By Richard Kallet, MS, RRT, FCCM

Director of Quality Assurance, Respiratory Care Services, Department of Anesthesia, San Francisco General Hospital

Mr. Kallet reports he is a major stockholder in the Asthma & Allergy Prevention Company and receives grant/research support from Nihon-Kohden.

SYNOPSIS: Different ventilator modes used for a spontaneous breathing trial affected a patient’s work of breathing (WOB) variously and differed regarding WOB measured after extubation. The clinical relevance of these differences is uncertain.

SOURCE: Sklar MC, Burns K, Rittayamai N, et al. Effort to breathe with various spontaneous breathing trial techniques. Am J Respir Crit Care Med 2017;195:1477-1485.

This meta-analysis of 16 prospective clinical studies with 239 adult subjects examined the effect of six approaches to conducting a spontaneous breathing trial (SBT), which included: pressure support (PS) of 5-10 cm H2O, continuous positive airway pressure (CPAP) of 5-8 cm H2O, T-piece (T-P), mimicking T-P on the ventilator (i.e., PS = 0/CPAP = 0 cm H2O), automatic tube compensation, and neutrally adjusted ventilatory assist. The meta-analysis included studies that were randomized crossover (n = 8), non-randomized crossover (n = 7), and a parallel group randomized, controlled trial. The primary finding was that SBTs performed with either low level PS or CPAP produce a comparable effect on work of breathing (WOB). Moreover, an SBT with PS reduces WOB by 30% compared to T-P and 46% compared to post-extubation measurements. The authors concluded that performing an SBT using either T-P or its correlate while attached to a ventilator represents “the optimal method for evaluating weaning readiness.”


Sklar et al nobly attempted to determine if the mode used for SBTs produces meaningful differences in WOB that might affect the success of an extubation trial. Unfortunately, their conclusion is highly suspect. A meta-analysis is ill-suited for assessing these studies and drawing meaningful conclusions given the methodological vagaries related to measuring and assessing respiratory muscle function in patients recovering from acute respiratory failure.

The maximal sustainable WOB level in patients recovering from acute respiratory failure has never been established. Years ago, Petros et al successfully extubated patients from CPAP and T-P with WOB levels (1.13 and1.35 J/L, respectively) that previously were considered incompatible with successful weaning (i.e., > 0.75 J/L).1 Physiologic studies of WOB require a uniform baseline (i.e., measurements made at rest) to be meaningful; humans tend to breathe at 0.3-0.5 J/L. Yet, it’s precarious to assume these WOB levels accurately reflect what is encountered during activities of daily living, let alone provide particularly useful information for making clinical decisions.

Moreover, a meta-analysis cannot account for important differences in measurement techniques when an endotracheal tube is present and post-extubation (when a tight-fitting mask or mouth piece is necessary). Even setting aside technological issues related to measuring flow and volume between intubated and non-intubated conditions, it’s highly improbable that “breathing behavior” in a conscious subject with an artificial airway is the same as when a clinician is holding a mask over the patient’s face tightly or when a patient tries to hold a mouth piece firmly in place. This problem was illustrated nicely by an intriguing contradiction in one of the cited studies. Despite a cross-sectional glottic area almost three times greater than the endotracheal tube (140 mm2 vs. 50 mm2), inspiratory effort paradoxically increased by 33%, WOB increased by approximately 150%, and mean tidal volume decreased by 23%. Even in a meticulously executed study, vexing findings can result and likely reflect these inherent methodological limitations.

Loss of interest in measuring WOB coincided with the publication of numerous SBT studies reaffirming the long-held impression that at least 70% of patients successfully resume unassisted breathing by simply withdrawing mechanical support. One study cited in the meta-analysis reported that extending an SBT by one hour to evaluate patients while breathing through a T-P offered no advantage compared to a minimum PS trial (i.e., ~ 7 cm H2O). Moreover, a large randomized trial found a significantly higher SBT success rate with PS of 7 cm H2O vs. T-P (86% vs. 78%, respectively) without significant differences in those remaining extubated after 48 hours (70% vs. 63%, respectively).2 It’s important to note that patients often exhibit significant buildup of biofilm and/or a particularly tortuous upper airway anatomy to which the pliable endotracheal tube conforms over time. Using low-level PS provides some offloading for these occult factors and is a matter of clinical expediency. Thus, the authors’ conclusion that T-P is the “optimal method” for conducting a SBT should be greeted with skepticism.

Regarding weaning, it’s helpful to apply Occam’s razor, which can be stated as follows: Patients sufficiently recovered from acute respiratory failure breathe without assistance when their minute ventilation demand decreases to the point that their respiratory muscles can handle the increased workload imposed by residual abnormalities in chest mechanics.


  1. Petros AJ, Lamond CT, Bennett D. The Bicore pulmonary monitor. A device to assess the work of breathing while weaning from mechanical ventilation. Anesthesia 1993;48:985-988.
  2. Esteban A, Alia I, Gordo F, et al. Extubation outcome after spontaneous breathing trials with T-tube or pressure support ventilation. The Spanish Lung Failure Collaborative Group. Am J Respir Crit Care Med 1997;156:459-465.