Differences in clinical presentation of pulmonary embolism in women and men

Journal of Thrombosis and Haemostasis, 8: 693 698 DOI: 10.1111/j.1538-7836.2010.03774.x ORIGINAL ARTICLE Differences in clinical presentation of pulmonary embolism in women and men H. ROBERT-EBADI,* G. LE GAL, M. CARRIER,à F. COUTURAUD, A. PERRIER, H. BOUNAMEAUX* and M. RIGHINI* *Division of Angiology and Hemostasis, Geneva University Hospital and Faculty of Medicine, Geneva, Switzerland; Department of Internal Medicine and Chest Diseases, EA 3878 (GETBO), Brest University Hospital, Brest, France; àdivision of Hematology, Department of Medicine, University of Ottawa, Ottawa, Canada; and Department of Internal Medicine, Geneva University Hospital and Faculty of Medicine, Geneva, Switzerland To cite this article: Robert-Ebadi H, Le Gal G, Carrier M, Couturaud F, Perrier A, Bounameaux H, Righini M. Differences in clinical presentation of pulmonary embolism in women and men. J Thromb Haemost 2010; 8: 693 8. Summary. Background: The risk of recurrence of pulmonary embolism (PE) is higher in men than in women. Differences in clinical presentation of deep vein thrombosis (DVT) have been reported between the two genders but comparative data on PE are lacking. Objectives: To compare clinical characteristics between women and men with suspected and confirmed PE and their impact on clinical probability prediction scores and on diagnostic work-up of PE, and to assess whether differences at presentation could account for the increased recurrence rate in men. Methods: Combined data from three prospective cohort studies including a total of 3414 outpatients with suspected PE were analyzed retrospectively. Clinical characteristics, pretest probability of PE, diagnostic yield of non-invasive tests and VTE recurrence rate were compared between genders. Results: The overall prevalence of PE was similar among women and men (22.3% vs. 23.1%; P =0.55). The clinical probability prediction scores (Geneva score and Wells score) performed equally well in both genders. A noninvasive diagnostic work-up was possible more often in men than in women. The proportion of PE-associated proximal DVT was higher in men than in women (43% vs. 33%; P = 0.009). VTE recurrence rate was also higher in men than women with PE (5.0% vs. 2.3%; P = 0.045). Conclusion: In spite of some differences in the clinical presentation of PE between women and men, clinical probability prediction scores perform equally in both genders. A higher prevalence of PEassociated proximal DVT in men could possibly indicate greater severity of PE episodes and partly account for the higher VTE recurrence rate in men. Keywords: clinical presentation, diagnostic tests, gender, prediction scores, pretest probability, pulmonary embolism. Introduction Venous thromboembolism (VTE) is a common and potentially fatal disorder. The estimated annual incidence of VTE is around 1 per 1000 in the general population, with half of the episodes presenting as pulmonary embolism (PE) [1]. Several studies have assessed the relationship between gender and incidence of VTE [2,3]. Given the diverging results and the small differences observed, it seems that the incidence of VTE is overall similar in both genders [4]. However, VTE recurrence rate has consistently been shown to be higher in men [5 7], suggesting that some VTE characteristics may differ between women and men. A recent study has indeed shown differences in the clinical presentation of DVT between women and men [8]. This prompted us to study whether clinical differences exist at initial presentation of PE. Also, as clinical probability assessment has long proven to be accurate to guide further investigations for the diagnosis of PE [9,10] and its systematic use is strongly supported by a recent position paper from international experts [11], we evaluated whether these differences in clinical presentation could have an impact on the performance of clinical probability prediction scores and on the diagnostic yield of