Nov. 11, 2010 – For the first time, a study suggests that a change in diet can influence the process leading to type 1 diabetes in infants at high risk for the disease.
There's no proof that the dietary change -- using a predigested infant formula instead of normal formula when a mother stops breastfeeding -- actually prevents type 1 diabetes.
But compared to babies fed normal formula, those who got the special formula were 50% less likely to develop the kind of antibodies that predict diabetes, reports Mikael Knip, MD, PhD, of the University of Helsinki, Finland.
Predigested formula -- also called hydrolyzed formula -- is a formula in which the protein content has been broken down into smaller proteins that can be digested more easily, according to the American Academy of Pediatrics.
The study findings apply only to infants that carry genetic factors indicating high diabetes risk, and not to children in the general population.
"My interpretation is that the dietary intervention reduces the initiation of the disease process leading to type 1 diabetes," Knip tells WebMD.
Not so fast, warn University of Massachusetts diabetes experts David M. Harlan, MD, and Mary M. Lee, MD. While they praise Knip's team for a carefully conducted study, they note that this small pilot study was far from definitive.
"It is very difficult to draw conclusions from this data," Lee, a pediatric endocrinologist, tells WebMD. "Their conclusion that there is a suggestion that the [predigested] formula might be protective is exactly that: a suggestion."
"Just by chance, those kids randomized to the [predigested] formula got that formula a little bit later. They were consuming breast milk a little bit longer, and maybe that explains it," Harlan tells WebMD. "This study adds further interest to the association between early diet and diabetes, but it is not the final answer. I don't want anybody to draw the conclusion that this formula prevents diabetes."
Sixteen of the study's 230 children went on to develop diabetes over a median follow-up period of 7.5 years. Seven of these kids were in the group scheduled to receive the predigested formula. But Knip notes that only four of them actually got the special formula.
Knip is more positive about the findings than either Harlan or Lee.
"I think that if you have a diabetic family member and have a newborn baby, the first recommendation is to breastfeed as long as possible," Knip says. "But then our recommendation is you should use a [predigested] formula. But I would not recommend this to the general population -- only to children with a first-degree relative who has type 1 diabetes."
Harlan and Lee agree with only part of this advice: That parents from families at higher risk of diabetes should breastfeed for as long as possible.
Phyllis W. Speiser, MD, chief of pediatric endocrinology at Cohen Children's Medical Center of New York and professor of pediatrics at Hofstra University, agrees with Harlan and Lee that it's premature to recommend predigested formula, which costs significantly more than regular formula.
"These formulas are fairly expensive, it would be a very costly thing to undertake," Speiser tells WebMD. "This is not exactly ready for prime time."
The predigested formula used in the Knip study was Nutramigen, from Mead Johnson. The manufacturer provided the formula for the study but was not otherwise involved in funding or directing the study.
A much larger and more tightly controlled study, which will show whether Nutramigen actually can prevent type 1 diabetes, already is under way. Preliminary findings are expected in early 2013, but final results won't be in until 2017.
Type 1 Diabetes Trigger Still Unknown
The diabetes marker that is the focus of the Knip study is the appearance of antibodies to insulin-producing beta cells in the pancreas. When enough of these cells die, diabetes results.
Most experts now think these antibodies are a result, and not the main cause, of an immune response that is killing off beta cells.
There are five of these antibodies, and the appearance of any one is a bad sign. It's not clear exactly how bad, but Knip says that kids who have two or more of these antibodies have about a 90% chance of eventually developing type 1 diabetes.
People with a certain genetic makeup are at increased risk of autoimmune diseases, including diabetes. But only a small percentage of people with this genetic signature get diabetes.
Exactly why children develop type 1 diabetes remains a mystery. Genetic susceptibility plays a major role, but it's not the whole story. A person with an identical twin shares the same genetic makeup, yet if one twin has type diabetes there's only a 50% chance that the other will, too.
Something in the environment is triggering diabetes. What is it? A huge international study -- The Environmental Determinants of Diabetes in the Young or TEDDY study -- is trying to find out.
Study investigators are looking for people willing to enroll their newborns in the 15-year study. Both families with and without diabetes are eligible for the study. More information on the study is available at the ClinicalTrials.gov web site.
The Knip report -- and an editorial by Harlan and Lee -- appear in the Nov. 11 issue of the New England Journal of Medicine.