Author + information
- ↵*Reprint requests and correspondence:
Dr. Gary S. Francis, Cleveland Clinic Foundation, Department of Cardiovascular Medicine, F25, 9500 Euclid Avenue, Cleveland, Ohio 44195, USA.
Over the past decade we have witnessed major strides in our ability to use drugs to impede or even reverse the progression of heart failure (HF). With impressive results from well-designed clinical trials, several new drugs have been introduced into the armamentarium against chronic HF, and evidence-based medicine has emerged as the vehicle by which we deliver new therapy. However, the construct of “evidence-based medicine” has three major assumptions: 1) it assumes a timely and accurate delivery of new knowledge from research studies to practicing physicians; 2) it assumes competency of the practicing physicians to adequately “assimilate” new knowledge and apply this knowledge appropriately to their own practice; and 3) it assumes that the findings from the research studies will apply in real-world scenarios. With more than 10,000 new randomized controlled trials published in the past decade, it is no small task for physicians to filter and assimilate detailed, new knowledge.
Spironolactone, a nonspecific aldosterone receptor antagonist, is an example of a drug that has recently been “re-introduced” as a treatment for HF, following the positive results observed in a landmark clinical trial (1). At a time when high-dose angiotensin-converting enzyme inhibitor and beta-adrenergic receptor blockers were at the high ground of medical therapy for patients with chronic HF, the Randomized Aldosterone Evaluation Study (RALES) demonstrated a marked mortality reduction of 30% in patients with advanced HF (New York Heart Association [NYHA] functional class III to IV) when only a modest dose of 12.5 to 25 mg/day dose of spironolactone was added to these two classes of drugs (1). Importantly, there was also a dose-dependent improvement in ventricular remodeling and exercise tolerance during treatment with spironolactone (2). As the authors of the RALES study note, spironolactone should be started at 12.5 mg/day and cautiously increased to 25 mg/day. Patients in the study required frequent follow-ups so that serum potassium and creatinine levels could be carefully monitored. The patients who entered RALES were all NYHA functional class IV or III patients who were previously class IV.
Since the publication of the RALES trial, it is apparent that the practicing community has widely embraced spironolactone to treat patients with HF (3). The RALES trial was first presented at the American Heart Association 71st Scientific Session on November 11, 1998. The striking results sparked an extensive dialogue of how spironolactone should be incorporated into the treatment of HF. However, the widespread acceptance of spironolactone was unique, as this was an old drug, simple to use, and physicians had experience with it. However, as indicated in the report by Bozkurt et al. (4)in this issue of the Journal, many patients who were being treated with spironolactone were quite dissimilar to the patients in the RALES trial. Bozkurt et al. (4)noted an increase in complication rates immediately following the prepublication of the RALES trial in July 1999, including inappropriate patient selection, inaccurate prescriptions, and poor monitoring of adverse events. In particular, injudicious use of spironolactone has led to serious cases of renal insufficiency and hyperkalemia. This was likely to be particularly a problem in patients with diabetes and in patients in whom angiotensin-converting enzyme inhibitor or angiotensin receptor blocker doses were not adjusted for the concomitant use of spironolactone. Nonsteroidal anti-inflammatory drugs also complicate the use of spironolactone. We have witnessed a very similar experience to Bozkurt et al. (4), albeit completely anecdotal.
The findings from this study uncover an emerging problem in the current age of polypharmacy and the increasingly more complex elderly patients. There seems to be an impediment to the translation of clinical evidence into real-world practice. It would be naive to simply put blame on those practicing physicians who have generated these medical “errors.” Indeed, the key findings from the Institute of Medicine’s report on medical errors indicate that errors may often be caused by system failures rather than individual negligence or incompetence (5). The early adoption of spironolactone (within the first few months of the publication) observed by Bozkurt et al. (4)demonstrates the good intentions of these practicing physicians. The challenge here is to resolve potential system failures in order to fulfill the assumptions for an effective practice of evidence-based medicine. Several issues can be raised from this study.
The first issue is the clarity of the publication. Although careful dose-finding studies were done before RALES, especially with regard to the hyperkalemia and renal insufficiency problems (6), most readers of the RALES study may not be familiar with these findings and discussions. Bozkurt et al. (4)have pointed out that diabetic patients (who are known to be at risk for such adverse outcomes) were not mentioned in the RALES publication. We all need to pay more attention to the clarity of the publication, particularly regarding the potential impact of patient entry and exclusion criteria.
The second issue is the limitations of the current paradigm of acquisition of new knowledge by practicing physicians. Although this was not the first report of adverse events in using spironolactone following the publication of the RALES trial (7,8), the authors picked a critical study period (within the first few months of the prepublication release of RALES) to amplify the notion that we have to pay more attention to the “learning curve” phenomenon. As inappropriate patient selection and application appeared to be a common theme in these reports (4,8,9), questions can also be raised as to whether busy physicians were even aware of the patient selection and dose criteria outlined in the RALES publication. In fact, a recent Canadian survey of 521 practicing physicians revealed that less than half of the respondents take the effort to review primary research studies (either reading the primary publication or indirectly from published summaries) in an attempt to practice evidence-based medicine (10). It is conceivable that only a minority of busy practicing physicians would have carefully studied the paper (or the accompanying editorial) before prescribing spironolactone to their patients with HF, although there is no direct evidence to support this. Therefore, there is a need to rethink how the process of scientific publication can be better adapted to facilitate appropriate assimilation of new knowledge. Specialty organizations and individual specialists should also take more responsibilities in disseminating clinical evidence to their peers. To extend this issue further, busy clinicians may not have the appropriate time or skills to “assimilate” new knowledge from published studies. The current paradigm of continuing medical education and recertification for medical boards may need to be reconsidered in the setting of the exponential increase in medical information and the proportionally less time to learn.
The third issue concerns the expertise of practicing physicians in applying a new therapeutic strategy to a patient with HF, especially when there is a relatively small margin of error. The differences among specialties in the appropriateness of spironolactone prescriptions have previously been reported (9). However, this does not directly translate to the notion that cardiologists and HF specialists should treat all patients with HF. Instead, the implementation of a multidisciplinary team approach such as the one described by the authors should facilitate an effective implementation of a drug such as spironolactone.
Finally, the need for anticipation of potential adverse events is necessary in any form of therapy, whether old or new. Patients in real-world practice are universally older, taking more drugs for more comorbid conditions, and may have confounding nonmedical issues (such as drug compliance) that affect their clinical outcomes. The false sense of security in using a well-known drug such as spironolactone has also likely contributed to its injudicious use and subsequent adverse outcomes. New drugs have the luxury of a wide range of monitoring requirements from the Food and Drug Administration. We encourage researchers and policy-makers to install national or local drug utilization evaluation schemes to learn how to maximize the benefits and minimize the adverse effects of a new treatment strategy. For example, a multidisciplinary drug utilization program has been initiated to monitor the use of neseritide at our institution, and we have carefully examined the indications, outcomes, and adverse effects for the first 100 patients. Such monitoring or surveillance programs require time, effort, and a team approach, but can allow objective evaluation of safety and efficacy of a new treatment strategy.
Taking care of patients is the doctor’s greatest privilege. As physicians, we should take great care in understanding the benefits and risks of every therapeutic maneuver that we perform. As researchers, we should acknowledge that the publication of our findings is only the end of the beginning, and the ultimate goal should only be our patients’ well-being. All’s well only if it ends well.
↵* Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
- American College of Cardiology Foundation
- Bozkurt B.,
- Agoston I.,
- Knowlton A.A.
- Kohn L.T.,
- Corrigan J.M.,
- Donaldson M.S.