From Medical Skeptic to Researcher: The View From the Other Side

John M. Mandrola, MD


May 13, 2019

Critiquing evidence for the past few years has grown my interest in how evidence is made.

My academic colleagues have often said that it's easier to critique science than do science.

I understood this, in theory. But having been only a consumer of medical evidence for most of my career, I had not felt the difficulties.  Though I've had only a small dose of research thus far, I can testify: it's damn hard.

At the 2019 Heart Rhythm Society Scientific Sessions, I was proud to be part of the His-SYNC trial,[1] a randomized controlled pilot trial comparing His-corrective pacing vs biventricular pacing in patients with heart failure.

Since I am a His-bundle pacing enthusiast, I felt a wave of emotion upon seeing my name and hospital logo on the slides in the large auditorium. Having a simultaneous publication in a big journal only added to the excitement.

Here's the problem: When I was telling electrophysiology legend Eric Prystowsky about "our" study at the convention center, he said, "Mandrola, if this wasn't your study, you'd tear it to shreds."

Senior author Rod Tung, MD, from the University of Chicago, presented His-SYNC in the first late-breaking trials session. This was a pilot randomized controlled trial (RCT) conceived and pre-registered in 2016. When Rod asked me to participate, I was thrilled to do so—because the skeptics of His-bundle pacing correctly point to the need for evidence before we adopt an unproven technique.

The skeptics know that beauty is intoxicating. Indeed, (successful) His-bundle pacing produces stunning electrocardiography (ECG)  images of QRS narrowing. I often run around the hospital showing ECGs to my colleagues, saying, look at how cool this is! Twitter, too, overflows with success stories.

But as Suneet Mittal and colleagues have shown, His-bundle pacing has downsides: sensing issues, higher thresholds, and a higher rate of dislodgements.[2] As a medical conservative,[3] I believe in adopting new techniques only if the evidence for benefit is strong.

Thus far, we cannot say the evidence for His-bundle pacing is strong. Most of the data comes from nonrandomized observational studies.

The lack of trial evidence is why the His-SYNC trial was so important. Remember, for patients with heart failure and left bundle-branch block, His-bundle pacing competes against a proven winner, cardiac resynchronization therapy (CRT).

His-SYNC was the first prospective head-to-head RCT to assess the feasibility and efficacy of His-CRT as a first-line strategy compared with biventricular (BiV) CRT. This was an investigator-initiated and self-funded trial.

Industry, of course, has zero interest in funding His-bundle pacing trials. If this technique prevents pacing-induced cardiomyopathy and corrects left bundle-branch block, it will decrease the market for implantable cardioverter defibrillators and CRT devices. Industry does not fund research that could decrease profits.

Now to the super-hard part—the results and conclusion of His-SYNC: "His-corrective pacing compared to biventricular pacing did not demonstrate significant improvements in electrocardiographic or echocardiographic parameters as compared to BiV- CRT."

My first taste of serious research, in a technique that I love, and the results did not reach statistical significance. Ah, the irony.

Our trial had the usual problems seen in device trials: small numbers of patients and high rates of crossover. It would have been unethical to leave a patient without resynchronization.

I spent many hours and numerous phone calls discussing the language to describe these results. On my mind was a paper I recently coauthored on the high prevalence of spin in the cardiovascular literature.[4] Spinning the results was not an option.

I even had the thought: Did we do all this work and learn nothing?

No. I believe the results of this trial advance the field. For instance, we learned that many of the crossovers in the His-bundle pacing arm had atypical left bundle-branch block. A very recent study showed that these patients cannot be corrected with His-bundle pacing.[5] The next His-bundle pacing trial, therefore, needs to exclude these patients.

These data also inform the expected rate of crossovers, which guides the power calculations of the next trial.  

Two Final Take-Homes

Firsthand experience makes it easier to understand the emotions of science and its reward structure. Consider that I work in private practice. Research is extra; it has no bearing on my main job of being a doctor.

Yet, I felt the rollercoaster of joy in getting published and seeing my work presented on the big stage, but then the sadness of results that are not statistically significant.

These emotions underscore the human-nature aspects of science. For instance, I've railed against financial conflicts of interest in the past. No doubt, these are relevant.

But perhaps the less tangible biases of science's reward structure are as strong. Gosh, I wanted this trial to be "positive."

One thing that hasn't changed in my mind: Critical appraisal needs to continue in earnest. Critical appraisal is not personal. Just because science is hard and emotional, that does not preclude it from skeptical interpretation.

If evidence is to guide medical practice, the consumers of evidence must approach it with dispassionate rigor.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: