Viral Load, Detection Rate and Sensitivity of Nasopharyngeal and Oropharyngeal Swab Sampling for Diagnosis of COVID-19

The rapid spread of the COVID-19 pandemic has led to a major public health crisis. Accurate screening methods for COVID-19 infection is essential and crucial for case detection, isolation, prevention and control of the current pandemic. At present, nasopharyngeal and oropharyngeal swabs are typically used as the method of choice for the diagnosis of SARS-CoV-2 infection. We carried out a review on the accuracy of the two different sampling sites, the nasopharyngeal and oropharyngeal swab sampling, focusing on the viral load, detection of positive cases and sensitivity in real-time polymerase chain reaction (RT-PCR) assay in diagnosing COVID-19. A total of 25 articles related to the topic were selected out of 5221 articles searched online using Scopus, PubMed and Medline, Embase, Web of Science, and Google scholar with the keywords COVID-19, SARS-CoV-2, nasopharyngeal swab, oropharyngeal swab, nasal swab and throat swab. All full text original articles were obtained and reviewed. Nasopharyngeal swab had Ulum Islamiyyah Journal | Special Issue: Healthcare in Pandemic Era: “The New Norm” Vol.2 36 significantly higher SARS-CoV-2 load than oropharyngeal swab (mean Ct value ranging from 24.3-37.8, higher detection of positive rate (highest rate 62.5%) and sensitivity (highest sensitivity 98.3%, P<0.05) in RT-PCR assay compared to oropharyngeal swab. Based on the scientific literature review, both nasopharyngeal and oropharyngeal swabs were reported to have 30% probability of yielding false negative results; thus clinically suspicious patients with negative results should be viewed with concern. In conclusion, although several methods of COVID-19 screening and type of specimen are available, nasopharyngeal swab is the best option for large scale screening as it yields significantly higher viral load, higher detection of positive rate among cases and higher sensitivity in RT-PCR assay compared to oropharyngeal swab in detecting SARS-CoV-2.


Introduction
The first case of Coronavirus disease 2019  was identified in Wuhan, China in December 2019. The patient presented with atypical pneumonia by an unrecognized pathogen. The agent causing the unusual pneumonia was from the Coronaviridae family and was formerly named 2019 novel coronavirus (2019-nCoV) (WHO, 2020). Early epidemiological reports of the viral transmission involved the population who lived in or had visited Wuhan and it was suggestive of human-to-human transmission. The World Health Organization (WHO) declared COVID-19 as a pandemic on 11 March 2020 (WHO, 2020). In Malaysia, the first cases of COVID-19 were reported on 25 January 2020, among visitors from China who entered the country via Singapore. Since then, the number of COVID-19 cases have rapidly increased in numbers with appearance of new clusters leading to the implementation of the Movement Control Order (MCO) on 18 March 2020 (Reuters, 2020).
The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) virus is spread from human to human via respiratory droplets. According to the WHO, most of the patients infected by SARS-CoV-2 commonly presented with symptoms of fever, non-productive cough and fatigue. A research on common clinical manifestations of COVID-19 by Guan (2020) among thousands of COVID-19 patients in Wuhan showed that most (88.7%) developed fever during hospitalisation, more than half (67.8%) suffered from a cough, about 43.8% had fever on admission, and one-third (38.1%) had fatigue. The same study also found that 14.9% experienced myalgia, 13.9% sore throat, 0.8% conjunctivitis and 13.6% of patients experienced headache. Besides, less than 20% of the patients had shortness of breath, and only one-third (33.7%) of all patients were producing sputum. Other clinical manifestations included lymphopenia, radiographic findings suggestive of pneumonia, acute respiratory distress syndrome (ARDS) and acute respiratory failure in severe cases (Guan et al., 2020).
The screening of COVID-19 infection is essential and crucial to control the spread of the pandemic. Researchers in most screening centres across the world continue to study and understand COVID-19 disease. They use nasopharyngeal (NP) and oropharyngeal (OP) swabs as the methods of choice to detect SARS-CoV-2. Currently, real-time reverse transcriptasepolymerase chain reaction (rRT-PCR) of NP and OP swabs generally has been used to screen for SARS-CoV-2 infection. The specimens collected from different anatomical sites for rRT-PCR might affect the test performance thus yielding different results. According to previous studies, there was significant difference in the viral loads of different specimens depending on where and when the samples were collected in the respiratory tract.
This review was performed based on available original articles to analyse NP and OP swabs sampling in terms of viral load, rate of detection of positive cases and the sensitivity of the rRT-PCR assay in diagnosing COVID-19.

Search strategy
Relevant articles published in English from 2000 to 2021were identified using the search engines Scopus, PubMed, Medline, Embase, Web of Science, and Google Scholar from 17 August 2020 to 25 February 2021. We identified 25 articles from a total of 5221 articles related to the topic using the keywords COVID-19, SARS-CoV2, nasopharyngeal swab, oropharyngeal swab, nasal swab, throat swab and screening.

