COMMENTARY

Apr 24, 2020 This Week in Cardiology Podcast

John M. Mandrola, MD

Disclosures

April 24, 2020

Please note that the text below is not a full transcript and has not been copyedited. For more insight and commentary on these stories, subscribe to the This Week in Cardiology podcast.

In This Week’s Podcast

For the week ending April 24, 2020 John Mandrola, MD comments on the following news and features stories.

Epidemiology

When I recorded last week there were 675,000 cases in the United States. Today there are 880K cases. The rate of rise three weeks ago was 3-fold; two-weeks ago it was 2-fold, last week it rose by 1.35 times and this week it rose by 1.30 times. A decreasing rate of rise despite increased testing is good.

When I recorded last week there were 30,000 deaths in the United States, today there are close to 50,000. The rate of rise 3 weeks ago was 5-fold, two weeks ago about 3-fold, last week, 1.8-fold, and this week, 1.6. The rate of rise of death is falling.

I spoke with a colleague from the New York City area this week. The story he told is utterly the opposite to what we feel here in the Midwest.

He said their large 1000-bed hospital had been taken over by COVID patients. Roughly 600 beds are filled, nearly all with patients with COVID-19 and 100 in the ICU. About one-third of the hospital was closed for potential surges; so, they had space for more COVID patients if necessary.

In Louisville, Kentucky, our situation is 180 degrees different. Our 500-bed hospital has about 30 patients with COVID, about 10 in the ICU, and these numbers haven’t budged much in the last two weeks.

We are not an isolated case. In a Tweet from UCSF, Dr. Bob Wachter reports even smaller numbers of COVID-19 patients in San Francisco – a decidedly large city. As of this morning, Wachter reports that at UCSF they have had 14 inpatients, 5 intubated and only 1 death.

The variance is perplexing and must influence interventions. Why would the response in New York be the same as in Kentucky or Kansas or Missouri? I am not talking about politicians’ choices to open restaurants and the like, but hospitals’ decisions to start delivering real healthcare. Outside New York, New Jersey, Detroit, and New Orleans, there are going to be cities and towns where lack of normal care exceeds morbidity and mortality from the virus. This week I find myself worrying more about too little care.

Notes on Testing

The PCR test, which measures the actual virus, has problems with sensitivity or true positive rate. Even in non-hot spot Louisville, Kentucky, I have seen an obvious COVID-19 patient test negative and then on repeat testing, test positive. The first test therefore was a false negative.

This is important. For example, in the obstetrics study I discussed last week, 13% tested positive. If the false negative rate is 20%, then the true prevalence of COVID-19 is actually higher.

But the biggest news on PCR testing this week was a preprint from Yale suggesting saliva had a higher rate of virus. Given the ease of collection for sputum this might be a huge advance.

This week also brought some sobering news on antibody testing. Medscape has a great video with Dr John Whyte and Dr Eric Topol on testing and tracking of the virus. Topol mentioned that there were about 90 different tests serology tests and none, I believe, have been adequately validated.

Here the problem is specificity, or the true negative rate. Specificity is the rate of true negatives divided by the sum of true negatives and false positives. False positives can occur because of overlap in detecting IgG of the other coronaviruses, which so many of us have. False positives become a huge problem when the prevalence in the population is low.

The Santa Clara study found a 3% population-weighted rate of seropositivity, based on a test with a specificity rate of about 99.5%. But if the test is 97% specific, statisticians and epidemiologists point out that 3% seropositivity is possible even when the actual prevalence is 0.

Specificity plays into yesterday’s news from Andrew Cuomo, the governor of New York who announced preliminary results of serology studies from the area. A random sample of about 3000 antibody tests returned a seropositive result of 21% in New York City. But without knowing the specificity of the test, it could actually not be that high. My friend Venk Murthy has estimated that the specificity of the test was probably north of 97%. But even if so, it’s not good news because, as Harvard virologist Trevor Bedford argued on Twitter, this leads to an infection fatality rate (IFR) of about 1%. That is not good, because it is many-fold greater than influenza.

A final note on asymptomatic status. Eric Topol Tweeted a nice summary slide from Scripps showing many of the studies of PCR testing. From these studies, it is clear that the asymptomatic rate of this disease is ≈ 40%.

Therapeutics

A paper in JAMA on 5700 inpatients in New York City delivered extremely sobering news on those who get mechanical ventilation. Outcomes were assessed for about half the cohort who were discharged or had died at the study end point. About 14% were in the ICU and the take-home finding was that for the 320 patients or 12.2% who had mechanical ventilation, mortality was 88%. People had hoped that the dismal rates of death on ventilation from 2 studies from China 81 and 97% respectively would be better here.

Cameron Kyle-Siddel was right, and Medscape was right to bring him on early. We need to find ways to support people without ventilators when possible. If you read one paper this week on pulmonary medicine, read Mathew Sibua’s paper in CHEST on Treading Lightly in a Pandemic. He and his co-authors have coined the term “minimally invasive, maximally attentive care.” They are not recommending anything other than good doctoring. Recognize the physiology of hypoxemia and support these patients as much as possible with oxygen, proning, non-invasive ventilation means.

A Therapeutic Conundrum

A preprint study on hydroxychloroquine (HCQ) has come out. It is an observational study done at the VA. They looked at about 370 patients; some got HCQ, some both HCQ and azithromycin (AZ), and others got neither. These were not randomized groups. Doctors chose. So the authors did the normal things; they adjusted for as many confounders as possible. The study found no evidence that use of hydroxychloroquine, either with or without azithromycin, reduced the risk of mechanical ventilation in patients hospitalized with COVID-19. An association of increased overall mortality was identified in patients treated with hydroxychloroquine alone.

Should one use the VA observational study to pump the brakes on HCQ? The VA study cannot make causal claims that HCQ and AZ are ineffective. To the authors’ credit they did not. And we shouldn’t.

One of the most egregious leaps of faith I saw this week was to use of tPA for ARDS-like oxygenation issues without documentation of clots. That this story got into the New York Times makes it even worse because no matter the qualifying statements, other docs may try it and patients might ask for it.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
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:

processing....