Normal d-dimer levels do not exclude the diagnosis of cerebral venous sinus thrombosis.

Abstract

Sirs: Cerebral venous sinus thrombosis (CVT) is a life threatening condition with a broad range of clinical neurological presentations and a large differential diagnosis [1]. Early diagnosis may be crucial to the neurological outcome, which can often be favourable. Treatment with heparin may improve survival and disability [2] and patients who continue to deteriorate may be suitable for local thrombolysis [3]. CVT must be considered in any patient with sub-acute headache, seizures, disorders of consciousness or papilloedema. The condition can mimic stroke, abscess, tumour, encephalitis or idiopathic intracranial hypertension. Similarly, there is a wide range of underlying aetiologies including inherited thrombophilias, oral contraception, pregnancy, local intracranial sepsis and systemic inflammatory diseases such as Behcet’s disease. Therefore, the diagnosis cannot be reliably made on clinical grounds alone and is often delayed, particularly when out of hours neuroimaging is unavailable. MRI and MR angiography are the accepted methods of choice, though the sensitivity and specificity of this technique have never been compared with catheter angiography [4]. In other venocclusive conditions such as deep vein thrombosis (DVT) and pulmonary embolism (PE) diagnostic accuracy has been improved by estimation of d-dimer levels in plasma [5]. Because d-dimers may rise as part of the acute phase response, raised levels do not constitute diagnostic proof of thrombosis. However, a normal d-dimer level is used as a test of exclusion. We have performed a pilot study to investigate if, by analogy with deep venous thrombosis (DVT) and pulmonary embolism (PE), estimation of plasma d-dimers could aid the diagnosis of patients with possible CVT. We used the MDA D-dimer (Organon Teknika Corporation, Durham, USA) a quantitative immunoassay using latex microparticles coated with mouse monoclonal antibody to D-dimer (8–8G). These latex particles thus agglutinate in the presence of D-dimer and this process can be monitored using the photo-optical detection system of the MDA 180 automated coagulometer. The rate of agglutination is proportional to the Ddimer concentration and can be interpolated from a reference curve. Assays were complete within 15 minutes. The primary aim of this study was to establish if a normal level of d-dimers could reliably exclude the diagnosis of CVT. Successive patients with radiologically proven (MRI) venous sinus thrombosis presenting to a regional neurology unit underwent plasma d-dimer estimation before anticoagulant therapy was started. The following brief clinical descriptions serve to illustrate that these patients were representative of patients with sinus thrombosis as a whole: • Patient 1 was a 17-year-old female who presented with a six-day history of occipital headache associated with nausea and vomiting progressing to collapse with motor and sensory loss of the left side of her body. She was taking the oral contraceptive pill. MRI/MRV showed a right-sided lateral and sigmoid sinus thrombosis. D-dimer levels were 430 μg/l (normal value less than 500μg/l). She was subsequently found to have a strongly positive lupus anticoagulant test (DRVVT). She was anticoagulated and made a full recovery. • Patient 2, a 55-year-old female with a history of chronic schizophrenia, was admitted having collapsed. There was a background of one week of increasing lethargy and slurring of speech and loss of hearing in the left ear. MRI/MRV revealed left lateral and sigmoid sinus thrombosis and left mastoiditis, which was surgically drained. The d-dimer level was 392μg/l. She made a full neurological recovery. • Patient 3, a 31-year-old male from Saudi Arabia, had a background history of pulmonary embolus and hepatic vein thrombosis and had unilaterally stopped taking warfarin six months prior to this episode. There was a history of venous thrombosis in his mother and sister. He presented with two generalised seizures on the background of a one week history of occipital headache and MRI revealed extensive thrombosis of the sagittal sinus. D-dimer levels were 1392μg/l. He made a full recovery. • Patient 4 was of Afro-Caribbean origin and had a long standing history of intravenous heroin use. He presented with an acute confusional state and was found to be in hyperosmotic non-ketotic diabetic coma. With treatment of this his conscious level improved and he began to complain of headache. Contrast enhanced CT of the brain showed sagittal sinus thrombosis. Ddimer levels were 919μg/l and he was found to have Sickle cell trait. He made a full neurological recovery. LETTER TO THE EDITORS