Immunonutrition and Upper Gastrointestinal Surgery: What Really MattersDepartment of Surgery, The Miriam Hospital, Providence, Rhode Island Correspondence: Harry C. Sax, MD, Department of Surgery, The Miriam Hospital, 164 Summit Avenue, Providence, RI 02906. Electronic mail may be sent to hsax{at}lifespan.org. Patients with upper gastrointestinal cancer or other surgical diseases are at high risk for both mechanical and infectious complications. The disease process itself places these patients at increased risk for malnutrition, with subsequent postoperative complications. Because these surgeries are semielective, a period of time is available to optimize the patient's condition. There are several trials that suggest preoperative immunonutrition supplementation improves outcomes and is cost-effective by reducing complications. However, the most important choice a patient or clinician can make may well be by whom and where the surgery is performed. Surgeons can perform complex procedures that potentially cure patients of their disease. With this extent of intervention, complications are not infrequent, and the clinician will seek any aid to potentially improve outcomes. Some of these decisions may actually be based on data; others are pure empiricism. What is increasingly being recognized is that multiple factors are involved in optimizing patient outcomes after upper gastrointestinal surgery. Perioperative nutrition support is but one of these factors. The patient with intrinsic disease of the foregut (esophagus, stomach, duodenum, pancreas) is at increased risk for malnutrition from several factors: mechanical obstruction or limitation of food intake, tumor-induced cachexia, obstruction of pancreaticobiliary secretions leading to malabsorption, and ongoing blood loss. The onset of symptoms is slow and insidious, often leading to diagnoses at more advanced stage of disease. The problems are exacerbated by the extent of surgery that needs to be performed in an attempt to cure the patient. It is important to review the specific operations so that analysis of the nutrition data can be taken in context. Because lower gastrointestinal disease (diverticulitis, colon cancer) is less often associated with significant malnutrition and the surgical interventions tends to be more straightforward, immunonutrition does not seem to play as significant a role. For the nonsurgeon, a brief review of surgical options in the therapy of upper gastrointestinal disease may be helpful.
Patients with esophageal carcinoma or intractable strictures requite excision of the esophagus, with reconstruction involving the creation of a gastric tube or intestinal interposition. There are multiple anastomoses that can leak, and at times the esophagogastric connection is under tension. Both abdominal and thoracic incisions may be necessary, leading to postoperative splinting and increased pulmonary complications. Postoperative leaks and fistulae significantly prolong hospitalization and increase cost. The morbidity associated with gastric resection, although less than esophagogastrectomy, is still significant. Dependent on the disease process, 30%–100% of the stomach must be removed and continuity restored with either a loop of jejunum (Billroth II), a direct anastomosis to the duodenum (Billroth I), or a Roux-en-Y limb to the esophagus or cardia. Leaking or necrosis of the duodenal stump is a feared complication because the proximity to the ampulla leads to a large amount of peritoneal soilage with bile and pancreatic enzymes. Postoperatively, the patients must consume small portions of food and are subject to dumping syndrome, similar to bariatric surgery (see the review on dumping syndrome by Ukleja in this issue1). Exclusion of the duodenum in Billroth II reconstructions or with a Roux limb leads to poor absorption of iron and exclusion of several feet of absorptive capacity. Perhaps the most complex procedure performed by a gastrointestinal oncology surgeon is the Whipple procedure, or pancreaticoduodenectomy. Used for treatment of carcinoma of the head of the pancreas, it involves resection of the duodenum, pancreas, and distal common bile duct. Reconstruction involves 3 anastomoses, any one of which can leak and fistulize. Pancreatic juices leaking into the surrounding tissues and blood vessels can induce fatal hemorrhage. Even if patients do not have a mechanical complication of surgery, the incisions lead to pulmonary insufficiency and increase the likelihood of perioperative pneumonias. Numerous monitoring devices disrupt the skin and provide direct access to the bloodstream. The number of cases done yearly at a specific institution also affects outcome because there is an increased likelihood that experienced staff will identify complications early and have skill in treating them. Because malnutrition is common in upper GI cancer patients, the role of perioperative nutrition has undergone extensive study. With the advent of immune-enhancing formulas, the role of these formulas is being identified. Although more expensive than conventional supplements, they may save money by reducing complications.2,3
