Thrombosis Research 133 (2014) 762–767
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
Risk factors for recurrent venous thromboembolism in young and middle-aged women Maria Ljungqvist a,⁎, Kristina Sonnevi a, Annica Bergendal b, Margareta Holmström c, Helle Kieler b, Gerd Lärfars a a b c
Department of Clinical Science and Education, Internal medicine, Södersjukhuset, Karolinska Institutet, Stockholm, Sweden Centre for Pharmacoepidemiology, Department of Medicine, Solna, Karolinska Institutet, Stockholm, Sweden Coagulation Unit, Hematology Centre Karolinska, Department of Medicine Solna Karolinska Institutet, Stockholm, Sweden
a r t i c l e
i n f o
Article history: Received 28 November 2013 Received in revised form 15 January 2014 Accepted 19 February 2014 Available online 26 February 2014 Keywords: Venous Thromboembolism Recurrence Risk Factors Estrogens Women
a b s t r a c t Background: It is a matter of debate whether women with an episode of VTE associated with estrogen have a lower risk of recurrence than women with an unprovoked VTE. Objectives: To identify risk factors for recurrent VTE in women and to assess the risk of recurrent VTE associated with combined oral contraceptives (CHC) or menopausal hormone treatment (HT), compared to surgery-related and unprovoked VTE. Patients/Methods: A cohort of 974 women aged 18–64 years with a first episode of VTE were followed-up for a median time of 5.2 years. All women were previously included as cases in the Swedish nation-wide case–control study “Thrombo Embolism Hormone Study” (TEHS). Hazard ratios for recurrence were calculated using univariable and multivariable Cox proportional hazards model. Results: A total of 102 patients (10%) suffered from recurrent VTE. The annual rate of recurrence was 1.0% in patients with surgery/cast, 2.0% in patients with CHC/HT and 3.2% in patients with unprovoked first VTE. Adjusted hazards ratio (HRa) for recurrence was 0.35 (95% CI 0.20-0.61) in women with VT provoked by surgery/cast while women with estrogen-associated VTE had a HRa of 0.70 (95% CI 0.43-1.20) compared to women with unprovoked VTE. Conclusion: Women 18–64 years are at low risk of recurrent VTE. Women with hormone associated VTE had a lower risk of recurrence than women with unprovoked VTE, but not as low as surgery/cast provoked VTE. © 2014 Elsevier Ltd. All rights reserved.
Introduction After a first episode of venous thromboembolism (VTE), there is a considerable risk of recurrence. The risk is highest within the first year after discontinuation of anticoagulant treatment with a cumulative incidence of 20-25% after 5 years in an unselected population [1,2]. Continuous anticoagulant treatment can prevent most episodes of recurrent VTE but decisions on extended treatment must be balanced against the risk of major bleeding complications [3–5]. The risk of recurrence after an unprovoked VTE is higher than if the VTE was provoked by a transient risk factor like surgery, cast or
Abbreviations: VTE, venous thromboembolism; CHC, combined hormonal contraceptives; HT, menopausal hormone treatment; DVT, deep vein thrombosis; PE, pulmonary embolism; TEHS, Thrombo Embolism Hormone Study; ICD, International Classification of Diseases; BMI, body mass index; HR, hazard ratios; CI, confidence intervals. ⁎ Corresponding author at: Department of Internal Medicine, Södersjukhuset, SE- 118 81 Stockholm, Sweden. Tel.: +46 8 6163462; fax: +46 8 6163115. E-mail address: [email protected] (M. Ljungqvist).
http://dx.doi.org/10.1016/j.thromres.2014.02.017 0049-3848/© 2014 Elsevier Ltd. All rights reserved.
pregnancy [2,6,7]. The risk of recurrent episodes of VTE is also influenced by gender as several studies have shown that men have at least a two-fold higher risk of recurrence than women [7–12]. A lower risk of recurrent VTE in women where the first episode was associated with estrogen treatment has been suggested as a possible explanation for this difference. Data on this matter is however still conflicting and it is debated whether hormonal treatment with combined oral contraceptives (CHC) or menopausal hormone treatment (HT) should be regarded as a provoking factor or not [9,13–17]. The current international guidelines recommend extended anticoagulant treatment in patients with a high risk of recurrence, such as after unprovoked proximal deep vein thrombosis (DVT) or unprovoked pulmonary embolism (PE) [18]. As women have a lower risk of recurrence than men there is a need to better understand the risk factors for recurrence in female patients, including the risk related to hormone associated VTE. The overall aim of this study was to assess the risk of recurrent VTE in women with hormone related thrombosis, VTE related to surgery or cast and unprovoked VTE. Furthermore, we wanted to identify the
M. Ljungqvist et al. / Thrombosis Research 133 (2014) 762–767
763
most important risk factors for recurrent VTE in young and middle aged women. Materials and Methods Population and Study Design This cohort study called TEHS-follow-up comprised women 18 to 64 years of age. All participants were previously included as cases in the “Thrombo Embolism Hormone Study” (TEHS), a population based case–control study on risk factors for VTE that recruited patients and controls from 43 larger or medium sized hospitals spread geographically in Sweden. In the TEHS- follow up study all cases in TEHS who accepted participation were followed-up concerning recurrent VTE. TEHS The methodology of the TEHS has been described elsewhere [19]. Briefly, 1433 women with a first episode of PE or DVT of the lower limb or pelvis were included as cases, from 2003 to 2009. The diagnosis of VTE was based on venography or Doppler ultrasound in DVT and CTscan or perfusion-ventilation scintigraphy in PE. Patients, who had been pregnant the past three months, had a current malignant disease or a history of malignancy within the past five years were not eligible for the TEHS. Baseline data was obtained through a telephone interview with the participants in a period of three months after the diagnosis of VTE and included information on body weight and height, risk factors for VTE, medical history, use of combined hormonal contraceptives (CHC) or menopausal hormone treatment (HT) and family history of VTE. Blood samples for DNA-analyses were collected from the women at time of inclusion. TEHS-follow-up In 2011, all cases included in TEHS who were still living in Sweden were followed-up by a questionnaire sent by mail. The questionnaire covered different medical aspects, as current medication, duration of previous and current anticoagulant treatment, hormone therapy with CHC or HT and recurrent events of VTE. Within one month a reminder was sent if the questionnaire was not returned. Out of the 1435 women included in the TEHS-study, 38 were not alive at the time of follow-up. The cause of death was retrieved from Cause of Death Register in Sweden, held by the National Board of Health and Welfare [20] and none was related to VTE. Thirteen women died from cancer, 11 from cardiovascular disease and the remaining 14 from other causes. Three patients had emigrated from Sweden leaving 1394 patients who were asked to participate in the follow-up study. Women on continuous anticoagulant treatment were not eligible for the analysis of recurrent VTE in the follow-up since they were not considered to be at risk of recurrence. A total of 345 patients rejected participation, 53 were on continuous anticoagulant treatment and 22 patients did not answer the question about current anticoagulant treatment, leaving 974 women to follow-up, Fig. 1. We obtained information on recurrent VTE both from the questionnaire and from data recorded in the Swedish Patient Register, a nationwide register held by the National Board of Health and Welfare in Sweden. From 1987 the Patient Register includes all hospital-based care in Sweden and from 2001 it also includes outpatient visits. The register cover data on age, sex, national registration number (a unique identification number assigned to every resident in Sweden) and up to 6 diagnoses and 6 surgical procedures including for each patient. Diagnoses are recorded according the current International Classification of Diseases (ICD) version 10 (ICD-10) at each out-clinic visit or hospitalization [21]. To identify potential recurrent events all study participants recorded in the register with diagnose of venous thrombosis according to ICD 10
Fig. 1. Flow-chart of the study cohort.
