Development of a nomogram for postoperative pneumonia after total hip or knee arthroplasty: a retrospective cohort study

Abstract Background: Postoperative pneumonia (POP) is a perioperative complication with a high incidence and mortality after total hip arthroplasty (THA) or total knee arthroplasty (TKA). This study aims to develop a prediction nomogram facilitating early detection of POP in patients undergoing THA or TKA. Method: The study conducted a retrospective analysis of adult patients who underwent THA or TKA at Fujian Provincial Hospital from January 1, 2012, to December 31, 2021. Patients were randomly divided into either the training or validation cohort. We employed the least absolute shrinkage and selection operator (LASSO) regression and multivariate logistic regression model to identify independent variables and develop a nomogram. Both discrimination and calibration were assessed through bootstrapping with 1000 resamples. Result: A total of 1344 patients with THA or TKA were included in this study. A regression analysis revealed the independent predictors for POP were age (OR 1.051, 95% CI 1.023-1.080), congestive heart failure (CHF) (OR 263.239, 95% CI 9.895-7003.231), chronic obstructive pulmonary disease (OR 55.248, 95% CI 7.249-421.092), hypertension (OR 3.318, 95% CI 1.497-7.358), albumin level (OR 0.784, 95% CI 0.732-0.839), and operative blood loss (OR 1.007, 95% CI 1.005-1.009).The area under the receiver operating characteristic (ROC) curve (AUC) and the concordance index in the training cohort were 0.861 and 0.893 respectively. The Hosmer−Lemeshow test signaled a good fit for the model (P=0.190). The calibration curve of the nomogram leaned closely towards the ideal diagonal line. Moreover, the decision curve analysis (DCA) portrayed a satisfactory positive net benefit. The validation cohort's results also affirmed the reliability of this prediction nomogram Conclusion: A nomogram was developed based on perioperative factors, to anticipate the occurrence of POP in patients who undergo THA or TKA. This model could provide a precise and highly effective predictive tool for early identification and intervention of POP.