The effect of anesthetic technique on mortality and major morbidity after hip fracture surgery: a retrospective, propensity-score matched-pairs cohort study

Introduction Spinal anesthesia (SA) has physiological benefits over general anesthesia (GA), but there is insufficient evidence regarding a mortality benefit. We performed a retrospective propensity score-matched cohort study to evaluate the impact of anesthetic technique on mortality and major morbidity in patients undergoing hip fracture surgery. Materials and methods Clinical, laboratory and outcome data were extracted from electronic databases for patients who underwent hip fracture surgery over a 13-year period at the University Health Network in Toronto, Ontario, Canada. The anesthetic technique was documented (SA or GA), and the primary outcome was 90-day mortality. Secondary outcomes included mortality at 30 and 60 days, hospital length of stay, pulmonary embolism (PE), major blood loss and major acute cardiac events. A propensity-score matched-pair analysis was performed following a non-parsimonious logistic regression model. Results Of the 2591 patients identified, 883 patients in the SA group were matched to patients in the GA group in a 1:1 ratio. There was a weak association between SA and lower 90-day mortality (risk ratio (RR) 0.74, 95% CI 0.52 to 0.96, 99% CI 0.48 to 1.00, p=0.037). SA was also associated with a lower incidence of both PE (1.3% vs 0.5%, p<0.001) and major blood loss (7.7% vs 4.8%, p<0.001) and a shorter hospital length of stay by about 2 days (median 11.9 vs 10 days, p=0.024). There was no difference in major cardiac events or mortality at 30 and 60 days. Discussion This propensity-score matched-pairs cohort study suggests that SA is weakly associated with a lower 90-day mortality following hip fracture surgery. SA was also associated with improved morbidity evidenced by a lower rate of PE and major blood loss and a shorter hospital length of stay. Given the retrospective nature of the study, these results are not proof of causality.