Respiratory Physiology of COVID-19-Induced Respiratory Failure Compared to ARDS of Other Etiologies

Domenico Luca Grieco; Filippo Bongiovanni; Lu Chen; Luca S. Menga; Salvatore Lucio Cutuli; Gabriele Pintaudi; Simone Carelli; Teresa Michi; Flava Torrini; Gianmarco Lombardi; Gian Marco Anzellotti; Gennaro De Pascale; Andrea Urbani; Maria Grazia Bocci; Eloisa S. Tanzarella; Giuseppe Bello; Antonio M. Dell'Anna; Salvatore M. Maggiore; Laurent Brochard; Massimo Antonelli


Crit Care. 2020;24(529) 

In This Article


COVID-19 Cohort

Thirty patients were enrolled. Demographics and relevant clinical characteristics are reported in Table 1. Twenty-three (77%) patients met the criteria for moderate ARDS, and 7 (23%) for severe ARDS.

Matched Cohorts

Thirty patients from the historical ARDS cohort were successfully matched to COVID-19 patients (individual data provided in Supplementary Table 1, demographics in Table 1). At low PEEP, median PaO2/FiO2 was 119 mmHg [101–142] in COVID-19 patients and 116 mmHg [87–154] in patients with ARDS from other etiologies. FiO2 was not different between cohorts (p = 0.51), while tidal volume was slightly higher, and PEEP lower, in patients with COVID-19 than controls. Although statistically significant, mean differences between cohorts were clinically negligible: 0.3 ml/kg [CI95% 0–0.6] and 0.9 cmH2O [CI95% 0.1–1.7], respectively. All COVID-19 patients were studied within 24 h from endotracheal intubation. Patients in the historical cohort underwent study procedures after a median time from ICU admission of 5 [3–10] days.

Respiratory Mechanics

These results are displayed in Figure 1 and Table 2.

Figure 1.

Respiratory mechanics. Individual values, medians, and interquartile range showing the distribution of respiratory variables in the matched cohorts. For each variable, COVID-19 patients' values are compared to those for matched non-COVID-19 ARDS patients, as detailed in text. PaO2/FiO2 ratio, PaCO2, respiratory system compliance, driving pressure, ventilatory ratio, and standardized minute ventilation were measured at low PEEP. Ventilatory ratio, compliance, and its predicted body weight-indexed value were slightly higher in COVID-19 patients than in ARDS of other etiologies. PBW, predicted body weight

Airway closure in COVID-19 cohort was less frequent than in ARDS from other etiologies: 2 (7%) vs. 10 (30%) patients (p = 0.021). At low PEEP, inter-individual variability of respiratory system compliance was 36% in COVID-19 patients and 31% in ARDS from other causes. Average respiratory system compliance and respiratory system compliance/PBW were slightly higher in patients with COVID-19 than in those affected by other ARDS: mean differences were 7 ml/cmH2O [CI95% 0–14] and 0.11 ml/cmH2O/kg [CI95% 0.2–0.20], respectively. This was not associated to statistically significant differences in the driving pressure (p = 0.098). In both cohorts, twenty-four (80%) patients showed driving pressure equal or lower than 14 cmH2O. In the COVID-19 cohort, respiratory system compliance (r = 0.52, p = 0.003) and respiratory system compliance/PBW (r = 0.53, p = 0.002) were linearly related to PaO2/FiO2 (Figure 2).

Figure 2.

Relationships between PaO2/FiO2, respiratory system compliance, and ventilatory ratio in COVID-19 ARDS patients. In the COVID-19 cohort, respiratory system compliance and its PBW-indexed value were linearly related to PaO2/FiO2 ratio (upper panels). Ventilatory ratio was not related to PaO2/FiO2 ratio nor to respiratory system compliance (lower panels). CRS, respiratory system compliance; PBW, predicted body weight

In COVID-19 cohort, ventilatory ratio was higher than in ARDS from other etiologies (mean difference 0.3 [CI95% 0–0.6], p = 0.032). In COVID-19 cohort, ventilatory ratio was not related to PaO2/FiO2 (p = 0.42) nor to respiratory system compliance (p = 0.46).

