As of the latest update by the World Health Organization (WHO), the cumulative number of coronavirus disease-19 (COVID-19) cases has exceeded 760 million, with a total of approximately 6.9 million deaths reported globally since the start of the pandemic [1]. A large number of studies have been conducted to understand the COVID-19 pathogenesis and the risk factors of disease severity and mortality. Severe cases and death of COVID-19 are associated with older age, comorbidities, organ dysfunction, lymphopenia, high cytokines, hypoalbuminemia, and weak immune responses [2, 3]. Laboratory parameters such as C-reactive protein, lactate dehydrogenase, neutrophil-to-lymphocyte ratio, fibrinogen, D-dimer, interleukin-6, platelet count, and lymphocyte count have shown promise as more objective measures for assessing risk assessment in COVID-19 patients [4]. These parameters can be valuable in predicting the prognosis of the disease, which, in turn, can be utilized to establish guidelines for treatment and prevention strategies.

The rapid and accurate diagnosis of COVID-19 is crucial, and real-time reverse transcription-polymerase chain reaction (RT-PCR)-based diagnostic tests are considered the gold standard for detecting severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections. The cycle threshold (Ct) value in RT-PCR is the number of amplification cycles needed for the fluorescence signal to surpass the set threshold level. As a result, the Ct value is inversely correlated with the viral load in the sample [5]. Several studies have analyzed the association between Ct values and disease severity, mortality, and infectivity, demonstrating conflicting results [6]. Most studies on this topic have included a limited number of patients, and there is a lack of research specifically focusing on Korean patients. As of now, the exact relationship between Ct values and disease severity in COVID-19 patients remains unclear [5].

Daegu, South Korea, experienced the first COVID-19 outbreak outside Wuhan, China. The number of infected patients grew rapidly since February 18, 2020, and by April 30, 2020, 6,933 confirmed cases were reported in the region [7]. In the present study, the correlation between initial Ct values of real-time RT-PCR for SARS-CoV-2 and clinical severity in hospitalized patients in Daegu was explored.

1. Patients

We analyzed the epidemiological survey data for the first large COVID-19 outbreak in Daegu collected by the Daegu Metropolitan Government. All patients were confirmed with a positive test result using real-time RT-PCR for SARS-CoV-2 at a screening clinic in a health center or a hospital. The Korean government classifies COVID-19 cases by severity to prioritize the treatment of more severe cases at hospitals, while asymptomatic patients and those with mild symptoms receive medical treatment and monitoring at community treatment centers. Patients with more disease severity are admitted to infectious disease hospitals or nationally designated treatment facilities, depending on the patient’s condition, for immediate in-hospital treatment. The present study enrolled a total of 750 hospitalized COVID-19 patients admitted from February 17, 2020, to April 29, 2020, in 8 hospitals in Daegu. All patients underwent real-time RT-PCR with upper and/or lower respiratory tract (URT/LRT) specimens at admission.

The clinical severity of the COVID-19 patients was assessed based on an 8-category modified scale by the Korea National Committee for Clinical Management of COVID-19, a modification of the WHO 10-category ordinal scale [8], as follows: 1) no limitation in performing daily activities; 2) limitation in performing daily activities but no need for supplemental oxygen therapy; 3) need for supplemental oxygen therapy via a nasal cannula; 4) need for supplemental oxygen therapy via facial mask; 5) need for high-flow supplemental oxygen therapy or noninvasive mechanical ventilation; 6) need for invasive mechanical ventilation; 7) multi-organ failure or the need for extracorporeal membrane oxygenation (ECMO) therapy; and 8) death. Out of 750 patients, we obtained classification results for 716 patients according to clinical severity. We grouped these patients into non-severe (categories 1 to 4) and severe groups (categories 5 to 8). The real-time RT-PCR kits used in these patients are shown in Table 1. This study was approved by the Institutional Review Board of the Daegu Catholic University Medical Center (IRB no. CR-20-052).

