key: cord-0883120-cb1n1862
authors: Morikawa, Kazuhiko; Misumi, Shigeki; Igarashi, Takao; Fujimori, Ayako; Ogihara, Akira; Akao, Ryo; Hasumi, Jun; Watanabe, Takashi; Fujii, Yuriko; Ojiri, Hiroya; Mori, Shohei
title: Clinical significance of chest CT for the exclusion of COVID-19 in pre-admission screening: is it worthwhile using chest CT with reverse-transcription polymerase chain reaction test?
date: 2022-05-08
journal: Respir Investig
DOI: 10.1016/j.resinv.2022.04.007
sha: 8972f8b75a3f0990f318e4dad1361b713366c573
doc_id: 883120
cord_uid: cb1n1862

BACKGROUND: A single reverse-transcription polymerase chain reaction (RT-PCR) test is not sufficient to exclude COVID-19 in hospital pre-admission screening. However, repeated RT-PCR tests are time-consuming. This study investigates the utility of chest computed tomography (CT) for COVID-19 screening in asymptomatic patients. METHODS: Between April 2020 and March 2021, RT-PCR testing and chest CT were performed to screen COVID-19 in 10 823 asymptomatic patients prior to admission. Chest CT findings were retrospectively evaluated using the reporting system of the Radiological Society of North America. Using RT-PCR results as a reference, we assessed the diagnostic efficacy of chest CT during both the low- and high-prevalence periods of the COVID-19 pandemic. RESULTS: Following a positive RT-PCR test, 20 asymptomatic patients (0.18%) were diagnosed with COVID-19; in the low-prevalence period, 5 of 6 556 patients (0.076%) were positive; and in the high-prevalence period, 15 of 4 267 patients (0.35%) were positive. Of the 20 asymptomatic COVID-19 positive patients, chest CT results were positive for COVID-19 pneumonia in 8 patients. Chest CT results were false-positive in 185 patients (1.7% false-positive rate, and 60% false-negative rate). Pneumonia that was classified as a “typical appearance” of COVID-19 reported as false-positives in 36 of 39 patients (92.3%). Across the study period, the diagnostic efficacy of “typical appearance” on chest CT were characterized by a sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) of 15%, 99.7%, 99.7%, 7.7%, and 99.8%; 20%, 99.6%, 99.6%, 4%, and 99.9%; and 13.3%, 99.7%, 99.7%, 14.3%, and 99.7%, in the entire study, low-, and high-prevalence periods, respectively. CONCLUSIONS: Addition of chest CT to RT-PCR testing provides no benefit to the detection of COVID-19 in asymptomatic patients.

On March 12, 2020, the World Health Organization declared a pandemic of a new coronavirus that had not been previously observed in humans, known as severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), which causes COVID-19 [1] . The first wave of the COVID-19 pandemic in Japan was confirmed in February 2020, the fourth wave was confirmed in early 2021, and the fifth was confirmed in July 2021 [2] .

The diagnosis of COVID-19 is confirmed through a positive reversetranscription polymerase chain reaction (RT-PCR) test. However, this test has several shortcomings including limited availability, a long turnaround time, and an imperfect diagnostic performance, with a sensitivity of 89% and positive predictive value (PPV) of 95.4-99.8% according to a pooled meta-analysis [3, 4] . The Radiological Society of North America (RSNA) introduced standardized COVID-19 reporting language [5] , which led to publication of the COVID-19 Reporting and Data System (CO-RADS) [6] .

is not necessarily inferior to that of RT-PCR, may be important for COVID -19 diagnosis. Chest CT might complement RT-PCR testing.

However, the incidence of normal chest CT findings in asymptomatic patients with COVID-19 is considerably high (estimated 46%) [10] , and chest CT findings are often normal in the early stage of the disease (i.e., the first 4 to 5 days after the onset of symptoms), even in symptomatic cases (13.9-33.3%) [3, 11] . Based on these factors, chest CT is currently not recommended as a routine screening tool for COVID-19, especially in asymptomatic patients [3, [12] [13] [14] .

