点击显示 收起
the Department of Neurosciences and Mental Health (P.C., J.M.F.), Hospital Santa Maria, Lisbon, Portugal
the Department of Neurology (A.G.L.), University Hospital, Lund, Sweden
the Department of Neurology (M.-G.B.), Hpital Lariboisière, Paris, France
the Department of Neurology (J.S.), Academic Medical Centre, Amsterdam, The Netherlands
the Department of Neurology (F.B.), Instituto Nacional de Neurologia y Neurocirurgia, México City, México.
Abstract
Background and Purpose— The causes of death of patients with cerebral venous thrombosis (CVT) have not been systematically addressed in previous studies. We aimed to analyze the causes and predictors of death during the acute phase of CVT in the International Study on Cerebral Vein and Dural Sinus Thrombosis (ISCVT) to identify preventable or treatable causes.
Methods— ISCVT is a multinational, prospective, observational study including 624 patients with CVT occurring between May 1998 and May 2001, in which 27 patients (4.3%) died during the acute phase, 21 (3.4%) within 30 days from symptom onset. Inclusion forms and a questionnaire assessing the causes of death were analyzed. A logistic regression analysis was performed to identify the predictors of death within 30 days from symptom onset of CVT.
Results— Median time between onset of symptoms and death was 13 days and between diagnosis and death, 5 days. Causes of death were mainly transtentorial herniation due to a unilateral focal mass effect (10 patients) or to diffuse edema and multiple parenchymal lesions (10 patients). Independent predictors of death were coma (odds ratio [OR], 8.8; 95% confidence interval [CI], 2.8 to 27.7), mental disturbance (OR, 2.5; 95% CI 0.9 to 7.3), deep CVT thrombosis (OR, 8.5; 95% CI, 2.6 to 27.8), right intracerebral hemorrhage (OR, 3.4; 95% CI, 1.1 to 10.6), and posterior fossa lesion (OR, 6.5; 95% CI, 1.3 to 31.7). Worsening of previous focal or de novo focal deficits increased the risk of death.
Conclusions— The main causes of acute death were neurologic, the most frequent mechanism being transtentorial herniation.
Key Words: cerebral veins cerebrovascular circulation death models, statistical prognosis sinus thrombosis
Introduction
Cerebral vein and dural sinus thrombosis (CVT) is an infrequent stroke type often described as having an unpredictable outcome.1 In the past, CVT was diagnosed almost exclusively at autopsy and thought to be almost always fatal.2–4 In early angiographic series, mortality ranged between 30% and 50%.5 In recent series, widely discrepant proportions of case fatality ranging from 4% to 33% were reported.6–11 In the International Study on Cerebral Vein and Dural Sinus Thrombosis (ISCVT), 4.3% of patients died during the acute phase of CVT and 3.4% within 30 days from symptom onset.12 In the ISCVT, a small percentage of patients remained dependent, and acute death was a determinant in the outcome "death or dependency" at the end of the follow-up, making it worthwhile to perform a separate analysis of acute death. Furthermore, the cause of death was seldom evaluated, especially in larger samples of patients. If causes of death could be identified, specific treatments could be planned to prevent fatality. The objectives of the present study were to (1) analyze case fatality during the course of CVT in the ISCVT; (2) describe the main causes of death; and (3) identify predictors of death.
Methods
Study Population
This study comprised patients with proven CVT who were included in the ISCVT, described in detail elsewhere.12 Briefly, the ISCVT is a prospective, multinational observational study that included 624 consecutive patients (age >15 years) with symptomatic CVT occurring between May 1998 and May 2001. The diagnosis of CVT was confirmed by conventional angiography, computed tomography venography, magnetic resonance imaging (MRI) combined with MR venography, or at surgery or autopsy, according to established diagnostic criteria.1
Data Collection
The following data were obtained in the ISCVT study: demographics; dates of symptom onset; hospital admission and diagnosis; symptoms and signs from onset to diagnosis; Glasgow Coma Scale (GCS) score on admission; location of the thrombus; number, size, and location of parenchymal lesions; risk factors; type of worsening (decreased consciousness, new focal deficits, worsening of previous focal deficit, seizures, other); treatment; and outcome. Presenting syndromes were dichotomized as isolated intracranial hypertension (any combination of headache, vomiting, and papilledema with or without visual loss or VI nerve paresis, without other neurologic symptoms or signs), and other presenting syndromes.
