Testing for COVID19 has become so much the norm in society that in casual conversation you might ask someone if they have been tested. Chances are you know someone who has and even more likely that it is many people at this point. If you have been following this blog for awhile you would know that one of the issues with testing is that it is extremely sensitive. The RT-PCR test can pick up even trace viral particles whether alive or dead and therefore a test in and of itself tells you one thing. There is virus present but whether it is alive or dead is another matter. Virus can be shed for weeks after symptoms develop so what information can we obtain that might tell us whether the patient is at risk of infecting another person. Moreover, when can we relax precautions around the patient themselves and no longer worry that they are a risk to others?

Winnipeg Researchers May Have the Answer

Given that I know many of the authors of the paper personally that I will discuss and the quality of the work they do I am delighted to cover this important work. The paper Predicting Infectious Severe Acute Respiratory Syndrome
Coronavirus 2 From Diagnostic Samples
by Bullard J et al tackles the question above using cycle times and times to positivity after initial symptoms. First off it is important to understand the concept of cycle time. When you use RT-PCR to amplify pieces of RNA there are 40 cycles of amplification that a sample is put through. The lower the number of cycles required to detect the RNA the more viral material was present to begin with. As the cycle times get well into the 30s the possibility of there being just trace amounts of virus exists or that the patient themselves had no virus present but the sample was contaminated with a very small quantity of virus. The time to positivity is the amount of hours/ days it takes after initial symptoms develop until an RT-PCR test becomes positive.

The authors analyzed 90 samples either from naso-pharyngeal swab or ETT secretions with a median age of 45 years (I know not neonatal). Collected samples included from day 0 to day 21 after symptom onset. What the authors did that was interesting in this paper was that in addition to the samples being tested by RT-PCR, for each positive sample they went through the laborious task of performing tissue culture to attempt to grow the virus. By doing so the authors were then able to compare the time to positivity (STT) and Ct thesholds to determine if there were numbers for each that could be used to predict which samples would have live virus that could be grown,

The Results Please

The two main findings in the paper were that no patient after day 8 from symptom onset with a positive RT-PCR could actually have live virus grown in tissue culture. This is shown in Figure 3 from the paper

Secondly, no patient with a cycle time greater than 24 could also have live virus grown. In fact for every 1 unit increase of Ct the odds ratio for infectivity decreased by 32%.

The size of the study is fairly large when it comes to COVID19 studies like this but as the authors say the results are for adults and given how few children have been infected especially in our location in Canada we should be hesitant to generalize to children and in particular neonates.

What the study does give us

This study I felt was worth sharing with you for the reassurance that I think it may give especially if it leads to further validation by even larger studies including children and those who are immunocompromised who may have prolonged shedding. What it suggests though is that the next time you encounter a patient who you are told tests positive there are a couple important questions to ask from your lab. The first is how many days from the start of symptoms was the RT-PCR done? Secondly, ask them for the Ct value. As per the authors

“Receiver operating characteristic curves constructed using Ct vs positive culture showed an area of 0.91 (95% CI, .85–.97; P < .001) with 97% specificity obtained at a Ct of > 24. Similarly, STT vs positive culture showed an area of 0.81 (95% CI .73–.90; P < .001), with 96% specificity at > 8 days”

In other words, if the answers to those questions are >8 days and a value for Ct >24 you should be able to leave the patient with reasonable expectation that they are either no longer infectious or at worst almost clear of live virus.

Interesting work that I hope will be helpful to someone out there!