ISSN: 1899-0967
Polish Journal of Radiology
Established by prof. Zygmunt Grudziński in 1926 Sun
Current issue Archive Manuscripts accepted About the journal Editorial board Abstracting and indexing Contact Instructions for authors Ethical standards and procedures
Editorial System
Submit your Manuscript
SCImago Journal & Country Rank
1/2022
vol. 87
 
Share:
Share:
Chest radiology
Review paper

Update on the limited sensitivity of computed tomography relative to RT-PCR for COVID-19: a systematic review

Clarissa Martin
1
,
Nina Cheng
2
,
Bryant Chang
2
,
Namrata Arya
3
,
Michael Joseph Diaz
4
,
Keldon Lin
3
,
Muhammad Umair
5
,
Joseph Waller
2
,
Travis Henry
6

1.
University of Pennsylvania, Philadelphia, PA 19104, USA
2.
Drexel University College of Medicine, Philadelphia, PA 19129, USA
3.
Mayo Clinic Alix School of Medicine, Scottsdale, AZ 85259, USA
4.
University of Florida College of Medicine, Gainesville, FL 32610, USA
5.
Johns Hopkins Department of Radiology & Radiological Sciences, Baltimore, MD 21205, USA
6.
Department of Radiology, Duke University School of Medicine, USA
© Pol J Radiol 2022; 87: e381-e391
Online publish date: 2022/07/12
Article file
- Update.pdf  [0.24 MB]
Get citation
 
PlumX metrics:
 

Introduction

Epidemiology, aetiology, and management

The SARS-CoV-2 virus is a positive-sense single-stranded RNA virus of the coronavirus (CoV) family. CoV viruses can cause respiratory, enteric, hepatic, and neurologic diseases in humans, and their ability to cross species barriers is thought to be the main mechanism of transmission that occurred in Wuhan. Whereas CoVs have been known to cause illness including common colds and self-limiting upper respiratory infections in immunocompetent individuals, lower respiratory tract infections may also occur in the elderly and the immunocompromised [1]. Human-to-human transmission most likely occurs via respiratory droplets but may also occur through aerosolization. The incubation time has been shown to range from 3 to 14 days [1]. Great efforts have been made in managing COVID-19-induced pneumonia, aberrant coagulation, and the ‘cytokine storm’ that leads to lung, nervous, gastrointestinal, cardiovascular, and renal tissue damage [2]. Current management consists of supportive care, such as oxygen support, fluid therapy, and symptom relief.

Diagnostic tools

RT-PCR testing, serologic testing to detect antibodies against SARS-CoV-2, and non-contrast chest CT scans are some of the tools used to aid and make accurate and rapid diagnoses of COVID-19 [3]. In a previous review, we suggested that the true sensitivity of CT was overestimated, and in the clinical setting CT is better utilized as a supplementary diagnostic tool. Here we revisit our previous analysis with more current data to determine whether this same conclusion holds [4]. The previous analysis was of great interest and is listed among the top 10 most cited articles published by Investigative Radiology in 2020 [5].

This review aims to stratify studies investigating the sensitivities of RT-PCR and CT for COVID-19 based on bias, to see how biases affect the sensitivities reported. This is not a review comparing the diagnostic accuracy of RT-PCR with CT, because one cannot compare the sensitivity of 2 diagnostic tests when 1 of them (RT-PCR) serves as the reference standard for the other (CT). Thus, we refrain from making definitive statements comparing the sensitivities of CT with RT-PCR; we describe the sensitivity of CT as “limited” relative to RT-PCR, for example, as opposed to “less than”, because that kind of statement cannot be made until there are sufficient studies on the sensitivity of CT that use a different, more accurate reference standard such as repeat RT-PCR testing (see Methods for more information).

Material and methods

Data sources and searches

The PubMed database was queried between 1 January 2020, and 25 April 2021 using the Medical Subject Headings search terms (sensitivity and specificity, AND RT-PCR, AND coronavirus, AND SARS-CoV-2) OR the presence of keywords (CT AND COVID-19 OR severe acute respiratory syndrome coronavirus 2) in the title, abstract, or full-text publications (n = 670). Google Scholar dataset was also searched, using the keywords “COVID-19” AND “SARS-CoV-2” AND “sensitivity” AND “coronavirus” AND “RT-PCR” AND “chest CT” AND “imaging” anywhere in the text articles published between 1 January 2020, and 26 May 2021 (n = 1050). In total 980 results were identified, of which 62 met the inclusion criteria for this study. The studies were evaluated by the authors J.W. and B.C., and discrepancies were resolved by the senior author M.H.

Study selection and quality assessment

The study protocol was registered with PROSPERO, and QUADAS-2, a quality assessment tool for studies of diagnostic accuracy, was used to stratify papers from high to low risk of bias. QUADAS-2 assesses the risk of bias based on 4 key factors:

  1. Patient selection. Patient cohorts that included both symptomatic and asymptomatic patients were considered as low risk of bias. Inappropriate exclusion of symptomatic or asymptomatic patients would not reflect the general SARS-CoV-2-infected population and would introduce bias. Studies conducted on paediatric patients raise applicability concerns, and the authors chose to thus classify them as high risk of bias because adult patients were excluded.

  2. Index test. Studies in which positive chest CT findings were interpreted as indicative of COVID-19 infection without RT-PCR confirmation were categorized as high risk for bias. To be considered low risk, studies were required to include patient cohorts whose diagnoses were confirmed with RT-PCR (in accordance with the American College of Radiology’s [ACR] recommendation that RT-PCR should serve as the primary diagnostic tool for COVID-19). Studies in which patients with positive CT and negative RT-PCR tests were considered COVID-19-positive were considered as high risk of bias. These studies were not excluded in this review because they still included a clinically relevant cohort of patients whose COVID-19 status was unclear. Finally, the timing of when CT was done is relevant and is study dependent.

