Cerebral Venous and Sinus Thrombosis in Women
Background and Purpose— Little is known about the gender-specific manifestations of cerebral venous and sinus thrombosis, a disease that is much more common in women than men.
Methods— We used data of the International Study on Cerebral Vein and Dural sinus Thrombosis (ISCVT), a multicenter prospective observational study, to analyze gender-specific differences in clinical presentation, etiology, and outcome of cerebral venous thrombosis.
Results— Four hundred sixty-five of a total of 624 patients were women (75%). Women were significantly younger, had less often a chronic onset of symptoms, and had more often headache at presentation. There were no gender differences in ancillary investigations or treatment. A gender-specific risk factor (oral contraceptives, pregnancy, puerperium, and hormonal replacement therapy) was present in 65% of women. Women had a better prognosis than men (complete recovery 81% versus 71%l P=0.01), which was entirely due to a better outcome in female patients with gender-specific risk factors. Women without gender-specific risk factors are similar to men in clinical presentation, risk factor profile, and outcome. Logistic regression analysis confirmed that the absence of gender-specific risk factors is a strong and independent predictor of poor outcome in women with sinus thrombosis (OR, 3.7; CI, 1.9 to 7.4).
Conclusions— Our study identified important differences between women and men in presentation, course, and risk factors of cerebral venous and sinus thrombosis and showed that women with a gender-specific risk factor have a much better prognosis than other patients.
Cerebral venous and sinus thrombosis (CVST) is a relatively rare subtype of stroke.1,2 In young to middle-aged adults, CVST is much more common in women than men with a ratio of approximately 3 to one.3 This skewed gender ratio is usually attributed to gender-specific risk factors, especially oral contraceptives, and to a lesser extent pregnancy, puerperium, and hormone replacement therapy.1 Case-control studies showed a significant association between use of oral contraceptives and CVST,4,5 which was confirmed in a recent meta-analysis (pooled OR, 5.59).6 Furthermore, in age groups that lack these gender-specific risk factors such as children7 and elderly patients,8 there is no gender predilection. Additional evidence comes from the fact that epidemiological studies performed before the widespread use of oral contraceptives do not show a difference in gender distribution.9 Finally, a single-center retrospective study showed a shift in gender ratio over time toward women that parallels an increase in use of oral contraceptives.10
Studies in arterial stroke have identified important differences in clinical manifestations and outcome between men and women.11,12 After adjustment for possible confounders, women were found to receive less ancillary investigations and have a worse prognosis than men. Furthermore, there are indications that women may respond differently to secondary prophylaxis treatment of arterial stroke than men.13 Several potential explanations for these differences have been offered, but no definitive conclusions can be drawn from the available data. A National Institute of Neurological Diseases and Stroke-sponsored multidisciplinary working group therefore recommended additional prospective studies regarding gender differences in stroke.14
Little is known about the differences between men and women regarding clinical manifestations and outcome of CVST. The scarce studies that do exist are mostly retrospective or comprise a small number of patients.5,15,16 The International Study on Cerebral Venous and dural sinus Thrombosis (ISCVT) is, to date, the largest prospective cohort study on CVST.3 Women comprise approximately three fourths of patients included in this cohort. To elucidate the specific manifestations of CVST in women, we have analyzed the ISCVT data based on gender, focusing on clinical manifestations, etiology, and outcome.
The organization of the ISCVT has been described in detail previously.3 Briefly, the ISCVT is a prospective international observational cohort study that included 624 consecutive adult patients with symptomatic CVST. In total, 89 centers from 21 countries participated. A complete list of all participating centers is available in a previous publication.3 The inclusion ran from May 1998 until May 2001. Diagnosis of CVST was confirmed by MR venography, conventional angiography, CT venography, surgery, or autopsy in all patients. A list of risk factors for CVST was attached to the inclusion form to assist investigators with the etiologic workup. Genetic thrombophilia screening was recommended for protein S and C deficiency, antithrombin deficiency, prothrombin mutation, factor V Leiden, and methylenetetrahydrofolate reductase mutation.