(I80.1-9, I82.1-9, I26.0, 9, I67.6, O22.3, O22.5) after their initial VTE were reviewed. All events, recorded in the register, were objectively verified by a review of the medical records. To be considered as a recurrent event the diagnosis was required to be confirmed by any of the same objective radiological methods as at inclusion of the TEHS-study. Furthermore, the event had to be regarded, by the treating physician, as having indication for resumed anticoagulant treatment for at least 3 months. The follow-up period started at the time of discontinuation of anticoagulant treatment after the first episode of VTE and patients were censored either at first recurrent VTE or at the date for answering the questionnaire. If information on the duration of anticoagulant treatment was missing, women with an index DVT were assumed to be treated with anticoagulants for 6 months and women with an index PE for 12 months according to the clinical praxis during the time of inclusion in Sweden. Both the TEHS and the TEHS follow-up study were approved by the research ethics committees in Sweden. All participants in TEHS and the TEHS follow-up study were included after a signed informed consent at inclusion of the study. Statistical Analyses Cumulative incidence of recurrent VTE was calculated using KaplanMeier survival analysis. All study participants were included with their maximum time available for follow-up and censored at time of recurrent event or end of follow-up. The localization of the VTE was classified as either distal DVT when localized below the popliteal vein, proximal DVT from the popliteal vein and more proximal including pelvic veins, or as PE. Patients with both DVT and PE were classified as having PE. To calculate the risk of recurrence according to characteristics of first VTE we classified patients in three groups according to provoking factor at the first VTE: unprovoked, provoked by surgery and or treatment with cast or provoked by HT/CHC treatment. CHC included oral tablets, dermal patches and intravaginal devices. HT included progestogen opposed and non-opposed oral formulations and dermal patches. Women with both surgery and HT/CHC treatment prior to the first episode of VTE were classified into the surgery/cast group as surgery/ cast has been shown to be a stronger risk factor than HT/CHC[19]. Information on weight and length were collected from time of inclusion in TEHS. Obesity was defined as body mass index (BMI) ≥ 30 kg/m2. Groups were compared using log-rank test. Cox proportional hazard model was used to calculate both crude and multivariable adjusted hazard ratios (HR) with their corresponding 95% confidence intervals (CI). The multivariable model included age, presence of factor V
764
M. Ljungqvist et al. / Thrombosis Research 133 (2014) 762–767
Leiden(G1691A) and/or prothrombin G20210A mutation (yes/no), BMI, location and extent of initial VTE (PE/proximal DVT or distal DVT) and whether the first event was provoked by surgery/cast, CHC/HT or unprovoked. For the analyses concerning recurrence by initial exposure the women with an unprovoked VTE constituted the comparison group. To estimate the HR for recurrence in women with a first VTE associated with CHC or HT, we analyzed women over and below the age of 50 years separately. The same classification as previously described was used to compare risk factors. Women 50 years or older who were using CHC at time of first VTE (n = 5) and women below 50 years using HT (n = 11) were excluded from these analyses, as were patients with missing data on use of CHC/HT at first VTE (n = 7). A sensitivity analysis was performed without exclusion of the mentioned patients without any major impact on the risk estimates. Two-sided P values b 0.05 were considered as indicating statistical significance. All statistical calculations were performed with the use of SPSS software, version 20.0 (SPSS Inc., Chicago, IL, USA). Results Characteristics of the 974 included women are shown in Table 1. The median age of women with VTE at inclusion in TEHS was 49 years (range 18–64). At the time of the first VTE 386 of all (40%) women were treated with hormones; 245 with CHC and the remaining 141 with HT. Of the CHC-users 48% used second generation progestogen (norethisterone, levonorgestrel) in combinations with estrogen, 45% used third generation progestogen (desogrestel, norgestimate, drospirenon) in combinations with estrogen and 7% used other compounds. A total of 230 patients (25%) were carriers of the FV Leiden mutation and 50 (5%) were carriers of the prothrombin G20210A mutation. In 350 women (36%) the VTE was provoked by surgery or cast and 224 of all included women (23%) were obese. Of all patients, 60% had a proximal DVT or PE as a first even of VTE. Risk Factors for Recurrence During a median follow-up of 5.2 years (range 0.1-9.1), 102 of the patients (10%) suffered from a recurrent episode of VTE (44 PE, 57 DVT). One patient had a cerebral vein thrombosis as recurrent VTE. The overall annual rate of recurrence was 2.0%. Recurrent VTE was diagnosed in 17 patients (4.9%) with a primary VTE provoked by surgery/ cast, 27 patients (10.0%) using CHC or HT and in 57 of the patients (16%) with a primary unprovoked VTE. This corresponds to a mean annual rate of recurrence of 1.0% in patients with surgery/cast, 2.0% in patients with CHC/HT and 3.2% in patients with unprovoked VTE. The follow-up consisted of 974 women corresponding to a total of 4.941 Table 1 Baseline characteristics of patients. All patients N = 974
Patients b 50 yrs N = 510
Patients ≥ 50 yrs N = 464
Age, median (years) CHC (%) HT (%) Missing data on CHC/HT (%) FV Leiden FII GA20210A Provoked by surgery/cast
49 (SD 13) 245 (25,2) 141 (14.5) 6 (0.72) 230 (23.6) 50 (5,1) 350 (35.9)
36 (SD 9) 240 (47.1) 11 (2.2) 2 (0.59) 132 (25.9) 27 (5.3) 181 (35.5)
58 (SD 4) 5 (1.1) 130 (28.0) 4 (0.86) 98 (21.1) 23 (5.0) 169 (36.4)
Site of index thrombosis Distal DVT Proximal DVT PE BMI (SD) BMI b 30 (%) BMI ≥ 30 (%)
389 (39.9) 293 (30.1) 292 (30.0) 26 (SD 5) 750 (77.0) 224 (23.0)
207 (40.6) 140 (27.5) 163 (32.0) 25 (SD 6) 401 (78.6) 109 (22.4)
182 (39.2) 153 (33.0) 129 (27.8) 27 (SD 5) 349 (75.2) 115 (24.8)
CHC = Combined hormonal contraceptives, HT = Menopausal hormone therapy, FII G20210A = Prothrombin G20210A mutation.
person-years and the cumulative incidence of recurrence according to provoking factor of first VTE is shown in Table 2. All patients had the highest rate of recurrence during the first year after cessation of anticoagulant treatment The crude HR for recurrence was 0.28 (95% CI 0.16-0.47) in women with a first VTE provoked by surgery/cast and 0.57 (95% CI 0.36-0.90) in women with a first VTE associated with hormone treatment, when compared to women with unprovoked VTE. In the multivariable models the risk estimates increased slightly and remained statistically significant for women with VTE provoked by surgery/cast but not for women having hormone treatment prior to their first VTE, Table 3. A plot of the cumulative incidence of recurrent VTE from multivariable Cox regression analysis in women with a first VTE provoked by HT/CHC, surgery/cast and unprovoked VTE is shown in Fig. 2. Obesity was correlated with an increased risk of recurrence with HR 1.8 (95% CI 1.2-2.7) compared to non-obese women. The increased risk persisted in multivariable adjustments. Carrier ship for factor V Leiden (G1691A) and/or the prothrombin G20210A was not significantly associated with recurrence and neither was a positive family history, Table 3. Age was not associated with the risk of recurrent events (data not shown). The localization of VTE was associated with the risk of recurrence. Patients with proximal DVT or PE had higher risk compared to distal DVT (HR 2.0, 95% CI 1.3-3.1), with consistent risk estimates in the multivariable analysis (HRa 2.0, 95% CI 1.3-3.2), Table 2. The initial localization of VTE was also predictive of the site of recurrent VTE. In patients with initial PE 68% of the recurrent events consisted of PE. Corresponding numbers for patients with proximal DVT and distal DVT was 51% and 63% respectively. About 1/3 of the patients with initial DVT had a recurrent event in form of PE (8 with initial distal DVT and 9 with initial proximal DVT). Risk Factors for Recurrence in Women b 50 years Hazard ratios and there 95% confidence intervals in women younger than 50 years of age at time of first VTE are shown in Table 4. Women with VTE provoked by surgery/cast had the lowest risk of recurrence. Obesity was associated with an increased risk of recurrent VTE with HRa 2.5 (95% CI 1.3-4.7) compared to non-obese patients in this agegroup. The presence of factor V Leiden and/or prothrombin G20210A mutation, a positive family history or age had no major impact on the risk estimate. Risk Factors for Recurrence in Women ≥ 50 years Table 5 show hazard ratios and there 95% confidence intervals in women ≥ 50 years at time of index VTE. Women with VTE provoked by surgery/cast had an even lower risk of recurrence, HRa 0.29 (95% CI 0.14-0.66) in this age-group. As in women b 50 years, the presence of factor V Leiden and/or prothrombin G20210A mutation, a positive family history or age had no major impact on the risk estimates. Obesity
Table 2 Cumulative incidence of recurrent VTE according to risk factor at first episode of VTE.