non-invasive tests. Finally, we sought to identify if differences at presentation of PE, especially a more proximal localization of PE or the presence of concomitant proximal DVT could account for the difference in recurrence rate between women and men reported in the literature. Correspondence: Marc Righini, Division of Angiology and Hemostasis, Department of Internal Medicine, Geneva University Hospital and Faculty of Medicine, 4 rue Gabrielle Perret-Gentil CH- 1211 Geneva 14, Switzerland. Tel.: +41 22 372 92 92; fax. +41 22 372 92 99. E-mail: marc.righini@hcuge.ch Received 14 September 2009, accepted 17 January 2010 Methods Patients and setting We analyzed combined data from three prospective cohort studies that included a total of 3414 outpatients with suspected PE. These outcome studies were all designed to evaluate

694 H. Robert-Ebadi et al diagnostic strategies for PE, combining clinical probability assessment, plasma D-dimer measurement (ELISA), lower limb venous compression ultrasonography (CUS) and chest computed tomography (CT) with some variations described below. All consecutive outpatients admitted to the emergency department were included if they had a clinical suspicion of PE defined as acute onset of new or worsening shortness of breath or chest pain without any other obvious etiology. Written informed consent was obtained from all patients. All three studies were multicenter studies conducted in Belgium, France and Switzerland [12 14]. Clinical probability assessment was performed using the Geneva score in the first two studies [15], and the revised Geneva score in the third study [16]. The first study was conducted between October 2000 and June 2002 and therefore single-detector CT was performed in the majority of patients [12]. The second study conducted between August 2002 and November 2003 used multi-detector CT [13]. Finally, the third study performed between January 2005 and August 2006 had a randomized non-inferiority design comparing a strategy with multi-detector CT with or without CUS [14]. Therapeutic anticoagulation was initiated in patients with confirmed PE, and the rate of thromboembolic events was assessed at 3 months in all patients. Data analysis First, we analyzed the prevalence of clinical characteristics (thromboembolic risk factors, symptoms and clinical signs) in women and men with suspected PE. Second, these characteristics were compared for each gender between patients with and without confirmed PE. v 2 -test was used to compare qualitative variables and StudentÕs t-test for quantitative variables. Third, we evaluated the influence of the presence or absence of risk factors, symptoms and clinical signs by computing their positive and negative likelihood ratios and their 95% confidence intervals for PE. Fourth, distribution of patients between clinical probability groups and prevalence of PE according to clinical probability was compared between women and men. As mentioned above, clinical probability was assessed prospectively in the three studies using the Geneva scores. We also evaluated clinical probability by retrospectively computing the three-level Wells score [17] in our database. Data were available for the latter analysis in 3319 out of 3414 patients (97.2%). In the remaining 95 patients (2.8%), information on the item Ôother diagnosis less likely than PEÕ from the Wells score was not available. Fifth, the contribution of each diagnostic test such as D-dimer, CUS and CT was compared between women and men using a v 2 -test. For the latter analysis, only the CUS arm of our randomized study was considered. Sixth, we analyzed differences in the most proximal localization of PE and in the rate of associated proximal DVT according to gender. Moreover, we compared the risk of recurrent VTE during the 3-month followup between women and men with confirmed PE. Finally, the same analyses were performed after excluding women with hormonal exposure. Results We analyzed data from a total of 3414 outpatients with suspected PE. The