February 27, 2019

EJRC - Esophageal Manometry for the estimation of Transpulmonary Pressure

Is esophageal pressure a valid surrogate of pleural pressure to estimate transpulmonary pressure in the clinical setting?


Transpulmonary pressure (PL) is a very valuable parameter in respiratory pathophysiology, as it allows the physician to separate the pressure delivered to the lung from the one acting on the chest wall in mechanically ventilated patients. Accordingly, a physiologically-based ventilation strategy should take the PL into account.

Although PL can be estimated at the bedside by measuring esophageal pressure (Pes) with a minimally invasive catheter, recent studies reveal it is hardly used in everyday practice [1,2]. Some concerns might limit a widespread use of this technique: 1) the validity of the esophageal pressure measurement itself; 2) the existence of two different approaches to estimate pleural pressure (Ppl) and, hence, PL (both requiring the measurement of Pes: one based in the absolute value of Pes [PL,es], and the other one based on the elastance ratio of chest wall to respiratory system [PL,ER]); and 3) a large vertical gradient of pleural pressure, particularly in heterogeneous lungs.

To shed light on this questions, Yoshida and collaborators designed a proof of concept study [3] in which inspiratory and expiratory estimates of PL derived from Pes were compared with the actual value of PL, in dependent and nondependent lung, by directly measuring Ppl with surgically-inserted pleural sensors in six anesthetised pigs and three human cadavers (respiratory mechanics comparable to values in patients with lung injury).

Electrical impedance tomography (EIT) data were recorded to estimate lung collapse and computed tomography (CT) scans were performed to estimate the superimposed pressure between pleural sensors.

Vertical Ppl gradient increased after induction of lung injury in the animal model at all PEEP levels (1.8-fold greater in injured vs normal lungs at PEEP of 10 cmH2O), and was 10±3 cmH2O in cadavers (at PEEP of 10 cmH2O).

Expiratory PL calculated from Pes reflected the actual values in mid and dependent lung regions at higher and lower PEEP values, respectively, in the animal model (correlation between measured PL in dependent zones and estimated PL: R2=0.94, p<0.01); and the actual values in midlung regions in the human model. CT scan images indicate that Pes reflects the local Ppl, adjacent to the esophageal balloon. EIT showed that PL just above zero was insufficient to keep the injured lung open (17% atelectasis at PEEP of 8 cmH2O, corresponding to PL just above 0 cmH2O).

Inspiratory PL calculated from Pes was well correlated with actual PL in non-dependent and dependent areas (R2=0.61, p<0.01 and R2=0.78, p<0.01, respectively, for PL,es; R2=0.67, p<0.01 and R2=0.66, p<0.01, respectively, for PL,ER). Nevertheless, Pes-derived PL better reflects the actual PL of the dependent to middle lung, whereas elastance ratio-derived PL better reflects the non-dependent lung.



Strengths of the study comprise the relevance of the study purpose (supporting the feasibility of esophageal manometry for the estimation of transpulmonary pressure), a good methodological approach (comparison of the technique with the gold standard, both in an animal and human model, and testing the two available methods for PL calculation), and the use of complementary examinations for a better physiological understanding of the findings (CT, EIT).

Limitations include a low sample size (thus, limiting external validity of the results), absence of randomisation (which is reasonable, given the type of study) and assumptions for PL,ER calculation which may not be true in all patients with ARDS (considering that Ppl is zero at functional residual capacity, and extrapolation of elastance ratio of chest wall from tidal breaths).


  • Esophageal manometry allows estimation of transpulmonary pressure in the clinical setting and is a useful monitoring tool for a physiologically-based ventilation strategy.
  • According to this study, the absolute values of Pes are accurate, and can reasonably reflect local PL for clinical purposes, if calibrated properly.
  • The vertical gradient of pleural pressure is increased in lung injury, so a protective ventilation targeting global parameters (VT, Pplat or PEEP) may not protect all the regions at the same time.
  • A personalised approach for lung protection may be feasible by using expiratory PL,es (to minimise atelectasis in dependent lung) and inspiratory PL,ER (to minimize overdistension in non-dependent lung).

This article review was prepared and submitted by ESICM NEXT member David Pérez Torres, Department of Intensive Care, Río Hortega University Hospital, Valladolid, Spain, on behalf of the ESICM Journal Review Club.

#NAHP #NEXT #ERJC #ICU #MechanicalVentilation #MV #RespiratoryMonitoring


1)  Akoumianaki, E. et al. (2014). The application of esophageal pressure measurement in patients with respiratory failure. American Journal of Respiratory and Critical Care Medicine, 189(5), 520-31.

2) Mauri, T. et al. (2016). Esophageal and transpulmonary pressure in the clinical setting: Meaning, usefulness and perspectives. Intensive Care Medicine, 42(9), 1360-1373.

3) Yoshida, T. et al. (2018). Esophageal Manometry and Regional Transpulmonary Pressure in Lung Injury. American Journal of Respiratory and Critical Care Medicine, 197(8), 1018-1026.

Comment on this news