Evidence-based medicine: pitfalls of overlooking safety

The results of large-scale long-term clinical trials can carry enormous weight, especially when the disease treated is common and warrants long-term intervention, and the results show significant benefit without serious adverse effects. However, the trial design needs to be studied with care, because many patients encountered in the clinical setting would not have met the inclusion criteria and the protocol would have included careful monitoring for serious adverse effects. A recent Canadian study of the impact of the Randomized Aldactone Evaluation Study (RALES)¹ highlights the possible adverse consequences of uncritically applying the results of a large scale clinical trial.²

In the RALES study, patients with severe heart failure (NYHA class III or IV) were randomised to spironolactone or placebo.¹ An ACE inhibitor was being used by 95% of the patients. After 24 months the death rate with spironolactone was 30% less than that with placebo. The difference was attributed to a lower risk of death from progression of heart failure and sudden death from cardiac causes with spironolactone. Patients with elevated serum creatinine or potassium were excluded and regular checks of serum potassium were conducted throughout the duration of the study.

The Canadian study of the impact of RALES² tracked a five-fold increase (p < 0.001) in spirono-lactone prescriptions among patients older than 65 years who had recently been hospitalised for heart failure and were taking an ACE inhibitor. Over the same period, there was a three-fold increase in hospitalisation for hyperkalaemia in the same patient population (to 11/1000 patients; p < 0.001). The death rate in hospital from hyperkalaemia also increased by a factor of three (to 2/1000; p < 0.001). There was no significant concurrent decline in hospitalisation for heart failure.

The authors of the Canadian study proposed the following possible explanations for the overall adverse outcome of application of the RALES study in the clinical setting:

• serum potassium not carefully monitored;
• baseline characteristics and conditions developing during therapy that predispose to hyperkalaemia overlooked;
• excessive doses of spironolactone used; and
• dietary intake of potassium increased.

Two other studies^3,4 have delivered cautionary messages for the application of RALES with one finding that 21% of attempts to treat heart failure with spironolactone were halted because of raised serum potassium or creatinine.³ It is noteworthy that of the 27 ADRAC reports of hyperkalaemia associated with the use of spironolactone in combination with an ACE inhibitor or an angiotensin II receptor antagonist, 19 were received since the publication of RALES.

Large scale clinical drug trials have become fundamental to the advancement of knowledge and improvement in clinical practice in medicine, but application to the individual patient requires consideration of the potential for serious adverse effects and assessment of the risk for that patient. Where there are risk factors for adverse reactions, initial use of the drug at low doses with monitoring for changes in key symptoms and laboratory parameters may allow benefit without a serious adverse event.

Reference

1. Pitt B, Zannad F, Remme W et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. New Engl J Med 1999;341:709-17.
2. Juurlink DN, Muhammad MM, Lee DS et al. Rates of hyperkalemia after publication of the Randomized Aldactone Evaluation Study. New Engl J Med 2004;251:543-51.
3. Witham MD, Gillespie ND, Struthers AD. Tolerability of spironolactone in patients with chronic heart failure - a cautionary message. Brit J Clin Pharmacology 2004;58:554-7.
4. Tamirisa KP, Aaronson KD, Koelling TM. Spironolactone-induced renal insufficiency in patients with heart failure. Amer Heart J 2004;148:971-8.

Reference
Australian Adverse Drug Reactions Bulletin
Volume 24, Number 2, April 2005