hormonal contraceptive methods and intrauterine devices (categories see Tab. 12).Source: U.S. Medical Eligibility Criteria for Contraceptive Use, 2010. Mod. from [110].

Condition COC/P/R POP DMPA Implants LNG-IUD Cu-IUD

Smoking

a) Age < 35 yrs 2 1 1 1 1 1b) Age ≥ 35 yrs i) < 15 Cigarettes/day 3 1 1 1 1 1 ii) ≥ 15 Cigarettes/day 4 1 1 1 1 1

Obesity

a) ≥ 30 kg/m2 BMI 2 1 1 1 1 1b) Menarche to <18 yrs and ≥ 30 kg/m2 BMI 2 1 2 1 1 1

Cardiovascular Disease 3/4 2 3 2 2 1(Multiple risk factors for arterial cardiovasculardisease such as older age, smoking, diabetes,and hypertension)

Hypertension

a) Adequately controlled hypertension 3 1 2 1 1 1b) Elevated blood pressure levels (properly takenb) measurements) i) Systolic 140–159 mmHg or diastolic 90–99 mmHg 3 1 2 1 1 1 ii) Systolic ≥ 160 mmHg or diastolic ≥ 100 mmHg* 4 2 3 2 2 1c) Vascular disease 4 2 3 2 2 1

History of high blood pressure during pregnancy 2 1 1 1 1 1(current blood pressure measurable and normal)

Deep venous thrombosis (DVT)/

pulmonary embolism (PE)

a) History of DVT/PE, not on anticoagulant therapy i) Higher risk for recurrent DVT/PE (≥ 1 risk factor) 4 2 2 2 2 1

– History of estrogen-associated DVT/PE– Pregnancy-associated DVT/PE– Idiopathic DVT/PE– Known thrombophilia, including antiphospholipid syndrome– Active cancer (metastatic, on therapy, or within 6 months after clinical remission), excluding non-melanoma skin cancer– History of recurrent DVT/PE

ii. Lower risk for recurrent DVT/PE (no risk factors) 3 2 2 2 2 1b) Acute DVT/PE 4 2 2 2 2 2c) DVT/PE and established on anticoagulant therapyc) for at least 3 months i) Higher risk for recurrent DVT/PE (≥ 1 risk factor) 4 2 2 2 2 2

– Known thrombophilia, including antiphospholipid syndrome– Active cancer (metastatic, on therapy, or within 6 months after clinical remission), excluding non-melanoma skin cancer– History of recurrent DVT/PE

ii) Lower risk for recurrent DVT/PE (no risk factors) 3 2 2 2 2 2d) Family history (first-degree relatives) 2 1 1 1 1 1e) Major surgery i) With prolonged immobilization 4 2 2 2 2 1 ii) Without prolonged immobilization 2 1 1 1 1 1f) Minor surgery without immobilization 1 1 1 1 1 1

Known thrombogenic mutations*

(e.g. factor V Leiden; prothrombin mutation;protein S, protein C, and antithrombin deficiencies) 4 2 2 2 2 1

Superficial venous thrombosis

a) Varicose veins 1 1 1 1 1 1b) Superficial thrombophlebitis 2 1 1 1 1 1

Current and history of ischemic heart disease* I C I C I C

4 2 3 3 2 3 2 3 1

Stroke* (history of cerebrovascular accidents) I C I C

4 2 3 3 2 3 2 1

Known hyperlipidemias 2/3 2 2 2 2 1

Rheumatic Diseases

Systemic lupus erythematosus* I C I C

a) Positive (or unknown) antiphospholipid antibodies 4 3 3 3 3 3 1 1b) Severe thrombocytopenia 2 2 3 2 2 2 3 2c) Immunosuppressive treatment 2 2 2 2 2 2 2 1d) None of the above 2 2 2 2 2 2 1 1

I =Initiation; C = Continuation; COC = combined oral contraceptive; P = combined hormonal contraceptive patch; R = combined hormonal vaginalring; POP = progestin-only pill; DMPA = depot medroxyprogesterone acetate; IUD = intrauterine device; LNG-IUD = levonorgestrel-releasing IUD;Cu-IUD = copper IUD; BMI = body mass index; DVT = deep venous thrombosis; PE = pulmonary embolism.* Condition that exposes a woman to increased risk as a result of unintended pregnancy.



– Dose reduction from 30–40 to 20 µgEE: Further reduction of the ethinylestradiol dose to 20 µg EE appears tolead to an additional although onlyslight reduction in VTE risk [9, 57].

– Dose reduction from 30–40 µg EEto non-EE COC: Lidegaard et al.[57] note that progestogen-onlypreparations are not associated with ahigher risk of VTE for women of fer-tile age. Mini-Pills with levonor-gestrel or norethisterone: VTE rateratio of 0.59 (0.33–1.03) (data basedon 65,820 woman-years), or with75 µg desogestrel: VTE rate ratio of1.12 (0.36–3.49) (data based on 9,044woman-years). (Authors’ note: Clearindications are not available as towhether these results can be trans-ferred without qualification to womenwith marked risk factors.)

– Risk assessment of COC with estra-diol or estradiol valerate: Comparedto ethinyl estradiol, estradiol and es-tradiol valerate lead to less liver en-zyme induction and less impact onhemostasis. It is currently unclearwhether this theoretical advantagealso actually leads to a lower inci-dence of VTE.

Progestogens and their dosage: Theinfluence of different progestogens onthe risk of VTE is disputed. Levonor-gestrel is usually taken as the referencefor comparisons between differentprogestogens. According to the bestcurrently available studies [9], VTE in-cidence for levonorgestrel-containingCOCs with less than 50 µg EE lies atapproximately 8 VTE per 100,000woman-years. While it describes therisk for the typical OC user population,this value is of only limited use on ac-count of the strong age-dependency ofrisk in individual cases. See Figure 8for age-dependency of risk.

Incidence rates for norethisterone, nor-ethisterone acetate, and norgestimate aresimilar to that for levonorgestrel [57].

Studies published in the mid-1990sshowed a higher risk for gestoden anddesogestrel, which are known as third-generation progestogens, compared tolevonorgestrel (see meta-analysis byKemmeren et al. [120]. Studies that ad-just for the temporal dependence of risk(higher in the initial months of expo-sure following first use or resumption

of use) and correctly adjust quantita-tively for age differences showed nosignificant differences between the newand established progestogens of thetime. However, the methodologicalshortcomings of all available studies donot allow clear conclusions to be drawnregarding causal connections. Resultson cyproterone acetate (CPA) are alsoconflicting. This applies even to resultsfrom the same working group that ana-lyzed the same data source at differentpoints in time with different method-ologies. Thus Lidegaard [121] found anincidence of 31 VTE per 100,000woman-years with a confidence inter-val of 13–49 using a Danish patient reg-istry, whereas six years later [57] the

point estimate for incidence was clearlyoutside the 2003 confidence interval at71 VTE per 100,000 woman-years. Inthe same period of time the relative riskcompared to levonorgestrel rose from0.7–1.9. These differences cannot eas-ily be explained by coincidence, andhighlight the considerable method-ological difficulties in carrying out andevaluating studies of VTE risk in OCusers.

Regarding chlormadinone acetate(CMA), there are no indications of ahigher VTE risk compared to levonor-gestrel-containing COCs [122]. Thisalso applies to ethinyl estradiol-contain-ing COCs with dienogest [117].

Figure 7. Cigarette smoking as risk factor for VTE with use of OCs. Mod from [112].

Figure 8. BMI and age as risk factors for VTE with OC use. Weight and age are independent risk factors with ad-ditive effect. Mod from [112].



Conflicting results have also beenfound for drospirenone. Two large-scale prospective cohort studies(Fig. 11) [9, 123] and a German case-control study [117] showed no higherrisk, while two studies published in2009 – a retrospective cohort study inDenmark [57] and a Dutch case-controlstudy [113] – showed a slightly higherrisk compared to levonorgestrel-con-taining preparations. The two latterstudies, however, exhibit substantialmethodological shortcomings [124,125]. The Dutch study was not statisti-cally significant, and also not represen-tative for either the cases or controls. Inthe Danish study, short-term and long-term use were misclassified to a consid-erable degree, and information aboutimportant risk factors was not available.In addition, an independent validationstudy showed that probably around30% of the diagnoses that the authorstook from the Danish patient registrywere incorrect [126]. In addition,shortly before this statement went topress, the Boston Collaborative DrugSurveillance Program published the re-sults from two retrospective case-con-trol studies in the USA and England us-ing the PharMetrics [127] and GPRDdatabases [128]. Both studies yieldedhigher risk estimates for drospirenone-containing COCs with 30 µg EE. Thesestudies too show considerable short-comings. The GPRD results, which arebased on confirmed VTE, are not statis-tically significant. The incidence rates,which are too low overall, show that thedatabase compiled only some of theVTE (possible ‘ascertainment bias’). Inaddition, the substantially different riskestimates for pulmonary embolism anddeep venous thrombosis (factor 4) indi-cate the presence of considerable differ-ential diagnostic bias. The PharMetricsstudy was based on non-confirmedVTE from a database used for calculat-ing benefits, which cannot provide a re-liable scientific basis unless the diag-noses are confirmed by health records.The study was not able to reproduceknown risks such as the dependence onduration of use (see above), and did nothave access to information on majorprognostic factors.

In sum, the VTE risk of drospirenone-versus levonorgestrel-containing COCscannot be conclusively ascertained. Thestudies with the best methodology do not

Figure 9. VTE risk over time following start of COC use. Mod. from [114].

Figure 10. Influence of short breaks in Pill use on VTE risk. Mod. from [114].

Figure 11. VTE risk factors with OC use: VTE risk of drospirenone-containing OCs. Mod. from [9].


Contraception and ThrombophiliaTable 14. Classification of contraceptives according to the risk of VTE in healthy women of reproductive age without additional risk factors(such as obesity, immobilization, positive family history of cardiovascular disease, cigarette smoking). [Rabe & Dinger 2011, personal commu-nication]

Risk Age Incidence Contraceptive method/ Published Studies Ongoing Studies(Years) (VTE/10,000 Population group

women years)

Reference ≤ 19 1–2 Healthy, non-pregnant women of child- Lidegaard 2009 [57]: INAS-OC and INAS-SCORE20–29 2–3 bearing age not using a contraceptive I = 3.0 (2.9–3.2); Ex = 4813 TWY (EURAS-type studies; end 2013 and 2014,30–39 3–4 respectively)40–49 5–7 Non-hormonal contraceptive methods Dinger 2007 [9]:

– tubal sterilization I = 4.4 (2.4–7.3); Ex = 65 TWY– condoms, spermicides– behavioral methods Review Article: Heinemann 2007

15–49 3–4 – copper IUDs

Unchanged 15–49 3–4 Progestin-containing contraceptives Lidegaard 2009 [57]:or (slight increased risk cannot completely Levonorgestrel-IUS EURAS-IUDSlightly be excluded in comparison to non-hormo- I = 3.4 (2.3–4.7); Ex = 101 TWY (EURAS-type study; ends 2012)increased nal contraceptive methods; therefore non- Progestin-only pill LASS

hormonal methods should be preferred for I = 2.0 (1.1–3.3); Ex = 75 TWY (EURAS-type study; ends 2011)women with a history of thrombophilia)– Levonorgestrel-IUS– Progestin-only pill– Progestin-only ovulation inhibitor– Progestin depot injections

Moderately ≤ 19 3–4 Combined oral contraceptives with < 50 µgincreased 20–29 5–8 Ethinyl estradiol and– Level 1 30–39 8–10 – Levonorgestrel (LNG), Norethisteron, Lidegaard 2009 [57]: LASS

40–49 15–22 – Norethisterone acetat or Norgestimate I = 5.5 (4.7–6.3); Ex = 367 TWY (EURAS-type study; ends 2011)– (NGM) (underestimation due to misclassifi-

15–49 6–10 cation of current duration of use andother reasons)

Dinger 2007 [9]:I = 8.0 (5.2–11.7); Ex = 31 TWY

– Chlormadinone acetate (probably no Waldmann-Rex 2009 [122]: No EURAS-type study– higher risk than with LNG-containing I = 2.4 (0.9–5.2); Ex = 25 TWY– COCs; however, a slightly higher risk (underestimation due to– cannot be excluded) methodological short-comings)

– Dienogest (probably no higher risk than Dinger 2010 [117]: INAS-SCORE– with LNG-containing COCs; however a OR vs LNG: 1.0 (0.6–1.8); 95 Ca/303 Cn (EURAS-type study; ends 2014)– slightly higher risk cannot be excluded)

– MPA depot injection (classification is van Hylckama, Vlieg et al. 2010 [131]: No EURAS-type study– based on a methodologically limited OR* 3.6 (1.8–7.1); 20 Ca/15 Cn– study with a limited number of cases (*vs. non-use of hormonal contra-– and controls; overestimation of risk ceptives; OR vs. LNG: ~1)– compared to other hormonal– contraceptives possible)

Combined oral contraceptive pills with INAS-SCOREEstradiol valerate and Dienogest (less (EURAS-type study; ends 2014)influence on hemostasis compared withEthinyl estradiol/Dienogest; however, riskassessment should be based on the VTE-incidence of Ethinyl estradiol/Dienogestas long as robust data are not available)

NuvaRing® (provisional classification based TASCon interim results from the TASC study) (EURAS-type study; ends 2012)

– Level 2 15–49 6–14 Combined oral contraceptives with < 50 µgEthinyl estradiol and– Drospirenone (DRSP) (inconsistent study Lidegaard 2009 [57]: LASS– results; in contrast to retrospective I = 7.8 (6.4–9.5); Ex = 131 TWY (EURAS-type study; ends 2011)– database studies 2 prospective cohort (no substantial missclassification of INAS OC– studies and 1 retrospective field study duration of use; potential overestimation (EURAS-type study; ends 2011)– did not show an increased risk compared of risk compared to LNG [see above])– to LNG-containing COCs; based on– currently available data a slightly Dinger 2007 [9]:– increased risk is possible) I = 9.1 (5.9–13.3); Ex = 29 TWY

Dinger 2010 [117]:OR vs LNG: 1.0 (0.5–1.8); 85 Ca/281 Cn

Jick et al. 2011 [127]:OR 2.2 (1.5–3.4); 166 Ca/550 Cn

Parkin et al. 2011 [128]:OR 2.9 (1.1–7.4); 57 Ca/176 Cn

– Desogestrel (DSG), Gestoden (GSD) or Lidegaard 2009 [57]: LASS– Cyproterone-acetate (CPA) (risk of VTE in DSG/GSD (EURAS-type study; ends 2011)– comparison with LNG-containing prepara- I = 6.8 (6.5–7.2); Ex = 2008 TWY– tions scientifically controversial, however CPA– a slightly to moderately higher risk is I = 7.1 (5.7–8.7); Ex = 127 TWY– possible) (less substantial underestimate of the

risk compared to LNG [see above])Case-control studies with and withoutadjustment for duration of use showedORs of ~ 2 and ~ 1. Numerous reviewarticles available; the most balancedrepresentation being the decision of theHigh Court of Justice in 2002 [132]

Evra contraceptive patch (VTE risk Dore 2010 [133]: No EURAS-type studycompared to COCs with LNG or NGM OR vs NGM: 2.0 (1.2–3.3); 102 Ca/353 Cncontroversial; a slightly to moderatelyincreased risk is possible) Jick 2010 [134]:

ORs vs LNG from 2 sources46 Ca/207 Cn & 97 Ca/382 Cn:2.0 (0.9–4.1) & 1.3 (0.8–2.1)

Substantially 15–49 20–30 Pregnancy and the first three months Heit 2005 [72]: No EURAS-type studyincreased after childbirth; risk after cesarean section I = 29.3 (23.8–35.6); Ex = 50 TWY

significantly higher than after spontaneousdelivery Lidegaard 2011 [135]: RR vs ‘non-pregnant

non-users’ 10.6 (9.4–12.0); 265 Cases

Ex = Exposure in 1000 woman-years; Ca = Number of relevant cases in a case control study; I = Incidence of VTE/10,000 WY with 95% CI; Cn = Number of relevant controls in a case controlstudy; OR = odds ratio; RR = relative risk; TWY = 1,000 woman-years; * cases/controls



show indications of a higher risk fordrospirenone, but these studies too havemethodological limitations and cannotexclude a slight increase in risk.