1 2 3 4 5 6
year years years years years years
All women N = 974
Women with Surgery/Cast N = 350
Women with CHC/HT N =272
Women with Unprovoked VTE N = 352
4% 6% 8% 9% 9% 11%
2% 3% 3% 3% 4% 6%
5% 7% 7% 8% 9% 10%
7% 10% 13% 15% 15% 18%
CHC/HT = treatment with either combined oral contraceptives or menopausal hormone therapy.
M. Ljungqvist et al. / Thrombosis Research 133 (2014) 762–767 Table 3 Risk for recurrence of venous thromboembolism in 974 women presented as HRs with 95% CI.
Unprovoked Surgery/cast CHC/HT BMI ≥ 30 Genetic factor V/II Family history PE/Proximal DVT
Recurrent events/person years
HR (95% CI)
HRa (95% CI)
57/1765 17/1776 27/1376 35/1137 33/1340 32/1229 75/2950
Reference 0.3 (0.2-0.5) 0.6 (0.4-0.9) 1.8 (1.2-2.7) 1.4 (0.9-2.1) 1.4 (0.9-2.2) 2.0 (1.3-3.1)
Reference 0.4 (0.2-0.7) 1 0.7 (0.4-1.1) 1 1.6 (1.0-2.4) 2 1.3 (0.9-2.0) 3 1.2 (0.8-1.8) 3 2.0 (1.3-3.2) 4
Adjusted for: 1Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, 2. Provoking factor, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, age 3Provoking factor, BMI, location of index VTE, age 4Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, provoking factor.
did not correlate with an increased risk of recurrent VTE in this agegroup. Discussion In this large Swedish long-term follow-up study of women 18–64 years with VTE we found that the overall risk of recurrent VTE was low. The risk after a first event of VTE associated with CHC or HT was lower than in cases with unprovoked VTE but higher than in VTE associated with surgery or cast treatment. The localization of the primary event and obesity were the two most important clinical risk factors for recurrent VTE in our study population. Several studies have reported lower risk of recurrent VTE in women who are using CHC or HT at the time of first VTE, compared to unprovoked VTE [7,11,13,22]. Only few studies included patients with provoked VTE and none compared the risk of recurrence in hormone associated VTE to surgery-provoked VTE [11,23]. The PREVENT-study found lower risk estimates of recurrence in women with CHC or HT associated VTE although the difference was not statistically significant [13]. Furthermore, in a study of 361 women less than 45 years of age in Austria, the use of CHC at first VTE had no impact on the risk of recurrence, HR 0.85 (95% CI 0.6-1.21) [16]. In the REVERSE-study, the risk of recurrence was not significantly lower than in unprovoked cases, and
765
Table 4 Risk for recurrence of venous thromboembolism in 503 women b 50 years presented as HRs with 95% CI.
Unprovoked Surgery/cast CHC BMI ≥ 30 PE/Proximal DVT
Recurrent events/person-years
HR
HRa
22/762 9/878 12/828 18/543 31/1.482
Reference 0.3 (0.2-0.7) 0.7 (0.3-1.2) 2.4 (1.3-4.3) 1.74 (0.9-3.3)
Reference 0.5 (0.2-0.9) 1 0.9 (0.2-1.1) 1 2.4 (1.3-4.7) 2 1.33 (0.7-2.6) 3
Adjusted for: 1Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, 2Provoking factor, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, age 3Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, provoking factor.
the authors suggest that hormone associated VTE cannot be regarded as provoked VTE [14]. Our large study on female VTE patients allows us to compare the risk of recurrence in women with CHC/HT to both unprovoked and surgery/ cast-provoked cases. Interestingly, our data suggest that the risk of CHC/HT associated VTE is not the same as either unprovoked or surgery-provoked VTE, but somewhere in-between. Furthermore, the clinical data available in this study allowed for multivariate adjustments to control for potential confounding by age, BMI and thrombophilia. This was not possible in previous large studies due to lack of data concerning BMI and thrombophilia[11]. The present study also allowed for subgroup analysis on women over and below 50 years to look at HT and CHC separately with consistent results concerning the risk of recurrence related to a first VTE in association with CHC/HT. We found that use of CHC or HT was associated with reduced risk of recurrence compared to unprovoked VTE in the crude analysis of the whole study population. This difference was no longer statistically significant after multivariate adjustments for BMI, localization of index VTE, age and thrombophilia which was probably due to lack of power. Nevertheless we believe that the true risk estimate is lower than in unprovoked VTE which is also indicated by the HR. In line with previously published data we found that obesity was significantly associated with increased risk of recurrence [24]. When we stratified the analysis to women over and under 50 years we noticed that this correlation was only valid in women under 50 years of age. In a previous study, Olie´ et al. showed that obesity was a risk factor for recurrence in women only and in their analysis of 349 women and 234 men the mean age of the women was 46 years and men 54 years [10]. Others, who have not found any correlation between obesity and recurrent VTE in women, have usually studied an older patient group [22] which suggests that obesity might be a more important risk factor in young women than in other populations. In a study of women under 45 years of age, BMI N 25 kg/m2 was the only clinical characteristic correlated with increased risk of recurrence [16]. As previously reported by others, the presence of any of the two most common genetic risk factors Factor V Leiden(G1691A) or Prothrombin G20210A mutation, did not affect the risk estimates of
Table 5 Risk for recurrence of venous thromboembolism in 449 women ≥ 50 years presented as HRs with 95% CI.
Unprovoked Surgery/cast HT BMI ≥ 30 PE/Proximal DVT Fig. 2. Comulative incidence of VTE based on multivariable Cox proportional Hazards model according to provoking factor at first VTE. Adjustments for BMI, localization of index VTE and carrier ship of Factor V Leiden/FII GA20210A.
Recurrent events/person-years
HR (95% CI)
HRa (95% CI)
35/909 8/860 10/537 16/561 41/1336
Reference 0.3 (0.1-0.5) 0.5 (0.3-1.1) 1.4 (0.8-2.4) 2.4 (1.3-4.6)
Reference 0.3 (0.1-0.7) 1 0.5 (0.2-1.0) 1 1.1 (0.6-2.0) 2 2.7 (1.3-5.5) 3
Adjusted for: 1Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, 2Provoking factor, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, age 3Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, provoking factor.
766
M. Ljungqvist et al. / Thrombosis Research 133 (2014) 762–767
recurrent VTE in any major way [23,25]. Neither did a positive family history of VTE, which has been shown to be an independent risk factor of first VTE [26,27]. This paradox could be explained by index event bias where carrier ship of thrombophilia at first event of VTE is associated with lower occurrence of other risk factors [28]. The risk of recurrence in patients with primary distal DVT was, as expected and previously reported, significantly lower than in proximal DVT or PE [2,5,29]. However 26 of 389 patients with initial distal DVT suffered from recurrence and 8 of these events were in form of PE. The strength of this study is its size; it is to our knowledge the largest follow up study of a defined study cohort of women with clinical data concerning the risk of recurrent VTE after an initial VTE in association with CHC or HT. The use of both the Swedish Patient Register and the questionnaires specifically designed for this study ensured that all recurrent events were identified. The thorough review of medical charts in patients with suspected recurrent events reassured that all recurrences were clinically relevant. A few limitations of this study must however be emphasized. First, patients with continuous anticoagulant treatment after the first episode of VTE were not included in the analysis. Patients being considered at high risk of recurrence already at the first VTE were therefore not included in our study, something that is hard to avoid without exposing patients to unnecessary risk. It must however be held in mind when interpreting the results. Data on hormonal treatment during follow up was assessed for all patients and 22 women resumed treatment with HT (n = 19) or CHC (n = 3) after first VTE. To avoid immortal time bias we didn´t exclude these patients in the follow-up. The low number of recurrent events in our study population resulted in some estimates of recurrence that were not statistically significant on the p b 0.05 level. The estimated risk is however likely to be close to the actual risk. The results of this study may have clinical implications. First of all, our study confirms that young and middle aged women have a low risk of recurrent VTE. The current guidelines advocate extended anticoagulant treatment with annual reassessment in unprovoked VTE, regardless of the gender of the patient [18]. The rate of major bleeding during anticoagulant treatment is 3-5% per year. Hence, in patients with a mean annual rate of recurrent VTE less than 5%, extended anticoagulant treatment could lead to higher risk of bleeding complications than favourable effect of protecting against VTE recurrence. Since all women in this study had an annual rate of VTE of less than 5%, and even women with unprovoked VTE had as low annual rate of recurrence as 3.2%, our results suggest that most women in this age group would not benefit from extended therapy. Our results also suggest that obesity is a risk factor for recurrent VTE in women below the age of 50 years, but not for women between 50 and 64. This needs to be evaluated in further studies. The overall risk of recurrence in this group is low but better tools to identify the women at increased risk are still warranted, particularly for women with unprovoked VTE. Several clinical prediction scores with data on d-dimer, and in some cases obesity, have been published [15,22,29]. The future challenge is to validate clinical scores in different subpopulations, including young and middle aged women. As different risk factors seems to be of varying importance in men and women, and also in different age groups, scoring systems might need to be adjusted according to the age and gender of the patient. In conclusion, we found that women 18–64 years have a low risk of recurrence in VTE. Women with hormone associated VTE had a lower risk of recurrence than women with unprovoked VTE, but not as low as surgery/cast provoked VTE. In women below 50 years, obesity was correlated with increased risk of recurrence, suggesting that information on BMI should be included in of the evaluation of risk in younger women. Conflict of Interest Statement No conflict of interest.
