#18: The Honeymoon Period in Type 1 Diabetes

One of my blog followers, Svet, asked if there were ways to extend the “honeymoon” period in type 1 diabetes (T1D). The “honeymoon” is defined as the period of time shortly after the clinical onset of T1D in which a transient fall in insulin requirements occurs due to improved beta-cell function. It is believed to occur as a result of reduction in blood glucose (BG) with insulin therapy which in turn relieves “glucose toxicity.” Glucose toxicity is a term used to describe the effect of prolonged hyperglycemia on suppressing insulin production by beta-cells. The honeymoon period is a natural remission of T1D that is usually temporary, ending with a gradual or abrupt increase in exogenous insulin requirements within 3 years of diagnosis in most young children, especially those with the HLA-DR3/4 genotype. Total remissions (not requiring exogenous insulin) have been reported in 2–12% and partial remissions (requiring reduced exogenous insulin) in 18–62% of young T1D patients during the honeymoon period. Older age and less severe initial presentation of T1D and low or absent islet-cell antibodies have been consistently associated with deeper and longer remission. Destruction of beta-cells is much slower and often only partial in older patients, 15% of whom have still some beta-cell function preserved 10 years after diagnosis. Most studies agree that preserved beta-cell function is associated with better glycemic control (lower HbA1c) and preserved alpha-cell glucagon response to hypoglycemia.

This study of 103 T1D children found that partial remission occurred in 71, being complete in three. The length of time until remission was 28.6 +/- 12.3 (mean +/- SD) days. The duration of remission was 7.2 +/- 4.8 months. They found that young age and severe disease at presentation was associated with decreased residual beta-cell function that is reflected by a lower incidence of partial remission. They did not investigate ways to extend the honeymoon period.

This study looked at factors which might predict honeymoon frequency and duration. “More than 80% of the children experienced a partial remission, which lasted more than 12 months in 41.7% and at least 24 months in 16.4% of cases. The mean duration of the remission period was 11.7 +/- 8.9 months. Age at diagnosis was the only pre-treatment factor which, on stepwise regression analysis, affected both partial remission duration and insulin requirement at the end of follow-up.” They too did not investigate ways to extend the honeymoon period.

This small observational study examined the influence of regular physical activity before and after the onset of T1D on the presence of a functional reserve of beta-cells. “One group (n = 8) exercised regularly (5 or more hours/week) before onset and continued doing so with the same regularity. The second group (n = 11) either did not perform physical activity or did so sporadically. The more active group debuted with and maintained significantly lower HbA(1c) levels and insulin requirements compared to the more sedentary group. C-peptide levels were only significantly higher in the active group at the moment of onset compared to the sedentary group. Altogether, the obtained results seem to indicate that physical activity allows a better control at the moment of onset regarding glycaemic control, residual endocrine pancreatic mass and subsequent insulin requirements.”

Prolongation of Honeymoon Period with a Low Carbohydrate Diet

Although I was unable to find any research studies testing the effect of a low carbohydrate diet on prolongation of the honeymoon period, Dr. Fran R. Cogen noted the following in 2008,

“The biggest question is how someone can prolong the honeymoon period. Clearly, the longer you prolong the honeymoon period, the longer you have control of blood sugars and the chance to delay potential, future complications related to high blood sugars. This is a huge question prompting much funding and research. Doctors and scientists do know that remaining on small amounts of insulin seems to prolong the honeymoon period (to decrease the “glucose toxicity” surrounding islet cells). Other suggestions include the initiation of a low carbohydrate diet. If one consumes low amounts of carbohydrate, less insulin will be required to enable glucose transport into the cells and theoretically cause less stress on the islet cells. Ironically, the low carbohydrate diet was the treatment of choice before the discovery of insulin to avoid the symptoms resulting from the inability of the islets to metabolize glucose. These children starved and looked extremely malnourished based on photographs included in past journal articles of the “founders” of insulin, Banting, Best, and Mcleod. I am aware of no studies that have positively demonstrated that a low carbohydrate diet alone will prolong the honeymoon period.”