The PhVWP (Pharmacovigilance Work-ing Party) of the European MedicinesAgency (EMA) [129] completed in May26th, 2011, a review of all available data,including some further re-analyses andinformation on additional analyses re-garding the risk of venous thrombo-embolism (VTE) associated with drospi-renone-containing combined oral con-traceptives (COC), such as Yasmin andYasminelle. Altogether seven epidemio-logical studies [9, 57, 113, 117, 123,127, 128] have analysed/evaluated an as-sociation between drospirenone-con-taining COC and VTE.

The assessment has not changed theconclusion that the risk of VTE withany COC (including those containingdrospirenone) is very small. ThePhVWP concluded that the data haveshown that drospirenone-containingCOCs are associated with a higher VTErisk than levonorgestrel-containingCOCs and that the risk may be similarto that for COCs containing desogestrelor gestodene. The PhVWP recom-mended that the product informationfor all drospirenone-containing COCsshould be updated to reflect these con-clusions. There is no reason for womento stop taking drospirenone-containingCOCs, such as Yasmin and Yasminelle,or any other COCs. In Germany theBfArM (Bundesinstitut für Arznei-mittel und Medizinprodukte) (2011)[130] published a similar statement attheir homepage.

What remains is that the use of anycombined oral hormonal contraception(COC) is always associated with ahigher risk of VTE. The risk depends onthe estrogen dose and is possibly modu-lated by the choice of progestogen.Whether there actually are clinically rel-evant differences in the thromboembo-lism risk for different progestogens isstill the object of discussion.

Table 14 compares the risk of thrombo-sis/VTE with different hormonal contra-ceptives for healthy women of fertile agewithout additional risk factors comparedto the risk for women in pregnancy andpuerperium.

for Evra® are 25% lower. Inter-individualvariability of ethinyl estradiol levels ishigher for Evra® than for OCs. It is notknown whether serious side effects aredue to pharmacokinetic differences(American RxList, September 8, 2010).

In fact there is little data in the literatureas to whether these qualitative pharma-cokinetic differences affect VTE risk.

On the individual study results: The firstepidemiological study was done by theBoston Collaborative Drug SurveillanceProgram (BCDSP). It showed that therisk of non-lethal VTE events in connec-tion with Evra® is comparable to therisks for COCs containing 35 µg ethinylestradiol and the progestogen norges-timate. New cases were added to thoseoriginally published. Point estimates forthe later cases are higher than those forthe earlier ones (Tab. 16). The total studysize comprises 162 cases and 626 con-trols. The odds ratio for the VTE risk ofEvra® versus that of a norgestimate-con-taining Pill was 1.23 (0.86–1.77).

The second study, which also had accessto patient records, was done by i3Ingenix. Its results showed a nearly two-fold higher risk of medically confirmedVTE events for the use of Evra® com-pared to a 35 µg ethinyl estradiol-con-taining Pill with the progestogen nor-gestimate [137]. However, at 1.1 the

4.2.1.1. Combined Contraceptive Patch(Evra®)Product description: A contraceptivepatch measuring 20 cm2 that releases theactive agents ethinyl estradiol and norel-gestromin, a metabolite of norgestimate.The ethinyl estradiol release was origi-nally given as 20 µg/day, but corrected inthe Summary of Product Characteristics(SPC) to 33.9 µg/day [136]. The patchhas an effective duration of 1 week; threeare used followed by a 1-week break.

Due to FDA recommendations, the firstreport was presented in the USA in No-vember 2005 on increased estrogen ex-posure from Evra® compared to oral hor-monal contraceptives with 35 µg ethinylestradiol. Additional information wasadded to the package leaflet in Septem-ber 2006 after the results of the first twoepidemiological studies on the VTE riskof Evra® were presented. In January2008 the results of a third epidemiologi-cal study were added to the package leaf-let.

A possibly higher thrombosis risk withEvra® is based on the following factors(Tab. 15):

For Evra® the area under the curve(AUC) for steady state levels of ethinylestradiol is approximately 60% higherthan for oral preparations. By contrast,peak concentrations of ethinyl estradiol

Table 15. Estimates (Odds Ratios) of Venous Thromboembolism Risk in CurrentUsers of Ortho Evra Compared to Oral Contraceptive Users (http://www.rxlist.com/ortho-evra-drug.htm) (per 4.5.2011)

Epidemiologic Studies Comparator Odds RatioEvra & VTE product (95 % CI)

i3 Ingenix NGM Study in Ingenix NGM/35 μg EEa 2.2* (1.2–4.0)b

(Cole et al. 2007 [137], Dore et al. 2010 [133],Cole et al. 2008 [138], Dore et al. 2009 [139])

BCDSPc NGM Study in Pharmetrics database NGM/35 μg EE 1.2 (0.9–1.8)d

(Jick et al. 2006 [140], Jick et al. 2007 [141],Jick et al. 2010 [142])

BCDSP LNGe Study in Pharmetrics database LNG/30 μg EE 2.0 (0.9–4.1)f

(Jick et al. 2010 [134])

Pharmetrics database LNG/30 μg EE 1.3 (0.8–2.0)g

(Jick et al. 2010 [134])

* Increase in risk of VTE is statistically significant; a NGM = norgestimate; EE = ethinylestradiol;b Pooled odds ratio from Cole et al. 2007 [137], Dore et al. 2010 [133]. [Initial 33 months ofdata: Odds Ratio (95 % CI) = 2.5† (1.1–5.5); Separate estimate from 24 months of data onnew cases not included in the previous estimate: Odds Ratio (95 % CI) = 1.4 (0.5–3.7)];c BCDSP = Boston Collaborative Drug Surveillance Program; d Pooled odds ratio from Jick etal. 2006 [140], Jick et al. 2007 [141], Jick et al. 2010 [142]. [Initial 36 months of data: OddsRatio (95 % CI) = 0.9 (0.5–1.6); Separate estimate from 17 months of data on new cases notincluded in the previous estimate: Odds Ratio (95 % CI) = 1.1 (0.6–2.1); Separate estimatefrom 14 months of data on new cases not included in the previous estimates: Odds Ratio(95% CI) = 2.4* (1.2–5.0)]; e LNG = levonorgestrel; f 48 months of data; g 69 months of data



lower confidence limit for the odds ratiois only slightly above 1.

A third study also done by BCDSP com-pared the risk of non-lethal VTE eventsfor Evra® with that for COCs with 30 µgethinyl estradiol and the progestogenlevonorgestrel [133]. It examined idio-pathic VTE cases and age-matched con-trols in the PharMetrics and MarketScandatabases. The VTE odds ratios forEvra® compared to levonorgestrel-con-taining Pills were 2.0 (0.9–4.1) and 1.3(0.8–2.1) For women > 40 years of agein the PharMetrics database, a statisti-cally significant higher risk was found,although no adjustment was made formultiple testing. The authors reached theconclusion that a higher risk cannot beexcluded for this age group.

Based on the study results described, the“Rote Liste” (German pharmaceuticaldirectory) as well as the German andAmerican summaries of product charac-teristics (SPC) contain an alert for a pos-sibly higher risk of VTE.

“Rote Liste” status on September 9,2010: “Data from a retrospective cohortstudy with women aged 15–44 yearssuggests that VTE incidence in womenwho have used Evra® is higher than thatin users of an oral contraceptive contain-ing levonorgestrel.”

German Summary of Product Character-istics (SPC) (003648-C647) (status Sep-tember 8, 2010): This contains the sametext as that on the “Rote Liste” above. Itnotes that the incidence is 1.4 timeshigher (95% CI 0.9–2.3) for women withand without other VTE risk factors and 1.5times higher for women without additionalVTE risk factors (95% CI 0.8–2.7).

American Summary of Product Charac-teristics (SPC): Three case-control stud-ies [133, 134, 137, 140, 141, 143] exam-ined the VTE risk for women aged 15–44 years who had used the Evra® contra-ceptive patch compared to that forwomen who had used oral contracep-tives with 30–35 µg ethinyl estradiol(EE) and norgestimate or levonorgestrel.Electronic patient data from “healthcareclaim” databases were used. Norgesti-mate is the pro-drug for norelgestromin,the progestogen in Ortho Evra®. Thesestudies (Tab. 15), which differed slightlyin design, yielded odds ratios between

0.9 and 2.5. Interpretations of these oddsratios extended from no higher risk toa nearly doubled risk. One study (i3Ingenix) examined the patient records ofVTE cases. In the three major clinicalstudies (n = 3,330 with 1,704 woman-years following exposure) there was onenon-lethal case of pulmonary embolismwith Ortho Evra®. A non-lethal post-operative case of pulmonary embolismwith Ortho Evra® was also reported(www.rxlist.com/ortho-evra-drug.htm).

Possible bias factors in VTE studies onEvra®:All three studies are based on databasesthat do not contain data or valid data forall relevant risk factors (e.g. BMI, familyhistory). Due to Evra®’s different methodof administration as well as market posi-tioning, it can be assumed that the userpopulations differ markedly, and there-fore also the risk profiles. It cannot beassumed, however, that the influence ofthese different risk profiles could befully considered in evaluating VTE risk.In addition, it is not unlikely that in theUSA, media reports as well as intensiveadvertisements and videos indicating atwo-fold higher VTE rate by lawyersseeking clients (e.g. YouTube) have en-couraged more frequent diagnostic mea-sures for Evra® patients, which in turnhas led to a higher detection rate of VTE(detection bias). On the other hand, it isalso likely that this has led to a morecareful exclusion of risk patients(healthy user effect). The net result ofthese opposing effects is difficult to as-sess.

WHO Medical Eligibility Criteria: Forknown thrombophilic mutations (e.g.factor V Leiden, prothrombin G20210Amutation, protein S, protein C or anti-thrombin deficiency):

Classification: grade 4 = contraindica-tion for Evra®.

Summary regarding VTE:– The data are not definitive, but indi-

cate what might be as much as a two-fold higher risk compared to levonor-gestrel-containing OCs. Until suffi-cient valid information is available, ahigher risk should be presumed in theinterest of patient safety.

– When Evra® is prescribed, patientsmust be informed of the potentiallyhigher risk.

– Evra® is contraindicated for patientswith a heightened risk of VTE.

4.2.1.2. Combined Vaginal Ring(NuvaRing®)Product description: Contraceptivevaginal ring that releases 0.120 mgetonogestrel (metabolite of desogestrel)and 15 µg ethinyl estradiol per day for 21days, and thus inhibits ovulation.

SPC: According to the manufacturer, thethromboembolism risk for NuvaRing® isequivalent to that for COCs. NuvaRing®

is a convenient form of contraceptionthat inhibits ovulation without daily Pillintake.

Studies: Unfortunately the results of theTransatlantic Active Surveillance onCardiovascular Safety of NuvaRing(TASC) are not yet available (http://cli-nicaltrials.gov/ct2/show/NCT00524771).Interim results on the basis of approxi-mately 30,000 woman-years and 29 VTEshow no higher risk for NuvaRing® com-pared to COC users – also after excludingdesogestrel- and gestoden-containingCOCs. The relevant Safety Monitoringand Advisory Council came to the con-clusion that the data provide no indicationof higher risk; however, robust conclu-sions regarding a slightly higher potentialrisk cannot yet be drawn at present[Dinger, personal communication].

Due to the known increased risk ofthrombosis with combined estrogen-containing preparations, an “existing orprevious venous thrombosis” is there-fore also considered a contraindica-tion for NuvaRing®.

Metabolic studies: The effects ofNuvaRing® on hemostatic parametershave been examined in an open com-parative study with 87 women [144].Thecomparison group received a combinedoral contraceptive (30 µg ethinyl estra-diol and 150 µg levonorgestrel). Thedata show that NuvaRing® exerts onlyslight, clinically non-relevant effects onthe coagulation parameters, comparableto those of levonorgestrel-containingoral contraceptives.

WHO Medical Eligibility Criteria:NuvaRing® is placed in the same cat-egory as combined oral contraceptives.For known thrombophilic mutations (e.g.factor V Leiden, prothrombin G20210A



mutation, protein S, protein C or anti-thrombin deficiency): Classification:grade 4 = contraindication.

Summary regarding VTE:– There are currently no definitive find-

ings on NuvaRing® that show an in-creased VTE risk compared to COCs.Interim results from the TASC studysuggest the provisional conclusionthat VTE risk is comparable to that forcombined oral contraceptives(Tab. 15).

– When NuvaRing® is prescribed, pa-tients should be informed of a higherVTE risk as with COCs.

– The SPC lists an existing or previousvenous thrombosis as a contraindica-tion.

