Massachusetts General Hospital Department of Anesthesia, Critical Care and Pain Medicine - Critical Care Research
Mass General Hospital  
 printemailA  |  A  |  A   

Research Overview

Critical Care Research

Pain and Neurosciences Research

Biomedical Innovation

Ethics and Health Policy

Clinical Research

Grant Funding


Guido Musch Laboratory

Summary Science Features

Relationship between regional and global gas exchange. Interest in investigating this relationship stems from two clinical observations:

1) In some patients with ALI, the degree of hypoxemia does not mirror the loss of lung aeration, as measured by conventional chest X-ray or computed tomography (CT);

2) Although the degree of hypoxemia is used to define the severity of ALI, it does not seem to have prognostic significance (Nuckton TJ et al. N Engl J Med 2002;346:1281-1286).

We therefore sought to better understand how the loss of aeration, which is related to outcome (Gattinoni et al. N Engl J Med 2006;354:1775-1786), is reflected in the degree of hypoxemia. In a lung-lavage model of ALI, we reproduced the clinical scenario in which there was large interanimal variability in shunt fraction even though all animals had similarly reduced gas fraction (i.e., loss of aeration). Despite this lack of correlation between global (i.e., whole-lung) shunt fraction and gas fraction, when the lung field was divided in horizontal regions-of-interest (ROIs), in which regional shunt and gas fractions were measured non-invasively by Positron Emission Tomography (PET), there was a tight systematic inverse correlation between regional shunt and gas fractions. This disconnect between regional and global behavior could be explained by taking into account the effect of the spatial distribution of perfusion and of the heterogeneous loss of aeration on regional shunt flow (Musch et al. Am J Respir Crit Care Med 2008;177:292-300). These results suggest that, in ALI, the relationship between loss of aeration and shunt is complex and that knowledge of the spatial distributions of perfusion and aeration is needed to properly interpret the degree of hypoxemia as a marker of ALI severity.

We have recently applied PET imaging of [13N]nitrogen (13N2), infused intravenously in saline solution, to study the relationship between regional perfusion, shunt and aeration in critically ill patients with ALI. Preliminary results suggest that our previous findings from animal models closely resemble the relationship observed in patients.

Effect of PEEP on gas exchange and VILI. To better understand the mechanism by which PEEP improves gas exchange in ALI, and may protect from VILI, we studied the effect of PEEP on regional perfusion, shunt, and aeration (Figure 1) as well as on the metabolic rate of neutrophils activated by VILI (Figure 2). To pursue these objectives, we employed PET imaging of, respectively, 13N2 and [18F]fluorodeoxyglucose ([18F]FDG). We found that PEEP reduced shunt in dependent regions of saline-lavaged lungs (Musch et al. Am J Respir Crit Care Med 2008;177:292-300) and delayed neutrophil activation during VILI. Despite this delay, neutrophil activation occurred even at high PEEP, and preceded other signs of VILI such as decreased compliance or histologic changes (Musch et al. Anesthesiology 2007;106:723-735).

Effect of RMs on gas exchange. Prompted by the counterintuitive clinical observation that, in a fraction of patients with ALI, oxygenation worsens after a RM, we sought to determine a possible mechanism for this worsening. In a polysorbate 80-saline lung lavage model, we showed that, when a RM fails to achieve stable and substantial alveolar recruitment, it can worsen oxygenation by diverting pulmonary blood flow toward poorly aerated regions, with the consequence that a larger fraction of the total pulmonary blood flow is shunted as it traverses these regions (Musch et al. Anesthesiology 2004;100:323-330).

Effect of prone positioning on gas exchange and VILI. Prompted by the clinical observation that only approximately 60 to 70% of patients with ALI improve their oxygenation when positioned prone, we sought to determine a possible mechanism for this intersubject variability. Based on our human (Musch et al. J Appl Physiol 2002;93:1841-1851) and animal (Richter et al. Am J Respir Crit Care Med 2005;172:480-487) studies, we hypothesize that the extent to which perfusion redistributes toward ventral regions in the prone position is a determinant of this variability. Currently, we are investigating whether the potential beneficial effects of prone positioning go beyond an improvement of gas exchange, and, in particular, if it can protect from VILI by reducing the activation of neutrophils.

Office Information

Department of Anesthesia,
Critical Care and Pain Medicine
Gray-Bigelow 444
55 Fruit Street
Boston, MA 02114

Public Transportation Access: Yes
Disabled Access: Yes