Acute myocardial infarction management by kidney function across hospitals in England

Compared with the general population, the risk of acute myocardial infarction (AMI) and associated worse outcomes in people with chronic kidney disease (CKD) is substantially increased. Guidelines recommend early invasive cardiac management strategies (angiography and percutaneous coronary intervention [PCI] if indicated) for most ST-elevation myocardial infarction (STEMI) cases and for non-STEMI cases judged to be at high risk of mortality, irrespective of the presence of CKD.

Some analyses suggest that people with reduced kidney function may benefit from invasive cardiac management despite the risks, but evidence from randomised controlled trials is lacking. This may partly explain the variation in invasive versus conservative management strategies, particularly for NSTEMI,  seen in hospitals serving the general population in England.

The present cross-sectional study by Bidulka and colleagues set out to describe variation in AMI management for people with reduced kidney function in England at the hospital and individual levels. Data captured in the Myocardial Ischaemia National Audit Project (MINAP) for people hospitalised for AMI (STEMI or NSTEMI) in English hospitals from 2014 to 2019 was included. Kidney function was defined using eGFR derived from the serum creatinine measured within 24 hours of AMI hospitalisation. 

In hospital-level analyses, 361,259 people with a first hospitalisation for AMI at 209 hospitals were included. Of these, 120 did not offer PCI, 38 offered it sometimes and 51 offered it all of the time. At individual-level analyses, 292,572 people with complete covariable data at 207 hospitals were included. 

At the hospital level, substantial variation was observed in the proportion reported as treated with invasive versus conservative cardiac management, particularly for people hospitalised with NSTEMI with reduced kidney function. There was little variation in AMI management for those hospitalised for STEMI with no evidence of reduced kidney function. 

At the individual level, there was a relative decrease in the odds of 

invasive versus conservative cardiac management by worsening kidney function in hospitals in which PCI was always available. After using multivariable logistic regression to derive adjusted predicted probabilities, lower adjusted predicted probabilities of being treated with invasive cardiac treatment were observed with decreasing eGFR range, particularly for NSTEMI cases. 

The findings demonstrate substantial variation in the treatment for NSTEMI of similar individuals, particularly those with reduced kidney function, depending on the hospital that they are admitted to. This needs to be explored further to understand the comparative effectiveness of invasive versus conservative NSTEMI management strategies in people with reduced kidney function and/or CKD.

The full article can be read here.