In a previous research report written in the early stages of the epidemic, the authors described that nearly half of their asymptomatic patients with COVID-19 had abnormal chest CT findings [10] . Therefore, we hypothesized that most asymptomatic patients with COVID-19 pneumonia could be detected using chest CT as a screening test during the high epidemic stage. Considering the sensitivity of the RT-PCR test, a single RT-PCR test could produce a false-negative result even if the patient already presents an abnormal chest CT finding that is highly suggestive of COVID-19 pneumonia. After experiencing domestic nosocomial clusters and COVID-19-related patient deaths in our hospital at the beginning of the epidemic in early 2020, as had other facilities in Japan [15, 16] , both RT-PCR testing and chest CT have been J o u r n a l P r e -p r o o f performed in our hospital to screen all the hospitalized patients, including asymptomatic patients, since the beginning of the epidemic in April 2020. Repeated RT-PCR tests are time-consuming, laborious, and burdensome for the patients. Therefore, we focused on chest CT as a screening tool for COVID-19 pneumonia, performed simultaneously with the RT-PCR test. We hypothesized that the high sensitivity of the chest CT would balance the limitations of the RT-PCR test and that we would be able to screen asymptomatic patients with COVID-19 who clear a single RT-PCR test. To the best of our knowledge, only a few studies had investigated the utility of chest CT as a screening method for the detection of COVID-19 in asymptomatic patients [13, 14, 17] ; however, none have compared the utility of chest CT for COVID-19 screening in asymptomatic patients between the low-prevalence and high-prevalence periods at the same institution. Therefore, the objectives of this study were: (1) to investigate the utility of chest CT as a screening method for the detection of COVID-19 in asymptomatic patients; (2) J o u r n a l P r e -p r o o f

Between April 2020 and March 2021, a total of 9 117 scheduled inpatients visited the pre-admission COVID-19 testing center at our institution 1 or 2 days before their scheduled admission. We excluded all patients with symptoms that suggested COVID-19 (fever of 37.1 °C or higher, cough, dyspnea, or other respiratory symptoms that are not explained by the current disease) and all patients who had an apparent history of close contact with a patient with COVID-19, 2 weeks prior to admission (n = 96). After exclusion, 9 021 asymptomatic patients who had traveled abroad within the 2 weeks prior to admission were enrolled in this retrospective study (Fig. 1) . For all patients, RT-PCR testing and chest CT were performed on the same day at the pre-admission COVID-19 testing center. We then excluded patients without acute pneumonia on chest CT or with unchanged chest CT findings compared to previous findings ( Fig. 1) . All of the patients were informed by their attending physicians that asymptomatic COVID-19 patients do exist, that pneumonia could be detected incidentally using chest CT, and that medical radiation exposure from chest CT has minimal impact on the individual. 

Chest CT scans were acquired using a multidetector CT scanner with 80 detector rows (Aquilion PRIME; Canon Medical Systems, Otawara, Japan). Patients were oriented in the supine position, and the entire chest (starting from the lung apices down to the posterior costophrenic sulci) was scanned using the following parameters: 1-mm collimation, 120 kV, and volume exposure control (standard deviation value: 17.0).

Images were reconstructed with a slice thickness of 5 mm in the axial plane without an interval gap, where appropriate. No intravenous contrast medium was administered.

Each chest CT examination was reviewed using chest images recorded on a picture archiving and communication system (PACS). Diagnostic imaging reports were prepared by radiologists with more than 6 years of experience, and who were blind to J o u r n a l P r e -p r o o f the results of the related RT-PCR tests. The presence of new pneumonia findings and the possibility of COVID-19 pneumonia were assessed for all CT images.

Two experienced radiologists reviewed the CT images from eligible patients using a PACS workstation monitor. The two radiologists had 25 and 12 years of experience, respectively, were blind to the RT-PCR test results and associated clinical information, and evaluated the images independently of each other. Any differences in evaluation results were resolved via consensus. In this study, the RSNA reporting system was used for image evaluation because it is easy to understand and apply, and it facilitates communication with physicians in other fields [7] . Chest CT findings were categorized into four patterns: "typical," "indeterminate," "atypical," and "negative" according to the reporting language proposed by the RSNA (Table 1 ) [5] . CT images that revealed preexisting lung lesions and no change in findings compared to previous images were categorized as "negative."

