Transpulmonary pressures in obese and non-obese COVID-19 ARDS

Background Data on respiratory mechanics of COVID-19 ARDS patients are scarce. Respiratory mechanics and response to positive expiratory pressure (PEEP) may be different in obese and non-obese patients. Methods We investigated esophageal pressure allowing determination of transpulmonary pressures (PL ) and elastances (EL) during a decremental PEEP trial from 20 to 6 cm H2O in a cohort of COVID-19 ARDS patients. Results Fifteen patients were investigated, 8 obese and 7 non-obese patients. PEEP ≥ 16 cm H2O for obese patients and PEEP ≥10 cm H2O for non-obese patients were necessary to obtain positive expiratory PL. Change of PEEP did not alter significantly ΔPL or elastances in obese patients. However, in non-obese patients lung EL and ΔPL increased significantly with PEEP increase. Chest wall EL was not affected by PEEP variations in both groups.


Background
Obesity, which is usually associated with better outcome for acute respiratory distress syndrome (ARDS) patients, is considered as a risk factor of acquiring a severe form of SARS COV-2-associated ARDS for SARS-CoV-2 [1]. Impact of obesity on respiratory mechanics of SARS-CoV-2-associated ARDS has not been investigated.
We hypothesized that respiratory mechanics including esophageal pressure (Pes) measurements might be different in obese and non-obese patients.
Therefore, the first objective of this study was to investigate transpulmonary pressures (P L ) in intubated SARS-CoV-2 patients according to their body mass index (BMI) during a decremental PEEP trial. Secondary objective was to assess lung and chest wall elastances (EL L and EL CW, respectively).

Patients
We conducted a prospective observational study in two intensive care units both in tertiary university hospitals (Hôpital de la Croix-Rousse, Hospices Civils de Lyon and Hôpital Nord, Assistance Publique-Hôpitaux de Marseille).
Patients were included in the study from 15 March to 15 April 2020 if they fulfilled inclusion criteria: adult admitted into the ICU for SARS-CoV-2, intubated and mechanically ventilated with moderate-to-severe ARDS criteria, sedated and paralyzed for clinical purpose and monitored by Pes catheter. As part of routine clinical management, we performed a decremental PEEP trial from 20 to 6 cm H 2 O by 2 cm H 2 O-steps in each patient during volume-controlled ventilation while other parameters were kept constant.
confirmed by presence of cardiac artifacts on the esophageal curve and by an occlusion test (expiratory hold on the ventilator) in passive conditions with gentle chest compression. The occlusion test was considered as positive when the correlation between ∆P es and ∆ airway pressure (Paw) was 0.8-1.2. To avoid overestimation or underestimation of esophageal pressures, we inflated the esophageal balloon with the minimal filling volume among the recommended range for each catheter which was within the flat portion of the volume-pressure curve of the balloon [2,3].
At each PEEP step, 2-s end-inspiratory occlusion pause allowed measurement of respiratory system (RS) and esophageal plateau pressure (Pplat and P es, insp , respectively), whereas 5-s end-expiratory occlusion pause allowed, respectively, measurement of RS and esophageal total PEEP (PEEP tot and P es,exp respectively). RS driving pressure (∆P RS ) was calculated as Pplat minus PEEP tot. RS, chest wall and lung elastances (EL RS , EL CW , and EL L , respectively) and elastance ratio were computed according standard formula. P L absolute values were calculated as airway pressures minus esophageal pressures during inspiration (P L,insp = Pplat-P es,insp ) and expiration (P L, exp = PEEP tot -P es,exp ), transpulmonary driving pressure (∆P L ) = P L,insp -P L,exp . Elastance ratio derived P L (P L,ER ) was calculated as Pplat x (EL L /EL RS ).

Statistics
Obesity was defined by a BMI ≥ 30. Results are reported as medians [interquartile range] or count (percentage) and compared between groups by Mann-Whitney U. Friedman test was used for repeated variables. p value < 0.05 was considered as significant.

Results
Fifteen patients were included in the study, 8 in the obese group (median BMI 34 [33-41]) and 7 in the non-obese group   Figure 1a represents Pplat and ∆P RS for each BMI group according to the PEEP levels. Figure 1b represents transpulmonary pressures and ∆P L . PEEP ≥ 16 cm H 2 O for obese patients and PEEP ≥ 10 cm H 2 O for non-obese patients were necessary to obtain positive P L, exp (Fig. 1b). At 16 cm H 2 O of PEEP, 71% of non-obese patients had P L , insp ≥ 20 cm H 2 O and 0% of obese patients, whereas with P L, ER was ≥ 20 cm H 2 O in, respectively, 86% of non-obese patients and 75% of obese patients. Change of PEEP did not alter significantly ∆P L or elastances in obese patients (Table 1). However, in non-obese patients EL RS and EL L increased significantly with PEEP increase. EL CW was not affected by PEEP variations in both groups.
Differences between obese and non-obese groups were significant at 18-20 cm H 2 O of PEEP with higher Pplat, ∆P RS, P L,insp , ∆P L in non-obese patients.

Discussion
During decremental PEEP trial, we found differences in transpulmonary pressures and respiratory mechanics in COVID-19 ARDS patients according to the presence of obesity. First, P L,insp , ∆P L were higher in non-obese patients at high PEEP (≥ 18 cm H 2 O), as Pplat and ∆P RS . Second, EL CW and EL L were not statistically different between groups. However, increase of PEEP was significantly associated with an increase of EL L in non-obese patients. Third, high PEEP levels (i.e., 16 cm H 2 O) were associated with potential injurious P L,ER (≥ 20 cm H 2 O).
Preliminary studies with CT-scan have reported diffuse and bilateral pulmonary lesions during COVID-19 ARDS, which is usually associated with recruitability by PEEP in ARDS [4].
Recent studies using the same method to assess recruitability (recruitment-to-inflation ratio) reported conflicting data in those patients [5][6][7] with range from 17 to 56% of highly recruitable patients and possible less recruitable patients at a more advanced time point of the disease [5].
However, the lower driving pressures (∆P RS and (∆P L ) observed for obese patients cannot discriminate lung recruitment from less lung overdistension without appropriate evaluation (CT-scan or electrical impedance tomography for instance).
We did not check for airway flow limitation in all patients that could have led to overestimate lung and respiratory system elastances, in particular in obese patients.
Finally, we were not able to compare respiratory mechanics of this cohort with non COVID-19 ARDS patients.
In conclusion, assessment of respiratory mechanics of COVID-19 ARDS patients with transpulmonary pressure monitoring might be useful when targets of protective lung ventilation could not be reached. The characteristics of obesity on respiratory mechanics airway opening pressure, recruitability of COVID-19 ARDS patients need further investigations.