  3. Reference standard. Per QUADAS-2, the method by which the reference standard is conducted may introduce bias. For RT-PCR, the reference standard of this study, a proper nasal swab must have been taken from the upper or lower respiratory tract (in compliance with ACR guidelines). Because RT-PCR cannot serve as both the reference standard and comparator test against CT, this study aims instead to determine the sensitivity of CT alone as a potential diagnostic tool, rather than direct comparison to RT-PCR. Finally, RT-PCR and CT accuracy depend on their temporal usage within the disease course, and may therefore have differing clinical utility in various clinical settings.

  4. Flow and timing. Studies that did not include all patients in the analysis were considered high risk for bias. Studies in which all patients did not undergo RT-PCR testing in the same manner (e.g. swabs not taken from the same location) were considered as high risk for bias.

Studies that did not provide information necessary for bias stratification were excluded as well as studies not published or available in English. Additional exclusion criteria included not specifying whether patients were symptomatic or asymptomatic, not indicating whether patients were adults or children, not indicating the presence of RT-PCR-confirmed SARS-CoV-2 cases within the cohort, and not including information regarding the source of the specimen used in RT-PCR. Preprints were not included in this analysis because they were not peer reviewed.

Reference standard

We refer to RT-PCR as the reference standard, as opposed to a “gold standard”, because the “gold standard” is a flawed concept given that it may be perceived as implying the test is perfect, which it is not. In fact, some may debate whether RT-PCR is even accurate enough at discriminating between COVID-19 and non-COVID-19 to call it the reference standard, given the technical challenges of sample acquisition affecting its sensitivity. Thus, without a very accurate reference test, approaches are needed like a composite reference standard (combination of several tests) or repeat RT-PCR testing.

Repeat RT-PCR testing was used as the reference standard when evaluating the sensitivity of RT-PCR (to compare the sensitivity of RT-PCR according to high-risk-of-bias studies with that of low-risk-of-bias studies), but given a lack of studies using this reference standard with CT, we employed single-use RT-PCR as the reference standard when evaluating the sensitivity of CT (and thus we refrain from making definitive statements comparing the sensitivity values reported for CT with that of RT-PCR, because one cannot compare the reference standard with the comparator test).

For CT studies, TP is defined as those in which the patient received a positive CT scan as well as a positive RT-PCR result, and FN is when the patient received a negative CT scan but a positive RT-PCR result. For RT-PCR studies, a TP is when the first RT-PCR result was positive as well as subsequent serial RT-PCR results, while a FN is one in which the first result was negative but subsequent serial RT-PCR results were positive.

Data synthesis and analysis

Compared to the inexpensive and high-throughput of RT-PCR, CT has demonstrated a specificity close to 80% for COVID-19, as well as comparatively higher cost and disparate accessibility around the world [6]. Nevertheless, CT has been proposed as a primary diagnostic test primarily due to its perceived superior sensitivity. Thus, specific focus was given to extracting sensitivity data rather than data concerning specificity, true positives, false negatives, etc. In addition, the specificity of CT for COVID-19 is well accepted, while the sensitivity is less clear because many studies report greatly varying sensitivity values, as this review will demonstrate.

Data regarding positive findings from CT and/or RT-PCR were extracted from all eligible studies. Duplicate extraction of data was performed by 2 authors. Random effects meta-analyses were performed, and the same random effects models were used to control for inter- and intra-study variability. Calculations were made in Microsoft Excel. Individual papers were sorted as biased (i.e. high risk of bias) and unbiased (i.e. low risk of bias). Heterogeneity between studies was evaluated by calculating I2, and forest plots based on the random effect models were created to demonstrate the sensitivity findings of each study for both CT and RT-PCR.

Results

Search results

A total of 1720 search results were identified, 740 of which were duplicates. Of nonduplicate results, 590 did not meet inclusion criteria, and 58 were excluded after further eligibility review. Hence, 62 papers were ultimately included for analysis. The number of articles excluded for not meeting a specific exclusion criterion are as follows: 310 for not having sufficient patient information to satisfy the QUADAS-2 patient selection criteria, 71 for index test, 144 for reference standard, and 65 for flow and timing (for a total of 590 papers excluded). The exclusion criteria used for this study are found in Figure 1. Review studies that summarized information already extracted from papers that met the inclusion criteria were also excluded.

Figure 1

Flow diagram of the study

/f/fulltexts/PJR/47518/PJR-87-47518-g001_min.jpg

RT-PCR as a reference standard to detect SARS-CoV-2 infection

Methodological biases

Despite RT-PCR’s current status as the putative diagnostic test for COVID-19, many studies investigating its sensitivity and specificity are undermined by methodological biases. For example, while Li et al. reported RT-PCR sensitivity for SARS-CoV-2 detection as 27.5% (n = 610), patients were assumed to be COVID-19 positive based on chest CT indications of viral pneumonia [7]. Because not all patients presenting with CT findings of viral pneumonia have COVID-19, this assumption compromises the accuracy of RT-PCR sensitivity. Similar limitations were identified in studies whose cohorts were not conclusively confirmed as COVID-19 positive, yielding inconsistent results regarding RT-PCR diagnostic performance. Wu et al. reported RT-PCR sensitivity among suspected but unconfirmed COVID-19 patients as 51% (n = 80), while Liu et al. found RT-PCR sensitivity for SARS-CoV-2 infection to be 38.25% (n = 4880) [8,9].

