BACKGROUND: ASA-PS is a widely used perioperative health assessment method, but with poor reproducibility. A novel objective, pediatric-specific risk classification system based on Neurological, Airway, Respiratory, Cardiovascular, Other categories and Surgical Severity (NARCO-SS) has been validated in only one US center. OBJECTIVE: Independent external validation of the NARCO-SS and comparison with the ASA-PS in predicting perioperative outcomes. METHODS: Preoperative ASA-PS and NARCO-SS scores were assigned to 387 children by pediatric anesthesia consultants at a tertiary care center in India and predefined perioperative adverse events and escalation of care recorded. Spearman's correlations determined the relationship between outcomes and scores and kappa statistics for interobserver reliability. The predictive performance of the ASA-PS and NARCO-SS was evaluated by the area under the receiver operating characteristics curves (AUC-ROC) for discrimination and Pearson's chi-square for calibration. RESULTS: NARCO-SS and ASA scores had significant Spearman's correlation coefficients with perioperative outcomes and moderate interobserver reliability. The NARCO-SS showed greater discrimination than the ASA-PS (AUC-ROC 0.778 vs 0.710 for escalation of care and 0.822 vs 0.724 for adverse events, P < 0.01). However, both scores had poor calibration (Pearson's chi-square, P < 0.0001). Individual NARCO-SS categories for neurological and airway lacked statistically significant Spearman's correlations. CONCLUSIONS: NARCO-SS is a valid risk stratification tool that is better than the ASA-PS in discriminating children with adverse perioperative outcomes. The poor calibration of both scores suggests neither can reliably predict perioperative outcomes in individual patients. Modification of neurological and airway categories may improve the predictive accuracy of the NARCO-SS.
BACKGROUND:ASA-PS is a widely used perioperative health assessment method, but with poor reproducibility. A novel objective, pediatric-specific risk classification system based on Neurological, Airway, Respiratory, Cardiovascular, Other categories and Surgical Severity (NARCO-SS) has been validated in only one US center. OBJECTIVE: Independent external validation of the NARCO-SS and comparison with the ASA-PS in predicting perioperative outcomes. METHODS: Preoperative ASA-PS and NARCO-SS scores were assigned to 387 children by pediatric anesthesia consultants at a tertiary care center in India and predefined perioperative adverse events and escalation of care recorded. Spearman's correlations determined the relationship between outcomes and scores and kappa statistics for interobserver reliability. The predictive performance of the ASA-PS and NARCO-SS was evaluated by the area under the receiver operating characteristics curves (AUC-ROC) for discrimination and Pearson's chi-square for calibration. RESULTS:NARCO-SS and ASA scores had significant Spearman's correlation coefficients with perioperative outcomes and moderate interobserver reliability. The NARCO-SS showed greater discrimination than the ASA-PS (AUC-ROC 0.778 vs 0.710 for escalation of care and 0.822 vs 0.724 for adverse events, P < 0.01). However, both scores had poor calibration (Pearson's chi-square, P < 0.0001). Individual NARCO-SS categories for neurological and airway lacked statistically significant Spearman's correlations. CONCLUSIONS:NARCO-SS is a valid risk stratification tool that is better than the ASA-PS in discriminating children with adverse perioperative outcomes. The poor calibration of both scores suggests neither can reliably predict perioperative outcomes in individual patients. Modification of neurological and airway categories may improve the predictive accuracy of the NARCO-SS.
Authors: Oguz Akbilgic; Max R Langham; Arianne I Walter; Tamekia L Jones; Eunice Y Huang; Robert L Davis Journal: PLoS One Date: 2018-01-19 Impact factor: 3.240