Rebuttal from Alan R. Schwartz and Philip L. Smith



Butler et al. present an interesting alternative to the Starling resistor model to account for the phenomenon of negative effort dependence (Butler et al. 2013). This phenomenon is characterized by a decrease in flow with increasing inspiratory effort, which can occur to a variable degree in sleeping individuals. Negative effort dependence is a feature commonly observed in pressure–flow relationships in collapsible conduits, whether they be biological or otherwise, and is illustrated in the classical descriptions of flow limitation to which Butler and co-workers referred.

Negative effort dependence is usually characterized by a plateauing of airflow at a submaximal level, rather than complete cessation of airflow as described by Butler et al. This flow plateau, like the maximal flow, is still determined by the gradient between upstream and critical pressure, and falls progressively as this gradient approaches zero. In other words, the Starling resistor models the pressure–flow regimen and the severity of upper airway obstruction similarly for maximal or submaximal flows alike, encompassing the phenomenon of negative effort dependence.

Several explanations could account for the unusual example of negative effort dependence characterized by the cessation of flow, as Butler et al. described. Redundant tissue can intussuscept during inspiration, blocking the airway completely, much as the wall of a vein can block the bevelled opening of a venipuncture needle when negative pressure is applied. Alternatively, the epiglottis can rarely flip down and occlude the laryngeal aperture during inspiration. While bulk tissue movement may obstruct flow completely in these examples, it would not account for snoring, which is more typically observed during periods of inspiratory airflow limitation. Snoring, like wheezing and stridor, is produced by vibration in airway tissues as flow oscillates around its maximal level. Rather than occluding the airway completely, tissues oscillate as pressure is dissipated by turbulent flows through a narrowed or critical orifice.

Furthermore, the ‘lumped’ tissue model described by Butler et al. does not offer the broad applicability of the Starling resistor in modelling upper airway flow dynamics during sleep across a spectrum from health to disease (Schwartz & Smith, 2013). Nor does it account for the induction of upper airway obstruction and sleep apnoea syndrome in normal individuals breathing at subatmospheric nasal pressure during sleep. Indeed, negative effort dependence may be more of an epiphenomenon than a key to understanding the pathogenesis and clinical spectrum of upper airway obstruction during sleep.

Call for comments

Readers are invited to give their views on this and the accompanying CrossTalk articles in this issue by submitting a brief comment. Comments may be posted up to 6 weeks after publication of the article, at which point the discussion will close and authors will be invited to submit a ‘final word’. To submit a comment, go to;591/9/2235



This work was supported by NIH HL50381.