4.2.2 Progestogen-Only Contraceptives4.2.2.1. Non-Estrogen Ovulation Inhibi-tor (Cerazette®)Product description: desogestrel-onlyPill (75 µg desogestrel per pill), whoseprogestogen level lies slightly above theovulation-inhibiting dose (60 µg/day fordesogestrel) and thus inhibits ovulation.Drawback: irregular menses occur insome patients. In contrast to the classicalmini-Pill, this preparation can be takenup to 12 hours delayed.

SPC: No information is provided by themanufacturers about VTE or CHD risks.No definitive studies are available.

Metabolic studies: A randomized, con-trolled, double-blind study comparedthe hemostaseological effects ofCerazette® with those of a mini-Pill with30 µg levonorgestrel. The authors con-cluded that the two preparations aresimilar and have favorable – with respectto VTE risk – effects on hemostasis[145]. In general it should be assumedthat the higher VTE risk from COCs isdue primarily to the estrogen.

Epidemiological studies: Lidegaard etal. [57] indicate that progestogen-onlypreparations for women of fertile agewithout pre-existing diseases are not as-sociated with an increased risk of VTE;mini-Pills with levonorgestrel or nor-ethisterone: rate ratio for VTE: 0.59(0.33–1.03) (study size: 65,820 woman-years) or with 75µg desogestrel rate ratio1.12 (0.36–3.49) (study size: 9,044woman-years).

WHO Medical Eligibility Criteria: Forknown thrombophilic mutations (e.g.factor V Leiden, prothrombin G20210Amutation, protein S, protein C or anti-thrombin deficiency): Classification:grade 2 (broadly usable method).

In general, at-risk women (e.g. carriersfor factor V Leiden mutation) should usea non-hormonal means of contraception.If this is not possible, one option undercertain circumstances is to use a pro-gestogen-only Pill following individualrisk-benefit consideration and extensiveinformation/counseling and documenta-tion. In such cases primary considerationcould be given to a levonorgestrel-con-taining “classical” mini-Pill, becausecurrent study results about it are basedon the most solid data.

Summary regarding VTE:– Cerazette® is a progestogen-only

preparation, which on account of itsrelatively high dose of the progesto-gen desogestrel compared to the“classical” mini-Pill, has an ovula-tion-inhibiting effect. Like the com-bined Pill, it can be taken up to 12hours delayed. In contrast to COCs, itcan also be taken during puerperium.Like with OCs, to ensure contracep-tive effectiveness a period of 36 hoursbetween two pills may not be ex-ceeded.

– Cerazette® does not have a heightenedrisk of thrombosis according to theLidegaard study [57], although thestudy’s relatively small number ofCerazette® users and the large confi-dence interval mean that a higher riskthan for the “classical” progeston-only Pill cannot be excluded.

4.2.3. “Classical” Progestogen-Only PillWithout ovulation-inhibiting effect;“classical” progestogen-only pill (POP)

Product description: The “classical”POP is a progestogen-only preparationwith norethisterone (dose: 0.35 mg/day)levonorgestrel (dose: 0.03 mg/day),whose progestogen level lies below theovulation-inhibiting dose and thus doesnot have an ovulation-inhibiting effect.It must be taken daily within a 3-hourwindow of time. Frequent intermen-strual bleeding is the major side effect.Only limited use is possible during puer-perium because the substances can crossover into breast milk.

SPC: No information from the manufac-turers about VTE or CHD risk.

Studies: No definitive studies or post-marketing studies.

Metabolic studies: A randomized, con-trolled, double-blind study compared thehemostaseological effects of Cerazette®

(see above) with those of a POP contain-ing 30 µg levonorgestrel. The authorsconcluded that the two preparations aresimilar and have favorable effects – withrespect to VTE risk – on hemostasis[145]. In general it should be assumedthat the higher VTE risk with COCs is dueprimarily to the estrogen.

Epidemiological studies: Lidegaard etal. [57] indicate that progestogen-onlypreparations for women of fertile agewithout pre-existing diseases are not as-sociated with a heightened risk of VTE;POPs with levonorgestrel or norethis-terone: rate ratio for VTE: 0.59 (0.33–1.03) (study size: 65,820 woman-years)or with 75 µg desogestrel: rate ratio 1.12(0.36–3.49) (study size: 9,044 woman-years).

WHO Medical Eligibility Criteria:For known thrombophilic mutations(e.g. factor V Leiden, prothrombinG20210A mutation, protein S, proteinC or antithrombin deficiency): Clas-sification: grade 2 (broadly usablemethod).

Primary consideration could be givenhere to a “classical” POP, because of theavailable studies.

If compliance (3-hour window) is notensured, specialists favor Cerazette® asthe next option. Ultimately, however,full decisional and educational responsi-bility lies with the prescribing gynecolo-gist.

Summary regarding VTE:According to the Lidegaard study [57],the classical POP does not have a higherthrombosis risk for women without pre-vious cardiovascular diseases. Thesample size for the two above-men-tioned progestogens is considerablyhigher than for the desogestrel-contain-ing Pill Cerazette®. In theory the favor-able study results could be distorted in sofar as primarily older women take thePOP after previous COC use and be-



cause women with an existing disposi-tion for VTE presumably have a lowerchance of receiving a POP prescription(“healthy user effect”). In quantitativeterms, however, these effects could onlydistort the results to a slight degree.Overall it can be assumed that the classi-cal POP does not lead to a higher risk ofthrombosis.

4.2.4. Subdermal Contraceptive Implant(Implanon ® )Product description: Implanon® is aprogestogen-only preparation (etono-gestrel), which is implanted subcutane-ously on the inner side of the upper arm,and has a contraceptive effect for threeyears based on continuous steroid re-lease. The maximum steroid levels in se-rum after 4 days lie at 814 pg/ml, after4–6 months in steady state at 200 pg/ml,and are still sufficient to inhibit ovula-tion after 3 years [146]. A new version ofImplanon® is currently coming onto themarket, namely Implanon® NTX, whichcan be localized by soft-tissue x-ray andwhose new inserter enables simpler andsafer subcutaneous implantation.

SPC: Contains no information from themanufacturer on increased risk of VTEor CHD.

Metabolic studies: A randomized, con-trolled, double-blind study compared thehemostaseological effect of Implanon®

with those of a corresponding levonor-gestrel-containing preparation. The au-thors concluded that the two prepara-tions had similar and overall only slighteffects on hemostasis [147].

Epidemiological studies: No epidemio-logical studies have been done on theCHD or VTE risks for Implanon®.

Attempts have therefore been made toextrapolate from epidemiological dataon oral intake of the correspondingprogestogen to the use of Implanon®.This assumes that the average serum lev-els are comparable. Complete compara-bility is not given, however, because it isunclear whether the constant steady statesteroid levels with Implanon® affect theVTE risk in similar ways to the strongdaily fluctuations in progestogen levelfrom oral intake.

Lidegaard et al. [57] conclude thatprogestogen-only preparations for

women of fertile age without pre-exist-ing diseases are not associated with ahigher risk of venous thromboembo-lism; mini-Pills with levonorgestrel ornorethisterone: VTE rate ratio of 0.59(0.33–1.03) (study size: 65,820 woman-years), or with 75 µg desogestrel: VTErate ratio of 1.12 (0.36–3.49) (study size:9,044 woman-years).

WHO Medical Eligibility Criteria: Forknown thrombophilic mutations (e.g.factor V Leiden, prothrombin G20210Amutation, protein S, protein C or anti-thrombin deficiency) (analogous tomini-Pill): Classification: grade 2(broadly usable method).

Summary regarding VTE:– Implanon® has not been studied with

respect to VTE.Extrapolation using the VTE datafrom the Lidegaard study (see com-ment above) does not yield a higherrisk of thrombosis, although thereis only limited comparability withoral preparations. In cases of doubt,therefore, preference should begiven to oral progestogen-onlypreparations.

– Compared to the depot progestogenswhich the WHO classifies as “broadlyusable” (grade 2), Implanon® can beimmediately removed from the bodyif complications arise.

– When counseling patients, the greaterrisk of menstrual disorders shouldbe explained; problems removingImplanon® should be solved by thenew inserter for Implanon® NTX.

4.2.5. Levonorgestrel IntrauterineSystem (Mirena ® )Product description: levonorgestrel-containing intrauterine system with aneffectiveness duration of 5 years. Themean release rate of levonorgestrel liesat 20 µg/24 hours the first year, andat approximately 10 µg/24 hours after5 years. The average release rate of levo-norgestrel over a period 5 years is 14 µg/24 hours. Systemic levonorgestrel levelsare lower than those for the mini-Pill.Menstrual volume and duration decreaseas well as rates of dysmenorrhea andtranscervical infections. The amenor-rhea rate for first use is 20%, and formultiple use up to 60%.

SPC: Contains no information from themanufacturer on the risk of VTE or CHD.

Metabolic studies: Hemostaseologicalstudies have been done on oral adminis-tration of levonorgestrel (see mini-Pill).

Epidemiological studies: Lidegaard etal. [57] found no heightened risk of VTEfor Mirena® users in a study based on101,351 woman-years: rate ratio 0.89(0.64–1.26). A limiting consideration isthat the authors were only able to esti-mate the average duration of use. Cross-reference with oral levonorgestrel prepa-rations (see mini-Pill) with consider-ation of the low systemic serum levelsfor Mirena® users, however, supports thesupposition that Mirena® use does notlead to higher VTE risk.

In a case-control study van HylckamaVlieg et al. [131] assessed the risk fac-tors for venous thrombosis. Premeno-pausal women aged 18 to 50 yearsusing depot-medroxyprogesterone ace-tate or levonorgestrel intrauterine de-vices were compared to nonusers ofhormonal contraceptives. No increasedrisk was associated with levonorgestrelintrauterine devices (odds ratio 0.3;95% CI 0.1–1.1).

WHO Medical Eligibility Criteria: Clas-sification as grade 2 (broadly usablemethod).

Special case – patients taking oralanticoagulants: Experts recommendthat insertion of an intrauterine systemshould take place 4 weeks at the earliestafter start of anticoagulant intake [148].

Summary regarding VTE:– The study by Lidegaard et al. [57]

shows no increase in VTE risk: rateratio 0.89 (0.64–1.26) (study size:101,351 woman-years).

– Specialists consider insertion 4 weeksafter start of anticoagulant intake pos-sible (warning: “off-label” use).

4.2.6. Depot ProgestogensProduct description:– Depot progestogens are progestogens

that are released over a period of 2–3months after i.m. or s.c. application.There are currently two older prepa-rations:

– Noristerat® from Bayer-Schering withnorethisterone enanthate (200 mg)

– Depo-Clinovir® from Pfizer with150 mg depot medroxyprogesteroneacetate, and the lower dosed and s.c.



administrable preparation Sayana®

(104 mg DMPA) introduced in 2009.

SPC: Contains no information from themanufacturer on VTE or CHD risk

Metabolic studies: An Egyptian studywith 30 women found no change inhemostatic parameters following 3-month and 15-month treatment withdepot-medroxyprogesterone [149].

Epidemiological studies: The WorldHealth Organization Collaborative Studyof Cardiovascular Disease and SteroidHormone Contraception (1998) [150]assessed the cardiovascular risk of oralhormonal contraceptives compared toprogestogen-only depot injections andcombined contraceptive depot injec-tions. It studied 3,697 women with car-diovascular diseases (59% stroke, 31%VTE and 10% acute myocardial infarc-tion). At the time, 53, 37, and 13 womenwere taking oral progestogen prepara-tions, injectable progestogen prepara-tions, and combined injectable contra-ceptives, respectively.

The overall adjusted odds ratios for allcardiovascular diseases for current useof oral progestogen-only preparations,injectable progestogen-only prepara-tions, and combined depot contracep-tives, compared to non-users of any typeof steroid hormonal contraceptives,were 1.14 (0.79–1.63), 1.02 (0.68–1.54), and 0.95 (0.49–1.86), respec-tively. No significant differences in theodds ratios for stroke, VTE, or myocar-dial infarction could be found in connec-tion with any type of contraception.However, a slight, non-significant in-crease was found in the odds ratios forVTE for current use of oral progestogen-only and for progestogen-only depotpreparations. For women with a historyof hypertension, the odds ratio for strokerose for users of all oral progestogenscompared to non-users of steroid hor-monal contraceptives who had no his-tory of hypertension, from 7.2 (6.1–8.5)to 12.4 (4.1–37.6).

The authors state that although the studyhas only a small number of disease cases(11 VTE cases and a total of 37 venousand arterial thromboembolic events), thedata suggest that there is no or only aslight increase in risk of stroke, VTE ormyocardial infarction for use of oral or

injectable progestogen-only or com-bined estrogen/progestogen-containingcontraceptives.

By contrast, the sub-analysis of a Dutchcase-control study [131] with 20 VTEcases and 15 controls among users ofdepot-medroxyprogesterone acetateyielded a three-fold VTE risk (95% confi-dence interval: 1.2–7.5) compared to thatfor non-users of hormonal contraception(i.e., the study found similar risks for de-pot-medroxyprogesterone acetate andlevonorgestrel-containing COCs). TheWHO study and especially the Dutchstudy, however, have considerable meth-odological shortcomings (e.g. insuffi-cient information on VTE risk factors toadequately adjust for preferential use ofnon-estrogen contraceptives by risk pa-tients, limited comparability of cases andcontrols). A conclusive statement aboutVTE and CHD risk is therefore not pos-sible on the basis of these studies.

Van Hylckama Vlieg et al. [131] exam-ined the risk of DVT for depot-medroxy-progesterone acetate compared to a levo-norgestrel-releasing intrauterine device.Multi-variant analyses considering envi-ronmental and genetic factors were per-formed in the course of a large case-con-trol study on risk factors for venousthrombosis. These analyses took pre-menopausal women aged 18–50 whowere neither pregnant nor within 4 weekspost-partum, and who were not takingoral contraceptives, for a total of 446cases and 1146 controls. Injectable depot-medroxyprogesterone acetate was associ-ated with a 3.6-fold (1.8–7.1) higher riskof venous thrombosis compared to thatfor non-users of hormonal contracep-tives. No higher risk was found for thelevonorgestrel-releasing intrauterine de-vice (odds ratio 0.3; 0.1–1.1). Becausetoo few women used a contraceptivepatch or implant, unfortunately no riskestimate could be given here. The risk ofvenous thrombosis was higher for depot-medroxyprogesterone acetate, whereas ahigher risk for the levonorgestrel-releas-ing intrauterine device could be excluded.This device is considered one of the safestcontraceptive options with respect tovenous thrombosis.