Follow-up visits were performed at 6 months, 12 months, and yearly thereafter and were predominantly done by direct interview and observation by the local investigators. During follow-up, disability, death, and other relevant symptoms and events were recorded. Disability was classified according to the modified Rankin scale (mRS, 0=complete recovery, 6=dead). In the current study, we used the score on mRS at 6 months as the primary outcome measure. For patients who missed the 6-month evaluation, we imputed the score of either the mRS at discharge or at 1-year follow-up, whichever was worse. The 3 outcome measures that are provided are: complete recovery (defined as a mRS score of 0 or 1), death or dependency (mRS 3 to 6), and mortality (mRS 6).
In the primary analysis, we compared demographics, clinical manifestations, etiology, and outcome between women and men. Because we hypothesized that clinical characteristics and outcome are different between female patients with gender-specific risk factors (GSRF) and those without these risk factors, we performed a second analysis in which we subdivided female patients into 3 subgroups based on the presence or absence of these risk factors. The following risk factors were defined as gender-specific: use of oral contraceptives, pregnancy, puerperium, and hormone replacement therapy.
Categorical data were analyzed with the χ2 test. Because continuous data were not normally distributed, we used the Mann-Whitney (for comparison of 2 groups) or Kruskal-Wallis test (for comparison of 3 groups). For each set of analyses (eg, demographics, symptoms, etiology, or outcome), uncorrected probability values are provided. In addition, we indicate in the tables if the probability values remain significant after Bonferroni correction for multiple testing. We used a stepwise backward logistic regression analysis to examine predictors of death or dependency at 6-month follow-up. The following variables were entered in to the multivariate analysis: age, thrombosis of the deep venous system, thrombosis of the superior sagittal sinus, mental status disorder, any paresis, coma, intracranial hemorrhage, any focal lesion on CT, central nervous system infection, and any malignancy. These variables were selected because they were significantly associated with poor outcome in a previous analysis of the entire ISCVT cohort.3 We added “absence of a GSRF” to the regression analysis as one of our hypotheses of this study. We calculated ORs and 95% CIs for the retained variables. All data were analyzed with SPSS 16.0.
Baseline Characteristics in Men and Women
Demographics and clinical characteristics at baseline of women and men are depicted in Table 1. Of the 624 included patients, 465 were women (75%). Women were significantly younger than men at the time of diagnosis (median age, 34 versus 42, respectively). There was no significant difference between percentages of patients from developing countries (20% versus 15%). Women less often had a chronic onset of symptoms (5% versus 13%) and the interval between onset of symptoms and admission was shorter in women compared with men. Duration of hospital admission was also shorter for female patients.
Headache was significantly more common in women compared with men (91% versus 82%), but frequency of other clinical symptoms and signs at baseline were similar. Distribution of occluded sinuses and veins was also similar in both groups. The superior sagittal sinus was occluded in 62% of women and 64% of men. The deep venous system was involved in 12% and 8%, respectively, whereas in both groups, 50% of patients had more than one occluded sinus (all differences not significant). Treatment regimens also did not differ between women and men. Approximately 83% of all patients, regardless of gender, received heparin in a therapeutic dose (Table 2). The majority of remaining patients received prophylactic dose heparin or antiplatelet drugs with no gender-related differences (data not shown).
Etiology and Outcome
Risk factors for CVST are shown in Table 2. Three hundred one of 465 women had a GSRF for sinus thrombosis (65%). Oral contraceptive users comprised the largest group (213 patients, 46% of all women). Seventy-seven women were pregnant or had recently given birth (17%) and 14 women used hormone replacement therapy (3%).