One of the classic studies on perioperative nutrition is the Veterans Affairs cooperative study.4 Surgical patients were randomized to immediate operation or 7 days of preoperative parenteral nutrition (PN). Overall complications were similar between the groups. In subgroup analysis, the most severely malnourished had a reduction in complication related to wound and anastomotic healing. In the mildly malnourished group, infectious complications were higher, suggesting an immunocompromising effect of the PN. This may well be due to the high-fat/high-glucose PN solution used during that period. It was shortly after this study that interest in enteral nutrition as a perioperative adjunct arose. There was a move to early postoperative enteral nutrition as opposed to PN in trauma and surgical patients. The emergence of "immune-enhancing diets" (IED) containing pharmacologic doses of arginine, fish oil, glutamine, or nucleotides raised hopes that patients could not only be maintained around surgery, but actually be given a better chance at a complication-free postoperative course. The Daly et al5 early report of the use of an IED or standard diet in a group of patients after upper gastrointestinal tract (UGI) surgery demonstrated improved in vitro measures of immune response. There was a trend toward reduced complications and length of stay (LOS). Kudsk et al6 randomized patients who had had severe traumas (ATI >25) to either an IED or isonitrogenous standard high-protein enteral diet. There were 17 patients in the IED group and 18 patients in the control group, and feedings were started about 1.5 days after admission and continued for 9–10 days. The IED was high in arginine, contained free glutamine, and had canola oil–derived  -3 fatty acids. The authors also followed a third group who served as
unfed controls; these patients were not randomized to be unfed: they were
patients who would have been eligible for the study but for some reason did
not have enteral access. The authors state the "2 patients died in the
treatment group and were dropped from the study"; otherwise, no data on
mortality are provided. Compared with patients receiving the standard formula,
patients receiving the IED had fewer septic complications (including fewer
intra-abdominal abscesses) and shorter overall hospital LOS, but trends toward
shorter ICU LOS and fewer days on the ventilator did not reach significance.
Complications included both local (abdominal abscess) and distant (pneumonia,
UTI, line infection) problems. It may be that the canola oil is less effective
than fish oil because the desaturases required for metabolism of the canola to
eicopentaenoic acid (EPA) and decahexaenoic acid (DHA) are down-regulated in
critically ill individuals.
In a meta-analysis, Heyland et
al7 state that
"immune-enhancing" diets offer no advantages with regard to
mortality or infections for the critically ill, and in fact, suggest that
there may be an increased rate of death among those who get the
"immune-enhancing" diet. He analyzed 22 trials and approximately
200 patients. Immunonutrition led to a nonsignificant increased risk of
mortality (RR, 1.10; CI, 0.93–1.31), but a significantly decreased risk
of infectious complications (RR, 0.66; CI, 0.54–0.80). Heyland's group
focused on the arginine content of the IEDs and suggested that the formulas
with the higher arginine content were the most efficacious. Most of the
studies in the critically ill or trauma patients used Impact (Novartis,
Minneapolis, MN), which also contains
The equivocal results in trauma and sepsis are not surprising. Any attempt
at mediating the inflammatory response is most likely to be successful if the
modulating agent is on board at the time of the insult. There appears to be a
period of time before the operative insult that can be used to mediate the
inflammatory response by allowing polyunsaturated fatty acids (PUFAs) to be
incorporated into the cell membrane and subsequently be cleaved for
prostaglandin synthesis. By the same token, maintenance of intracellular
glutamine stores before insult and catabolism could reduce lean body mass
loss. The ideal group to allow this type of study is patients undergoing major
elective surgery. Unlike trauma patients, who tend to be healthy at the time
of injury, patients with malignancy or in need of major GI or vascular surgery
are already at increased risk. Even if they do not manifest weight loss, they
will have a major, controlled traumatic insult that induces catabolism and
raises the risk of infectious and wound complications. The classic study from
Buzby4 on the use of
preoperative PN only showed benefit with a reduction in wound complications in
the most malnourished patients. This was offset by increased infectious
complications in the less malnourished patients. One hypothesis is that in