study population comprised 1940 women (57%; mean age 60 ± 17) and 1474 men (43%; mean age 60 ± 20). The diagnosis of PE was confirmed in 773 patients (22.6%): 432 out of 1940 women and 341 out of 1474 men (22.3% vs. 23.1%; P = 0.55). Thromboembolic risk factors, symptoms and clinical signs according to gender are presented in Table 1. A personal or family history of VTE, or varicose veins was more commonly found in women, whereas cancer and chronic obstructive pulmonary disease (COPD) were more frequent in men. Both spontaneously reported calf pain and pain at calf palpation were significantly more frequently found in women. Displayed in Table 2 for each gender is the comparison of risk factors, symptoms and clinical signs between patients with and without PE. We tested the interaction of gender in the association between each risk factor, symptom, clinical sign on one hand and the diagnosis of PE on the other hand. The Table 1 Clinical characteristics of patients with suspected pulmonary embolism Characteristics (n = 1940) (n = 1474) P-value Risk factors Previous DVT/PE 377 (19.4) 231 (15.7) 0.004 Family history of VTE 306 (15.8) 130 (8.9) < 0.0001 COPD 134 (6.9) 212 (14.4) < 0.0001 Cancer 133 (6.9) 158 (10.7) < 0.0001 Surgery within 1 month 113 (5.8) 75 (5.1) 0.35 Varicose veins 471 (24.5) 228 (15.6) < 0.0001 Oral contraception 204 (10.5) 0 (0.0) Hormone replacement therapy 157 (8.1) 0 (0.0) Chronic heart failure 119 (6.6) 124 (9.0) 0.01 Symptoms Chest pain 1327 (68.5) 998 (67.7) 0.63 Dyspnea 1394 (71.9) 996 (67.6) 0.007 Syncope 338 (17.4) 278 (18.9) 0.29 Recent cough 419 (21.6) 321 (21.8) 0.9 Calf pain 332 (17.1) 202 (13.7) 0.007 Anxiety 275 (20.0) 120 (11.5) < 0.0001 Hemoptysis 80 (4.1) 83 (5.6) 0.04 Clinical signs Unilateral lower limb edema 137 (11.4) 112 (11.6) 0.87 Pain at calf palpation 158 (13.2) 87 (9.1) 0.002 Respiratory rate 20.7 (6.5) 20.8 (8.4) 0.83 Heart rate 87.3 (19.3) 88.0 (31.7) 0.41 Systolic blood pressure 139 (24) 139 (23) 0.73 Diastolic blood pressure 79 (26) 81 (15) 0.001 Alternative diagnosis more likely than PE 1102 (58.2) 914 (63.5) 0.002 DVT, deep vein thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism; COPD, chronic obstructive pulmonary disease. Data are presented as numbers (%); continuous variables are presented as mean (±range).

Gender and presentation of PE 695 Table 2 Prevalence of symptoms, risk factors and clinical signs according to gender and presence/absence of pulmonary embolism (n = 1940) (n = 1474) Characteristics PE No PE P PE No PE P Interaction P Risk factors Previous DVT/PE 135 (31.3) 242 (16.0) < 0.0001 105 (30.8) 126 (11.1) < 0.0001 0.004 Family history of VTE 89 (20.7) 217 (14.5) 0.002 47 (13.8) 83 (7.4) < 0.0001 0.27 Cancer 37 (8.6) 96 (6.4) 0.11 50 (14.7) 108 (9.5) 0.007 0.54 COPD 18 (4.2) 116 (7.7) 0.01 29 (8.5) 183 (16.2) < 0.0001 0.82 Surgery (within 1 month) 45 (10.4) 68 (4.5) < 0.0001 23 (6.7) 52 (4.6) 0.11 0.13 Varicose veins 131 (30.4) 340 (22.8) 0.001 77 (22.7) 151 (13.4) < 0.0001 0.21 Chronic heart failure 30 (7.4) 89 (6.4) 0.47 23 (7.3) 101 (9.5) 0.25 0.18 Symptoms Chest pain 254 (55.8) 1073 (71.2) < 0.0001 209 (61.3) 789 (69.6) 0.004 0.29 Dyspnea 370 (85.6) 1024 (67.9) < 0.0001 268 (78.8) 728 (64.3) < 0.0001 0.14 Syncope 77 (17.8) 261 (17.3) 0.82 71 (20.8) 207 (18.3) 0.29 0.54 Calf pain 123 (28.5) 209 (13.9) < 0.0001 97 (28.4) 105 (9.3) < 0.0001 0.03 Anxiety 49 (15.8) 226 (21.2) 0.04 24 (10.0) 96 (11.6) 0.49 0.52 Hemoptysis 16 (3.7) 64 (3.2) 0.62 24 (7.0) 59 (5.2) 0.20 0.22 Clinical signs Unilateral lower limb edema 61 (22.4) 76 (8.2) < 0.0001 57 (26.4) 55 (7.4) < 0.0001 0.24 Pain at calf palpation 69 (27.7) 89 (9.6) < 0.0001 49 (22.8) 38 (5.1) < 0.0001 0.008 Heart rate > 100 bpm 114 (26.4) 283 (18.8) 0.001 100 (29.3) 234 (20.7) 0.001 0.89 Alternative diagnosis more likely than PE 93 (22.1) 1009 (68.5) < 0.0001 98 (30.2) 816 (73.2) < 0.0001 0.3 DVT, deep vein thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism; COPD, chronic obstructive pulmonary disease. Data are presented as numbers (%); continuous variables are presented as mean (±range). Table 3 Diagnostic values of clinical characteristics Characteristics LR+ LR) LR+ LR) P-value Risk factors Previous DVT/PE 2.0 (1.6 2.3) 0.8 (0.8 0.9) 2.8 (2.2 3.5) 0.8 (0.7 0.8) 0.004 Family history of VTE 1.4 (1.1 1.8) 0.9 (0.9 1.0) 1.9 (1.3 2.6) 0.9 (0.9 1.0) 0.27 Cancer 1.3 (0.9 1.9) 1.0 (0.9 1.0) 1.5 (1.1 2.1) 0.9 (0.9 1.0) 0.54 Surgery within 1 month 2.3 (1.6 3.3) 0.9 (0.9 1.0) 1.5 (0.9 2.4) 1.0 (0.9 1.0) 0.13 Varicose veins 1.3 (1.1 1.6) 0.9 (0.8 1.0) 1.7 (1.3 2.2) 0.9 (0.8 1.0) 0.21 Oral contraception 1.1 (0.8 1.5) 1.0 (1.0 1.1) Hormone replacement therapy 1.5 (1.1 2.1) 1.0 (0.9 1.0) Chronic heart failure 1.2 (0.8 1.7) 1.0 (1.0 1.0) 0.8 (0.5 1.2) 1.0 (1.0 1.1) 0.18 Symptoms Chest pain 0.8 (0.8 0.9) 1.4 (1.2 1.6) 0.9 (0.8 1.0) 1.3 (1.1 1.5) 0.29 Dyspnea 1.3 (1.2 1.3) 0.4 (0.4 0.6) 1.2 (1.1 1.3) 0.6 (0.5 0.7) 0.14 Syncope 1.0 (0.8 1.3) 0.9 (0.9 1.0) 1.1 (0.9 1.4) 1.0 (0.9 1.0) 0.54 Calf pain 2.1 (1.7 2.5) 0.8 (0.8 0.9) 3.1 (2.4 3.9) 0.8 (0.7 0.8) 0.03 Anxiety 0.7 (0.6 1.0) 1.1 (1.0 1.1) 0.9 (0.6 1.3) 1.0 (1.0 1.1) 0.52 Hemoptysis 0.9 (0.5 1.5) 1.0 (1.0 1.0) 1.4 (0.9 2.1) 1.0 (1.0 1.0) 0.22 Clinical signs Unilateral lower limb edema 2.7 (2.0 3.7) 0.8 (0.8 0.9) 3.6 (2.6 5.0) 0.8 (0.7 0.9) 0.24 Pain at calf palpation 2.7 (2.0 3.6) 0.8 (0.8 0.9) 4.5 (3.0 6.6) 0.8 (0.8 0.9) 0.008 Heart rate > 100 bpm 1.4 (1.2 1.7) 0.9 (0.9 1.0) 1.4 (1.2 1.7) 0.9 (0.8 1.0) 0.89 Alternative diagnosis more likely than PE 0.7 (0.6 0.8) 1.4 (1.3 1.6) 0.7 (0.7 0.8) 1.4 (1.2 1.6) 0.62 DVT, deep vein thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism. Data are presented as positive likelihood ratios and negative likelihood ratios (95% confidence interval) for PE in the presence or absence of each characteristic. difference in the prevalence of calf pain between patients with and without PE was significantly more pronounced in men than in women (interaction P-values 0.03 and 0.008 for spontaneous calf pain and calf pain at palpation, respectively). Also, the association between a history of previous VTE and PE was significantly stronger in men (P = 0.004). Corresponding positive and negative likelihood ratios in women and inmenaredisplayedintable3. We also compared the contribution of each diagnostic test to the diagnosis of PE between genders. The distribution of

696 H. Robert-Ebadi et al patients in the three categories of clinical probability was not different between women and men (P = 0.31). Moreover, the proportion of confirmed PE in each clinical probability group (i.e. low, intermediate or high) assessed prospectively using the Geneva score was similar in both genders (Table 4). Interestingly, the proportion of confirmed PE was also similar between women and men in the low, intermediate and high probability groups assessed retrospectively using the Wells score (Table 4). As shown in Table 5, there were differences in the diagnostic yield of D-dimer, CUS and CT between women and men. In men, CUS and D-dimer were more useful in ruling in and out the diagnosis, respectively. Overall, 64% of women but only 57% of men had to undergo a CT to achieve a final diagnosis (P = 0.001). We also assessed disease severity in women and men using the most proximal level of PE and the presence of associated proximal DVT as indicators of higher clot burden and greater severity of the episode. Among PE diagnosed by CT, the most proximal level was troncular in 29%, lobar in 37%, segmental in 30% and multiple subsegmental in 4%, without any significant difference between genders. However, the proportion of PE-associated proximal DVT was higher in men than in women: 43% of men compared with 33% of women had an associated proximal DVT (P = 0.009). Finally, we compared the VTE recurrence rate between women and men with PE during the 3 months follow-up, i.e. while patients were still on anticoagulation. Interestingly, 17 out of 341 men (5.0%) vs. 10 out of 432 women (2.3%) Table 4 Distribution of clinical probability groups and pulmonary