The distribution of ventilation during bronchoconstriction is patchy and bimodal: A PET imaging study

https://doi.org/10.1016/j.resp.2005.05.023Get rights and content

Abstract

Recent PET imaging data from bronchoconstricted sheep (Vidal Melo et al., 2005) showed that V˙/Q˙ distributions were bimodal and topographically patchy, but including a substantial heterogeneity at scales <2.2 ml. In this paper, we reanalyze the experimental data to establish the contribution of ventilation (V˙r) heterogeneity to the bimodality in V˙/Q˙. This analysis demonstrates that the distribution of V˙r during bronchoconstriction was bimodal with large patches of severe hypoventilation occupying an average of 41% of the imaged lung. The degree of hypoventilation to these regions was highly correlated with the degree of oxygenation impairment, but was quite variable amongst animals in spite of consistent degrees of mechanical obstruction. Remarkably, those regions were found to be hyperventilated before methacholine and their degree of hyperventilation was correlated with their degree of hypoventilation during bronchoconstriction. These data suggest that improving the uniformity of ventilation at baseline may be a desirable therapeutic target if the risk of severe hypoxemia during asthma attacks is to be minimized and/or the distribution of inhaled pharmaceuticals is to be optimized.

Introduction

V˙/Q˙ distribution in asthmatics and bronchoconstricted animals has been shown to be bimodal (Rubinfeld et al., 1978, Wagner et al., 1978, Wagner et al., 1987) and regionally patchy (de Siqueira et al., 1997, Horsley et al., 1985, King et al., 1998). In a recent report, we presented PET derived V˙/Q˙ distributions obtained from bronchoconstricted sheep (Vidal Melo et al., 2005). These V˙/Q˙ distributions were bimodal and showed a patchy distribution of regional ventilation (V˙r). Further analysis demonstrated a large fraction of the measured heterogeneity in V˙/Q˙ at sub-resolution levels (<2.2 ml) and that sub-resolution heterogeneity contributed significantly to the bimodality in V˙/Q˙ in all animals studied. Although it is expected that the V˙/Q˙ heterogeneity caused by bronchoconstriction should be mostly caused by V˙r heterogeneity, our previous analysis did not determine the extent to which the reported heterogeneity and bimodality of the V˙/Q˙ distribution was due to spatial heterogeneity in V˙r or Q˙r, nor the contribution of the observed patchiness in ventilation to global heterogeneity in V˙r. It also remained unclear whether regions of reduced ventilation during inhaled methacholine had any recognizable predisposition for bronchoconstriction as compared to the other lung regions. In this paper, we reanalyze the previous experimental data to establish the contribution of V˙r heterogeneity to the bimodality in V˙/Q˙ and to characterize the distributions of V˙r and Q˙r inside and outside the regions of hypoventilation. In addition, we test whether the regions of reduced ventilation following methacholine induced bronchoconstriction had different V˙r,Q˙r or lung aeration than the rest of the lung prior to the delivery of methacholine.

Section snippets

Methods

We analyze data from six normal sheep weighing 17 kg (range 15–21 kg) with a study protocol described in detail (Vidal Melo et al., 2005). The protocol was approved by the Committee on Animal Care of the MGH. Briefly, the animals were anesthetized, intubated, mechanically ventilated and placed in the prone position. Mechanical ventilation was set at an inspired oxygen fraction (FIO2) = 0.49 ± 0.02, positive end-expiratory pressure (PEEP) = 5 cmH2O, tidal volume (VT) = 17 ± 2 cm3/kg and inspiratory time of

Results

Significant bronchoconstriction was achieved in all animals. Ppeak during methacholine inhalation was 2.4 ± 04 times that at baseline, as required by protocol design. Also, arterial partial pressure of O2 was significantly reduced and alveolar–arterial PO2 gradient ((Aa)PO2) and PaCO2 was significantly increased during bronchoconstriction (Table 1).

PET imaging scans showed nearly complete tracer washout in control conditions and significant amounts of residual tracer remained in the lungs at the

Discussion

In a previous analysis of this set of data, we demonstrated that in this animal model of severe bronchoconstriction the distribution of V˙/Q˙ was clearly bimodal (Vidal Melo et al., 2005). We also showed that the bimodality was in great part the result of heterogeneity in V˙/Q˙ at length scales smaller than the spatial resolution of our imaging instrument (2.2 cm3) and that assessment of such “sub-resolution” heterogeneity was not important in normal lungs but was critical for accurate

Acknowledgments

This work was funded by NIH grant HL068011. T. Schroeder was supported in part by the German Academic Exchange Service (DAAD).

References (14)

There are more references available in the full text version of this article.

Cited by (0)

View full text