In addition, a questionnaire was sent to the investigators for each patient who was reported to have died during the course of CVT. The causes of death were classified as (1) cerebral transtentorial herniation secondary to diffuse edema or multiple bilateral lesions (hematoma or infarct) or to a unilateral focal mass effect (hematoma or infarct)13; (2) pulmonary embolism14; (3) neurogenic pulmonary edema15; (4) generalized status epilepticus; (5) underlying disease; (6) sudden (<1 hour) unwitnessed death; (7) any combination of the above; or (8) other.
Outcome
All patients who died during hospitalization for the inclusion episode of CVT were analyzed, and their causes of death were assessed. Deaths occurring during follow-up were not the subject of the present analysis. We compared the causes of early death (before the median time between the onset of symptoms and death) with those of late death (after the median between the onset of symptoms and death).
For analysis of the predictors of death, the outcome of interest was mortality within 30 days of onset of CVT symptoms. Because the duration of hospitalization was variable, we analyzed follow-up data to identify patients who died after discharge but within 30 days after onset of symptoms.
Statistical Analysis
Descriptive statistics were performed to describe the CVT patients who died. For continuous variables, means, standard deviations, medians, and ranges were calculated. For categorical variables, numbers and percentages for each category were tabulated.
Comparisons were made between CVT patients who died and survivors. Bivariate analysis was performed for the outcome "death within 30 days" with the 2 (with Yates’ correction when necessary) or Fisher exact test for categorical data and with Student’s t test for continuous data. We performed a logistic regression analysis (backward method) and calculated odds ratios (ORs) and 95% confidence intervals (CIs) for the retained variables associated with the outcome "death" in the bivariate analyses (P<0.10). The specificity and sensitivity of the model for prediction of death were calculated. Data were analyzed with SPSS 11.0 for Windows.
Results
Description of Patients
Six hundred twenty-four adult patients were included in the ISCVT from 89 centers in 21 countries. Twenty-seven patients (4.3%) died during the inclusion episode of CVT, 21 (3.4%) within 30 days after symptom onset. None of these deaths occurred after discharge.
Death occurred a median of 13 days (mean, 21.2 days; SD, 24.5) after symptom onset, a median of 5 days (mean, 11.3 days; SD, 15.8) after diagnosis. Concerning patients who died within 30 days from symptom onset, death occurred a median of 9 days (mean, 10.6 days; SD, 6.4) after symptom onset and a median of 4 days (mean, 5.3 days; SD, 4.5) after diagnosis. Table 1 describes baseline characteristics and risk factors for CVT in patients who died during the course of CVT.
No difference was found in the time from onset of symptoms to admission or to the diagnosis of CVT in patients who died compared with those who survived. Mental status disorders (P<0.001), impaired consciousness (P<0.001), and seizures (P=0.002) at admission were more frequent in patients who died, whereas isolated intracranial hypertension syndrome was less frequent (P=0.023). Thrombosis of the superior sagittal sinus (P=0.023), cortical veins (P=0.004), deep cerebral veins (P<0.001), parenchymal lesions (P=0.002), hemorrhagic lesions (P=0.002), right hemorrhagic lesions (P=0.001), and posterior fossa lesion (P=0.004) were more frequent at admission in patients who died. Size of parenchymal lesions was significantly higher in patients who died during the acute phase. Twelve of 25 (48%) had parenchymal lesions >5 cm in their larger diameter, in contrast to 106 of 558 surviving patients (19%; OR, 3.9; 95% CI, 1.8 to 8.9; P<0.001).
All patients who died deteriorated during the course of the inclusion episode (median of 2 days after admission; mean, 6.2; SD, 10.6; Table 2). Among patients who died within 30 days from symptom onset, the following types of worsening were more frequent: altered mental state (P<0.001), worsening of previous focal deficit (P<0.001), and new focal deficit (P<0.001). New lesions on subsequent neuroimaging examinations were more frequent among those who died (P<0.001), either infarct or edema (P=0.007), or hemorrhage (P<0.001).
