Intraoperative hypotension and in-hospital outcomes in older adults undergoing procedure-defined hip or femoral surgery with a same-admission S72 diagnosis: a retrospective cohort study using INSPIRE

Intraoperative hypotension may contribute to adverse outcomes in older adults undergoing procedure-defined hip or femoral surgery with a same-admission S72 diagnosis, but its association with in-hospital outcomes remains uncertain. We conducted a retrospective cohort study using INSPIRE version 1.4.2. Patients aged at least 65 years undergoing orthopaedic hip or femoral procedures with a same-admission S72 diagnosis were eligible; the first eligible operation per patient was retained. The exposure was the time-weighted average (TWA) depth of mean arterial pressure (MAP) below 65 mmHg during anaesthesia, using arterial MAP when available and non-invasive MAP otherwise. The primary outcome combined in-hospital death, ICU admission within 24 h of surgery, or total hospital stay longer than 14 days; death or ICU admission was secondary. Logistic models used six covariates and patients with classifiable outcomes and complete covariates. Supplementary analyses examined monitoring source, postoperative stay, additional confounding and uncertain outcomes. The cohort included 432 patients. The primary outcome was positive in 139, negative in 261 and indeterminate in 32; death or ICU admission was positive in 32, negative in 347 and indeterminate in 53. Among the 139 primary-positive cases, 86 (61.9%) reflected prolonged total stay with both death and ICU admission known negative. All patients had an eligible MAP series. The adjusted OR per 1-mmHg TWA increase was 1.037 (95% CI 0.861–1.250; P = 0.700; 394 patients) for the primary composite and 0.935 (95% CI 0.642–1.361; P = 0.725; 373 patients) for death or ICU admission. Sensitivity analyses, including scenarios allowing for outcomes missing not at random, did not establish a clear association. We did not observe a clear association between TWA MAP below 65 mmHg and either composite in this procedure-defined cohort. Interpretation is further limited because the primary composite was driven largely by prolonged stay. Imprecision, uncertain event timing and limited fracture classification preclude conclusions that hypotension is harmless.