Published in the BMJ at the end of 2016, results of a contest between two enemies fated to duke it out for (seemingly) all eternity.
The first ever diet trial to 100% resolution of prediabetes with diet is neither low fat nor low carb. It’s high protein. (I wish it was also low-PUFA)
Study participants ran a roughly 500 KCal per day deficit, meaning it was 500 KCal below the amount of calories they needed to avoid weight loss. It was also least 1200 Kcal per day less than the participants had probably been consuming (the average American gets 3000-3600 Kcal daily). That degree of calorie restriction forces the body to start burning body fat.
Although both study groups were eating the same amounts of calories, the different macronutrient compositions in the two groups led to very different body composition and biomarker changes. Once again, showing that as much as Coke and the sugar industry would like us to believe a calorie is a calorie, the fact is not all calories are equal in the eyes of your body.
As a side note, but an important one, the degree of calorie restriction was equivalent to skipping a meal every day, or fasting every three days. Keep in mind that these fasting-related strategies may actually be easier and healthier because of the superior insulin-lowering and fat-burn promoting effects compared to the frequent small meals plus snacks strategy employed here.
Here’s the study summary:
Who: 24 overweight men and women with prediabetes and an average A1c of 6. (Nearly 200 possible study subjects were excluded due to metformin use, abnormal thyroid, triglycerides, LDL, or blood pressure, and a handful of other reasons.)
Diet composition: Lean meats and fish, vegetables, nuts, whole grains, fruits, legumes and margarine (yikes), in varying amounts depending on which group–see below:
High protein diet: 30% kcals protein, 40% kcals CHO, 30% kcals fat. So for the average participant on 1800 kcal, the high protein diet provided 180g of carb and 135g of protein,
High carb diet: 15% kcals protein, 55% kcals CHO, 30% kcals fat. For the average participant on 1800kcl, the high carb diet provided 247.5g of carb. and 67.5g of protein.
Result highlights after the six month study period ended, better performing diet in green:
Weight loss (% of initial body weight)
High Carb: 11.3 | High Protein: 9.8
Lean Mass Change (% of initial body weight)
High Carb: -3.02 | High Protein: +2.55
Fat Mass Change (% of initial body weight)
High Carb: -3.55 | High Protein -2.49
A1c Change from Baseline (%mmol/mol)
High Carb -0.2 | High Protein -0.54
HOMA IR Change from Baseline – Marker of Insulin Resistance (bigger drop is is better)
High Carb: -1.4 | High Protein -3.21
ISI – Marker of Insulin Sensitivity (bigger rise is better)
High Carb: +1.1 | High Protein +4.2
Triglyceride Change from Baseline (mg/dL)
High Carb: -11.4 | High Protein -69.4
LDL Change from Baseline (mg/dL)
High Carb: -4.3 | High Protein 23.0
Blood Pressure Change from Baseline (systolic/diastolic)
High Carb: -8/-7 | High Protein -14/-9
TNF alpha Change from Baseline (pg/mL) – Marker of inflammation
High Carb: -2.9 | High Protein -8.0
IL-6 Change from Baseline (pg/mL) – Marker of inflammation
High Carb -1.63 | High Protein: -4.02
Reactive Oxygen Species Change From Baseline (umol/L) – Marker of oxidative stress
High Carb -0.4 | High Protein: -1.3
A key marker that didn’t change much in either group: HDL
Here’s my take:
- Bariatric surgeons lose. Those bariatric surgeons who claim that only by doing a gastric bypass at the cost of $30K can one cure pre-diabetes can no longer legitimately make that claim.
- Calories count but there’s more to it. The fact that both groups lost weight shows that yes calories do count. But the kind of calories plays a big role in where the weight loss comes from.
- High carb + low cal is dangerous. The high carb group lost an alarming amount of lean mass, not specified if from bone, muscle or other organs but all are possible, even brain mass loss, and have been shown to decline in other studies.
- Protein supports lean tissue. The fact that the high protein group gained lean mass suggests they were possibly exercising more than the other group, which would be very interesting if that were the case. Another fascinating possible explanation is that the high protein intake at every meal stimulated muscle growth, per the predictions of Dr Leyman.
- Carbs keep LDL up. LDL barely budged in the high carb group, in spite of greater weight loss.
- Carbs promote insulin resistance. The markers of insulin sensitivity and resistance barely budged in the high carb group in spite of a degree of weight loss far in excess of the 5% that doctors promise patients will lead to improved health outcomes such as reduced risk of complications of insulin resistance.
- …Or maybe protein promotes insulin sensitivity? The high protein diet reduced both markers of oxidative stress quite impressively. Oxidative stress is controlled in part by antioxidant enzyme systems, which are composed of protein and amino acids.
- Vegetable oils damage HDL. HDL did not improve because the fats in the diet came primarily from refined vegetable oils, which damage lipoproteins and low HDL is the canary in the coal mine for damaged lipoproteins. If they’d used natural fat sources like butter instead of margarine, I think HDL would have improved more.
- The most interesting group went missing. If they’d included a third group with carbs at 100, protein at 80 and allowed the balance to come from natural fat, I think that would have been very interesting.
Both arms of this study are nothing like the diet that I recommend, being too high in carb and unnatural fat. Still, this offers testament to the fact that simply switching from overeating to restricting calories will bring some benefits, which is why we see so many diets out there that can legitimately claim to be healthier than the standard American 3000+ calorie diet.
“Remission of PreDiabetes to Normal Glucose Tolerance in Obese Adults With High Protein Versus High Carbohydrate Diet” Randomized Control Trial. Frankie B Stentz; Amy Brewer; Jim Wan; Channing Garber; Blake Daniels; Chris Sands; Abbas E Kitabchi BMJ Open Diabetes Res Care. 2016;4(1)