More Questions about COVID-19 Answered by the Experts

As a retired physician, I feel that I should be doing something helpful during this pandemic—like being out on the front lines saving lives or, at least, taking SARS-CoV-2 swabs in a tent somewhere. Since I was trained in biochemistry before training as a doctor, perhaps I could even help run assays on those swabs or work in a lab developing new tests. But I practiced pediatrics, which seems to be the specialty least affected by the virus. And these days, being at retirement age, I am among the people at elevated risk.

However, I still find myself still going back to the medical journals to learn everything I can about SARS-CoV-2 (the virus) and COVID-19 (the disease). So I’ll offer this second installment of common questions about the coronavirus answered, as best as possible, directly from the medical literature. (Read part one here.)

Electron microscope image of MERS coronavirus attached to respiratory lining, captured and color-enhanced at the NIAID Integrated Research Facility in Fort Detrick, Maryland. NIAID-RML

Does COVID-19 Cause People to Lose Their Sense of Smell?

Sources:

“Neurological Manifestations of Hospitalized Patients with COVID-19 in Wuhan, China: A Retrospective Case Series Study” (Ling Mao, et al., Tongji Medical College, Wuhan) (link)

“Non-Neural Expression of SARS-CoV-2 Entry Genes in the Olfactory Epithelium Suggests Mechanisms Underlying Anosmia in COVID-19 Patients” (Brann, D. et al., Harvard Medical School) (link)

When I started hearing anecdotal reports about people losing their ability to smell during COVID-19 infection, I just had to find out whether it was a real symptom and, if so, how reliable it might be as an indicator. This is the sort of concern that keeps people awake at night but causes physicians to roll their eyes in frustration when no one is looking. Next to high fever and difficulty breathing, anosmia, as the loss of the sense of smell is called in the medical literature, is a trivial thing. But could it be significant anyway?

I traced some online news reports of loss of smell with COVID-19 back to two articles that seem legitimate. Neither has yet been published in a major medical journal, but both were produced by knowledgeable researchers at respected medical schools. The two together are pretty good evidence that, yes, anosmia can be a symptom of SARS-CoV-2 infection.

The first paper, from a group of neurologists in China, is based on a study of 214 patients from three hospitals in Wuhan during the early stages of the pandemic. The patients’ charts were screened for all neurological symptoms, which included everything from mild complaints to severe events. The intent of the paper is clearly to try to anticipate which patients are at highest risk for severe neurologic consequences, like stroke or loss of consciousness. Because the study included many elderly, acutely ill, hospitalized patients, it is not terribly surprising that five patients had strokes during their illness. But the researchers also screened for less dramatic symptoms, which they mention briefly: dizziness in 16 percent of patients, headache in 13 percent, hypogeusia (decrease in sense of taste) in 5 percent, and hyposmia (decrease in sense of smell) in 5 percent.

There is no way to know how diligently these 214 patients were tested for anosmia or hyposmia. Since the data was obtained by merely reviewing medical records, some patients must have reported loss of sense of smell, and their physicians recorded it in their medical history. But I doubt anyone had time to actually go around asking people specifically about their ability to smell. It is likely that more patients had anosmia than are recorded in the study, but they didn’t notice it or consider it important, what with all the coughing and trouble breathing they must have been experiencing. The researchers do not speculate about this at all, being much more interested in the possibility of severe neurologic complications. So I would guess the actual incidence of loss of smell is probably higher than 5 percent.

The second paper, from Harvard Medical School, is by a team of neurobiologists who operated under the assumption that loss of the sense of smell is, in fact, a real symptom of COVID-19 and attempted to explain biologically why that might occur. The paper is quite long and intricate, and I must admit that even as a physician, I did not understand absolutely every detail.

Basically, however, the virus is known to bind to cells using a surface protein called ACE2 with the aid of an enzyme (protease) called TMPRSS2. The researchers studied the DNA of cells in the lining of the nose from both people and mice, looking for the ability to produce (or “express”) these two substances. They found that the cells in the nose responsible for detecting scents (olfactory cells) do not produce ACE2 or TMPRSS2 except, possibly, in very young mice. Therefore, it seems that SARS-CoV-2 cannot attack these cells directly. But there are other cell types surrounding the olfactory cells that do express ACE2 and TMPRSS2. The researchers speculate, at length, on ways the virus might interfere with the sense of smell by damaging these surrounding cells. More research is required, they say. I’m not holding my breath.