Etiologic workup was complete in approximately 80% of patients (no gender difference). Seventy-five percent of the participating centers performed a complete genetic screening for thrombophilia in all patients. Overall, frequency of genetic thrombophilia was similar in women and men (22% versus 25%; P=0.4). The prevalence of the specific prothrombotic mutations as well as the total number of patients that were screened for each mutation is provided in Supplemental Table I, available online at http://stroke.ahajournals.org. With the exception of the antithrombin deficiency, which was more common in men (0.3% versus 2.7%; P=0.02), there were no significant differences between women and men.
Infections were less common among women (10% versus 21%; P<0.001), mainly due to a lower frequency of ear, nose, and throat infections. Malignancies and mechanical causes, including trauma, were also less frequent in women. Finally, women more often had multiple risk factors (47% versus 33%).
The 6-month follow-up rate was high with data available for 93% of patients. For missing cases (14 men, 31 women), the mRS score was imputed as described in the “Methods” section. A significantly larger proportion of women had a complete recovery (81% versus 71%; P=0.01). Accordingly, death or dependency was less frequent among women (12% versus 20%; P=0.008). All-cause mortality was also lower (6% versus 10%), but this did not reach statistical significance (P=0.12). Omission of the missing values yielded essentially the same results. Outcome at last follow-up (median follow-up, 16 months) also showed similar results (death or dependency 11% versus 20%; complete recovery 81% versus 74%).
Role of GSRF in CVST
Female patients were divided into 3 subgroups. There were 143 women with only GSRF and no other risk factor (Group A1), 158 women with both GSRF and another risk factor (Group A2), and 164 women without GSRF (Group B). The results of the subgroup analysis are shown in Table 3. For reference purposes, the results of men are also shown, but statistical testing is between female patients only.
The 2 groups of women with GSRF are similar to each other in most respects but differ considerably from women without GSRF. Women with GSRF are younger, had shorter hospital admissions, and were more often included from a country with low or middle-low gross national income according to the World Bank definition (www.worldbank.org). Furthermore, headache was almost uniformly present in women with GSRF (97% and 94%) but less frequent in women without GSRF. There was no difference in frequency of heparin treatment between the subgroups of female patients. Women without GSRF did receive steroids more often (31%) than patients with GSRF (both 20%; P=0.03).
Irrespective of the presence or absence of other risk factors, women with GSRF had significantly better outcomes than those without GSRF; 85% had fully recovered at the 6-month follow-up visit. Mortality and “death or dependency” were less frequent in women with GSRF. A post hoc analysis in which all 3 groups were compared with each other confirmed that the difference in outcome is solely caused by a worse outcome in women without a GSRF. Differences in outcome at the last follow-up visit were similar (data not shown). The outcome in all categories of the mRS at 6 months for the 4 groups of patients is shown in the Figure.
Because the subdivision of female patients is based on risk factors, comparison of etiologic factors between the female subgroups is inapplicable, but comparison of the risk factors of women without GSRF (Group B) and men is valid (Table 4). This comparison shows highly similar frequencies with the exception of a higher rate of infections in men (21% versus 15%; P=0.1).
Logistic regression analysis confirmed that the absence of a GSRF is a strong, independent predictor of poor outcome among women (OR, 3.7; P<0.001; Table 5). Other risk factors that were associated with death or dependency are intracranial hemorrhage, mental status disorder, thrombosis of the deep venous system, coma, and central nervous system infection. Age and malignancy, which were independently associated with poor outcome in the overall analysis of the ISCVT population,3 were not retained in the current model. When both predictors were forced into the model, the absence of a GSRF remained significantly associated with poor outcome (OR, 3.3; P=0.003). Age and malignancy were not independently associated with poor outcome (P=0.9 and P=0.1, respectively).
Our study shows important differences between women and men with CVST and between women with and without GSRF, mostly the use of oral contraceptives. The most conspicuous finding is the far better prognosis of women, which is exclusively caused by a substantially better outcome in female patients with GSRF. Previous retrospective studies reported that CVST in pregnant women has a relatively favorable prognosis.15,17 These results are now confirmed in a large, prospective cohort study and extended to women with all GSRF.