All but 2 of the 27 patients who died were treated with heparin (Table 2), a similar proportion compared with those who survived. Thrombolytic therapy was administered in 5 patients who deteriorated despite other treatments, locally in the thrombus in 4 patients, and systemically in 1 patient.
Causes of Death
The investigators returned 21 of 27 questionnaires. Two independent investigators adjudicated the cause of death by using the questionnaires and the inclusion form data (Table 2). The most frequent cause of death was transtentorial herniation, due to either a focal mass effect (10 patients) or to multiple lesions and edema (10 patients).
Causes of Early and Late Death
The causes of early death (13 patients) were different from those of late death (14 patients). Early deaths were due to transtentorial herniation because of multiple lesions, diffuse edema (7 patients), and a focal mass effect (6 patients). Late deaths were less frequently due to transtentorial herniation (7 patients): cardiopulmonary arrest in a patient with leukemia; sudden death in a patient with multiple cerebral hemorrhages and respiratory distress; underlying disease in an HIV-infected patient; pulmonary embolism and intracranial hypertension due to diffuse edema in 1 patient who was treated with heparin; sepsis in 2 patients; and unclassified in 1 patient.
Predictive Models of Death
A logistic multivariate analysis identified the following independent variables predicting death within 30 days from symptom onset: seizure, mental status disorder, GCS score <9 at admission, deep CVT, and right hemorrhage and posterior fossa lesions on the initial MRI or computed tomography scan (Table 3).
We added to the previous predictive model the variables corresponding to the different types of worsening except "worsening of consciousness" to avoid including the outcome in the predictor. The variables retained in this model were seizure, mental status disorder, GCS <9 at admission, deep CVT, posterior fossa lesion, worsening of focal signs, or occurrence of new focal signs after admission. (Table 3).
Discussion
ISCVT is the largest prospective series of patients with CVT collected in different centers and countries. The most common cause of death was transtentorial herniation due to a unilateral hemorrhagic lesion or diffuse edema and bilateral lesions. Main predictors of death within 30 days were seizure, mental status disturbances, coma (GCS <9), deep CVT, and right hemorrhage and posterior fossa lesions.
ISCVT reported a lower case fatality than the majority of previous studies, including those performed in specific settings such as pregnancy or puerperium. Table 4 depicts the percentages of death and their 95% CIs in recent case series of CVT with >20 patients.7,11–12,16–29 To decrease potential ascertainment bias in ISCVT, investigators were repeatedly asked to search for cases through the Imaging Department, Intensive Care Unit, and other hospital departments.
Causes of death were not addressed systematically previously. Early autopsy series gave details about the location of occluded veins and sinus and parenchymal lesions, but overall they did not provide the cause of death. Pulmonary embolism, heart disease, cachexia and marasmus, and intracranial lesions were considered the cause of death in autopsy studies.2–5 Many cases were associated with infectious diseases, which are currently less common. We have identified only 1 case of death due to pulmonary embolism, which is less than previously suggested,30 probably because of the generalized treatment with heparin in ISCVT patients. Death could have been due to pulmonary embolism in another patient with sudden death and respiratory distress, although this was not confirmed.
In more recent series of CVT, causes of death were seldom ascertained. In the VENOPORT study, cerebral edema with or without seizures, cerebral anoxia due to seizure, and sudden cardiopulmonary arrest were the major causes of death.11 Other series reported transtentorial herniation due to hemorrhagic infarct,31 intubation accident leading to cardiopulmonary arrest,31 and septic multiorgan failure.21 In the present study, the main causes of death were neurologic, most frequently transtentorial herniation, due to either focal hemorrhagic lesion or multiple lesions with diffuse edema. This distinction may be important when selecting therapy for individual patients. Decompressive craniectomy was recommended for such patients many years ago5; however, this intervention has only rarely been reported in recent years.32 In light of our findings, decompressive craniectomy should be reconsidered for patients with progressive herniation.