It seems likely that, for some, anosmia is indeed a symptom of SARS-CoV-2 infection. Unfortunately, however, the association of this symptom with COVID-19 is not strong enough for it to be a predictive diagnostic tool. Even more people with COVID-19 have headaches and dizziness, but because these are so common, they clearly cannot be used to diagnose it. Anosmia is rarer, but it can also be caused by other things, like allergies and regular colds.

Once there are enough test kits available, I might recommend that someone who notices a decreased sense of smell get tested. However, we would not be able to predict, from the presence of anosmia alone, whether or not the person has COVID-19, nor can anosmia reliably tell us anything about the eventual severity of the disease. The severity of COVID-19 is determined only by progressive difficulty breathing, which, at this point, remains the most important symptom to monitor.

How Many People Will Get the Virus but Never Develop Symptoms?

Source: “Asymptomatic and Presymptomatic SARS-CoV-2 Infections in Residents of a Long-Term Care Skilled Nursing Facility—King County, Washington, March 2020” (Kimball, A. et al., Morbidity and Mortality Weekly Report, April 3, 2020)

Because testing for SARS-CoV-2 has been done, by necessity, only on the sickest patients, I have suspected (and hoped) that the full spectrum of COVID-19 is not being represented by the nightly reports of “new cases” and deaths. In fact, now that some sites have enough test kits to test people who are not quite so sick, the fatality rate among diagnosed cases has been decreasing steadily. This is not because we are doing better at treating the disease but because we are detecting larger numbers of less severe cases, meaning the cases resulting in death represent a smaller percentage of the total.

In even the earliest reports from China, Germany, and the Diamond Princess cruise ship, there were people who tested positive for SARS-CoV-19 but did not have symptoms. Some of these positives were presymptomatic at the time they were tested, meaning that they weren’t yet ill but would become so later. There are also reliable reports of people who continue to test positive for days after their symptoms have subsided. From a public health standpoint, this is very bad news, since it makes it very difficult to determine who needs to be quarantined and for how long.

Scattered through these reports are hints of people who never became symptomatic at all, but their numbers have not been confirmed. It is fairly easy to find data on the presence or absence of symptoms at the time of testing for SARS-CoV-2, but not on the occurrence of symptoms in the weeks before or after the test. I was disappointed to find that, even with a captive patient population aboard the Diamond Princess, no one seems to have tracked the symptoms of those people who were positive. This paper finally reports symptoms over the weeks before and after testing and provides follow-up on a few people who’d had no symptoms at the time of testing.

In the study, published in the CDC’s Morbidity and Mortality Weekly Report, researchers from the Seattle Department of Health tested almost everyone in a skilled nursing facility, regardless of the presence of symptoms. The residents at the facility were exposed by a health-care worker who worked several days while feeling ill before testing positive for SARS-CoV-2. After two residents were hospitalized, a total of 76 residents were, I imagine, happy to be tested. (However, a few refused and were not included in the study).

In addition to the presence or absence of the virus, the researchers were also able to measure the amount of virus present. There were 23 positive tests (30 percent of the tested residents), and all of the samples showed high enough virus levels to be considered infectious. To assess condition, the researchers asked about a long list of possible symptoms, including the usual fever, worsening cough, and muscle aches as well as atypical symptoms such as confusion and headache. (Alas, they did not ask about anosmia.) They found 13 people who’d had no changes at all in their health in the preceding two weeks. The amount of virus was no different between residents with and without symptoms at the time of testing, indicating that people with no symptoms are just as likely to spread the disease as people without them. But the researchers did not stop there—they continued to monitor symptoms. Three people who tested positive never developed any symptoms at all.

The intent of the paper is to warn other skilled nursing facilities that mere monitoring of symptoms is not enough to control an outbreak. If there is a positive test, they say, everyone in a facility must be quarantined until the results of testing are available. Health authorities are all pretty much agreed on this by now. But I find it interesting that, even in this high-risk population, 3 out of 23 positives never had any symptoms.

This is a very small study, so it cannot be scaled up to predict the number of asymptomatic cases that might be out there. But it certainly seems likely that people who have had a documented close exposure, like living with a known positive case, may well have had a SARS-CoV-2 infection even if they never had any symptoms. It remains to be seen whether these people are now immune to the virus. I sincerely hope so.

How Promising Is Remdesivir for Treating COVID-19?

Source: “Compassionate Use of Remdesivir for Patients with Severe COVID-19,” (Brainard, D.M. and about sixty others, New England Journal of Medicine, April 10, 2020)

With COVID-19 firmly established as a potentially fatal, widespread disease, possible miracle cures are being reported at an alarming rate both on the internet and on the evening news. Many of these are not based on any facts, but there are drugs out there, both new and established, that might be effective based on their known mechanisms of action. Unfortunately, their effectiveness is theoretical, since they have been tested almost exclusively in animal cell cultures. So yes, there are promising medications, but they are, for the most part, early in development. None of them are anywhere near ready to be called cures.