The ISCVT is, to date, the largest prospective cohort study on CVST.3 Participating centers included hospitals from all over the world and investigators committed themselves to include data on consecutive patients. Furthermore, CVST was confirmed in all patients using standard radiological techniques and completeness of follow-up data was high. For these reasons, the data presented in this study are likely to be representative of CVST in general. It must be noted, however, that the ISCVT included only adult patients. Generalization to pediatric patients must therefore be cautioned. Furthermore, few patients from the African and Asian continents were included. Risk factor profiles may be different in patients from these continents, including a higher rate of infection- and pregnancy-related CVST and lower rates of oral contraceptive-related CVST.18
An obvious explanation for the better prognosis in women with GSRF would be that they are younger. In the overall analysis of the ISCVT data (including male patients), age was indeed an independent, although relatively weak, predictor of outcome (hazard ratio, 2.0; CI, 1.2 to 3.3).3 However, in our present multivariate analysis of female patients only, age was not a significant predictor. In this model, the absence of a GSRF both strongly and independently predicted poor outcome (OR, 3.7; CI, 1.9 to 7.4) in women.
Another reason for the better prognosis of female patients with GSRF could be that they are in better general condition. Women using oral contraceptives are checked by doctors in some countries, and women with known thrombotic risk factors or other medical problems may avoid oral contraceptives. Also, pregnancy might be less likely or be delayed in women with health problems. These explanations are supported by our finding that women without GSRF have a risk factor profile that is almost identical to that of men with the exception of infections. Finally, the better outcome in women could be the result of the fact that women seek medical attention sooner than men as might be indicated by the shorter time between onset of symptoms and admission. The fact that men more often have a chronic onset of symptoms might also support this hypothesis, although, based on our data, we cannot exclude the possibility that the more insidious onset in women is the result of gender-specific pathophysiological differences. Furthermore, we found no evidence that women were in a better clinical condition at presentation compared with men (Table 1).
Our study did not demonstrate any gender-specific differences regarding diagnosis or treatment of CVST, which also could have explained the differences in outcome. Approximately 80% of both male and female patients received full-dose heparin, which is considered standard treatment.19 Similarly, we found no difference in diagnostic procedures. In both genders, 80% of patients received complete etiologic workup for thrombotic risk factors. This is an important finding, because previous studies have reported gender-specific differences in management of both ischemic heart disease20 and arterial stroke.11 In those conditions, women underwent less ancillary investigations and were less likely to receive aggressive therapy compared with men. Nevertheless, based on our data, we cannot exclude the possibility of underdiagnosis in female patients, because by definition, the ISCVT included only patients who underwent CT, MRI, or angiography.3
Interestingly, women with GSRF more often had headache at presentation as compared with women in Group B and men. Headache in CVST can be caused by several mechanisms: increased intracranial pressure, hemorrhagic infarcts, small sulcal hemorrhages, and innervation of pain-sensitive fibers in the dura mater (eg, by distension of the sinus walls).21 We found no indication of more extensive thrombosis or a different pattern of thrombosis in women with GSRF. Frequency of baseline intracranial hemorrhages and infarcts was also similar in subgroups of female patients. Furthermore, the fact that we found a similar rate of papilledema in all groups might indicate that there was no difference in frequency of intracranial hypertension. However, because lumbar puncture was not routinely performed, an increased frequency of intracranial hypertension in women with GSRF cannot be excluded. Alternatively, the higher prevalence of headache in female patients could be the result of a mechanism independent of CVST. It has been shown that women with arterial stroke also more often have headache than men.22 Finally, the higher rate of headache in women with GSRF could be the effect of age. Only 65% of elderly patients with CVST (>65 years) reported headache at presentation as shown in a previous study.8 The women in Group B, who less often had headache, were indeed older than women with GSRF.
