Clinical characteristics of elderly hip fracture patients with chronic cerebrovascular disease and construction of a clinical predictive model for perioperative pneumonia

Abstract

IntroductionThe recovery of cerebrovascular disease (CVD) will increase the incidence of perioperative pneumonia (POP). However, there is limited research on POP in elderly patients with hip fractures complicated by CVD. Therefore, our research focuses on the following two issues: (1) What are the clinical features of elderly patients with hip fractures combined with CVD? (2) What are the predictive factors for the occurrence of POP in such patients? HypothesisMale, femoral neck fracture and hypoalbuminemia can be predictive factors for the development of POP after hip fracture in CVD patients. Material and methodsThis is a nested case-control study that included patients aged 65 to 105 years with CVD who had a hip fracture between January 2021 and January 2023. According to the occurrence of POP, they were divided into case group and control group. Collecting data includes demographic information, clinical data, and surgical information. Least absolute shrinkage and selection operator (LASSO) regression and multivariate logistic regression analyses were used to select variables. The constructed predictive model was transformed into a nomogram. Predictive performance was assessed using the area under the receiver operating characteristic curve (AUC), calibration curves and decision curve analysis (DCA). ResultsWe ultimately included 714 patients, 69.3% female, with a median age of 80 years. Asymptomatic cerebral infarction (ACI) is the most common CVD (55.7%). More patients developed intertrochanteric fractures than femoral neck fractures (57.1 vs. 42.9%). In total, 606 patients (84.9%) underwent surgery. The most common perioperative complications were anemia (76.9%) and hypoalbuminemia (71.8%). POP (20.0%) was more common preoperatively (89.5%). Factors such as fracture type, surgical wait time, implant used for surgery, and anesthesia type did not differ between the presence or absence of postoperative pneumonia. 143 patients with POP served as the case group. Five hundred and seventy one patients did not develop POP and served as the control group. The predictors of POP were male (OR 1.699,95%CI 1.150–2.511, p<0.05), femoral neck fracture (OR 2.182,95%CI 1.491–3.192, p<0.05), and hypoalbuminemia (OR 3.062, 95%CI 1.833–5.116, p<0.05). This model has good discrimination, calibration, and clinical practicality. DiscussionIn this study, we constructed a clinical prediction model for the occurrence of POP in CVD combined with hip fracture in the elderly, with risk factors including gender, fracture type and perioperative hypoproteinemia. Therefore, we can take effective preventive measures against the occurrence of POP in patients with these factors in our clinical work. Level of proofIV; nested case-control study.