COVID PCR Test Problems? This Is a Big Deal…
This past week, Elon Musk tweeted, “Something extremely bogus is going on. Was tested for covid four times today. Two tests came back negative, two came back positive. Same machine, same test, same nurse. ” While I went about my business and rolled my eyes at our testing situation, it wasn’t until this week that I found a study on this issue that blew me away. I’d like to share it today as it related to Musk’s tweet and is VERY concerning for EVERYONE. Let’s dig in.
What is PCR Testing?
At this point, you’ve likely heard the term “PCR” but may not be sure what this means. First, we now have three different types of COVID tests. The first is a rapid antigen test that detects a coronavirus specific protein in the nose, the second is a serology test that detects antibodies made by your body against the virus, and the third is PCR. We’ll focus on PCR today, even though Elon’s tweet was about an antigen test.
PCR is short for a polymerase chain reaction and it’s basically a test looking for snippets of RNA from the coronavirus. What you need to know is that it works by amplifying that RNA through cycles. So for example, let’s look at the image below:
So if we begin with one coronavirus RNA strand, each PCR cycle doubles that number. So by the third cycle, we have eight. Now anyone who has ever played with logarithmic math knows that the power of doubling is staggering. For example, if we take the test out 10 cycles, we now have 1,024 strands. By 20 cycles, we have 1,048,576 strands. In fact, every single cycle after that doubles again, so the 21st cycle produces 2,097,152 strands!
That’s the point of PCR, it’s the ability to take a very small amount of RNA and amplify it to a much bigger number of strands so that they can be easily detected. This is why this test is considered the gold standard in detecting COVID-19.Request a Regenexx Appointment
Back in June, NPR did a nice piece on what Zebra Mussels can tell us about diagnosing coronavirus cases (1). The story is about a researcher who would test lakes using PCR for the presence of Zebra Mussel DNA rather than slogging around in the lakes trying to find these invasive striped clams. He soon realized that by actually getting into the lakes to check, that his test was yielding false positives. Meaning the PCR test was positive that there was mussel DNA in the lake water, but he could find none in that lake. That got me concerned, but again, I was busy as we were just returning to the clinic, so I had better things to do than to research this any further.
Then a few months later, I heard a story about how the number of PCR cycles was impacting testing, but again, by that time I was back to being a full-time medical director, doctor, and other jobs, so I blew it off. However, yesterday I read a comment by an avid reader of this blog about a European court case on the issue. That caught my eye, as scientific matters presented in court require a careful study of the evidence. So let’s begin there and that will quickly lead to the study that blew me away and that should deeply concern you.
A Portuguese Court and PCR
A few weeks ago a court of appeals in Lisbon Portugal found that a local lockdown was illegally imposed because the PCR test that was being used to quarantine people was unreliable (2). I frankly wasn’t concerned about the lockdown or the decision, but what data was used to make this decision. After some digging, that paper is quite real, so let’s see what that’s all about.
The Paper that Blew My Mind
What the court used to make its decision is a real research paper contained in a letter to the editor published in the journal of Clinical Infectious Diseases (3). The authors were supported by a French Government grant and performed 250,566 COVID PCR tests on 179,151 patients and found 13,161 that were positive. They then decided to try to grow out the coronavirus itself from 3,790 samples. However, only about half of those samples grew any virus. They then looked at the number of PCR cycles being run and found out that if they used only 25 cycles as the cut-off point for a positive test that about 70% of patients had a sample that grew a live virus. However, if they increased that to 35 cycles, then only 3% of the samples grew out a live virus. This very paper was also highlighted in the publication Science, so I’m not the only one talking about this issue (4).
Why is this is VERY BIG deal? Because most PCR tests in the US and Europe use 35 or more cycles! For example, I Googled to find out data on tests approved by the US FDA (Emergency Use Authorizations) and quickly found the following:
- PCR test kit #1-45 cycles
- PCR test kit #2-45 cycles
- PCR test kit#3-38 cycles
- PCR test kit #4->35 cycles
So, HOUSTON, WE HAVE A PROBLEM! Our entire pandemic response rests on a test that’s being performed incorrectly. You can’t make this stuff up.
What Does this Mean?
When this happens with a diagnostic test it’s called a false positive. That means that the test shows positive when there is no disease present. In this case, there may be a few RNA strands in the patient’s nose, but they are not infected as they have no living virus capable of replicating itself. This would mean that the world-wide tests have grossly overestimated the number of COVID-19 positive cases.
What Can You Do?
First, the terminology to know is “CT Value” where CT means “cycle threshold”. Second, if you’re a healthcare provider ordering a PCR test on a patient or employee you expect may have COVID-19 based on symptoms or exposure, insist that the test have a CT value of no more than 30 cycles. This may require staff calling around to different testing companies. As a patient, you want the same info. This value should be 30 or less.
So how tough is it to find out this information? I called our Quest Diagnostic rep hotline. I got a person who was clueless about what I wanted. She tried to transfer me to a technical person, but I got nowhere. Another rep took my name for a technical call back, which I’ll update here when I get the call. In the meantime, I can’t easily find this information online for Quest. It’s also not easy to find for LabCorp. Most disturbing is the fact that both companies seem to have benchmarked their tests against other tests rather than growing out viable virus samples. In addition, the LabCorp documents seem to be much more concerned about false negatives than false positives.
So as I publish this blog this morning, as a physician, I can’t easily find how many cycles the two biggest testing companies in America (Quest and LabCorp) are using.
The upshot? Yikes! As a physician, I can’t tell you how disturbed I am that the test that we’re using to drive public policy and quarantine decisions is likely being performed incorrectly and that serious people have known about this issue since September. I’ll update this blog as I find more. In the meantime, I remain blown away.
11/26/20 Update-I have had no call back from LabCorp, which is very disappointing given that I am a clinic customer of the company.
(1) National Public Radio. What Zebra Mussels Can Tell Us About Errors In Coronavirus Tests. https://www.npr.org/sections/health-shots/2020/06/15/871186164/what-zebra-mussels-can-tell-us-about-errors-in-coronavirus-tests. Accessed 11/25/20
(2) The Portugal News Online. Court decides that quarantine in state of alert is illegal. https://www.theportugalnews.com/news/2020-11-21/court-decides-that-quarantine-in-state-of-alert-is-illegal/56830. Accessed 11/25/20
(3) Jaafar R, et al. Correlation Between 3790 Quantitative Polymerase Chain Reaction–Positives Samples and Positive Cell Cultures, Including 1941 Severe Acute Respiratory Syndrome Coronavirus 2 Isolates, Clinical Infectious Diseases, ciaa1491, https://doi.org/10.1093/cid/ciaa1491
(4) Science. One number could help reveal how infectious a COVID-19 patient is. Should test results include it? https://www.sciencemag.org/news/2020/09/one-number-could-help-reveal-how-infectious-covid-19-patient-should-test-results Accessed 11/25/20