In a study of exclusively symptomatic patients, Fang et al. calculated RT-PCR sensitivity for COVID-19 detection as 71% (n = 51, p < 0.001) [10]. Exclusively symptoma-tic patient cohorts were similarly utilized in numerous studies (Table 1), and a major limitation of this design is the inability to generalize results to asymptomatic COVID-19-positive individuals [10-17]. Alternatively, the study cohorts recruited by Sun et al. and Clerici et al. were limited to cancer patients and clinically recovered patients, respectively [18,19]. Ultimately, the study of suspected but unconfirmed COVID-19 patients may underestimate true RT-PCR sensitivity for SARS-CoV-2 infection. Finally, in a study of patients having undergone 2 consecutive SARS-CoV-2 RT-PCR tests, Xiao et al. reported that 78.6% (n = 70) yielded at least 1 positive result, but they did not elaborate regarding the diagnostic sensitivity following the first RT-PCR test [20]. Because the sensitivity of the initial RT-PCR alone is likely to be lower than that of 2 consecutive tests, a definitive conclusion concerning overall RT-PCR diagnostic sensitivity cannot be made from the Xiao study.

Table 1

Sensitivities of initial reverse transcriptase polymerase chain reaction (RT-PCR) for diagnosing COVID-19 infection in high-risk-of-bias studies

Biased studyNo. of patientsPositive result, n (%)Study limitations
R. Liu et al.48801854 (38)Determined patients had COVID-19 based on typical symptoms or contact with COVID-19
J. Wu et al.8041 (51)Patient cohort was not COVID-19 confirmed
Y. Fang et al.5136 (71)No asymptomatic patients (all patients had fever or acute respiratory symptoms)
C. Long et al.3630 (83)No asymptomatic patients (only included patients with fever > 38°C and COVID-19 pneumonia suspicion)
A.T. Xiao et al.7055 (79)These data were for 2 consecutive RT-PCR tests
T. Xu et al.5149 (96)No asymptomatic patients
R. Sun et al.3524 (69)Clinically positive cases were determined by symptoms and chest CT
B. Clerici et al.393300 (77)Only clinically recovered patients
N. Hanif et al.7835 (45)No asymptomatic patients
C. Thomas et al.8479 (94)No asymptomatic patients
S. Schalekamp et al.536497 (93)No asymptomatic patients
N. Sverzellati et al.248190 (77)No asymptomatic patients
V.R. Bollineni et al.5136 (71)Only patients with respiratory distress presenting to ED
D. Chen et al.2114 (67)No asymptomatic patients
Z. Wen et al.8837 (42)RT-PCR collected from different tissues

[i] The study limitations are aspects of the study’s methods that prevent generalizing the reported sensitivity to the broader SARS-CoV-2-infected population.

[ii] RT-PCR – real-time polymerase chain reaction, SARS-CoV-2 – severe acute respiratory syndrome coronavirus-2.

RT-PCR: What do low-risk-of-bias studies tell us?

This analysis only included studies that collected RT-PCR samples from individual patients, excluding studies involving pooled clinical RT-PCR samples. All studies indicated as low risk for bias (Table 2) evaluated initial RT-PCR sensitivity via repeat testing of the entire patient cohort. Nonetheless, RT-PCR sensitivity findings varied from 70% to 97% for SARS-CoV-2 infection [7,21]. He et al. reported the sensitivity of initial RT-PCR detection of SARS-CoV-2 as 79% (n = 34) [22]. Meanwhile, a national survey of 26 French hospitals by Herpe et al. found the RT-PCR sensitivity to be 87% (n = 2225) [23]. Finally, Ducray et al. reported that 278 of 287 COVID-19- confirmed patients in their cohort tested positive on their initial RT-PCR test, yielding a sensitivity of 97% [21]. We speculate that factors such as variations in disease severity and the time of RT-PCR testing relative to disease course may account for the heterogeneity of these results, as summarized in Figure 2. Finally, this review refrains from commenting on pooled RT-PCR sensitivity because such studies were excluded.

Figure 2

Forest plot of reverse transcriptase polymerase chain reaction (RT-PCR) studies showing the sensitivity of each study using a random effects model to control for heterogeneity and showing subgroups by bias in the studies

/f/fulltexts/PJR/47518/PJR-87-47518-g002_min.jpg
Table 2

Sensitivities of Iinitial reverse transcriptase polymerase chain reaction (RT-PCR) for diagnosing COVID-19 infection in low-risk-of-bias studies

Low-risk-of-bias studyNo. of patientsPositive result, n (%)Main topic of study
Y. Li et al.241169 (70)*RT-PCR testing of hospitalized SARS-CoV-2 patients
W. Wang et al.12791 (72)Investigation of different types of RT-PCR specimens
A. Bernheim et al.10290 (88)Serial chest CT findings of symptomatic COVID-19 patients
H.Y.F. Wong et al.6458 (91)Correlation of chest CT findings with RT-PCR tests for COVID-19 patients
J.L. He et al.3427 (79)Comparison of CT and initial RT-PCR in diagnosing COVID-19
G. Herpe et al.25642225 (87)Efficacy of chest CT for COVID-19
V. Ducray et al.287278 (97)Chest CT for triage of COVID-19 patients

RT-PCR – real-time polymerase chain reaction, SARS-CoV-2 – severe acute respiratory syndrome coronavirus-2

* Although this study was originally classified as biased for assuming that patients with pneumonia have COVID-19, we were able to correct for this by only using the 241 patients who were eventually confirmed positive on RT-PCR in our calculation.

