Coronary disease & ACSCohort study
Implications of a Race Term in Glomerular Filtration Rate (GFR) Estimates Used to Predict Acute Kidney Injury (AKI) After Coronary Intervention
- Retrospective cohort study
- registry-based (National Cardiovascular Data Registry (NCDR) CathPCI)
Authors' conclusion
Incorporating a GFR estimate without a Black race term into the NCDR AKI risk prediction model yielded more accurate prediction of AKI risk for Black patients, which has important implications for reducing disparities and benchmarking.
Study context
- Population
- Patients undergoing percutaneous coronary intervention (PCI) in 2020, NCDR CathPCI registry
- Participants
- 455,806 PCI encounters
- Exposure
- AKI risk models using GFR equations with vs without a race term, or serum creatinine alone
- Outcome
- Performance of AKI risk models (2014 NCDR AKI model) predicting AKI after PCI
Main finding
As reported in the abstract
Calibration in Black patients (intercept)
| Group | Calibration intercept |
|---|---|
| Race-agnostic | −0.01 |
| Race term | 0.15 |
- Race-agnostic models reclassified Black patients into higher-risk categories
- 6%
- With a race-agnostic model, AKI occurred in Black patients more often than expected
- 18%
Observational data: associations, not effects.
Original publication
Implications of a Race Term in GFR Estimates Used to Predict AKI After Coronary Intervention. JACC Cardiovasc Interv. Published 25 September 2023. doi:10.1016/j.jcin.2023.07.031
Summarised from the published abstract; every number is checked against it