Eligibility criteria
Eligible studies included cohort, case-control studies, retrospective, cross-sectional and descriptive studies reporting the diagnostic accuracy of the NP and OP swab sampling in terms of the viral load by both methods, detection of positive cases and sensitivity rate in rRT-PCR assay in diagnosing COVID-19. We excluded articles with sample sizes less than five. Studies that did not include both NP and OP swab sampling analysis, published in non-English language or conducted on non-human samples were omitted.

Data extraction and bias assessment
Publications identified were screened for their title and abstracts according to the eligibility criteria, and further shortlisted for full-text screening. Out of 5221, only 25 articles that met the eligibility criteria were shortlisted and included after the full text screening. The findings of only 12 articles which included comparison of both NP and OP swabs are summarised in Table 1.

SARS-CoV-2 Load in Nasal and Throat Swab Specimens
Viral load plays a significant role in detecting SARS-CoV-2 infection. The trend of viral nucleic acid shedding in COVID-19 patients is different from SARS-CoV and MERS-CoV, which had minimal shedding in the early stages and peaked approximately 10 days after occurrence of symptoms (Peiris et al., 2003).
The recent study reported by Zou et al. among 18 patients infected by SARS-CoV-2 suggested that NP swab yielded a higher viral load than OP swab (Zou et al., 2020). This finding was supported the study by Wang W et al., where NP swab had a higher viral load than OP swab; the NP swab had mean cycle threshold (Ct) value of 24.3 (1.4 × 10 6 copies/mL) while mean Ct value of OP swab was more than 30 (<2.6 × 10 4 copies/mL). A cycle threshold value less than 40 was interpreted as positive for COVID-19 . Cycle threshold was inversely proportionate to the amount of target nucleic acid in the sample. This study showed that the lower the Ct value, the higher the viral load but contrary to this finding, the result of a study by Yu et al., (2020) showed that the OP swab had significantly higher viral load than NP swab (2552 copies/test [ SD=1965], p < 0.001) and (651 copies/test [SD= 501], p < 0.001) respectively. Wang H. et al. (2020) conducted a study on the difference in SARS-CoV-2 viral load in 120paired NP and OP specimens by rRT-PCR analysis. The Ct values were used to determine the SARS-CoV-2 RNA expression with lower Ct values corresponding to higher viral load . They found that NP swab had significant low Ct value compared to OP swab (p< 0.001) which suggested a higher viral load in NP swab samples. In among 57 positive COVID-19 patients, majority (91.2%) had lower Ct value of NP swab compared to OP swab which were 35.3% and 38.7%, respectively. Therefore, it indicates that the viral load in NP swab is significantly 10 times higher than OP swab. This finding was aligned with the result of studies conducted by Patel et al. (2021).
A study by Hernes et al. (2011) involving detection of other multiple respiratory viruses also found that NP swabs had lower Ct values than OP swabs (mean difference Ct 4.25, 95% CI [2.43, 6.07], p<0.001) indicating higher viral load in NP swab specimens which was 19 times higher (95% CI [5.4, 67.2] than OP swab specimens. Similar findings were reported by Peter et al. (2019).