retrospect, because significant calories were given as Heys et al8 performed a meta-analysis of the use of immunonutrition in patients with both critical illness and cancer. He found an overall odds ratio of 0.47 (CI 0.30–0.73) for the development of major infectious complications in those patients receiving immune nutrition. This finding was repeated for the subgroup of patients with GI cancers. Nosocomial pneumonias and overall mortality were not affected. Perhaps due to the decreased infectious complications, patients receiving target IED had a short length of hospital stay of 2.5 days (CI 4.0–1.0 days). Braga et al9 entered almost 200 malnourished, elective surgical patients into 3 groups: "standard" enteral feedings immediately postoperative, standard feedings both pre- and postoperatively, or an immune-enhancing diet, both pre- and postoperatively. The operations were primarily gastric and pancreatic, with the remainder esophageal and colorectal. The group receiving perioperative IED had fewer infections and a shorter LOS than subjects receiving the postoperative, standard diet. The perioperatively fed, but nonenhanced diet was intermediate between the 2 for infectious complications but also had a decreased LOS compared with control. This study emphasizes the importance of not only correcting preoperative deficits but also "loading" with immune-modulating nutrients before the insult. The same group published a study limited to colorectal cancer patients.10 It is unclear if some patients from previous studies were included. The groups were slightly different: IED before and after surgery; IED only preoperatively; isocaloric isonitrogenous non-IED only preoperatively; or no pre- or postoperative supplements. Diet was initiated when GI function returned. Gut microperfusion was quantified intraoperatively, standard in vitro measures of macrophage response were assayed, and infectious complications followed. The groups receiving immunomodulating nutrients, including arginine, had increased gut perfusion and decreased postoperative infectious complications (12% preoperatively, 10% perioperatively, 32% conventional preoperatively, 35% standard). It is clear that to modulate the inflammatory and immune cascades, the agent must be on board at the time of insult. The elective situation afforded by the patient requiring UGI surgery provides the opportunity to study perioperative nutrition, including the use of IEDs to "beef up" the immune response before intervention. Braga et al11 used an IED that can be consumed orally then continued via jejunostomy postoperatively. In a phase 3 trial of over 200 patients receiving perioperative support, the group receiving IED had fewer infectious complications such as pneumonia, wound infection, or abscess. There was no difference in noninfectious complications, but 15 patients (7.5%) were removed from the study due to the major complication of anastomotic leak. The IED group also had lower levels of inflammatory markers, C-reactive protein, and IL-6. A follow-up to this study in well-nourished patients was published by the same group in 2002.12 It concentrated on only preoperative administration of IED. In this study, no difference was seen between perioperative and preoperative-only therapies. Once again, infectious complications were reduced, and later analysis supported cost effectiveness of the use of preoperative IED loading, even in well-nourished patients.2
In almost all studies reported, even the control arms had reasonable outcomes with relatively low mortalities. The volumes of patients necessary to carry out these studies suggested that the surgeons and institutions had high-volume experience. The reality is that all surgeons and all hospitals are not the same. Birkmeyer et al13,14 published sentinel studies reporting wide variation in operative mortality for high-risk, highly complex operations, such as esophagectomy, gastrectomy, and pancreatic resection. High-volume hospitals had an adjusted odds ratio of death of 0.36, 0.72, and 0.20, respectively, compared with results from hospitals with the lowest volumes of these operations. Taken another way, there is a fivefold difference in your chance of dying after a Whipple procedure if the hospital does not do them routinely. Surgeons' experience is another factor. Once again, the adjusted odds ratios for death were from 1.24 to 3.61 between high- and low-volume surgeons. The Leapfrog group has identified the importance of volume in surgical outcomes as one of its major criteria in assessing outcomes.15 These wide variations in outcome are far in excess of what might be induced by perioperative immunonutrition. However, the studies reported came from specialized centers and as such, show an improvement over already good results. The concern is that these data will be extrapolated to the general surgical population, with IEDs offering the promise of reductions in morbidity and mortality that would approach those of high-volume centers.
It is clear that patients at nutrition risk should be offered preoperative optimization. Specific recommendations regarding IEDs have been published in the Journal of Parenteral and Enteral Nutrition supplements.16,17 In short, the greatest benefit of IEDs appears to be with the perioperative use of IEDs in patients undergoing major upper abdominal surgery. The benefits have not been as clearly defined in colonic surgery, in part because anastomoses are not always required (colostomy, ileostomy). The IEDs do offer some benefit when provided preoperatively in with a target of 5–7 days as a 2–3 times/day supplement. No amount of immunonutrition however, can reverse the effects of sub-optimal surgical technique, judgment, and perioperative care.
Nutrition in Clinical Practice, Vol. 20, No. 5,
540-543 (2005)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Top
Surgical Options
Nutrition Support in the...