In addition, attempts can be made to ex-trapolate from epidemiological datagathered for oral use to depot use of thecorresponding progestogen. As already

mentioned, Lidegaard et al. (2009) [57]indicate that progestogen-only prepara-tions in women of fertile age withoutprevious cardiovascular disease are notassociated with an increased risk ofVTE; progestogen-only pills with levo-norgestrel or norethisterone: rate ratiofor VTE: 0.59 (0.33–1.03) (study size:65,820 woman-years) or with 75 µgdesogestrel rate ratio 1.12 (0.36–3.49)(study size: 9,044 woman-years). Ex-trapolation assumes that average serumlevels correspond. Complete compara-bility is not given, however, as it is notclear whether the constant steroid levelsin the steady state under depot progesto-gens have effects on VTE risk similar tothose of the strongly fluctuating levelswith oral progestogens.

Overall, there is no sufficient demonstra-tion of higher VTE or CHD risk for depotprogestogens. However, a slightly higherVTE risk cannot be fully excluded.

WHO Medical Eligibility Criteria: Forknown thrombophilic mutations (e.g.factor V Leiden, prothrombin G20210Amutation, protein S, protein C or anti-thrombin deficiency) (analogy: mini-Pill): Classification: grade 2 (broadlyusable method). Due to the possibleconversion of norethisterone enanthateto ethinyl estradiol, the author team rec-ommends classifying the norethisteroneenanthate-containing depot injection(Noristerat) as grade 3. This statement isbased on the study by Chu et al. (2007)[151] showing that norethisterone ac-etate is converted to ethinyl estradiol fol-lowing oral intake of 10–40 mg per dayat a rate of 0.2–0.33%.

Summary:– Increased VTE risk has not been suffi-

ciently demonstrated, although it alsocannot be fully excluded.

– According to the WHO criteria, thesepreparations have a grade 2 classifica-tion for use by VTE risk patients. Theauthors of this paper, however, adviseagainst doing so because if unex-pected complications arise, the pro-gestogen can remain systemically ac-tive for 2–3 months and in some casesfor up to 12 months.

4.2.7. Postcoital PillsPost-coital hormonal methods of shift-ing or suppressing ovulation followingunprotected sexual relations.



Product descriptions:Yuzpe method (1.0 mg levonorgestreland 100 µg ethinyl estradiol): Venousand arterial thromboembolism in authori-zation studies: None published. No VTEcases occurred in a study by Vasilakis etal. [152] of two doses of 100 µg ethinylestradiol and 0.5 mg levonorgestrel cov-ering 12,416 woman-years.Metabolic studies: A randomized clini-cal study by van Rooijen et al. [153] ex-amined the effects of emergency contra-ception with levonorgestrel alone and incombination with ethinyl estradiol. Bothmethods led to accelerated thrombingeneration in the Hemker test, with astronger effect shown by the combina-tion with ethinyl estradiol.

Levonorgestrel (0.75 or 1.5 mg): TheFDA (www.fda.gov/ohrms/dockets/ac/03/briefing/4015B1_12_FDA-Tab%205-1-Medical%20Officer%20Review.doc) (retrieved 9.9.2011) lists data from22 studies with approximately 15,000study participants Good overall toler-ability is reported, with no venous or ar-terial thromboembolism.Metabolic studies: A randomized clini-cal study by van Rooijen et al. [153] ex-amined the effects of emergency contra-ception with levonorgestrel alone and incombination with ethinyl estradiol. Bothmethods led to accelerated thrombingeneration in the Hemker test, with astronger effect shown by the combina-tion with ethinyl estradiol.

ellaOne® (30 mg ulipristal acetate):No venous or arterial thromboembolismwas observed in an authorization study(www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMate-rials/Drugs/ReproductiveHealthDrugsAdvisoryCommittee/UCM215510.pdf)(retrieved 9.9.2011) with 4,636 partici-pants. In clinical studies with several1000 participants no VTE had been ob-served. In addition, there was no phar-macovigilance reporting of VTE for themore than 270,000 women who havealready taken ellaOne® in the post-mar-keting phase [Schuller, personal com-munication based on Periodic SafetyUpdate Report July 2011].Metabolic studies: Hemostaseologicalstudies have not been done.

4.2.8. Copper IUDsProduct description: Copper-contain-ing intrauterine devices that inhibit

sperm ascension and ability to fertilizethe egg cell via continuous release ofcopper in the uterine cavity. Also inhibitimplantation.

WHO Medical Eligibility Criteria:Classification as grade 1 (method can al-ways be used).

Restrictions: If there are contraindica-tions for intrauterine devices (see SPC):e.g. inflammatory genital diseases,anomalies in the uterine cavity (e.g.myomas, septation) etc. For nulliparouswomen, insertion only after careful risk/benefit analysis.

4.2.9. Additional OptionsFor women: Additional current, non-hormonal alternatives include chemicalmethods (e.g. spermicides), barrier meth-ods (e.g. cervical caps, diaphragms), andnatural family planning methods (seerelevant literature). A permanent methodconsists of uterine tube sterilization.

For men: Feasible methods includecondoms, as well as the permanent op-tion of vasectomy.

4.3. Contraception for SpecificPatient Groups

4.3.1. Contraception for Women TakingAnticoagulants on a Short- or Long-Term BasisThe WHO Medical Eligibility Criteriarecommend the use of progestogen-onlypreparations for women undergoing anti-coagulation treatment (grade 2 classifi-cation). In Germany, this recommenda-tion can be viewed in more nuancedform, for the following reasons:– The INR can be monitored at any time

and at short intervals in Germany. Ifmonitoring (at least every 2 weeks)confirms levels in the accepted range,anticoagulation treatment should notprohibit use of an ovulation inhibitor.

– Use of an ovulation inhibitor preventsovulation or endometrium-relatedblood discharge.

However, when anticoagulation isstopped, the ovulation inhibitor shouldbe discontinued approximately 6 weeksbeforehand because of the residual co-agulation-promoting effect.

Hypermenorrhea is a not infrequentproblem with ongoing anticoagulation,

and should be actively discussed withpatients. Signs of anemia and iron defi-ciency should also be watched for andtreated. Ways of treating hypermenor-rhea under anticoagulation include:– Mirena®

– Ovulation inhibitors– COC intake without interruption (“off

label” recommendation)– If family planning is concluded: e.g.

endometrial ablation

COC use should never be restarted with-out a risk evaluation following cessationof vitamin K antagonists.

Anticoagulation treatment (phenprocou-mon, warfarin) carries a higher risk ofteratogenesis, especially if taken in the6–8th weeks of gestation (OR: 3.9). Themiscarriage rate is 42%, with higherrates also for premature births (16%) andintrauterine growth retardation. A suit-able safe means of contraception istherefore needed, and should be activelydiscussed with the patient [154].

Ideally, these women should plan preg-nancies with care and possibly switchfrom oral anticoagulants to low molecu-lar-weight heparin about 2 cycles beforeconception.

4.3.2. Contraception before SurgicalOperationsSee section 2.4.3.

4.3.3. Contraception During PuerperiumWomen should not take combined con-traceptives for three weeks post-partum.If they are no longer nursing, they canthen use a COC. It should be considered,however, that although VTE risk is high-est within the first 3 weeks post-partum,it remains elevated for a period of ap-proximately 3 months (Lidegaard, 2007:w w w. l i d e g a a r d . d k / S l i d e s / O C %20epidem/PP%2007-11-20%20en.pdf;slide 16)

Progestogen-only preparations can betaken before day 21 post-partum, andalso while nursing.

4.3.4. Emergency ContraceptionFor healthy women: No restriction inselecting a preparation.

For women with increased VTE risk:Given the extensive data on levonorges-trel-only preparations, and due to the



thrombogenic potential of estrogens, theauthors advise against progestogen/estrogen combinations and in favor of alevonorgestrel-only preparation.

4.3.5. WHO Recommendations forAt-Risk PatientsThe World Health Organization has pub-lished contraception classification crite-ria for women with specific medicalrisks. The “WHO Medical EligibilityCriteria” are available on the Internet(Tabs. 12,13).

Table 13 shows contraceptive recom-mendations for known thrombophilicfactors (e.g. factor V Leiden mutation,prothrombin G20210A mutation, pro-tein S/protein C/antithrombin deficien-cies), and different risk situations affect-ing lupus erythematosus.

The authors of this paper, however, holdthat depot progestogens should only berecommended to risk patients followingthe most careful of risk/benefit analyses.

5. Summary and Recom-

mendations for Use

Preamble with liability disclaimer: Allrecommendations regarding the use ofcontraceptives by non-healthy individu-als are “off-label” recommendationswhich the author team has compiled inconsensus to the best of its knowledgeand belief on the basis of the availableliterature. The physician must makeeach individual decision on the basis ofan individual risk/benefit analysis. Theinformation contained herein should as-sist in that process, but neither the au-thors nor the publishers assume any typeof liability individually or collectivelywith respect to the information or its use(see disclaimer).

5.1. Women without Thrombo-philia Risk who Desire Contra-ceptionContraceptive counseling: The follow-ing alternatives depend on age, familysize, and planned duration of contra-ception: the Pill (COC), vaginal ring,Mirena®, contraceptive patch, hormonalimplant, mini-Pill, depot progestogen.Each case must be decided individually,with consideration of the specific lifesituation, risk factors, “non-contracep-tive benefits”, and also “shared decision-making” aspects.

Estrogen dose: The lower the dosage ofethinyl estradiol, the higher the rate ofintermenstrual bleeding; whereas thehigher the dose, the greater the effect onthe coagulation system and the probabil-ity of thromboembolitic complications.In the absence of specific VTE risk fac-tors, a preparation with the lowest pos-sible dosage of ethinyl estradiol (≤ 30 µgestradiol) should be selected. If contra-indications are present, non-estrogenmethods should be used.

Ethinyl estradiol versus estradiol orestradiol ester: The extent to which fur-ther risk can be further minimized by theuse of estradiol or its ester needs to beshown by epidemiological studies.

Progestogen selection: Progestogensare chosen with consideration ofwhether signs of androgenization arepresent (e.g. acne vulgaris, seborrhea,hirsutism). In this case, androgen-an-tagonist progestogens are prescribed,e.g. cyproterone acetate, chlormadinoneacetate, dienogest or drospirenone.

Although a higher VTE risk for theprogestogens desogestrel, gestoden andcyproterone acetate has not been estab-lished beyond doubt, a somewhat higherrisk than for other progestogens should beassumed in benefit/risk considerations inthe interest of patient protection.

5.2. Women with HeightenedThrombophilia Risk whoDesire ContraceptionFamily history:– Family history of CHD: Conditions/

events in parents before the age of 45(some sources use 50): myocardial inf-arction in the mother; stroke, thrombo-sis, thromboembolism in either parent.Diseases in the parents and siblings ofthe parents, as well as in the siblingsof the patient, can be added to the as-sessment.For CHD risk above and beyond VTErisk, metabolic conditions includinglipid metabolic disorders, diabetesmellitus, hypertension etc. also play arole.

– In the study by Dinger [9], approxi-mately 2% of women of fertile agehave a positive first-degree familyhistory of fatal myocardial infarction/stroke before the age of 50; similarly,first-degree family history of deepvenous thrombosis and pulmonary

embolism is reported by approxi-mately 3% of women of fertile age.

– For positive family histories of car-diovascular disease, further clarifica-tion may be needed by laboratorytesting, possibly family screening.

– In clinical practice, family history isoften more useful than laboratorytests for thrombophilia in assessingthe risk of venous thrombosis [31].

Thrombophilia testing (laboratory):– Thrombophilia parameters can help in

assessing the risk of VTE. However,laboratory tests show thrombophilia inonly approximately 50–60% of pa-tients who have had VTE. Negativethrombophilia results, therefore, donot exclude a higher risk of VTE. Thisis important to note when determiningrisk, especially if lab results are nega-tive but family history is positive.

5.3. Contraceptive Counselingwith Consideration of VTERisk FactorsNon-hormonal contraceptive methods:Barrier methods, chemical methods,natural family planning, etc. Contracep-tive effectiveness is inferior to that ofhormonal methods.

Copper IUDs are the only method thatthe WHO also recommends for womenwith a higher risk of thrombosis (grade 1classification), but are not consideredacceptable by all women.

Progestogen-only pills or non-estrogenovulation inhibitors: Epidemiologicalstudies show low risk of VTE [57].

Progestogen-only contraceptives do notappear to be associated with higher risk,and can therefore be used by thrombo-philic individuals or post-thromboembo-lism. The same applies to Mirena®.

The WHO classifies them as possible sec-ond-choice methods (grade 2). Consistentintervals (3-hour time window per day)are needed for the classical mini-Pill, thewindow for the non-estrogen ovulationinhibitor (Cerazette) is up to 12 hours.Higher occurrence of menstrual disorderscompared to COCs.

VTE risk data for the desogestrel-onlycontraceptive Cerazette are less robustthan those for levonorgestrel-only con-traceptives. It is therefore unclear



whether there are slight differences be-tween these non-estrogen preparations.

Hormonal implants: Implanon®: Noepidemiological studies on VTE risk. Noavailable data from authorization stud-ies.

Risk assessment similar to mini-Pill; incontrast to depot progestogens, Impla-non can be promptly removed if side ef-fects should occur.

Depot progestogens: No available datafrom authorization studies.

The WHO classifies them as a possiblesecond-choice method (grade 2); the au-thors of this paper advise caution, how-ever, because depending on the prepara-tion they can have long-term effects of 3–12 months following the last injection.

Combined oral hormonal contracep-tives, vaginal ring, and contraceptivepatch:– Absolute contraindication: Previ-

ous arterial or venous thromboembo-lism (without current therapeuticanticoagulation).

– Relative contraindication on indi-vidual case-by-case basis.

5.4. Limitation of the WHOContraception PrescriptionRecommendationsA limitation of the WHO contraceptionprescription recommendations [110,155] for women with specific medicalrisks is that these recommendations donot differentiate for thrombogenic muta-tions and whether these are homozygousor heterozygous (e.g. factor V Leidenmutation, prothrombin G20210A muta-tion). They also do not cover polymor-phisms or the degree of protein C, pro-tein S and antithrombin deficiency.

The WHO criteria only list the thrombo-philia results, without further differen-tiation:

“Known thrombogenic mutations(e.g., factor V Leiden; prothrombin mu-tation; protein S, protein C, and anti-thrombin deficiencies).”