Acknowledgements We like to thank Jacob Odeberg for the genotyping in TEHS-study and Hans Pettersson at Karolinska Institutet, Södersjukhuset for helpful discussions on statistical analyses. This study was funded by Janssen-Cilag,Novartis,Organon, Schering, Wyeth, AFA, Center for Gender Medicine KI and by the authors’ affiliations. The sponsors were not involved in the design and conduct of the study; collection, management, analysis, nor interpretation of the data; and preparation, review, nor approval of the manuscript.
References [1] Prandoni P, Lensing AW, Cogo A, Cuppini S, Villalta S, Carta M, et al. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med 1996;125:1–7. [2] Hansson PO, Sorbo J, Eriksson H. Recurrent venous thromboembolism after deep vein thrombosis: incidence and risk factors. Arch Intern Med 2000;160:769–74. [3] Kearon C, Gent M, Hirsh J, Weitz J, Kovacs MJ, Anderson DR, et al. A comparison of three months of anticoagulation with extended anticoagulation for a first episode of idiopathic venous thromboembolism. N Engl J Med 1999;340:901–7. [4] Linkins LA, Choi PT, Douketis JD. Clinical impact of bleeding in patients taking oral anticoagulant therapy for venous thromboembolism: a meta-analysis. Ann Intern Med 2003;139:893–900. [5] Schulman S, Rhedin AS, Lindmarker P, Carlsson A, Larfars G, Nicol P, et al. A comparison of six weeks with six months of oral anticoagulant therapy after a first episode of venous thromboembolism. Duration of Anticoagulation Trial Study Group. N Engl J Med 1995;332:1661–5. [6] Schulman S, Lindmarker P, Holmstrom M, Larfars G, Carlsson A, Nicol P, et al. Postthrombotic syndrome, recurrence, and death 10 years after the first episode of venous thromboembolism treated with warfarin for 6 weeks or 6 months. J Thromb Haemost 2006;4:734–42. [7] Baglin T, Luddington R, Brown K, Baglin C. Incidence of recurrent venous thromboembolism in relation to clinical and thrombophilic risk factors: prospective cohort study. Lancet 2003;362:523–6. [8] Kyrle PA, Rosendaal FR, Eichinger S. Risk assessment for recurrent venous thrombosis. Lancet 2010;376:2032–9. [9] Kyrle PA, Minar E, Bialonczyk C, Hirschl M, Weltermann A, Eichinger S. The risk of recurrent venous thromboembolism in men and women. N Engl J Med 2004;350:2558–63. [10] Olie V, Zhu T, Martinez I, Scarabin PY, Emmerich J. Sex-specific risk factors for recurrent venous thromboembolism. Thromb Res 2012;130:16–20. [11] Douketis J, Tosetto A, Marcucci M, Baglin T, Cosmi B, Cushman M, et al. Risk of recurrence after venous thromboembolism in men and women: patient level meta-analysis. BMJ 2011;342:d813. [12] McRae S, Tran H, Schulman S, Ginsberg J, Kearon C. Effect of patient's sex on risk of recurrent venous thromboembolism: a meta-analysis. Lancet 2006;368:371–8. [13] Cushman M, Glynn RJ, Goldhaber SZ, Moll S, Bauer KA, Deitcher S, et al. Hormonal factors and risk of recurrent venous thrombosis: the prevention of recurrent venous thromboembolism trial. J Thromb Haemost 2006;4:2199–203. [14] Le Gal G, Kovacs MJ, Carrier M, Do K, Kahn SR, Wells PS, et al. Risk of recurrent venous thromboembolism after a first oestrogen-associated episode. Data from the REVERSE cohort study. Thromb Haemost 2010;104:498–503. [15] Rodger MA, Kahn SR, Wells PS, Anderson DA, Chagnon I, Le Gal G, et al. Identifying unprovoked thromboembolism patients at low risk for recurrence who can discontinue anticoagulant therapy. CMAJ 2008;179:417–26. [16] Laczkovics C, Grafenhofer H, Kaider A, Quehenberger P, Simanek R, Mannhalter C, et al. Risk of recurrence after a first venous thromboembolic event in young women. Haematologica 2007;92:1201–7. [17] Le Moigne E, Delluc A, Tromeur C, Nowak E, Mottier D, Lacut K, et al. Risk of recurrent venous thromboembolism among young women after a first event while exposed to combined oral contraception versus not exposed to: a cohort study. Thromb Res 2013;132:51–5. [18] Kearon C, Akl EA, Comerota AJ, Prandoni P, Bounameaux H, Goldhaber SZ, et al. Antithrombotic therapy for VTE disease: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e419S–94S. [19] Bergendal A, Bremme K, Hedenmalm K, Larfars G, Odeberg J, Persson I, et al. Risk factors for venous thromboembolism in pre-and postmenopausal women. Thromb Res 2012;130:596–601. [20] Socialstyrelsen [National Board of Health and Welfare]. Cause of Death Register [Dödsorsakregistret]. http://www.socialstyrelsen.se/register/dodsorsaksregistret. [21] Socialstyrelsen [National Board of Health and Welfare]. National Inpatient Register [Patientregistret]. http://www.socialstyrelsen.se/register/halsodataregister/ patientregistret. [22] Tosetto A, Iorio A, Marcucci M, Baglin T, Cushman M, Eichinger S, et al. Predicting disease recurrence in patients with previous unprovoked venous thromboembolism: a proposed prediction score (DASH). J Thromb Haemost 2012;10:1019–25. [23] Marchiori A, Mosena L, Prins MH, Prandoni P. The risk of recurrent venous thromboembolism among heterozygous carriers of factor V Leiden or prothrombin G20210A mutation. A systematic review of prospective studies. Haematologica 2007;92:1107–14.
M. Ljungqvist et al. / Thrombosis Research 133 (2014) 762–767 [24] Eichinger S, Hron G, Bialonczyk C, Hirschl M, Minar E, Wagner O, et al. Overweight, obesity, and the risk of recurrent venous thromboembolism. Arch Intern Med 2008;168:1678–83. [25] Ho WK, Hankey GJ, Quinlan DJ, Eikelboom JW. Risk of recurrent venous thromboembolism in patients with common thrombophilia: a systematic review. Arch Intern Med 2006;166:729–36. [26] Bezemer ID, van der Meer FJ, Eikenboom JC, Rosendaal FR, Doggen CJ. The value of family history as a risk indicator for venous thrombosis. Arch Intern Med 2009;169:610–5.
767
[27] Sonnevi K, Bergendal A, Adami J, Larfars G, Kieler H. Self-reported family history in estimating the risk of hormone, surgery and cast related VTE in women. Thromb Res 2013;132:164–9. [28] Dahabreh IJ, Kent DM. Index event bias as an explanation for the paradoxes of recurrence risk research. JAMA 2011;305:822–3. [29] Eichinger S, Heinze G, Jandeck LM, Kyrle PA. Risk assessment of recurrence in patients with unprovoked deep vein thrombosis or pulmonary embolism: the Vienna prediction model. Circulation 2010;121:1630–6.