Sensitivity of chest CT for SARS-CoV-2 infection

Methodological biases

Methodologically biased patient cohorts contributed to a lack of generalizability in several studies (Table 3). For instance, although Ai et al. reported a 97% (n = 601; 95% CI: 95-98%) chest CT sensitivity for SARS-CoV-2 detection using RT-PCR as a reference standard, this may be an overestimation because the cohort consisted of symptomatic pneumonia-presenting patients [24]. Guan et al. reported a lower sensitivity of chest CT at 82.1% (n = 877), but because this patient cohort included COVID-19 patients with severe adverse outcomes (i.e. ICU admission, need for mechanical ventilation, death), these conclusions should also be interpreted with caution [25]. As such, it is clear that the reported sensitivity in both studies mentioned above may overestimate the true sensitivity of chest CT.

Table 3

Sensitivities of initial chest CT for diagnosing COVID-19 infection in high-risk-of-bias studies

Biased studyNo. of patientsPositive results, n (%)Study limitations
D. Wang D et al.3014 (47)No adult patients.
X. Lu et al.170111 (65)No adult patients.
F. Zheng et al.2416 (67)No adult patients.
Y. Wang et al.5537 (67)All asymptomatic patients.
Z. Hu et al.2417 (71)All asymptomatic patients.
W.J. Guan et al.877720 (82)Only patients who were admitted to an ICU, used a ventilator, or died were included.
G. Huang et al.3026 (87)Implied all patients were symptomatic (grouped by time of symptom onset).
Z. Chen et al.9891 (93)No asymptomatic patients.
W. Zhu et al.3230 (94)No asymptomatic patients.
K. Wang et al.114110 (96)No asymptomatic patients.
T. Ai et al.601583 (97)Used a cohort of patients with pneumonia.
D. Caruso et al.6260 (97)No asymptomatic patients (only included patients with respiratory symptoms).
C. Long et al.3635 (97)No asymptomatic patients (patients all had fever >38°C and COVID-19 pneumonia suspicion).
Y. Fang et al.5150 (98)No asymptomatic patients (all patients had fever or acute respiratory symptoms).
J. Chen et al.249243 (98)No asymptomatic patients.
X. Xu et al.6261 (98)No asymptomatic patients (patients with nonspecific respiratory symptoms were included).
J.J. Zhang et al.135134 (99)No asymptomatic patients.
T. Xu et al.5151 (100)No asymptomatic patients.
Z. Zhou et al.6262 (100)No asymptomatic patients.
X. Zhao et al.8080 (100)No asymptomatic patients.
H. Shi et al.8181 (100)No asymptomatic patients.
R. Han et al.108108 (100)No asymptomatic patients (involved mild patients but they all have COVID-19 associated pneumonia).
D. Wang et al.138138 (100)No asymptomatic patients.
C. Wu et al.201201 (100)No asymptomatic patients.
Ravikanth R.481470 (98) *No asymptomatic patients *(calculation corrected)
H.A. Gietema et al.8374 (89)No asymptomatic patients
R. Sun et al.2218 (82)Clinically positive cases were determined by symptoms and chest CT
V.R. Bollineni et al.144144 (100)Only patients with respiratory distress presenting to ED
A. Orlacchio et al.344313 (91)No asymptomatic patients
N. Sverzellati et al.190156 (82)No asymptomatic patients
S. Schalekamp et al.536493 (92)No asymptomatic patients
C. Thomas et al.8070 (88)No asymptomatic patients
N. Hanif et al.3835 (92)No asymptomatic patients
Z. Wen et al.8882 (93)No asymptomatic patients

[i] The study limitations are aspects of the study’s methods that prevent generalizing the reported sensitivity to the broader SARS-CoV-2-infected population.

[ii] CT – computed tomography, ICU – intensive care unit, SARS-CoV-2 – severe acute respiratory syndrome coronavirus-2

The most prevalent limitation within CT sensitivity studies was the lack of inclusion of asymptomatic COVID-19 patients [10-13,15,16,26-43]. Because asymptomatic patients account for a large subset of SARS-CoV-2-positiveindividuals, their inadequate representation in COVID-19-related research undermines the applicability of findings to the entire COVID-19 patient population. In a study whose cohort consisted solely of patients displaying respiratory symptoms, chest CT sensitivity was observed as 97% (n = 62) [30]. This finding was replicated in a smaller study of 36 patients, all of whom presented with fever alongside pneumonia [11]. In fact, several studies that reported near-perfect chest CT sensitivities (98-100%) utilized exclusively symptomatic patient cohorts [10,32,37]. Meanwhile, numerous studies citing markedly lower chest CT sensitivities (47-67% and 67-71%) utilized patient cohorts consisting of altogether paediatric and asymptomatic patients, respectively [38,44-47]. Inui et al., comparing the diagnostic sensitivity of CT between symptomatic and asymptomatic COVID-19 patients aboard the Diamond Princess cruise ship, found that CT exhibited significantly higher sensitivity in detecting SARS-CoV-2 infection in symptomatic patients (79%, n = 28) over their asymptomatic counterparts (54%, n = 76, p = 0.023) [48]. These differential results imply that inclusion of COVID-19 patients of varying disease severity may be essential to an accurate overall assessment of chest CT sensitivity and diagnostic utility in SARS-CoV-2 detection.

CT: What do low-risk-of-bias studies tell us?

