To our knowledge, there are no long-term community-based cohort studies designed to evaluate or compare the risk of inherited or acquired risk factors for venous thromboembolism (VTE) in women under the age of 50 years, except for a Danish cohort study which, at least initially, targeted general at cardiovascular risk factors and not VTE specifically (DNA was obtained later during follow-up) . Moreover, this study was specifically focused on factor V Leiden.
It was our aim to evaluate or compare the absolute risk and risk ratio of established clinical or genetic risk markers for VTE. In the past, most risk factor studies for VTE were restricted to clinically available markers such as age, BMI, previous VTE, family history, or acute factors (immobilization, surgery, accidents, pregnancy/puerperium, and hormonal contraceptive use) and based on clinical or cross-sectional, observational studies or analyses in administrative databases. Many observational studies or cohort studies in young women did not consider inherited factors (overview about incidence and risk factor studies in [7, 8]). Cohort studies in the population rarely included or reported genetic markers for thrombophilia and acquired, lifestyle-related risk factors, except the Physicians Health Study for example – the latter however only for males over 40 years of age , or the above mentioned Danish cohort study .
Other studies with focus on markers for hereditary thrombophilia were performed in patients (e.g. in anticoagulant clinics), in relatives of carriers of genetic mutations but not in the general population [6, 10–12]. Point estimates for thrombosis-free survival in carriers of major thrombophilic states are often restricted to the selected cohort of family members only (overview in Crowther ). Moreover, the evaluation of genetic markers often does not consider the impact of clinically available risk factors and the design was mainly restricted to clinical or case-control studies.
Thirty-four VTE cases, classified as definite or probable, occurred within this period, which is equivalent with about 10 per 10,000 WYs. At the first glance, this incidence seems to be high. However, this might be the result of the specifics of our study: We put great effort on the detection of potential cases and – even more important – we included all definite and probable cases, whereas most of the reported incidence rates in young women refer only to "confirmed" and so-called "idiopathic VTE", i.e. excluded all cases that occurred in temporal relationship to other potential reasons such as pregnancy/puerperium, surgery, or immobilization, for example. A similar overall incidence rate of 12.3 events/104 person-years was observed in the Danish cohort study [calculated from – 1 -], which however covers both gender and a higher mean age (45 years in the Danish study vs. 26 years in our study).
Idiopathic VTE, however, reflects only a smaller part of all confirmed VTE cases . In our cohort study we found roughly 53 % so-called "idiopathic" (primary) VTE cases, and the other roughly 47% of cases had a previous VTE in their history, or pregnancy/puerperium, surgery, or other reasons for immobilization/long bed-rest shortly prior to the VTE event. Thus, the incidence of "idiopathic VTE" observed in this study can be estimated to be about 5 per 10,000 WY. This is in agreement with other reported incidence rates in the general population [1, 7]. The incidence estimates for definite VTE ranges between 1 to 6 per 104 WY in non-users or oral contraceptives (OC) and 2 to 10 per 104 WYs in OC users. Older studies depicted almost-always higher incidence rates than more recently performed studies (see overview in ). A recent systematic review  came to a pooled incidence of definite VTE for the general population of 5 per 10,000 person years, similar in males and females, and found that around 40% of VTE cases were "idiopathic". We conclude that our data can be generalized for the female population of this region in the fertile age range. This conclusion is supported by results of a prospective, community-based cohort study  that found a VTE incidence rate of 2.7 "primary VTE cases" per 104 person-years (equal to idiopathic: no previous VTE history, no cancer, no surgery or trauma), however, in males aged 40–49 years.
The absolute risks (incidence rates) varied markedly among those exposed or non-exposed by genetic and acquired VTE risk factors in our cohort study (see table 3). The relatively small numbers of women exposed to genetic VTE markers (see also table 2) should be taken into account before drawing conclusions. Thus, we have to be worry about statistically robust results in most of the sub-groups.
The crude, not adjusted comparison between "exposed" (= risk factor present) and "non-exposed" (= risk factor absent) showed incidence ratios (OR) ranging from 0.6 (AT deficiency) to 13.5 (personal history of VTE). The highest VTE incidence rate was found for women with a history of previous VTE (125.8 VTE cases per 104 WY – compared with 9.3 in women without VTE history), i.e., a 13.5fold increased incidence rate ratio (see table 3). Other significant incidence rate ratios were observed for higher age, elevated BMI, family history of VTE (and for varicose veins).
After fully controlling for all other risk factors (logistic regression) only age, personal and family history of VTE remained significant risk factors (see table 4). A significant role of the family history of VTE has been reported in several studies [15–17]. It is surprising that in our study the aggregate variable "family history" was more important than the individual genetic or related markers.
