[ Jeremy R. Beitler is a clinical research fellow in pulmonary and critical care medicine at Brigham and Women's Hospital and Harvard Medical School. His research in Dr. Malhotra's laboratory focuses on respiratory mechanics and cardiopulmonary interaction during permissive hypercapnia for ARDS. Rolf D. Hubmayr is a professor of physiology and medicine and director of the critical care service at Mayo Clinic. He also serves as chair of the NHLBI ARDS Network pathogenesis committee. His research focuses on mechanisms and biologic sequelae of lung deformation in ventilator-induced lung injury and ARDS. Atul Malhotra is an associate professor in pulmonary, critical care, and sleep medicine at Brigham and Women's Hospital and Harvard Medical School. He also serves as secretary-treasurer of the American Thoracic Society. His research interests include respiratory mechanics in ARDS and metabolic/cardiovascular complications of sleep disorders.]
Current ventilator strategies for acute respiratory distress syndrome (ARDS) aim to impact lung function and clinical outcomes in several ways: low tidal volumes to minimize overdistension, titrated positive end-expiratory pressure (PEEP) to prevent alveolar derecruitment, and recruitment manoeuvres to promote parenchymal homogeneity. Among these interventions, only low tidal volume has been shown definitively to improve mortality from ARDS (Malhotra, 2007). During low tidal volume ventilation, practice varies substantially on whether to allow some degree of alveolar hypoventilation with incidental hypercapnic acidosis (Amato et al. 1998), or to increase respiratory rate to maintain alveolar ventilation, often requiring respiratory rates >30 breaths min-1 (Brower et al. 2000). As the independent effect of hypercapnia has never been tested in humans, it remains unclear whether such permissive hypercapnia has protective or deleterious effects in ARDS.
Two major questions surrounding permissive hypercapnia include (1) whether the mechanical effects of limiting respiratory rate per se yield clinical benefit, and (2) what the effects are of hypercapnia, independent of minute ventilation, on end-organ injury.
Low tidal volumes, which may contribute to hypercapnia, minimize mechanical lung injury. Ventilation at high tidal volumes causes alveolar overdistension, leading to alveolar epithelial and capillary endothelial cell junction breaks, cell detachment from the basement membrane, intracapillary blebs, and alveolar and interstitial oedema (Fu et al. 1992). Further contributing to regional wall stress are shear forces created by inflation of normal alveoli adjacent to atelectatic or flooded alveoli (Perlman et al. 2011). Additionally, atelectrauma occurs from cyclic recruitment and collapse of atelectatic lung units during respirations. Resultant formation and destruction of foam bubbles contributes interfacial stress that disrupts intracellular plasma membrane–cytoskeletal adhesions, leading to bleb formation and plasma membrane disruption (Oeckler et al. 2010). Low tidal volumes may reduce atelectrauma if lung units with high opening pressures remain closed throughout the respiratory cycle.
In response to deformation-induced cell strain, alveolar epithelial cells rapidly translocate lipid molecules to the plasma membrane to increase cell surface area and reduce strain, thereby minimizing risk of membrane rupture and facilitating repair when injury does occur (Vlahakis et al. 2002). Alveolar overdistension and cyclic atelectasis also induce extensive cytokine release that contributes further to end-organ injury (Imai et al. 2003). Low tidal volume ventilation directly reduces cyclic overdistension and thus risk of lung injury.
Additionally, limiting end-inspiratory plateau pressure is used commonly to minimize barotrauma. However, airway pressures do not delineate the often substantial contribution of the chest wall to respiratory system mechanics (Talmor et al. 2006). Plateau pressures may underestimate alveolar distension in patients with negative pleural pressures, such as during spontaneous respirations. In a recent study demonstrating reduced mortality in severe ARDS with neuromuscular blockade (Papazian et al. 2010), the incidence of barotrauma halved with paralysis compared to placebo despite similar initial airway plateau pressures. This may suggest that high transpulmonary pressures (defined as the pressure difference between the airway opening and pleural space) during spontaneous respirations worsened lung injury. Similarly, in a preclinical model of ARDS, lung injury worsened despite unchanged plateau pressures when spontaneous breathing generated high transpulmonary pressures (Yoshida et al. 2012). To account for patient differences in pleural pressures and chest wall mechanics, some advocate tailoring mechanical ventilation in ARDS to transpulmonary pressure by estimating pleural pressure with oesophageal manometry (Talmor et al. 2008).