For treating certain types of illnesses,gynecologists depend on hormonalmethods, which can include combinedoral hormonal contraceptives.

Table 16 differentiates between twogroups, of lower and higher VTE riskrespectively, for use of estrogen-con-taining contraceptives. Taking accountof these two risk groups, the flow dia-gram in Figure 12a, b shows treatmentrecommendations for estrogen-contain-ing COCs in connection with familyhistory, patient history, and thrombo-philia diagnoses. Ultimately, only in thecase of positive family history, negativepatient history, and negative throm-bophilia lab results can COCs be usedfor appropriate indications (see dis-claimer).

A relevant thrombosis risk with heredi-tary thrombophilia very probably onlyarises when in addition to the actual (ge-netic) feature, e.g. heterozygous factor VLeiden mutation, other factors are alsopresent which are currently not measur-able. They can be seen in anamneses: pa-tient history of thrombosis or embolism,family history, and/or multiple miscar-riages. This also applies to the data forcounseling pregnant women who show“only” heterozygous factor V Leidenmutation but no clinical prehistory.

Testing for thrombophilia should beprompted by positive patient or familyhistory; otherwise the laboratory resultsalone are of relatively little use in esti-mating individual risk.

For all patients with known thrombo-philia and who do not show a positiveown or family history: if gynecologistsprescribe hormonal contraception, theymust undertake careful risk/benefitanalyses and discuss them with the pa-tient – also regarding psycho-social is-sues in the event of unwanted pregnancy– in particular for girls/young women –as well as with respect to the consider-ably higher risk of VTE.

Also with these patients, alternatives tohormonal contraception should alwaysbe discussed in detail. In each case anindividual decision must be made withcareful discussion of the benefits andrisks and with “shared decision-making”considerations. The authors can onlypresent selection models here; liabilityin each individual case is borne entirelyby the attending physician.

It remains to note that this informationrefers to individual case decisions that

are not backed by studies and not consid-ered in the WHO recommendations.

The following thrombophilia constella-tions are a contraindication for combinedoral hormonal contraceptives, even if thepatient’s own and family history are nega-tive:– factor V Leiden homozygous,– prothrombin G20210A homozygous– factor V Leiden heterozygous + pro-

thrombin G20210A mutation– factor V Leiden homozygous + pro-

thrombin G20210A mutation– antithrombin deficiency6

– protein C deficiency6

– protein S deficiency6

6 No clear cut-off exists to separate severe casesfrom non-severe cases of inhibitor deficiencies(Luxembourg 2011, personal communication)

Table 16. VTE risk assessment for dif-ferent thrombophilia constellationswith class A lower and class B higher

Risk profile A

Moderately elevated DVT riskHeterozygous factor V Leiden mutationG1691A without own or family history ofDVT, and without additional risk factorsHeterozygous prothrombin G20210Amutation without own or family historyof DVT, and without additional risk factorsHeterozygous protein C deficiency with-out own or family history of DVT, andwithout additional risk factorsHeterozygous protein S deficiency with-out own or family history of DVT, andwithout additional risk factors

Risk profile B

Strongly elevated DVT riskPrevious thrombosis or arterial occlusionThrombophilia and positive family historyThrombophilia and additional risk factorssuch as– smoking– obesity– varicose veins– regular corticoid intake– chronic intestinal inflammationHomozygous factor V Leiden mutationG1691A, with or without DVT historyHomozygous prothrombin G20210Amutation with or without DVT historyCombined heterozygous factor V Leidenmutation + prothrombin G20210A muta-tion with or without DVT historyDemonstrated antiphospholipid syndromeAntithrombin deficiencyOther thrombophilia combinations

Insufficient study results for lipoprotein (a)No contraception limitations for MTHFRpolymorphism



For the following pathological results, aspecialist in internal medicine or ahemostaseologist specializing in throm-bophilia should be consulted:– presence of antiphospholipid anti-

bodies

5.5. Previous thrombosis com-bined with exogenous event(e.g. accident, immobilization)without laboratory confirmedthrombophiliaNo data confirm that COCs are safer forpatients with a history of reversible risk-

associated thrombosis than for thosewith a history of idiopathic thrombosis.Decisions as to whether the recurrencerisk following risk-associated VTE islow should not be made by the gynecolo-gist.

Instead, decisions should be made withan angiologist and hemostaseologist. Inany event, such decisions can only bemade on an individual case basis follow-ing careful risk/benefit analysis and pa-tient counseling regarding treatment op-tions and associated risks.

Even for women with a negative familyhistory and no laboratory confirmedthrombophilia, who have had a thrombo-sis in connection with an external non-hormonal dependent event, no estrogen-containing hormonal contraceptivesshould be prescribed because there arecurrently no data for this situation show-ing that estrogen-containing COCs donot increase the risk of recurrent throm-bosis. According to the WHO, the firstpreference should be a copper IUD (or ifno more children are planned, sterili-sation for either the man or the woman);as a less preferred alternative, consider-ation could be given to a classical mini-Pill with norethisterone or levonorges-trel or a levonorgestrel-containing IUD.In such cases the possible use of depotprogestogen injections or a desogestrel-containing mini-Pill (non-estrogen ovu-lation inhibitor), categorized by theWHO as grade 2, should be viewed withcaution. In relationships with rare sexualintercourse a combination of condomsand spermicides should be discussed, es-pecially if condoms are currently beingused for sexually transmissible diseaseprevention. But with unsure methods theconsiderably higher risk of VTE and ofunplanned pregnancy should also beconsidered.

However, if serious treatment consider-ations (e.g. severe menstrual distur-bances, acne vulgaris with severe dis-tress following exhaustion of dermato-logical means of treatment, etc.) requirethe use of an estrogen-containing COCin individual cases, the patient must beextensively informed and counseledabout the risk-benefit analyses and pos-sible consequences of treatment options,and this must be confirmed with a writ-ten declaration of informed consent aspart of the medical records.

Figure 12 a, b. Flow diagram for assessing VTE risk (based on family/patient history and thrombophilia tests) inchoice of contraceptive. Source: T. Rabe.

a

b



This statement is based on the followingobservations and studies: The risk of re-current thrombosis following an initialVTE is approximately 20–30% within10 years [156–159]. The lifetime risk ofrecurrent thrombosis is higher for menthan for women, with the latter showinga longer interval between first-time andrecurrent thromboses. Lijfering et al.(2009) [160] trace this to women havingfirst-time thromboses at a younger agedue to hormonal factors. In contrast tothis, Douketis et al. (2011) [161] showno difference between men and womenin recurrence with or without hormone-associated VTE. Christiansen et al.(2010) [162] indicate that the risk profilefor recurrent thrombosis differs consid-erably from that for first-time thrombo-sis.

Only one study group (49 women takingOCs, 18 postmenopausal women takingHRT) showed a lower risk of recurrentVTE for women with a non-provokedestrogen-dependent VTE, but the risk isalso not significantly lower than forwomen without estrogen intake [163].This small study could not be consideredin the above-described risk assessment.

Finally, it is unclear whether previousthrombosis/embolism with trigger fac-tors and without estrogen influence in-creases the risk of recurrent thrombosiswith estrogen intake. For this reason as ageneral principle women who have had athromboembolic event should not useestrogen-containing contraceptives.

5.6. Approach for SpecificRisk Groups– Women taking oral anticoagulants

should be actively offered contracep-tion on account of teratogenesis and asubstantially higher rate of pregnancycomplications. Every type of contra-ception is possible here, given stableadjustment of anticoagulation therapy.Ethinyl estradiol-containing contra-ceptives need to be discontinued 6weeks before anticoagulation isstopped. Careful INR adjustment andregular INR testing should be done atleast every two weeks.

– Regarding the choice of contracep-tion for patients with risk factors, seethe 2009 WHO Medical EligibilityCriteria for contraceptive use (2010update) [110], which analyze differ-ent patient health situations and rec-

ommend suitable contraceptive meth-ods (Tabs. 12, 13).

Depending on the seriousness of thelaboratory results – both separatelyand in combination – decisions mustbe made on an individual basis as towhether the woman should have to dowithout hormonal contraception.

5.7. General TreatmentRecommendations– If combined hormonal oral contra-

ceptives are used, then preparationswith lower EE doses – depending ontolerability and menstrual pattern –and progestogens with low VTE risk(Tab. 14).

– Reduce cigarette smoking if appli-cable, or better yet stop altogether

– Reduce weight if applicable– Avoid additional risk factors such as

exsiccosis, immobilization/inactivityduring extended travel.

5.8. Further Notes on Contra-ception DecisionsCombined oral hormonal contraceptivescan be required for “non-contraceptivebenefits”, namely for conditions thatcannot be treated otherwise.– certain types of acne vulgaris (fol-

lowing dermatological consultationand exhaustion of non-hormonalmeans of treatment). An alternativefor severe otherwise treatment-resist-ent acne as an “off-label” recommen-dation in individual cases: chlorma-dinone acetate (e. g. Chlormadinon2 mg fem Jenapharm®) (day 1–5) plustransdermal estrogen (e.g. Estreva®

gel, Gynokadin® gel) then 3-day break(Rott 2011, personal communication).

– irregular menses: following unsuc-cessful treatment with selected phyto-pharmaceuticals

– dysmenorrhea: following unsuccess-ful treatment with selected phyto-pharmaceuticals

In the event of positive patient or familyhistory, the decision as to whether to pre-scribe a hormonal contraceptive in anyindividual case must follow clinical con-sideration and individual, well docu-mented counseling.

Non-hormonal contraceptives can beconsidered as alternatives:– spermicides, condoms– copper IUD

– tubal sterilization– vasectomy

Follow-up exams for oral hormonalcontraceptives:Case history: If VTE risk factors arepresent, query for early symptoms ofcardiovascular disease (non-specific pain[see “ACHES” checklist]), migraines,impaired vision, medication history, legsymptoms.

Clinical testing: Blood pressure/pulserate at rest.

Laboratory tests:Only with relevant indication: e.g.HbA1c, lipids, HOMA index, liver values.

D-dimer test: D-dimer concentrationscan be helpful in assessing the recur-rence risk of venous thrombosis.

Paraclinical testing:Ultrasound only with relevant indica-tion: Androgenization with suspicion ofpolycystic ovaries, ovarian cysts, endo-metriosis, etc.

5.9. Notes on Evaluating Medi-cal Studies on VTE and Contra-ceptivesFor all case-control studies, it is impor-tant that the control group is representa-tive of the user population. In addition,cases and controls should be comparablewith respect to prevalence of majorknown prognostic factors. In theory,controls can be matched to cases. Inpractice, however, it is seldom possibleto find controls for each VTE case whomatch more than 2 or 3 of the prognosticfactors. It is therefore necessary to adjustfor the differences in “non-matched”prognostic factors. This means that in-formation on the major prognostic fac-tors – sex, age, weight (BMI), durationof exposure, cigarette smoking, familyhistory, and thrombophilia diagnoses –has to be available. Because it is knownthat approximately one half of VTEcases show pathological thrombophiliatest results, information about familyhistory at least should be available forboth cases and controls. This is generallyonly the case with field studies. Withoutknowledge of thrombophilia diagnosticsor family history, as was the case for allstudies in the 1990s on comparative VTErisk for second and third-generationprogestogens, it is not possible to con-



clusively evaluate VTE risk. Studiesshould also record duration of Pill usewhen thromboses occur as well as Pill-free intervals and previous use of otherOC preparations.

Reports on VTE and contraceptives inGermany are compiled by the Arznei-mittelkommission der deutschen Ärzte-schaft or by the Bundesinstitut für Arz-neimittel und Medizinprodukte. Thistype of spontaneous reports, however,cannot provide the basis for comparativerisk assessment of different prepara-tions. Such reports are better suited foridentifying signs of risk that should thenbe systematically investigated.

For Europe, data are compiled andevaluated by the European MedicinesAgency (EMA) in London, and for theUSA by the FDA.

A spontaneous reporting system yieldsonly a low percentage of even seriousundesired events. This percentage isstrongly influenced by external factors.Spontaneous reporting procedures byhealthcare professionals can be nega-tively influenced by possible liability is-sues. This is particularly the case in theUSA. On the other hand, aggressive ad-vertising by lawyers there can encouragepatients to report undesired results andthus establish causal connections be-tween conditions and potential factors.Independently of the individual country,however, the greatest influence is ex-erted by the respective mass media.

Hasty risk assessments on the basis of in-dividual studies are dangerous – whereascompilation of prospective data by large-scale studies with clear end points (e.g.the EURAS study) provides a more reli-able basis for evaluating the VTE risk ofhormonal contraceptives. For other ques-tions, it can often be cost-effective andtime-saving to use databases. When ex-amining the VTE risk for hormonal con-traception, however, it should be notedthat the data in these databases is not gath-ered for the primary purpose of scientificstudy. Diagnoses and tests for suspectedconditions are used to justify therapeuticand diagnostic measures – which meansthat diagnostic validity is often limited,only a small percentage of risk factors aredocumented, and exposure data are basedon prescriptions which in the case of OCsare not at all equivalent with the actual

start of exposure. Not infrequently, thisresults in erroneous assignment of severeundesired events and exposure.

Conflict of Interest

The author team reserves the right to up-date this statement in 12 months if newstudies lead to changes in relevant mate-rial.

T. Rabe has held talks for Jenapharm,Bayer-Schering Pharma, HRA, andMSD, receiving payment and in somecases travel expenses. Some advisoryboard activity. Chairman of theDeutsche Gesellschaft für Gynäkologi-sche Endokrinologie und Reproduk-tionsmedizin e.V. (DGGEF).

The DGGEF e.V. receives unrestrictededucational grants from Jenapharm,MSD, Bayer-Schering Pharma, HRA.

M. Ludwig has held events togetherwith Bayer-Schering Pharma, Grünen-thal and Jenapharm and received pay-ment for presentation or consultingfunctions. He is a member of a privatelaboratory and medical group that pro-vides gynecological-endocrinologicaland hemostaseological services.

H. Rott receives no funding from thepharmaceutical industry; she is a mem-ber of a private medical and laboratorygroup specializing in hemostaseologicalservices.

J. Dinger directs a private institute (Ber-lin Center for Epidemiology and HealthResearch) that performs epidemiologicalstudies financed by unconditional grantsfrom manufacturers of COCs – Bayer-Schering Pharma and MSD. Some ofthese studies are cited in this statement.No paid presentation or consulting work.

