Comparison of Siriraj Stroke Score and the WHO criteria in the clinical classification of stroke subtypes

Clinical distinction between cerebral haemorrhage (CH) and cerebral infarction (CI) is important in the management of stroke patients in areas where CT scan facility is lacking or access limited by cost and distance. This distinction is necessary in our environment where an increasing proportion of patients are suspected to have haemorrhagic stroke. This study compares Siriraj stroke score (SSS) and the WHO criteria for the acute stroke syndrome as simple tools for this purpose. The computerised tomography (CT) brain scans of all patients referred with clinical diagnosis of stroke at the University College Hospital (UCH), Ibadan, and RADMED diagnostic centre. Lagos were retrieved and reviewed as well as the case notes of these patients at the referral hospitals. Relevant clinical data were extracted from the case records. The patients were classified into either CI or CH using the WHO criteria for acute stroke syndrome and the SSS. This classification was compared with the CT scan diagnosis using the latter as the gold standard. Data analysis was performed with Epi-info software, and Kappa statistics (k value) for comparability test with 95% confidence interval was used to compare the two clinical criteria with the gold standard. Ninety-six patients had complete clinical records and CT scan features consistent with the diagnosis of stroke, of which 52 were diagnosed as CI and 44 as CH. SSS had sensitivity of 50% for haemorrhage and 58% for infarction with an overall accuracy of 54.2%. The WHO criteria for the acute stroke syndrome had sensitivity of 73% for haemorrhage and 69% for infarction with an overall accuracy of 71%. The kappa coefficient was 0.18 for the SSS and 0.41 for the WHO criteria. The diagnostic accuracy of the WHO criteria for the acute stroke syndrome is higher than that of the Siriraj stroke score. The WHO criteria showed moderate agreement (k=0.41) with the CT scan, while the SSS showed no agreement (k=0.18). When CT scan is not affordable or its use is limited by distance, the WHO criteria for acute stroke syndrome could be more useful. A prospective study with a larger sample size is suggested for definitive conclusion.

Misdiagnosis of stroke--a computerised tomography scan study

Misdiagnosis of stroke has significant implication for definitive therapy. This study assessed the frequency of misdiagnosis of stroke using computerised tomography (CT) scan of the brain. One hundred and fifty-six patients admitted over a five year period (1991-1996) with clinical features suggestive of stroke had their CT brain scan reviewed. Only 89 (57%) had neuroradiological features consistent with stroke, of which 59 (66%) had cerebral infarction while 30 (34%) had cerebral haemorrhage. In 67 (43%) of the cases, there were no features of cerebro-vascular accident (CVA) on the CT scan. In this group, cerebral atrophy was the commonest radiological abnormality (21/67), followed by brain tumor (10/67) and subdural haematoma (9/67). The CT scan was normal in 25 patients. The misdiagnosis of surgically treatable conditions in as many as 21 (13.5%) of the 156 patients calls for better neurological evaluation of patients admitted for stroke in tertiary and non-tertiary hospitals. Where available, patients with clinical diagnosis of stroke should have CT scan evaluation to ensure that patients who can be helped surgically are identified early and appropriately treated.