Comparison of eligible randomized patients with two groups of ineligible patients: can the results of the VA Total Parenteral Nutrition clinical trial be generalized?

Abstract

The recently reported VA Cooperative Study "A Randomized Clinical Trial of Total Parenteral Nutrition (TPN) in Malnourished Surgical Patients" randomized 395 pre-operative patients to TPN treatment or control. The study concluded that the use of perioperative TPN should be limited to the most severely malnourished patients. The study also followed 233 patients eligible for the study who refused to give informed consent for randomization (Eligible Refusers) as well as 1220 patients who were ineligible because they were not sufficiently malnourished (Index Group). Patients in the Index Group were determined to be significantly healthier than those in the two eligible groups of patients. Those in the Eligible Refuser group were shown to be slightly less malnourished than the Randomized Patients. The 395 patients randomized to the study (Randomized Patients) showed the highest rate of septic complications at 30 days and at 90 days (10% and 13% respectively) with rates for the Eligible Refusers slightly lower (8% and 9%) and Index Group rates still lower (4% and 4%). Nonseptic complication rates showed the same pattern (19% and 22% for the Randomized group, 12% and 12% for Eligible Refusers, and 10% and 10% for the Index Group). Because (a) the beneficial effect of TPN is attained only in severely malnourished patients, (b) there is increased risk of septic complications with TPN use in patients not severely malnourished, (c) Index Group patients, and presumably the population of patients from which they are drawn, are not severely malnourished, it follows that unless specifically indicated, TPN should not be used in nonseverely malnourished patients.