10th June 2023, Dr Chee L Khoo
It’s all very confusing, isn’t it? Low fat, low carbohydrate, Mediterranean diet, Paleo diet, Nordic diet or DASH diet. There are many other diets which we haven’t even heard of. Are they any good? Do they cause harm? How can we advise our patients if we don’t know what those diets are? Which diet should our patients with type 2 diabetes be on? Well-designed dietary recommendations and nutrition therapy are essential to improve both life expectancy and quality. However, the flood of nutrition information available is of variable quality, creates controversy regarding the best approaches, and is likely to confuse both people with diabetes and health professionals.
The Diabetes and Nutrition Study Group (DNSG) of the European Association for the Study of Diabetes (EASD) recently commissioned a comprehensive systematic review and meta-analysis of published scientific literature out there to come up with evidence-based recommendations and commentary on macronutrients, foods, dietary patterns, and the broader lifestyle context of type 2 diabetes management. I will attempt to summarise the essential bits that is relevant to our patients in primary care. I will not be providing references for the recommendations as they are all in the consensus statement (1).
We will have a look at the three major macronutrients (carbohydrates, fat and protein), other food groups (nuts, lentils etc) and traditional diets (Mediterranean, Nordic etc).
Carbohydrates intake in the management of T2D
Carbohydrates intake in T2D is one of most complicated nutrients to get “right”. The amount, the proportion and the type of carbohydrate patients can have has been studied extensively. In an average person in Australia (including patients with T2D), carbohydrates make up between 45-55% of total daily energy intake (TDEI). Many studies consider lower carbohydrate diet (LCD) as diets with < 40% while very low carbohydrate diet (VLCD) as diets with < 30% of TDEI.
Meta-analyses have shown that when nutrient complete LCDs are compared head-to-head, they are equally effective as higher carbohydrate (but low-fat) diets, without any clinically significant long-term differences in glycaemic levels, lipids, blood pressure or weight management.
Extreme carbohydrate restriction has been associated with increased low-density lipoprotein cholesterol (LDL-C) levels, hypoglycaemia, ketoacidosis, and vitamin and mineral inadequacies. If patients choose to follow these extremely diet, it is important that this is done with health professional support with close follow up and monitoring. These diets are difficult to follow in the long run and there is lack of evidence on long-term effects.
Carbohydrate counting may be useful as a technique for determining mealtime insulin dose in people with type 1 diabetes. Carbohydrate counting has been associated with improved HbA1c in adults, with no detrimental effect on severe hypoglycaemia or quality of life.
In addition to the recommendations on the percentage of total daily energy intake and based on moderate to high level of evidence:
- Foods naturally high in dietary fibre should be encouraged.
- Dietary fibre intake should be at least 35 g per day (4 g per 1000 kJ).
- Minimally processed whole grains, vegetables, legumes, seeds, nuts and whole fruits should be recommended as sources of dietary fibre.
- Fibre-enriched foods and fibre supplements should be considered when sufficient intakes cannot be obtained from diet alone.
- Diets with a low glycaemic index or low glycaemic load can be recommended, provided their composition is consistent with overall diet recommendations for dietary fibre, sugars, saturated fats and protein.
- Intakes of free or added sugars should be below 10% of total energy intake.
- Non-nutritive sweeteners (NNS) can be used to replace sugars in foods and beverages.
Dietary fat intake in diabetes management
These recommendations are supported by systematic reviews and meta-analyses and are in international and WHO guidelines aimed at the general population. Naturally, they are relevant to patients with diabetes as well. They promote the consumption of foods containing primarily plant-based mono- and polyunsaturated fats (i.e. non-hydrogenated non-tropical vegetable oils e.g. olive oil, rapeseed/canola oil, soybean oil, sunflower oil, linseed oil and consumption of seeds, nuts, fish and avocado) rather than saturated or trans-fats (i.e. fats from meats and processed meats, butter, coconut products or palm oil). Cardiovascular outcomes trials indicate that mixed nuts (30 g/day) added to a Mediterranean dietary pattern reduces major cardiovascular events.
The reduction of saturated fats in the diet is recommended mainly due to their potential to elevate LDL-C which exhibit a causal relationship with atherogenesis and CVD. There is some evidence that an association between trans-fats and coronary heart disease (CHD) incidence may be due to the intake of industrially produced trans-fat rather than those naturally present in ruminant fat. Replacing saturated fats with polyunsaturated or mono-unsaturated fats result in reductions of HbA1c, fasting glucose concentration and liver fat content and improved measures of insulin resistance.
