Clinical diagnosis of intracerebral haemorrhage: validation of a simple scoring tool in West Africans.

OBJECTIVE

To validate the use of a simple stroke scoring tool in the clinical diagnosis of intracerebral haemorrhage (ICH).

DESIGN

Prospective observational study .

METHODS

184 patients with acute stroke were consecutively evaluated within 24 hours of admission with a simple clinical tool- Benin Stroke Score (BSS), designed to diagnose ICH before cranial Computed tomography (CT) scans were done and the results were compared with CT findings as the gold standard. BSS is a simple 3-item tool with a minimum score of 0 and a maximum score of 3.5, that scores age, Glasgow Coma Scale (GCS) and pre-intervention admission diastolic blood pressure on an assigned score of either 0 or 1 and up to 1.5 for diastolic blood pressure. A model was devised and the accuracy was tested on a receiver operated characteristics (ROC) curve. Intra-rater and inter- rater reliability were tested with Pearson's correlation. BSS and CT were compared with kappa statistics.

RESULTS

(1) BSS cut-off value of 2.5 and above was positive for ICH with sensitivity, specificity, positive and negative predictive values and likelihood ratio of 70.20%, 89.00%, 74.10%, 87.00% and 6.38 (AUC of ROC curve = 0.847, p < 0.001). (2) The inter-rater and intra-rater reliability coefficients of BSS were, r =0.90, 0.95 and 0.95 (p < 0.001, <0.001 and < 0.001) respectively. (3) BSS agreed with cranial CT findings in 60% of cases (p < 0.001).

CONCLUSION

BSS is a simple 3-item tool with a reliable level of accuracy that could be used early after admission in the clinical diagnosis of ICH where neuroimaging is not available.