COMPUTERIZED TOMOGRAPHY OF CHEST AND ABSOLUTE LYMPHOCYTE COUNT AS PREDICTOR OF POSITIVE SARS-COV-2 POLYMERASE CHAIN REACTION

Schanz, Olivia; Galvis, Jensen; Fleitas Sosa, Derlis; Yu, Daohai; Salerno, Daniel

TOPIC: Chest Infections TYPE: Original Investigations PURPOSE: Severe acute respiratory syndrome coronavirus 2 (SARS-COV2) is the virus responsible for the coronavirus disease 2019 (COVID-19) which has caused a pandemic generating over half a million deaths in the United States. Lymphopenia is common at presentation in COVID-19 and has been associated with disease severity. Furthermore, patients with severe infection are more likely to have multifocal, bilateral, peripheral ground glass opacities on chest computerized tomography (CT). COVID-19 is diagnosed with a polymerase chain reaction (PCR) test using a nasopharyngeal swab (NPS) for SARS-COV2. The turnaround time for this test has improved;however, as the pandemic continues there have been shortages of reagents leading to continued challenges in timely testing. The aim of our study is to evaluate whether CT chest findings combined with absolute lymphocyte count (ALC) can predict positive COVID-19 PCR swab result. METHODS: We included 3544 patients who presented to the emergency department at a large, urban academic center between March and September 2020. All patients had a NPS for the presence of SARS-CoV-2 using the Luminex NxTAG CoV Extended Panel, a low-resolution chest CT scan without contrast, and a complete blood count with differential to determine ALC. CT scans were classified into three standardized categories by a board-certified chest radiologist based on features of multifocal pneumonia. Category 1 is consistent with multifocal pneumonia, category 2 is indeterminate for multifocal penumonia, and category 3 is inconsistent with multifocal pneumonia. In our analysis we combined categories 2 and 3 into one group. Using the Youden index (J) method, a joint test was performed on CT scan and lymphocyte count. Logistic regression was used determine the parameters of our model. RESULTS: We found that patients with category 1 CT scans and ALC < 1.6 K/mm