In approximately half of all female patients, more than one thrombotic risk factor was identified. It is generally accepted that thrombosis is a multicausal disease, which requires various risk factors to develop.23 Previous studies have shown that use of oral contraceptives and a genetic prothrombotic condition increase the risk of thrombosis in a multiplicative way,4,24 and a similar synergistic effect has been observed in CVST.6 Workup for thrombophilia is therefore advisable, even in the presence of a GSRF. Although the acute management of CVT is not altered by the presence of thrombophilia, it may be important at a later stage for counseling and management in subsequent high-risk situations such as pregnancy.
In conclusion, we have indentified important differences between women and men in the presentation, course, and risk factors of cerebral venous and sinus thrombosis and showed that women with a GSRF have a much better prognosis than other patients.
We thank all investigators who contributed to the ISCVT study. Their names and institutions are listed in a previous publication.3
Sources of Funding
This study was supported by PRAXIS grant C/SAU/10248/1998 from the “Fundação para a Ciência e Tecnologia.” J.M.C. is supported by “The Netherlands Organisation for Scientific Research” (NWO), grant number 021.001.045.
- Received November 26, 2008.
- Revision received February 16, 2009.
- Accepted March 19, 2009.
Ferro JM, Canhao P, Stam J, Bousser MG, Barinagarrementeria F. Prognosis of cerebral vein and dural sinus thrombosis: results of the International Study on Cerebral Vein and Dural Sinus Thrombosis (ISCVT). Stroke. 2004; 35: 664–670.
de Bruijn SF, Stam J, Koopman MM, Vandenbroucke JP. Case-control study of risk of cerebral sinus thrombosis in oral contraceptive users and in [correction of who are] carriers of hereditary prothrombotic conditions. The Cerebral Venous Sinus Thrombosis Study Group. BMJ. 1998; 316: 589–592.
Dentali F, Crowther M, Ageno W. Thrombophilic abnormalities, oral contraceptives, and risk of cerebral vein thrombosis: a meta-analysis. Blood. 2006; 107: 2766–2773.
Ferro JM, Canhao P, Bousser MG, Stam J, Barinagarrementeria F. Cerebral vein and dural sinus thrombosis in elderly patients. Stroke. 2005; 36: 1927–1932.
Di CA, Lamassa M, Baldereschi M, Pracucci G, Basile AM, Wolfe CD, Giroud M, Rudd A, Ghetti A, Inzitari D. Sex differences in the clinical presentation, resource use, and 3-month outcome of acute stroke in Europe: data from a multicenter multinational hospital-based registry. Stroke. 2003; 34: 1114–1119.
Glader EL, Stegmayr B, Norrving B, Terent A, Hulter-Asberg K, Wester PO, Asplund K. Sex differences in management and outcome after stroke: a Swedish national perspective. Stroke. 2003; 34: 1970–1975.
Bushnell CD, Hurn P, Colton C, Miller VM, del ZG, Elkind MS, Stern B, Herrington D, Ford-Lynch G, Gorelick P, James A, Brown CM, Choi E, Bray P, Newby LK, Goldstein LB, Simpkins J. Advancing the study of stroke in women: summary and recommendations for future research from an NINDS-Sponsored Multidisciplinary Working Group. Stroke. 2006; 37: 2387–2399.
Cantu C, Barinagarrementeria F. Cerebral venous thrombosis associated with pregnancy and puerperium. Review of 67 cases. Stroke. 1993; 24: 1880–1884.
Estanol B, Rodriguez A, Conte G, Aleman JM, Loyo M, Pizzuto J. Intracranial venous thrombosis in young women. Stroke. 1979; 10: 680–684.
Khealani BA, Wasay M, Saadah M, Sultana E, Mustafa S, Khan FS, Kamal AK. Cerebral venous thrombosis: a descriptive multicenter study of patients in Pakistan and Middle East. Stroke. 2008; 39: 2707–2711.
Tentschert S, Wimmer R, Greisenegger S, Lang W, Lalouschek W. Headache at stroke onset in 2196 patients with ischemic stroke or transient ischemic attack. Stroke. 2005; 36: e1–e3.