Abstract Background and aims Despite major advances in stroke diagnosis, autopsy studies continue to demonstrate discrepancies between premortem clinical diagnoses and postmortem findings. Accurate determination of the cause of death is essential for the development of stroke care and treatment. This review focuses on the diagnostic value of autopsy in stroke-related deaths by examining the concordance between clinical diagnoses and neuropathological findings. Methods A systematic literature review was conducted in PubMed, Embase, and the Cochrane Library. Eligible studies from 1985 onwards reported original autopsy findings and compared them with premortem clinical stroke diagnoses. Study selection and data extraction were performed independently by two reviewers using Covidence. Extracted data included study characteristics, sample size, autopsy rates, and reported diagnostic discrepancies. Results Of 291 identified studies, 10 studies were included, comprising 1642 patients who died following a clinical stroke diagnosis, of whom 1068 patients underwent autopsy. Misclassification or undiagnosed stroke aetiology or stroke-related causes of death was reported in 6 studies. Missed intracranial infarcts or hemorrhagic transformation was identified in 5 studies. Clinically unrecognized systemic complications, most commonly pneumonia, pulmonary embolism, myocardial infarction, sepsis, and occult malignancy were reported in 7 studies. Inaccuracies in cause-of-death registration were described in 2 studies, and imaging–pathology discrepancies were reported in 4 studies. Conclusions Autopsy remains a valuable tool for diagnostic clarification in stroke by identifying clinically unrecognized pathology and correcting misclassified causes of death, despite advances in neuroimaging and clinical assessment. Autopsy should be considered if the cause of death and stroke aetiology are not certain or is atypical. Conflict of interest Nothing to disclose
Cetinalp et al. (Fri,) studied this question.