We divided the research period into two parts: low-prevalence of COVID-19 (April 2020 to November 2020 with a peak prevalence of 17.41 / 100 000 people in Tokyo) and high-prevalence of COVID-19 (December 2020 to March 2021 with a peak prevalence of 88.94 / 100 000 people in Tokyo) [18, 19] . The frequency of chest CT J o u r n a l P r e -p r o o f findings was evaluated for each period.

Using RT-PCR test results as the reference, the diagnostic efficacy of chest CT was evaluated for each prevalence period by calculating the sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), and negative likelihood ratio (LR-). The statistical analysis was performed using SPSS version 25 (IBM Corp., Armonk, NY, USA). Dichotomous variables were analyzed using Fisher's exact test. For all statistical tests, a P-value of <0.05 was considered statistically significant.

A total of 9 021 asymptomatic patients were included in the study, of which 5 191 were men and 3 830 were women (mean age: 62 years; range: 12-103 years). Of the total patient pool, 5 581 patients (3 247 men and 2 334 women; mean age: 60 years; range: Table 2 summarizes the demographic characteristics and comorbid diseases in those patients with chest CT findings that suggested acute pneumonia (including COVID-19).

Of the 10 823 patients, 20 asymptomatic patients (0.18%) were diagnosed with COVID-19 due to a positive RT-PCR test. COVID-19 was ruled out in the remaining 10 803 patients (99.82%), and none of these patients were diagnosed with COVID-19 after hospitalization. Table 3 summarizes the chest CT findings during the study period.

During the low-prevalence period, 5 of 6 556 patients (0.076%) were diagnosed with COVID-19. Of the 25 patients with abnormal chest CT findings corresponding to a "typical appearance" of COVID-19 pneumonia, only 1 (4%) was diagnosed with COVID-19. Of the 69 patients with "indeterminate appearance" on chest CT, only 1 (1.4%) was diagnosed with COVID-19. In all 16 patients with "atypical appearance" on chest CT, COVID-19 was ruled out. For the remaining 6 446 patients, chest CT findings were "negative"; however, following a positive RT-PCR test, 3 of these patients (0.047%) were diagnosed with COVID-19. We diagnosed COVID-19 using chest CT in 8 patients. However, there were 185 patients who were diagnosed as false-positive for COVID-19 pneumonia during the entire study period (1.7% false-positive rate; 60% false-negative rate). From 36 of the false-positive patients with a "typical appearance" on chest CT, COVID-19 pneumonia was suspected in 13 patients due to lack of comparative images; these 13 patients were later diagnosed as having interstitial pneumonia (either non-specific interstitial pneumonia, desquamative interstitial pneumonia, or smoking-related interstitial lung Of the 149 false-positive patients with "indeterminate appearance" and "atypical appearance" on chest CT, incidental findings suggestive of COVID-19 pneumonia were as follows: diagnosis and treatment for infectious pneumonia (n = 14 patients), old interstitial pneumonia (n = 5 patients), suspected drug-induced pneumonia (n = 3 patients), non-tuberculosis mycobacterium infection (n = 3 patients).

Of the 185 false-positive patients, delay in admission or treatment occurred in 56 patients (30%), with a median delay of 28 days, and a range of delay of 1-180 days.

Suspension of hospitalization occurred in 6 patients (3.2%). Among the 56 falsepositive patients, 10 patients were treated for bacterial pneumonia, 3 were treated for drug-induced lung injury, and 2 were treated for acute exacerbation of interstitial pneumonia.

In this study, the incidence of chest CT findings leading to suspected COVID-19 pneumonia in asymptomatic pre-hospital patients was 1.8%. These results were much lower than those reported in similar studies that targeted pre-operative asymptomatic patients (7-9.6%) [13, 14, 17] .

Using the RSNA reporting system, our results suggest that chest CT findings that are consistent with "typical appearance" were more likely to be COVID-19 pneumonia than those consistent with the other two patterns. This finding is consistent with the results of previous studies [3, 5, 7, 8] ; however, most of the cases were falsepositive. Some studies have demonstrated the usefulness of chest CT even in a lowprevalence region [21, 22] . These studies reported that chest CT may be useful for early diagnosis of COVID-19 in symptomatic patients in the early stages of disease with false-negative RT-PCR results [20, 21] ; however, one meta-analysis indicated that screening patients with suspected disease using chest CT was associated with low PPV (range: 1.5-8.3%) in a low-prevalence region and low-to-moderate PPV (range: 24.3-44.8%) even in a high-prevalence region [4] . In this study, we compared the diagnostic efficacy of chest CT for screening in asymptomatic patients between the low-prevalence J o u r n a l P r e -p r o o f and high-prevalence periods of the pandemic. Our findings indicated that the PPV was 4% in the low-prevalence period and 14.3% in the high-prevalence period, in accordance with values reported in previous studies [4, 13] .

