RACHS-2: A New Risk Stratification Tool for Pediatric Congenital Heart Surgery

Surgeons performing open heart surgery in modern operation room
Researchers sought to develop a publicly available Risk Stratification for Congenital Heart Surgery (RACHS-2) tool that uses ICD-10 administrative data.

Risk Stratification for Congenital Heart Surgery (RACHS-2) is an accurate method of risk stratification for pediatric cardiac surgery, utilizing International Classification of Disease-10th Revision (ICD-10) administrative data. These findings were published in the Journal of the American College of Cardiology.

Researchers sought to develop a method for pediatric congenital heart surgery risk stratification to predict operative mortality utilizing empirical administrative data that is available publicly. To optimize sensitivity and specificity, researchers repeatedly refined the RACHS-2 stratification system via a training data set of Pediatric Health Information Systems claims compared with locally held clinical data from the Society of Thoracic Surgeons-Congenital Heart Surgery (STS-CHS) from 4 pediatric hospitals.

Locally held clinical validation data was obtained from STS-CHS from the New York Congenital Heart Surgeons Collaborative for Longitudinal Outcomes and Utilization of Resources, excluding repetitive data from New York Presbyterian/Morgan Stanley Children’s Hospital and New York-Presbyterian/Komansky Children’s Hospital. New York State Medicaid claims were used as a second administrative data source validation for RACHS-2. With only 1.0% false positive, 99.6% of total congenital heart surgery registry cases were captured by RACHS-2.

The investigators used statistical logistic regression to compare the ability of RACHS-2 to predict operative mortality vs STAT Mortality Categories. Clinical data included patients aged less than 18 years who had received cardiopulmonary bypass or non-cardiopulmonary bypass congenital heart surgery between October 1, 2015, and December 31, 2019. Administrative data came from the same time period from any patients with an ICD-10 code indicating cardiac surgery.

RACHS-2 accurately predicted operative mortality in training data (0.76, adjusted for RACHS-2; 95% CI, 0.72-0.80), and validation data (0.84, adjusted for RACHS-2; 95% CI, 0.80-0.89), and for operative mortality in training data adjusted for STAT Mortality Categories (0.75; 95% CI, 0.71-0.79), and validation data adjusted for STAT Mortality Categories (0.84; 95% CI, 0.79-0.89).

The researchers urge caution when using the RACHS-2 code for the liberal assessment of individual procedures, as not all ICD-10 codes map with precision.

The statistical code for RACHS-2 is publicly available from the corresponding study author upon request. “This work represents a novel identification and risk stratification method for pediatric cardiac surgery using ICD-10 administrative data,” the study authors wrote. “Creation and sharing of this tool expand researchers’, policy makers’, and hospitals’ ability to use administrative data sources to assess and improve outcomes, resource requirements, reimbursement, and health disparities for children with congenital heart disease.”  


Allen P, Zafar F, Mi J, et al. Risk stratification for congenital heart surgery for ICD-10 administrative data (RACHS-2). J Am Coll Cardiol. Published online February 8, 2022.