The impact of individual genetic markers on VTE risk (FVL, PTM, MTHFR) was not statistically significant in the logistic regression analysis (table 4) – possibly also due to the small number of incident cases. This is also true for Protein C and AT deficiency, i.e. when using the lower 5th percentile as usual definition for deficiency . Even FVL mutation did not show significantly increased VTE risk, although the point estimate (OR = 2.0) – at face-value – is compatible with the estimate of another, but much larger cohort study , but lower than reported from several case-control studies. This difference between risk estimates for FVL in cohort and case-control studies is probably due to methodological reasons .
A similarly weak impact of genetic markers was also observed in follow-up studies of selected groups: Carriers of genetic polymorphisms have been followed prospectively and found a low absolute annual incidence of VTE [10, 12, 19]. Another prospective cohort observed a low incidence of VTE in otherwise healthy thrombophilic children  or asymptomatic family members who are carriers of factor V Leiden [11, 12] or other family studies . Deficiency of AT and PC activity had also no significant impact on thrombotic risk. We cannot exclude however that a considerable part of such unexpected results for PC and AT activity may be related to pre-analytical conditions in our field study. From a laboratory perspective in a few women polymorphisms leading to decreased levels of analytical results but not to clinical manifest thrombophilia are quite possible. An explanation might be that silent mutations and polymorphisms cause reduced activities in the laboratory assays not reflecting a potential clinical problem [22, 23]. Without a positive personal or family history of VTE such results should at least be confirmed with additional laboratory tests and family examinations before informing the patients of a potential thrombophilic diathesis or before recommending respective medical prophylaxis.
There is an increasing debate about the role of genetic factors in the prediction of future VTEs and thus sustaining a controversy about genetic screening. This issue is a current controversy in the literature [24, 25]. Clinical reports often suggest a high VTE recurrence rate in patients with previous VTE , but we found this phenomenon only in 4 of our 34 incident VTE cases. A recently published community-based cohort study of FVL carriers & non-carriers observed no significant difference on VTE incidence between both groups, except for women ≥ 60 years of age .
In general, our results support the notion that genetic parameters alone are relatively weak long-term risk factors; the occurrence of VTE requires interaction of both inherited and acquired risk factors or gene-gene-interactions .
The clinical VTE risk factors with significantly elevated incidence rate ratios such as advanced age and elevated BMI, but also history of previous VTE, family history of VTE or family history of varicose veins are not new. Most physicians are aware of these risk factors and consider them in practice. No incidence difference was also found for ever smoking which is rarely considered as risk factor for VTE. Hypertension was not analyzed. No significant prognostic impact was found for ever use of hormonal treatment/contraception at baseline. This is plausible because sex hormones effects are not general characteristics but short-term acting factors, which need another statistical approach, and was not the aim of this study.
The influence of more acutely acting risk factors – such as immobilization, surgery, long-haul flights, and use of drugs (e.g. OCs) was intentionally excluded from this analysis, although interactions between hereditary factors and acute, environmental pattern are known [21, 29, 30]. Only parameters that were available at baseline and likely to affect the long-term development were eligible for this analysis. Other influential risk factors or preventive measures have to be considered when discussing activities to reduce a predicted increased risk in the medical practice. It was not the aim of the study and data are neither available to test the effect of preventive measures nor the effect of additional risk factors in the immediate period before the event occurred. This would require another study design and a separate study with sufficient power for such questions.
It is a limitation of this long-term cohort study, however, that the number of incident, confirmed (definitive and probable) VTE cases was still small in absolute numbers (n = 34) in this cohort observation period of 32,508 years of observation. The low incidence can be explained by the young average age (26 years at entry). Thus, incidence & and risk estimates have wide confidence intervals and conclusions are limited. Rare combinations of risk markers have not yet materialized in one single VTE case. This makes it difficult to further divide into cases that occurred in presumably exposed or unexposed sub-groups. This is particularly true if the potential risk factors (exposure) are infrequent, as it is for genetic markers. It is a problem of the study design that no attempt was made to confirm of deficiencies (second blood sample) and no family study was planned to assess inheritance of deficiencies such as AT or PC.
In case of FVL mutation the adjusted risk is apparently increased about 2fold (non-significant) but only 4 cases had a positive test of FVL mutation (for prothrombin mutation only 2 cases). If one then adjusts for 14 other potential factors, there is obviously a statistical problem: Resulting risk estimates might be unstable, i.e. drifted in either direction. Even if one focuses on the crude OR, which is similar to the adjusted estimate in this case, careful interpretation is warranted. It cannot be excluded that the "true risk" of the mutation markers is 2fold increased, although the risk estimates do not favor of such a conclusion. Insofar, future analyses will benefit from an improved point of departure (longer observation, more cases).