The relationship between carbohydrate intake and cardiometabolic risk remains controversial in people with type 1 diabetes (T1D). In a systematic review of dietary carbohydrate (CHO) content [ranging from very-low (<10% of daily caloric intake) to high (≥45%kcal)], we performed a linear spline meta-regression on glycemic control, total daily insulin dose (TDD), and blood lipids. Inclusion was met by 101 studies with 45,972 participants (54% female, 45% pediatric). CHO intake was non-linearly associated with HbA1c (Figure 1) and TDD. Below the knot of 225 g CHO/day (mean intake), CHO intake was positively associated with HbA1c (+0.52% per 50 g, p<0.001) and TDD (+0.12 IU/kg/day per 50 g, p<0.001). Above the knot, CHO intake was positively associated with higher standard deviation on continuous glucose monitoring (+11.6 mg/dL per 50 g, p=0.031), but not associated with HbA1c or TDD. CHO intake was not associated with LDL-C, HDL-C or triglycerides. The ADA target of HbA1c <7% was achieved in 10% of high-, 24% of moderate- (26-44%kcal), 78% of low- (10-25%kcal), and 100% of very-low-CHO studies (p<0.001). In summary, lower CHO intake was associated with improved glycemic control, without - unlike previously described - altering blood lipids. While observational in nature, these findings suggest that dietary CHO reduction may help reduce cardiometabolic risk in T1D.
A.P. Koutnik: Advisory Panel; Simply Good Foods, Nutrishus Brands. T. Wood: None. K. Storoschuk: None. B.S. Lennerz: None.
NIH NIDDK (R01DK135884, K23DK119546, R03DK123541)