embolism prevalence according to clinical probability using the Geneva score and the three-level Wells score in women and men (n = 1940) (n = 1474) P-value Geneva score CP Low 835 (43) 672 (46) Intermediate 988 (51) 713 (48) 0.31 High 117 (6) 89 (6) PE prevalence according to CP Low CP 59 (7) 59 (9) Intermediate CP 271 (27) 200 (28) High CP 102 (87) 82 (92) Global prevalence 432 (22.3) 341 (23.1) 0.55 (n = 1886) Wells score CP Low 948 (50.3) 781 (54.5) Intermediate 844 (44.8) 582 (40.6) High 94 (5.0) 70 (4.9) PE prevalence according to CP Low CP 70 (7.4) 69 (8.8) Intermediate CP 285 (33.8) 206 (35.4) High CP 64 (68.1) 50 (71.4) CP, clinical probability; PE, pulmonary embolism. Data are presented as numbers (%). (n = 1433) P value 0.048 Table 5 Contribution of each diagnostic test to rule in and to rule out PE experienced an objectively confirmed recurrent VTE event during follow-up, corresponding to an odds ratio of 2.2 (1.0 4.9; P = 0.045). Of note, after excluding women with hormonal risk factors, i.e. oral contraception(n = 204), hormone replacement therapy (n = 157) and women who were in the post-partum period (n = 28), the prevalence of PE in women was 21.3%, and there were no relevant differences with the data obtained from the analysis including all women. Discussion (n = 1486) PE ruled out, n 1159 824 (n = 1090) P value Negative DD in non high CP 435 (37.5) 353 (42.8) Negative CT 706 (60.9) 456 (55.3) 0.03 Other negative tests (V/Q scan, PA) 18 (1.6) 15 (1.8) PE ruled in, n 327 266 Positive proximal CUS in 107 (32.7) 115 (43.2) either positive DD or high CP patients Positive CT 213 (65.1) 145 (54.5) 0.04 Other positive tests (V/Q scan, PA) 7 (0.5) 6 (0.6) Data are presented as numbers (%). Among outpatients with suspected PE, symptoms and clinical signs differ slightly between genders (Table 1). Regarding clinical signs and symptoms, our data suggest that women might have a higher awareness of VTE disease (Table 1) and a lower threshold for reporting PE-related symptoms. This could account for the higher prevalence of reported anxiety, shortness of breath and calf pain. It could also explain the less frequent association of these symptoms with confirmed PE in women. Contrarily, men seem to have a lower awareness of VTE disease which might explain why signs suggestive of PE and/or thrombosis are less frequently reported but are more frequently associated with a positive diagnosis of PE in men. The prevalence of several risk factors also differs between women and men with PE (Tables 1 and 2). Interestingly, women with and without PE more often report a history of VTE in their family. One possible explanation could be a better recall of events or a greater interest in their family membersõ medical condition in women than in men. As displayed in Table 2, only history of previous VTE and calf pain have a higher impact on the likelihood of PE in men than in women. Few other studies have assessed the differences in clinical presentation of PE between women and men. Analysis of data from the International Cooperative Pulmonary Embolism Registry (ICOPER), a cohort of 1454 consecutive patients with PE, showed similar prevalence of risk factors in both genders. Chest pain and hemoptysis were

Gender and presentation of PE 697 significantly more frequent in men and dyspnea more frequent in women. In our study, hemoptysis was more frequent in men than in women with PE but also without PE, and therefore influenced the likelihood ratio of PE similarly in both genders. The same finding applies to dyspnea, being more frequent in our study in women with and without PE, with a similar impact on likelihood of PE in both genders [18]. Another study compared differences between 496 women and 406 men with suspected PE from the PIOPED trial (the Prospective Investigation of Pulmonary Embolism Diagnosis). Prevalence of PE was higher in men than women (34% vs. 25%) mainly because of increased prevalence of PE in men under 50 years not explained by differences in risk factors. Regarding