Prevention & lipidsCohort study

Frailty assessment and the risk of major adverse cardiovascular events after elective surgery

Eur Heart J Acute Cardiovasc CarePublished

  • retrospective cohort study
  • propensity-matched 1:1

Authors' conclusion

Frailty before non-cardiac surgery was associated with increased 30-day MACE. Incorporating frailty into RCRI improves risk prediction, particularly for patients with low baseline cardiac risk.

Study context

Population
Older US adults insured through Medicare undergoing non-cardiac surgery between January 2015 and September 2021mean age 73 years, 43% male, 22.4% had intermediate/high frailty
Participants
6,399,535 adults
Exposure
intermediate/high frailty vs low-frailty
Outcome
MACE:
  • Mortality
  • Myocardial infarction
  • Stroke
  • Heart failure
Follow-up
30-day

Main finding

As reported in the abstract

Odds ratio

1.19

95% CI 1.17–1.21

P < 0.001

intermediate/high frailty vs low-frailty
OR 1.19 (95% CI 1.17–1.21); P < 0.001
experienced 30-day major adverse cardiovascular events (MACE) (overall cohort)
2.97%

Observational data: associations, not effects.

Further findings

OutcomeOR (95% CI)P value
Mortality · 30-day · propensity-matched cohort1.53 (1.49–1.58)P < 0.001
30-day MACE by baseline cardiac risk (subgroup) — intermediate/high frailty vs low-frailty · RCRI 0-11.28 (1.25–1.31)
30-day MACE by baseline cardiac risk (subgroup) — intermediate/high frailty vs low-frailty · RCRI 2-41.06 (1.03–1.09)
OutcomeEstimateP value
Predictive performance for 30-day MACE, RCRI with frailty vs. RCRI — RCRI with frailty vs RCRIC-index 0.773 vs. 0.758P < 0.001
Predictive performance for 30-day MACE, RCRI with frailty vs. RCRI — RCRI with frailty vs RCRIBrier score 0.0277 vs. 0.0290
  • 30-day MACE by baseline cardiac risk (subgroup): Pinteraction < 0.001

Original publication

Frailty assessment and the risk of major adverse cardiovascular events after elective surgery. Eur Heart J Acute Cardiovasc Care. Published 19 August 2026. doi:10.1093/ehjacc/zuag108

Summarised from the published abstract; every number is checked against it