Table 1 . SARS-CoV-2 real-time RT-PCR kit used in the present study and the number of patients by clinical severity group

Test kit (Manufacturer)	Target genes	Cut-off Ct value	Number of patients
			Non-severe group	Severe group	Total
Allplex SARS-CoV-2 Assay (Seegene, Seoul, Korea)	E, RdRP, N	< 40	511	11	522
PowerChek 2019-nCoV Real-time PCR Kit (KogeneBiotech, Seoul, Korea)	E, RdRP	< 35	139	6	145
DiaPlexQ Novel Coronavirus Detection Kit (SolGent, Daejeon, Korea)	ORF1a, N	< 40	40	1	41
Real-Q 2019-nCoV Detection Kit (BioSewoom, Seoul, Korea)	E, RdRP	< 38	8	0	8

Abbreviations: Ct, cycle threshold; RT-PCR, reverse transcription polymerase chain reaction.

2. Real-time RT-PCR test of SARS-CoV-2

The entire real-time RT-PCR test process, including specimen collection and transport, test procedure, interpretation, and test reporting, was conducted according to the published guidelines for diagnosing COVID-19 by the Korean Society for Laboratory Medicine (KSLM) and the Korea Centers for Disease Control and Prevention (KCDC) [9]. The reagents used for real-time RT-PCR in the eight hospitals (Table 1) were as follows: Allplex SARS-CoV-2 Assay (Seegene, Seoul, Korea), PowerChek 2019-nCoV Real-time PCR Kit (KogeneBiotech, Seoul, Korea), DiaPlexQ Novel Coronavirus Detection Kit (SolGent, Daejeon, Korea) and Real-Q 2019-nCoV Detection Kit (BioSewoom, Seoul, Korea). Each reagent targets two or three virus-specific genes out of the envelope (E), RNA-dependent RNA polymerase (RdRP), open reading frame 1a (ORF1a), and nucleocapsid (N) genes. KSLM and the KCDC recommend a positive result determination only when all target genes are detected. If one or more, but not all, genes included in the kit were amplified, the results were interpreted as indeterminate. All results were reported after being confirmed by laboratory physicians. We collected Ct values for each target gene of real-time RT-PCR at the time of admission.

3. Statistical analyses

Continuous variables are expressed as mean±standard deviation values, while categorical variables are presented as frequency and percentile. We used the Wilcoxon signed-rank and Kruskal-Wallis tests to compare the Ct values according to sample types, target genes, and severity groups. Data analyses were performed with the R software version 4.0.0.

A P-value of <0.05 was considered statistically significant.

1. Demographic data

The mean age was 55.5±20.7 years, and 62.1% (N=466) of the patients were female (Table 2). Those over the age of 50 accounted for 64.1%. The number of cases in participating hospitals was the highest in the F hospital, a designated hospital for COVID-19 in Daegu. The patients were hospitalized approximately 3 days after being confirmed with COVID-19. The total of 716 patients was classified according to clinical severity as follows: 523 (73.0%) in category 1, 60 (8.4%) in category 2, 81 (11.3%) in category 3, 35 (4.9%) in category 4, and 17 (2.4%) in categories 5 to 8.

Table 2 . Demographic data of the 750 COVID-19 patients in Daegu, South Korea

Characteristics
Total Number		750
M:F ratio		0.61:1
Mean age (yr)		55.5 ± 20.7
Age distribution (yr)		N (%)
	1-10	12 (1.6)
	11-20	19 (2.5)
	21-30	101 (13.5)
	31-40	48 (6.4)
	41-50	89 (11.9)
	51-60	143 (19.1)
	61-70	140 (18.7)
	71-80	123 (16.4)
	81-99	75 (10.0)
Hospital	A	14 (1.9)
	B	14 (1.9)
	C	109 (14.5)
	D	124 (16.5)
	E	36 (4.8)
	F	314 (41.9)
	G	109 (14.5)
	H	30 (4.0)
Days from diagnosis to admission (N = 613)	3.0 ± 6.6

