Hydroxychloroquine for Coronavirus: Truth, Again, Between Opposing Corners?

My native aptitude, if I may presume to say so, is seeing the whole elephant in the room. I have great respect for the targeted expertise many of my colleagues apply to the trees of their special interest and scrutiny. I am an inveterate forest guy.
Image of a melting antique brass alarm clock with old-fashioned bells on top, face glass broken, little hand on the 12, big hand broken and lying on the glass on the table.

That explains my positioning with regard to the coronavirus pandemic in general. I happily defer to others for rarefied erudition in pandemic response, virology, or mathematical risk modeling. My interest, rooted in the fundamentals of preventive medicine, public health, and general epidemiology- is in recognizing the full range of potential dangers, and the minimization of total harm. I stand by the contention that there is more than one way for this contagion to harm people, both directly through infection, and indirectly through the consequences of social upheaval, and both are bad. The national path through the crisis should be paved with the explicit intention of minimizing both.

This same thinking prompts me now to comment on the use of hydroxychloroquine for coronavirus, a promising but far from adequately tested treatment the president has apparently taken to touting as something of a panacea. I commend to the attention of those wanting more particulars on this topic the excellent work of my friend and colleague, Dr. James Hamblin, who writes for The Atlantic, and can draw on training as a physician with a masters degree in public health.

In his reflections on hydroxychloroquine, Dr. Hamblin is absolutely correct that our president has no business offering an opinion on the matter. The president, who generally conveys a disdain for science rather than any working knowledge of it, is an utterly ill-suited source. He should stop dispensing medical advice without a license, to say nothing of basic understanding, and defer to those qualified to do so.

But that doesn’t necessarily mean he’s wrong, if only by happenstance. A broken clock is wrong 99.86% of the time, but is correct twice a day.

There may be promise in hydroxychloroquine for severe coronavirus infection, according both to some early published studies and case reports, and per anecdotes conveyed to me by friends who have lived through an ICU experience with a loved one. There is, as well, potential danger in using the drug- including some risk of death by cardiac dysrhythmia.

The acuity of need, the paucity of data, and the trade-off between risk and benefit accentuate three considerations. First, absence of evidence is not evidence of absence. In other words, lacking decisive evidence in this early going that hydroxychloroquine is effective doesn’t mean it is ineffective- it just means we don’t yet reliably know.

Second, clinical decisions, especially when circumstances are dire, often need to be made despite uncertainty. The next, best option after textbook offerings have been exhausted will require judgment informed by something other than clinical trials yet to be conducted (see Table 2 in particular).

And finally, everything in clinical practice comes down to best assessments of benefit and risk in ratio to one another.

Many years ago, during my training in Internal Medicine, I was the senior resident responsible for the intensive care unit (ICU). My beeper chimed at 2AM or so, and it was my fellow resident, responsible for the general medical floors, calling to say a patient in pulmonary edema needed transfer to intensive care.

I arrived to find the patient, who had advanced kidney, lung, and heart disease- on the brink of death. She was desperate for air, frothing at the mouth, and her blood pressure was low and falling.

I administered an acute dose of intravenous morphine- which, among other things, very rapidly relaxes the smooth muscle in arteries. The aim in this is to reduce left ventricular afterload- the pressure the left ventricle must overcome to pump blood out and empty. If successful, the lightened work load relieves a failing left ventricle, the heart can empty, and the fluid backed up into the lungs can be mobilized. An effect is discernible in some number of seconds- which was all we had in this instance.

My fellow resident was not convinced by my call, noting that the patient was already hypoxic (deficient in blood oxygen), and that morphine could depress respiration; that the patient was already hypotensive (i.e., had abnormally low blood pressure), and morphine could drop blood pressure further, leading to shock, and death.

I knew all this as well as she did, and agreed. But there was this: absent immediate, effective treatment, the patient was going to die for sure. Whatever the risks of doing something, they were less under the circumstances, than the risks of not doing something. In general, we might say that when we have eliminated guaranteed death, whatever is left, however risky, is better.

As it happened, by the time we got the patient on her gurney to the ICU, her heart had been so unburdened, and her lungs so effectively relieved, that her blood pressure went up, not down. She was sitting up, chatting amiably about how much better she felt. In fact, I wound up in the dog house with the ICU nurses, who rolled their eyes at me for wasting an ICU bed on a patient who did not apparently need critical care.

Let’s be clear: this anecdote is a tale of neither brilliance, nor heroism. It was pretty much the standard operating procedure of acute care medicine. You may have no good options, and you almost never know how things are going to go. You generally have less than perfect information. You are left to take such information as you can get, estimate the ratio of potential benefit to possible risk among the options in front of you, and choose the best you can.

I tell this tale because some of you have been through a coronavirus ICU experience with a loved one; others are going through one now; and much as I regret to say it, others among us yet will.  This is all happening too fast to hope for the comprehensive clinical trial results we would favor to inform clinical decisions. You should expect, and get, the best ratio of benefit/risk the medical team can provide, riddled with uncertainty though it may be.

We don’t yet know the best way to treat SARS-CoV-2, and there is surely no panacea in the mix of promising options. There are promising options, however, hydroxychloroquine among them, and you should have recourse to those as clinical urgency dictates. Get that guidance from someone genuinely qualified to offer it, however, and avoid telling time with a broken clock.


This article was first published on LinkedIn.