AIMS: To determine the incidence, severity and timing of hyperkalaemia after initiation
of Mineralocorticoid Receptor Antagonists (MRAs), evaluate the predictive value of
common equations for estimation of glomerular filtration rate and to assess compliance
to local prescribing and monitoring guidelines.
METHODS: A retrospective study of patients commenced on an MRA and referred to
the Heart Failure clinic at the Royal Brisbane and Women’s Hospital in the past three
years. The proportion of patients who developed hyperkalaemia over 180 days was
determined, along with a time-to-event analysis. The predictive values of eGFR versus
Cockcroft Gault (CG) for hyperkalaemia were assessed by area under the ROC curve
RESULTS: The study included 149 patients. 118 prescribed spironolactone and 31
eplerenone. Forty-six patients (31%) developed hyperkalaemia, 19 (41%) of whom
recorded a severe result. The median (IQR) time to hyperkalaemia was 37 (16, 87) days.
Analysis of ROCAUC showed no significant differences overall, but sub-analyses
demonstrated CG, using lean body weight in the equation, possessed greater
discriminatory power in patients who developed severe hyperkalaemia (p = 0.035).
CONCLUSIONS: This study highlights the importance of regular potassium monitoring
following commencement of MRAs that should occur at least at monthly intervals and
continue for six months duration. The use of eGFR or CG to estimate GFR is clinically
appropriate however, CG may be the better choice in select populations.