Response to PEEP—Gas Exchange and Respiratory Mechanics

These results are displayed in Table 2 and Figures 3 and 4.

Figure 3.

Response to PEEP. Before-and-after plots showing the effects of high PEEP on PaO2/FiO2 ratio (upper left panel), ventilatory ratio (upper right panel), respiratory system compliance (lower left panel), and driving pressure (lower right panel). In both groups, PaO2/FiO2 ratio increased at increasing PEEP, but the increase was significantly higher in the COVID-19 cohort (see also Table 2). Ventilatory ratio could either increase, decrease, or remain unchanged, with no significant difference between cohorts. At high PEEP, compliance increased and driving pressure decreased in non-COVID-19 patients, while no changes were detected in COVID-19 patients. Black dots represent individual patients before and after the increase in PEEP, and individual changes are traced by dotted lines. CRS, respiratory system compliance

Figure 4.

Potential for lung recruitment in COVID-19 ARDS patients. In COVID-19 patients, recruitment-to-inflation ratio was inversely related to respiratory system compliance at low PEEP (upper left panel, Pearson's correlation and linear regression), meaning that patients with lower baseline compliance displayed the highest potential for lung recruitment. Importantly, lung recruitability was not related to changes in respiratory system compliance (and driving pressure) induced by PEEP. With PEEP, compliance could either increase, decrease, or remain unchanged (change in compliance was defined clinically relevant when >5 ml/cmH2O), independently from the recruitment-to-inflation ratio, as shown in the upper right panel. The changes in PaO2/FiO2 induced by PEEP were independent from recruitability (lower left panel), while PEEP-induced PaCO2 changes were weakly but significantly related to the recruitment-to-inflation ratio (lower right panel,). CRS, respiratory system compliance

High PEEP yielded improvement in PaO2/FiO2 in both cohorts. The PEEP-induced improvement in PaO2/FiO2 was greater in COVID-19 patients than among subjects with non-COVID-19 ARDS: median PEEP-induced changes in PaO2/FiO2 were 42 mmHg [24–62] vs. 27 mmHg [5–42], respectively (p = 0.005). In 27 (97%) patients of COVID-19 cohort, high PEEP yielded improvement in PaO2/FiO2.

In both cohorts, PEEP did not affect PaCO2 nor ventilatory ratio. In COVID-19 patients, respiratory system compliance and driving pressure did not change with high PEEP: in ARDS from non-COVID-19 etiology, high PEEP reduced compliance by 5 ml/cmH2O [CI95% 2–8] and increased driving pressure by 2 cmH2O [CI95% 1–3].

In both cohorts, high PEEP caused reduction in systolic arterial pressure, without affecting heart rate and diastolic pressure.

Response to PEEP—Recruitability

Median recruitment-to-inflation ratio (i.e., recruitability) was 0.73 [0.43–1.04] in COVID-19 patients and 0.55 [0.35–1.08] in ARDS from other causes (p = 0.39). Inter-individual variability of the recruitment-to-inflation ratio was 72% and 64%, respectively. Recruitment-to-inflation ratio was greater than 0.5 in 22 COVID-19 patients (73%) vs. 17 patients (57%) from the non-COVID-19 ARDS cohort (p = 0.28).

In COVID-19 patients, recruitment-to-inflation ratio had a weak but significant inverse correlation with respiratory system compliance recorded at low PEEP (r = − 0.37, p = 0.04). Recruitment-to-inflation ratio had no relationship with the change in PaO2/FiO2 caused by high PEEP (p = 0.45) but had a reverse linear relationship with the change in PaCO2 due to PEEP (r = − 0.46 p = 0.010) (Figure 4).

Recruitment-to-inflation was not different between patients who showed increased, decreased, or unchanged respiratory system compliance (and driving pressure) with high PEEP (p = 0.14).

Clinical Outcome

At 28 days, 19 (63%) patients in the COVID-19 cohort had died and 3 (10%) were still on mechanical ventilation.