Intrasinus thrombolysis in cerebral venous sinus thrombosis: Single-center experience in 19 patients
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.63800
Source of Support: None, Conflict of Interest: None
Background: Evidence from small case series suggests that the intrasinus thrombolysis (IST) is relatively safe and effective in rapid recanalization of thrombosed sinuses and reversal of neurological deficits in patients with cerebral venous and sinus thrombosis (CVST) However, in the absence of randomized controlled trials, the exact role of IST in the management of CVST is unclear, Aim: To study the safety and efficacy of IST in patients with CVST. Materials and Methods: Adult patients with CVST who received IST during a two-year period (January 2003-December 2004) were included. Data regarding demographic, clinical and radiological features were collected. Follow-up data were obtained at 3-6 months. Magnetic resonance venography (MRV) was repeated to assess the recanalisation of venous sinuses. Results: Nineteen patients (11 women) with a mean age of 32 years (range 17-46 years) received IST during the study period. Common clinical features at presentation included headache, altered consciousness and seizures. Indications for thrombolysis included clinical deterioration despite adequate anticoagulation and rapid worsening of consciousness or neurological deficits. Thirteen patients (68%) had dural sinus thrombosis alone and six others had coexisting deep venous system involvement. Venous infarcts were present in 13 patients. At discharge, 15 patients (79%) had good outcome and were either asymptomatic or had only mild deficits and were independent for activities of daily living. Three patients died and one survived with severe neurological deficits. Angiographic improvement (as per digital subtraction angiography) was noted in 12 patients (complete in five and partial in seven) and seven patients had poor or no recanalization of the involved venous sinuses. At a median follow-up of 6.3 months, 14 (74%) patients had no or mild neurological deficits. Conclusion: IST is safe and effective in patients with CVST who fail to respond to conventional medical treatment. However, the subgroup of patients who are likely to benefit the most from this procedure is not clear from our data. Large randomized controlled trials are required to further clarify this issue.
Keywords: Cerebral venous thrombosis, efficacy, outcome, thrombolysis, safety, urokinase
Cerebral venous and sinus thrombosis (CVST) accounts for 0.5-1% of all strokes  and is one of the common causes of young stroke in India,  about 20% in people aged 40 years or less.  The standard therapy of CVST is anticoagulation with intravenous unfractionated heparin or subcutaneous low molecular weight heparin (LMWH). Outcomes are generally good. Even though there is a steady decline in the mortality, recent reports still indicate a mortality of 5-30%.  The predictors of poor outcome include: coma, intracerebral hemorrhage (ICH), rapidly progressing clinical deficits, posterior fossa lesion and involvement of the deep venous system.  Presence of any of the poor prognostic features or failure to respond to adequate systemic anticoagulation have been considered as possible indications for local intrasinus thrombolysis (IST),  a procedure that requires technical expertise and is not widely available. Evidence from small case series suggests that the thrombolytic therapy is relatively safe and effective in rapid recanalization of thrombosed sinuses and reversal of neurological deficits.  However, in the absence of randomized controlled trials, the exact role of IST in the management of CVST is unclear, especially with regard to patient selection, optimal time to intervene and contraindications to this therapy.  In order to achieve clarity on these issues, we analyzed the data of patients with CVST who recieved IST in our stroke unit.
All patients above the age of 18 years with a confirmed diagnosis of CVST during the study period, January 2003-December 2004, were included in the study. All patients were clinically examined by a neurologist and all those suspected to have CVST underwent brain imaging, magnetic resonance imaging (MRI) with venography (MRV) to confirm the diagnosis of CVST.
