During the past decade of research on acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) there has been heightened interest in measuring physiologic dead-space fraction (VD/VT). Since ARDS first was described 44 years ago, the primary focus on gas exchange abnormalities concerned oxygenation. In fact, until the publication in 2002 by Nuckton et al,1 pulmonary dead-space ventilation was considered relevant only in the subacute or fibroproliferative phase.2 This impression probably reflected preoccupation with the problem of improving oxygenation. In contrast, it was relatively easy to achieve normocapnia simply by raising minute ventilation. In retrospect, the advent of lung-protective ventilation and permissive hypercapnia focused more attention on the issue of impaired carbon dioxide excretion in patients with ALI/ARDS.
Interest in measuring VD/VT grew rapidly following publication of the ARDS Network ARMA study3 and the aforementioned study by Nuckton et al.1 Lung-protective ventilation was the first therapy ever shown to reduce mortality in ARDS. Moreover, elevated VD/VT can be a sign of lung over-distention.4 A markedly elevated VD/VT can develop early in the course of ARDS, and it has a strong association with higher mortality. One issue involved how much of the elevated pulmonary dead space reflected an injurious mechanical ventilation strategy or more severe lung injury.
Since 2002, numerous studies have explored the prognostic, pathophysiologic, and therapeutic utility of measuring VD/VT in ALI/ARDS. Its value in predicting mortality has been reaffirmed by several studies.5–8 The specific value of measuring VD/VT to increase our understanding of the pathophysiology of ARDS is based upon carbon dioxide’s relatively high diffusability across tissue membranes, compared to oxygen.9 Thus VD/VT is a more perfusion-sensitive variable that may be useful as an indirect marker for pulmonary endothelial injury. In ALI/ARDS, endothelial injury is associated with coagulation disorders and fibroproliferation, both of which are associated with increased mortality.10 For example, a recent study found that elevated VD/VT, coupled with biomarkers of pulmonary endothelial injury, was a strong predictor for mortality in patients with ALI/ARDS.11
Other recent studies have used VD/VT to assess the therapeutic impact of pharmacologic, positional, and ventilator interventions. In the current issue of the Journal, Raurich and colleagues12 report that VD/VT was elevated markedly in the subacute phase of ARDS, and decreased significantly in response to treatment with methylprednisolone. These findings provide indirect evidence regarding the mechanism by which methylprednisolone improves pulmonary function in ARDS: namely, its ability to decrease both collagen formation and platelet adhesion that in part may be responsible for pulmonary microcirculatory obstruction.13 In addition, Raurich et al’s findings of reduced VD/VT are similar to those who reported the effects of activated protein C in patients with ALI/ARDS,14,15 as well as those studying the effects of heparin therapy on microvascular thrombosis in patients undergoing cardiac surgery.16
Measurement of VD/VT also has been valuable in assessing the effects of lung recruitment, both in patients with severe ARDS17–20 and in animal models of ALI.21,22 Of particular interest is that measuring VD/VT may be superior to oxygenation indices in assessing lung recruitment.19 Likewise, deterioration in VD/VT is more sensitive than oxygenation indices in detecting lung de-recruitment during a PEEP decrement trial.21
Previously we had speculated that an elevated VD/VT in ALI/ARDS in part represents pulmonary endothelial damage and reflects the severity of both pulmonary arterial and microvascular obstruction.1,5 However, others23,24 have correctly pointed out that the carbon dioxide-based dead-space measurement is sensitive to any ventilation-perfusion imbalance. Moreover, this ventilation-perfusion ratio imbalance appears skewed, particularly toward the mid-range, as opposed to the high-range and pure dead-space ventilation. Although this does not negate the likely contribution of vascular injury, it does emphasize that these lung abnormalities do not account for all of the increase in the VD/VT in ALI/ARDS patients. One could reasonably conclude that VD/VT is just more representative of overall lung damage and subsequent mortality risk, regardless of the relative distribution of physiologic and pathogenetic factors that contribute to it. On this issue, others appear to be in agreement also.22–24
Over 2 decades ago, Murray and colleagues25 created the first quantitative measurement of lung injury severity. This score integrated abnormalities in respiratory-system compliance, oxygenation, and chest radiograph with the level of PEEP. Although the lung injury score has been valuable in quantifying the severity of lung dysfunction, it has not been particularly useful in predicting mortality.26 The cumulative evidence reviewed here is remarkably consistent in demonstrating the value of VD/VT in both the study and management of ARDS. Therefore, we should perhaps consider updating the 4-point lung injury score to incorporate VD/VT, making it a 5-point scoring system. The expectation would be that this modification may provide a more complete assessment of injury severity and may improve its overall prognostic value. A reasonable scoring scheme for VD/VT might be: ≤ 0.39 (0), 0.40–0.49 (1), 0.50–0.59 (2), 0.60–0.69 (3), ≤ 0.70 (4).
Although variation in measurement technique presents some important technical considerations (most notably controlling for the effects of compression volume contamination),27 these are relatively minor. Regardless, both volumetric capnography and indirect calorimetry now are readily available in most intensive care units. Therefore, measuring VD/VT is practical and could be incorporated in assessing patients with ALI/ARDS in both academic and community based intensive care units.
The past decade of research on dead-space ventilation has been impressive in the new insights it has generated about both the pathophysiology of ALI/ARDS and the utility of measuring VD/VT for assessing new therapies. We eagerly look forward to the possibility that the next 10 years of research on dead-space ventilation will further enrich our understanding.