Evidence-based medicine is a term that attendees often use in their lectures, and medical students like to incorporate it prominently into their assessments and plans. There is so much clinical trial data that it is impossible to keep up to date with the latest guidelines and the latest landmark trials, and all the time is spent scrutinizing the results. We rarely delve into the rigorous analysis of the metrics used to quantify our results, as we do about the results themselves. Such scrutiny is necessary to answer the most basic and fundamental question (that we think about so many times): How much evidence is there in evidence-based medicine?
It is a well-established statement that angiotensin receptor-neprilysin inhibitors (ARNIs) benefit heart failure (HF) patients with reduced ejection fraction. This data comes from his PARADIGM-HF trial and is the main positive headline conclusion of that study. However, looking objectively at the evidence presented in that trial, ARNI was effective against moderate-dose enalapril, but also ramipril and valsartan alone. There is also fairly new data from the LIFE trial in 2021 that concluded that ARNI was no better than valsartan alone, but collectively ARNI could not be better than valsartan alone, even if the evidence might tell a different story. Keep it soaked.
Subgroup analysis of LIFE by Vader et al. We investigated how patients tolerated ARNI and found that the main factors for intolerance were primarily determined by hypotension, symptoms of hypotension (dizziness, fatigue), and renal disease. , they were able to predict ARNI intolerance by stratifying patients with hypotension, valvular disease, insulin use, electrolyte disturbances, or no ACEi/ARB use. With four or more of these predictors, he increased the predicted intolerance rate to over 50%.
We hypothesize that this dichotomy is due to the fact that HF trials generally recruit patients of appropriate socioeconomic status and healthy enough to attend the clinic. This selective recruitment differs significantly from real high-risk patients with multiple comorbidities who were included in LIFE. Together with a subgroup analysis by Vader et al., his highly regarded ARNI demonstrated no benefit in these patients, many of whom did not tolerate it in the first place. Anecdotally, ARNI is usually initiated on an admission basis for her HF exacerbation in a vulnerable patient. Still, evidence of its benefits comes primarily from outpatient clinic patients, mostly white males of high socioeconomic status. It is a reminder that many outpatients do not inherit the same benefits from ARNI as outpatients.
The concept that high-risk patients are less likely to benefit from stapling does not apply only to the setting of heart failure. There seems to be a bias toward generalizing medical therapy guidelines extrapolated from such trials, with an emphasis on trials that recruit optimal patient profiles, rather than representing realistic patients with comorbidities. but seem to ignore them. Take, for example, CORONA, the 2005 AURORA trial, which used statins in patients with HF, and the 2009 4D trial, which examined the use of statins in ESRD. All three of these trials investigated high-risk patients with multiple comorbidities, and all of them showed no benefit from statins. Evidence reveals that the assumption that high-risk patients will benefit from common treatments and interventions is often false.
This means that we call our actions “e“Evidence-Based Medicine”:
- Watchman did not meet non-inferiority to warfarin for the primary endpoints of stroke and systemic embolism in the PREVAIL trial.
- Provides a watchman when patients intolerant to anticoagulant therapy are excluded in trials of benefit.
- There is no convincing evidence that AF ablation reduces clinical outcomes in CABANA, and there are currently no placebo-controlled trials examining quality of life.
But somehow these interventions weigh heavily in the absence of sufficient evidence to support their eventual use.
There is a meta-analysis published by the Journal of Clinical Epidemiology in 2022, but I feel it has not received the attention it deserves. This analysis looked at the data behind over 1,500 Cochrane interventions and objectively analyzed whether there was high-quality evidence of clear benefit. The authors specifically used an intervention with a Cochrane review because it is widely practiced. They analyzed the data according to his GRADE criteria, which required the primary endpoint to be unbiased, statistically significant, and to provide effective treatment. With such high standards, one would think that a significant number of these interventions meet the stringent criteria to justify their widespread use. Shockingly, only 5% of these 1,500 interventions were firmly rooted in ‘evidence-based medicine’. Additionally, 8% of interventions showed statistically significant harm.
There is tremendous bias, especially when talking about evidence-based medicine. Perhaps it’s the financial side of healthcare or outright confirmation bias, but when there are trials that show limited benefit in interventions that favor clinicians to do, we tend to look the other way. So the next time you want to justify using something by labeling it as an “evidence-based” drug, stop and think about how much evidence goes into that designation. please look. The results may surprise you.
Benjamin Borokhovsky is a medical student.