Anticoagulation with conventional heparin was the initial treatment in all the patients. Partial thromboplastin time (PTT) monitoring was performed to adjust the dose of heparin. Patients were offered IST when there was: (1) lack of adequate clinical response to heparin (worsening of clinical symptoms, such as headache, visual acuity or weakness, and also confirmed by computed tomography (CT)/MRI of the brain in suspected cases); (2) rapid worsening of neurological deficits; and (3) severe papilledema and patients were considered to be at a high risk of losing their vision. Informed consent for IST was taken from the patient or the immediate relative of the patient after explaining the risks and benefits of the procedure. All patients had screening for for thrombophilic risk factors. Decompressive hemicraniectomy was performed for those patients with severe hemispheric mass effect due to hemorrhagic infarction.
Collection of data
Data prospectively collected in the case records included: time from symptom-onset to admission, initiation of treatment, clinical features at presentation, reasons for initiating thrombolytic therapy, duration of treatment, MRI and MRV features and digital subtraction angiography (DSA) findings preceding and after completion of thrombolysis and clinical status at discharge and at follow-up.
IST in all the patients was performed by the interventional radiologist of the team (JV). Using the transfemoral approach, cerebral venogram was performed and individual sinuses were selectively catheterized and a bolus of 100,000 IU of urokinase was administered. The catheter was later placed in the involved sinus and urokinase was continued at a rate of 100,000 IU/h as continuous infusion (up to a maximum of 600,000 IU total dose). Mechanical thrombectomy was attempted when possible using a balloon. Check angiograms were performed at 24 h, 36 h, 48 h or beyond as required and at termination. Thrombolysis was terminated when there was radiological dissolution of the thrombus or the maximum dose of urokinase was reached or in the presence of clinical deterioration.
Patients were clinically followed-up for at least 6 months and the outcome was measured using modified Rankin Scale (mRS; 0 = no impairment, 6 = death). Treatment response was assessed as good improvement (patients with no or only minimal residual neurological deficits, mRS 0 or 1), partial improvement (patients with functional disability but independent for Activities of daily living (ADL), mRS 2) and poor (severe deficits, mRS 3 and 4 and death). MRV was performed to assess the radiological improvement at the 3-6 months follow-up.
A total of 42 patients (24 women) were admitted with a diagnosis of CVST during the 2-year study period. Of these, 19 patients (45.2%) received IST therapy. There were 11 women (57.9%) in the thrombolysed group. The mean age of the thrombolysed patients was 32 years (range 17-46 years). The most common presenting symptom was headache (noted in 14 patients). Three of these patients developed rapidly evolving visual loss unresponsive to intravenous heparin therapy. Other presenting clinical features are given in [Table 1]. Abnormal thrombophilia profile was the most common abnormality, noted in eight (42%) patients. Other causes included oral contraceptive pill usage (2), L-asparaginase therapy (1), postpartum state (2), polycythemia (2), dehydration (1) and antiphospholipid antibody syndrome (1). No cause could be identified in two patients. Only two patients had presented with a second episode of CVST; in all others, it was their first episode. The relatively lower proportion of postpartum CVST in our series is because our hospital does not receive many postpartum cases (possibly due to referral bias). Serum homocysteine was not routinely performed in these cases.
Thirteen patients (68%) had features of dural sinus thrombosis. The remaining six patients (32%) had combined dural sinus and deep venous system thrombosis. Superior sagittal sinus (SSS) (alone or in combination with other venous sinuses) was the most commonly affected sinus, 15 (79%) patients. The sinuses affected are listed in [Table 2]. Thirteen patients (68%) had single or multiple infarcts (hemorrhagic in six), while in six, no parenchymal lesions were seen [Table 3]. A single large hematoma or infarct in proximity of a thrombosed vein or sinus was the most common observation. Multiple lesions were found to be associated with the occlusion of deep veins. Presence of hemorrhage, infarct size or number of infarcts did not reflect in clinical outcome [Table 3].
The indications for thrombolysis included clinical deterioration despite adequate anticoagulation therapy with conventional heparin infusion (seven patients), rapid deterioration in sensorium (five patients) and rapidly progressive neurological deficits within 24 h of hospitalization (seven patients). The mean interval between hospitalization and IST was 28 h (6-54 h).