In addition to reducing tidal stretch, limiting respiratory rate may prevent biomechanical injury by reducing the frequency of cyclic recruitment and collapse of damaged alveoli and of repetitive shear stress in the heterogeneously diseased lung. The cumulative effects of such repetitive stress may lead to frequency-dependent membrane failure known as fatigue behaviour (Marini et al. 2003), analogous to bending a paperclip back and forth repeatedly to cause its eventual breakage. Indeed, lower respiratory rate reduces lung injury in preclinical models of ARDS (Hotchkiss et al. 2000). Moreover, dynamic hyperinflation may occur with high respiratory rate, increasing risk of regional overdistension (Marini, 2011).
However, it is entirely unclear whether resultant hypercapnia offsets or augments the advantageous biomechanical effects of low tidal volume and respiratory rate. Among the proposed additional benefits, hypercapnia attenuates oxygen free radical formation, NF-κB activation, and TNFα, IL-1, IL-6 and IL-8 cytokine production (Taylor & Cummins, 2011) increases sympathetic tone with associated haemodynamic improvements (Wang et al. 2008), and theoretically reduces the oxygen delivery required to meet cellular energetic demand (Hillered et al. 1984).
Many of these theoretical benefits of hypercapnia may be deleterious when translated into the clinical realm. The immunomodulatory effects of hypercapnia delay bacterial clearance and worsen lung injury in pulmonary sepsis models (O’Croinin et al. 2008), of particular concern because pneumonia is a leading risk factor for ARDS (Gajic et al. 2011). Permissive hypercapnia's adrenergic effects should be viewed with caution in light of recent data demonstrating increased morbidity and mortality in ARDS with β-agonist therapy from unclear mechanisms (Matthay et al. 2011; Gao Smith et al. 2012). Ongoing β3 adrenergic receptor stimulation may also lead to progressive deterioration in cardiac function (Gauthier et al. 1996). Moreover, hypercapnic pulmonary vasoconstriction augments hypoxic pulmonary vasoconstriction and may worsen right heart strain (Mekontso Dessap et al. 2009). The reduced oxygen consumption from hypercapnia is at least partly due to mitochondrial dysfunction, shown to impair alveolar epithelial cell proliferation (Vohwinkel et al. 2011).
Hypercapnia additionally may have deleterious effects directly on the injured lung. While most alveolar cells survive deformation-induced injury by translocating lipids to the injured plasma membrane (Vlahakis et al. 2002), hypercapnic acidosis impairs plasma membrane wound healing in ventilator-injured lungs in a pH-dependent fashion (Caples et al. 2009). Hypercapnia, independent of pH, also impairs alveolar fluid reabsorption by inhibiting epithelial Na+-K+-ATPase (Briva et al. 2007).
Finally, the cardiopulmonary response to hypercapnia may lead further to harm. Heightened ventilatory drive during hypercapnic acidosis may require deeper sedation or paralysis to prevent patient–ventilator dyssynchrony, excess tidal excursions and double-triggering of the ventilator. If not blunted, this heightened ventilatory drive may generate more negative pleural pressures (Yan et al. 1993), thereby increasing transpulmonary pressure and alveolar stress. Enhanced pulmonary blood flow via hypercapnia-induced catecholamine release may worsen alveolar damage, capillary leak and oedema formation (Broccard et al. 1998). Even during moderate alveolar overdistension insufficient to cause lung injury alone, increasing pulmonary blood flow leads to lung injury that otherwise would not occur (Guery et al. 1998). The combined effects of occult increased transpulmonary pressure and increased pulmonary blood flow from permissive hypercapnia may create precisely this environment wherein occult lung injury is likely to occur.
As investigation of permissive hypercapnia transitions to humans, focus must be placed on distinguishing the effect attributable to minimizing biomechanical injury from that attributable to hypercapnia itself. Whether the independent effect of hypercapnia per se is harmful or protective is unknown. At the very least, there is sufficient concern to give pause. Most data suggesting improved outcomes associated with permissive hypercapnia are probably attributable primarily to the mechanical benefits of limiting tidal volume and respiratory rate.
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 http://jp.physoc.org/letters/submit/jphysiol;591/11/2767
Conflict of interest
J. R. Beitler and R. D. Hubmayr have no conflicts of interest to declare. A. Malhotra previously received consulting and/or research income from Philips, SGS, SHC, Apnex, Apnicure and Pfizer, but has relinquished all outside personal income since May 2012.