presenting symptoms and signs, hemoptysis was again more frequent in men (21% vs. 10%; P < 0.02) as well as leg swelling (36% vs. 24%; P < 0.04) [19]. In summary, as recently reported for DVT [8], clinical presentation of PE shows some slight differences between women and men. Although these differences might be important when assessing pretest probability of PE, each individual symptom has a limited value in the diagnosis of PE, both in women and in men [20]. This suggests that an overall clinical assessment using a validated score remains a highly valuable way to assess clinical probability of PE, and this point will be further discussed. In the study by Andreou et al.[8], among patients referred for suspicion of DVT, the prevalence of DVT was higher in menthaninwomen(14.4%vs.9.4%;p = 0.001), whereas prevalence of PE was similar in both genders (23.1% vs. 22.3%; P = 0.55) in our study. The distribution in different pretest probability groups did not differ between genders in both studies. However, in Andreou et al. study [8], prevalence of DVT in the low and intermediate probability groups was significantly higher in men than women (6.9% vs. 3.5%, P = 0.025 and 16.8% vs. 8.7%, P = 0.004, respectively), whereas it was similar in the high probability group (44% vs. 40.2%, P = 0.59). In our analysis, the overall prevalence of PE was similar in both genders, as well as the prevalence in each pretest probability group using either the Geneva score or the Wells score (Table 4). Therefore, although the accuracy of pretest probability assessment for DVT seems to be lower for men than for women in those with low to intermediate risk, this does not seem to be the case for PE. This important finding in our study confirms the fact that in spite of the slight clinical differences in presentation of PE between genders, clinical probability scores perform equally well in the two genders. The performance of diagnostic tests according to gender was previously analyzed by Stein et al. [21] using data from the Prospective Investigation of Pulmonary Embolism Diagnosis II (PIOPED II), a study designed to investigate the accuracy of multidetector CT pulmonary angiography alone or combined with CT venography for the diagnosis of PE. Overall, there was no significant difference in the sensitivity and specificity of CT pulmonary angiography alone or combined with CT venography, and no difference in the proportion of conclusive test results between women and men [22]. The only exception was a higher specificity of CT pulmonary angiography in women, but specificity was high in both genders (97% vs. 93%; P = 0.015). The performance of these tests in diagnosing PE was therefore considered equivalent in women and men [22]. An interesting finding in our study is the greater diagnostic yield of noninvasive tests such as D-dimer and CUS, for ruling out or ruling in PE, in men than in women (Table 5). This is reflected by a larger proportion of women necessitating a CT to achieve a final diagnosis (64% vs. 57%, P = 0.001). As the diagnostic algorithms and the prevalence of PE were the same for women and men in the studies used for our analysis, the lower prevalence of proximal DVT in women with suspected PE cannot be accounted for by a lower threshold for performing CUS in women. The higher proportion of proximal DVT in men than women with suspected PE is consistent with the fact that calf pain is more strongly associated with PE in men than in women. Finally, assessment of severity of VTE episodes is interesting in regard to the difference in recurrence rates of VTE between genders previously reported in several studies. Recent robust evidence that gender influences the risk of recurrence of VTE raises the hypothesis that some VTE characteristics may differ between women and men [5,6]. Indeed, one could hypothesize that greater severity of the first episode in men could partly account for a greater risk for recurrent events. For DVT, the findings by Andreou et al. [8] did not support this