R. Bauersachs does not state any conflictof interest with respect to this statement.

A. O. Mueck has given talks for Bayer-Schering Pharma, MSD, Jenapharm, andHRA, receiving payment and in somecases travel expenses. Advisory boardactivities for the companies named.President of the Deutsche Gesellschaftfür Menopause.

B. Luxembourg works at the Depart-ment of Molecular Hemostaseology for

the German Red Cross Blood DonationService (DRK-Blutspendedienst) ofBaden Württemberg-Hessen and theUniversity Hospital Frankfurt. No otherpersonal or business sources for conflictof interest.

C. Albring is President of the Berufs-verband der Frauenärzte e.V. (GermanAssociation of Obstetricians and Gyne-cologists) and Chairman of the AG Hor-mone. He has no financial or other con-nection with any pharmaceutical indus-try.

References:

1. Luxembourg B, Krause M, Lindhoff-Last E. Basiswissen Ge-rinnungslabor; cme.aerzteblatt.de/kompakt, 2007; 14–5.

2. Cohen AT, Agnelli G, Anderson FA, Arcelus JI, Bergqvist D,Brecht JG, Greer IA, Heit JA, Hutchinson JL, Kakkar AK,Mottier D, Oger E, Samama MM, Spannagl M; VTE Impact As-sessment Group in Europe (VITAE). Venous thromboembolism(VTE) in Europe. The number of VTE events and associatedmorbidity and mortality. Thromb Haemost 2007; 98: 756–64.

3. Heit JA, Petterson T, Farmer S, Bailey K, Melton L. Trendsin Incidence of deep vein thrombosis and pulmonary embo-lism: a 35-year population-based study. Blood 2006; 108: 430.

4. Cushman M, Albert W, Tsai RH, White G, Susan R, HeckbertS, et al. Deep Vein Thrombosis and Pulmonary Embolism inTwo Cohorts: The Longitudinal Investigation of Thromboem-bolism Etiology. Am J Med 2004; 117: 1925.

5. Engbers MJ, van Hylckama Vlieg A, Rosendaal FR. Venousthrombosis in the elderly: incidence, risk factors and riskgroups. J Thromb Haemost 2010; 8: 2105–12.

6. Heit JA, Melton L, Lohse C, Petterson T, Silverstein M,Mohr D, OFallon W. Incidence of venous thromboembolism inhospitalized patients versus community residents. Mayo ClinProc 2001; 76: 1102–10.

7. Heit JA, OFallon W, Petterson T, Lohse C, Silverstein M,Mohr D, Melton L. Relative impact of risk factors for deepvein thrombosis and pulmonary embolism: a population-based study. Arch Intern Med 2002; 162: 1245–8.

8. Nicolaides AN, Fareed J, Kakkar AK, et al. Prevention andtreatment of venous thromboembolism. International Consen-sus Statement (guidelines according to scientific evidence).Int Angiology 2006; 25: 101–61.

9. Dinger JD, Heinemann LAJ, Kühl-Habich D. The safety of adrospirenone-containing oral contraceptive: final results fromthe European active surveillance study on oral contraceptivesbased on 142,475 women-years of observation. Contraception2007; 75: 344–54.

10. Rosendaal FR, Van Hylckama V, Tanis BC, Helmerhorst FM.Estrogens, progestogens and thrombosis. J Thromb Haemost2003; 1: 1371–80.

11. Wu O, Robertson L, Langhorne P, Twaddle S, Lowe GD,Clark P, Greaves M, Walker ID, Brenkel I, Regan L, Greer IA.Oral contraceptives, hormone replacement therapy, thrombo-philias and risk of venous thromboembolism: a systematic re-view. The Thrombosis: Risk and Economic assessment ofthrombophilia screening (TREATS) study. Thromb Haemost2005; 94: 17–25.

12. Gohil R, Peck G, Sharma P. The genetics of venous throm-boembolism. A meta-analysis involving ~120,000 cases and180,000 controls. Thromb Haemost 2009; 102: 360–70.

13. Scottish Intercollegiate Guidelines Network (2002). Pro-phylaxis of Venous Thromboembolism. A national clinicalguideline. Edinburgh (www.sign.ac.uk).

14. Scottish Intercollegiate Guidelines Network (2005). Pro-phylaxis of Venous Thromboembolism. A national clinicalguideline. Edinburgh (www.sign.ac.uk).

15. Moerchel C, Kroeger K. Prophylaxe tiefer Bein- und Be-ckenvenenthrombose. Dt Ärztebl 2007; 104: A2886-2893.

16. Merriman L, Greaves M. Testing for thrombophilia: an evi-dence-based approach. Postgraduate Med J 2006; 82: 699–704.

17. Dinger JD. Personal communication 2010.



18. Heit JA, Mohr D, Silverstein M, Petterson T, OFallon W,Melton L. Predictors of recurrence after deep vein thrombosisand pulmonary embolism: a population-based cohort study.Arch Intern Med 2000; 160: 761–8

19. Schulman S, Lindmarker P, Holmstrom M, Larfars S, Carls-son A, Nicol P, Svensson E, Ljungberg B, Viering S, NordlanderS, Leijd B, Jahed K, Hjorth M, Linder O, Beckmann M. Post-thrombotic syndrome, recurrence, and death 10 years afterthe first episode of venous thromboembolism treated withwarfarin for 6 weeks or 6 months. J Thromb Haemost 2006; 4:732–42.

20. Wells PS, Anderson DR, Bormanis J, Guy F, Mitchell M,Gray L, Clement C, Robinson KS, Lewandowski B: Value of as-sessment of pretest probability of deep-vein thrombosis inclinical management. Lancet 1997; 350: 1795–8.

21. Wells PS, Anderson DR, Bormanis J, Guy F, Mitchell M,Gray L, Clement C, Robinson KS, Lewandowski B. Value ofclinical management. Lancet 1997; 350: 1795–8.

22. Vandenbroucke JP, Koster T, Briët E, Reitsma PH, BertinaRM, Rosendaal FR. Increased risk of venous thrombosis inoral-contraceptive users who are carriers of factor V Leidenmutation. Lancet 1994; 344: 1453–7.

23. Vandenbroucke JP, van der Meer FJM, Helmerhorst FM,Rosendaal FR. Factor V Leiden: should we screen oral contra-ceptive users and pregnant women? BMJ 1996; 313: 1127–30.

24. Briët E, van der Meer FJM, Rosendaal FR, Houwing-Duistermaat JJ, van Houwelingen HC. The family history andinherited thrombophilia. Br J Haematol 1994; 87: 348–52.

25. Cosmi B, Legnani C, Bernardi F, Coccheri S, Palareti G.Value of family history in identifying women at risk of venousthromboembolism during oral contraception: observationalstudy. BMJ 2001; 322: 1024–5.

26. Aznar J, Mira Y, Vaya A, Ferrando F, Villa P. Is family his-tory sufficient to identify women with risk of venous throm-boembolism before commencing the contraceptive pill? ClinAppl Thromb Hemost. 2002; 8: 139–41.

27. Caprini JA, Goldshteyn S, Glase CJ, Hathaway K. Throm-bophilia testing in patients with venous thrombosis. Eur JVasc Endovasc Surg 2005; 30: 550–5.

28. van Sluis GL, Söhne M, El Kheir DY, Tanck MW, GerdesVEA, Büller HR. Family history and inherited thrombophilia.J Thromb Haemost 2006; 4: 2182–7.

29. Dowling NF, Austin H, Dilley A, Whitsett C, Evatt BL,Hooper WC. The epidemiology of venous thromboembolism inCaucasians and African-Americans: the GATE Study. J ThrombHaemost 2003; 1: 80–7.

30. Noboa S, Le Gal G, Lacut K, et al; for the EDITH Collabora-tive Study Group. Family history as a risk factor for venousthromboembolism. Thromb Res 2008; 122: 624–9.

31. Bezemer ID, van der Meer FJM, Eikenboom JCJ, Rosen-daal FR, Doggen CM. The Value of Family History as a Risk In-dicator for Venous Thrombosis. Arch Intern Med 2009; 169:610–15.

32. Anderson FA Jr, Spencer FA. Risk factors for venousthromboembolism. Circulation 2003; 107 (Suppl 1): I9–16.

33. Goodacre S, Sutton AJ, Sampson FC. Meta-analysis: thevalue of clinical assessment in the diagnosis of deep venousthrombosis. Ann Intern Med 2005; 143: 129–39.

34. Kuipers S, Cannegieter SC, Middeldorp S, et al. The abso-lute risk of venous thrombosis after air travel: a cohort studyof 8,755 employees of international organisations. PLoS Med2007; 4: e290.

35. Kuipers S, Schreijer AJ, Cannegieter SC, et al. Travel andvenous thrombosis: a systematic review. J Intern Med 2007;262: 615–34.

36. Geerts WH, Bergqvist D, Pineo GF, Heit JA, Samama CM,Lassen MR, & Colwell CW. Prevention of venous thromboem-bolism: American College of Chest Physicians Evidence-BasedClinical Practice Guidelines (8th Edition). Chest 2008; 133:381S-453S.

37. Kuipers S, Cannegieter SC, Middeldorp S, et al. The abso-lute risk of venous thrombosis after air travel: a cohort studyof 8,755 employees of international organisations. PLoS Med2007; 4: e290.

38. World Health Organization. WHO Research into GlobalHazards of Travel (WRIGHT) project: Final Report of Phase I.Geneva (Switzerland): World Health Organization; 2007[cited 2008 May 30]. Available from: http://www.who.int/cardiovascular_diseases/wright_project/en.

39. Hirsh J, Guyatt G, Albers GW, Harrington R, SchünemannHJ. American College of Chest Physicians Evidence-BasedClinical Practice Guidelines (8th Edition). Chest 2008; 133:71S–109S.

40. Schobersberger W, Toff WD, Eklöf B, Fraedrich G, GungaHC, Haas S, Landgraf H, Lapostolle F, Partsch H, Perschler F,Schnapka J, Schobersberger B, Scurr JH, Watzke H. Traveller’sthrombosis: International consensus statement. VASA 2008;37: 311–7.

41. Geerts WH, Heit JA, Clagett P, Pineo GF, Colwell CW,Anderson AF Jr, Brownell Wheeler H. Prevention of VenousThromboembolism. Chest 2001; 119: 132S–175S.

42. Geerts WH, Pineo GF, Heit JA, Bergqvist D, Lassen MR,Colwell CW, Ray JG. Prevention of venous thromboembolism:the Seventh ACCP Conference on Antithrombotic and Throm-bolytic Therapy. Chest 2004; 126: 338S–400S.

43. Encke A, Haas S, Sauerland S, Abholz HH, Beckmann MW,et al. S3-Leitlinie Prophylaxe der venösen Thromboembolie(VTE). Finale Version vom 18. März 2009. Eur J Vasc Med2009; 38 (Suppl 76): 1–131.

44. Snow V, Qaseem A, Barry P, Rodney H, Rodnick JE, et al.Management of Venous Thromboembolism: A clinical PracticeGuideline from the American College of Physicians and theAmerican Academy of Family Physicians. Ann Intern Med2007; 146: 204–10.

45. Rogers FB, Cipolle MD, Velmahos G, Rozycki G. A practicemanagement guidelines for the management of venousthromboebolism in trauma patients. EAST Practice ParameterWorkgroup for DVT Prophylaxis (http://www.east.org/tpg/dvt.pdf retrieved 2011-05-09)

46. Venous thromboembolism (surgical) (replaced by CG92):Venous thromboembolism: reducing the risk of venous throm-boembolism (deep vein thrombosis and pulmonary embolism)in inpatients undergoing surgery (http://www.nice.org.uk/CG46 retrieved 9 May 2011)

47. Royal College of Obstetricians and Gynaecologists.Venous Thromboembolism and Hormonal Contraception;Green-top Guideline No. 40 (http://www.rcog.org.uk/files/rcog-corp/GTG40VenousThromboEmbolism0910.pdf retrieved2010-05-11).

48. Reid R, Leyland N, Wolfman W, Allaire C, Awadalla A,Best C, Dunn S, Lemyre M, Marcoux V, Menard C, Potestio F,Rittenberg D, Singh S, Senikas V; Society of Obstetricians andGynaecologists of Canada. SOGC clinical practice guidelines:Oral contraceptives and the risk of venous thromboembolism:an update: no. 252, December 2010. Int J Gynaecol Obstet2011, 112: 252–6.

49. SOGC clincal practice guideline. Thromboembolism: AnUpdate. JOGC 2010; 1192–7.

50. Shah SH, Becker RC. Genetics of Thrombosis. In: AskariAT, Lincoff AM (eds). Antithrombotic Drug Therapy in Cardio-vascular Disease. Humana Press, Inc, Totowa, NJ; 2009.

51. Trégouët DA, Heath S, Saut N, Biron-Andreani C, SchvedJF, Pernod G, et al. Common susceptibility alleles are unlikelyto contribute as strongly as the FV and ABO loci to VTE risk:results from a GWAS approach. Blood 2009; 113: 5298–303.

52. Bezemer ID, Bare LA, Doggen CJ, Arellano AR, Tong C,Rowland CM, et al. Gene variants associaed with deep veinthrombosis. JAMA 2008; 299: 1306–14.

53. Martinelli I, Sacchi E, Landi G, Taioli E, Duca F, MannucciPM. High risk of cerebral-vein thrombosis in carriers of a pro-thrombin-gene mutation and in users of oral contraceptives.N Engl J Med 1998; 338: 1793–7.

54. Emmerich J, Rosendaal FR, Cattaneo M, Margaglione M,de Stefano V, Cumming T, Arruda V, Hillarp A, Reny J. Com-bined effect of factor V Leiden and prothrombin 20210A onthe risk of venous thromboembolism. Thromb Haemost 2001;86: 809–16.

55. Rosendaal FR, Vessey M, Rumley A, Daly E, Woodward M,Helmerhorst FM, et al. Hormonal replacement therapy, pro-thrombotic mutations and the risk of venous thrombosis. BrJ Haematol Mar 2002; 116: 851–4.

56. Juul K, Tybjaerg-Hansen A, Schnohr P, Nordestgaard BG.Factor V Leiden and the risk for venous thromboembolism inthe adult Danish population. Ann Intern Med 2004; 140: 330–7.

57. Lidegaard Ø, Løkkegaard E, Svendsen AL, Agger C. Hor-monal contraception and risk of venous thromboembolism:national follow-up study. BMJ 2009; 339: b2890.

58. Dearborn JT, Hu SS, Tribus CB, Bradford DS. Thromboem-bolic complications after major thoracolumbar spine surgery.Spine 1999; 24: 1471–6.

