Answers are grounded in this paper and cite their sources.
Background Non-contrast computed tomography (NCCT) hypodensities may identify patients with spontaneous intracerebral hemorrhage (ICH) at increased risk of hematoma expansion (HE), but the consistency and certainty of the evidence remain unclear. Methods We searched Web of Science, PubMed, Embase, and the Cochrane Library from inception to August 11, 2026. Two reviewers independently screened studies, extracted data, and assessed risk of bias using the Quality in Prognosis Studies tool. Adjusted odds ratios (ORs) were synthesized using a prespecified random-effects model. The review was registered in PROSPERO (CRD420261468570) and reported according to PRISMA 2020. Results Source data included retrospective observational cohorts alongside post-hoc analyses of prospectively enrolled trial participants, totalling 12 reports with 9,639 reported participants before accounting for possible cohort overlap. The overall certainty of evidence was LOW. The primary meta-analysis included five studies that evaluated hypodensities, used the conventional HE threshold (>6 mL or >33%), and reported adjusted ORs. Hypodensities were associated with higher odds of HE (pooled OR, 4.23; 95% CI, 2.47–7.26; p < 0.001), with moderate heterogeneity ( I 2 = 67.7%). A broader exploratory synthesis of 12 study estimates yielded a pooled OR of 4.35 (95% CI, 2.99–6.33; I 2 = 84.7%), but this estimate combined different marker definitions, HE thresholds, effect estimates, and potentially overlapping source cohorts. Ten studies (83.3%) were judged to have a low overall risk of bias. Exploratory tests did not detect marked funnel-plot asymmetry, although their power was limited. Conclusion Baseline NCCT hypodensities were associated with higher odds of subsequent HE, but multiple methodological limitations constrain the evidence certainty (rated as LOW). These markers may support multimodal risk assessment; standardized definitions and prospective validation are required before routine clinical implementation. Systematic review registration https://www.crd.york.ac.uk/PROSPERO/view/CRD420261468570 .
No takes yet. Share an insight, caveat, or question.
Wang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: