University of Virginia Health System
Nutrition Support E-Journal Club
Spring weather has been fickle this year, teasing us with a nice day or two (usually weekdays) and then turning back to cold and rain for a spell. Luckily the traineeship was scheduled near the end of the month, and our guests were able to enjoy several days of mild weather. Our trainees in April hailed from Pittsburg, PA, Martinsburg, WV, Abilene, TX, and British Columbia, Canada.
Pohl M, Mayr P, Mertl-Roetzer M, et al. Glycemic control in patients with type 2 diabetes mellitus with a disease-specific enteral formula: stage II of a randomized, controlled multicenter trial. JPEN J Parenter Enteral Nutr. 2009 Jan-Feb;33(1):37-49.
This was a randomized, double-blind, multi-center study of 105 patients residing at either neurological rehabilitation facilities or a nursing home requiring long-term tube feeding due to dysphagia as a result of neurological disorders. Patients were randomized to receive continuous enteral feeding with either a formula intended for diabetic patients (decreased carbohydrate, increased monounsaturated fats, with supplemental chromium, fish oil, and antioxidants - Diben; Fresenius Kabi, Bad Homburg, Germany) or an isocaloric, isonitrogenous standard formula.
The primary outcomes were the total daily insulin requirement as compared to baseline, fasting glucose and HgbA1C. Secondary outcomes were serum lipids, afternoon blood glucose, tube feeding tolerance (nausea, vomiting, diarrhea) and "safety" (routine safety of EN feeding, laboratory, adverse events at any time).
Inclusion and Exclusion Criteria were:
Patients 40 years or older with insulin-treated type 2 DM or with a fasting blood glucose of 120mg% and a HbA1C>7.0% and/or elevated fasting blood glucose (FG >120 mg/dL) and indication for long-term tube feeding due to dysphagia caused by neurological disorders (e.g., stroke, traumatic brain injury, hypoxic brain damage).
Type 1 DM, known allergy to ingredients of the enteral products, intake of other enteral diets, or use of parenteral nutrition, severe liver disease, renal failure, congestive heart failure, human immunodeficiency virus, or systemic glucocorticoid therapy within the last 2 weeks before and/or after study admission
Major Results reported by authors:
- The median change in insulin requirement from baseline was -8 units for the test formula and +2 units for the control formula at 56 days; at 84 days the median change in insulin requirement was -21.4 units for the test formula and +4.15 for control. (p<0.0001 both time points).
- The median change in fasting glucose after 56 days was -1.42 mmol/L (-25.5 mg/dL) for the test formula and -0.46 mmol/L (-8.3 mg/dL) in the control group (p<0.0021). After 84 days, median change in fasting glucose was -2.17 mmol/L (-39.0 mg/dL) in the test group and -0.67 mmol/L (-12.1 mg/dL) in the control group (p<0.0001).
- Median HgbA1C was not significantly different between the groups after 84 days (-1.30% under test treatment and -1.20% in the control group).
- The median change in afternoon blood glucose after 84 days was -2.36 mmol/L (-42.5 mg/dL) for the test formula and -0.49 mmol/L (-8.9 mg/dL) in the control group (p<0.0001).
- There were no significant differences between groups with regards to hyperglycemia, hypoglycemia, or feeding tolerance "safety issues."
"Long-term tube feeding with a disease-specific enteral formula was safe and well tolerated in type 2 diabetic patients with neurological disorders. When compared with a standard diet, total insulin requirement decreased significantly with less hypoglycemia. In addition, fasting glucose and afternoon glucose were significantly lowered, resulting in improved glycemic control."
Some positive aspects of this study are that this was a multicenter study, subjects were randomly allocated into groups, and it was double-blind so that neither the patients nor their caregivers knew which formula they were receiving.