One explanation for the high false-positive rate of chest CT in asymptomatic patients is that many diseases can radiologically mimic COVID-19 pneumonia, including other forms of viral pneumonia [22, 23] , atypical pneumonia [24] , interstitial pneumonia [3, 8, 12, 21, 22] , drug-induced pneumonia [21, 25, 26] , and traumatic changes such as rib fractures [27] . Our results suggested that unknown early or mild interstitial pneumonia and bilateral lower lobe dorsal lung collapse due to poor inspiration were difficult to differentiate from mild COVID-19 pneumonia, which supports the findings of previous reports [3, 8, 12, 21, 22] . Based on these results, chest CT used as an adjunct to RT-PCR testing for COVID-19 screening in asymptomatic patients did not contribute to the exclusion of COVID-19, even in the high-prevalence period.

At the end of the study period, the Alpha strain as a variant of concern (VOC) was the dominant origin of COVID-19 infections across Japan [28] [29] [30] . A shift to the Delta strain had been reported, following a trend that was also developing in other countries [28, 31, 32] . Several studies (including preprints) reported that the time taken from viral exposure to the result of a positive RT-PCR test result may be shorter in J o u r n a l P r e -p r o o f patients with the Delta strain than in those without VOCs, that the Delta strain may multiply faster and be more infectious in the initial stages, and that the Delta strain may prove to be more virulent than non-VOCs [33, 34] . Several studies have reported that COVID-19 vaccinations remain highly effective against COVID-19 hospitalization and death, including those caused by the Delta variant [35] ; however, breakthrough infection in vaccinated patients remains the focus of attention [36] and newer VOCs have been defined in early 2022 that have contributed to a global re-epidemic [18, 28, 37, 38] . We did not find chest CT to be an effective pre-admission screening tool for asymptomatic patients because it was not effective even during the high-prevalence period when the vaccine was not widely available. Moreover, considering the increased medical costs associated with chest CT compared with those of RT-PCR, we do not recommend using chest CT as an effective tool to screen for COVID-19.

This study has several limitations. First, it had a single-center retrospective design with a heterogenous pre-admission patient group. Second, we cannot exclude the possibility that some asymptomatic and infected patients passed all pre-admission examinations. Third, we did not investigate vaccination status as a potential confounder in this study. Lastly, since the Omicron strain is currently at the center of the epidemic [18, 28, 37, 38] , the results of this study obtained from April 2020 to March 2021 were J o u r n a l P r e -p r o o f based on the data at a time when other mutants, such as the Delta variant, were endemic.

The current situation in which the Omicron variant is prevalent is different to the circumstances in which this study was conducted with previous COVID-19 variants.

In conclusion, chest CT is not adequately sensitive for COVID-19 screening in asymptomatic patients compared with RT-PCR, even in high-prevalence periods.

Moreover, 1.7% of patients in our study had false-positive results, even with the use of the RSNA reporting system, which deterred them from treatment and hospitalization.

Our findings indicate that addition of chest CT to RT-PCR testing does not provide additional benefit for COVID-19 screening in asymptomatic patients. commercial, or not-for-profit sectors.

The authors have no conflicts of interest.

