Read on Twitter
One aspect of testing that I’m interested in (but absolutely no expert) is the correlation between +ve qPCR, and subsequent viral culture. The idea being that if you test +ve by PCR but can’t culture it, you are significantly less (or not) infectious. Sounds too easy.
This paper just came out in NEJM which is why I think it’s turned up on my timeline recently. There are many papers like this one out there but this one seems easy enough to demonstrate the concept. https://www.nejm.org/doi/full/10.1056/NEJMc2027040
Just one figure: Y axis is the Ct value (lower number implies more starting material i.e. more virus). Red dots are the samples that were able to be cultured … which makes sense that it correlates with the lower Ct value. All good.
But... how sure are we that those blue dots are not infectious based on culture? Would I let a blue dot (-ve culture), 7d after symptom onset with a Ct value 26-30, visit us? I’ve wondered to what extent you can really be sure of that, because it ‘sounds’ like a tricky assay.
I’ve been tagged in a couple of interesting discussions on the topic recently so wanted to share those. Turns out that there are MANY pitfalls with this assay, and that a person who tests positive by PCR that has no positive culture may indeed be infectious.
It sounds like we would run into people testing +ve for PCR and then not being able to culture the virus due to technical problems. I have no doubt that these things will correlate, namely high viral load, low Ct, more infectious, but seems hard to reliably perform en masse.
It's been suggested that we should integrate viral culture into our testing infrastructure (see paper below).
Based on the opinions above it sounds like it could get us into trouble. There are many technical challenges that seem prohibitive. https://academic.oup.com/cid/advance-article/doi/10.1093/cid/ciaa1764/6018217
Based on the opinions above it sounds like it could get us into trouble. There are many technical challenges that seem prohibitive. https://academic.oup.com/cid/advance-article/doi/10.1093/cid/ciaa1764/6018217
Is it a pipe dream to do this on a national level under such intense time pressure? It seems like it may be too error prone a technique to be useful in a pandemic response.
Would like to hear more.... post it up.
Would like to hear more.... post it up.
