ABSTRACT
INTRODUCTION: Prone positioning is a cornerstone therapy in COVID-19 associated severe Acute Respiratory Distress Syndrome (ARDS). Ventilatory ratio (VR) calculates dead space ventilation, which is independently associated with an increased risk of mortality in ARDS patients. However, no studies so far have demonstrated the changes in VR with proning in such patients. We evaluated VR as a new tool to predict survival in prone positioned COVID -19 ARDS patients. METHODS: This retrospective study included 24 COVID-19 positive ARDS patients admitted to SUNY Downstate Medical Center ICU between April 4, 2020 to May 15, 2020, who had PaO2/FiO2 (PF) ratio < 150, and were proned. Data points of total number of proning days, PaCO2, PF ratio, and VR, were collected before proning, on day 1, and on the last day of proning, along with the mortality data. Statistical analysis was performed in R using student t-test and linear regression. RESULTS: Of the 24 patients included, 80% were male. The median age of the patients was 68 years. With proning, the PF ratio increased by a mean of 0.32 (IQR 0.04,0.54 p<0.01) on day 1 and 0.592 (IQR 0.17,0.91 p<0.001) on last day of proning. VR decreased by a mean of -0.156 (IQR -0.837,+0.09 p=0.15) on day 1 and -0.388 (IQR -0.725,- 0.056 p= 0.15) on last day of proning. PaCO2 is predictive of VR before proning (adjusted R2 =0.69, p<0.001), and on last day of proning (adjusted R2 =0.5, p<0.001). The length of proning had no relationship with change in VR or PF ratio at any time point. VR of < 2 before proning was predictive of overall survival (p < 0.05) with an average survival of 25 days vs 9.8 days for VR > 2. The mortality rate for all patients included was 80% (20/24). CONCLUSIONS: COVID -19 ARDS patients demonstrate impaired ventilation and hypoxemia. Commonly used PF ratio, in proning protocols, has limited prognostic value but deadspace fraction is a powerful predictor of mortality in ARDS. Proning improved PF ratio and VR, however, the latter did not reach statistical significance. PF ratio was not predictive of length of survival, however, VR > 2 before proning is a significant predictor for overall mortality. Hence, VR can function as a new bedside predictive tool with proning in COVID-19 ARDS patients.