Clinical improvement - Immediate postthrombolysis
At discharge, 11 patients (58%) had good outcome (Rankin 0). This group included the three patients who had progressive visual impairment unresponsive to anticoagulation. These patients regained normal visual acuity within 1-4 days following thrombolysis. Four patients (21%) had mild residual deficits (Rankin 2). Two patients with recurrent CVST had slow resolution of symptoms after thrombolysis and one of them required thecoperitoneal shunt following which the patient was symptom free. Four patients (21%) deteriorated while undergoing thrombolysis. One had expansion in hematoma size with mass effect and required emergency decompressive hemicraniectomy. This patient had a stormy hospital course and significant residual weakness and cognitive deficits (case 3). Hemorrhagic transformation of multiple deep venous infarcts during thrombolysis resulted in the death of two patients due to transtentorial herniation and one patient died due to an unrelated cause (ventricular fibrillation).
Radiological outcome - Immediate postthrombolysis
Fifteen patients had no new lesions on immediate postthrombolysis MRI. Four patients developed new lesions: one patient developed significant expansion of existing hematoma and three patients with deep system involvement developed multiple new or additional hemorrhages in the midbrain and thalamus.th Angiographic (DSA) improvement was noted in 12 (63%) patients. Complete angiographic recanalization was seen in five patients and partial in another seven patients. Seven patients had poor or no recanalization. Of these, two had an earlier CVST event (case 1) and three had poor outcome. The location and number of occluded veins did not correlate with the extent and pace of response to treatment. Clinical recovery was slow in four of 13 patients, with occlusions restricted to superficial dural sinuses. Only four of six patients who had extensive involvement of both systems deteriorated. Notably, these four patients had multiple deep-seated hemorrhagic infarctions before thrombolysis.
All patients received oral anticoagulants for at least 6 months after discharge. Patients with normal thrombophilia profiles at 6 months did not receive further anticoagulation. At the last clinic follow-up (median: 6.3 months, range 6-8 months), 14 patients (74%) were either asymptomatic or had only minor subjective symptoms.
Case 1: A 20-year-old college student was receiving maintenance immunosuppressive therapy following renal transplantation since September 2002. In January 2003, he developed focal seizures and was diagnosed to have CVST on MRI of the brain. Oral anticoagulant therapy was commenced, which he discontinued 6 months later. He presented to our hospital in September 2003 with focal seizures, left hemiparesis and increasing somnolence of 36 h duration. MRV showed features of SSS and right transverse sinus thrombosis. Anticoagulation with heparin infusion did not show desired therapeutic response. IST with urokinase for 82 h in two sittings was attempted with partial response. He later required lumboperitoneal shunt for increasing papilledema and impaired vision. He was asymptomatic at the time of discharge.
Case 2: A 38-year-old lady with fibroid uterus, pulmonary stenosis and history of mild headaches off and on for 2 months presented with acute onset severe headache, vomiting and blurred vision of 2 days duration. On examination, she had grade 4 papilledema and reduced vision but no focal neurological deficits. MRV revealed features of extensive CVST involving SSS, right transverse sinus and deep venous system [Figure 1]. Visual acuity continued to deteriorate even after 8 h of anticoagulation with heparin. Symptoms and papilledema rapidly resolved following thrombolysis. Angiogram showed recanalization of SSS and deep venous system [Figure 2]. She received oral anticoagulant therapy for 6 months.th
Case 3: A 44-year-old man, recently diagnosed as hypertensive, presented with a history of one episode of generalized seizure and acute onset excruciating headache of 8 h duration. On examination, he had papilledema and minimal left pyramidal signs. MRV revealed right transverse sinus and straight sinus thrombosis and right temporoparietal hematoma. The patient continued to deteriorate despite treatment with heparin infusion. IST with urokinase was started. Check angiogram 12 h later revealed incomplete recanalization and urokinase infusion was further continued. Three hours later, the headache worsened and he slipped into coma. Urgent CT scan brain demonstrated extension of hemorrhage with intraventricular extension. He underwent emergency decompressive hemicraniectomy and required mechanical ventilation for 2 weeks. At discharge, the patient had left hemiparesis and cognitive dysfunction. Six months later he achieved functional independence but could not return to his profession.
