Performance of the Ventilatory Ratio to Predict Intubation in Patients With COVID-19 Acute Respiratory Distress Syndrome.

OBJECTIVES: Ventilatory ratio (VR) could be a useful marker of COVID-19 acute respiratory distress syndrome (ARDS) worsening, since its value has been shown to correlate with dead space and severity of non-COVID-19 ARDS. In this study, we sought to: 1) compare VR to the estimated intrapulmonary shunt and the Pao2/Fio2 ratio for the association with the risk of intubation and 2) define the best VR threshold to discriminate between intubated patients and nonintubated patients.

DESIGN: Multicenter, prospective experimental study.

SETTING: ICUs from University and General Hospitals.

PATIENTS: Patients with moderate-to-severe COVID-19 ARDS.

INTERVENTIONS: We conducted standardized continuous positive airway pressure sessions with an Fio2 of 100% and positive end-expiratory pressure of 5 cm H2O to calculate the Pao2/Fio2 ratio, the estimated intrapulmonary shunt (indexed on cardiac output), and VR. These sessions were repeated until intubation or ICU discharge. A time-dependent joint multivariable model was performed.

MEASUREMENTS AND MAIN RESULTS: Fifty-two patients were included, 25 of whom required intubation (48%). Among studied variables including the Pao2/Fio2 ratio and the estimated intrapulmonary shunt, VR was most associated with the longitudinal risk of intubation with a hazard ratio (HR, 3.30; 97.5% CI, 1.42-8.91; p = 0.001) compare to the Pao2/Fio2 ratio with HR (0.98; 97.5% CI, 0.97-0.99; p = 0.002) and the estimated intrapulmonary shunt with HR (2.31; 97.5% CI, 0.37-24.01; p = 0.394). The best VR threshold values were identified at day 1 (VR = 2.1, sensitivity, 85.7%; specificity, 60.1%; area under the receiver operating characteristic curve [AUC] = 0.810), day 3 (VR = 2.3; sensitivity, 92.0%; specificity, 91.2%; AUC = 0.973), and day 5 (VR = 2; sensitivity, 92%; specificity, 97%; AUC = 0.979) to discriminate intubated and nonintubated patients.

CONCLUSIONS: The study provides valuable insights into COVID-19 ARDS, highlighting VR as a reliable and objective predictor of intubation. Increased VR could be a critical marker of lung injury progression in patients with ARDS.