- Why did I suspect COVID-19 in mid-January?
- When I started to doubt my COVID-19 status
- My COVID-19 Antibody Test Results
- Are COVID-19 antibody tests accurate?
- Why did I have SARS-CoV-2 IgM antibodies at all?
- Where do I go from here?
As someone living in New York City when the coronavirus pandemic was heating up, and especially as someone who had been in close contact with neighbors and friends recently returned from China in late December and early January, there was a time I believed that the illness I experienced in mid-January, a strange flu like stomach bug, was indeed the dreaded strain of coronavirus known as COVID-19 .
After taking a COVID-19 antibody test, that now seems a lot less likely.
Let me tell you why.
Why did I suspect COVID-19 in mid-January?
Partly because I wanted to believe. Even though the protective effect of antibodies on COVID-19 reinfection is not conclusively established, a prior infection adds the peace of mind that you’ve had the virus and fought it off.
So, to be fair, part of my COVID-19 suspicion was wishful thinking. But, part of it was certainly rooted in reality as well.
I got very sick, and in a strange way.
I experienced gastrointestinal symptoms (both vomiting and diarrhea) and that evolved into a low fever and some shortness of breath. My oxygen saturation, as measured by the amazing local nurse who took care of me, was on the low side at 90%, and using an oxygen machine seemed to help.
At first, I chalked everything up to altitude sickness.
After all, Jackson does sit at about 6,200 feet above sea level, quite a few notches higher than my usual low altitude perch in New York City. However, the altitude angle faded a bit for me as a an explanation after altitude and anti-nausea meds didn’t make a lasting impact. My nurse, Ali Kalenak, gave me a small dose of dexamethasone, toradol, and zofran, but the impact, usually very effective for altitude, was only a small help.
I was sick for about two weeks. After week one, my condition improved, but skiing or major physical exertion put me back to square one.
One of the emerging symptoms of COVID-19 is gastrointestinal. As many as one third of patients who contract the novel coronavirus experience diarrhea, nausea and loss of appetite. For some, these digestive troubles are their only symptoms. When I saw studies like this one from the journal Gastroenterology documenting the gastrointestinal side of COVID-19, I had my lightbulb moment.
I must have already had this thing!
When I started to doubt my COVID-19 status
For awhile there, I was confident I’d already had COVID-19.
My circle in New York included friends from China, NYC was the epicenter of the illness, and my symptoms matched up.
However, we now know that the timing didn’t quite fit.
Although there were lives lost to COVID-19 as early as February 6 in Northern California, which is a sign of some spread weeks before that date, the likelihood is that the virus wasn’t rapidly spreading in New York City until March (although it’s possible the first cases were present in the city as early as late January). While I was in New York for a few weeks in February as well, I left the city to spend time with family before the virus gained maximum velocity.
My reverse “aha moment” came when I saw data from the Seattle Flu Study, a rolling “pandemic surveillance program” that tracks the spread and severity of flu in Washington state.
Researchers at the University of Washington retrospectively tested previously collected samples from patients with confirmed respiratory illness for COVID-19.
They didn’t get “COVID-19 hits” until February 24, 2020. This means community spread in Washington state, and adjacent states like California, probably wasn’t widespread in mid-January, when I was sick.
For a detailed rundown on the COVID-19 retrospective analysis conducted by the Seattle Flu Study and what it could mean for the timing of community spread, particularly on the west coast, see this excellent Twitter thread.
We tested 3600 samples collected in Jan 2020 for COVID-19 status and found zero positives. We tested 3308 samples collected in Feb 2020 and found a first positive on Feb 21 with a total of 10 samples testing positive in Feb. 5/18 pic.twitter.com/Jcx0qzdfx8
— Trevor Bedford (@trvrb) April 12, 2020
I also had this Twitter conversation with investor and philanthropist Chris Sacca who reported seeing mostly negative antibody test results from work he funded in New York state.
We’ve particularly seen a lot of tests from NYC, Washington State, SF, and LA. I’m telling you the numbers aren’t that big. Obviously higher than Nebraska and Wyoming. But it’s just not that widespread.
— Chris Sacca (@sacca) April 19, 2020
Bottom line is that, while it was possible my illness in January was COVID-19, the Seattle Flu Study pointed more in the direction of traditional flu, or even food poisoning.
My COVID-19 Antibody Test Results
Even though my optimism was fading, I went ahead and had my blood drawn by a local doctor here in Jackson, Wyoming, my adopted home city while I ride out this pandemic. My physician used test kits from Ray Biotech in Atlanta. The Ray Biotech kits utilize ELISA technology (which we will see in a minute is an important detail).