This is a case report of the effect of a low carbohydrate diet on prolonging the honeymoon period.

“We report a patient with a prolonged honeymoon phase for four years while maintaining low carbohydrate intake and healthy life style habits. As type 1 DM is an autoimmune disease, reducing insulin secretion through a low carbohydrate diet and optimizing insulin sensitivity through exercise in this case may reduce antigen exposure to the immune system. This could play a role in maintaining a longer honeymoon period than usual.”

There have been several clinical trials of immunosuppressants to stop the immune destruction of beta-cells in T1D in animals and humans. However, the trials in humans have not been very fruitful unfortunately and this remains an active area of research. Additionally, replacement of beta-cells with islet cell transplants has also been attempted but with only partial and short-term improvements.

For the time being, I believe a low carbohydrate diet and exercise with maintenance of near-normal blood glucose to avoid the “glucose toxicity” that suppresses remaining beta-cell function is the best option for not only prolonging the honeymoon period but for treatment of T1D in general.


  1. Svet Pavlovsky

    Thank you for providing such a useful information! I am trying to do a low carb diet (for my kid) and our first A1C after diagnosis was already below 6. The doctor was surprised and “concerned” that he does not get enough carbs (and almost no insulin..) Reading through your blog, I am more peaceful with our new lifestyle. I told her that our body can create energy from fat. And actually the ketogenic diet is used to treat some other diseases the doctor told me. The doctors know about the ketogenic diet but do not recomend nor encourage it. Sadly…
    Hope our honeymoon will last for long with that diet 🙂


      • Svet Pavlovsky

        Dear Prabhu, I don’t know for sure if my son is still on the honeymoon phase, probably not. He is on low carb diet and it helps a lot to obtain normal blood sugars most of the times. We avoid all grains, no sugar drinks, and basically avoid food that contain high amount of carbohydrates. I do bake him bread from coconut flour and he eats mainly home food: meat, cheese, vegetables and nuts.


  2. Stefan

    This is an interesting study stating that the honeymoon period can be much longer than we expect, >40 years! I am pretty sure that this happened to me, I never had a C-peptide test (until 1 year ago when it was negative) but I think that since my diagnosis >30 years ago at age 13, I produced a significant amount of insulin until about 10 years ago (I’m guessing wildly at 20-40% of daily need). I think so because at that time I could eat lots of bad things like soft drinks and chocolate without any real high blood sugars, I don’t have any HbA1c values left from this time but I think my values was in the range of 6.5 – 7.0, I was never taking any blood test tests, not adjusting doses much and eating/drinking things that diabetics really should not.
    At about 10 years ago, my values shoot up really bad, I had to stop drinking sugared soft drinks and cut down all the other sweet stuff I was eating, but my values still got worse. I have discussed this with all my diabetic doctors I have had since then (a few 😦 ) but none have even acknowledged that this is possible.
    Now eating low carb/ketogenic, I really wonder what would have happened if I had used this diet then, It might have been possible for me to take a basal dose only and cover meals with my own produced insulin, which I think kind of happened when I covered normal meals with bolus insulin and my own produced insulin covered the spikes I induced by my gorging on sweet stuff giving me a decent blood sugar the next morning.


    • Keith Runyan, MD

      Thanks for sharing Stefan. Your hypothesis sounds reasonable. There are several possible mechanisms by which a ketogenic diet might extend the honeymoon period. Eliminating wheat (gluten) and dairy from one’s diet might remove potential antigen(s) that stimulate insulin and beta-cell antibody production. The lower carb content of the ketogenic diet may reduce the demand on the fewer beta-cells that remain. The reduction in systemic inflammation may also help preserve beta-cells. This topic really needs to be studied as well as the effect of ketones on beta-cells and their effect on preventing symptoms of hypoglycemia. Apparently, NIH is not interested in funding very many studies of diet therapy for diabetes.