Emergency laparotomy (EL) is an increasingly common procedure in the elderly. Factors associated with mortality in the subpopulation of frail patients have not been thoroughly investigated. Sarcopenia has been investigated as a surrogate for frailty and poor prognosis. Our primary aim was to evaluate the association between easily measured sarcopenia parameters and 30-day postoperative mortality in elderly patients undergoing EL. Length of stay (LOS) and admission to an intensive care unit were secondary end points.
We conducted a retrospective cohort study, over a 5-year period, of patients aged 65 y and older who underwent EL at a tertiary university hospital. Sarcopenia was evaluated on admission computed tomography scan by two methods, first by psoas muscle attenuation and second by the product of perpendicular cross-sectional diameters (PCSDs). The lowest quartile of PCSDs and attenuation were defined as sarcopenic and compared with the rest of the cohort. Attenuation was stratified for the use of contrast enhancement. Multivariant logistic regression was performed to determine independent risk factors.
During the study period, 403 patients, older than 65 y, underwent EL. Of these, 283 fit the inclusion criteria and 65 (23%) patients died within 30 d of surgery. On bivariate analysis, psoas muscle attenuation, but not PCSDs, was found to be associated with 30-day mortality (OR = 2.43, 95% CI = 1.34-4.38, P = 0.003) and longer LOS (35.7 d versus 22.2 d, Δd 13.5, 95% CI = 6.4-20.7, P < 0.001). In a multivariate analysis, psoas muscle attenuation, but not PCSDs, was an independent risk factor for 30-day postoperative mortality (OR = 2.35, 95% CI = 1.16-4.76, P = 0.017) and longer LOS (Δd = 14.4, 95% CI = 7.7-21.0, P < 0.001). Neither of the sarcopenia parameters was associated with increased admission to an intensive care unit.
Psoas muscle attenuation is an independent risk factor for 30-day postoperative mortality and LOS after EL in the elderly population. This measurement can inform clinicians about the operative risk and hospital resource utilization.