ARRS 2022 Abstracts


1327. COVID Pneumonia: Can a Chest Radiography Scoring System be Used to Predict a Favorable Outcome in Patients Receiving Remdesivir
Authors * Denotes Presenting Author
  1. Shaimaa Fadl *; Virginia Commonwealth University Health System
  2. khalid Shaqdan; Virginia Commonwealth University Health System
  3. E. Proffitt; Virginia Commonwealth University Health System
  4. Frank Dana; Virginia Commonwealth University Health System
  5. Leila Rezai Gharai; Virginia Commonwealth University Health System
  6. Mark Parker; Virginia Commonwealth University Health System
We hypothesize that the application of a qualitative radiographic parenchymal severity scoring system on conventional chest radiography may help predict either a favorable or unfavorable response to Remdesivir compared to receiving the standard of care therapy without Remdesivir. If our hypothesis is correct, such a scoring system may be helpful in predicting the evolution of COVID-19 pneumonia, potential complications related to such, or in predicting a given patient’s clinical outcome during their hospital course.

Materials and Methods:
Following Institutional Review Board (IRB) approval, we conducted a HIPPA-compliant retrospective review of consecutive patients admitted and diagnosed with confirmed positive real-time reverse transcriptase polymerase chain reaction (rRT-PCR) test for SARS-CoV-2 infection, presenting between March 1 and September 30, 2020. In those patients who received chest imaging, a minimum of 3 fellowship-trained thoracic radiologists performed a consensus read and comparative analysis of the chest x-ray findings. Radiologists were blinded to patients who were COVID-19 positive and did not receive Remdesivir. Radiologists reviewed the chest radiographs at the time of initial admissions and any follow-up x-rays. A 4-point qualitative scoring system was applied by the radiologists to each chest radiograph. Each lung was divided into three zones (upper, mid, and lower zones). A “0 score” was assigned for no abnormalities; “1 score” was given for minimal doubtful findings; “2 score” was given for interstitial opacities; and “3 score” was assigned for alveolar air space opacities. The frequency and percentages for categorical variables were summarized as means and standard deviations for continuous variables.

There were 125 of 453 hospitalized patients with COVID-19 positive who enrolled in the Remdesivir trial; 89 of these patients received imaging studies including one or more single view chest radiographs at admission, during their hospital course, and/or at the time of discharge. Out of the 89 patients with imaging studies 36 patients received Remdesivir and 53 patients did not. Nineteen of 36 (52.78%) patients treated with Remdesivir improved radiographically based on their x-ray severity score and 24/53 (45.48%) patients improved in the absence of Remdesivir. The mean age for patients receiving and not receiving Remdesivir was 57 years (SD 12.5) and 51 years (SD20.9), respectively. An equal number of men and women were enrolled. Most enrollees were African American (56%), and 90% were immunocompetent. By the completion of our analysis, 85% (326/380) of patients entered the recovery phase whereas 5% (18/380) required pulmonary rehabilitation and 10% (36/380) died of COVID-related complications.

Unfortunately, the application of our chest radiographic severity scoring system of hospitalized patients with COVID-19 pneumonia could not predict those patients who may or may not favorably respond to Remdesivir therapy. We discuss potential reasons for this short-coming in the presentation.