* 66% of the patients had malignant disease. Note: NR: not reported only dextrose-saline (6 kcal/kg) over the same period. Patients in both groups had similar weight loss before operation (1.8 to 5.9 kg). Increased morbidity — impaired wound healing — was documented in four of the five patients in the control group. Because of the small number of patients, statistical significance was not obtained.
The first study to demonstrate significant beneficial effects was performed by Heatley et al., who randomized 74 patients with stomach or esophageal cancer to preoperative total parenteral nutrition (TPN) plus an oral diet versus an oral diet alone for 7 to 10 d.97 All patients with esophageal anastomoses were given postoperative TPN, regardless of randomization. The overall complication rate, mortality rate, and hospital length of stay were not different between the groups. Subset analysis revealed a significantly lower percentage of wound infections in the treated group; and the difference was magnified if a second subset of patients with low serum albumin (< 3.5 gm/dl) were included.
Mullen et al.98 compared the effect of 10 d of preoperative TPN (n = 50) with that of a standard oral hospital diet (n = 95) in patients undergoing surgery for intraabdominal malignancy. The groups were matched for nutritional status and assessed by serum albumin < 3.5 gm/dl, negative responses to skin antigens, and weight loss > 5 kg in the preceding 12 months. Although this study demonstrated a significant decrease in major complications and mortality in the group receiving TPN, the intake of those receiving oral nutrition was not controlled. In a similar study, Muller et al. randomized 160 upper GI and colorectal cancer patients to 10 d of preoperative TPN plus oral diet versus oral diet alone.99 This clinical trial reported a significant reduction in the major complication rate (intra-abdominal abscess, peritonitis, anas-tomotic leakage, ileus) and mortality in the treatment group. There are significant concerns with this trial. When the surgeons in this study began stapling their anastomoses, anastomotic leakage — which was the major cause of the mortality and morbidity — was substantially reduced, whether or not patients received nutritional supplementation.
These studies have been criticized because of small numbers of patients, lack of randomization, use of unsuitable control groups, differences in nutritional status before surgery, and inadequate quantity or duration of nutrition. Detsky et al.100 carried out a meta-analysis of trials of perioperative nutritional support and, despite the limitations of many of the studies, found that nutritional supplementation reduced the morbidity rate by 21% and the mortality rate by 32% after major surgery, although it was not possible to identify clearly which particular patients would benefit. Major iatrogenic complications associated with TPN were documented in 7% of patients.
Because of problems in the methodology and design of previous randomized clinical trials evaluating the impact of perioperative TPN in the reduction of postoperative morbidity and mortality, Buzby et al.101 performed a critical review of previous randomized trials and determined that the defects in these studies fell into four major categories: defects in statistical design (inadequate sample size, inappropriate procedures for randomization); inappropriate patient selection (presence of malnutrition not an eligibility requirement for study participation); inappropriate treatment regimens (inadequate TPN regimen in treated patients, use of forced enteral feeding in control subjects); inadequate definition of endpoint criteria (lack of objective criteria defining presence of complications).
These considerations led to the development of the Veterans Affairs Total Parenteral Nutrition Cooperative Group study.82 This was a prospective, multicenter study that followed 395 patients undergoing thoracic or abdominal surgery who were randomized to 7 to 15 d of preoperative and 3 d of postoperative TPN versus no intravenous nutritional support. All patients were given an oral diet if clinically indicated (mean daily caloric intake in the TPN group was 2944 kcal and in the control group was 1280 kcal). Sixty-six percent of the patients were cancer patients. The overall complication rate and mortality were not different between the groups, suggesting no benefit to routine perioperative support. In addition, the incidence of infectious complications and the length of hospital stay were greater in the TPN group. However, subgroup analysis revealed that if patients were severely malnourished (as defined by the Nutrition Risk Index or the Subjective Global Assessment),89 they had a significantly lower noninfectious complication rate without an increase in the infectious complications. Noninfectious complications included major complications, such as anastomotic leakage, bronchopleural fistula, and organ dysfunction.
Two prospective randomized studies have examined the role of perioperative TPN in pancreatic cancer and hepatocellular carcinoma with differing conclusions. Brennan et al.102 randomized 117 patients with pancreatic cancer undergoing major pancreatic resection to receive postoperative parenteral nutrition (mean 12 d) or standard intravenous fluids. The majority of patients demonstrated mild malnutrition based on weight loss and serum albumin. Major complications that could potentially be influenced by the addition of TPN (fistula, abscess, obstruction, anastomotic leakage, reoperation) were significantly higher in the group that received parenteral nutrition. There were no differences in postoperative mortality or length of hospital stay between the two groups. Fan et al.103 studied 124 patients with hepatocellular carcinoma who underwent major hepatic resection randomized to 7 d of preoperative and 7 d of postoperative TPN versus intravenous fluids and an oral diet. A significant reduction in the postoperative complication rate was demonstrated primarily due to a reduction in the subset of infectious complications. Again, there were no significant differences in the mortality or hospital length of stay.
In assessing the clinical implications of these various studies, it appears that routine perioperative TPN is not associated with a significant benefit in the surgical cancer patient, regardless of the type of malignancy or the magnitude of the operation. The use of TPN can be associated with an increased risk of infectious complications. However, subset analyses seem to indicate that if after conducting a nutritional assessment of the patient, whether by clinical or objective measures, a severe level of malnutrition is determined, then these patients may benefit from perioperative TPN, as the risks of nutritionally related complications outweigh the risks of TPN-related complications.
perioperative enteral nutrition
Several prospective randomized studies have also examined the role of enteral feeding in the perioperative period in cancer patients, and these are listed in Table 12.4. The proposed advantages of enteral nutrition over parenteral include the following:
• Maintenance of gut mucosal mass
• Maintenance of brush border enzyme activity
• Support of gut immune function
• Preservation of gut mucosal barrier function
• Maintenance of a balanced luminal microflora environment
Was this article helpful?