The Relative Impact of Specific Postoperative Complications on Older Patients Undergoing Hip Fracture Repair

Hip fractures affect a vulnerable population and are associated with high rates of morbidity, mortality, and resource utilization. Although postoperative complications are a known driver of mortality and resource utilization, the comparative impacts of specific complications on outcomes is unknown. This study assessed which complications are associated with the highest effects on mortality and resource utilization for older patients who undergo hip fracture repair.

Patients ≥ 65 years of age who underwent hip fracture repair during 2016-2017 included in the Hip Fracture Targeted ACS NSQIP (American College of Surgeons National Surgical Quality Improvement Program) database populated the data set. Prolonged hospitalization (≥ 75th percentile) and 30-day mortality and readmission were the primary outcomes. Population attributable fractions (PAFs) were used to quantify the anticipated reduction in the primary outcomes that would result from complete prevention of 10 postoperative complications.

Of 17,755 patients across 117 hospitals, 70.9% were female, 26.0% were over age 90, 22.8% had an American Society of Anesthesiologists (ASA) score of 4-5, and 53.9% presented with an intertrochanteric fracture. Postoperative delirium affected 29.8% of patients and was associated with death (PAF 18.0%; 95% confidence interval [CI] = 13.2-22.5), prolonged hospitalization (PAF 14.3%; 95% CI = 12.7-15.8), and readmission (PAF 15.0%; 95% CI = 11.3-18.6). Pneumonia affected 4.1% of patients and was associated with death (PAF 10.9%; 95% CI = 8.9-12.8), prolonged hospitalization (PAF 4.0%; 95% CI = 3.5-4.5), and readmission (PAF 9.1%; 95% CI = 7.5-10.7). The impact of the other eight complications was comparatively small.

Postoperative delirium and pneumonia are the highest-impact complications for older hip fracture repair patients. These complications should be prioritized in quality improvement efforts that target this patient population.