Low saturated- and trans-fat intakes should comprise <10% and <1% of total energy, respectively. Saturated fat should not be replaced with rapidly digested carbohydrates such as sugars and simple starches. The evidence for use of n-3 supplements in diabetes treatment has shown no or little benefit for CVD reduction. There is, however, some support for recommending regular consumption of fatty fish as an additional means of reducing CHD risk in people with diabetes.
Protein intake in diabetes management
How much protein should patients with T2D consume? The recommendations have also been pretty confusing. “Some but not too much” doesn’t help our us nor our patients as to how much protein they should be consuming daily. To clarify what is discussed in the literature, protein intakes >20% of total energy is considered to be “high-protein” intake while protein intake of < 10% is considered to be “low-protein” intake. Energy-reduced weight-loss diets have been extensively studied and high-protein diets are frequently recommended for weight loss and to try to minimise loss of muscle mass. High protein intake cannot be recommended as it has not been studied long term in people with type 2 diabetes. Low protein intake risks protein insufficiency.
Thus, for weight-stable, normal-weight people with diabetes a protein intake of 10–20% total energy is recommended for people under the age of 65 years with an estimated glomerular filtration rate (eGFR) >60 ml/min per 1.73m2. Higher intakes (15–20% total energy) are recommended for those aged 65 years or older. For people with T2D who have overweight or obesity with an eGFR >60, a protein intake of 23–32% may be recommended in the short term (up to 12 months) in the context of a weight-loss diet. For people with moderate diabetic nephropathy (stage 3a: eGFR <60 but >45 ml/min per 1.73m2) a protein intake of 10– 15% is recommended.
The sources of protein in the diet are supposed to be dairy and dairy substitutes, legumes with complementary whole grains, eggs, fish, poultry and lean meat. Insufficient evidence exists from clinical trials in people with type 2 diabetes to indicate preferential intake of either animal or plant protein.
A word of warning though, the evidence underpinning the above recommendations in relation to protein intake is rated as low.
Traditional diets
Mediterranean, Nordic, Vegetarian, DASH, Portfolio and many other traditional diets pretty much all emphasise the increased consumption of whole grains, whole vegetables and fruit, legumes, nuts, seeds and nonhydrogenated non-tropical vegetable oils, while minimising the consumption of meat (especially red and processed meat), sugar-sweetened beverages, sweets and refined grains. They all lead to reductions in fasting plasma glucose, body weight, LDL-cholesterol, triacylglycerols, blood pressure and reduced need for anti-hyperglycaemic medications. In addition, they are associated with lower risk of total CVD, CVD mortality, cancer mortality, CHD and stroke. When we talk about whole grains, we are talking about brown rice, whole wheat, rye, oats and barley. Processing and milling whole grains lower their capacity to reduce glycaemia.
The rapidly digestible starch and sugar (i.e. high glycaemic index) provided by whole vegetables and fruit may be a problem in some patients with diabetes if intakes are very high. For patients who like fruit juices, the evidence for benefits for 100% fruit juice appears restricted to intake from a single piece of fruit (≤150 ml). Higher intakes of green leafy vegetables rather than root vegetables or intakes of temperate fruit such as berries, citrus fruit, and apples and pears may be useful.
Legumes include pulses (e.g. beans, peas, chickpeas and lentils), oil-seed legumes (e.g. soy, peanuts) and fresh legumes (e.g. peas, string beans) and are associated with lower risk of total CVD, CHD and hypertension, and obesity incidence. Type 2 diabetes incidence was not reduced by higher intake of legumes in meta-analyses.
Nuts include tree nuts such as almonds, walnuts, pistachios, pecans, Brazil nuts, cashews, hazelnuts, macadamia nuts and pine nuts. Peanuts are also often included in assessments of nuts. Despite the concerns of the high calories in nuts, Results of RCTs of intermediate risk factors indicate reductions in established lipid targets (LDL-cholesterol, total cholesterol, triacylglycerols) [133, 142, 143] and markers of glycaemia (HbA1c, fasting blood glucose) [144] at median intakes of 50–67 g/day.
In summary, in patients with T2D, a discussion of diet is a must irrespective of the stage of their disease. The above review by the EASD will assist your discussions with your patients with T2D.
Reference:
Evidence-based European recommendations for the dietary management of diabetes. The Diabetes and Nutrition Study Group (DNSG) of the European Association for the Study of Diabetes (EASD). Diabetologia. Diabetologia (2023) 66:965–985. https://doi.org/10.1007/s00125-023-05894-8