Driving pressure in obese ventilated patients: another brick in the (chest) wall
The incidence of obesity is increasing worldwide, and obese or severely obese patients often present with several comorbidities. Obesity is associated with significant complications, and an increase in the mortality of obese patients after hospital admission compared to non-obese patients is justified . The proportion of obesity in patients presenting with acute respiratory distress syndrome (ARDS) has not been investigated systematically, but data from the influenza A (H1N1) pandemic in 2009 suggested that obese patients were overrepresented compared to the general population . The management of ARDS in patients with obesity is a special challenge, since obesity and severe obesity are associated with marked derangements in lung and chest wall mechanics .
Characteristics of respiratory mechanics in obese mechanically ventilated patients (findings from clinical studies)
Static respiratory system compliance (Cst,rs) (mL cmH2O−1)
Lung compliance (Cst,L) (mL cmH2O−1)
Chest wall compliance (Cst,w) (mL cmH2O−1)
Pelosi 1996 
20 pts., general anaesthesia
BMI < 25 = 66 ± 14
BMI > 40 = 35 ± 5
107 ± 32
55 ± 15
191 ± 45
112 ± 47
Pelosi 1998 *
24 pts., general anaesthesia
BMI < 25 ≈ 75
BMI 25–40 ≈ 55
BMI > 40 ≈ 46
Normal findings in spontaneously breathing adult subjects 
Static respiratory system elastance (cmH2O L−1)
Lung elastance (cmH2O L−1)
Chest wall elastance (cmH2O L−1)
Driving pressure (∆P) (cmH2O)
Chiumello 2016 
101 ARDS patients
BMI < 25 = 25 (19–31)
BMI 25–30 = 27 (21–36)
BMI > 30 = 26 (24–32)
In a large retrospective analysis of the data of 3562 ARDS patients from nine studies, Amato et al.  demonstrated that elevated driving pressure (∆P), a parameter expressing the relation between tidal volume and functional lung size (compliance of the respiratory system) = VT/Cst,rs, was an independent ventilation variable strongly associated with increased mortality, while changes in tidal volume or positive end-expiratory pressure were not independent predictors of the outcome. An actual systematic review  reconfirmed the retrospective data of Amato et al.  by analysing the data of four prospective randomized studies (3252 patients): a higher ∆P was associated with a significantly higher mortality in ARDS patients.
Surprisingly, this important finding seems not to be of value for obese ARDS patients. The retrospective analysis presented by de Jong et al.  on the impact of ∆P on 90-day mortality in 100 obese (BMI ≥ 30) and 262 non-obese ARDS patients revealed a significant relationship between ∆P and mortality only in non-obese patients. The ∆P on day 1 was significantly lower in non-obese ARDS patients (11.9 ± 4.2 cmH2O) who had survived 90 days compared to non-obese non-survivors (15.2 ± 5.2 cmH2O, p < 0.001). In obese patients, the ∆P did not differ between survivors and non-survivors. Additionally, in a multivariate Cox analysis, ∆P was identified as an independent predictor of day 90 mortality in non-obese, but not in obese patients. Interestingly the mortality did not differ between non-obese (47%) and obese ARDS (46%) patients. Although the current study has some limitations (the monocentric and retrospective design), an important finding was that the day 1 mean values for ∆P were similar in the survivors (13.7 ± 4.5 cmH2O) and in the non-survivors (13.2 ± 5.1 cmH2O, p = 0.41), and on a level which has been identified as being within a ‘safe’ border. Amato et al.  described a ∆P of 15 cmH2O as a threshold discriminating between a good outcome of ARDS patients (relative risk of death ≤ 1 with ∆P ≤ 15 cmH2O) and a higher risk of mortality with ∆P > 15 cmH2O. The study by de Jong et al.  presented ‘safe’ ∆P levels (≤ 15 cmH2O) for the majority of obese ARDS patients. Is obesity per se a lung protective precondition generating low(er) ∆P values? If, yes, which are the specific (patho)physiologic mechanisms for such a distinction from non-obesity?
The reduction in Cst,w is a characteristic finding in the respiratory mechanics of mechanically ventilated obese patients (Table 1), and this distinctiveness could be responsible for the lower ∆P values, as seen in the ‘obesity paradox’. A meta-analysis on the causal relation between obesity and mortality in ARDS  evaluating the combined data of nine studies found that obesity was significantly associated with a reduced risk of ARDS mortality (pooled OR 0.63, 95% CI 0.64–0.84). In consequence, further studies should carefully investigate the potential role of chest wall dynamics, pleural pressure and transpulmonary pressure in obesity with the aim of examining the hypothesis that adiposity offers a certain protection against ventilator-induced lung injury. Furthermore, the shape of obesity is various, and in recent years it was demonstrated that the pattern of body fat distribution (general or abdominal adiposity) has a great impact on outcomes. Reis et al.  reported that in middle-aged healthy persons, higher values of the waist-to-thigh ratio (in both sexes) or the waist-to-hip ratio (in women) were positively associated with increased mortality, while the BMI exhibited U- or J-shaped associations. In intensive care, no data exists on the pattern of obesity, and in future researchers should investigate not only BMI but also the effects of abdominal obesity on outcomes.
Are the results of the study by de Jong et al.  a setback for those who found in the parameter ∆P (easily assessed at the bedside!) a new tailor-made solution for an individualized lung protective ventilation strategy in all ARDS patients? No, quite the opposite! The data from de Jong et al. provides an important step forward to a personalized strategy in intensive care medicine . Again, it becomes clear that a simple ‘one variable fits all’ will not work in critical care.
Compliance with ethical standards
Conflicts of interest
The author declares no conflict of interest.
- 2.Morgan OW, Bramley A, Fowlkes A, Freedman DS, Taylor TH, Gargiullo P, Belay B, Jain S, Cox C, Kamimoto L, Fiore A, Finelli L, Olsen SJ, Fry AM (2010) Morbid obesity as a risk factor for hospitalization and death due to 2009 pandemic influenza A (H1N1) disease. PLoS One 5:e9694CrossRefPubMedPubMedCentralGoogle Scholar
- 10.de Jong A, Cossic J, Verzilli D, Monet C, Carr J, Conseil M, Monnin M, Cisse M, Belafia F, Molinari M, Chanques G, Jaber S (2018) Impact of the driving pressure on mortality in obese and non-obese ARDS-patients: a retrospective study of 362 cases. Intensive Care Med. https://doi.org/10.1007/s00134-018-5241-6