Journal Club Review: “Detection of SARS-CoV-2 in Different Types of Clinical Specimens.”
The goal of COVID Reviews is to provide a series of reviews related to recent analytics-focused articles on COVID-19. Our hope is to provide short summaries and critiques in a journal club-oriented format that can be quickly digested to assess the methodologies used, populations/outcomes assessed, impact of work, and strengths/weaknesses of each article.
Article
Review by: David Ferguson MT(ASCP); Wade Schulz, MD, PhD
Authors’ Aim
The author’s aim was to explore the distribution of SARS-CoV-2 RNA in various clinical samples in patient’s who were diagnosed as having COVID-19 infection in 3 Chinese hospitals in Hubei and Shandong provinces andBeijing, China, from January 1 through February 17, 2020. The authors’ aim was to determine the specimen types that provide the best diagnostic accuracy to identify patients infected with COVID-19 and which sites may harbor virus.
Why Article Was Selected
The article was selected as it is one of the earliest discussions of COVID-19 testing accuracy related to COVID-19. The numbers from this peer-reviewed article are often cited, despite the low number of patients and samples in many categories. The topic is crucial to provide insights into test quality and sample selection for COVID-19 testing, and also provides information as to possible physiologic sites of infection. As highly accurate diagnostic tests are critical to controlling the spread of COVID-19 disease both in hospitals and communities, knowing which tests and specimens provide the best information regarding infection is crucial to guide appropriate testing and response.
Methods Employed by Authors
The authors correlated sample type (eg, nasal swab, throat swab) with laboratory results of Real-Time Reverse Transcriptase Polymerase Chain Reaction (rRT-PCR), or RNA-based, testing. RNA was extracted from clinical samples from patients with multiple sample types (nasal swabs, throat swabs, stool, blood, BAL, and fibrobronchoscope brush biopsy) and compared positive PCR findings and cycles to threshold (Ct) values in an attempt to determine the viral load in the various sample types.
Results and Conclusions
There were 1070 specimens collected from 205 patients with COVID-19 who were a mean age of 44 years (range: 5–67 years). The authors found a range of sensitivities in these tests that varied by specimen type, ranging from high sensitivity in bronchoalveolar lavage (93%) to no detection in urine (0%). The most commonly used swabs for testing showed variable sensitivity, with nasal swabs positive in only 63% of cases (n=8) and pharyngeal swabs present in only 32% of cases (n=398). Interestingly, Ct values were lowest in the nasal swabs, indicating likely higher viral titers, despite the lower number of positive samples.
Strengths of Article
The aim of the article is very timely and of great utility to the diagnostic virology community in general. It covered the main sample types that are typically used to diagnose viral respiratory disease and in that respect was comprehensive. The use of CT values as a metric to approximate viral load was also very useful. Some data sets had sufficient numbers and clear results to be evaluated. For instance, viremia was rare (3/307) viruria was non-existent (0/72). So a conclusion that testing either of these source material is of little clinical utility can be drawn.
Limitations of Article
There are several major limitations of this article that require further review and investigation:
- The method used for gold standard classification to confirm SARS-CoV-2 infection is not listed — for example, radiographs plus positive from any site?
- The results state 5 of 8 nasal swabs were positive, but in Figure 1, which shows patients with multiple simultaneous specimens, 6 nasal swabs are shown to be positive in just a subset of patients based on Ct cutoffs.
- The samples collected were random and not systematic so it is difficult to ascertain truly what samples will provide the most diagnostic utility. Saliva was not collected as one of the specimen types.
- Sample size was very small for some significant sample types (i.e. nasal swabs) and thus were of questionable significance. Similarly, only 20 patients had specimens from multiple sites collected at the same time.
- No association with clinical presentation and the source of the positive samples. It is difficult to associate the patient’s clinical status with sample positivity or CT value.
- The assumption of “live virus” in stool samples and thus infectivity via this source was not substantiated by methods or other findings and seems to be tenuous at best.
Take-Home Message
The subject is of great significance and utility to the world of diagnostic virology. This study points out that several physiologic sites will likely yield less sensitivity and that non-respiratory sites are of questionable clinical use (although saliva was not tested). While very important initial information, more structured evaluation with specified collection times and types, with clear definition of the positive cohort are needed. Additionally, data that correlates patients with varying degree of disease process and viral burden may be useful in monitoring and predicting outcomes.
