Additional Insulin Is Required in Both the Early and Late Postprandial Periods for Meals High in Protein and Fat: A Randomized Trial

A randomised controlled trial of additional bolus insulin using an insulin-to-protein ratio compared with insulin-to-carbohdrate ratio alone in people with type 1 diabetes following a carbohydrate-restricted diet

AIMS

Postprandial hyperglycemia can be problematic for people with type 1 diabetes (T1DM) following carbohydrate-restricted diets. Bolus insulin calculated for meal protein plus carbohydrate may help. This study evaluated the effect of additional bolus insulin using an insulin-to-protein ratio (IPR) on glycaemic control.

MATERIALS AND METHODS

Participants with T1DM aged ≥18-years were randomly allocated (1:1) to either carbohydrate and protein-based, or carbohydrate-based insulin dosing alone for 12 weeks while following a carbohydrate-restricted diet (50-100 g/day). Measurement of HbA1c and continuous glucose monitoring occurred at baseline and 12 weeks, with assessment of participant experience at 12 weeks.

RESULTS

Thirty-four participants were randomised, 22 female, mean(SD): age 39.2 years (12.6) years; diabetes duration 20.6 years (12.9); HbA1c 7.3 % (0.8), 56.7 mmol/mol (9.2). Seven in each group used insulin pump therapy. HbA1c reduced at 12 weeks with no difference between treatments: mean (SD) control 7.2 % (1.0), 55.7 mmol/mol (10.6); intervention 6.9 % (0.7), 52.3 mmol/mol (7.2) (p = 0.65). Using additional protein-based insulin dosing compared with carbohydrate alone, there was no difference in glycaemic variability, time spent in euglycemic range (TIR), or below range. Participants using IPR reported more control of their diabetes, but varying levels of distress.

CONCLUSIONS

Additional bolus insulin using an IPR did not improve glycaemic control or TIR in patients with well controlled T1DM following a carbohydrate-restricted diet. Importantly, the use of the IPR does not increase the risk of hypoglycemia and may be preferred.

A new proposal for a second insulin bolus to optimize postprandial glucose profile in adolescents with type 1 diabetes

AIMS

To evaluate whether a second insulin bolus, calculated with a new approach, could improve postprandial glucose (PPG) after the intake of real-life high-fat (HF) and high-protein (HP) mixed meals.

METHODS

Fifteen adolescents with T1D treated with non-automated insulin pumps and CGM were enrolled. Patients received standard, HF and HP mixed meals treated with one pre-meal insulin bolus; based on differences in PPG between standard, HF and HP meals, correction boluses were calculated (30% and 60% of pre-meal bolus for HF and HP meals, respectively). Then patients received the same HF or HP meal treated with pre-meal bolus plus second insulin bolus after 3 h. Differences between postprandial variables after HF and HP meals treated with one or two insulin boluses were assessed by paired Student's t-test.

RESULTS

Treating HF and HP meals with two insulin boluses significantly reduced the postprandial BG-AUC (21% and 26% respectively, p < 0.05), increased %TIR (from 52.5 to 78.3% for HF meal; from 32.7 to 57.1% for HP meal; p < 0.01), and reduced mean BG and %TAR (p < 0.01), with no differences in %TBR.

CONCLUSIONS

The new way to calculate and administer correction boluses 3 h after HF and HP meals is effective and safe in reducing PPG and the hypoglycemia risk.

Protein Ingestion in Reducing the Risk of Late-Onset Post-Exercise Hypoglycemia: A Pilot Study in Adolescents and Youth with Type 1 Diabetes

