Summary: In two small acute randomised crossover trials in adults with type 1 diabetes, neither the type of dietary protein (egg, beef, chicken, salmon or whey; n=16) nor the glycaemic index of carbohydrate (high vs low GI bread; n=8) significantly altered postprandial glucose. Responses varied widely between individuals, but all between-source and between-GI comparisons were non-significant (p>0.05), and the authors concluded that insulin dosing algorithms likely do not need to distinguish protein source or GI in high-fat, high-protein meals.
PICO Summary
| Element | Detail |
|---|---|
| Population | Adults with type 1 diabetes; two acute randomised crossover trials conducted in Australia (Study 1: n=16; Study 2: n=8). Insulin dosed by each participant’s usual individualised insulin-to-carbohydrate ratio. |
| Intervention | Study 1: five test meals differing by protein type (egg, beef, chicken, salmon or whey, each 30 g protein) served with fried rice (45 g carbohydrate). Study 2: high-GI vs low-GI bread (GI 76% vs 52%) with peanut butter (19 g protein, 30 g fat). |
| Comparison | Within-subject crossover: each protein source compared against the others (Study 1); high-GI compared against low-GI carbohydrate (Study 2). Capillary glucose measured from 30 min pre-meal to 5 h postprandially. |
| Outcome | Study 1: no significant difference in glucose iAUC between protein sources (chicken 203±66, egg 263±100, beef 309±89, salmon 338±83, whey 397±115 mmol·min/L; all p>0.05); hypoglycaemia (≤3.5 mmol/L) occurred at least once with every protein but did not differ by source (p>0.05). Study 2: glucose curves were virtually identical for high vs low GI, with iAUC non-significant at 1 h (p=0.185), 3 h (p=0.538) and 5 h (p=0.694). No confidence intervals or NNT were reported. |
Expert Commentary
This is a null result, and it should be read as such rather than as evidence that protein or carbohydrate quality is irrelevant. Across both crossover trials, no statistically significant difference in postprandial glycaemia was detected between protein sources or between high and low glycaemic-index carbohydrate, and the numerically higher excursion seen with whey did not reach significance. The headline finding is therefore the absence of a measurable signal, set against substantial individual variability. The principal limitation is power: with sixteen and eight participants respectively and a wide spread of responses, these acute single-meal studies are likely underpowered to exclude clinically meaningful effects, so a non-significant p-value cannot be equated with proven equivalence. The food-based interventions could not be blinded, and the comparisons reflect one meal in controlled conditions rather than habitual eating. Can I use this with my patients? Cautiously, yes, for adults with type 1 diabetes already counting carbohydrate, it offers reassurance that obsessing over protein source or bread glycaemic index within a high-fat, high-protein meal is unlikely to be the main driver of post-meal swings, while individualised insulin adjustment and glucose monitoring remain central. Larger, adequately powered trials are needed before these reassurances are written into dosing algorithms; until then, clinicians should keep tailoring advice to the patient in front of them.
References
Li X, Wainwright A, Fio CZ, et al. Do the Types of Dietary Carbohydrate and Protein Affect Postprandial Glycemia in Type 1 Diabetes? Nutrients. 2025;17(11):1868. doi:10.3390/nu17111868