hypothesis, as greater extent of DVT was found in women than in men. Although our data did not show any difference between genders in the most proximal level of the pulmonary tree affected, PE was more often associated with concomitant proximal DVT in men than in women. We could not find any valuable explanation to the greater prevalence of proximal DVT in men than in women with PE. Indeed, the only associated disorders that were more prevalent in men than in women with PE were cancer and COPD. Obviously, cancer is known to be a major risk factor for VTE. However, despite the higher prevalence of cancer in men, this had no effect on the prevalence of PE which was similar in both genders. A specifically increased risk of DVT only due to cancer seems quite unlikely, except in cases of local compression in the pelvis. However, < 3% of our patients had invasive pelvic neoplasia. Information about the nature of previous VTE episodes would have been of help as we know that DVT tends to recur as DVT and PE tends to recur as PE. Unfortunately, this information was not available, and Ôprevious VTEÕ was mentioned as a whole in the three studies. Associated presence of proximal DVT in patients with confirmed PE could indicate a higher residual clot burden in the legs and partly account for the higher VTE recurrence rate during the initial 3 months of follow-up observed in our study. This important finding supports the hypothesis that greater severity of VTE episodes in men (as assessed by a higher clot burden) contributes to the higher VTE recurrence rate in men during the initial period of anticoagulation as well as after discontinuation of anticoagulation as previously reported in the literature. Interestingly, the presence of DVT in patients with PE has also been shown to be

698 H. Robert-Ebadi et al associated with a larger PE extent [23] and a poorer prognosis [24]. Conclusion Although some differences exist between women and men in the clinical presentation of PE, these differences can be considered marginal. Only history of previous VTE and calf pain have a higher influence on likelihood of PE in men than in women. The overall prevalence of PE and the prevalence for each probability group are similar in both genders. Importantly, the lack of gender difference in the prevalence of PE in all pretest probability groups confirms that clinical prediction scores perform equally in women and men. Finally, the initial hypothesis that some differences in presentation of PE could explain different recurrence rates between genders is confirmed. Indeed, the increased prevalence of PE-associated proximal DVT in men possibly points to a higher clot burden and could partly account for the higher recurrence rate in men. Disclosure of Conflict of interests The authors state that they have no conflict of interest. References 1 White RH. The epidemiology of venous thromboembolism. Circulation 2003; 107: I4 8. 2 Kniffin WD Jr, Baron JA, Barrett J, Birkmeyer JD, Anderson FA Jr. The epidemiology of diagnosed pulmonary embolism and deep venous thrombosis in the elderly. Arch Intern Med 1994; 154: 861 6. 3 Silverstein MD, Heit JA, Mohr DN, Petterson TM, OÕFallon WM, Melton LJ III. Trends in the incidence of deep vein thrombosis and pulmonary embolism: a 25-year population-based study. Arch Intern Med 1998; 158: 585 93. 4 Moores L, Bilello KL, Murin S. Sex and gender issues and venous thromboembolism. Clin Chest Med 2004; 25: 281 97. 5 McRae S, Tran H, Schulman S, Ginsberg J, Kearon C. Effect of patientõs sex on risk of recurrent venous thromboembolism: a metaanalysis. Lancet 2006; 368: 371 8. 6 Kyrle PA, Minar E, Bialonczyk C, Hirschl M, Weltermann A, Eichinger S. The risk of recurrent venous thromboembolism in men and women. NEnglJMed2004; 350: 2558 63. 7 Baglin T, Luddington R, Brown K, Baglin C. High risk of recurrent venous thromboembolism in men. JThrombHaemost2004; 2: 2152 5. 8 Andreou ER, Koru-Sengul T, Linkins L, Bates SM, Ginsberg JS, Kearon C. Differences in clinical presentation of deep vein thrombosis in men and women. J Thromb Haemost 2008; 6: 1713 9. 