Studies considered low risk for bias demonstrated slightly lower sensitivity findings for chest CT in SARS-CoV-2 detection. Among low-risk-of-bias studies, CT sensitivity for COVID-19 detection ranged from 44% to 92% (Table 4) [49,50]. Kassem et al. found chest CT to be significantly more sensitive for SARS-CoV-2 detection at more advanced diseases stages, reporting sensitivities of 50% (n = 50) and 100% (n = 53) for early and progressive COVID-19 stages, respectively, with an overall sensitivity of 76% (n = 103) [51]. Interestingly, studies with larger sample sizes reported higher sensitivity findings for chest CT [21,23,50,52-55]. For example, Falaschi et al. reported chest CT sensitivity for COVID-19 infection as 91% (n = 462), with significantly lower sensitivity in patients under 50 years old [53]. Similarly, Herpe et al. found that CT sensitivity for SARS-CoV-2 detection was 90% (n = 2564), with no significant differences across geographic regions of varying disease prevalence [23]. Likewise, Hermans et al. shared similar results in a prospective cohort study of Dutch patients, reporting chest CT sensitivity to be 90% (n = 133) [52]. Overall, our random effects meta-analyses report the mean sensitivity of chest CT as 91% (n = 5377 patients/34 studies; 95% CI: 87-96) in high-risk-of-bias studies and 78% (n = 4568/17 studies; 95% CI: 71-86) in low-risk-of-bias studies (Figure 3).

Figure 3

Forest plot of computed tomography (CT) studies showing the sensitivity of each study using a random effects model to control for heterogeneity and showing subgroups by bias in the studies.

/f/fulltexts/PJR/47518/PJR-87-47518-g003_min.jpg
Table 4

Sensitivities of initial chest CT for diagnosing COVID-19 infection in low-risk-of-bias studies

Low-risk-of-bias studyNo. of patientsPositive results, n (%)Main topic of study
A. Bernheim et al.3616 (44)Serial chest CT findings of symptomatic COVID-19 patients
H. Qiu et al.3619 (53)Clinical presentation of paediatric COVID-19 patients
S. Inui et al.10463 (60)CT findings of Diamond Princess COVID-19 patients
J. Wu et al.8055 (69)CT and laboratory findings of imported COVID-19 patients
K. Li et al.7856 (72)Comparison of CT imaging and COVID-19 clinical features
Y.H. Xu et al.5041 (82)CT findings of COVID-19 patients
Z. Ling et al.295246 (83)CT findings in asymptomatic SARS-CoV-2 patients
W. Yang et al.149132 (89)CT imaging and clinical findings in COVID-19 patients
H.Y.F. Wong et al.2825 (89)Correlation of chest CT findings with RT-PCR tests for COVID-19 patients
W. Zhao et al.10193 (92)Correlation between COVID-19 CT imaging findings and clinical features
Z. Falaschi et al.462419 (91)Chest CT performance in diagnosing COVID-19
J.L. He et al.3426 (77)Comparison of CT and initial RT-PCR in diagnosing COVID-19
M.N.E. Kassem et al.10378 (76)CT clinical findings in COVID-19
J.J.R. Hermans et al.133120(90)Comparison of chest CT with initial RT-PCR in COVID-19
A. Mirahmadizadeh et al.2819 (68)CT sensitivity and specificity for COVID-19
G. Herpe et al.25642319 (90)Efficacy of chest CT for COVID-19
V. Ducray et al.287259 (90)Chest CT for triage of COVID-19 patients

[i] CT – computed tomography, SARS-CoV-2 – severe acute respiratory syndrome coronavirus-2

Discussion

Whereas many factors must be considered in the overall evaluation of a test’s clinical utility, such as specificity and cost, the present study focuses on diagnostic sensitivity, a perceived strength of chest CT for the disease process studied. Our analyses identify several sources of bias conducive to overestimation of the true sensitivity of chest CT in SARS-CoV-2 detection. Among studies deemed high risk for methodological bias, many recruited exclusively symptomatic cohorts. Such approaches lend insight into chest CT sensitivity for SARS-CoV-2 infection in symptomatic individuals but fail to account for its diagnostic utility in asymptomatic patients. Consequently, conclusions regarding the sensitivity of CT from such studies cannot be reliably generalized to the entire COVID-19 patient population. Likewise, chest CT sensitivity is overestimated when patients presenting with pneumonia are at once presumed to be positive for COVID-19 [24]. Overall, these assumptions ultimately limit the generalizability of positive findings. RT-PCR sensitivity is also underestimated in studies where patients are assumed to be positive for SARS-CoV-2 infection simply on the basis of symptomatic presentation or knowledge of having been in contact with an individual confirmed for SARS-CoV-2 infection [9]. As discussed, it must be considered that not all COVID-19-positive individuals display symptoms and, conversely, not all patients presenting with COVID-19-like symptoms are true-positives for SARS-CoV-2 infection.

While RT-PCR has been instrumental in facilitating COVID-19 detection and diagnosis, a few challenges remain for its use in large-scale detection. Amidst the urgency of the pandemic, molecular diagnostic tests for SARS-CoV-2 detection were expeditiously formulated and authorized for emergency use, with limitations on the extent of validation and optimization that might otherwise have been afforded in the development process [56]. Moreover, false-negative RT-PCR results may occur due to inappropriate time of testing, inadequate sample collection, and low viral load [57,58]. Finally, RT-PCR often requires a waiting time of several hours for results to become available [59]. Conversely, RT-PCR has been commended for its high specificity for SARS-CoV-2 detection, a notable advantage over chest CT [60].

Limitations

One limitation was our exclusion of articles not published or available in English. In addition, the high I2 statistic (99% for the low-risk- and high-risk-of-bias RT-PCR studies) reflects a significant level of heterogeneity between the studies included, and range from differences clinically (such as the patient cohort) and methodologically (such as study design). Also, while we consider the absence of asymptomatic patients in a study as a bias given that many COVID-19 patients are asymptomatic, we recognize that it is challenging to include asymptomatic patients on a diagnostic test given that the tests are intended to be used on patients suspected of COVID-19 (because the same technique can be used as both a diagnostic test and a screen).

