A Comparative Study of Clinical Scoring System of Cerebrovascular Accident with CT Scan

OBJECTIVE: Stroke is the second most common cause of death and major cause of disability worldwide. Low and middle income countries account for 85.5% of total stroke deaths worldwide and the number of disability-adjusted life years in these countries is seven times that in high income countries.

CT scan is a safe and non-invasive procedure for differentiating between cerebral infarction and haemorrhage. However, in India and other developing countries the availability of CT scan facilities in rural and peripheral centres is low but cost of scanning precludes its routine usage by the poorer society. So this study aims at differentiating ischaemic and haemorrhagic strokes at bedside by using clinical scoring systems and to compare the accuracy of clinical scoring with the CT scan findings.

METHODS: This was a prospective clinical study on 100 patients who presented to the hospital within 72 hours of onset of symptoms of stroke. Siriraj and Allen stroke score were calculated using various parameters, subsequently all of these patients underwent CT scan of brain and accuracy of clinical scoring system with the CT scan was compared.

RESULTS: Most number of affected patients belonged to the age-group 61-70 with a male preponderance. 66% of the stroke cases were ischaemic, most common presenting symptom was hemiplegia followed by altered sensorium. Midline shift in CT scan was seen more in haemorrhagic stroke patients. On comparison of Siriraj score with CT scan findings correct correlation was seen in 82% of cases while Allen score showed correct correlation only in 67% of cases so the diagnosis of stroke subtype was best predicated by Siriraj score as compared to Allen score.

CONCLUSION: Present study shows that Siriraj score had good sensitivity and positive predictive value for ischaemic stroke (91.93% and 90.47%) while for haemorrhagic stroke sensitivity and positive predictive value was less (80.6% and 83.33%). Allen's score also showed good sensitivity and positive predictive value (96.2% and 91.22%) for ischaemic stroke however for haemorrhagic stroke it had poor sensitivity and positive predictive value (75% and 88.23%). So overall Siriraj score is a better tool for bedside early clinical diagnosis of stroke and to start anti-ischaemic therapy where immediate CT scan facilities are not available but these scores cannot be used for deciding thrombolysis using recombinant tissue plasminogen activators (r-tPA) which requires exclusion of haemorrhage by CT scan.