Prediction of perioperative cardiac complications in patients with proximal femoral fracture

Abstract

Listen

Translate article

Background. Prediction of cardiac complications following orthopaedic and trauma surgery is necessary to improve the quality of treating the elderly patients. Objective. To assess the effectiveness of prognostic scores of perioperative cardiac risk in patients with proximal femoral fracture. Material and methods. We retrospectively reviewed 918 hospital patients with proximal hip fracture from January, 1 2018 to December, 31 2019. Perioperative cardiac risks were assessed using the Goldman Risk Index, Revised Cardiac Risk Index (Lee Index) and Gupta Perioperative Cardiac Risk Index. Results. Cardiac complications occurred in 7 (0.76%) of 918 patients, 6 (0.65%) patients developed acute myocardial infarction, 1 (0.11%) suffered from complete atrioventricular block. Receiver operating characteristic (ROC) curve analysis of the relationship between the time from injury to surgery and cardiovascular complications did not give statistically significant results (AUC (area under a curve) = 0.574, 95% CI (confidence interval): 0.352–0.796). When compared the presence of cardiac complications with the Lee Criteria predictions, significant differences were revealed (p = 0.007), and the Goldman Index data were not statistically significant (p = 0.151). The area under the ROC curve of the corresponding relationship between the prognosis of cardiac complications and the Gupta Index was 0.782 with 95% CI: 0.574–0.991 (p = 0.017), the sensitivity and specificity of the model were 83.3% and 70.4%, respectively. Conclusion. The Goldman Index and Lee Index have no significant value for predicting perioperative cardiac complications in patients with proximal femoral fracture. The Gupta Index has an acceptable level of sensitivity and specificity in predicting cardiac complications.