Several models predicting the outcome "death or dependency" have been reported,10,24,33 but no predictive model of death has been previously described. The individual time course is highly variable in venous stroke. This partly explains why our predictive model has a low sensitivity. By adding "clinical course after admission" to admission variables, we identified more patients at high risk of death, such as those developing new focal signs or showing a worsening of a previous focal deficit. However, there are still limitations in predicting individual survival outcome.
The results of this study have important implications. First, although CVT has a low case fatality, it is possible to predict some patients who are at increased risk of death. These patients should be closely monitored, and worsening of their clinical condition should be regarded as an indication for more aggressive treatment. Second, given the potential for neurologic recovery after CVT, there is a case for assessing decompressive craniectomy in patients who are deteriorating due to a parenchymal lesion producing a mass effect.
Acknowledgments
This study was supported by PRAXIS grant C/SAU/10248/1998 from the Fundao para a Ciência e Tecnologia. We wish to thank the investigators who participated in the ISCVT (International Study on Cerebral Vein and Dural Sinus Thrombosis). Their names and centers have been listed in a previous publication.12
References
Bousser MG, Russell RR. Cerebral venous thrombosis. In: Warlow CP, Van Gijn J, eds. Major Problems in Neurology. London: WB Saunders; 1997; 33: 27–29.
Garcin R, Pestel M. Thrombophlébites Cérébrales. Paris: Masson et Cie; 1949.
Barnett HJM, Hyland HH. Non-infective intracranial venous thrombosis. Brain. 1953; 76: 36–49.
Kalbag RM, Woolf AL. Cerebral Venous Thrombosis. London: University Press; 1967.
Krayenbuhl H. Cerebral venous and sinus thrombosis. Clin Neurosurg. 1967; 14: 1–24.
Bousser MG, Chiras J, Bories J, Castaigne P. Cerebral venous thrombosis: a review of 38 cases. Stroke. 1985; 16: 199–213.
Cantú C, Barinagarrementeria F. Cerebral venous thrombosis associated with pregnancy and puerperium: review of 67 cases. Stroke. 1993; 24: 1880–1884.
deVeber G, Andrew M, Adams C, Bjornson B, Booth F, Buckley DJ, Camfield CS, David M, Humphreys P, Langevin P, MacDonald EA, Gillett J, Meaney B, Shevell M, Sinclair DB, Yager J; Canadian Pediatric Ischemic Stroke Study Group. Cerebral sinovenous thrombosis in childhood. N Engl J Med. 2001; 345: 417–423.
Einhupl KM, Masuhr F. Cerebral venous and sinus thrombosis: an update. Eur J Neurol. 1994; 1: 109–126.
de Bruijn SFTM, de Haan RJ, Stam J; for the Cerebral Venous Sinus Thrombosis Study Group. Clinical features and prognostic factors of cerebral venous sinus thrombosis in a prospective series of 59 patients. J Neurol Neurosurg Psychiatry. 2001; 70: 105–108.
Ferro J, Correia M, Pontes C, Baptista M, Pita F (VENOPORT). Cerebral vein and dural sinus thrombosis in Portugal: 1980–1998. Cerebrovasc Dis. 2001; 11: 177–182.
Ferro JM, Canho P, Stam J, Bousser M-G, Barinagarrementeria F; for the ISCVT Investigators. 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.
Plum F, Posner JB. The Diagnosis of Stupor and Coma, 3rd ed. Philadelphia: F.A. Davis Co; 1980.
Guidelines on diagnosis and management of acute pulmonary embolism: Task Force on Pulmonary Embolism, European Society of Cardiology. Eur Heart J. 2000; 21: 1301–36.
Simon RP. Neurogenic pulmonary oedema. Neurol Clin. 1993; 11: 309–323.
Einhupl KM, Villringer A, Meister W, Mehraein S, Garner C, Pellkofer M, Haberl RL, Pfister HW, Schmiedek P. Heparin treatment in sinus venous thrombosis. Lancet. 1991; 338: 597–600.