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
Risk factors for recurrent venous thromboembolism in young and middle-aged women Maria Ljungqvist a,⁎, Kristina Sonnevi a, Annica Bergendal b, Margareta Holmström c, Helle Kieler b, Gerd Lärfars a a b c
Department of Clinical Science and Education, Internal medicine, Södersjukhuset, Karolinska Institutet, Stockholm, Sweden Centre for Pharmacoepidemiology, Department of Medicine, Solna, Karolinska Institutet, Stockholm, Sweden Coagulation Unit, Hematology Centre Karolinska, Department of Medicine Solna Karolinska Institutet, Stockholm, Sweden
a r t i c l e
i n f o
Article history: Received 28 November 2013 Received in revised form 15 January 2014 Accepted 19 February 2014 Available online 26 February 2014 Keywords: Venous Thromboembolism Recurrence Risk Factors Estrogens Women
a b s t r a c t Background: It is a matter of debate whether women with an episode of VTE associated with estrogen have a lower risk of recurrence than women with an unprovoked VTE. Objectives: To identify risk factors for recurrent VTE in women and to assess the risk of recurrent VTE associated with combined oral contraceptives (CHC) or menopausal hormone treatment (HT), compared to surgery-related and unprovoked VTE. Patients/Methods: A cohort of 974 women aged 18–64 years with a first episode of VTE were followed-up for a median time of 5.2 years. All women were previously included as cases in the Swedish nation-wide case–control study “Thrombo Embolism Hormone Study” (TEHS). Hazard ratios for recurrence were calculated using univariable and multivariable Cox proportional hazards model. Results: A total of 102 patients (10%) suffered from recurrent VTE. The annual rate of recurrence was 1.0% in patients with surgery/cast, 2.0% in patients with CHC/HT and 3.2% in patients with unprovoked first VTE. Adjusted hazards ratio (HRa) for recurrence was 0.35 (95% CI 0.20-0.61) in women with VT provoked by surgery/cast while women with estrogen-associated VTE had a HRa of 0.70 (95% CI 0.43-1.20) compared to women with unprovoked VTE. Conclusion: Women 18–64 years are at low risk of recurrent VTE. Women with hormone associated VTE had a lower risk of recurrence than women with unprovoked VTE, but not as low as surgery/cast provoked VTE. © 2014 Elsevier Ltd. All rights reserved.
Introduction After a first episode of venous thromboembolism (VTE), there is a considerable risk of recurrence. The risk is highest within the first year after discontinuation of anticoagulant treatment with a cumulative incidence of 20-25% after 5 years in an unselected population [1,2]. Continuous anticoagulant treatment can prevent most episodes of recurrent VTE but decisions on extended treatment must be balanced against the risk of major bleeding complications [3–5]. The risk of recurrence after an unprovoked VTE is higher than if the VTE was provoked by a transient risk factor like surgery, cast or
Abbreviations: VTE, venous thromboembolism; CHC, combined hormonal contraceptives; HT, menopausal hormone treatment; DVT, deep vein thrombosis; PE, pulmonary embolism; TEHS, Thrombo Embolism Hormone Study; ICD, International Classification of Diseases; BMI, body mass index; HR, hazard ratios; CI, confidence intervals. ⁎ Corresponding author at: Department of Internal Medicine, Södersjukhuset, SE- 118 81 Stockholm, Sweden. Tel.: +46 8 6163462; fax: +46 8 6163115. E-mail address: [email protected] (M. Ljungqvist).
http://dx.doi.org/10.1016/j.thromres.2014.02.017 0049-3848/© 2014 Elsevier Ltd. All rights reserved.
pregnancy [2,6,7]. The risk of recurrent episodes of VTE is also influenced by gender as several studies have shown that men have at least a two-fold higher risk of recurrence than women [7–12]. A lower risk of recurrent VTE in women where the first episode was associated with estrogen treatment has been suggested as a possible explanation for this difference. Data on this matter is however still conflicting and it is debated whether hormonal treatment with combined oral contraceptives (CHC) or menopausal hormone treatment (HT) should be regarded as a provoking factor or not [9,13–17]. The current international guidelines recommend extended anticoagulant treatment in patients with a high risk of recurrence, such as after unprovoked proximal deep vein thrombosis (DVT) or unprovoked pulmonary embolism (PE) [18]. As women have a lower risk of recurrence than men there is a need to better understand the risk factors for recurrence in female patients, including the risk related to hormone associated VTE. The overall aim of this study was to assess the risk of recurrent VTE in women with hormone related thrombosis, VTE related to surgery or cast and unprovoked VTE. Furthermore, we wanted to identify the
M. Ljungqvist et al. / Thrombosis Research 133 (2014) 762–767
763
most important risk factors for recurrent VTE in young and middle aged women. Materials and Methods Population and Study Design This cohort study called TEHS-follow-up comprised women 18 to 64 years of age. All participants were previously included as cases in the “Thrombo Embolism Hormone Study” (TEHS), a population based case–control study on risk factors for VTE that recruited patients and controls from 43 larger or medium sized hospitals spread geographically in Sweden. In the TEHS- follow up study all cases in TEHS who accepted participation were followed-up concerning recurrent VTE. TEHS The methodology of the TEHS has been described elsewhere [19]. Briefly, 1433 women with a first episode of PE or DVT of the lower limb or pelvis were included as cases, from 2003 to 2009. The diagnosis of VTE was based on venography or Doppler ultrasound in DVT and CTscan or perfusion-ventilation scintigraphy in PE. Patients, who had been pregnant the past three months, had a current malignant disease or a history of malignancy within the past five years were not eligible for the TEHS. Baseline data was obtained through a telephone interview with the participants in a period of three months after the diagnosis of VTE and included information on body weight and height, risk factors for VTE, medical history, use of combined hormonal contraceptives (CHC) or menopausal hormone treatment (HT) and family history of VTE. Blood samples for DNA-analyses were collected from the women at time of inclusion. TEHS-follow-up In 2011, all cases included in TEHS who were still living in Sweden were followed-up by a questionnaire sent by mail. The questionnaire covered different medical aspects, as current medication, duration of previous and current anticoagulant treatment, hormone therapy with CHC or HT and recurrent events of VTE. Within one month a reminder was sent if the questionnaire was not returned. Out of the 1435 women included in the TEHS-study, 38 were not alive at the time of follow-up. The cause of death was retrieved from Cause of Death Register in Sweden, held by the National Board of Health and Welfare [20] and none was related to VTE. Thirteen women died from cancer, 11 from cardiovascular disease and the remaining 14 from other causes. Three patients had emigrated from Sweden leaving 1394 patients who were asked to participate in the follow-up study. Women on continuous anticoagulant treatment were not eligible for the analysis of recurrent VTE in the follow-up since they were not considered to be at risk of recurrence. A total of 345 patients rejected participation, 53 were on continuous anticoagulant treatment and 22 patients did not answer the question about current anticoagulant treatment, leaving 974 women to follow-up, Fig. 1. We obtained information on recurrent VTE both from the questionnaire and from data recorded in the Swedish Patient Register, a nationwide register held by the National Board of Health and Welfare in Sweden. From 1987 the Patient Register includes all hospital-based care in Sweden and from 2001 it also includes outpatient visits. The register cover data on age, sex, national registration number (a unique identification number assigned to every resident in Sweden) and up to 6 diagnoses and 6 surgical procedures including for each patient. Diagnoses are recorded according the current International Classification of Diseases (ICD) version 10 (ICD-10) at each out-clinic visit or hospitalization [21]. To identify potential recurrent events all study participants recorded in the register with diagnose of venous thrombosis according to ICD 10
Fig. 1. Flow-chart of the study cohort.
