BioScience Trends. 2018;12(2):193-200. (DOI: 10.5582/bst.2018.01061)
A reliable grading system for prediction of hematoma expansion in intracerebral hemorrhage in the basal ganglia.
Huang YW, Zhang Q, Yang MF
Hematoma expansion (HE) is an independent predictor of poor outcome and secondary neurological deterioration in intracerebral hemorrhage (ICH) and is associated with high morbidity and mortality. Noncontrast computed tomography (NCCT) may identify the sites of active extravasation. Therefore, we have attempted to (1) devise a reliable and easy-to-use prediction score to predict the risk of HE in ICH and (2) validate the accuracy of this grading system and perform an independent analysis of HE predictors. We included patients in whom an intracerebral hemorrhage (ICH) occurred in the basal ganglia between Jan. 2015 and Jan. 2018. These patients had undergone a baseline CT scan at Qinghai Provincial People's Hospital within 24 hours after the onset of ICH symptoms. Two observers independently assessed the presence of the island sign, blend sign, or swirl sign on an NCCT scan during patient selection. Patients underwent a baseline NCCT scan and 24-hour NCCT follow-up for analysis of HE. The accuracy of this grading system was assessed. Independent predictors of HE were identified using multivariable regression. Of 266 patients with ICH, 61 (22.93%) presented with the island sign, 63 (23.68%) presented with the blend sign, and 50 (18.80%) presented with the swirl sign. The overall incidence of HE was 37.22% (99/266). Of 125 patients (46.99%) who underwent a baseline CT scan within 6 hours of onset, 141 (53.01%) underwent a scan in 6-24 hours. Multivariable logistic regression analysis identified the hematoma volume (OR, 0.974; P = 0.042), intraventricular hemorrhage (IVH) extension (OR, 3.225; P = 0.003), time from onset to the baseline CT scan (OR, 0.986; P < 0.001), and anticoagulant use or an international normalized ratio (INR) > 1.5 (OR, 3.362; P = 0.006) as closely associated with HE. In conclusion, the grading system demonstrated reliable accuracy at predicting HE. The grading system demonstrated acceptable accuracy in an independent single-institution study. The role of the grading system in predicting HE and poor outcome in patients with ICH is significant. NCCT imaging markers may serve as key markers for HE prediction.