When is a Controversy Not a Controversy? Escaping Never-Never LandGastroenterology–Hepatology–Nutrition Section, Department of Veterans Affairs Medical Center, and Georgetown University School of Medicine, Washington, DC Correspondence: Timothy O. Lipman, MD, GI-Hepatology-Nutrition Section (151W), DVAMC, 50 Irving Street NW, Washington, DC 20422. Electronic mail may be sent to timothy.lipman{at}med.va.gov.
When a thing ceases to be a subject of controversy, it ceases to be a subject of interest. William Hazlitt (1778–1830) Some 5 years ago or so at an American Society of Parenteral and Enteral Nutrition (A.S.P.E.N.) Clinical Congress, Dr Patrick Twomey, a longtime member of A.S.P.E.N., honored me and several others by induction into the A.S.P.E.N. Curmudgeon's Club (curmudgeon: a crusty irascible cantankerous old person full of stubborn ideas). Members of the club were "elder statesmen" in A.S.P.E.N., definitely crusty and irascible, probably not cantankerous, and imbued with definite, if not stubborn, ideas. What was the genesis of the club? It was an enumeration of the many passing artificial feeding fads that we had witnessed over the 20-some years of a professional organization. Dr Twomey listed many more than I can now remember, including protein-sparing therapy; the myriad benefits of "elemental" diets; branched-chain amino acids for liver failure and catabolic states; the philosophy that if some is good, more is better; anthropometrics; skin testing; postpyloric feeding; and bacterial translocation. Perhaps longtime members of A.S.P.E.N can add other topics to this list. I was reminded of the club when I saw the list of topics in this issue of Nutrition in Clinical Practice dealing with nutrition support controversies. I feel, as did Yogi Berra, déjà vu all over again. Many of these are old friends: Do we count protein calories as part of total calories? How should we use albumin, both in diagnosis and therapy? How do we make a diagnosis of nutrition status? And does anything we measure have clinical meaning? I have some brief and extended comments. The many faces of albumin have been with us since the beginning. David Seres1 makes the compelling case that it, and many other "markers" of nutrition status, are at best nonuseful surrogates: little benefit in actually making a diagnosis, less benefit in directing therapy, and no benefit in terms of measuring clinical benefit. There have been many excellent reviews of albumin over the years, often relooking at the same issues. The overview by Mendez et al2 is a fine review of the metabolism and therapeutic uses of albumin in clinical practice. They note that there is no support for the use of IV albumin infusion as an artificial feeding product and that serum albumin levels should not be used as a marker of adequacy of artificial feeding. I think that it is time we give "albumin" a well-deserved rest. I certainly teach that as a diagnostic marker, serum albumin less reflects nutrition status than serves as a "negative catabolic" or nonspecific "illness" marker, that too many nonnutrition factors contribute to its level in the serum, and that low levels rise when patients' clinical conditions improve but not in response to artificial feeding. At best, albumin can be taken as a prognostic factor, but not one that reflects any response to intervention with artificial feeding. Finally, infusion of albumin has no role in artificial feeding. I certainly hearken to Dr Seres' conclusion: "... surrogates such as albumin, prealbumin, transferrin, IGF-1, delayed hypersensitivity, total lymphocyte count, and others should no longer be referred to as `nutrition' markers." Like albumin, these markers are multifactorial in etiology when they are abnormal and do not necessarily respond to artificial feeding therapies. They, too, should be put to rest and not used in clinical evaluation, clinical follow-up, or public policy. Ron Koretz3 takes us on a whirlwind tour of the pitfalls of worshipping p values and the mistakes of interpretation that may arise when "statistical significance" is either incorrect or without clinical meaning. His paper is a must-read for anyone interested in understanding how all that is "statistically significant" is not created equal. (Truth in writing: Ron is a longtime friend, and we think alike on many issues.) The review on evidence-based medicine brings me to the title of my editorial. Hise et al's4 description of the basic components of answering a question using evidence-based medicine is clear cut: "The first... step... begins with a clearly formulated question. Each question should be carefully structured in order to accurately access and evaluate the relevant literature. The question should be framed to focus upon a specific patient population, and should include a predictor (intervention treatment), and outcome variable (expected result of the intervention treatment) and the comparator (population without intervention treatment or with a different intervention treatment, ie, the control group)." Dr Hise goes on to give the example of asking, in an intensive care unit (ICU) patient population, "does feeding by the enteral route (predictor) reduce morbidity or mortality (outcome variables) when compared with parenteral nutrition (comparator)?" Later in the paper, she further describes several ICU patients with acute pancreatitis and the evidence-based clinical approach to decide that enteral feeding of such patients provides enhanced clinical benefit compared with parenteral nutrition (PN). So where is the controversy in this? Very few in the artificial feeding community want to address the possibility that no interventions are necessary in the above scenarios. The fundamental issue facing us, which has been addressed by evidence-based medicine, is that we do not know whether intervention works or not. Controversial? Not if we do not wish to confront the topic. Ron Koretz5 