High SARS-CoV-2 viral load is associated with a worse clinical outcome of COVID-19 disease

Introduction

Coronavirus SARS-CoV-2 emerged in the human population in 2019 is the causal agent of the new pandemic disease COVID-19 (1). The virus has spread rapidly worldwide, and at the time of this writing there are 46,403,652 confirmed cases, and 1,198,569 deaths in 219 countries worldwide, according to the WHO (https://covid19.who.int/); these numbers are increasing daily. Evolution of a virus in a specific host is defined by a number of closely related parameters such as viral load, replication rate, genetic heterogeneity and viral fitness that may influence virus adaptability, viral pathogenesis and disease progression (2, 3). The replicative capacity of a virus is clinically relevant because it largely determines the viral load in infected individuals, and viral load influences disease manifestations (4). In the case of SARS-CoV-2, some studies have correlated viral load with disease severity whereas in others this correlation was not clear. A positive correlation was reported in a cohort of SARS-CoV-2-infected patients from China, showing that the viral load detected in the respiratory tract was positively linked to lung disease severity (5). In a related study, the analysis of the viral RNA level in upper respiratory tract samples from 76 patients with COVID-19 revealed significantly lower Ct values (cycle threshold which is inversely correlated with viral RNA level), and longer virus-shedding periods in those patients classified as severe, as compared with those that exhibited mild disease (6). Additionally, a prospective study in a large hospitalized cohort of 1,145 infected patients documented a significant lower probability of survival in patients with high viral load than in those with low viral load (7). In contrast, in other studies, it was observed that the diagnostic viral load level was lower in hospitalized than in non-hospitalized patients, resulting in a lack of correlation of viral load with admission to intensive care unit (ICU), length of oxygen support, and overall patient survival (8). Thus, the dynamics of viral load and its connection with different clinical parameters remain poorly characterized, and large studies with additional cohorts worldwide are needed to define the possible association and the predictive value of the viral load regarding disease progression and mortality. Viral RNA load calculated from nasopharyngeal swabs might be added to other predictive parameters to complete an early risk stratification of COVID-19 patients (9).

Results and Discussion

Given the disparate results trying to correlate viral load with COVID-19 disease progression, we have addressed this question with a large cohort of 448 patients admitted to the Fundación Jiménez Díaz Hospital (FJD, Madrid, Spain) from April 3 to 29, 2020 coinciding with the first COVID-19 outbreak in Spain. At the time of admission, all patients had clinical symptoms, and were confirmed to be positive for SARS-CoV-2 by a specific real-time RT-PCR (VIASURE Real Time PCR). The Ct values analyzed correspond to the first available COVID-19 PCR-positive sample for each patient. Details regarding the clinical classification are described in Materials and Methods. Mean Ct values for mild, moderate and severe COVID-19 patients were 35.75 ± 0.45, 32.69 ± 0.37, and 29.58 ± 0.70, respectively. A univariate analysis showed statistically significant differences among viral load values according to the infection severity (p<0.0001; ANOVA test) (Figure 1). Specifically, average Ct values were significantly lower in the severe group as compared with the moderate and the mild disease groups, and also for the comparison between moderate and mild clinical categories (p<0.001 in all cases) (t-test with Bonferroni correction). Thus, Ct values (as a measurement of viral load) in SARS-CoV-2-infected patients correlates positively with the disease progression and poor prognosis.

• Download figure
• Open in new tab

Figure 1.

Correlation between viral load [measured by the Ct (number of PCR amplification cycles needed to cross the threshold detection level) value] and the severity of COVID-19 disease.

Regarding host factors, it has been established that age greater than 65 is a risk factor to develop acute respiratory distress syndrome (ARDS), a major complication of COVID-19 pneumonia, and that the risk of death increases with advanced age (10). As a second factor, disease severity and mortality for males is significantly higher than for females, being older men the population most at risk (11, 12). Additionally, several co-morbidities have been potentially associated with poor outcome including hypertension (high blood pressure), cardiovascular disease, diabetes, obesity, and lung disease (asthma, EPOC) (13). We have included these risk factor data in our patient cohort to assess their alignment with our disease severity-viral RNA load correlation. Interestingly, a significant association between viral load and infection severity was still observed after adjusting for age, sex, hypertension, cardiovascular disease, diabetes, obesity, asthma and chronic obstructive pulmonary disease (COPD) (see ANCOVA tests in Table 1). The significant difference in viral load between the three groups was not attributable to the percentage of hospitalization, the percentage of ICU admission or the percentage of the different types of oxigenotherapy (see ANCOVA tests in Table 1). Thus, our results with a SARS-CoV-2 population from a cohort which is different from those in previous studies on viral load influence on disease, suggest a positive correlation between viral load and COVID-19 disease severity. This conclusion is also in agreement with two studies showing a longer persistence of high viral load viruses in respiratory samples of patients with severe disease than those with mild disease, suggesting that the viral load may be a prognostic parameter (14, 15).

To monitor the average number of days between the symptom onset and the sample extraction in our cohort, we collected these data for 102 patients. On average NP swabs were obtained at 7.84 ± 6.40 days after symptoms onset, which is within the time interval of active infection. It has been described that the highest SARS-CoV-2 viral load in throat swabs -and consequently the highest transmissibility peaks- is around 5-6 days after the symptom onset (16), but this is an average that depends on each patient; the transmissibility window extends from a few days before symptom onset up to 30 days in patients with severe disease (14, 17).

A correlation between high viral load and disease severity has been also found in other viral infections. Among children naturally infected with respiratory syncytial virus (RSV), increased viral load was associated with clinical severity of disease defined as an increased risk for intensive care, prolonged hospitalization, or the development of respiratory failure (18). Examination of hepatitis A virus (HAV) RNA from sera by real-time PCR resulted in higher initial viral load in patients with severe outcomes such as fulminant hepatitis and severe acute hepatitis than in patients with less severe infection (19). Viral load in patients infected with the pandemic type A influenza virus H1N1 (2009) who suffered pneumonia was higher than in patients with milder disease (those with bronchitis or upper respiratory tract infection), suggesting that the viral load is also an important predictive value in influenza infection (20). Other instances of a connection of viral load with disease progression have been reviewed (2).

In terms of viral quasispecies dynamics, a key question regarding clinical implications is that a population with a large viral load will contain a broader mutant repertoire than a population with low viral load. Even if each individual mutant might be present at the same frequency in both populations (as it corresponds to mutant frequency being an “intrinsic” property), the mutant repertoire is an “extrinsic” property of the population, a key distinction that has been numerically studied (21). Thus, viral load is a potential source of variants that enhance the probability of infection to alternative cell types in the course of infection which is increasingly viewed as a step-wise adaptation process (2-4, 21).

Methods

Authors contributions

CP, IG, RF-R and JE designed the study. AM-V, IG, RF-R and JE collected samples and validated RT-PCRs. MES, BM-G, RL-V and LV-S analyzed Ct and clinical data. MC, RL-R, BA, PM, AH, JCT, JR-H, CA and CP classified clinically SARS-CoV-2 patients. IM performed statistics. JMA, JR-H, CA and CP supervised analysis. CP, MES, BM-G, RL-V and LV-S wrote the manuscript. All authors reviewed the manuscript.