One of my favorite books growing up was “Charlie and the Chocolate Factory”. I revelled in the comeuppance meted out to the likes of Augustus Gloop, Mike Teavee and Veruca Salt. It’s so rare in life to see folks get what they deserve. And like all children, I loved candy and was fascinated by how it was made. Candy was delicious. Candy was magical. And as described in the book, candy was equally magical in its manufacture. Knowing that it was made by mysterious “Oompa Loompas” made it that much sweeter.


Soon after I first read “Charlie…” I was lucky enough to go to the Hershey Park. This was at a time before it became a mere theme park, and you could still visit the actual chocolate production lines. “Oompa Loompas! I will finally see Oompa Loompas!”

The aroma from the factory was fantastic. There were a lot of cool machines and molten chocolate. But there were no Oompa Loompas to be seen. Sad as I was at first, over time I grew to appreciate that the chocolate tasted just as good as before. It was just as sweet and was just as satisfying when made by machines or the hands of Oompa Loompas. The magic of the chocolate was in the experience.

Years later, as a pediatric cardiologist, I was pulled into an odd situation: dermatologists began asking for my guidance in prescribing propranolol. Odd, I thought. Why would dermatologists need to control heart rate or blood pressure? Did their treatments induce arrhythmias?

But then I read about how it was recently described to cause involution of infantile hemangiomas. Discovered quite by accident, it had quickly supplanted the then standard therapies of watchful waiting, corticosteroids, interferon, or vincristine (!) as more efficacious and much safer. Almost magical. This observation was later followed up by a multiple case series, small placebo-controlled trials, and ultimately a large (460 infant!) randomized controlled trial. The results were dramatic. As described by the investigators: “Treatment with propranolol at a dose of 3 mg per kilogram per day for 6 months resulted in a significantly higher success rate (primary outcome) as compared with placebo (60% vs. 4%).”

How does it work? If you Google around, you will see a lot of hand waving about “vasoconstriction” or “VEGF” signalling, or whatnot. It was quite fascinating for me in particular, as I spent the early part of my career as basic scientist trying to understand cardiovascular biology. To this day the mechanisms are still being investigated, using tools as various as in vitro tumor cell lines, biochemical binding assays, and mice with transplanted tumors. No satisfactory answers have shown themselves. Although, to be fair, we don’t have a great idea of how it is that infantile hemangiomas appear in the first place.

In a similar manner, we are witnessing this with the recent clinical trials of SGLT2 inhibitors. Developed with a mechanism in mind—forcibly excreting glucose through the urine—these drugs were first used in diabetes. Somewhat surprisingly, in large cardiovascular safety studies, they showed a cardiovascular benefit over and above what would have been expected given their degree of glucose control. Given their dramatic, and somewhat unexpected results, their use was expanded to heart failure outside of diabetes. And the good results have continued.

A great deal has been written speculating on how it is that SGLT2 inhibitors work in heart failure. Is it from mere diuresis? It there a metabolic component to heart failure? Are there special receptors in the heart that respond to this medication?

We don’t really know. But then, what do we really know about heart failure? There are many folks clamoring to know the mechanism of how drugs work before accepting the clear results of randomized trials. We were taught to accept and appreciate mechanisms in school. But how well do we really understand the mechanisms of the diseases we treat? It was not that long ago that beta-blockers—now a mainstay of heart failure therapy—were verboten.

It may sound harsh, but most of what we think we know about disease are “Just So” stories. Look at the recent experience with ORBITA, or any number of other “negative” trials that put our pathophysiological understanding to the test. In the end, I believe that the wise physician will accept that it’s better to just enjoy the chocolate and not work so much about if Oompa Loompas were involved.

+ posts

1 Comment

  1. Very nice piece. In many cases, drugs have been developed based without prior knowledge of mechanism. That said, understanding of mechanism enables development of follow-on drugs with greater effectiveness, reduced off target effects, more convenient dosing, etc. I would not dismiss the importance of these. After all how many patients are on lovastatin today?

Leave a Reply