It is worth noting here that the COVID-19 antibody tests are looking for signs of a past infection. These are not the tests we give to individuals who are currently very ill. These people are given the nasal swap PCR tests to see if they need to quarantine for a period of time to prevent inadvertent spread of the illness.
By contrast, antibody tests look for an immune system reaction to a past infection. Many of you will have had this type of test for Epstein Barr virus. Epstein Barr is one of the world’s most common viruses and the types of antibodies we use to identify a past infection are identical to what we now seek to find with COVID-19 infections.
IgM antibodies signal a recent infection.
IgG antibodies signal a more distant infection.
I was tested for both IgM and IgG antibodies using a traditional blood draw.
There are two basic methods for testing for COVID-19 antibodies: the finger prick method (which doesn’t show you amounts of antibodies, only positive / negative) and a full blood draw, which shows you the amount of each antibody in the serum.
Positive results show values greater than 1.0 AU/mL.
My results were:
- 0.046 IgG AU/mL
- 0.315 IgM AU/mL
My test results came back negative, but as you can see, I did have some IgM antibody activity to COVID-19.
And was the test I took accurate?
Are COVID-19 antibody tests accurate?
This is a matter of some controversy. Early serology studies from Miami, Los Angeles and New York suggest that the COVID-19 pandemic has been in our midst long before cases were officially reported.
In fact, it has been estimated that as much as 25% of the population in New York City has already been infected by COVID-19. The antibody data helps us arrive at an accurate fatality rate since, at present, the troves and troves of asymptomatic cases are not yet factored in.
However, there have been reports of both false positive and false negatives in antibody testing.
How prevalent are these inaccuracies?
Specificity vs. Sensitivity
In COVID-19 antibody testing, a low rate of false negative is called high sensitivity, whereas a low rate of false positive is described as high specificity. 1
It’s tough to say whether antibody tests are accurate as a group because there are so many tests out there. Although some tests show close to 100% specificity (low rate of false positives), a good estimate at present is to bake in a 10% chance that antibody results are wrong in either direction.
My specific test was not FDA approved, and yet it relied on ELISA technology that is reported by John’s Hopkins to have 87.5% sensitivity from other labs. This means that ELISA antibody tests are more likely to register a false negative than they are a false positive.
The ELISA test administered by Ortho Clinical Diagnostics is said to have 100% specificity, so the results are much more reliable if you get a positive hit.
Based on this data, there is a greater than one in ten chance that my antibody results were inaccurate.
Why did I have SARS-CoV-2 IgM antibodies at all?
As you can see from the table above detailing my test results, assuming I didn’t get a false negative, my IgG results were obviously and conclusively negative. I didn’t register any of these antibodies to COVID-19.
But why did I have some IgM antibodies? Remember it is the IgM antibodies that peak soon after battling the illness and then give way to the IgG antibodies that begin to rise in the weeks after recovery.
Let’s list the possible scenarios.
Was I IgG deficient?
I discussed this issue in my piece on lab testing for celiac disease.
Some people are deficient overall in the antibodies the immune system makes in response to viral invaders. 2 As such, when they’ve had an illness like COVID-19, no antibodies show because none are present in high enough amounts even under normal circumstances to register as a lab value.
Out of curiosity, I’ve had my serum IgA levels tested in years past and they came back normal. Food sensitivity tests, which measure IgG responses to various food with varying degrees of accuracy, also showed some positive results, so it’s very unlikely I am IgG deficient.
Is it possible I had COVID-19 when taking the test?
I ask this question because of the small, but seemingly non-trivial, amounts of IgM antibodies I had to COVID-19. The positive range is > 1 AU/mL and I registered at 0.315.
When I called the lab, they said the numbers themselves had no inherent value, that they were only relevant against the positive and negative controls. Interesting.
When do IgG and IgM antibodies develop and do they stay elevated?
In the Journal Nature, a study conducted on 285 COVID-19 patients showed 100% IgG activity after 19 days after symptom onset.
Consider this visual from a study that mapped antibody levels in COVID-19 patients over time.
It’s pretty clear from this visual that those who have had COVID-19 reliably (but not uniformly) produce antibodies to the illness and that as the days from infection increase, IgM goes down while IgG goes up.
The odds are that, barring an immune system deficiency, if you’ve had COVID-19, you will see IgG antibody activity. However, keep in mind that a test with low sensitivity could produce a false negative.
Where do I go from here?
As I’ve detailed in this post, it’s unlikely I have had COVID-19. Having said that, the ELISA test that produced my results has low sensitivity and wasn’t FDA approved. As such, I will take another test and update this post with the results in an attempt to rule out a false negative.