Viral Load and the Delta Variant

Last November I recorded a talk for the Association of Molecular Pathology (AMP) with my friend and clinical ID colleague, Dr. Inessa Gendlina. The topic of the discussion related to the utility of Ct values as it relates to SARS-Cov2 RT-PCR and framing a clinical understanding of where Ct values are/are not useful as it relates to clinical patient care. Recently, I continue to find literature relating the higher viral loads (lower Ct values) associated with infections with the Delta variant. This is obviously concerning, and while I agree with the data which demonstrates lower average Ct values, I believe the interpretation that these patients simply carry more virus may not be the entire story. In fact, another explanation of this finding must be considered as it may have some broad infection control implications.

For the average patient the diagnosis of a COVID infection is brought to light by a single PCR result. Additional testing throughout the course of this patient's infection is not necessary (as it does not typically guide care) nor is it recommended as individuals may retain PCR positivity for some time. Using PCR positivity for things like return to work can limit availability of essential personnel who are no longer carriers of replication-competent virus, yet they may retain viral RNA positivity by PCR. The single data point obtained from most infections made truly assessing the viral kinetics in these patients difficult but a typical rise and fall of viral RNA could be assumed and plotted overall from strong data collected correlating time from symptom onset and viral burden. 

A great service was performed by the professional sports leagues in that their performance of rigorous, continual monitoring of athletes allowed for accurate (see Kissler et al) assessment of viral kinetics. The rise and fall (or fall and rise using Ct values) could now be plotted definitively. Additionally, at this same time several studies had been looking at assessing infectivity of a patient by correlating both time from infection and secondary attack rates (Cheng HW, JAMA) and by correlating viral loads of a patient samples with isolation of replication competent virus (Bullard J, CID). These data and many more are the firm basis on which CDC produced their recommendations regarding the duration and cessation of isolation using a symptom and time based approach.

I believe that it is important to consider that a perfectly rational explanation of the finding that patients infected with Delta appear to have, on average, more abundant viral loads in their samples is not simply that Delta causes more virus to be present. It is entirely possible that there are definite differences in the variant viral kinetics. A broader replicative peak would mean that individuals who present slightly later in their infections, no longer present on the declining portion of their individual infection curve (timepoint D in image), but at a portion of a peak which still presents with high viral loads and low Cts. Given the variability in when a patient presents for testing this can easily explain the finding of higher average viral loads. Looking at the graphic one can immediately see that even if the maximum viral loads are not any higher in Delta infections, the mean viral burden would appear higher (and the mean Ct lower) in those cases given the larger window of peak infection.

With that said there is good news and bad. The good news is that we can produce experimental evidence which will teach me to get my head out of the clouds. The sports franchise data would easily support or completely disprove this hypothesis. The bad news, however, is that if found to be correct a reassessment of the timelines associated with the current symptom-based cessation of isolation may require re-evaluation.