Prehospital death in aneurysmal subarachnoid haemorrhage. A systematic review and meta-analysis of population-based studies
Le résumé fourni par la source
A substantial share of people with aneurysmal subarachnoid haemorrhage (SAH) die before they reach hospital. These deaths are absent from hospital series and from treatment trials, so estimates of how lethal SAH is, and of how much survival has improved, rest on the surviving minority of patients. The only previous synthesis of this question pooled 18 population-based studies and estimated a rate of 12 percent (Huang and van Gelder, Neurosurgery 2002). Several nationwide register studies with linkage to national death records have appeared since, and the number of eligible study bases has more than doubled. We conduct a systematic review and meta-analysis of the proportion of incident SAH cases that die before hospital admission. The estimand is prehospital deaths divided by all incident SAH cases arising in a defined source population. This is not case fatality, and denominators restricted to admitted patients are ineligible. Our central question concerns ascertainment. Identifying an SAH death that occurs in the community usually requires a postmortem examination, and postmortem examination rates in the general population have fallen in most countries over recent decades. Reported prehospital death rates may therefore be lower than the true rates, and any apparent decline over time may be an artefact of falling postmortem examination rates. No previous review has tested this. We search PubMed, Scopus and the Cochrane Library without language, date or country limits, and hand-search reference lists. We pool proportions with a random-effects model, assess risk of bias in four domains covering out-of-hospital ascertainment, diagnostic accuracy, outcome measure and study base definition, and examine the community postmortem examination rate and calendar time as moderators in meta-regression. We also pool within-study-base comparisons by sex, by aneurysm location, and by age at the event. We expect the pooled rate to be higher than previously reported, to increase with the completeness of out-of-hospital ascertainment, and to show no decline over time once ascertainment is accounted for.
Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.