Can rapid antibody tests and chest computed tomography really substitute real-time polymerase chain reaction in COVID-19?

To the Editor

We read with great interest the recently published study by Ozturk et al. about the evaluation of rapid antibody tests and chest computed tomography (CT) in COVID‐19 patients.

At the beginning of the COVID‐19 pandemic, several physicians combined chest CT and rapid antibody test to confirm the diagnosis of COVID‐19, mainly because of the suboptimal accuracy of the first real‐time‐polymerase chain reaction (RT‐PCR) screening test and of the paucity of reagents.

The authors should be commended for their efforts in evaluating the rapid antibody test and chest CT findings of COVID‐19 in a consistent number of patients.

The high sensitivity (77.8%) reported in the paper confirms the role of chest CT as a fundamental diagnostic tool for the early diagnosis of COVID‐19 infection in symptomatic patients.

Nevertheless, some concerns could be raised, especially about the methods adopted in this retrospective study which, in our opinion, suggest taking with caution at least some of the conclusions drawn by the Authors.

First of all, since only 46 patients out of 320 showed a positive result at RT‐PCR, it is not clear how the remaining 274 patients could be classified as infected by COVID‐19. The authors reported that patients were classified as COVID‐19 positive on the basis of clinical evaluation and imaging findings. Hence, we think that the diagnosis of COVID‐19 in those 274 patients should have not been used as a reference, even more, if it was used to evaluate the accuracy of CT, itself employed for COVID‐19 diagnosis. The use of such reference standard could have biased all the results and should at least have been reported among the study limitations.

Moreover, it has been previously reported that patients with a monolateral lung involvement at chest CT could have a falsely negative RT‐PCR. Since almost 50% of those included in this study did not show a bilateral lungs involvement at chest CT, in this group RT‐PCR should have been performed on the bronchoalveolar lavage to confirm the diagnosis of COVID‐19 infection.

Second, antibodies production after an infection has a variable “window period" that depends on the time required for seroconversion. Indeed, Long et al. reported that the positive rate of virus‐specific immunoglobulin G reached 100% after 17–19 days after symptoms onset, and positivity of virus‐specific immunoglobulin M reached a peak of 94.1% 20–22 days after symptom onset. Therefore, if serum samples were collected within 0–7 days from COVID‐19 diagnosis, it is reasonable that a not irrelevant part of the population was in that “window period” and, therefore, tested negative.

Finally, it is stated that chest CT was evaluated by an “infection and clinical microbiologist.” In our opinion, to have a more reliable identification of radiological signs of the CT scans, images should have been reviewed by at least one radiologist expert in thoracic imaging.

In summary, while we agree with the conclusions that chest CT and rapid antibody test can be useful diagnostic tools for clinicians in the setting of the COVID‐19 pandemic, it should be highlighted that the multiple biases of this retrospective study could affect the robustness of the conclusions drawn by the authors.

CONFLICT OF INTEREST

The authors declare that there are no conflict of interests.