tetano
Editor, Senior Moderator
AJR Am J Roentgenol
. 2020 Jun 1;1-9.
doi: 10.2214/AJR.20.23391. Online ahead of print.
Systematic Review and Meta-Analysis on the Value of Chest CT in the Diagnosis of Coronavirus Disease (COVID-19): Sol Scientiae, Illustra Nos
Hugo J A Adams[SUP] 1 [/SUP], Thomas C Kwee[SUP] 2 [/SUP], Derya Yakar[SUP] 2 [/SUP], Michael D Hope[SUP] 3 4 [/SUP], Robert M Kwee[SUP] 5 [/SUP]
Affiliations
Abstract
OBJECTIVE. The purpose of this article is to systematically review and meta-analyze the diagnostic accuracy of chest CT in detecting coronavirus disease (COVID-19). MATERIALS AND METHODS. MEDLINE was systematically searched for publications on the diagnostic performance of chest CT in detecting COVID-19. Methodologic quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool. Meta-analysis was performed using a bivariate random-effects model. RESULTS. Six studies were included, comprising 1431 patients. All six studies included patients at high risk of COVID-19, and five studies explicitly reported that they included only symptomatic patients. Mean prevalence of COVID-19 was 47.9% (range, 27.6-85.4%). High or potential risk of bias was present throughout all QUADAS-2 domains in all six studies. Sensitivity ranged from 92.9% to 97.0%, and specificity ranged from 25.0% to 71.9%, with pooled estimates of 94.6% (95% CI, 91.9-96.4%) and 46.0% (95% CI, 31.9-60.7%), respectively. The included studies were statistically homogeneous in their estimates of sensitivity (p = 0.578) and statistically heterogeneous in their estimates of specificity (p < 0.001). CONCLUSION. Diagnostic accuracy studies on chest CT in COVID-19 suffer from methodologic quality issues. Chest CT appears to have a relatively high sensitivity in symptomatic patients at high risk of COVID-19, but it cannot exclude COVID-19. Specificity is poor. These data, along with other local factors such as COVID-19 prevalence, available real-time reverse transcriptase-polymerase chain reaction tests, staff, hospital, and CT scanning capacity, can be useful to healthcare professionals and policy makers to decide on the utility of chest CT for COVID-19 detection in the hospital setting.
Keywords: COVID-19; CT; coronavirus disease; infection; lung; viral infections.
. 2020 Jun 1;1-9.
doi: 10.2214/AJR.20.23391. Online ahead of print.
Systematic Review and Meta-Analysis on the Value of Chest CT in the Diagnosis of Coronavirus Disease (COVID-19): Sol Scientiae, Illustra Nos
Hugo J A Adams[SUP] 1 [/SUP], Thomas C Kwee[SUP] 2 [/SUP], Derya Yakar[SUP] 2 [/SUP], Michael D Hope[SUP] 3 4 [/SUP], Robert M Kwee[SUP] 5 [/SUP]
Affiliations
- PMID: 32478562
- DOI: 10.2214/AJR.20.23391
Abstract
OBJECTIVE. The purpose of this article is to systematically review and meta-analyze the diagnostic accuracy of chest CT in detecting coronavirus disease (COVID-19). MATERIALS AND METHODS. MEDLINE was systematically searched for publications on the diagnostic performance of chest CT in detecting COVID-19. Methodologic quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool. Meta-analysis was performed using a bivariate random-effects model. RESULTS. Six studies were included, comprising 1431 patients. All six studies included patients at high risk of COVID-19, and five studies explicitly reported that they included only symptomatic patients. Mean prevalence of COVID-19 was 47.9% (range, 27.6-85.4%). High or potential risk of bias was present throughout all QUADAS-2 domains in all six studies. Sensitivity ranged from 92.9% to 97.0%, and specificity ranged from 25.0% to 71.9%, with pooled estimates of 94.6% (95% CI, 91.9-96.4%) and 46.0% (95% CI, 31.9-60.7%), respectively. The included studies were statistically homogeneous in their estimates of sensitivity (p = 0.578) and statistically heterogeneous in their estimates of specificity (p < 0.001). CONCLUSION. Diagnostic accuracy studies on chest CT in COVID-19 suffer from methodologic quality issues. Chest CT appears to have a relatively high sensitivity in symptomatic patients at high risk of COVID-19, but it cannot exclude COVID-19. Specificity is poor. These data, along with other local factors such as COVID-19 prevalence, available real-time reverse transcriptase-polymerase chain reaction tests, staff, hospital, and CT scanning capacity, can be useful to healthcare professionals and policy makers to decide on the utility of chest CT for COVID-19 detection in the hospital setting.
Keywords: COVID-19; CT; coronavirus disease; infection; lung; viral infections.