Although there are multiple presentations, respiratory disease is the primary manifestation of a severe COVID-19 infection. For managing severe ARDS due to COVID-19, current guidelines from the United States National Institutes of Health recommend using “low tidal volume ventilation (VT 4–8 mL/kg of predicted body weight) over higher tidal volumes (VT >8 mL/kg)” and “prone ventilation for 12 to 16 hours per day over no prone ventilation” for hypoxia refractory to optimized ventilatory settings [
6]. Chest CT is an instrumental part of the workup and assessment of disease severity with the most common findings being bilateral ground glass opacities. COVID-19 infection has already shown lung parenchymal changes that progressed to large bulla, which coalesce and enlarge over time leading to bullous emphysematous disease and subpleural blebs [
2]. This could predispose this population to developing pneumothorax or pneumomediastinum, which have already been reported in the literature [
7]. One explanation for the pneumomediastinum could be the Macklin effect, where air from ruptured alveoli dissects along bronchovascular planes and into the mediastinum [
8,
9]. This is in contrast to a pneumothorax where the alveolar rupture is involved with injury to the visceral pleura resulting in air accumulation in the pleural space. A small asymptomatic primary pneumothorax can typically be managed with observation and sequential imaging to ensure stabilization or re-expansion of the lung. However, barotrauma from positive pressure ventilation is a well-known cause of iatrogenic pneumothorax and the current recommendation for critically ill patients who are intubated in the ICU is tube thoracostomy placement [
3]. To the best of our knowledge, there are currently no recommendations specifically for managing a small pneumothorax in a COVID-19 positive patient. In each of these cases of pneumothorax due to barotrauma we decided to pursue non-operative management. Factors contributing to this decision were the small size of the pneumothoraces and the fact that the patient would need to be disconnected from the ventilator to allow the lung to drop to avoid parenchymal injury, which could be detrimental to an already critical patient. For the majority of our cases we decreased the PEEP as much as possible and recommend this practice to others attempting expectant management. However, further studies are needed to assess if this truly makes a difference. An additional benefit was limiting viral exposure to hospital staff. The limitations of this study is that there were only 4 cases encountered making it difficult to apply to a larger group, and that the pneumothoraces were small (< 3 cm from apex to cupola), leaving no guidance for those with moderate to large sized pneumothoraces. We have not managed any other COVID-19 patients with pneumothorax to date. While no institutional guidelines exist, we would have a lower threshold for pleural drainage in moderate to large pneumothoraces. Expectant management can be quite precarious in this population, but we believe these cases show that observation in ventilated COVID-19 positive patients with a small pneumothorax may be an appropriate option. It has been shown in post-mortem studies that COVID-19 causes diffuse alveolar damage and lymphocytic or cellular fibromyxoid exudate [
10]. We hypothesize that this severe exudative process may contribute to sealing the alveolar and pleural injury, allowing for small pneumothoraces to remain stable during positive pressure ventilation. As the COVID-19 pandemic continues to unfold we will learn more about the exact pathophysiology that occurs.