Discussion
Among symptomatic hospitalised and outpatient COVID-19 cases at the beginning of the pandemic in Italy, we observed that disease severity, death, 6-month sequelae and the number of signs and symptoms at diagnosis distributed according to the amount of nasal-pharyngeal SARS-CoV-2 detected within the first week from disease onset, independently from other known determinants of COVID-19 severity.
While for other viral illnesses initial viral load has been associated with disease severity 17,18,25, a consensus has not been reached in COVID-19. To the best of our knowledge, 11 studies have shown that nasal-pharyngeal SARS-CoV-2 Ct are not associated to nor predict COVID-19 severity 13–15,19,26–32. However, only six of them analysed samples with >100 cases, the majority included only hospitalised patients, and the follow-up was relatively short 13–15,19,26–32. Significantly, only a minority took into account the time between COVID-19 onset and the time of swab collection. As example, an inverse correlation between viral load and pulmonary lesions at CT scan has been described: the higher the viral load in nasopharynx, the milder was the pneumonia32. Considering viral kinetics and immune-pathology dynamics 5,27,29,33, the lower viral load could have been associated with worse imaging as in this group SARS-CoV-2 may have had longer time to replicate and diffuse in the lower respiratory airways, stimulating the noxious inflammatory response. Patients with the highest viral load in the upper respiratory airways may have been detected at an earlier stage, when lung involvement had not yet had time to reach its zenith 32.
If it is not a matter of snapshot issues, it could be a matter of grey scale. Some of the studies supporting no difference in disease severity according to viral load did not include the whole spectrum of COVID-19 and compared either asymptomatic with very mildly symptomatic patients or not hospitalized with hospitalized patients, with the latter ranging from mild to critical symptoms 14,15.
On the opposite, studies where a positive association between nasal-pharyngeal SARS-CoV-2 Ct values and COVID-19 severity and/or outcomes was detailed are flourishing 6,7,9–12. Most of them included samples of more than 100 individuals with the reference swab represented by the diagnostic one, despite only one reported the time between disease onset and swab collection 7. As compared to previous studies, the one here described has the longest follow-up and included non-hospitalized patients. Furthermore, our retrospective design at 6 months from COVID-19 onset and the availability of data on time between disease onset and swab collection allowed us to adjust for this factor as well as to analyse outcomes and parameters of the infection at a time when its evolution was surely over.
Lower Ct values were associated with more signs and symptoms at diagnosis and a more frequent pattern of respiratory and systemic complaints. Our findings are in line with those recently published by others 8,12,34. As for the latter34, we did observe higher viral load in febrile patients compared to afebrile ones: a greater burden of virus could induce higher inflammatory response primarily manifesting as fatigue, malaise, fever and headache. In accordance with this, plasma and nasopharyngeal SARS-CoV-2 RNA has been correlated with pro-inflammatory cytokines and inflammation biomarkers levels, such as IL-6 and CRP/serum amyloid A ratio 19,20.
On the contrary, we did not observed any association between nasal-pharyngeal viral load and prevalence of olfactory/taste disorder as described by others and plausibly explained by higher local amount of virus and related inflammation 34. Nevertheless, we did not quantify the dysfunction with an objective scale and even others failed in finding such an association 35. Olfactory and gustatory dysfunction represented the second most common sequelae (7.7%) still present at 6 months from the infection, after dyspnoea and just before chronic cough. Other studies reported persistent olfactory and gustatory dysfunction at 6 weeks and at 3 months in the 28.2-16.7% (smell-taste loss) and 10.3% of individuals 35,36. Up to 6 months, the dysfunction may persist in a small proportion and, together with other sequelae was more common in patients with lower Ct values regardless of age, gender, comorbidities and infection severity.
Our study relied on data collected by surveys and electronic medical records of hospital admitted patients but also outpatient cases; thus, it could be less controlled compared to studies focusing on inpatient cases only, lacks of data on treatments, and did not differentiate further the category of patients requiring from low-flow oxygen to reservoir mask. However, in March, most of the potentially impactful treatments were not available, while recall bias may have affected signs and symptoms reliability, but less likely that of the other outcomes. A classification of COVID-19 severity based on the worst oxygen support during hospitalisation should be carefully used as single parameter, since biological variables depending on disease severity such as the ratio of arterial oxygen partial pressure to fractional inspired oxygen may not directly correlate with the provided oxygen support, especially in case of intubation and CPAP in elderly or comorbid patients. Lastly, we faced a 13% of non-responder to the survey that partially unbalanced the representativeness of our sample compared to the Ct distribution of the sampling frame.
The currently emerging evidence is that early Ct values from nasal-pharyngeal swab correlate with disease susceptibility (age and comorbidities, such as smoking and hypertension, both associate with potential differential expression of ACE2 37,38) and clinical presentations and predict disease severity, survival and sequelae.
Whether a better definition of these patterns could help at triaging newly diagnosed cases is pending, as well as potential application of SARS-CoV-2 viral load in routine clinical practice. Reporting predictive virological parameters, as already done for other infections, could inform clinicians management and strategies for monitoring and allocating resources; this may be especially useful during the climax of SARS-CoV-2 waves when hospital beds and resources are limited and require programmed and evidence-based selections and timely risk stratifications.
The interpretation of a single Ct value should still be performed cautiously as it may be affected by sample collection, analysed gene, adopted assay, and analytic limits 39. Longitudinal studies should evaluate further viral dynamics to better interpret Ct values in accordance with the moment of swab collection along COVID-19 course. Indeed, quantitative assays based on RT-PCR and other techniques are under development to exactly quantify SARS-CoV-2 RNA and may be soon introduced in routine clinical management, integrating qualitative tests by overcoming many of these limitations.