(I80.1-9, I82.1-9, I26.0, 9, I67.6, O22.3, O22.5) after their initial VTE were reviewed. All events, recorded in the register, were objectively verified by a review of the medical records. To be considered as a recurrent event the diagnosis was required to be confirmed by any of the same objective radiological methods as at inclusion of the TEHS-study. Furthermore, the event had to be regarded, by the treating physician, as having indication for resumed anticoagulant treatment for at least 3 months. The follow-up period started at the time of discontinuation of anticoagulant treatment after the first episode of VTE and patients were censored either at first recurrent VTE or at the date for answering the questionnaire. If information on the duration of anticoagulant treatment was missing, women with an index DVT were assumed to be treated with anticoagulants for 6 months and women with an index PE for 12 months according to the clinical praxis during the time of inclusion in Sweden. Both the TEHS and the TEHS follow-up study were approved by the research ethics committees in Sweden. All participants in TEHS and the TEHS follow-up study were included after a signed informed consent at inclusion of the study. Statistical Analyses Cumulative incidence of recurrent VTE was calculated using KaplanMeier survival analysis. All study participants were included with their maximum time available for follow-up and censored at time of recurrent event or end of follow-up. The localization of the VTE was classified as either distal DVT when localized below the popliteal vein, proximal DVT from the popliteal vein and more proximal including pelvic veins, or as PE. Patients with both DVT and PE were classified as having PE. To calculate the risk of recurrence according to characteristics of first VTE we classified patients in three groups according to provoking factor at the first VTE: unprovoked, provoked by surgery and or treatment with cast or provoked by HT/CHC treatment. CHC included oral tablets, dermal patches and intravaginal devices. HT included progestogen opposed and non-opposed oral formulations and dermal patches. Women with both surgery and HT/CHC treatment prior to the first episode of VTE were classified into the surgery/cast group as surgery/ cast has been shown to be a stronger risk factor than HT/CHC[19]. Information on weight and length were collected from time of inclusion in TEHS. Obesity was defined as body mass index (BMI) ≥ 30 kg/m2. Groups were compared using log-rank test. Cox proportional hazard model was used to calculate both crude and multivariable adjusted hazard ratios (HR) with their corresponding 95% confidence intervals (CI). The multivariable model included age, presence of factor V
764
M. Ljungqvist et al. / Thrombosis Research 133 (2014) 762–767
Leiden(G1691A) and/or prothrombin G20210A mutation (yes/no), BMI, location and extent of initial VTE (PE/proximal DVT or distal DVT) and whether the first event was provoked by surgery/cast, CHC/HT or unprovoked. For the analyses concerning recurrence by initial exposure the women with an unprovoked VTE constituted the comparison group. To estimate the HR for recurrence in women with a first VTE associated with CHC or HT, we analyzed women over and below the age of 50 years separately. The same classification as previously described was used to compare risk factors. Women 50 years or older who were using CHC at time of first VTE (n = 5) and women below 50 years using HT (n = 11) were excluded from these analyses, as were patients with missing data on use of CHC/HT at first VTE (n = 7). A sensitivity analysis was performed without exclusion of the mentioned patients without any major impact on the risk estimates. Two-sided P values b 0.05 were considered as indicating statistical significance. All statistical calculations were performed with the use of SPSS software, version 20.0 (SPSS Inc., Chicago, IL, USA). Results Characteristics of the 974 included women are shown in Table 1. The median age of women with VTE at inclusion in TEHS was 49 years (range 18–64). At the time of the first VTE 386 of all (40%) women were treated with hormones; 245 with CHC and the remaining 141 with HT. Of the CHC-users 48% used second generation progestogen (norethisterone, levonorgestrel) in combinations with estrogen, 45% used third generation progestogen (desogrestel, norgestimate, drospirenon) in combinations with estrogen and 7% used other compounds. A total of 230 patients (25%) were carriers of the FV Leiden mutation and 50 (5%) were carriers of the prothrombin G20210A mutation. In 350 women (36%) the VTE was provoked by surgery or cast and 224 of all included women (23%) were obese. Of all patients, 60% had a proximal DVT or PE as a first even of VTE. Risk Factors for Recurrence During a median follow-up of 5.2 years (range 0.1-9.1), 102 of the patients (10%) suffered from a recurrent episode of VTE (44 PE, 57 DVT). One patient had a cerebral vein thrombosis as recurrent VTE. The overall annual rate of recurrence was 2.0%. Recurrent VTE was diagnosed in 17 patients (4.9%) with a primary VTE provoked by surgery/ cast, 27 patients (10.0%) using CHC or HT and in 57 of the patients (16%) with a primary unprovoked VTE. This corresponds to a mean annual rate of recurrence of 1.0% in patients with surgery/cast, 2.0% in patients with CHC/HT and 3.2% in patients with unprovoked VTE. The follow-up consisted of 974 women corresponding to a total of 4.941 Table 1 Baseline characteristics of patients. All patients N = 974
Patients b 50 yrs N = 510
Patients ≥ 50 yrs N = 464
Age, median (years) CHC (%) HT (%) Missing data on CHC/HT (%) FV Leiden FII GA20210A Provoked by surgery/cast
49 (SD 13) 245 (25,2) 141 (14.5) 6 (0.72) 230 (23.6) 50 (5,1) 350 (35.9)
36 (SD 9) 240 (47.1) 11 (2.2) 2 (0.59) 132 (25.9) 27 (5.3) 181 (35.5)
58 (SD 4) 5 (1.1) 130 (28.0) 4 (0.86) 98 (21.1) 23 (5.0) 169 (36.4)
Site of index thrombosis Distal DVT Proximal DVT PE BMI (SD) BMI b 30 (%) BMI ≥ 30 (%)
389 (39.9) 293 (30.1) 292 (30.0) 26 (SD 5) 750 (77.0) 224 (23.0)
207 (40.6) 140 (27.5) 163 (32.0) 25 (SD 6) 401 (78.6) 109 (22.4)
182 (39.2) 153 (33.0) 129 (27.8) 27 (SD 5) 349 (75.2) 115 (24.8)
CHC = Combined hormonal contraceptives, HT = Menopausal hormone therapy, FII G20210A = Prothrombin G20210A mutation.
person-years and the cumulative incidence of recurrence according to provoking factor of first VTE is shown in Table 2. All patients had the highest rate of recurrence during the first year after cessation of anticoagulant treatment The crude HR for recurrence was 0.28 (95% CI 0.16-0.47) in women with a first VTE provoked by surgery/cast and 0.57 (95% CI 0.36-0.90) in women with a first VTE associated with hormone treatment, when compared to women with unprovoked VTE. In the multivariable models the risk estimates increased slightly and remained statistically significant for women with VTE provoked by surgery/cast but not for women having hormone treatment prior to their first VTE, Table 3. A plot of the cumulative incidence of recurrent VTE from multivariable Cox regression analysis in women with a first VTE provoked by HT/CHC, surgery/cast and unprovoked VTE is shown in Fig. 2. Obesity was correlated with an increased risk of recurrence with HR 1.8 (95% CI 1.2-2.7) compared to non-obese women. The increased risk persisted in multivariable adjustments. Carrier ship for factor V Leiden (G1691A) and/or the prothrombin G20210A was not significantly associated with recurrence and neither was a positive family history, Table 3. Age was not associated with the risk of recurrent events (data not shown). The localization of VTE was associated with the risk of recurrence. Patients with proximal DVT or PE had higher risk compared to distal DVT (HR 2.0, 95% CI 1.3-3.1), with consistent risk estimates in the multivariable analysis (HRa 2.0, 95% CI 1.3-3.2), Table 2. The initial localization of VTE was also predictive of the site of recurrent VTE. In patients with initial PE 68% of the recurrent events consisted of PE. Corresponding numbers for patients with proximal DVT and distal DVT was 51% and 63% respectively. About 1/3 of the patients with initial DVT had a recurrent event in form of PE (8 with initial distal DVT and 9 with initial proximal DVT). Risk Factors for Recurrence in Women b 50 years Hazard ratios and there 95% confidence intervals in women younger than 50 years of age at time of first VTE are shown in Table 4. Women with VTE provoked by surgery/cast had the lowest risk of recurrence. Obesity was associated with an increased risk of recurrent VTE with HRa 2.5 (95% CI 1.3-4.7) compared to non-obese patients in this agegroup. The presence of factor V Leiden and/or prothrombin G20210A mutation, a positive family history or age had no major impact on the risk estimate. Risk Factors for Recurrence in Women ≥ 50 years Table 5 show hazard ratios and there 95% confidence intervals in women ≥ 50 years at time of index VTE. Women with VTE provoked by surgery/cast had an even lower risk of recurrence, HRa 0.29 (95% CI 0.14-0.66) in this age-group. As in women b 50 years, the presence of factor V Leiden and/or prothrombin G20210A mutation, a positive family history or age had no major impact on the risk estimates. Obesity
Table 2 Cumulative incidence of recurrent VTE according to risk factor at first episode of VTE.
