The Time-Effect Relationship between Time to Surgery and In-Hospital Postoperative Pneumonia in Older Patients with Hip Fracture.

Abstract

Introduction: Pneumonia is a common and devastating complication following hip fracture surgery in older patients. Time to surgery is a potentially modifiable factor associated with improved prognosis, and we aim to quantify the time-effect relationship between time to surgery and in-hospital postoperative pneumonia (IHPOP) and identify the effect of delayed surgery on the risk of IHPOP.

Methods: We analyzed clinical data of older hip fracture patients (≥60 years) undergoing surgical treatments at a tertiary referral trauma center between 2015 and 2020. Restricted cubic spline (RCS) was used to fit the time-effect relationship between time to surgery and IHPOP. Based on the results of RCS, we divided patients into two groups of "early surgery" and "delayed surgery." A 1:1 propensity score matching (PSM) analysis and multivariate conditional logistic regression analysis were performed to minimize the selection bias and determine the association magnitude. Subgroup analysis was conducted to assess potential interaction effects between delayed surgery and common risk factors for IHPOP.

Results: 3,118 eligible patients were included. The RCS curve showed an inverse S-shape trend and the relative risk of IHPOP decreased in the range of days 2-3 and increased on day 1 and day 3 or more post-injury, with the lowest point on day 3. PSM yielded 1,870 matched patients and delayed surgery (>3 days) was identified to be independently associated with IHPOP (relative ratio, 1.66; 95% confidence interval, 1.12-2.46; p value, 0.011). We observed positive interaction effects between delayed surgery and age of 80 years or more, female gender, COPD, heart disease, ASA score ≥3, anemia, and hypoproteinemia.

Conclusion: The relative risk of IHPOP decreased in the range of 2-3 days and increased on day 1 and day 3 or more post-injury. Delayed surgery (>3 days) was identified to be independently associated with a 1.66-fold increased risk of IHPOP.