Evaluation of the clinical utility of any diagnostic tool is dependent on a variety of test statistics, including specificity, precision, and accuracy. Because the scope of this study is limited to the evaluation of sensitivity, an overall assessment of the diagnostic value of chest CT in identifying SARS-CoV-2 infection is precluded without discussion of additional test characteristics. Future studies should also address holistic factors like accessibility and cost, to facilitate a more nuanced understanding of the clinical value of CT diagnostics in the context of the COVID-19 pandemic.

CT and RT-PCR sensitivity are also influenced by additional factors, including disease onset and symptom severity. Nonetheless, several studies included in this review refrained from presenting such data, thus limiting the breadth of the sensitivity analyses conducted within this study. As such, the independent effects of COVID-19 patient characteristics such as time of onset and severity of symptoms on the sensitivity of chest CT diagnosis, respectively, remain to be elucidated.

Lastly, concerning sensitivity analyses conducted among low-risk-of-bias studies (Table 4), we acknowledge that the selective inclusion of patients with solely positive RT-PCR results (the refence standard) may slightly bias the results of this study against chest CT. This is due to the inability of CT imaging to distinguish patients with false-negative RT-PCR results. Thus, this review intentionally abstains from offering any definitive comparison statements between the diagnostic sensitivities of RT-PCR with that of CT. Rather, the primary inquiry of this study focuses on the diagnostic utility of chest CT with respect to its sensitivity in the identification of SARS-CoV-2 infection according to low- and high-risk-of-bias studies.

Conclusions

This review is a follow-up to our previous analysis investigating the methodologies of studies on the sensitivity of RT-PCR and CT for SARS-CoV2 infection. CT has been shown to be a highly sensitive diagnostic test in diagnosing SARS-CoV2, but biased methodology inflates sensitivity values and limits generalizability to non-symptomatic populations. While the sensitivity for CT decreased after adjusting for biases, it did not decrease as much as in our original study, which incorporated fewer studies. Conversely, biased methodology may underestimate the sensitivity of RT-PCR by assuming a SARS-CoV2 diagnosis based on symptomology or contact. After adjusting for this bias, when compared to our prior review, the sensitivity values increased for both CT and RT-PCR. Furthermore, incorporating recent data increased the reliability of sensitivity values by expanding the sample size of the low-risk-of-bias studies analysis. It is important to note that CT and RT-PCR were not compared directly to each other because RT-PCR serves as a reference and not a comparator/gold standard. Further large-scale comparator studies using low-risk-of-bias methodologies are still needed to more accurately evaluate these diagnostic tools.

References

1 

Li Q, Guan X, Wu P, et al. Early transmission dynamics in Wuhan, China, of novel coronavirus-infected pneumonia. N Engl J Med 2020; 382: 1199-1207.

2 

Huang C, Wang Y, Li X, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet 2020; 395: 497-506.

3 

Li C, Zhao C, Bao J, et al. Laboratory diagnosis of coronavirus disease-2019 (COVID-19). Clin Chim Acta 2020; 510: 35-46.

4 

Waller JV, Allen IE, Lin KK, et al. The limited sensitivity of chest computed tomography relative to reverse transcription polymerase chain reaction for severe acute respiratory syndrome coronavirus-2 infection: a systematic review on COVID-19 diagnostics. Invest Radiol 2020; 55: 754-761.

5 

Runge VM, Heverhagen JT. Scientific advances, investigative radiology 2020 (and beyond). Invest Radiol 2021; 56: 271-273.

6 

Islam N, Ebrahimzadeh S, Salameh J, et al. Thoracic imaging tests for the diagnosis of COVID-19. Cochrane Database Syst Rev 2021; 3: CD013639.

7 

Li Y, Yao L, Li J, et al. Stability issues of RT-PCR testing of SARS-CoV-2 for hospitalized patients clinically diagnosed with COVID-19. J Med Virol 2020; 92: 903-908.

8 

Wu J, Liu J, Zhao X, et al. Clinical characteristics of imported cases of coronavirus disease 2019 (COVID-19) in Jiangsu province: a multicenter descriptive study. Clin Infect Dis 2020; 71: 706-712.

9 

Liu R, Han H, Liu F, et al. Positive rate of RT-PCR detection of SARS-CoV-2 infection in 4880 cases from one hospital in Wuhan, China, from Jan to Feb 2020. Clin Chim Acta 2020; 505: 172-175.

10 

Fang Y, Zhang H, Xie J, et al. Sensitivity of Chest CT for COVID-19: Comparison to RT-PCR. Radiology 2020; 296: E115-E117.

11 

Long C, Xu H, Shen Q, et al. Diagnosis of the Coronavirus disease (COVID-19): rRT-PCR or CT? Eur J Radiol 2020; 126: 108961.

12 

Xu T, Chen C, Zhu Z, et al. Clinical features and dynamics of viral load in imported and non-imported patients with COVID-19. Int J Infect Dis 2020; 94: 68-71.

13 

Hanif N, Rubi G, Irshad N, et al. Comparison of HRCT chest and RT-PCR in diagnosis of COVID-19. J Coll Physicians Surg Pak 2021; 30: S1-S6.

14 

Thomas C, Naudin M, Tasu JP, et al. Efficacy of chest CT scan for COVID-19 diagnosis in a low prevalence and incidence region. Eur Radiol 2021; 31: 8141-8146.

15 

Schalekamp S, Bleeker-Rovers CP, Beenen LFM, et al. Chest CT in the Emergency Department for diagnosis of COVID-19 pneumonia: Dutch experience. Radiology 2021; 298: E98-E106.