9 WellsPS,GinsbergJS,AndersonDR,KearonC,GentM,TurpieAG, Bormanis J, Weitz J, Chamberlain M, Bowie D, Barnes D, Hirsh J. Use of a clinical model for safe management of patients with suspected pulmonary embolism. Ann Intern Med 1998; 129: 997 1005. 10 Perrier A, Bounameaux H, Morabia A, de Moerloose P, Slosman D, Didier D, Unger PF, Junod A. Diagnosis of pulmonary embolism by a decision analysis-based strategy including clinical probability, D-dimer levels, and ultrasonography: a management study. Arch Intern Med 1996; 156: 531 6. 11 Stein PD, Sostman HD, Bounameaux H, Buller HR, Chenevert TL, Dalen JE, Goodman LR, Gottschalk A, Hull RD, Leeper KV Jr., Pistolesi M, Raskob GE, Wells PS, Woodard PK. Challenges in the diagnosis of acute pulmonary embolism. Am J Med 2008; 121: 565 71. 12 Perrier A, Roy PM, Aujesky D, Chagnon I, Howarth N, Gourdier AL, Leftheriotis G, Barghouth G, Cornuz J, Hayoz D, Bounameaux H. Diagnosing pulmonary embolism in outpatients with clinical assessment, D-dimer measurement, venous ultrasound, and helical computed tomography: a multicenter management study. Am J Med 2004; 116: 291 9. 13 Perrier A, Roy PM, Sanchez O, Le Gal G, Meyer G, Gourdier AL, Furber A, Revel MP, Howarth N, Davido A, Bounameaux H. Multidetector-row computed tomography in suspected pulmonary embolism. NEnglJMed2005; 352: 1760 8. 14 Righini M, Le Gal G, Aujesky D, Roy PM, Sanchez O, Verschuren F, Rutschmann O, Nonent M, Cornuz J, Thys F, Le Manach CP, Revel MP, Poletti PA, Meyer G, Mottier D, Perneger T, Bounameaux H, Perrier A. Diagnosis of pulmonary embolism by multidetector CT alone or combined with venous ultrasonography of the leg: a randomised non-inferiority trial. Lancet 2008; 371: 1343 52. 15 Wicki J, Perneger TV, Junod AF, Bounameaux H, Perrier A. Assessing clinical probability of pulmonary embolism in the emergency ward: a simple score. Arch Intern Med 2001; 161: 92 7. 16 Le Gal G, Righini M, Roy PM, Sanchez O, Aujesky D, Bounameaux H, Perrier A. Prediction of pulmonary embolism in the emergency department: the revised Geneva score. AnnInternMed2006; 144: 165 71. 17 Wells PS, Anderson DR, Rodger M, Ginsberg JS, Kearon C, Gent M, Turpie AG, Bormanis J, Weitz J, Chamberlain M, Bowie D, Barnes D, Hirsh J. Derivation of a simple clinical model to categorize patients probability of pulmonary embolism: increasing the models utility with the SimpliRED D-dimer. Thromb Haemost 2000; 83: 416 20. 18 McHughKB,VisaniL,DeRosaM,CovezzoliA,RossiE,Goldhaber SZ. Gender comparisons in pulmonary embolism (results from the International Cooperative Pulmonary Embolism Registry [ICOPER]). Am J Cardiol 2002; 89: 616 9. 19 Quinn DA, Thompson BT, Terrin ML, Thrall JH, Athanasoulis CA, McKusick KA, Stein PD, Hales CA. A prospective investigation of pulmonary embolism in women and men. JAMA 1992; 268: 1689 96. 20 West J, Goodacre S, Sampson F. The value of clinical features in the diagnosis of acute pulmonary embolism: systematic review and metaanalysis. QJM 2007; 100: 763 9. 21 Stein PD, Fowler SE, Goodman LR, Gottschalk A, Hales CA, Hull RD, Leeper KV Jr, Popovich J Jr., Quinn DA, Sos TA, Sostman HD, Tapson VF, Wakefield TW, Weg JG, Woodard PK. Multidetector computed tomography for acute pulmonary embolism. N Engl J Med 2006; 354: 2317 27. 22 Stein PD, Beemath A, Quinn DA, Olson RE, Goodman LR, GottschalkA,HalesCA,HullRD,LeeperKVJr,SostmanHD,WegJG, Woodard PK. Usefulness of multidetector spiral computed tomography according to age and gender for diagnosis of acute pulmonary embolism. Am J Cardiol 2007; 99: 1303 5. 23 Galle C, Papazyan JP, Miron MJ, Slosman D, Bounameaux H, Perrier A. Prediction of pulmonary embolism extent by clinical findings, D- dimer level and deep vein thrombosis shown by ultrasound. Thromb Haemost 2001; 86: 1156 60. 24 Wicki J, Perrier A, Perneger TV, Bounameaux H, Junod AF. Predicting adverse outcome in patients with acute pulmonary embolism: a risk score. Thromb Haemost 2000; 84: 548 52.