Detection Rate and Sensitivity by Nasopharyngeal and Oropharyngeal Swabs Specimens in Diagnosing COVID-19
A descriptive study among 301 hospitalised COVID-19 patients showed different positive rates between NP and OP swabs specimens (Xiao et al., 2020). The researchers had performed 1113 tests using NP and OP swab specimens for SARS-CoV-2 detection by rRT-PCR analysis. Seventy-four (74) tests consisting of 37 pairs of both OP and NP swabs collected at the same time showed results of positive NP swab and negative OP swab in 12 pairs of tests (32.4%), but only two pairs of tests (5.4%) showed results of positive OP swab and negative NP swab. More than half (62.2%) pairs of tests had similar results of OP and NP swabs, of which four pairs had both positive results and 19 pairs had both negative results. They also did 46 swab samplings of both NP and OP at different points of time, where OP swabs were collected first, followed by NP swab. Unfortunately, the time frame between the OP and NP was not mentioned. The study found that 41.3% (19/46) of samples had negative OP swab followed by a positive NP swab which indicates high probability of false negative in a single collection of OP swab in the early stage of the illness. Therefore, it is estimated that about half (41.3%) of OP swab samples had false negative results of SARS-CoV-2 rRT-PCR assays compared to confirmed diagnosis of COVID-19 using NP swab. This study suggested that specimens for SARS-CoV-2 rRT-PCR assay should be obtained from NP swab as it was more sensitive and reliable for the assay compared to OP swab. Wang H. et al. (2020) conducted a study among 120 COVID-19 patients to compare the detection rate for both NP and OP swabs. They defined the detection rate as the percentage of positive results from the total specimens. The sampling was done between the 3rd and 49th day from the onset of symptoms, at a median of 27.0 days (IQR 23.0-31.5). Among these 120 COVID-19 patients, the detection rate of NP swabs was nearly half (46.7%) of the total cases whereas the detection rate of OP swabs was only 10.0% from the total of positive cases. It showed that NP swab had a significantly higher detection rate of SARS-CoV-2 than OP swab (p < 0.001). The NP swab also had a significantly higher detection rate (p< 0.001) after 21 days following the onset of symptoms .
Meanwhile, another study by Lin et al. (2020) which compared the detection rates of COVID-19 from sputum specimens and OP swab specimens showed that the positive detection rate was significantly higher in the sputum group compared to the throat swabs which were 76.9% and 44.2%, respectively (Lin et al., 2020). Similarly, Wang H et al. (2020) found that sputum had a higher positive rate, followed by NP swab and OP swab which were 72%, 63% and 32%, respectively.
In a study involving 213 COVID-19 patients, 866 samples from 61 respiratory tracts of the patients were obtained including NP swab and OP swab . The samples were obtained upon admission and at different points of time throughout the course of the disease. Sample collection dates were classified into three groups which were '0 to 7', '8 to 14' and '≥15' days after the onset of symptoms. The patients were also classified into 'severe' and 'mild' groups. A total of 205 OP swabs and 490 NP swabs were obtained. For the '0 to 7'days after symptom onset group, the positive rate of NP swabs (73.3% and 72.1% for severe and mild case, respectively) were higher than the OP swabs (60.0% and 61.3% for severe and mild case, respectively). In the group '8 to 14' days after symptom onset it was noted that the positive rate of 125 OP swabs was only 50% in severe and 29.6% in mild cases as compared to NP swab of 72.3% and 53.6% in the severe and mild groups, respectively. In the '≥15' after symptom onset group, it was revealed that NP swabs had a much higher positive rate than the OP swabs among the mild cases. This finding suggested that NP swabs are more reliable samples for virus detection. The OP swabs are not recommended for the SARS-CoV-2 virus screening, especially when the samples are obtained during 8 to14 days and more than 14 days after onset of symptoms in mild cases as a significant proportion of false negative outcomes may result from this. However, in the same study by Yang et al. (2020), most of the specimens were obtained after antiviral therapy, which may have influenced viral shedding. Furthermore, care should be taken as a single negative result does not indicate that the patient is not infected. The sensitivity of the test depends on whether the sample collected by NP and OP swab is sufficient as well as the technological sensitivity of the test itself which has various sensitivity measures .
In another study by Patel et al. (2021), they reported only a slightly higher sensitivity for NP swabs in pairs of specimens collected less than seven days after symptoms onset compared to OP swabs.
However, OP swab sensitivity was comparatively low in swab pairs collected after seven days of illness onset (median, 12 days; IQR, 9-19); while 14 (23.7%) NP swabs tested positive, and10 (17.0%) OP swabs tested positive (p = .045). Wang H et al. (2020) analysed the diagnostic sensitivity by using NP and OP swab specimens among 57 positive COVID-19 patients. They defined the sensitivity as the percentage of true positives correctly identified by both methods. The results showed that the sensitivity of NP swab (98.3%) specimen was significantly higher than OP swab (21.1%) specimen. In order to explain what medical conditions could influence the disparity in sensitivity, they categorized patients based on clinical features and laboratory values. In all medical conditions, except for febrile patients, NP swab sensitivity was substantially higher than OP swab sensitivity (p < 0.05). Among the seven febrile patients, there was no significant difference in sensitivity between NP and OP swabs, which can be explained by the limited sample size. This discovery showed that the NP swab was more reliable diagnostically than the OP swab.
A similar observation was reported by Bwire et al. (2021) in which the positive detection rate for NP swab and OP swab were 45.5% and 7.6%, respectively. A study by Zhang et al. (2020) found that the positive rate in NP swabs (37.5%) was higher compared to OP swabs (20.8%). However, the result was not significant, likely due to the small sample size. In contrast, a study by Calame et al. (2021) showed there was significant correlation between OP swabs and NP swabs specimen with regards to the analytical sensitivity at the quantitative level.