16 

Sverzellati N, Ryerson CJ, Milanese G, et al. Chest radiography or computed tomography for COVID-19 pneumonia? Comparative study in a simulated triage setting. Eur Respir J 2021; 58: 2004188.

17 

Chen D, Jiang X, Hong Y, et al. Can chest CT features distinguish patients with negative from those with positive initial RT-PCR results for coronavirus disease (COVID-19)? AJR Am J Roentgenol 2021; 216: 66-70.

18 

Sun R, Achkar S, Ammari S, et al. Systematic screening of COVID-19 disease based on chest CT and RT-PCR for cancer patients undergoing radiation therapy in a coronavirus French hotspot. Int J Radiat Oncol Biol Phys 2021; 110: 947-956.

19 

Clerici B, Muscatello A, Bai F, et al. Sensitivity of SARS-CoV-2 detection with nasopharyngeal swabs. Front Public Health 2020; 8: 593491.

20 

Xiao AT, Tong YX, Zhang S. False negative of RT-PCR and prolonged nucleic acid conversion in COVID-19: rather than recurrence. J Med Virol 2020; 92: 1755-1756.

21 

Ducray V, Vlachomitrou AS, Bouscambert-Duchamp M, et al. Chest CT for rapid triage of patients in multiple emergency departments during COVID-19 epidemic: experience report from a large French university hospital. Eur Radiol 2021; 31: 795-803.

22 

He JL, Luo L, Luo ZD, et al. Diagnostic performance between CT and initial real-time RT-PCR for clinically suspected 2019 coronavirus disease (COVID-19) patients outside Wuhan, China. Respir Med 2020; 168: 105980.

23 

Herpe G, Lederlin M, Naudin M, et al. Efficacy of chest CT for COVID-19 pneumonia diagnosis in France. Radiology 2021; 298: E81-E87.

24 

Ai T, Yang Z, Hou H, et al. Correlation of chest CT and RT-PCR testing for coronavirus disease 2019 (COVID-19) in China: a report of 1014 cases. Radiology 2020; 296: E32-E40.

25 

Guan WJ, Ni ZY, Hu Y, et al. Clinical characteristics of coronavirus disease 2019 in China. N Engl J Med 2020; 382: 1708-1720.

26 

Huang G, Gong T, Wang G, et al. Timely diagnosis and treatment shortens the time to resolution of coronavirus disease (COVID-19) pneumonia and lowers the highest and last ct scores from sequential chest CT. AJR Am J Roentgenol 2020; 215: 367-373.

27 

Chen Z, Fan H, Cai J, et al. High-resolution computed tomography manifestations of COVID-19 infections in patients of different ages. Eur J Radiol 2020; 126: 108972.

28 

Zhu W, Xie K, Lu H, et al. Initial clinical features of suspected coronavirus disease 2019 in two emergency departments outside of Hubei, China. J Med Virol 2020; 92: 1525-1532.

29 

Wang K, Kang S, Tian R, et al. Imaging manifestations and diagnostic value of chest CT of coronavirus disease 2019 (COVID-19) in the Xiaogan area. Clin Radiol 2020; 75: 341-347.

30 

Caruso D, Zerunian M, Polici M, et al. Chest CT features of COVID-19 in Rome, Italy. Radiology 2020; 296: E79-E85.

31 

Chen J, Qi T, Liu L, et al. Clinical progression of patients with COVID-19 in Shanghai, China. J Infect 2020; 80: e1-e6.

32 

Xu XW, Wu XX, Jiang XG, et al. Clinical findings in a group of patients infected with the 2019 novel coronavirus (SARS-Cov-2) outside of Wuhan, China: retrospective case series. BMJ 2020; 368: m606.

33 

Zhang JJ, Dong X, Cao YY, et al. Clinical characteristics of 140 patients infected with SARS-CoV-2 in Wuhan, China. Allergy 2020; 75: 1730-1741.

34 

Zhou Z, Guo D, Li C, et al. Coronavirus disease 2019: initial chest CT findings. Eur Radiol 2020; 30: 4398-4406.

35 

Zhao X, Liu B, Yu Y, et al. The characteristics and clinical value of chest CT images of novel coronavirus pneumonia. Clin Radiol 2020; 75: 335-340.

36 

Shi H, Han X, Jiang N, et al. Radiological findings from 81 patients with COVID-19 pneumonia in Wuhan, China: a descriptive study. Lancet Infect Dis 2020; 20: 425-434.

37 

Han R, Huang L, Jiang H, et al. Early clinical and CT manifestations of coronavirus disease 2019 (COVID-19) pneumonia. AJR Am J Roentgenol 2020; 215: 338-343.

38 

Wang D, Hu B, Hu C, et al. Clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus-infected pneumonia in Wuhan, China. JAMA 2020; 323: 1061-1069.

39 

Wu C, Chen X, Cai Y, et al. Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in Wuhan, China. JAMA Intern Med 2020; 180: 934-943.

40 

Wen Z, Chi Y, Zhang L, et al. Coronavirus disease 2019: initial detection on chest CT in a retrospective multicenter study of 103 Chinese patients. Radiol Cardiothorac Imaging 2020; 2: e200092.

41 

Ravikanth R. Diagnostic accuracy and false-positive rate of chest CT as compared to RT-PCR in coronavirus disease 2019 (COVID-19) pneumonia: a prospective cohort of 612 cases from India and review of literature. Indian J Radiol Imaging 2021; 31(Suppl 1): S161-S169.

42 

Gietema HA, Zelis N, Nobel JM, et al. CT in relation to RT-PCR in diagnosing COVID-19 in The Netherlands: a prospective study. PLoS One 2020; 15: e0235844.

