Wednesday, September 13, 2006
The Journal of Parenteral and Enteral Nutrition has a provocative article showing a mortality benefit of parenteral nutrition in advanced cancer. Reading this study is like taking a trip through bizarro world (more on this in a second). This was a randomized trial of 150 patients with advanced cancer (mostly rectal, esophageal, and gastric) and 'malnutrition' (BMI less than 20 or a 5% loss of body mass in 3 months). All were undergoing anti-cancer therapies (chemo, radiation) at a cancer center in Germany. They were randomized (details of the randomization are not disclosed) to either supplemental parenteral nutrition (PN) (through a central venous line) or oral nutritional supplementation (calorie shakes). Mean age was ~58 years and BMI 23. The idea was that the PN patients would get 30% of their daily nutritional needs via the PN and the rest via regular oral diet (50-60%) and calorie shakes (10-20%). The non-PN patients would get their daily needs via calorie shakes (20-40%) and the rest via regular oral diet. Total calorie and protein intake, on average, was the same between the groups during the study (albeit from different sources). Patients were followed for a mean of 11 months. The study's primary aim was purportedly the effect of PN on quality of life and body composition (there doesn't seem to be a set time frame for the primary outcome...e.g. at 6 months? 12 months? etc.).
What they found was that the PN group had higher BMI's, albumin, and quality of life scores. Remarkably they found that the PN group's median survival was 12.5 months vs. 9 months in the non-PN group (this difference was statistically significant).
Let me repeat that: they found that PN prolonged life by about 3 months in this group of patients with advanced, incurable malignancies.
And so we enter bizarro world. The first question anyone should ask here is why is this study which shows there is a therapy that can prolong life by 3 months (when your expected survival is 10 months that's pretty darn good) published in the (with all due respect) obscure JPEN and not Journal of Clinical Oncology or the NEJM for that matter? This, if true, would be a super-big deal. Second, and this is why reading this study was weird, is that they bury this finding after multiple tables of subgroup analyses about total body weight etc. etc. as if anyone cares (maybe the eds & readers of JPEN care but the rest of the world doesn't). It's as if the authors aren't really excited about their finding.
So what's going on? I don't know. The article doesn't disclose several things one would like to see (the nature of the randomization, etc.) and you sort of get a sense that this isn't the most rigorous research ever conducted (alternatively the editors of JPEN may not have cared about this stuff). Also, if one is to believe these findings it would imply that it's not that the patients were benefitted by increased calorie/protein intake but by the fact that the nutrition was delievered via an IV as protein/calorie intake was the same between groups. All this being said, these people took 150 sick people, ostensibly randomized them, and found out that one group lived ~25% longer than the other, and that's a finding that needs to be investigated. I would really like Pallimed readers to check this article out and leave comments if you have any further insight into this.