Dietary protein causes dose-dependent hyperglycemia in individuals with type 1 diabetes (T1D). This study investigated the effect of consuming 50 g of protein on overnight blood glucose levels (BGLs) following late-afternoon moderate-intensity exercise. Six participants (3M:3F) with T1D, HbA1c 7.5 ± 0.8% (58.0 ± 8.7 mmol/mol) and aged 20.2 ± 3.1 years exercised for 45 min at 1600 h and consumed a protein drink or water alone at 2000 h, on two separate days. A basal insulin euglycemic clamp was employed to measure the mean glucose infusion rates (m-GIR) required to maintain euglycemia on both nights. The m-GIR on the protein and water nights during the hypoglycemia risk period and overnight were 0.27 ± 043 vs. 1.60 ± 0.66 mg/kg/min (p = 0.028, r = 0.63) and 0.51 ± 0.16 vs. 1.34 ± 0.71 mg/kg/min (p = 0.028, r = 0.63), respectively. Despite ceasing intravenous glucose infusion on the protein night, the BGLs peaked at 9.6 ± 1.6 mmol/L, with a hypoglycemia risk period mean of 7.8 ± 1.5 mmol/L compared to 5.9 ± 0.4 mmol/L (p = 0.028) on the water night. The mean plasma glucagon levels were 51.5 ± 14.1 and 27.2 ± 10.1 ng/L (p = 0.028) on the protein and water night, respectively. This suggests that an intake of protein is effective at reducing the post-exercise hypoglycemia risk, potentially via a glucagon-mediated stimulation of glucose production. However, 50 g of protein may be excessive for maintaining euglycemia.

Efficacy of insulin dosing algorithms for high-fat high-protein mixed meals to control postprandial glycemic excursions in people living with type 1 diabetes: A systematic review and meta-analysis

Optimizing postprandial blood glucose (PPG) levels after mixed meals that contain high fat and protein remains a challenge in the treatment of type 1 diabetes. This study evaluated the efficacy of different algorithms used for dosing insulin based on counting units of high fat and high protein (HFHP) meals with the current conventional method of counting carbohydrates alone to control PPG excursions. The MEDLINE, EMBASE, and Cochrane electronic databases were searched, with the analysis restricted to randomized control trials (RCTs). The primary outcome was the PPG (mean and standard deviation) at 240 min. The pooled final estimate was the mean difference (MD) of the PPGs at 240 min using random effect models to account for heterogeneity. In total, 15 studies were identified and included in the systemic review, of which 12 were RCTs, and three studies were non-randomized trials. The pooled MD of the PPG at 240 min was in favor of additional insulin doses in HFHP meals compared to the carbohydrate counting alone. The statistically significant results favored the combined bolus (30:70) that split over 2 h in insulin pump therapy with pooled MD of the PPG, 240 min of -24.65; 95% CI: -36.59, -8.41; and heterogeneity, 0%. Other statistically significant results favored the additional insulin added to insulin to carb ratio (ICR) of meal bolus (25-60% ICR) in multiple daily injections therapy with the pooled MD of PPG at 240 min, -21.71; 95% CI: -38.45, -4.73; and heterogeneity, 18%. Insulin treatment based on fat and protein content, in addition to carbohydrate counting, is more effective than the carbohydrate counting method alone; however, further research is warranted to determine the best equation for fat and protein counting, particularly in people with multiple daily injections.

Postprandial glucose variability in type 1 diabetes: The individual matters beyond the meal

AIM

To explore intraindividual (between-meals) and interindividual (between-subjects) variability of postprandial glucose response (PGR) in type 1 diabetes (T1DM).

METHODS

Data were taken from five cross-over trials in 61 subjects with T1DM on insulin pump wherein the effects of different dietary components or the intraindividual-variability of PGR to the same meal were evaluated by CGM. Predictors (type of meal or nutrient composition) of early (iAUC_0-3h), late (iAUC_3-6h), total (iAUC_0-6h), and time-course of postprandial blood glucose changes (iAUC_{3-6hminus0-3h}) were evaluated using two mixed-effect linear regression models considering the patient's identification number as random-effect.

RESULTS

High-glycemic-index (HGI) and low-glycemic-index meals were the best positive and negative predictors of glucose iAUC_0-3h, respectively. A Low-Fat-HGI meal significantly predicted iAUC_{3-6hminus0-3h} (Estimate 3268; p = 0.017). Among nutrients, dietary fiber was the only significant negative predictor of iAUC_0-3h (Estimate -550; p < 0.001) and iAUC_0-6h (Estimate -742; p = 0.01) and positive predictor of iAUC_{3-6hminus0-3h} (Estimate 336; p = 0.043). For all models, the random-effect patient was statistically significant (p < 0.001 by ANOVA).