2. Results of real-time RT-PCR at the time of admission

All 750 hospitalized patients underwent SARS-CoV-2 real-time RT-PCR with URT and/or LRT specimens. A total of 716 patients underwent RT-PCR using URT samples. Of these, 633 (88.4%) patients were reported as positive, 79 (11.0%) as indeterminate, and 4 (0.6%) as negative. RT-PCR with LRT samples was performed only in 192 patients. Of these, 158 (82.3%) patients were reported as positive, 26 (13.5%) as indeterminate and 8 (4.2%) as negative. Table 3 shows the Ct values of real-time RT-PCR according to the specimen at the time of admission. The Ct values in the E and RdRP genes showed variations based on the sample type. The Ct values of the LRT samples were significantly lower than those of the URT samples. There was no difference in Ct values between the E and RdRP genes (P=1.000); however, there was a statistically significant difference in Ct values among the four genes (P<0.001).

Table 3 . Ct values of SARS-CoV-2 real-time RT-PCR according to the sample type at the time of admission

Target gene	Cycle threshold
	Upper respiratory tract	Lower respiratory tract	P
E	30.04 ± 5.45	28.37 ± 6.19	0.001
RdRP	31.56 ± 5.34	29.75 ± 6.23	< 0.001
ORF1a	37.16 ± 2.40	36.86 ± 1.56	0.4637
N	32.32 ± 5.03	31.68 ± 6.28	0.6333

3. Disease severity and Ct value of SARS-CoV-2 real-time RT-PCR

The correlation between the Ct value in the URT sample and clinical severity was analyzed for 716 patients with known clinical severity. The Ct values of the E, RdRP, and N genes showed significant differences among the eight severity categories (P<0.001, P<0.001, and P<0.001, respectively). Patients in the severe group exhibited lower Ct values for the E (P<0.001), RdRP (P=0.029), ORF1a (P=0.009), and N genes (P=0.003) compared to those in the non-severe group. When analyzing patient data using the Alleplex SARS-CoV-2 Assay reagent kit, the Ct values for the E, RdRP and N gene according to clinical severity are presented in Table 4, showing significant differences (P=0.002, P=0.003, and P=0.001, respectively).

Table 4 . Ct values of Allplex SARS-CoV-2 Assay according to clinical severity

Target gene	Cycle threshold
	Non-severe group	Severe group	P
E	28.19 ± 5.08	21.68 ± 6.47	0.002
RdRP	30.40 ± 5.10	24.20 ± 6.38	0.003
N	31.06 ± 4.88	24.62 ± 6.29	0.001

In the present study, we analyzed the correlation between Ct values of real-time RT-PCR and clinical severity of SARS-CoV-2 infections in 716 patients in the Daegu outbreak, South Korea. The Ct value at the time of admission was significantly lower in patients with severe disease.

Notably, the Ct value is assumed as a surrogate marker for viral load [10] and is inversely related to the amount of the virus. During previous pandemics of MERS-CoV and SARS-CoV, studies reported a similar association between low Ct values and the severity of infection [11, 12]. Additionally, in patients with influenza B infection, higher viral loads were detected in hospitalized patients compared to ambulatory patients [13].

Numerous prior studies have suggested that the Ct value obtained from initial RT-PCR can predict COVID-19 patient outcomes during hospitalization. Moreover, a correlation exists between higher SARS-CoV-2 viral load and increased severity of illness and mortality in COVID-19 patients [14, 15]. Tsukagoshi et al. showed that the viral load in fatal cases was significantly higher than in symptomatic or asymptomatic cases [16]. They suggested that a high viral load of SARS-CoV-2 in elderly patients at an early stage of the disease results in a poor outcome. Furthermore, Kociolek et al. conducted a study involving 817 children, showing that asymptomatic children had lower viral loads in their URT than children with symptoms [17]. Therefore, the amount of virus can be considered as a prognostic factor regardless of age. Pujadas et al. showed that a high viral load correlated with a higher mortality rate and suggested using quantitative analyses for patient risk stratification [18]. According to a study, even mild patients had lower Ct values at the time of community treatment center entry compared to asymptomatic patients [19]. This suggests that the Ct values obtained from SARS-CoV-2 RT-PCR tests could potentially serve as a factor to predict the severity of COVID-19.