59. Faunø P, Suomalainen O, Rehnberg V, Hansen TB, KronerK, Soimakallio S, Nielsen E. Prophylaxis for the prevention ofvenous thromboembolism after total knee arthroplasty. Acomparison between unfractionated and low-molecular-weight heparin. J Bone Joint Surg Am 1994; 76: 1814–8.

60. Grady D, Wenger NK, Herrington D, Khan S, Furberg C,Hunninghake D, Vittinghoff E, Hulley S. Postmenopausal hor-mone therapy increases risk for venous thromboembolic dis-ease. The Heart and Estrogen/progestin Replacement Study.Ann Intern Med 2000; 132: 689–96.

61. Greenfield LJ, Proctor MC, Rodriguez JL, Luchette FA,Cipolle MD, Cho J. Posttrauma thromboembolism prophylaxis.J Trauma 1997; 42: 100–3.

62. Harenberg J, Roebruck P, Stehle G, Habscheid W,Biegholdt M, Heene DL. Heparin Study in Internal Medicine(HESIM): design and preliminary results. Thromb Res 1992;68: 33–43.

63. Leizorovicz A, Simonneau G, Decousus H, Boissel JP. Com-parison of efficacy and safety of low molecular weight hep-arins and unfractionated heparin in initial treatment of deepvenous thrombosis: a meta-analysis. BMJ 1994; 309: 299–304.

64. Lindqvist P, Dahlback B, Marsal K. Thrombotic risk duringpregnancy: a population study. Obstet Gynecol 1999; 94: 595–9.

65. Tincani E, Piccoli M, Turrini F, Crowther MA, Melotti G,Bondi M. Video laparoscopic surgery: is out-ofhospitalthromboprophylaxis necessary? J Thromb Haemost 2005; 3:216–20.

66. Turpie AG, Bauer KA, Eriksson BI, Lassen MR. Fondaparinuxvs enoxaparin for the prevention of venous thromboembolismin major orthopedic surgery: a meta-analysis of 4 randomizeddouble-blind studies. Arch Intern Med 2002; 162: 1833–40.

67. Lowe GD, Haverkate F, Thompson SG, Turner RM, BertinaRM, Turpie AG, Mannucci PM. Prediction of deep vein throm-bosis after elective hip replacement surgery by preoperativeclinical and haemostatic variables: the ECAT DVT Study. Euro-pean Concerted Action on Thrombosis. Thromb Haemost 1999;81: 879–86.

68. Mantilla CB, Horlocker TT, Schroeder DR, Berry DJ, BrownDL. Risk factors for clinically relevant pulmonary embolismand deep venous thrombosis in patients undergoing primaryhip or knee arthroplasty. Anesthesiology 2003; 99: 552–60.

69. Selby R, Geerts WH. Venous thromboembolism: risk fac-tors and prophylaxis. Semin Respir Crit Care Med 2000; 21:493–501.

70. Dinger J et al. Oral contraceptive effectiveness accordingto body mass index, weight, age, and other factors. Am JObstet Gynecol 2009; 201: 263.

71. Dinger J, Thai DM, Buttmann N, Bardenheuer K. Effective-ness of oral contraceptive pills in a large U.S. cohort compar-ing progestogen and regimen. Obstet Gynecol 2011; 117: 33–40.

72. Heit JA, Kobbervig CE, James AH, Petterson TM, BaileyKR, Melton LJ. Trends in the incidence of venous thromboem-bolism during pregnancy or postpartum: A 30-year population-based study. Ann Intern Med 2005; 143: 697–706.

73. Rees D. The population genetics of factor V Leiden(Arg506Gln). Br J Haematol 1996; 95: 579–86.

74. Rosendaal FR, Koster T, Vandenbroucke JP, Reitsma PH:High risk of thrombosis in patients ho-mozygous for factor VLeiden (activated protein C resistance). Blood 1995; 85: 1504–8.

75. Kurnik D, Lubetsky A. Genetic variants and risk for venousthromboembolic events: summing up the evidence. ThrombHaemost 2009; 102: 183–4.

76. Segal JB, Brotman DJ, Necochea AJ, Emadi A, Samal L,Wilson LM, et al. Predictive value of factor V Leiden and pro-thrombin G20210A in adults with venous thromboembolismand in family members of those with a mutation: a systematicreview. JAMA 2009; 301: 2472–85.

77. Wu O, Robertson L, Twaddle S, et al. Screening for throm-bophilia in high-risk situations: a meta-analysis and cost-ef-fectiveness analysis. Br J Haematol 2005; 131: 80–90.

78. Luxembourg B, Delev D, Geisen C, Spannagl M, Krause M,Miesbach W, Heller C, Bergmann F, Schmeink U, GrossmannR, Lindhoff-Last E, Seifried E, Oldenburg J, Pavlova A. Mo-lecular basis of antithrombin deficiency. Thromb Haemost2011; 105 [Epub ahead of print].

79. van Vlijmen EF, Brouwer JL, Veeger NJ, Eskes TK, de GraeffPA, van der Meer J. Oral con-traceptives and the absolute riskof venous thromboembolism in women with single or multiplethrombophilic defects: results from a retrospective family co-hort study. Arch Intern Med 2007; 167: 282–9.

80. Luxembourg B, Schmitt J, Humpich M, Glowatzki M,Seifried E, Lindhoff-Last E. Intrinsic clotting factors in depen-dency of age, sex, body mass index, and oral contraceptives:definition and risk of elevated clotting factor levels. BloodCoagul Fibrinolysis 2009; 20: 524–34.



81. Legnani C, Cini M, Cosmi B, Poggi M, Boggian O, PalaretiG. Risk of deep vein thrombosis: interaction between oral con-traceptives and high factor VIII levels. Haematologica 2004;89: 1347–51.

82. Bloemenkamp KW, Helmerhorst FM, Rosendaal FR,Vandenbroucke JP. Venous thrombosis, oral contraceptivesand high factor VIII levels. Thromb Haemost 1999; 82: 1024–7.

83. Wahl DG, Guillemin F, de Maistre E, Perret-Guillaume C,Lecompte T, Thibaut G. Meta-analysis of the risk of venousthrombosis in individuals with antiphospholipid antibodieswith-out underlying autoimmune disease or previous thrombo-sis. Lupus 1998; 7: 15–22.

84. de Groot PG, Lutters B, Derksen RH, Lisman T, Meijers JC,Rosendaal FR. Lupus anticoagulants and the risk of a first epi-sode of deep venous thrombosis. J Thromb Haemost 2005; 3:1993–7.

85. Galli M, Luciani D, Bertolini G, Barbui T. Anti-beta 2-glyco-protein I, antiprothrombin antibodies, and the risk of thrombo-sis in the antiphospholipid syndrome. Blood 2003; 102: 2717–23.

86. de Laat B, Pengo V, Pabinger I, Musial J, Voskuyl AE,Bultink IE, Ruffatti A, Rozman B, Kve-der T, de Moerloose P,Boehlen F, Rand J, Ulcova-Gallova Z, Mertens K, de Groot PG.The association between circulating antibodies against do-main I of beta2-glycoprotein I and thrombosis: an interna-tional multicenter study. J Thromb Haemost 2009; 7: 1767–73.

87. Giannakopoulos B, Passam F, Ioannou Y, Krilis SA. Howwe diagnose the antiphospholipid syndrome. Blood 2009; 113:985–94.

88. Urbanus RT, Siegerink B, Roest M, Rosendaal FR, de GrootPG, Algra A. Antiphospholipid antibodies and risk of myocardialinfarction and ischaemic stroke in young women in the RATIOstudy: a case-control study. Lancet Neurol 2009; 8: 998–1005.

89. Metjian A, Lim W. ASH evidence-based guidelines: shouldasymptomatic patients with an-tiphospholipid antibodies re-ceive primary prophylaxis to prevent thrombosis? HematologyAm Soc Hematol Educ Program 2009: 247–9.

90. de Bree A, van der Put NM, Mennen LI, Verschuren WM,Blom HJ, Galan P, Bates CJ, Herrmann W, Ullrich M, Dierkes J,Westphal S, Bouter LM, Heine RJ, Stehouwer CD, Dekker JM,Nijpels GN, Araújo F, Cunha-Ribeiro LM, Refsum H, Vollset S,Nygard O, Ueland PM. Prevalences of hyperhomocysteinemia,unfavorable cholesterol profile and hypertension in Europeanpopulations. Eur J Clin Nutr 2005; 59: 480–8.

91. den Heijer M, Lewington S, Clarke R. Homocysteine,MTHFR and risk of venous thrombosis: a meta-analysis ofpublished epidemiological studies. J Thromb Haemost 2005;3: 292–9.

92. Kelly PJ, Rosand J, Kistler JP, Shih VE, Silveira S,Plomaritoglou A, Furie KL. Homocysteine, MTHFR 677C—>Tpolymorphism, and risk of ischemic stroke: results of a meta-analysis. Neurology 2002; 59: 529–36.

93. Martinelli I, Battaglioli T, Burgo I, Di Domenico S, MannucciPM. Oral contraceptive use, thrombophilia and their interac-tion in young women with ischemic stroke. Haematologica2006; 91: 844–7.

94. den Heijer M, Willems HP, Blom HJ, Gerrits WB, CattaneoM, Eichinger S, Rosendaal FR, Bos GM. Homocysteine lower-ing by B vitamins and the secondary prevention of deep veinthrombosis and pulmonary embolism: A randomized, placebo-controlled, double-blind trial. Blood 2007; 109: 139–44.

95. Martí-Carvajal AJ, Solà I, Lathyris D, Salanti G. Homocys-teine lowering interventions for pre-venting cardiovascularevents. Cochrane Database Syst Rev 2009; 7(4): CD006612.

96. Jamison RL, Hartigan P, Kaufman JS, Goldfarb DS, WarrenSR, Guarino PD, Gaziano JM; Veterans Affairs Site Investiga-tors. Effect of homocysteine lowering on mortality and vascu-lar disease in advanced chronic kidney disease and end-stagerenal disease: a randomized con-trolled trial. JAMA 2007;298: 1163–70.

97. Ray JG, Kearon C, Yi Q, Sheridan P, Lonn E; Heart Out-comes Prevention Evaluation 2 (HOPE-2) Investigators. Ho-mocysteine-lowering therapy and risk for venous thromboem-bolism: a randomized trial. Ann Intern Med 2007; 146: 761–7.

98. Franco RF, Araújo AG, Guerreiro JF, Elion J, Zago MA.Analysis of the 677 C—>T mutation of the methylenetetra-hydrofolate reductase gene in different ethnic groups. ThrombHaemost 1998; 79: 119–21.

99. Xin XY, Song YY, Ma JF, Fan CN, Ding JQ, Yang GY, ChenSD. Gene polymorphisms and risk of adult early-onset ischemicstroke: A meta-analysis. Thromb Res 2009; 124: 619–24.

100. Lewis SJ, Ebrahim S, Davey Smith G. Meta-analysis ofMTHFR 677C—>T polymorphism and coronary heart disease:does totality of evidence support causal role for homocysteineand preventive potential of folate? BMJ 2005; 331: 1053.

101. Slooter AJ, Rosendaal FR, Tanis BC, Kemmeren JM, vander Graaf Y, Algra A. Prothrombotic conditions, oral contracep-tives, and the risk of ischemic stroke. J Thromb Haemost2005; 3: 1213–7.

102. Pezzini A, Grassi M, Iacoviello L, Del Zotto E, Archetti S,Giossi A, Padovani A. Inherited thrombophilia and stratifica-tion of ischaemic stroke risk among users of oral contracep-tives. J Neurol Neurosurg Psychiatry 2007; 78: 271–6.

103. Marchiori A, Mosena L, Prins MH, Prandoni P. The risk ofrecurrent venous thromboembolism among heterozygous carri-ers of factor V Leiden or prothrombin G20210A mutation. Asystematic review of prospective studies. Haematologica2007; 92: 1107–14.

104. Brouwer JL, Lijfering WM, Ten Kate MK, Kluin-NelemansHC, Veeger NJ, van der Meer J. High long-term absolute riskof recurrent venous thromboembolism in patients with heredi-tary deficiencies of protein S, protein C or antithrombin.Thromb Haemost 2009; 101: 93–9.

105. Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE,Comerota AJ; American College of Chest Physicians. Anti-thrombotic therapy for venous thromboembolic disease:American College of Chest Physicians Evidence-Based ClinicalPractice Guidelines (8th Edition). Chest 2008; 133 (6 Suppl):454S–545S.

106. Heinemann LAJ, Dinger JC. Range of published esti-mates of VTE incidence in young women. Contraception 2007;75: 328–36.

107. Bruce FC, Berg CJ, Hornbrook MC, Whitlock EP, CallaghanWM, Bachman DJ, Gold R, Dietz PM. Maternal morbidity ratesin a managed care population. Obstet Gynecol 2008; 111:1089–95.

108. Pomp ER, Rosendaal FR, Doggen CJ. Smoking increasesthe risk of venous thrombosis and acts synergistically withoral contraceptive use. Am J Hematol 2008; 83: 97–102.

109. Salonen Ros H, Lichtenstein P, Bellocco R, Petersson G,Cnattingius S. Increased risks of circulatory diseases in latepregnancy and puerperium. Epidemiology 2001; 12: 456–60.

110. WHO. Medical eligibility criteria for contraceptive use:4th ed. 2009. WHO 2010.

111. Lidegaard Ø, Edstrom B, Kreiner S. Oral contraceptivesand venous thromboembolism: a five-year national case-con-trol study. Contraception 2002; 65: 187–96.

112. Dinger JD. Abstract auf dem Congress der European So-ciety of Contraception, Prague 2008.

113. van Hylckama FM, Vlieg A, Helmerhorst FM, VandenbrouckeJ P, Doggen CJM, Rosendaal FR. The venous thrombotic riskof oral contraceptives, effects of oestrogen dose and progesto-gen type: results of the MEGA case-control study. BMJ 2009;339: b2921.

114. Dinger J, Moehner S, Do Minh T. Early use effects on therisk of venous thromboembolism after initiation of oral contra-ceptive use. Eur J Contracept Reprod Health Care 2010; 15(Suppl 1): 43.

115. Gomes MP, Deitcher SR. SO, Risk of venous thromboem-bolic disease associated with hormonal contraceptives andhormone replacement therapy: a clinical review. Arch InternMed 2004; 164: 1965–76.