1 2 3 4 5 6
year years years years years years
All women N = 974
Women with Surgery/Cast N = 350
Women with CHC/HT N =272
Women with Unprovoked VTE N = 352
4% 6% 8% 9% 9% 11%
2% 3% 3% 3% 4% 6%
5% 7% 7% 8% 9% 10%
7% 10% 13% 15% 15% 18%
CHC/HT = treatment with either combined oral contraceptives or menopausal hormone therapy.
M. Ljungqvist et al. / Thrombosis Research 133 (2014) 762–767 Table 3 Risk for recurrence of venous thromboembolism in 974 women presented as HRs with 95% CI.
Unprovoked Surgery/cast CHC/HT BMI ≥ 30 Genetic factor V/II Family history PE/Proximal DVT
Recurrent events/person years
HR (95% CI)
HRa (95% CI)
57/1765 17/1776 27/1376 35/1137 33/1340 32/1229 75/2950
Reference 0.3 (0.2-0.5) 0.6 (0.4-0.9) 1.8 (1.2-2.7) 1.4 (0.9-2.1) 1.4 (0.9-2.2) 2.0 (1.3-3.1)
Reference 0.4 (0.2-0.7) 1 0.7 (0.4-1.1) 1 1.6 (1.0-2.4) 2 1.3 (0.9-2.0) 3 1.2 (0.8-1.8) 3 2.0 (1.3-3.2) 4
Adjusted for: 1Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, 2. Provoking factor, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, age 3Provoking factor, BMI, location of index VTE, age 4Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, provoking factor.
did not correlate with an increased risk of recurrent VTE in this agegroup. Discussion In this large Swedish long-term follow-up study of women 18–64 years with VTE we found that the overall risk of recurrent VTE was low. The risk after a first event of VTE associated with CHC or HT was lower than in cases with unprovoked VTE but higher than in VTE associated with surgery or cast treatment. The localization of the primary event and obesity were the two most important clinical risk factors for recurrent VTE in our study population. Several studies have reported lower risk of recurrent VTE in women who are using CHC or HT at the time of first VTE, compared to unprovoked VTE [7,11,13,22]. Only few studies included patients with provoked VTE and none compared the risk of recurrence in hormone associated VTE to surgery-provoked VTE [11,23]. The PREVENT-study found lower risk estimates of recurrence in women with CHC or HT associated VTE although the difference was not statistically significant [13]. Furthermore, in a study of 361 women less than 45 years of age in Austria, the use of CHC at first VTE had no impact on the risk of recurrence, HR 0.85 (95% CI 0.6-1.21) [16]. In the REVERSE-study, the risk of recurrence was not significantly lower than in unprovoked cases, and
765
Table 4 Risk for recurrence of venous thromboembolism in 503 women b 50 years presented as HRs with 95% CI.
Unprovoked Surgery/cast CHC BMI ≥ 30 PE/Proximal DVT
Recurrent events/person-years
HR
HRa
22/762 9/878 12/828 18/543 31/1.482
Reference 0.3 (0.2-0.7) 0.7 (0.3-1.2) 2.4 (1.3-4.3) 1.74 (0.9-3.3)
Reference 0.5 (0.2-0.9) 1 0.9 (0.2-1.1) 1 2.4 (1.3-4.7) 2 1.33 (0.7-2.6) 3
Adjusted for: 1Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, 2Provoking factor, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, age 3Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, provoking factor.
the authors suggest that hormone associated VTE cannot be regarded as provoked VTE [14]. Our large study on female VTE patients allows us to compare the risk of recurrence in women with CHC/HT to both unprovoked and surgery/ cast-provoked cases. Interestingly, our data suggest that the risk of CHC/HT associated VTE is not the same as either unprovoked or surgery-provoked VTE, but somewhere in-between. Furthermore, the clinical data available in this study allowed for multivariate adjustments to control for potential confounding by age, BMI and thrombophilia. This was not possible in previous large studies due to lack of data concerning BMI and thrombophilia[11]. The present study also allowed for subgroup analysis on women over and below 50 years to look at HT and CHC separately with consistent results concerning the risk of recurrence related to a first VTE in association with CHC/HT. We found that use of CHC or HT was associated with reduced risk of recurrence compared to unprovoked VTE in the crude analysis of the whole study population. This difference was no longer statistically significant after multivariate adjustments for BMI, localization of index VTE, age and thrombophilia which was probably due to lack of power. Nevertheless we believe that the true risk estimate is lower than in unprovoked VTE which is also indicated by the HR. In line with previously published data we found that obesity was significantly associated with increased risk of recurrence [24]. When we stratified the analysis to women over and under 50 years we noticed that this correlation was only valid in women under 50 years of age. In a previous study, Olie´ et al. showed that obesity was a risk factor for recurrence in women only and in their analysis of 349 women and 234 men the mean age of the women was 46 years and men 54 years [10]. Others, who have not found any correlation between obesity and recurrent VTE in women, have usually studied an older patient group [22] which suggests that obesity might be a more important risk factor in young women than in other populations. In a study of women under 45 years of age, BMI N 25 kg/m2 was the only clinical characteristic correlated with increased risk of recurrence [16]. As previously reported by others, the presence of any of the two most common genetic risk factors Factor V Leiden(G1691A) or Prothrombin G20210A mutation, did not affect the risk estimates of
Table 5 Risk for recurrence of venous thromboembolism in 449 women ≥ 50 years presented as HRs with 95% CI.
Unprovoked Surgery/cast HT BMI ≥ 30 PE/Proximal DVT Fig. 2. Comulative incidence of VTE based on multivariable Cox proportional Hazards model according to provoking factor at first VTE. Adjustments for BMI, localization of index VTE and carrier ship of Factor V Leiden/FII GA20210A.
Recurrent events/person-years
HR (95% CI)
HRa (95% CI)
35/909 8/860 10/537 16/561 41/1336
Reference 0.3 (0.1-0.5) 0.5 (0.3-1.1) 1.4 (0.8-2.4) 2.4 (1.3-4.6)
Reference 0.3 (0.1-0.7) 1 0.5 (0.2-1.0) 1 1.1 (0.6-2.0) 2 2.7 (1.3-5.5) 3
Adjusted for: 1Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, 2Provoking factor, carrier ship of factor V Leiden and/or FII G20210A, location of index VTE, age 3Age, BMI, carrier ship of factor V Leiden and/or FII G20210A, provoking factor.
766
M. Ljungqvist et al. / Thrombosis Research 133 (2014) 762–767
recurrent VTE in any major way [23,25]. Neither did a positive family history of VTE, which has been shown to be an independent risk factor of first VTE [26,27]. This paradox could be explained by index event bias where carrier ship of thrombophilia at first event of VTE is associated with lower occurrence of other risk factors [28]. The risk of recurrence in patients with primary distal DVT was, as expected and previously reported, significantly lower than in proximal DVT or PE [2,5,29]. However 26 of 389 patients with initial distal DVT suffered from recurrence and 8 of these events were in form of PE. The strength of this study is its size; it is to our knowledge the largest follow up study of a defined study cohort of women with clinical data concerning the risk of recurrent VTE after an initial VTE in association with CHC or HT. The use of both the Swedish Patient Register and the questionnaires specifically designed for this study ensured that all recurrent events were identified. The thorough review of medical charts in patients with suspected recurrent events reassured that all recurrences were clinically relevant. A few limitations of this study must however be emphasized. First, patients with continuous anticoagulant treatment after the first episode of VTE were not included in the analysis. Patients being considered at high risk of recurrence already at the first VTE were therefore not included in our study, something that is hard to avoid without exposing patients to unnecessary risk. It must however be held in mind when interpreting the results. Data on hormonal treatment during follow up was assessed for all patients and 22 women resumed treatment with HT (n = 19) or CHC (n = 3) after first VTE. To avoid immortal time bias we didn´t exclude these patients in the follow-up. The low number of recurrent events in our study population resulted in some estimates of recurrence that were not statistically significant on the p b 0.05 level. The estimated risk is however likely to be close to the actual risk. The results of this study may have clinical implications. First of all, our study confirms that young and middle aged women have a low risk of recurrent VTE. The current guidelines advocate extended anticoagulant treatment with annual reassessment in unprovoked VTE, regardless of the gender of the patient [18]. The rate of major bleeding during anticoagulant treatment is 3-5% per year. Hence, in patients with a mean annual rate of recurrent VTE less than 5%, extended anticoagulant treatment could lead to higher risk of bleeding complications than favourable effect of protecting against VTE recurrence. Since all women in this study had an annual rate of VTE of less than 5%, and even women with unprovoked VTE had as low annual rate of recurrence as 3.2%, our results suggest that most women in this age group would not benefit from extended therapy. Our results also suggest that obesity is a risk factor for recurrent VTE in women below the age of 50 years, but not for women between 50 and 64. This needs to be evaluated in further studies. The overall risk of recurrence in this group is low but better tools to identify the women at increased risk are still warranted, particularly for women with unprovoked VTE. Several clinical prediction scores with data on d-dimer, and in some cases obesity, have been published [15,22,29]. The future challenge is to validate clinical scores in different subpopulations, including young and middle aged women. As different risk factors seems to be of varying importance in men and women, and also in different age groups, scoring systems might need to be adjusted according to the age and gender of the patient. In conclusion, we found that women 18–64 years have a low risk of recurrence in VTE. Women with hormone associated VTE had a lower risk of recurrence than women with unprovoked VTE, but not as low as surgery/cast provoked VTE. In women below 50 years, obesity was correlated with increased risk of recurrence, suggesting that information on BMI should be included in of the evaluation of risk in younger women. Conflict of Interest Statement No conflict of interest.