Ameri A, Bousser MG. Cerebral venous thrombosis. Neurol Clin. 1992; 10: 87–111.
Barinagarrementeria F, Cantu C, Arredondo H. Aseptic cerebral venous thrombosis: proposed prognostic scale. J Stroke Cerebrovasc Dis. 1992; 2: 34–39.
Daif A, Awada A, al-Rajeh S, Abduljabbar M, al Tahan AR, Obeid T, Malibary T. Cerebral venous thrombosis in adults: a study of 40 cases from Saudi Arabia. Stroke. 1995; 26: 1193–1195.
Brucker AB, Vollert-Rogenhofer H, Wagner M, Stieglbauer K, Felber S, Trenkler J, Deisenhammer E, Aichner F. Heparin treatment in acute cerebral sinus venous thrombosis: a retrospective clinical and MR analysis of 42 cases. Cerebrovasc Dis. 1998; 8: 331–337.
Bergui M, Bradac GB, Daniele D. Brain lesions due to cerebral venous thrombosis do not correlate with sinus involvement. Neuroradiology. 1999; 41: 419–424.
de Bruijn SFTM, Stam J; for the Cerebral Venous Sinus Thrombosis Study Group. Randomized, placebo-controlled trial of anticoagulant treatment with low-molecular-weight heparin for cerebral sinus thrombosis. Stroke. 1999; 30: 484–488.
Baumgartner RW, Studer A, Arnold M, Georgiadis D. Recanalisation of cerebral venous thrombosis. J Neurol Neurosurg Psychiatry. 2003; 74: 459–461.
Breteau G, Mounier-Vehier F, Godefroy O, Gauvrit J-L, Mackowiak-Cordoliani M-A, Girot M, Bertheloot D, Hénon H, Lucas C, Leclerc X, Fourrier F, Pruvo JP, Leys D. Cerebral venous thrombosis. 3-year clinical outcome in 55 consecutive patients. J Neurol. 2003; 250: 29–35.
Mehraein S, Schmidtke K, Villringer A, Valdueza JM, Masuhr F. Heparin treatment in cerebral sinus and venous thrombosis: patients at risk of fatal outcome. Cerebrovasc Dis. 2003; 15: 17–21.
Soleau SW, Schmidt R, Stevens S, Osborn A, MacDonald JD. Extensive experience with dural sinus thrombosis. Neurosurgery. 2003; 52: 534–544.
Sanchetee PC, Dhamija RM, Roy AK, Venkataraman S. Peripartum cerebral venous thrombosis. J Assoc Physicians India. 1992; 40: 664–666.
Hamouda-M’Rad I, Mrabet A, Ben Hamida M. Thromboses veineuses et infarctus artériels cérébraux au cours de la grossesse et du post-partum. Rev Neurol (Paris). 1995; 151: 563–568.
Nagaraja DD, Haridas TT, Taly AB, Veerendrakumar MM, Subbukrishna DK. Puerperal cerebral venous thrombosis: therapeutic benefit of low dose heparin. Neurol India. 1999; 47: 43–46.
Diaz JM, Schiffman JS, Urban ES, Maccario M. Superior sagittal sinus thrombosis and pulmonary embolism: syndrome rediscovered. Acta Neurol Scand. 1992; 86: 390–396.
Rondepierre P, Hamon M, Leys D, Lederc X, Mournier-Vehrer F, Godefroy O, Janssens E, Pruvo JP. Thromboses veineuses cérébrales: étude de l’évolution. Rev Neurol. 1995; 151: 100–104.
Stefini R, Latronico N, Cornali C, Rasulo F, Bollati A. Emergent decompressive craniectomy in patients with fixed dilated pupils due to cerebral venous and dural sinus thrombosis: report of three cases. Neurosurgery. 1999; 45: 626–30.
Ferro JM, Lopes MG, Rosas MJ, Ferro MA, Fontes J; for the Cerebral Venous Thrombosis Portuguese Collaborative Study Group (VENOPORT). Long-term prognosis of cerebral vein and dural sinus thrombosis: results of the VENOPORT Study. Cerebrovasc Dis. 2002; 13: 272–278.