Isolated intracranial hypertension as the only sign of cerebral venous thrombosis

Methods.

A total of 160 consecutive patients with CVT (94 female, 66 male; age range 11 to 77) were systematically examined by our group between 1975 and 1998. CVT was diagnosed with MRI, magnetic resonance venography (MRV), or conventional angiography. Baseline characteristics of the patients and pertinent clinical information such as headache, papilledema, seizures, focal neurologic symptoms or signs, and level of consciousness were obtained on first encounter and during follow-up. Neuroimaging results, CSF analysis, etiologic factors, treatment, and prognosis of patients were also systematically recorded. Patients were separated into two groups according to their clinical presentation: patients with isolated ICH, who never developed any other neurologic symptoms or signs (Group I), and patients with other neurologic symptoms or signs (Group II). Comparison of Groups I and II were performed using chi-square and Student’s t-tests.

Results.

Fifty-nine of 160 patients (37%) had ICH as the only sign of CVT (Group I), and 101 (63%) had other neurologic symptoms or signs (Group II). Detailed characteristics of patients in both groups are presented in the table.

Among our 59 patients with ICH, 59 (93%) had headache (including 8 without papilledema), and 51 of 59 (86%) had papilledema (including 4 without headache). Nine patients (15%) had unilateral (7) or bilateral (2) sixth nerve palsies. Twenty-four patients (40.6%) had headaches for at least 1 month before correct diagnosis and appropriate treatment without developing focal neurologic symptoms or signs. Neuroimaging showed involvement of more than one sinus in 35 of 59 patients (59%). The superior sagittal sinus was involved in 32 patients (54%) (isolated in 7) and lateral sinuses in 47 (80%) (isolated in 17). The straight sinus was thrombosed in eight patients, cortical veins were involved in two, and deep cerebral veins in three, always in association with a thrombosis in the superior sagittal sinus or lateral sinuses. A majority of patients with isolated ICH (54%) had a normal brain CT. Lumbar puncture was performed in 44 patients of Group I (75%). Opening pressure was ≥200 mm in 25 of 32 patients (78%) in whom it was measured. Eleven patients (25%) had abnormal CSF content, among whom 6 had elevated red cells as the only abnormality. Etiologies and risk factors for CVT in Group I included local causes in 7 patients (12%), abdominal surgery in 1 (2%), systemic inflammatory diseases in 18 (30.5%) (including Behçet’s disease in 13, MS in 1, chronic uveomeningitis in 1, systemic lupus erythematosus in 2, and systemic vasculitis in 1), bacterial infection in 2 (3%), cancer in 1 (2%), postpartum state in 1 (2%), coagulopathies in 11 (19%), oral contraception in 7 (12%), and no apparent cause in 11 (19%). Treatment consisted of anticoagulants in 41 of 59 (69.5%), steroids or acetazolamide in 26 of 59 (44%), lumbar puncture to reduce intracranial pressure in 44 of 59 (75%), and lumbo-peritoneal shunt in 1 of 59. Six patients improved spontaneously and did not receive any treatment. Prognosis was good in the group of patients with isolated raised intracranial pressure because only 3 of 59 patients developed optic atrophy with severe visual loss, 1 died from carcinomatous meningitis, and 55 of 59 (93%) had complete recovery. However, formal visual field testing was not systematically performed.

Results of the comparison of Group I and Group II are detailed in the table. The only statistically significant differences were the higher frequency of patients with lateral sinus thrombosis in Group I and of cortical vein thrombosis in Group II, the higher frequency of abnormal CTs and of abnormal CSF content in Group II, the higher frequency of inflammatory diseases in Group I and of postpartum state in Group II, and the worse prognosis of patients from Group II. As shown in the table, 82% of the patients who developed focal neurologic symptoms or signs had clinical features of ICH, which were usually the first symptoms of CVT. Fourteen of the 101 patients (14%) with other neurologic symptoms or signs had isolated ICH for at least 1 month before developing focal symptoms or signs.

Discussion.

One third of our patients with CVT presented with isolated signs of intracranial hypertension, clinically mimicking IIH. In numerous reports, the criteria for the diagnosis of IIH were based on clinical, CT, and CSF findings only.1,4,7,8 CT was normal in 54% of our patients with isolated ICH and therefore should not be used to rule out CVT. CSF studies showed normal CSF content in 75% of our patients with isolated raised intracranial pressure, and showed elevated red cells as the only abnormality in 6 of the 11 patients with abnormal CSF content. These findings demonstrate that CVT should still be considered in patients with isolated raised intracranial pressure in whom both CT and CSF content are completely normal. Therefore, MRI, specifically looking for evidence of CVT, should be ordered systematically in patients with isolated raised intracranial pressure, even when CT and CSF content are normal. MRI is now the method of choice for the diagnosis and follow-up of CVT because it shows the thrombus itself typically as an increased signal on T1- and T2-weighted images.1,2,9,10 Furthermore, the signal of the thrombus changes over time, thus illustrating the natural history of the thrombotic process (figures 1 to 4⇓⇓⇓).1,2,9,10 The interpretation of MRI findings in CVT may, however, sometimes be difficult, and false-negative images have been described, either in the very early stage in the first 2 or 3 days or at late stage after 1 month (see figures 1 to 4⇓⇓⇓). In such cases it is crucial to perform MRV, which will demonstrate the absence of flow in the thrombosed sinus.1,2,9,10 When MRI is not available or is contraindicated, conventional angiography should be considered.1,4,9,10 Indeed, in one prospective study of 24 consecutive patients presenting with typical features of “IIH,” angiography disclosed CVT in 6.4

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Thrombosis of the left lateral sinus on day 2 (acute stage). On the sagittal T1-weighted image (A) the fresh thrombus appears isointense (arrow). MR venography of the same patient (B) shows the absence of flow signal within the left lateral sinus.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. Thrombosis of the superior sagittal sinus on day 2 (acute stage). On the sagittal T1-weighted image (A) the fresh thrombus appears isointense (arrows), whereas on the T2-weighted image (B) the occluded sinus is strongly hypointense, simulating a normal flow void.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3. Thrombosis of the left lateral sinus on day 10 after the onset of clinical symptoms. The sagittal T1-weighted image shows hyperintense thrombus signal in the occluded sinus (arrow).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4. Thrombosis of the left lateral sinus in a patient with chronic cerebral venous thrombosis. On the sagittal T1-weighted image (A) the occluded sinus appears heterogeneous (arrow). Magnetic resonance venography (B) demonstrates the absence of flow in the thrombosed sinus (arrow).

A previous study,4 in which patients with CVT and with IIH were compared, showed a large female preponderance and a high prevalence of obesity in patients with IIH. Indeed, female preponderance and obesity or recent weight gain are well-known features of IIH,6 suggesting that CVT should be particularly looked for in males and nonobese females presenting with ICH. Among our 59 patients with isolated ICH, 33 (56%) were female and 26 were male. We did not record the weight of our patients with CVT presenting with ICH, but being a young obese woman obviously does not protect a patient from developing CVT and therefore should not be used on an individual basis to rule out CVT.

Some conditions associated with CVT, such as local causes, systemic inflammatory diseases, infection, cancer, pregnancy and the postpartum state, or coagulopathies, were found in our patients with isolated raised intracranial pressure (see table). When any of the many conditions predisposing to CVT are present, this should certainly prompt extensive investigations to detect CVT, and restricting those investigations to such cases has been suggested.8 However, in view of the high prevalence (19%) of CVT without recognizable etiology, even after an extensive workup, the absence of such conditions should not obviate a search for CVT.

One could argue that it is not so important to rule out CVT in patients presenting with features of IIH (i.e., isolated raised intracranial pressure with normal CT and normal CSF content) because the management would be the same—reduction of intracranial pressure by medical or surgical means. However, we think that, given the difference in prognosis of these two conditions, they should be managed differently. Indeed, the prognosis of IIH is often so good (i.e., patients do not die from IIH) that it has been called “benign intracranial hypertension”6-8 (in fact not so benign, given the risk of permanent visual loss caused by prolonged papilledema and secondary optic atrophy6-8,11). However, CVT is still associated with a 10 to 30% mortality from multiple venous infarctions, uncontrolled seizures, pulmonary embolism, or, more frequently, from the underlying disease such as cancer or infection.1,12-15 Intracranial hypertension is one of the prevailing clinical features of CVT and was present initially in 82% of our 101 patients from Group II, who subsequently developed other neurologic symptoms or signs (see table). A large majority of patients with CVT have headache as the first symptom of CVT, and it is impossible to predict which patients will have severe complications from CVT and which patients will only manifest persistent ICH. Indeed, 14% of our patients from Group II had symptoms of ICH for at least 1 month before developing focal symptoms or signs. Moreover, comparison of Group I and Group II only showed obvious differences such as a higher frequency of cortical vein thrombosis, abnormal CTs, and abnormal CSF content in patients from Group II (see table). Other reports have shown that CVT can recover spontaneously and that such a good outcome is more frequent in patients who present with ICH than in those who have focal signs or parenchymal lesions on imaging.1 Indeed, 10% of our 59 patients with CVT and isolated raised intracranial pressure improved spontaneously. However, the outcome of patients presenting with CVT is totally unpredictable, and as already strongly emphasized in the 1960s,15 “in sinus thrombosis, the patient is in danger of cerebral infarction at any moment from spread of thrombosis to the cerebral veins.” Moreover, patients with CVT are at risk for pulmonary embolism, which has been reported in as many as 11.3% of patients with acute CVT.14 Therefore, we recommend that patients with isolated raised intracranial pressure from CVT should be managed not only like patients with IIH, but also acutely with heparin. Our approach is to lower intracranial pressure by one lumbar puncture (usually performed as part of the diagnostic or etiologic workup) and subsequently with the use of acetazolamide or steroids to prevent permanent visual loss. In addition, we anticoagulate all patients with demonstrated CVT, provided that there is no general contraindication to the use of heparin.1,2,16 Indeed, there is evidence that patients with CVT treated with anticoagulants have more favorable outcomes and that anticoagulation is safe, even in patients with hemorrhagic cerebral venous infarction.1,16-19

Patients with isolated raised intracranial pressure from CVT have a good prognosis for vision because only three patients (4.5%) in Group I had severe visual loss from secondary optic atrophy. Conversely, up to 25% of patients with IIH have severe visual loss.11 This difference has already been emphasized,3,4 and it has been suggested that the higher frequency of visual loss in IIH compared with CVT may be explained by a more chronic onset and a longer time to diagnosis and treatment in patients with IIH. A complete neuro-ophthalmologic evaluation was performed in 20 patients with CVT and ICH,3 and it suggested that most patients with CVT maintain good vision. However, as emphasized in series of patients with IIH,11 visual loss secondary to papilledema is often insidious, with progressive constriction of the visual fields and relative sparing of central visual acuity, suggesting that visual acuity is not a sensitive marker in the evaluation of vision loss in these patients.