One consideration that our group discussed was the large number of subjects that dropped out of the study, with a disproportionate number of dropouts between the two groups. At day 56, nearly 50% of the control group, but less than 30% of the treatment group had dropped out. Only 40% of those patients randomized to control completed the study, while 65% of those patients randomized to the test treatment completed the study. It is notable that the most common reason for study discontinuation was resumption of ability to swallow (followed by transfer to nursing home). Overall, 21% of patients randomized to control regained ability to swallow compared to 7.7% of those randomized to treatment. The authors do not provide statistics for outcomes such as discharge and recovery of swallow function. Although it would appear that those patients receiving the standard formula had improved recovery of swallow function, considering the relatively small number of patients (11 versus 4 for recovery of swallow function) this may be a random difference not related to the feeding formulas. It would require a much larger study to answer outcome questions, but does raise the question about the need for specialized formulas if overall outcomes are not significantly different.
The dropouts are important, because despite referring to the groups as "intention to treat," the major results presented are based on data that was available on days 56 and 84. This means that the number of patients analyzed was small (62 total patients day 56, 55 patients day 84) which increases the chance of a Type 1 error (thinking the treatment works, when it really does not). The authors do report on data analyzed with all patients, with missing values replaced by the last value carried forward. However, they only mention that a combination of all endpoints together reaches statistical significance, and do not provide details on the individual endpoint values.
The other aspect that we discussed is the fact that the disease-specific formula contained a large amount of chromium (400mcg in 1500mL) that goes beyond usual dietary intakes. It is unclear how much of the effect of the formula was related to the chromium alone, since other studies of chromium supplementation have reported an insulin-sparing effect (1).
Although the authors mention a reduction in hypoglycemic events in the abstract with the test formula, it should be noted that there was no significant difference between hypoglycemic episodes between the two groups.
Our Take Home message:
The disease-specific formula tested appeared to have a medication sparing effect (approx. 25 units insulin/day at 84 days) and resulted in an improvement in fasting glycemic control when analyzed on the relatively small number of patients that completed the study. A larger study with true intention to treat analysis that includes improvement in clinical outcomes as well as cost-effectiveness should be completed before routine use can be endorsed.
- 1) Balk EM, Tatsioni A, Lichtenstein AH, et al. Effect of chromium supplementation on glucose metabolism and lipids: a systematic review of randomized controlled trials. Diabetes Care. 2007;30(8):2154-63.
We also discussed the following two articles:
- Holick MF, Biancuzzo RM, Chen TC, et al. Vitamin D2 is as effective as vitamin D3 in maintaining circulating concentrations of 25-hydroxyvitamin D. J Clin Endocrinol Metab. 2008 Mar;93(3):677-81.
- Armas LAG, Hollis B, Heaney RP 2004 Vitamin D2 is much less effective than vitamin D3 in humans. J Clin Endocrinol Metab 89:5387-5391.
regarding supplementation of Vitamin D2 vs D 3. It is too much to cover here, but our conclusions were that it does not matter which form you use, but most of the commercially available vitamin D is D3 anyway...
- Our next Weekend Warrior 2 day mini-traineeship program is scheduled for Saturday and Sunday, July 11 - July 12, 2009. If you know anyone who might not be able to get away for our full week traineeship, please let them know about our weekend program, and to check out our website for full information--we have several slots left!
- Our next 2 webinars are: June 16th: Adult Parenteral Nutrition--Kate Willcutts MS, RD, CNSD and July 14th: Nutrition Support in Critical Illness--Joe Krenitsky MS, RD.
- Check out the full schedule of webinar programs at:
See the latest Practical Gastroenterology articles:
- Rosner M. Metabolic Acidosis in Patients with Gastrointestinal Disorders: Metabolic and Clinical Consequences. Practical Gastroenterology 2009;XXXIII(4):42.
- Hise ME, Fuhrman MP. The Effect of Diabetes-Specific Enteral Formulae on Clinical and Glycemic Indicators. Practical Gastroenterology 2009;XXXIII(5):20.
- Available at:
Joe Krenitsky MS, RD
Carol Rees Parrish MS, RD
PS - Please feel free to forward this on to friends and colleagues.