Lung regional metabolic activity and gas volume changes induced by tidal ventilation in patients with acute lung injury

Abstract

Rationale: During acute lung injury (ALI), mechanical ventilation can aggravate inflammation by promoting alveolar distension and cyclic recruitment-derecruitment. As an estimate of the intensity of inflammation, metabolic activity can be measured by positron emission tomography imaging of [(18)F]fluoro-2-deoxy-D-glucose.

Objectives: To assess the relationship between gas volume changes induced by tidal ventilation and pulmonary metabolic activity in patients with ALI.

Methods: In 13 mechanically ventilated patients with ALI and relatively high positive end-expiratory pressure, we performed a positron emission tomography scan of the chest and three computed tomography scans: at mean airway pressure, end-expiration, and end-inspiration. Metabolic activity was measured from the [(18)F]fluoro-2-deoxy-D-glucose uptake rate. The computed tomography scans were used to classify lung regions as derecruited throughout the respiratory cycle, undergoing recruitment-derecruitment, and normally aerated.

Measurements and main results: Metabolic activity of normally aerated lung was positively correlated both with plateau pressure, showing a pronounced increase above 26 to 27 cm H(2)O, and with regional Vt normalized by end-expiratory lung gas volume. This relationship did not appear to be caused by a higher underlying parenchymal metabolic activity in patients with higher plateau pressure. Regions undergoing cyclic recruitment-derecruitment did not have higher metabolic activity than those collapsed throughout the respiratory cycle.

Conclusions: In patients with ALI managed with relatively high end-expiratory pressure, metabolic activity of aerated regions was associated with both plateau pressure and regional Vt normalized by end-expiratory lung gas volume, whereas no association was found between cyclic recruitment-derecruitment and increased metabolic activity.

Figures

Figure 1.

Representative images obtained by computed tomography (CT; gray scale on left) obtained at end-expiration, end-inspiration, and by positron emission tomography (PET; the image shows the last frame acquired during dynamic imaging, between 47 and 57 min after injection of the bolus of [18F]fluoro-2-deoxy-D-glucose [18FDG], gray scale on right). After an automated preidentification, we visually confirmed the presence of regions derecruited both at end-expiration and end-inspiration (blue line) and regions undergoing cyclic recruitment–derecruitment (red line). End-expiratory lung volume (EELV) of the normally aerated tissue (green line) and of the poorly aerated tissue (yellow line) were calculated on the CT scan obtained at end-expiration, whereas the Vt distending these regions was computed as the gas volume difference between the end-expiratory and end-inspiratory scan. These regions were later applied on PET scan to compute the respective 18FDG influx rate constant (Ki). Same computations were performed for the poorly aerated tissue. The gaps between the regions of interest are displayed to increase figure clarity.

Figure 2.

(A) Individual values of [18F] fluoro-2-deoxy-D-glucose (18FDG) influx rate constant (Ki) for lung regions derecruited both at end-expiration and end-inspiration (derecruited) and regions undergoing cyclic recruitment–derecruitment (recruiting–derecruiting). Open circles represent the values from each patient, solid circles indicate the mean. No statistical difference could be observed, even if the values were normalized by the density (recorded at mean airway pressure, reflecting the average density along the respiratory cycle) of the regions (B).

Figure 3.

Comparison of variables potentially associated with ventilator-induced lung injury between patients displaying an [18F]fluoro-2-deoxy-D-glucose (18FDG) influx rate (Ki) of the lung regions undergoing cyclic derecruitment–recruitment higher (Kirecruiting–derecruiting/Kiderecruited > 1) or lower (Kirecruiting–derecruiting/Kiderecruited < 1) than regions derecruited throughout the respiratory cycle. No statistically significant difference was observed among the variables analyzed. MV = mechanical ventilation; R-D = recruiting–derecruiting.

Figure 4.

(A) The plot of [18F]fluoro-2-deoxy-D-glucose (18FDG) influx rate of the normally aerated tissue (Kinormally-aerated) versus the plateau airway pressure shows a statistically significant correlation between the variables, with a nonlinear shape characterized by a steep increase of Kinormally-aerated for values of plateau pressure higher than 26 cm H2O (dotted line). (B) The correlation was still tight after normalizing Kinormally-aerated by 18FDG influx rate of the nonaerated tissue, excluded from tidal ventilation (Kinormally-aerated/Kinonaerated).

Figure 5.

[18F]fluoro-2-deoxy-D-glucose (18FDG) influx rate of the normally aerated tissue (A) before (Kinormally-aerated) and (B) after (Kinormally-aerated/Kinonaerated) normalization by 18FDG influx rate of nonaerated tissue not exposed to tidal ventilation was tightly correlated with the Vt expanding this lung regions (VtCT, normally-aerated) normalized by the end-expiratory lung volume (EELVCT, normally-aerated), over which the Vt is distributed.