点击显示 收起
Division of Gastroenterology and Hepatology Northwestern University Medical School 676 North St Clair Street, Suite 880 Chicago, IL 60611 E-mail: a-buchman{at}northwestern.edu
Dear Sir:
Boelens et al state that no data exist showing that intestinal atrophy results in bacterial or endotoxin translocation, and therefore they may have failed to understand the rationale for my previous study. I agree that no such human data are available; however, there are many studies in rodent models, including those I cite here (1–3). Our hypothesis in the 1995 study (undertaken in 1993) was that total parenteral nutrition is not associated with clinically significant intestinal atrophy in humans (4). We did not study bacterial or endotoxin translocation in the healthy volunteers because it would have been unethical to do so. We found mild but clinically insignificant intestinal hypoplasia and no atrophy after 2 wk of total parenteral nutrition in this study in which patients had normal intestines at baseline. Although we reported only plasma glutamine concentrations (4), we also measured mucosal glutamine concentrations. Mucosal glutamine was undetectable both at baseline and after 2 wk of total parenteral nutrition. These data were not reported because at the time we were unsure whether this was an error or whether the human jejunum has a significant amount of glutaminase and glutamine was rapidly metabolized. To my knowledge, there have been no reported measurements of glutamine in human jejunum, only in duodenum. Regardless of whether it is appropriate, glutamine "deficiency" has been accepted as a diagnosis in the presence of decreased plasma glutamine concentrations alone.
Boelens et al were concerned with 2 points: first, that I focused solely on the hypothesis that glutamine exerts its effect via intestinal morphology and, second, that I stated that Houdijk et al (5) suggested that the decreased infectious complications observed in their study may have been related to reduced bacterial translocation (6).
Bacterial translocation was the only mechanism suggested in Houdijk et al's article. They never stated that increased arginine may have been the reason for the fewer infections observed with glutamine supplementation. Boelens et al suggest a role for taurine as well. Taurine may mediate the glutamine-associated decrease in extracellular fluid, but this has no proven clinical significance.
The decrease in infections observed by Houdijk et al (5) in association with glutamine supplementation is novel and appears valid. However, this study should be carefully evaluated before clinical recommendations are made. The "inexplicably and inappropriately high" infection rate in the control group may have accounted for the difference in the infection rate alone, rather than the glutamine supplement (7). Additionally, in several other studies I cited in my review, glutamine was not associated with decreased infection risk.
Contrary to their statement, I covered such issues as glutamine status during various medical and surgical conditions; intestinal permeability; intestinal immune function; extracellular fluid retention; nitrogen, energy, and electrolyte balance; effects on the liver; infectious complications; T cell function; tumor necrosis factor concentrations; clinical severity of Crohn disease; diarrhea; quality of life; and hospital length of stay (6).
Although glutamine may increase arginine synthesis, the clinical sequelae of this in humans are unknown. In addition, Boelens et al state that arginine is important in wound healing, but this has not been shown in humans. In the only related human study of which I am aware, arginine supplementation resulted in increased hydroxyproline content in healing artificially created wounds, but did not actually enhance or hasten wound healing (8).
Although arginine may be important for the immune system in humans, no human evidence supports that theory. There are also no data showing that arginine supplementation improves the human immune response. Responses to concanavalin A and phytohemagglutinin cannot be described as valid and reproducible tests of human immune function (8). Neither a glutamine-free diet nor a glutamine-supplemented diet has any affect on intestinal immunoglobulin A production or on gut-associated lymphoid tissue in humans (9,10), and such effects have not been universally shown in rats (11). Houdijk et al (5) showed that after 7 d of feeding a glutamine-supplemented diet, plasma arginine and glutamine concentrations were not significantly different between supplemented and control patients fed for a mean of 12 d each. The observed decrease in sepsis prevalence occurred during a time when plasma glutamine and arginine status may have equilibrated in both the experimental and treatment groups. Houdijk et al reported all episodes of sepsis to have occurred within the first 7 d of feeding.
Boelens et al suggest that glutamine supplementation leads to increased arginine, which in turn leads to increased nitric oxide production. If that is true, it is possible glutamine could be detrimental. Nitric oxide impairs the intestinal endothelial response to endotoxinemia in rats (12). In addition, glutamine supplementation could increase intestinal inflammation because of stimulated mucosal T cell function and increased nitric oxide synthesis (13,14). In this fashion, glutamine supplementation may have a further negative effect on mesenteric blood flow in critical illness.
With regard to the meta-analysis of Novak et al (15), such reports do not replace properly performed investigations. The validity of any meta-analysis is in part related to the rigor of each of its included studies, as well as their heterogeneity. Rather than meta-analysis, what is needed are properly designed, controlled, and analyzed clinical studies of glutamine supplementation. For the most part, these have not been performed to date. Unfortunately, clinical decisions with clinical consequences and expense are being made on the basis of animal studies alone in the absence of appropriate human investigations. That is why parenteral glutamine is not approved by the Food and Drug Administration for use in the United States. When a nutrient is used as a drug, it should be treated as a drug and regulated as a drug.
REFERENCES