Clinical profile of cerebral venous sinus thrombosis and the role of imaging in its diagnosis in patients with presumed idiopathic intracranial hypertension

Retrospective descriptive study reporting the rate of occurrence of cerebral venous sinus thrombosis (CVST), highlighting the role of magnetic resonance imaging (MRI) and magnetic resonance venography (MRV) in patients with presumed idiopathic intracranial hypertension (IIH). Study was conducted in the department of neuro-ophthalmology at a tertiary eye care center in South India. Data from 331 patients diagnosed with IIH from June 2005 to September 2007 was included. Inclusion criteria were: Elevated opening cerebrospinal fluid (CSF) pressure of more than 200 mm of water on lumbar puncture, normal CSF biochemistry and microbiology, and normal neuroimaging as depicted by computed tomography(CT) scan. Exclusion criteria were: Space-occupying lesions, hydrocephalus, meningitis, intracranial pressure within normal range, abnormal CSF biochemistry and microbiology. The remaining patients were evaluated with MRI and MRV. CVST was present in 11.4% of patients who were presumed to have IIH (35/308). MRI alone identified 24 cases (68%) of CVST, while MRI used in combination with MRV revealed an additional 11 cases (32%). Risk factors associated with CVST were identified in nine out of 35 patients (26%). CVST may be misdiagnosed as IIH if prompt neuroimaging by MRI and MRV is not undertaken. Risk factors of CVST may not be apparent in all the cases and these patients are liable to be missed if CT scan alone is used for neuroimaging, hence MRI, combined with MRV should be undertaken to rule out CVST.

prior to the MRI era [2] CT scan alone may be an insuffi cient tool for accurate diagnosis of IIH. MRI, along with MRV wherever necessary should be the modality of neuroimaging for accurate diagnosis. [3,4] Risk factors for cerebral venous sinus thrombosis (CVST) may not be apparent in all the cases, hence it is difficult to exclude CVST clinically in these patients. Magnetic resonance imaging (MRI) in combination with magnetic resonance venography (MRV) is recommended to correctly diagnose CVST in these patients. [2][3][4] We report the rate of occurrence of CVST in patients with presumed IIH as well as the associated risk factor profi le, which prompted subsequent MRV.

Materials and Methods
The study was conducted in the department of neuroophthalmology at a tertiary eye care center in South India. Patients diagnosed with IIH between June 2005 and September 2007 were included in this study. Inclusion criteria were: Elevated opening CSF pressure of more than 200 mm of water on lumbar puncture, normal CSF biochemistry and microbiology, and normal neuroimaging as depicted by CT scan.
Patients were excluded if they had space-occupying lesions, hydrocephalus, meningitis, ICP within normal range, abnormal CSF biochemistry and microbiology or bilateral disc edema from other causes like ischemic optic neuropathy, diabetic papillopathy. The remaining patients were evaluated with MRI, and MRV. MRV was performed only when MRI was not supportive in diagnosis of CVST, or there was strong clinical suspicion of CVST due to associated risk factors, namely male sex, non-obesity, post-surgery, use of oral contraceptives, deep vein thrombosis, hypercoagulable states in the presence of normal MRI.

Results
Data of 331 patients diagnosed with IIH was retrieved. Twentythree patients were lost to follow-up and hence excluded. Thirty-fi ve patients out of 308 (11.4%) were found to have CVST. The study population included 240 females and 68 males. Mean age was 31.4 years (standard deviation 10.6 years).
Among the CVST group there were nine females and 26 males (age range 17-58 years). Though MRI was diagnostic in 24 patients out of 35 (68%) showing hypointense lesion [ Fig.  1] in the venous sinuses it was suggestive/suspicious in the remaining 11 cases. MRV showing signal void was diagnostic of CVST in all these 11 cases [ Fig. 2]. Fig. 3 shows normal MRI for comparison. Thirty-one patients had thrombosis involving the superior sagitt al sinus (88%). Four patients had isolated involvement of sigmoid and transverse sinus.
Underlying risk factors in the form of pregnancy (n ϭ 2) post caesarean section (n ϭ 2), deep vein thrombosis (n ϭ 1), oral contraceptives (n ϭ 2), hyperhomocystenemia (n ϭ 1), leukemia (n ϭ 1) were present only in nine cases (26%). Patients were referred to the neurosurgeon and were treated with anticoagulants, and acetazolamide. None of the patients with negative MRI/ MRV developed CVST in the subsequent follow-up of one year.