Discussion
Most of the reviewed articles related to the detection of SARS-CoV-2 infection showed that nasal swab had a significantly higher viral load than throat swab (Zou et al., 2020;Wang H. et al., 2020 andPatel et al., 2020). It was found that SARS-CoV-2 load was reported to be 10 times significantly higher for NP swab than OP swab with Ct value ranging from '24.3 to 37.8' and '29.9 to 39.4', respectively. The difference of Ct value by both methods was 3.3 (95% CI [2.2, 4.5], p < 0.001) which indicates that NP swab had higher viral load compared to OP swab. The Ct is inversely proportionate to the amount of target nucleic acid in the sample which means the lower the Ct, the higher the viral load. In contrast, we found only one study which reported Bwire, et al. 2020 University that OP swabs had significantly higher viral load than NP swabs . It can be theorised that higher viral load in NP specimens is due to higher amount of SARS-CoV-2 viral replication in the nasopharynx than in the oropharynx after infection sets in. Secondly, it might be due to larger contact surface when collecting the specimens in the nasopharynx that can allow more virus collection.
Although it has been demonstrated that sputum has higher viral load and detection rate as compared to NP and OP swabs, Zhang et al., 2020) the procedure to obtain samples from lower respiratory tract will induce aerosol production and expose the healthcare staff to a high risk of viral transmission. Therefore, upper respiratory tract specimen is preferable in practice. Since the viral load increases with time, early phase of infection may be missed by NP and OP swab tests; thus requiring a repeat test . If we have to assess a population that is asymptomatic by swabbing, we should expect that the false negative rate will increase proportionally. The kinetics of viral RNA should be considered because in the early stages of SARS-CoV-2 infection, the viral load is very low thus yielding a negative result . As reported by Xiao et al. (2020), there is an estimate of 41.3% false negative rate of OP swabs. In fact, about 30% of clinically symptomatic patients' swabs give a false negative result (Loeffelholz et al., 2020).
Due to its ease of acquisition, NP swab has been preferred and used for diagnosis and dynamic observation of COVID-19 patients. Two successive negative SARS-CoV-2 RNA detections of NP swab specimens have been recognised as a prerequisite for hospital discharge or quarantine release. Nonetheless, the risk of false negative findings is a drawback of the NP swab, raising the question that the persistence of viral shedding could be present in the lower respiratory tract (Winichakoon et al., 2020). A study by  of 68 COVID-19 patients who underwent both NP swab and sputum test reported that 20.6% (n=14) patients who initially experienced negative NP swab samples subsequently registered positive sputum specimen during follow-up. They found that the median duration of viral shedding from sputum specimens was significantly longer compared to NP swab specimen (p<0.001).
Meanwhile, a study on COVID-19 survivors by Zhou et al. (2020) reported that there was a median of 20 days and a mean of 37 days of viral shredding in the OP swab cases. However, in cases screened by the NP swab, viral shredding had a longer median period of 25 days and a longer mean duration of 41 days. These results suggested that after the onset of symptoms, the NP swab could detect SARS-CoV-2 for a longer duration.
NP and OP swabs are the recommended specimen types for COVID-19 diagnostic testing. Nevertheless, it may cause complications such as pain and can cause bleeding, especially in patients with low platelet counts (Chan et al., 2020). Other reported complications by the patients from NP and OP swabs were teary eyes, rhinorrhoea and emesis (Chan et al., 2020). This review has some limitations. First, some studies did not define specifically the exact anatomical site for "nasal" and "pharyngeal' swab. The nasal swabs can be taken from anterior nares, mid turbinate or nasopharynx. Secondly, some studies did not provide details of clinical severity or grading, timing of sampling, very wide range of patient illness, therefore, the viral load detection rate and sensitivity may vary and unable to be correlated with disease course. Third, some studies did not pair the NP and OP swab specimens; hence the correlation may be less accurate. Finally, the sample size and number of positive results were small in majority of the studies, hence it may affect the precision of the estimate of the test sensitivity. This review has provided good supporting evidence of the results of the comparisons between the two sampling techniques that are commonly used in large COVID-19 screening centres.

Conclusion
All currently available testing methods including NP and OP swab have their rates of sensitivity and specificity in diagnosing COVID-19. In our review, we can conclude that NP swab had a higher detection rate of positive SARS-CoV-2 viral load and higher sensitivity in RT-PCR assay compared to OP swab for diagnosis of COVID-19. Nevertheless, there is a possibility that both NP and OP swabs may yield false negative result thus investigators should be concerned of patients with clinical suspicion with negative NP and OP swab results. Clinicians should take note that NP swabs have longer duration of viral shredding while OP swabs contribute to higher percentage of false negative results.
Based on this review, NP swab is the best sampling technique to be practised as it is more reliable than OP swab in detecting SARS-CoV-2 infection. A single NP or OP swab sample during the early onset of the disease is not recommended as it has a probability of false negative result which can lead to misdiagnosis and increase the rate of SARS-CoV-2 transmission. For discharge criterion, although NP swab had a longer viral shredding than OP swab, it is recommended to take sputum sample specimen as it has the longest duration of viral shredding than both methods. Nevertheless, studies with larger sample sizes are required to effectively develop and establish a more specific diagnostic approach to improve detection of COVID-19 infection.