Acknowledgements We like to thank Jacob Odeberg for the genotyping in TEHS-study and Hans Pettersson at Karolinska Institutet, Södersjukhuset for helpful discussions on statistical analyses. This study was funded by Janssen-Cilag,Novartis,Organon, Schering, Wyeth, AFA, Center for Gender Medicine KI and by the authors’ affiliations. The sponsors were not involved in the design and conduct of the study; collection, management, analysis, nor interpretation of the data; and preparation, review, nor approval of the manuscript.
References [1] Prandoni P, Lensing AW, Cogo A, Cuppini S, Villalta S, Carta M, et al. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med 1996;125:1–7. [2] Hansson PO, Sorbo J, Eriksson H. Recurrent venous thromboembolism after deep vein thrombosis: incidence and risk factors. Arch Intern Med 2000;160:769–74. [3] Kearon C, Gent M, Hirsh J, Weitz J, Kovacs MJ, Anderson DR, et al. A comparison of three months of anticoagulation with extended anticoagulation for a first episode of idiopathic venous thromboembolism. N Engl J Med 1999;340:901–7. [4] Linkins LA, Choi PT, Douketis JD. Clinical impact of bleeding in patients taking oral anticoagulant therapy for venous thromboembolism: a meta-analysis. Ann Intern Med 2003;139:893–900. [5] Schulman S, Rhedin AS, Lindmarker P, Carlsson A, Larfars G, Nicol P, et al. A comparison of six weeks with six months of oral anticoagulant therapy after a first episode of venous thromboembolism. Duration of Anticoagulation Trial Study Group. N Engl J Med 1995;332:1661–5. [6] Schulman S, Lindmarker P, Holmstrom M, Larfars G, Carlsson A, Nicol P, et al. Postthrombotic syndrome, recurrence, and death 10 years after the first episode of venous thromboembolism treated with warfarin for 6 weeks or 6 months. J Thromb Haemost 2006;4:734–42. [7] Baglin T, Luddington R, Brown K, Baglin C. Incidence of recurrent venous thromboembolism in relation to clinical and thrombophilic risk factors: prospective cohort study. Lancet 2003;362:523–6. [8] Kyrle PA, Rosendaal FR, Eichinger S. Risk assessment for recurrent venous thrombosis. Lancet 2010;376:2032–9. [9] Kyrle PA, Minar E, Bialonczyk C, Hirschl M, Weltermann A, Eichinger S. The risk of recurrent venous thromboembolism in men and women. N Engl J Med 2004;350:2558–63. [10] Olie V, Zhu T, Martinez I, Scarabin PY, Emmerich J. Sex-specific risk factors for recurrent venous thromboembolism. Thromb Res 2012;130:16–20. [11] Douketis J, Tosetto A, Marcucci M, Baglin T, Cosmi B, Cushman M, et al. Risk of recurrence after venous thromboembolism in men and women: patient level meta-analysis. BMJ 2011;342:d813. [12] McRae S, Tran H, Schulman S, Ginsberg J, Kearon C. Effect of patient's sex on risk of recurrent venous thromboembolism: a meta-analysis. Lancet 2006;368:371–8. [13] Cushman M, Glynn RJ, Goldhaber SZ, Moll S, Bauer KA, Deitcher S, et al. Hormonal factors and risk of recurrent venous thrombosis: the prevention of recurrent venous thromboembolism trial. J Thromb Haemost 2006;4:2199–203. [14] Le Gal G, Kovacs MJ, Carrier M, Do K, Kahn SR, Wells PS, et al. Risk of recurrent venous thromboembolism after a first oestrogen-associated episode. Data from the REVERSE cohort study. Thromb Haemost 2010;104:498–503. [15] Rodger MA, Kahn SR, Wells PS, Anderson DA, Chagnon I, Le Gal G, et al. Identifying unprovoked thromboembolism patients at low risk for recurrence who can discontinue anticoagulant therapy. CMAJ 2008;179:417–26. [16] Laczkovics C, Grafenhofer H, Kaider A, Quehenberger P, Simanek R, Mannhalter C, et al. Risk of recurrence after a first venous thromboembolic event in young women. Haematologica 2007;92:1201–7. [17] Le Moigne E, Delluc A, Tromeur C, Nowak E, Mottier D, Lacut K, et al. Risk of recurrent venous thromboembolism among young women after a first event while exposed to combined oral contraception versus not exposed to: a cohort study. Thromb Res 2013;132:51–5. [18] Kearon C, Akl EA, Comerota AJ, Prandoni P, Bounameaux H, Goldhaber SZ, et al. Antithrombotic therapy for VTE disease: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e419S–94S. [19] Bergendal A, Bremme K, Hedenmalm K, Larfars G, Odeberg J, Persson I, et al. Risk factors for venous thromboembolism in pre-and postmenopausal women. Thromb Res 2012;130:596–601. [20] Socialstyrelsen [National Board of Health and Welfare]. Cause of Death Register [Dödsorsakregistret]. http://www.socialstyrelsen.se/register/dodsorsaksregistret. [21] Socialstyrelsen [National Board of Health and Welfare]. National Inpatient Register [Patientregistret]. http://www.socialstyrelsen.se/register/halsodataregister/ patientregistret. [22] Tosetto A, Iorio A, Marcucci M, Baglin T, Cushman M, Eichinger S, et al. Predicting disease recurrence in patients with previous unprovoked venous thromboembolism: a proposed prediction score (DASH). J Thromb Haemost 2012;10:1019–25. [23] Marchiori A, Mosena L, Prins MH, Prandoni P. The risk of recurrent venous thromboembolism among heterozygous carriers of factor V Leiden or prothrombin G20210A mutation. A systematic review of prospective studies. Haematologica 2007;92:1107–14.
M. Ljungqvist et al. / Thrombosis Research 133 (2014) 762–767 [24] Eichinger S, Hron G, Bialonczyk C, Hirschl M, Minar E, Wagner O, et al. Overweight, obesity, and the risk of recurrent venous thromboembolism. Arch Intern Med 2008;168:1678–83. [25] Ho WK, Hankey GJ, Quinlan DJ, Eikelboom JW. Risk of recurrent venous thromboembolism in patients with common thrombophilia: a systematic review. Arch Intern Med 2006;166:729–36. [26] Bezemer ID, van der Meer FJ, Eikenboom JC, Rosendaal FR, Doggen CJ. The value of family history as a risk indicator for venous thrombosis. Arch Intern Med 2009;169:610–5.
767
[27] Sonnevi K, Bergendal A, Adami J, Larfars G, Kieler H. Self-reported family history in estimating the risk of hormone, surgery and cast related VTE in women. Thromb Res 2013;132:164–9. [28] Dahabreh IJ, Kent DM. Index event bias as an explanation for the paradoxes of recurrence risk research. JAMA 2011;305:822–3. [29] Eichinger S, Heinze G, Jandeck LM, Kyrle PA. Risk assessment of recurrence in patients with unprovoked deep vein thrombosis or pulmonary embolism: the Vienna prediction model. Circulation 2010;121:1630–6.