A vial of remdesivir from a Gilead Sciences advertisement

Remdesivir is a new antiviral medication that has been receiving a lot of hopeful press recently. (I’d never heard of it myself until it was mentioned in the March 16–23 issue of Time magazine.) It has been shown to inhibit SARS-CoV-2 in cell cultures by blocking an enzyme the virus needs to invade cells, so it potentially could work against COVID-19.

According to this paper from early April, remdesivir has been used only for 500 Ebola patients, in whom it had a “favorable clinical safety profile.” This means, I think, that it didn’t obviously make anyone sicker than Ebola did. The drug manufacturers therefore proceeded with this study, a trial of in human volunteers with COVID-19. There are innumerable enthusiastic reports about it on Google. But don’t get your hopes up too much.

This paper demonstrates everything that is troubling about the publication of medical research. It was published by the New England Journal of Medicine, one of the most prestigious journals anywhere, but there are many red flags that add up to a high level of possible bias. For one thing, the research was funded and conducted by Gilead Sciences, the manufacturer of the drug. Inquiries about the study are to be directed to a Dr. Brainard, who is an employee of Gilead. And the article was written by ghostwriters, also employed by Gilead. They got what they paid for: a beautifully written article with lots of lovely charts and graphs, all serving to disguise an almost complete lack of meaningful content.

Basically, the company “compassionately” sent a free 10-day supply of remdesivir to anyone who inquired about it. There are enough names listed as authors that it is quite possible everyone who used the drug as directed now has one more publication to add to their academic résumé. In return, the company received reports about the patients’ progress. This included daily liver and kidney function tests, which are commonly monitored when trying a new drug. Gilead sent its drug to 61 patients in nine countries but used data from only 53 because of “missing post-baseline information.” Those 53 patients were all given “at least one” dose of the drug. These are all red flags.

The company collected a lot of data on patient progress, most of which is carefully arranged and easy to interpret. Interestingly, the data for liver and kidney toxicity is available only in a supplemental report, though the study does mention that about 20 percent had signs of liver failure and 10 percent had signs of kidney failure. To be fair, the people given this drug were the sickest possible COVID-19 patients, with 30 receiving invasive mechanical ventilation and 4 on complete heart/lung bypass (ECMO) at the time they were given the drug. These carry their own risks of liver and kidney damage.

The data on respiratory function seems impressive. It shows that the patients who improved did so about four days after beginning the drug. This improvement was faster in patients who were younger or needed less ventilatory support; this isn’t actually much of a surprise, but their overall mortality rate does look pretty good. Seven patients died, but, given how sick they were, this is not surprising either—the authors point out that their mortality rate of 13 percent is better than reported for severe cases elsewhere, including a study done on a rival drug, which reported a 22 percent mortality rate even though their patients were supposedly not as sick. This could mean that remdesivir actually had a positive effect.

The biggest problem is that there were no control patients in this study. Without people from the same institutions receiving similar care for similar levels of illness, but not receiving remdesivir, we have no idea whether the drug did anything at all. If we had reliable overall statistics for survival of COVID-19, which we do not, we might be able to make a valid conclusion, but without any comparison group, there is no way to tell how well the drug actually performs. The authors do admit to this in the last paragraph, concluding by informing us that randomized, placebo-controlled studies are already underway. A quick internet search reveals many excited reports about this as well, from which I conclude that Gilead Sciences also has a very good publicity department.

I hope that there is a drug somewhere that will make a difference, either alone or in combination with other drugs, against COVID-19. Remdesivir may well be one of them. I am comforted by the fact that so many minds are working on the problem, even if some of them publish before they have anything to report. I suspect that we will not find anything that makes much of a difference in patients whose lungs are already overwhelmed by the virus, but even small improvements would be welcome.

My biggest hope for the antivirals is that one of these drugs will prove effective as a preventive that can be given soon after exposure, much like Tamiflu for influenza. I’m sure the people of Gilead Sciences feel the same, and not just because they would make a ton of money. Unfortunately, a preventive drug is even harder to study than a drug for severe illness, requiring a much safer drug in a much bigger trial. We would first need to establish the number of asymptomatic or mildly symptomatic people, as well as analyze a large amount of outcome for all the people who test positive for SARS-CoV-2. This won’t be possible until there are enough test kits to test everyone who has been exposed.

In the meantime, I find myself encouraged by the skunk that is hanging around my garage, because it provides an easy way to make sure my nose is still working.