Outcome in patients not receiving intrasinus thrombolysis
During the study period, 23 patients with CVST were treated with conventional treatment (anticoagulation with heparin infusion in 19 and LMWH in five patients) only. Mortality in the conventional treatment group was 30% (7/23), as compared to 16% (3/19) in the IST group. At the last follow-up, 52% (12/23) had good outcome (mRS 0-1) in the nonthrombolysed group, as compared to 74% (14/19) in the thrombolysed group.
Use of local pharmacological thrombolysis in CVST was first reported in 1988 by Scott et al. They infused urokinase directly into the SSS over 11 h via a midline frontal burr hole.  The patient in coma with decerebrate posturing, improved rapidly over three days and had only mild aphasia and memory impairment at four weeks. Subsequently, several authors have reported case reports and small case series (involving five to 18 patients) highlighting the usefulness of IST in CVST. ,,, A systematic review of all published cases of thrombolysis for CVST up to July 2001 identified 146 patients.  Six larger trials have been published, the largest case series including 20 patients of IST. 
The criteria for IST in this study were rapid worsening of symptoms, presence of at least one factor for potentially unfavorable outcome  or lack of response to systemic anticoagulation. There are no established criteria in the literature regarding the indications for thrombolysis in CVST. Buccino and colleagues suggested the following criteria: risk of death, involvement of deep venous system and poor clinical response despite full heparinization,  whereas Stam and colleagues considered IST in patients with altered mental status, coma, straight sinus thrombosis or large space-occupying lesions.  Rapid and sustained recovery observed in about three-fourth of our patients is similar to previous published reports. In a recently published case series, 12/20 (60%) of patients had an excellent recovery after IST.  Rapid resolution of neurological dysfunction, particularly in patients unresponsive to intravenous anticoagulation with heparin, indicates the importance of infusing an adequate concentration of fibrinolytic agent at the site of thrombosis. The need for superselective catheterization and infusion of thrombolytic agent at the site of thrombosis and the need to continue the infusion for several hours have been highlighted by others too.  Early therapeutic intervention may lead to successful clot lysis while it is still in a friable condition. On the contrary, chronic lesions prove to be resistant to lysis, as is seen in our patients with recurrent thrombosis, one of whom required, in addition, a thecoperitoneal shunt before complete clinical recovery.
Adverse outcome in five patients (about 25%) on the other hand emphasizes the unpredictability of this treatment modality and the importance of identifying factors that predict adverse response. Of these, two patients who were at a high risk for systemic anticoagulation had catastrophic hemorrhage following thrombolysis. Others have reported similar rates of hemorrhage after thrombolysis. In a recent series of the 20 patients, 5 (25%) patients had hemorrhage following thrombolysis (minor in three and significant in two cases).  It is quite possible that use of other alternate treatments such as mechanical or rheolytic thrombolysis might have avoided death and disability in our patients.  There were three (16%) deaths in our series (two due to hemorrhage) and six (30%) (only one due to increased hemorrhage) deaths in the other study.  Thus death appears to be an uncommon direct complication of thrombolysis in CVST. 
This study highlights the benefit of thrombolysis, particularly in patients unresponsive to anticoagulation. The group that received IST had a better outcome (lower mortality and higher proportion of patients with mild or no deficits) as compared to the group that received conventional treatment. Authors believe that excellent resolution of visual deficits and intracranial hypertension in these patients is related to intervention early in the clinical course.As mentioned in other reports, there is no correlation between extent of recanalization and clinical outcome. Patients improve despite the lack of angiographic recanalization. The major limitation of our study is that it is an observational study and not a randomized study and the patients who received IST and who had not received IST were not comparable. Well designed randomized controlled trials comparing heparin infusion and IST alone or in combination with heparin are very much required.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3]