43 

Orlacchio A, Gasparrini F, Roma S, et al. Correlations between chest-CT and laboratory parameters in SARS-CoV-2 pneumonia: a single-center study from Italy. Medicine (Baltimore) 2021; 100: e25310.

44 

Lu X, Zhang L, Du H, et al. SARS-CoV-2 infection in children. N Engl J Med 2020; 382: 1663-1665.

45 

Zheng F, Liao C, Fan QH, et al. Clinical characteristics of children with coronavirus disease 2019 in Hubei, China. Curr Med Sci 2020; 40: 275-280.

46 

Wang Y, Liu Y, Liu L, et al. Clinical outcomes in 55 patients with severe acute respiratory syndrome coronavirus 2 who were asymptomatic at hospital admission in Shenzhen, China. J Infect Dis 2020; 221: 1770-1774.

47 

Hu Z, Song C, Xu C, et al. Clinical characteristics of 24 asymptomatic infections with COVID-19 screened among close contacts in Nanjing, China. Sci China Life Sci 2020; 63: 706-711.

48 

Inui S, Fujikawa A, Jitsu M, et al. Erratum: Chest CT findings in cases from the cruise ship “Diamond Princess” with coronavirus disease 2019 (COVID-19). Radiol Cardiothorac Imaging 2020; 2: e204002.

49 

Bernheim A, Mei X, Huang M, et al. Chest CT findings in coronavirus disease-19 (COVID-19): relationship to duration of infection. Radio-logy 2020; 295: 200463.

50 

Zhao W, Zhong Z, Xie X, et al. Relation between chest CT findings and clinical conditions of coronavirus disease (COVID-19) pneumonia: a multicenter study. AJR Am J Roentgenol 2020; 214: 1072-1077.

51 

Kassem MNE, Masallat DT. Clinical application of chest computed tomography (CT) in detection and characterization of coronavirus (Covid-19) pneumonia in adults. J Digit Imaging 2021; 34: 273-283.

52 

Hermans JJR, Groen J, Zwets E, et al. Chest CT for triage during COVID-19 on the emergency department: myth or truth? Emerg Radiol 2020; 27: 641-651.

53 

Falaschi Z, Danna PSC, Arioli R, et al. Chest CT accuracy in diagnosing COVID-19 during the peak of the Italian epidemic: a retrospective correlation with RT-PCR testing and analysis of discordant cases. Eur J Radiol 2020; 130: 109192.

54 

Ling Z, Xu X, Gan Q, et al. Asymptomatic SARS-CoV-2 infected patients with persistent negative CT findings. Eur J Radiol 2020; 126: 108956.

55 

Yang W, Cao Q, Qin L, et al. Clinical characteristics and imaging manifestations of the 2019 novel coronavirus disease (COVID-19): a multi-center study in Wenzhou city, Zhejiang, China. J Infect 2020; 80: 388-393.

56 

Wang X, Yao H, Xu X, et al. Limits of detection of 6 approved RT-PCR kits for the novel SARS-Coronavirus-2 (SARS-CoV-2). Clin Chem 2020; 66: 977-979.

57 

Lauer SA, Grantz KH, Bi Q, et al. The incubation period of coronavirus disease 2019 (COVID-19) from publicly reported confirmed cases: estimation and application. Ann Intern Med 2020; 172: 577-582.

58 

Afzal A. Molecular diagnostic technologies for COVID-19: limitations and challenges. J Adv Res 2020; 26: 149-159.

59 

China National Health Commission. Chinese Clinical Guidance for COVID-19 Pneumonia Diagnosis and Treatment (7th edition). March 4th 2020. Available at: http://kjfy.meetingchina.org/msite/news/show/cn/3337.html (Accessed: 02.04.2021).

60 

van Kasteren PB, van der Veer B, van den Brink S, et al. Comparison of seven commercial RT-PCR diagnostic kits for COVID-19. J Clin Virol 2020; 128: 104412.

61 

Bollineni VR, Nieboer KH, Doring S, et al. The role of CT imaging for management of COVID-19 in epidemic area: early experience from a University Hospital. Insights Imaging 2021; 12: 10.

62 

Wang W, Xu Y, Gao R, et al. Detection of SARS-CoV-2 in different types of clinical specimens. JAMA 2020; 323: 1843-1844.

63 

Wong HYF, Lam HYS, Fong AH, et al. Frequency and distribution of chest radiographic findings in patients positive for COVID-19. Radiology 2020; 296: E72-E78.

64 

Qiu H, Wu J, Hong L, et al. Clinical and epidemiological features of 36 children with coronavirus disease 2019 (COVID-19) in Zhejiang, China: an observational cohort study. Lancet Infect Dis 2020; 20: 689-696.

65 

Li K, Fang Y, Li W, et al. CT image visual quantitative evaluation and clinical classification of coronavirus disease (COVID-19). Eur Radiol 2020; 30: 4407-4416.

66 

Xu YH, Dong JH, An WM, et al. Clinical and computed tomographic imaging features of novel coronavirus pneumonia caused by SARS-CoV-2. J Infect 2020; 80: 394-400.

67 

Mirahmadizadeh A, Pourmontaseri Z, Afrashteh S, et al. Sensitivity and specificity of chest computed tomography scan based on RT-PCR in COVID-19 diagnosis. Pol J Radiol 2021; 86: e74-e77.

Copyright: © Polish Medical Society of Radiology This is an Open Access article distributed under the terms of the Creative Commons Attribution-Noncommercial-No Derivatives 4.0 International (CC BY-NC-ND 4.0). License allowing third parties to download articles and share them with others as long as they credit the authors and the publisher, but without permission to change them in any way or use them commercially.
 
Quick links
© 2024 Termedia Sp. z o.o.
Developed by Bentus.