Sang-Wook Lee1, Keon-Sik Kim2, Sung-Wook Park2, Jaewon Kim2, Jeong-Hyun Choi2, Sangho Lee3, Kyoung-Woon Joung3, In-Cheol Choi3. 1. Department of Anesthesiology and Pain Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea, sangwooklee20@gmail.com. 2. Department of Anesthesiology and Pain Medicine, Kyung Hee University Hospital, Kyung Hee University College of Medicine, Seoul, Republic of Korea. 3. Department of Anesthesiology and Pain Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea.
Abstract
BACKGROUND: Predicting preoperative frailty risk in emergency surgery is difficult with limited information because preoperative evaluation is not commonly performed properly. A recent study attempted to predict preoperative frailty risk using only diagnostic and surgical codes that can be extracted from the electronic medical records system. OBJECTIVE: This study aimed to validate whether the prediction model of preoperative frailty risk presented in the previous study is well applied to other medical hospitals' data. METHODS: This is a retrospective cohort study including 1,557 patients (≥75 years old) who were admitted to a single institution for emergency operations between January 1, 2010, and December 31, 2019, for study analysis. The Charlson comorbidity index, Hospital Frailty Risk Score, and the recently developed Operation Frailty Risk Score (OFRS) were calculated using the patient's diagnostic and operation codes. The predictive performances of these calculated risk scores and the American Society of Anesthesiologists-Physical Status classification for postoperative 90-day mortality were compared by using the receiver operating characteristic curve analysis. FINDINGS: The predictive performance of the OFRS, Charlson comorbidity index, American Society of Anesthesiologists-Physical Status, and Hospital Frailty Risk Score for postoperative 90-day mortality was 0.81, 0.630, 0.699, and 0.549 on a c-statistics basis, respectively. CONCLUSIONS: The OFRS using diagnostic and operation codes may show the best predictive performance for 90-day mortality compared to other risk scores, and it can be the clinically applicable model to evaluate the preoperative frailty risk in elderly patients undergoing emergency surgery.
BACKGROUND: Predicting preoperative frailty risk in emergency surgery is difficult with limited information because preoperative evaluation is not commonly performed properly. A recent study attempted to predict preoperative frailty risk using only diagnostic and surgical codes that can be extracted from the electronic medical records system. OBJECTIVE: This study aimed to validate whether the prediction model of preoperative frailty risk presented in the previous study is well applied to other medical hospitals' data. METHODS: This is a retrospective cohort study including 1,557 patients (≥75 years old) who were admitted to a single institution for emergency operations between January 1, 2010, and December 31, 2019, for study analysis. The Charlson comorbidity index, Hospital Frailty Risk Score, and the recently developed Operation Frailty Risk Score (OFRS) were calculated using the patient's diagnostic and operation codes. The predictive performances of these calculated risk scores and the American Society of Anesthesiologists-Physical Status classification for postoperative 90-day mortality were compared by using the receiver operating characteristic curve analysis. FINDINGS: The predictive performance of the OFRS, Charlson comorbidity index, American Society of Anesthesiologists-Physical Status, and Hospital Frailty Risk Score for postoperative 90-day mortality was 0.81, 0.630, 0.699, and 0.549 on a c-statistics basis, respectively. CONCLUSIONS: The OFRS using diagnostic and operation codes may show the best predictive performance for 90-day mortality compared to other risk scores, and it can be the clinically applicable model to evaluate the preoperative frailty risk in elderly patients undergoing emergency surgery.