The duck test for COVID

With good reason, the testing process for COVID has attracted a lot of attention. There's been news about the limited numbers of tests that have been distributed. Doctors have been frustrated getting tests for our patients. And if you aren't a celebrity or a tiger in the Bronx zoo, at least in April 2020 in most of the United States, you probably needed to be pretty sick to actually get tested.  

I wanted to spend this email talking about the way testing happens for people in the hospital as well as the strengths and weaknesses of our current state of testing. 

Each COVID patient begins their hospital stay as a PUI, a "person under investigation." Every hospital has a slightly different protocol, but most of them are variations of a general theme. When a patient comes to the hospital, they're screened for any potential COVID contacts and suspicious travel as well as symptoms, the most common of which are cough and fever. They also get asked about less common symptoms, such as diarrhea, vomiting, weakness/malaise, muscle aches, runny nose, sore throat. If there's any suspicion based on your symptoms and exposures that you're infected, you go into isolation and you get tested. This is all before you've seen a doctor.

Testing as of right now means that you get a nasal swab that is tested for viral RNA. RNA is the genetic material of the virus, which is structurally almost identical to DNA, but has one less oxygen atom. RNA stands for ribonucleic acid and DNA stands for deoxyribonucleic acid, de-oxy meaning devoid of oxygen. 

The RNA test that we have right now only detects the virus - a sign of active infection - and doesn't tell us anything about where someone is in their disease course, how sick they're going to get, or whether they were previously infected and are now immune. 

The swab is essentially a long, stiff q-tip that gets stuck deep into the back of your nose and wiggled around there. It's pretty unpleasant. And since it's based on physically collecting the viral genetic material, the accuracy is somewhat technique dependent. You really need to swab deeply to get to where you need to be.  

And after swabbing, there's the wait for results. Until pretty recently in most of America, the wait to get results was 48-72 hours. Things have gotten quite a bit better over the past week or two. Now many locations, but not all, have a rapid test, where you get results in about 45 minutes. This is, obviously, a huge improvement over 2-3 days.

Unfortunately, the current test really sucks.

The sensitivity of this test for viral RNA is about 75-80%, meaning that if you are infected, theres a 75-80% chance that the test will come back positive. This means that between 1 in 4 and 1 in 5 infected people will have a negative test. These numbers are kind of insane, to the point that if this were any other clinical situation, we would almost never use a test with this degree of inaccuracy.

Since the test really stinks, when there's a high clinical suspicion for COVID we end up testing patients multiple times. I have had plenty of patients with a negative first test who came back positive on a repeat, and even a patient who didn't test positive until the 4th swab, but we kept testing because the diagnosis made sense. 

In other words, if it walks like a duck and quacks like a duck, it's probably COVID.

Why do we have the suspicion to keep sending this test on patients multiple times? Well, because we're seeing a ton of COVID. And because most patients come in with similar symptoms and objective lab and imaging findings, a topic for another email. But also because almost everyone else is staying home. Heart attacks, appendicitis, strokes, we're seeing less of all of it. It's probably because people are staying home until late in their disease course, with deleterious consequences. 

So what would make the test better and where do we need to go? For one thing, a higher sensitivity would be great so that we don't have false negatives on testing that require repeat. But the real game changer here, at least in my eyes, is a test for immunity. 

You may have heard or seen this described as a serology or serologic test, which basically means that it looks at the blood serum (the clear-ish liquid part of blood that isn't red or white blood cells) for antibodies to the virus. A mature serologic test can evaluate the difference between antibodies that are made in the acute, or active, phase of infection, which are called IgM, and antibodies that indicate a past infection, suggesting immunity, which are called IgG. 

In a perfect world, we'd have a rapid, accurate, widely available, and cheap serologic test that would let us know who is immune and make the process of reintegrating into normal life much more straightforward. Until that time, there's just way too much guesswork involved for those who haven't been clearly infected to go back to their pre-social distancing lives, especially if they have conditions that put them at increased risk.