PNEUMOMEDIASTINUM AS A COMPLICATION OF COVID-19: MAY NOT NEED INTERVENTION

Abstract

SESSION TITLE: Late-breaking Abstract Posters SESSION TYPE: Original Investigation Posters PRESENTED ON: October 18-21, 2020 PURPOSE: Evaluate patients with pneumomediastinum (PTM) as a complication of coronavirus 2019 (Covid-19) infection and their outcomes. METHODS: Data was obtained by retrospective analysis of a database of Covid-19 patients, admitted to two New York hospitals from March 25th to April 26th, 2020. Identified 13 patients with PTM as a complication of Covid-19. RESULTS: Median age of patients was 67 years, 7 (54%) were men, and 1 patient had pre-existing emphysema. Presenting symptoms included dyspnea (13/13, 100%), fever (10/13, 77%), and cough (9/13, 69%); none had chest pain. Median presenting SaO2/FiO2 ratio was 359. Pneumomediastinum was diagnosed on presentation in 2 patients (15%), during hospitalization but prior to invasive mechanical ventilation in 4 patients (31%), and after invasive mechanical ventilation in 7 patients (54%). Six patients had CT scans confirming the diagnosis of PTM. Twelve of the 13 patients were initiated on mechanical ventilation at median 5.5 days. Four patients (31%) developed pneumothoraces and 1 patient required bilateral chest tube drainage. Tension pneumothorax, tension pneumopericardium, and chest wall compartment syndrome were not observed in our cohort. Pneumomediastinum self-resolved in 5 patients (38%) while on mechanical ventilation. (Fig. 1) Seven out of 12 intubated patients (58%) underwent prone ventilation without worsening of PTM. One patient was discharged home, 10 patients expired, and 2 patients remain hospitalized. CONCLUSIONS: Pneumomediastinum is a known sequelae of noncompliant lungs in severe acute respiratory distress syndrome (ARDS), typically in the setting of positive pressure delivered with invasive mechanical ventilation. Our cohort had a low prevalence of predisposing risk factors such as emphysema, and a high mortality. Pneumomediastinum was not strongly associated with mechanical ventilation, as nearly half (6/13) of the patients developed PTM prior to intubation. CLINICAL IMPLICATIONS: In our cohort, findings of PTM may be attributed to two main etiologies converging: damage from Covid-19 and self-inflicted increases in transpulmonary pressures. We postulate that impaired surfactant production from type II pneumocytes may predispose to atelectrauma leading to diffuse alveolar injury and risk of rupture, unlike in typical ARDS where barotrauma occurs in the context of poorly compliant lungs. Moreover, the increased work of breathing and pronounced cough seen in Covid-19 pneumonia may lead to increased transpulmonary pressures resulting in PTM. Reductions in transpulmonary pressures after intubation with initiation of sedation and neuromuscular blockade may explain the resolution of PTM in spite of invasive ventilation. Given the high mortality we observed, early identification and initiation of methods to reduce transpulmonary pressures is imperative to reduce risk of further complications. DISCLOSURES: No relevant relationships by Young Im Lee, source=Web Response No relevant relationships by Nan Li, source=Web Response No relevant relationships by Joseph Mathew, source=Web Response No relevant relationships by Lina Miyakawa, source=Web Response No relevant relationships by Paru Patrawalla, source=Web Response No relevant relationships by Paru Patrawalla, source=Web Response No relevant relationships by Adam Rothman, source=Web Response No relevant relationships by Vishad Sheth, source=Web Response No relevant relationships by David Steiger, source=Web Response No relevant relationships by Jigna Zatakia, source=Web Response