Nutritional Support and the Surgical Patient

Yvonne Huckleberry


Am J Health Syst Pharm. 2004;61(7) 

In This Article

Impact of NS on Patient Outcomes

While several studies have reported a favorable impact of NS on surrogate endpoints (e.g., weight maintenance, trends in visceral protein status),[10,11,12] evidence of an influence on clinical outcomes is conflicting. For example, a review sponsored by A.S.P.E.N., the National Institutes of Health (NIH), and the American Society for Clinical Nutrition (ASCN) summarized 22 prospective RCTs focusing on perioperative NS in general surgical patients.[8] The objective of the review was to assess the current medical literature and identify issues that warrant further investigation. Trials had to include patients randomized to receive PN for at least five days, with calories and nitrogen approximating daily requirements compared with controls (no NS). In addition, clinical outcomes (e.g., postoperative complications, mortality) had to be included as primary endpoints. The majority of the patients were described as moderately malnourished individuals with GI cancer. Pooled analysis of these trials suggested that 7-10 days of preoperative PN resulted in a 10% decrease in the risk of postsurgical complications but that early postoperative PN was associated with a 10% increase in risk. Both of these findings were statistically significant. A difference in mortality was not observed. These authors concluded that future research should focus on identifying subsets of patients who may benefit from preoperative NS.

A more recent systematic review, however, did not show a beneficial effect of preoperative NS or potential harm with postoperative nutrition.[9] This meta-analysis included 41 RCTs in over 2150 patients randomized to receive PN before or after surgery (nonprotein calories ≥10 kcal/kg/day plus a nitrogen source) or an ad libitum diet (controls). Endpoints included reports of patient morbidity or mortality and length of hospitalization. All RCTs excluded severely malnourished patients, and the overall nutritional status of patients included in the analysis was not described. Results indicated that PN had no significant impact on postsurgical complication rates (absolute risk difference, -6%; 95% CI, -13% to 1%), mortality (absolute risk difference, 0%; 95% CI, -2% to 2%), or duration of hospitalization. The difference between these results may be due to the different mix of RCTs evaluated, the adequacy of nutrients provided, the timing of intervention, or the preoperative nutritional status of the patient population.

Two systematic reviews of RCTs assessed whether NS, compared with no NS, impacts patient outcomes.[13,14] In regard to PN in surgical patients, the more recent meta-analysis included 27 trials involving 2907 patients.[13] Patients were described as either malnourished or not malnourished; the adequacy of calories and nitrogen provided in comparison to daily needs was not assessed. Aggregate results indicated no effect on mortality (RR, 0.97; 95% CI, 0.76 to 1.24) but a reduction in complications associated with PN (RR, 0.81; 95% CI, 0.65 to 1.01). However, this difference was not statistically significant, and heterogeneity was observed between groups. Subgroup analysis of trials published since 1989 and trials with higher-quality scores showed PN made no difference in complication rates (RR, 1.07; 95% CI, 0.86 to 1.32) or mortality (RR, 1.08; 95% CI, 0.81 to 1.43) compared with no NS. Consistent with the A.S.P.E.N.-NIH-ASCN review and others,[15,16,17,18] subgroup analysis suggested some potential benefit of PN in malnourished patients. The difference, however, was not statistically significant. Whether the composition, duration, or volume of PN influenced the results of any of these analyses is unclear. The conclusions, however, are consistent in suggesting that early postoperative PN does not improve clinical outcomes in surgical patients, with the exception of potential benefit for those who are severely malnourished.

EN has been promoted as the preferred route of NS. Enteral tube feeding offers the theoretical advantage over PN of improved maintenance of GI mucosal integrity via regular delivery of nutrients into the GI tract. This is suggested to reduce the risk of mucosal atrophy, bacterial translocation, infectious complications, multiorgan failure, and death. However, human evidence supporting these claims is sparse.[6,19,20,21,22] Enteral feeding is generally perceived as more physiological, safer, and less expensive than PN. However, Lipman[23] assessed available prospective RCTs comparing EN versus PN and challenged these perceptions. A cost advantage was suggested as the only evidence-based benefit observed for EN. A financial saving with postoperative EN compared with PN has been suggested by others as well.[24]

A more recent systematic review of prospective RCTs criticized the methodology used in Lipman's review.[25] These authors retrieved 27 RCTs involving 1828 patients to evaluate PN versus EN in terms of morbidity and mortality. Patients randomized to receive PN were given calories equal to or in excess of the estimated daily energy requirements. Nutrient goals for EN were not described. The results of this meta-analysis showed that patients receiving EN had a significantly lower risk of infectious complications than PN recipients (RR, 0.66; 95% CI, 0.56 to 0.79). Results were consistent in subgroup analysis, regardless of study quality score, year of publication, or the nutritional status of patients in the study. However, significant treatment effects were not observed for other complications or patient mortality.

Enteral tube feeding has also been compared with nothing-by-mouth status in terms of clinical outcomes. For example, Lewis and colleagues[26] performed a meta-analysis of 11 small RCTs involving 837 patients who underwent elective GI surgery. Patients were randomized to start EN within 24 hours of surgery or to receive the traditional postoperative routine of nothing by mouth progressing to a diet as tolerated. Results showed a significant reduction in the risk of infection (RR, 0.72; 95% CI, 0.54 to 0.98) and length of hospital-stay (-0.84 day; 95% CI, 0.36 to -1.33) in patients receiving early EN. There was an increased risk of vomiting associated with early EN, particularly when intraoperative feeding tube placement was not routine (RR, 1.21; 95% CI, 0.73 to 1.99). These authors concluded that, although the data were insufficient to prove an advantage of early EN, the trend toward a potential benefit was enough to warrant an adequately powered RCT.

In summarizing the evidence concerning the impact of NS on patient outcomes, it appears that severely malnourished patients are most likely to benefit from preoperative nutritional intervention. However, this would require a willingness of both the surgeon and the patient to delay surgery, establish access, and accept the risks of complications as well as the increased health care costs associated with preoperative nutritional rehabilitation.

Most patients tend to resume oral intake within 6-9 days after surgery,[5,6] suggesting that NS is usually not warranted in the early postoperative period for these patients. Whether the benefits of NS may outweigh the risks in patients with mild to moderate malnutrition preoperatively is less clear, but the delay for nutritional intervention should be less than would be recommended for well-nourished patients.

Despite challenges to the superiority of EN over PN, EN is preferred as at least the more cost-effective route of NS for patients with functioning GI tracts. A reduction in infectious complications may be an additional advantage. An adequately powered, prospective RCT is needed to better define the difference between EN and PN in terms of patient outcomes.

Recognizing the potential benefit of perioperative NS on those who are severely malnourished, patients should be assessed prior to surgery whenever possible. For those identified as malnourished or at risk of becoming malnourished, the timing and route of potential nutritional intervention should be planned accordingly. However, this early screening may not be feasible for many surgical patients. Therefore, a familiarity with nutritional-assessment variables appropriate for postoperative patients is essential.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.