has addressed the specific issue of pancreatitis. In acute pancreatitis, postpyloric intestinal artificial feeding results in modest decreases in infection, length of hospital stay, and potential requirement for surgery when compared with IV artificial feeding.6 However, this assumes that some form of interventional artificial feeding is beneficial (ie, has positive effects on disease outcome). When comparing the 2 interventions, we can only say that intervention by the intestinal route is better than intervention by the IV route. We cannot say that any intervention is beneficial in and of itself. Intestinal intervention may look good only because of deleterious effects from IV intervention. Unfortunately, and much to the discomfort of the artificial feeding community, there is little to no evidence that interventional artificial feeding, initiated within 1–2 weeks after an individual is unable to eat, has major positive clinical outcome benefits. Two major consensus conferences reviewing a large number of randomized, controlled trials failed to find benefit from interventional artificial feeding.7,8 A systematic review and meta-analysis of PN served as a technical review of PN for the American Gastroenterological Association.9 Again, no documented benefit, and even possible harm, was found to accrue as a result of IV artificial feeding. Also of importance, according to evidence found in the technical review, was the fact that the time frame for which artificial feeding intervention was needed should probably be measured in weeks, not the 5, 7, or 10 days usually stated by various "experts." A systematic review comparing intestinal artificial feeding with no feeding intervention found a modest reduction in infectious complications in surgical patients but little other evidence for benefit or harm.10 The pickings are slim for interventional artificial feeding and acute pancreatitis. To my knowledge, only 1 study has addressed IV artificial feeding, finding that recipients of IV feeding had a longer duration of hospitalization and possibly more infections as compared with non–artificially fed controls; that is, the intervention resulted in net harm.11 Only 1 study addresses intestinal artificial feeding in acute pancreatitis compared with no such intervention.12 In a short-term, primarily physiologic study, short-term intestinal artificial feeding neither ameliorated nor exacerbated the inflammatory response to acute pancreatitis, nor did it alter multiple organ dysfunction scores. If one wished to scour for negative aspects of intestinal artificial feeding in this report, those receiving the intervention had more abnormal intestinal permeability and more frequent nausea. It is possible that interventional artificial feeding may provide a clinical outcome benefit in the medical treatment of acute pancreatitis. It is possible that it may not. There is insufficient evidence in the literature to make a judgment. If we are to take an evidence-based approach, however, we could well extrapolate from other data that suggest that interventional artificial feeding started before 2 weeks would provide little benefit. Such a conclusion is anathema within the artificial feeding community. Why? I have argued previously that we are trapped by language.13 Language, what we name our actions, imposes meaning and affects what we do. We equate technology with eating. We fail to recognize that "nutrition support" is a medical intervention accompanied by potential risks and complications, not ingesting good food as we traditionally conceive of it. This failure to recognize the difference between healthy eating and tube infusion of limited nutrients becomes manifest, and we become intellectually paralyzed, when we are confronted with the "S" word, starvation. Rationality disappears. We are terrified by the "S" word; the response to evidence is, "How can you let your patient starve!" "Starvation" is associated with famine images and eating associated with happy images. We equate our interventions with Sunday brunch, Thanksgiving dinner, or a Passover Seder. When we insert a tube, we are initiating a medical intervention with risk and perhaps benefit. What we put into the tube is not the same as what we eat. The composition is limited and fixed, quite dissimilar to a healthy diet. For the past several years, I have been searching for good terminology to replace "nutrition support" and to express neutrally what we do. In this paper, I have talked about artificial feeding, be it intestinal or IV feeding. I am not 100% happy with these terms. Perhaps tube infusion of limited nutrients would be better; perhaps something else. Like everyone else, I am against starvation and for nutrition. However, if using the "S" word and calling what we do "nutrition support" engenders reflex responses and precludes rational thought and evaluation, we need to change our language so that we can compassionately but dispassionately evaluate the evidence and provide competent clinical care. If our language reflects that we are engaged in a medical procedure, we can then look at clinical evidence to determine if what we do has clinical benefit. If our language precludes thought, we are trapped in a world in which we are afraid of controversy and clinical evidence. Peter Pan never wanted to grow up. He wanted to remain a boy forever in Never-Never Land. Captain Hook was always looking over his shoulder for the ticking clock. The Journal of Parenteral and Enteral Nutrition was first published in 1977; Nutrition in Clinical Practice debuted in 1986. Perhaps we have a ticking time bomb following us if we choose not to confront reality but rather continue to live in our own fantasy land and avoid facing the proof. Are we ever going to grow up? Only time and the evidence will tell.
Nutrition in Clinical Practice, Vol. 20, No. 3,
291-293 (2005) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