116. Winkler UH. Hemostatic effects of third- and second-gen-eration oral contraceptives: absence of a causal mechanism fora difference in risk of venous thromboembolism. Contraception2000; 62 (Suppl): 11S-20S.

117. Dinger J, Assmann A, Möhner S, Do Minh T. Risk of venousthromboembolism and the use of dienogest- and drospirenone-containing oralcontraceptives: results from a German case-control study Fam Plann Reprod Health Care 2010; 36: 123–9.

118. World Health Organization Collaborative Study of Cardio-vascular Disease and Steroid Hormone Contraception. Venousthromboembolic disease and combined oral contraceptives:results of international multicentre case-control study. Lancet1995; 346: 1575–82.

119. Rosendaal FR, Helmerhorst FM, Vandenbroucke JP. Fe-male hormones and thrombosis. Arterioscler Thromb Vasc Biol2002; 22: 201–10.

120. Kemmeren J, Algra A, Grobbee D. Third generation oralcontraceptives and risk of venous thrombosis: meta-analysis.Br Med J 2001; 323: 131–4.

121. Lidegaard Ø. Absolute and attributable risk of venousthromboembolism in women on cyproterone acetate. J ObstetGynaecol Canada 2003; 25: 575–7.

122. Waldman-Rex S, Schramm G. VTE-Risiko unter oralenKontrazeptiva: Fundierte Datenlage bei Belara® (2 mg CMA/0,03 mg EE). Gyne 2009; 10: 33.

123. Seeger JD, Loughlin J, Eng P, Clifford C, Robin MS, CutoneJ, Walker A. Risk of Thromboembolism in Women Taking Ethi-

nylestradiol/Drospirenone and Other Oral Contraceptives.Obst Gynecol 2007; 110: 587–893.

124. Dinger J. Oral contraceptives and venous thromboembo-lism: old questions revisited. J Fam Plann Reprod Health Care2009; 35: 211–2.

125. Shapiro S, Dinger J. Risk of venous thromboembolismamong users of oral contraceptives: a review of two recentlypublished studies. J Fam Plann Reprod Health Care 2010; 36:33–8.

126. Severinsen MT, Kristensen SR, Overvad K, Dethlefsen C,Tjønneland A, Johnsen SP. Venous thromboembolism dischargediagnoses in the Danish National Patient Registry should beused with caution. J Clin Epidemiol 2010; 63: 223–8.

127. Jick SS, Hernandez RK. Risk of non-fatal thromboembo-lism in women using oral contraceptives containing drospiren-one compared with women using oral contraceptives contain-ing levonorgestrel: case-control study using United Statesclaims data. BMJ 2011; 342: d2151.

128. Parkin L, Sharples K, Hernandez RK, Jick SS. Risk ofvenous thromboembolism in users of oral contraceptives con-taining drospirenone or levonorgestrel: nested case-controlstudy based on UK General Practice Research Database. BMJ2011; 342: d2139.

129. www.ema.europa.eu/docs/en_GB/document_library/Re-port/2011/05/WC500106708.pdf

130. http://www.bfarm.de/DE/Pharmakovigilanz/risikoinfo/2011/drospirenon.html

131. van Hylckama Vlieg A, Helmerhorst FM, Rosendaal FR.The risk of deep venous thrombosis associated with inject-able depot – medroxyprogesterone acetate contraceptives ora levonorgestrel intrauterine device. Arterioscler Thromb VascBiol November 2010. DOI: 10.1161/ATVBAHA.110.211482

132. High Court of Justice. Approved judgement case No0002638. Neutral Citation No: [2002] EWHC 1420 (QB).http://www.hmcourts-service.gov.uk/judgmentsfiles/j1298/xyz_-v-schering.htm

133. Dore DD, Norman H, Loughlin J, Seeger JD. Extendedcase-control study results on thromboembolic outcomesamong transdermal contraceptive users. Contraception 2010,81: 408–13

134. Jick SS, Hagberg KW, Hernandez RK, Kaye JA. Postmar-keting study of ORTHO EVRA® and levonorgestrel oral contra-ceptives containing hormonal contraceptives with 30 mcg ofethinyl estradiol in relation to nonfatal venous thromboembo-lism. Contraception 2010; 81: 16–21.

135. Lidegaard Ø. Incidence rate of VTE among pregnant andpuerperal women, DK 1994–96. (http://www.lidegaard.dk/Slides/OC%20epidem/PP%2007-11-20%20en.pdf accessedAug. 9, 2011)

136. European Medicines Agency. EVRA – Procedural stepstaken and scientific information after the authorization. 2010:www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000410/WC500031512.pdf

137. Cole JA, Norman H, Doherty M, Walker AM. Venousthromboembolism, myocardial infarction, and stroke amongtransdermal contraceptive system users. Obstet Gynecol2007; 109: 339–46.

138. Cole JA, Norman H, Doherty M, Walker AM. Venousthromboembolism, myocardial infarction, and stroke amongtransdermal contraceptive system users. Obstet Gynecol2008; 111: 1449.

139. Dore D, Norman H, Seeger, J. Eligibility criteria in venousthromboembolism, myocardial infarction, and stroke amongtransdermal contraceptive system users. Letter to the Editor.Obstet Gynecol 2009; 114: 175.

140. Jick SS, Kaye JA, Russmann S, Jick H. Risk of nonfatalvenous thromboembolism in women using a contraceptivetransdermal patch and oral contraceptives containing nor-gestimate and 35 mcg of ethinyl estradiol. Contraception2006; 73: 223–8.

141. Jick S, Kaye JA, Jick H. Further results on the risk ofnonfatal venous thromboembolism in users of the contracep-tive transdermal patch compared to users of oral contracep-tives containing norgestimate and 35 µg of EE. Contraception2007; 76: 4–7.

142. Jick S, Hagberg K, Kaye J. ORTHO EVRA® and venousthromboembolism: an update. Letter to the Editor. Contracep-tion 2010; 81: 452–3.

143. Evra drug safety report. The risk of venous thromboembol-ism, myocardial infarction and ischemic stroke among womenusing the transdermal contraceptive system compared towomen using norgestimate-containing oral contraceptives with35 µg ethinylestradiol. Revised Final Report, January 2009.



144. Magnusdottir EM, Bjarnadottir RI, Onundarson PT,Gudmundsdottir BR, Geirsson RT, Magnusdottir DS, Dieben TO.The contraceptive vaginal ring (NuvaRing®) and hemostasis: acomparative study. Contraception 2004; 69: 461–7.

145. Winkler UH, Howie H, Bühler K, Korver T, Geurts TBP,Coelingh Bennink HJT. A randomized controlled double-blindstudy of the effects on hemostasis of two progestogen-onlypills containing 75 µg desogestrel or 30 µg levonorgestrel.Contraception 1998, 57: 385–92.

146. Bennink HJ. The pharmacokinetics and pharmacodynam-ics of Implanon, a single-rod etonogestrel contraceptive im-plant. Eur J Contracept Reprod Health Care 2000; 5 (Suppl 2):12–20.

147. Egberg N, van Beek A, Gunnervik C, Hulkko S, Hirvonen E,Larsson-Cohn U, Coelingh Bennink H. Effect of the hemostaticsystem and liver function in relation to Implanon® and Norplant®:a prospective randomized clinical trial. Contraception 1998;58: 93–8.

148. Winkler U. Patientin mit multipler Thrombose. WelcheAntikonzeption und wie abrechnen? Leser fragen Experten.Gynecol Tribune 20. Januar 2004; 5. Jahrgang, Nr. 1/2.

149. Fahmy K, Khairy M, Allam G, Gobran F, Alloush M. Effectof depo-medroxyprogesterone acetate on coagulation factorsand serum lipids in Egyptian women. Contraception 1991, 44:431–44.

150. World Health Organization Collaborative Study of Cardio-vascular Disease and Steroid Hormone Contraception. Cardio-

vascular Disease and Use of Oral and Injectable Progestogen-Only Contraceptives and Combined Injectable ContraceptivesResults of an International, Multicenter, Case-Control Study.Contraception 1998; 57: 315–24.

151. Chu MC, Zhang X, Gentzschein E, Stanczyk FZ, Lobo RA.Formation of ethinyl estradiol in women during treatment withnorethindrone acetate. J Clin Endocrinol Metabol 2007; 92:2205–7.

152. Vasilakis C, Jick SS, Jick H. The risk of venous thrombo-embolism in users of postcoital contraceptive pills. Contra-ception 1999; 59: 79–83.

153. van Rooijen M, Berntorp E, Bremme K. Thrombin genera-tion after emergency contraception. Thrombosis Research2009, 123: 152.

154. Schaefer C, Hannemann D, Meister R, Eléfant E, PaulusW, Vial T, Reuvers M, Robert-Gnansia E, Arnon J, De SantisM, Clementi M, Rodriguez-Pinilla E, Dolivo A, Merlob P. Vita-min K antagonists and pregnancy outcome. A multi-centreprospective study. Thromb Haemost 2006; 95: 949–57.

155. WHO. Medical eligibility criteria for contraceptive use:4th ed. 2009. WHO 2009.

156. Hansson PO, Sorbo J, Eriksson H. Recurrent venousthromboembolism after deep vein thrombosis: incidence andrisk factors. Arch Intern Med 2000; 160: 769–74.

157. Prandoni P, Lensing AW, Cogo A, Cuppini S, Villalta S,Carta M, Cattelan AM, Polistena P, Bernardi E, Prins MH. The

long-term clinical course of acute deep venous thrombosis.Ann Intern Med 1996; 125: 1–7.158. Baglin T, Luddington R, Brown K, Baglin C. Incidence ofrecurrent venous thromboembolism in relation to clinical andthrombophilic risk factors: prospective cohort study. Lancet2003; 362: 523–6.159. Christiansen SC, Cannegieter SC, Koster T, VandenbrouckeJP, Rosendaal FR. Thrombophilia, clinical factors, and recurrentvenous thrombotic events. JAMA 2005; 293: 2352–61.160. Lijfering W, Veeger NJ, Middeldorp S, Hamulyák K, PrinsMH, Büller HR, van der Meer J. A lower risk of recurrentvenous thrombosis in women compared with men is explainedby sex-specific risk factors at time of first venous thrombosisin thrombophilic families. Blood 2009; 114: 2031–6.161. Douketis J, Tosetto A, Marcucci M, Baglin T, Cosmi B,Cushman M, Kyrle P, Poli D, Campbell Tait R, Iorio A. Risk ofrecurrence after venous thromboembolism in men and women:patient level meta-analysis. BMJ 2011; 342: d813.162. Christiansen SC, WMJ Fering, FM Helmerhorst, FR Rosen-daal, SC Cannegieter. Sex difference in risk of recurrent venousthrombosis and the risk profile for a second event, J ThrombHaemost 2010; 8: 2159–68.163. Gal G, Kovacs MJ, Carrier M, Do K, Kahn SR, Wells PS,Anderson DA, Chagnon I, Solymoss S, Crowther M, Righini M,Lacut K, White RH, Vickars L, Rodger M. Risk of recurrentvenous thromboembolism after a first oestrogen-associatedepisode. Data from the REVERSE cohort study. Thromb Haemost2010; 104: 498–503.

Haftungsausschluss

Die in unseren Webseiten publizierten Informationen richten sich ausschließlich an geprüfte und autorisierte medizinische Berufsgruppen und entbinden nicht von der ärztlichen Sorg-faltspflicht sowie von einer ausführlichen Patientenaufklärung über therapeutische Optionen und deren Wirkungen bzw. Nebenwirkungen. Die entsprechenden Angaben werden von den Autoren mit der größten Sorgfalt recherchiert und zusammengestellt. Die angegebenen Do-sierungen sind im Einzelfall anhand der Fachinformationen zu überprüfen. Weder die Autoren, noch die tragenden Gesellschaften noch der Verlag übernehmen irgendwelche Haftungsan-sprüche.

Bitte beachten Sie auch diese Seiten:

Impressum Disclaimers & Copyright Datenschutzerklärung

Mitteilungen aus der Redaktion

e-Journal-AboBeziehen Sie die elektronischen Ausgaben dieser Zeitschrift hier.

Die Lieferung umfasst 4–5 Ausgaben pro Jahr zzgl. allfälliger Sonderhefte.

Unsere e-Journale stehen als PDF-Datei zur Verfügung und sind auf den meisten der markt-üblichen e-Book-Readern, Tablets sowie auf iPad funktionsfähig.

Bestellung e-Journal-Abo

Haftungsausschluss

Die in unseren Webseiten publizierten Informationen richten sich ausschließlich an geprüfte und autorisierte medizinische Berufsgruppen und entbinden nicht von der ärztlichen Sorg-faltspflicht sowie von einer ausführlichen Patientenaufklärung über therapeutische Optionen und deren Wirkungen bzw. Nebenwirkungen. Die entsprechenden Angaben werden von den Autoren mit der größten Sorgfalt recherchiert und zusammengestellt. Die angegebenen Do-sierungen sind im Einzelfall anhand der Fachinformationen zu überprüfen. Weder die Autoren, noch die tragenden Gesellschaften noch der Verlag übernehmen irgendwelche Haftungs-ansprüche.

Bitte beachten Sie auch diese Seiten:

Impressum Disclaimers & Copyright Datenschutzerklärung

Mitteilungen aus der Redaktion

e-Journal-AboBeziehen Sie die elektronischen Ausgaben dieser Zeitschrift hier.

Die Lieferung umfasst 4–5 Ausgaben pro Jahr zzgl. allfälliger Sonderhefte.

Unsere e-Journale stehen als PDF-Datei zur Verfügung und sind auf den meisten der markt-üblichen e-Book-Readern, Tablets sowie auf iPad funktionsfähig.

Bestellung e-Journal-Abo

Besuchen Sie unsere Rubrik

Medizintechnik-Produkte

InControl 1050 Labotect GmbH

Aspirator 3 Labotect GmbH

Philips Azurion: Innovative Bildgebungslösung

Neues CRT-D Implantat Intica 7 HF-T QP von Biotronik

Artis pheno Siemens Healthcare Diagnostics GmbH

https://www.kup.at/impressum/index.html

https://www.kup.at/impressum/disclaimer.html

https://www.kup.at/impressum/datenschutz.html

https://www.kup.at/cgi-bin/gratis-abo.pl

https://www.kup.at/perl/abbildungen.pl?q=Medizinprodukte+Labortechnik&ct=1&b=a

Documents

Journal für Reproduktionsmedizin und EndokrinologieLaboratory testing: General screening for thrombophilia prior to the prescription of oral contraceptives (OC) is not recommended