CONCLUSION

Beyond the meal characteristics (including glycemic index, fat and fiber content), individual traits significantly influence PGR. Specific interindividual factors should be further identified to properly predict glucose response to meals with different composition in individuals with T1DM.

Effects of Dietary Fat and Protein on Glucoregulatory Hormones in Adolescents and Young Adults With Type 1 Diabetes

CONTEXT

Dietary fat and protein impact postprandial hyperglycemia in people with type 1 diabetes, but the underlying mechanisms are poorly understood. Glucoregulatory hormones are also known to modulate gastric emptying and may contribute to this effect.

OBJECTIVE

Investigate the effects of fat and protein on glucagon-like peptide (GLP-1), glucagon-dependent insulinotropic polypeptide (GIP) and glucagon secretion.

METHODS

2 crossover euglycemic insulin clamp clinical trials at 2 Australian pediatric diabetes centers. Participants were 12-21 years (n = 21) with type 1 diabetes for ≥1 year. Participants consumed a low-protein (LP) or high-protein (HP) meal in Study 1, and low-protein/low-fat (LPLF) or high-protein/high-fat (HPHF) meal in Study 2, all containing 30 g of carbohydrate. An insulin clamp was used to maintain postprandial euglycemia and plasma glucoregulatory hormones were measured every 30 minutes for 5 hours. Data from both cohorts (n = 11, 10) were analyzed separately. The main outcome measure was area under the curve of GLP-1, GIP, and glucagon.

RESULTS

Meals low in fat and protein had minimal effect on GLP-1, while there was sustained elevation after HP (80.3 ± 16.8 pmol/L) vs LP (56.9 ± 18.6), P = .016, and HPHF (103.0 ± 26.9) vs LPLF (69.5 ± 31.9) meals, P = .002. The prompt rise in GIP after all meals was greater after HP (190.2 ± 35.7 pmol/L) vs LP (152.3 ± 23.3), P = .003, and HPHF (258.6 ± 31.0) vs LPLF (151.7 ± 29.4), P < .001. A rise in glucagon was also seen in response to protein, and HP (292.5 ± 88.1 pg/mL) vs LP (182.8 ± 48.5), P = .010.

CONCLUSION

The impact of fat and protein on postprandial glucose excursions may be mediated by the differential secretion of glucoregulatory hormones. Further studies to better understand these mechanisms may lead to improved personalized postprandial glucose management.

Substantial Intra-Individual Variability in Post-Prandial Time to Peak in Controlled and Free-Living Conditions in Children with Type 1 Diabetes

The optimal time to bolus insulin for meals is challenging for children and adolescents with type 1 diabetes (T1D). Current guidelines to control glucose excursions do not account for individual differences in glycaemic responses to meals. This study aimed to examine the within- and between-person variability in time to peak (TTP) glycaemic responses after consuming meals under controlled and free-living conditions. Participants aged 8–15 years with T1D ≥ 1 year and using a continuous glucose monitor (CGM) were recruited. Participants consumed a standardised breakfast for six controlled days and maintained their usual daily routine for 14 free-living days. CGM traces were collected after eating. Linear mixed models were used to identify within- and between-person variability in the TTP after each of the controlled breakfasts, free-living breakfasts (FLB), and free-living dinners (FLD) conditions. Thirty participants completed the study (16 females; mean age and standard deviation (SD) 10.5 (1.9)). The TTP variability was greater within a person than the variability between people for all three meal types (between-person vs. within-person SD; controlled breakfast 18.5 vs. 38.9 min; FLB 14.1 vs. 49.6 min; FLD 5.7 vs. 64.5 min). For the first time, the study showed that within-person variability in TTP glycaemic responses is even greater than between-person variability.