‘Absence of Evidence Is Not the Evidence of Absence’ - How Do We Translate Clinical Research Into Clinical Practice?

The title of this post is a quote, but who said it first is not entirely clear to me. It has been used in medical editorials as far back as the 1980s, television shows such as Dexter, and notably in a 2002 Donald Rumsfeld speech on defending the rationale for the ongoing search for infamous “weapons of mass destruction.”

However, I was reminded of this quote and a larger problem during a recent case in the cath lab. A 60-year-old male with a known focal proximal LAD lesion was referred for stable but persistent class III angina. He had chronic back pain and required frequent MRIs. He had preserved EF, but overall his systemic systolic pressures ran low at 90 mmHg. He could tolerate antiplatelet and statin therapies, but only 1 low-dose antianginal. Beta-blockers could not be started due to resting sinus bradycardia in the 50s. Additional antianginals were contraindicated due to relative systemic hypotension. Pacemaker implantation (so that beta-blockers could be added) was not considered due to his ongoing need for MRIs. And so, on this particular cath lab visit, we decided to proceed with PCI to his focal lesion.

As we were about to begin, a fellow piped up: “Wait a minute. How can we do PCI on this patient? He is a COURAGE patient!”

An important trial, COURAGE evaluated the role of PCI vs aggressive medical therapy in nearly 2,300 patients with stable angina. It demonstrated that PCI did not reduce the risk of death, MI, or MACE when added to optimal medical therapy. 

While the fellow brought up an important point, it made me realize the flip side of contemporary cardiology training, where the need for “evidence-based medicine” has sidelined clinical judgment. I recently attended a lecture by Dr. Anthony DeMaria, former JACC editor, who took this concept a step further by saying the loss of such judgment may lead to “loss of clinical freedom” for future cardiologists. 

Clinical evidence and clinical trials are integral parts of training and our future practice. But their limitations are as important as their strengths. I recently learned that despite such emphasis on deriving evidence from clinical trials and applying it to practice, only a very small subset of the patients we treat would actually be considered eligible for enrollment. For example, patients who have end-stage renal disease, are older than 70 or women, lack insurance (and hence regular follow-up), or belong to minority ethnic groups continue to be underrepresented in large-scale randomized clinical trials. These subgroups represent a significant proportion of the patients we take care of routinely.

COURAGE did enroll a large population; however, lost in a supplement to the paper is that nearly 36,000 patients were actually screened for the study. In other words, 93% of the patients were actually ineligible. What do these patients look like? No one knows as these data are seldom published. I suspect that most of the patients I care for on the cath lab table, like the man described above, are usually excluded from trials.

So how should we treat patients not represented in contemporary clinical trials? It begins with emphasizing that we truly are not crippled in treating them when specific correlates are absent. There are important examples in and out of medicine where routine practices are not supported by evidence. American armed forces continue to use parachutes for soldiers who jump out of planes despite the absence of a randomized clinical trial showing effectiveness. We use insulin for diabetic ketoacidosis, furosemide for congestive heart failure, and synchronized defibrillation for ventricular arrhythmias all in the absence of randomized clinical trial data. 

So on what foundation can we base our decisions? Clinical judgment. It’s okay to perform PCI on the patient who cannot tolerate any additional medical therapy despite the fact he does not fit into the clinical trial mold. That is my call. 

Amusingly, there is even evidence to support that my clinical judgment may be just as important, if not more so, as what can be learned from randomization in a clinical trial. The MASS II trial randomized patients with stable multivessel disease to medical therapy, PCI, or surgery. Prior to enrollment, the patients’ primary cardiologists were asked what the ideal treatment strategy (based on their clinical judgment) should be for each patient with each response sheltered from the trial investigators. At 1 year, adverse cardiovascular events were lower when patient randomization was concordant with the primary cardiologist’s clinical judgment.

The balance of clinical judgment and evidence-based medicine is tricky. Clinical trial data does not always apply to individual patients, and no single physician can have enough clinical observation to make clear, objective decisions. Fellows are crippled when encountering patients who do not fit the mold of a particular clinical trial, but they also lack the experience to make the decision without the stencil. Both approaches must be emphasized equivocally. The pendulum has swung far toward the science of medicine, but perhaps we should pull it back toward the art and have a balance of both. 

The data from randomized clinical trials represents the beginning of the decision making process, not the end. –Anthony DeMaria