J o u r n a l P r e -p r o o f Tables   Table 1. Standardized reporting language for findings of COVID-19 by computed tomography (CT), as defined by the Radiological Society of North America (RSNA) [5] Imaging pattern

Typical appearance Peripheral, bilateral, ground-glass opacity (GGO) with or without consolidation or visible intralobular lines ("crazy-paving") Multifocal GGO of rounded morphology with or without consolidation or visible intralobular lines ("crazy-paving") Reverse halo sign or other findings of organizing pneumonia (seen later in the disease)

Absence of typical features AND presence of: Multifocal, diffuse, perihilar, or unilateral GGO with or without consolidation lacking a specific distribution and are nonrounded or non-peripheral Few very small GGO with a nonrounded and non-peripheral distribution 

General's opening remarks at the media briefing on COVID-19

Labour and Welfare current situation in Japan. Situation report. Current situation in Japan

Chest CT in COVID-19: What the radiologist needs to know

Diagnostic performance of CT and reverse transcriptase polymerase chain reaction for coronavirus Disease 2019: A meta-analysis

Radiological Society of North America Expert Consensus Document on Reporting Chest CT Findings Related to COVID-19: Endorsed by the Society of Thoracic Radiology, the American College of Radiology, and RSNA

CO-RADS: A Categorical CT Assessment Scheme for Patients Suspected of Having COVID-19-Definition and Evaluation

Comparison of chest CT grading systems in coronavirus disease 2019 (COVID-19) pneumonia

The usefulness of chest CT imaging in patients with suspected or diagnosed COVID-19: A review of literature

Diagnostic accuracy of North America expert consensus statement on reporting CT findings in patients suspected of having COVID-19 infection: An Italian singlecenter experience

Chest CT Findings in Cases from the Cruise Ship "Diamond Princess

Temporal changes of CT findings in 90 patients with COVID-19 pneumonia: A longitudinal study

COVID-19 imaging: What we know now and what remains unknown

Preoperative computerized tomography screening for COVID-19 pneumonia in asymptomatic patients: Experiences from two centers

Preoperative chest computed tomography screening for coronavirus Disease 2019 in asymptomatic patients undergoing cardiac surgery

Nosocomial SARS-CoV-2 infections in Japan: A cross-sectional newspaper database survey

Clusters of COVID-19 in long-term care hospitals and facilities in Japan from 16

Chest CT screening for COVID-19 in elective and emergency surgical patients: Experience from a UK tertiary centre

Updates on COVID-19 in Tokyo

The National Institute of Infectious Diseases. Current situation of infection

Diagnostic performance of chest CT to differentiate COVID-19 pneumonia in non-highepidemic area in Japan

COVID-19 pneumonia: A review of typical CT findings and differential diagnosis

Performance of Radiologists in Differentiating COVID-19 from Non-COVID-19 Viral Pneumonia at Chest CT

CT manifestations of coronavirus Disease (COVID-19) pneumonia and influenza virus pneumonia: A comparative study

Differences among COVID-19, bronchopneumonia and atypical pneumonia in chest high resolution computed tomography assessed by artificial intelligence technology

Pulmonary drug toxicity: Radiologic and pathologic manifestations

Differentiation of COVID-19 pneumonitis and ICI induced pneumonitis

Evaluation of chest CT scan as a screening and diagnostic tool in trauma patients with coronavirus disease 2019 (COVID-19): A cross-sectional study

Labour and Welfare current situation in Japan. Variant of concern (VOC) cases

Predicted dominance of variant Delta of SARS-CoV-2 before Tokyo Olympic Games

World Health Organization. COVID-19 weekly epidemiological update

World Health Organization. COVID-19 weekly epidemiological update

Viral infection and transmission in a large, well-traced outbreak caused by the SARS-CoV-2 Delta variant

Clinical and Virological Features of SARS-CoV-2 Variants of Concern: A Retrospective Cohort Study Comparing B.1.1.7 (Alpha), B.1.315 (Beta), and B.1.617.2 (Delta)

COVID-19 vaccines and vaccination

Reporting Breakthrough Cases. Possibility of COVID-19 Illness after Vaccination

SARS-CoV-2 Variant Classifications and Definitions

World Health Organization. Tracking SARS-CoV-2 variants

) Malignancies including post-therapeutic or undertreatment state (lung cancer

Collagen disease (5) 2.6

Drugs that may affect patient's immunocompetence Corticosteroid (median 10 mg; range 2-30 mg) 10 5

NTM, non-tuberculosis mycobacteria; CKD, chronic kidney disease

MTX, methotrexate; ICI, immune checkpoint inhibitor

We would like to thank Editage (http://www.editage.com) for editing and reviewing this manuscript for English language.

This research did not receive any specific grant from funding agencies in the public,