DOI: 10.1097/01.CCM.0000269404.14491.AE
,
,
PMID: 17581388
Issn Print: 0090-3493
Publication Date: 2007/07/01
Facebook
Twitter
LinkedIn
Email
 |
Checking for direct PDF access through Ovid | |
Excerpt
First, for the five studied sheep, lung lavage injury was described as diffuse in three vs. localized, gravity-dependent, severe, and heterogeneous in two. We contend that the diffuse disease group more closely approximates the clinically encountered diffuse and heterogeneous disease characterized by a distribution of mechanical time constants. Alternatively, the gravity-dependent severe disease sheep are probably not representative of clinical lung injury. The authors’ description of the lung lavage model is incomplete on two counts: 1) the number of lavage repetitions in the individual sheep; and 2) the recovered-to-instilled fluid volumes with each lavage. We suspect that the described gravity-dependent lung injury form may be an experimental artifact of the imperfect passive drainage method used to recover the lavage fluid. This resulted in flooded vs. aerated lung regions, that is, a case of extreme heterogeneity approximated by two compartments with markedly different mechanics. The implications of this scenario on the perceived advantages of obtaining dynamic respiratory mechanics measurements for optimizing PEEP may be significant.
Second, we found the peak-to-peak ventilation pressure (VPpp) vs. PEEP data in Figure 5C to be particularly intriguing. Specifically, VPpp appears more systematically sensitive to PEEP effects than either EVW-derived variables (Elow and Rhet; Fig. 7) (1). Also, on visual inspection, VPpp seems highly correlated with the computed tomography-derived pressure-volume data (Fig. 2) (1). The authors did not provide the VPpp correlation to the computed tomography-derived lung disease heterogeneity variable (HUCV) as they did for Elow (R = 0.59) and Rhet (R = 0.70) (1). Readers are not told whether VPpp was measured using standard ventilator waveforms or EVW. If the preceding is true, and VPpp was SVW-derived, one may legitimately question whether EVW-derived variables offer new, clinically relevant information. Alternatively, if VPpp was derived via EVW, the question changes to whether the additional steps to partition the lung mechanical properties are even needed. Therefore, a comparison of VPpp derived via standard ventilator waveforms and EVW and how each correlates to HUCV are worthwhile additional considerations. Ironically, the authors’ proposed EVW technology would be even more appealing to users if EVW-derived VPpp proves to be the superior clinical variable.
Third, the absence of preinjury mechanics and computed tomography scan data in this article was noteworthy. Pre- vs. postlavage mechanics comparisons could shed more light on the nature of the lung injury. More, it would have been informative to ascertain whether the postinjury mechanics for the 15.0–17.5 cm H2O PEEP setting—where the trade-off between recruitment and overdistension is optimized—approached (and/or correlated with) the preinjury lung mechanics more closely than other settings.
We thank the authors for their consideration of our comments and congratulate them on their fine article.
