Carbohydrate Restriction for Type 2 Diabetes.

We are often told to look to the £m charity Diabetes UK for dietary and other advice on receiving a diagnosis of Diabetes. As a pre-diabetic (HbA1C 38 mmol/mol or 5.8 % in old money) I did this and prepared a short summary :-

Moving on, I looked for evidence based ways to reduce the symptoms and severity of the disease beyond the emphasis on weight loss. Weight loss may be helpful but it is somewhat long term and unreliable, so can we do anything useful with carbohydrate restriction alone ?

10 years ago a study was published on the effect of a 5% carbohydrate / 35% protein / 60% fat weight maintenance diet (1.3 * Resting metabolic rate) in normal weight healthy subjects.The run-in diet was 60% carbohydrate/ 30% fat and a rapid reduction in the 24h insulin concentration was seen on switching to the low carb diet :-

OK, they weren't diabetic but the effect is quick and was achieved in a weight maintenance diet so no confusion from calorie restriction or weight loss. Tim Noakes, Jason Fung and Joseph Kraft are all about the insulin being the problem, so this looks promising.

My next find was a much more recent study looking at the effect of moderate carbohydrate restriction in Japanese subjects who had a diagnosis of Type 2 diabetes for ~4 years on average. These people were also of a healthy weight but at ~60 were twice as old as the short term group above.

"Moderate carbohydrate restriction" involves omitting carbohydrates at both dinner and breakfast, at dinner alone,  or halving the amount at dinner - three levels of restriction. Carbohydrate snacks were forbidden but protein and fat were acceptable, grams of nutrients were not specified. By assigning those with the highest HbA1c (>=9.0%) to the greatest level of carbohydrate restriction, and those with the lowest(<=7.4%) to the least, the changes achieved after 6 months were :-

The large reductions in HbA1c were mainly seen in the high group (red circles) with the low group (green triangles) having much less scope for improvement. 25 subjects reduced or eliminated diabetes medication and 12 started or increased. These are shown as blue and orange lines respectively :-

This graph also illustrates the wide range and generally high carbohydrate intake at baseline, and the relatively moderate intake adopted of 70 - 230 grams/day. On average the energy intake dropped 400 kcal/day  leading to an average BMI reduction of 0.9 to 24.0. Reductions in HbA1c were not however correlated with weight loss (changes in BMI) :-

Weight loss is clearly not a prerequisite to improved blood glucose and insulin levels, as further demonstrated by Gannon & Nuttall's "diet only" approach to reducing sugars and starches in their Low Biologically Available Glucose (LoBAG) diet :-

Finally we turn to  a study funded by Diabetes UK but available behind a paywall :

The abstract gives nothing away about blood glucose control, bizarrely, but displays DBUK's obsession with weight loss and fat intake. The subjects here were obese diabetics with poor glucose control under supervision of hospital consultants, so in some respects more challenging cases. The "severe dietary carbohydrate restriction" adopted was advice to limit carbs to 70 g/day in the low carb arm, however the 12 week food diary revealed an intake mean of 109.5g of carbohydrate or 33.5% of the restricted energy intake of 1290 calories per day. The low fat arm was at 169g / 45% of energy.

33 vs 45% carbohydrate does not sound like a test of "severe" carbohydrate restriction, although the intended intake was about a third lower, and the non-significant difference in HbA1c reduction between the diets is perhaps predictable. Low carb reduced HbA1c by 0.55 average compared to 0.23 on low fat. Low carb did do significantly better in terms of weight loss and reducing Total/HDL cholesterol. In seeking to explain the lack of effect on glycaemic control the authors conducted a post hoc analysis of ~75% of subjects and revealed that 85% of LC subjects on insulin were able to reduce their dose.

Many other studies are summarised in the paper by Feinman et al -

"Dietary carbohydrate restriction as the first approach in diabetes management: Critical review and evidence base"