To explore the risk factors of postoperative noninfectious fever in patients with femoral intertrochanteric fracture, and to establish a model for predicting the risk.

The patients with intertrochanteric fractures who underwent proximal femoral nail antirotation (PFNA) surgery in Northern Jiangsu People’s Hospital from January 2019 to May 2021 were retrospectively screened. Inclusion criteria: preoperative diagnosis of unilateral closed intertrochanteric fracture of femur; PFNA operation was performed; no previous open hip trauma or surgical history; complete data. Exclusion criteria: preoperative fever; obvious infection before and after operation; preoperative diagnosis of febrile diseases; patients undergoing multi-sites combined surgery; pathological fracture. Sex, age, body mass index (BMI), hypertension, diabetes, smoking history, drinking history, anesthesia methods, American Association of Anesthesiologists (ASA)score, operation time, intraoperative blood loss, catheterization, the perioperative blood transfusion, preoperative nebulization, deep vein thrombosis, the preoperative hemoglobin, white blood cells, albumin, and C-reactive protein, as well as the postoperative hemoglobin( Hb), white blood cells, albumin, C-reactive protein on the first day after operation were selected and recorded. Uni-and multivariate logistic regression analyses were carried out to determine the independent risk factors, and a nomogram risk prediction model was established to predict the risk of postoperative noninfectious fever.

A total of 316 patients were collected, of which 103 had non-infectious fever. The results showed that male (χ²=8.509), general anesthesia (χ²=7.058), longer operation time (χ²=90.225), more blood loss (χ²=24.497), perioperative blood transfusion (χ²=17.005) and low albumin on the first day after operation (χ²=5.925) were independent risk factors leading to postoperative noninfectious fever (all P<0.05). Taking six factors into account, a nomogram risk prediction model was established. The result of calibration curve showed that the calibration curve fitted well with the ideal curve, and the predicted value was consistent with the measured value. C-index was used to evaluate the calibration curve, and the result showed the nomogram model had good prediction ability [C-index=0.870, 95% confidence interval(CI) ( 0.831, 0.910), the corrected C-index=0.858].

The influencing factors of noninfectious fever after PFNA mainly include male, general anesthesia, long operation time, large amount of bleeding, perioperative blood transfusion history and low albumin on the first day after operation. The nomogram model has a good ability to predict the risk of noninfectious fever after PFNA. However, these conclusions need the support of large samples and multi-center research.