However, various clinical studies with both small and large sample sizes have shown discrepancies in establishing a significant correlation between the Ct value and disease severity in COVID-19 [6]. Le Borgne et al. demonstrated that the nasopharyngeal viral load in emergency departments was not a predictor of disease severity and mortality in COVID-19 patients [20]. Similarly, a cohort study conducted by Lee et al. found no significant difference between viral load and the clinical course of the disease [21]. The meta-analysis of 34 studies revealed an uncertain relationship between COVID-19 severity and viral load [22]. These discrepancies could be attributed to variations in the study population, study design, sample type, sample collection time, sample preparation, RT-PCR reagent, test process, and other factors.

The present study has several limitations. Firstly, although the WHO uses Ct values as a surrogate for the level of viral load, it is important to note that the American Association for Clinical Chemistry (AACC) has emphasized that Ct values should not be used for managing patients with COVID-19 [23]. The lack of standardization of SARS-CoV-2 RT-PCR tests in different countries and variations in sample collection methods from the respiratory tract contribute to the differences observed in Ct values. Additionally, variations in the performance of equipment and reagents can also impact the results. However, the URT samples were collected by healthcare workers according to the guidelines in the present study, hence maintaining a certain level of consistency in the sampling process. Furthermore, to ensure consistency and reliability, all eight hospital laboratories adhered to the published guidelines provided by the KSLM and the KCDC throughout the entire test process, including sample transport, test procedures, and result interpretation [9]. By following these common guidelines, the potential variations between laboratories were minimized, enhancing the accuracy and comparability of the data obtained in the study. Upon analyzing patient data using the Alleplex SARS-CoV-2 Assay reagent kit, a notable difference in Ct values was observed for each target gene based on clinical severity. Consequently, if the same reagent is consistently used within a laboratory, the Ct value could serve as an informative indicator for predicting clinical severity. Although this study did not establish a specific cut-off due to data limitations, it may prove valuable to determine a severity cut-off for each reagent kit in future research. Second, the Ct value in the present study represents the value at the time of admission and may not correspond to the value at the time of diagnosis. Hospital admission was typically performed approximately 3 days after diagnosis owing to a lack of available beds, resulting in varying hospitalization times among patients. Delayed hospitalization time can affect the Ct values of real-time RT-PCR and thus influence patient severity. Nevertheless, the Ct values at the time of admission were significantly lower in patients who required intensive treatment such as high-flow supplemental oxygen therapy, noninvasive mechanical ventilation, invasive mechanical ventilation, multi-organ failure, ECMO therapy, or fatal cases compared to patients with relatively mild disease. Lastly, various factors have been identified as predictors of the severity of COVID-19, including old age, male sex, underlying health conditions, and specific laboratory parameters [3, 24]. Although the present study did not analyze all these co-factors, it is essential for medical staff to take them into account during hospitalization. In this context, objective parameters like Ct values from real-time RT-PCR can serve as valuable tools to complement the overall assessment and help in predicting the severity of the disease.

Ct values might be affected by several pre-analytic, analytic, and post-analytic variables such as collection technique, specimen type, sampling time, viral kinetics, transport and storage conditions, nucleic acid extraction, viral RNA load, primer design, real-time PCR efficiency, and the method used to determine the Ct value. However, given the absence of a singular definitive prognostic factor in COVID-19 patients, Ct values derived from real-time RT-PCR tests upon admission can serve as a supplementary indicator for prognostic prediction. Accurate prognosis prediction and appropriate treatment planning are pivotal for effectively managing COVID-19 patients with diverse clinical severities. Several studies have reported various prognostic markers [3, 24]; however, more accumulated data and evidence are required owing to the diverse presentation and outcomes observed in COVID-19 patients.

In the present study, we report an association between initial Ct values of real-time RT-PCR and disease severity on admission. Our data suggest that the Ct value can be a supportive parameter for predicting the severity of COVID-19 patients. Therefore, the initial Ct value reported on hospital admission may serve as a risk stratification tool.

None declared.