Abstract
Significant progress has been made in the prevention and management of hypercoagulation. Unfractionated heparin (UF) and low molecular weight heparin (LMWH) are indicated for acute cerebral vein thrombosis with or without intracranial hemorrhage (ICH). A recent meta-analysis of four trials comparing UF and LMWH aimed to evaluate the efficacy and safety of both agents in terms of disability, intracranial hemorrhage, and mortality. However, several methodological aspects of the meta-analysis warrant further discussion. It appears that the disability outcome was not sufficiently harmonized by design or statistical standardization, some inputs could not be validated, incorrect statistical analyses were performed, major results could not be replicated, and conclusions were not supported by the statistical results. The conclusion of a statistically significant reduction in mortality is not supported by the data.
•Disability scales were not harmonized or standardized.•Unable to replicate findings of mortality and disability.•Incorrect statistical analysis by selecting random effect model.•Non-significant heterogeneity shouldn't be used as surrogate for model selection.•Conclusion couldn't be supported by statistical results.