Discussion
The clinical presentation of CVST closely overlaps that of IIH. IIH is a diagnosis of exclusion according to Dandy's criteria [1] i.e. symptoms and signs att ributable to the raised ICP, abnormal CSF pressure on lumbar puncture, normal CSF composition, normal neuroimaging results, awake and alert patient and absence of localizing fi ndings on neurological examination.
The dynamics of ICP are described by the relationship: [5,6] Intracranial pressure ϭ cerebrospinal fl uid FR ϫ R O ϩ P SS , where Cerebrospinal fl uid FR ϭ cerebrospinal fl uid formation, R O ϭ cerebrospinal fl uid outfl ow resistance, and P SS ϭ outfl ow pressure in the superior sagitt al sinus.
It is apparent from this relationship that sagitt al sinus hypertension will cause a concomitant increase in ICP.
The type of neuroimaging has to be modified in view of associated risk factors for CVST, namely male sex, non-obesity, use of oral contraceptives, deep vein thrombosis, hypercoagulable states, lupus anticoagulant, infections of ear, nose and throat, mastoiditis and history of surgery in the region of head and neck. It is crucial to diff erentiate CVST from IIH because the management protocols are entirely diff erent. CVST is typically treated with anticoagulants and IIH with diuretics. CVST is life-threatening as it can cause stroke and death, unlike IIH, which stops with optic atrophy.
The combination of MRI and MRV is the diagnostic modality of choice for CVST. [7,8] MRV alone may be false-positive in cases of sinus aplasia or hypoplasia (seen as a fl ow gap). It can also mistake T2-weighted hypointense signal of deoxyhemoglobin and intracellular methemoglobin as fl ow void. [9] The dural sinuses most frequently aff ected are superior sagitt al sinuses, transverse and sigmoid sinuses.
Our results diff er from the study of Lee et al., [10] a smaller study that reviewed MRV results in 22 consecutive patients, all women who fi t the typical demographic profi le of IIH, and identifi ed none of the patients with CVST. Purvin et al. [11] suggested that clinical manifestations of CVST may be diff erentiated from IIH by the abrupt onset and marked severity of symptoms.
However, the presentation of CVST can vary considerably based on the extent and location of thrombosis. The presentation of most of our patients with CVST was not the same as that described by Purvin et al. and closely mimics that of IIH.
Our study identifi ed CVST in 11.4% of patients who were presumed to have IIH, and in the absence of correct diagnosis and lack of MRI and MRV these patients might continue to get treated as IIH, resulting in dismal visual prognosis, as well as signifi cant risk of stroke and death. MRV in combination with MRI is recommended to identify this subgroup of patients who present with associated risk factors.  Retinocytoma is a rare retinal tumor that is generally believed to be a benign variant of retinoblastoma. [1][2][3][4][5][6] The reported proportion of retinocytoma among the population with retinoblastoma has varied from 1.8 to 10%. [1,2,7] The ophthalmoscopic appearance of the retinocytoma, including the presence of a gray translucent mass, intralesional calcification, retinal pigment epithelial alteration and chorioretinal atrophy resembles the spectrum of retinoblastoma regression patt erns observed aft er irradiation. [2] It is important to recognize retinocytoma clinically and diff erentiate it from active retinoblastoma as it usually requires close observation rather than active treatment.
Simultaneous occurrence of retinocytoma and retinoblastoma is very rare [1,7,8] and most retinocytomas are stable and demonstrate no tendency to grow or metastasize. [2] We report here a unique presentation of a retinocytoma associated with bilateral retinoblastoma.

Case Report
A 3-year-old girl was referred to the oncology service because of left exotropia associated with an intraocular mass. Familial history was not contributory. Examination under anesthesia revealed an elevated semitranslucent mass measuring about 9 ϫ 8 mm in basal dimensions and 3 mm in thickness superior to the left fovea, surrounded by a margin of chorioretinal atrophy and retinal pigment epithelial alteration [ Fig. 1]. Some calcifi cation was evident in the central portion of the tumor. This retinal lesion was thought to be a retinocytoma. Five small translucent masses compatible with retinoblastoma were found in the same eye: Four in the posterior pole and the fi ft h near the ora serrata. Two subretinal seeds were found near the ora serrata as well. In the right eye, two small retinoblastomas were present near the optic disc and another one was found near the ora serrata [ Fig. 2]. Anterior segments of both the eyes were normal. No extraocular extension and central nervous system abnormality was found in systemic evaluation and