Showing papers by "James A. Frank published in 2000"

β-Adrenergic agonist therapy accelerates the resolution of hydrostatic pulmonary edema in sheep and rats

Abstract: To determine whether β-adrenergic agonist therapy increases alveolar liquid clearance during the resolution phase of hydrostatic pulmonary edema, we studied alveolar and lung liquid clearance in tw...

Diabetes mellitus: a negative predictor for the development of acute respiratory distress syndrome from septic shock.

Abstract: I n this issue of Critical Care Medicine, Dr. Moss and colleagues (1) provide the first prospective study that identifies a clinical disorder associated with a decreased incidence of acute respiratory distress syndrome (ARDS)— diabetes mellitus. Of 113 patients with septic shock, the authors identified 32 (28%) with a history of diabetes mellitus as defined by a prescription for insulin or use of an oral hypoglycemic agent. The incidence of ARDS was strikingly lower in the diabetic patients compared with the nondiabetic patients (25% vs. 47%; odds ratio 5 0.33, 95% confidence interval 0.12– 0.90). Until now, the potential effect of diabetes on the incidence of ARDS and mortality largely has been overlooked. In fact, the prevalence of diabetes mellitus is not mentioned in most studies of ARDS or sepsis. The discovery of a lower incidence of ARDS among diabetic patients has important implications for the design of future studies. The major strength of this study is the prospective, multicenter design and the use of multivariate logistic regression for statistical analysis. The inclusion of only patients with septic shock is an additional strength for two reasons. First, sepsis is the most common and most lethal cause of ARDS (2). Second, the diagnosis of sepsis was defined clearly by objective criteria. Also, the overall incidence of ARDS in this study was similar to other studies of septic patients. Therefore, the results of this study can be generalized with greater confidence. There are, however, some limitations to this study. First, the definition of diabetes mellitus may have been too restrictive, because patients with diabetes controlled by diet or undiagnosed type 2 diabetes may have been excluded. Furthermore, combining all patients with type I and II diabetes without specifying the need for insulin may be suboptimal, because the molecular and genetic basis for diabetes in these patients is heterogeneous (3). Consequently, the underlying factors influencing the development of ARDS may not be uniform among these diabetic patients. As the authors point out, it may have been preferable to determine hemoglobin A1 levels in all patients as an additional marker of disease and as a measure of diabetes control. Using a prior prescription for insulin or an oral hypoglycemic agent as the definition of diabetes also may select for patients who received better medical care before admission. A second potential limitation to this study is that the association between the incidence of ARDS and diabetes mellitus becomes weaker when the admission glucose values are compared. By using a plasma glucose concentration of 120 mg/dL as a threshold, the incidence of ARDS for patients with hyperglycemia was 33% (22/66) compared with 51% (24/ 47) in the patients with normal plasma glucose (p 5 .06). In a subgroup analysis of the 81 nondiabetic patients, nearly half of whom had an elevated admission plasma glucose, the incidence of ARDS in patients with hyperglycemia was 39% (15/39) compared with 55% (23/42) in the patients with normal plasma glucose (p 5 0.1). These differences are not statistically significant, but a trend toward a lower incidence of ARDS in hyperglycemic patients is apparent. This trend might have been less evident if blood glucose had been analyzed as a continuous variable rather than as a discontinuous variable, that is, above or below 120 mg/dL. Therefore, hyperglycemia may not be the only factor influencing the development of ARDS, although it is one of the most obvious abnormalities in diabetics. However, given the small sample size, a type II error may have occurred when only plasma glucose levels were considered. A larger number of patients, perhaps in association with hemoglobin A1 levels, might have sorted out this potential additional association. What are the mechanisms of protection against ARDS in diabetic patients? Are diabetics protected because of impaired neutrophil function or altered neutrophil-endothelial interactions, or are they protected by hyperglycemia and an associated hyperosmolar state? Perhaps both mechanisms are important. It is well known that neutrophils from diabetics have impaired biological responses, including reduced bactericidal activity and impaired chemotaxis (4, 5). Neutrophil adhesion to the endothelium also can be altered in diabetics (6, 7). These abnormalities could attenuate neutrophil-dependent lung injury in patients with septic shock. Alternatively, hyperglycemia may have an independent effect on neutrophil or endothelial properties, resulting in decreased injury in response to bacteria or endotoxin (4). Future studies could be designed to better distinguish these possibilities. These studies should include a more comprehensive definition of diabetes mellitus including diabetes controlled by diet alone and diabetes as indicated by hemoglobin A1 concentrations, plasma glucose concentrations, and serum osmolality. In addition, it might be useful to measure plasma markers of acute lung injury that have shown some promise in prior studies. For example, measurement of von Willebrand factor-antigen (8), intracellular adhesion molecule-1 (9), and tumor necrosis factor receptors may be valuable. These biological markers might have some additional positive or negative predictive value when coupled with the presence or absence of diabetes mellitus. Although Dr. Moss and colleagues (1 focus primarily on the incidence of ARDS among diabetic patients, the issue of mortality also should be considered. In the current study, mortality was not different between the diabetics and nondiabetics with ARDS (50% vs. 63%, p 5 .5). If decreased lung injury was responsible for the lower incidence of ARDS, one

Role in Lung Fluid Balance in Clinical Lung Injury

Abstract: This article reviews new insights into the basic mechanisms that regulate vectorial transport of salt and water across the alveolar epithelium of the adult lung. The first section provides evidence that active ion (sodium) transport is the primary mechanism that reg- ulates in vivo alveolar fluid clearance. This section includes a discussion of how catechol- amine-dependent and independent mecha- nisms can upregulate alveolar fluid transport. The second section briefly discusses evidence that implicates transcellular water channels in alveolar epithelial fluid transport. The third section discusses the transport function of the alveolar epithelial barrier under pathologic conditions that are relevant to clinical pulmo- nary edema and acute respiratory distress syndrome. Some of this information has been reviewed re~ently.5~