Imprecision nutrition? Duplicate meals result in unreliable individual glycemic responses measured by continuous glucose monitors across three dietary patterns in adults without diabetes

BackgroundContinuous glucose monitors (CGMs) are being used to characterize postprandial glycemic responses and thereby provide personalized dietary advice to minimize glycemic excursions. However, the efficacy of such advice depends on reliable CGM responses. ObjectiveTo explore within-subject variability of CGM responses to duplicate meals in an inpatient setting. MethodsCGM data were collected in two controlled feeding studies (NCT03407053 and NCT03878108) in 30 participants without diabetes capturing 1056 meal responses in duplicate [~]1 week apart from four dietary patterns. One study used two different CGMs (Abbott Freestyle Libre Pro and Dexcom G4 Platinum) whereas the other study used only Dexcom. We calculated the incremental area under the curve (iAUC) for each 2-h post-meal period and compared within-subject iAUCs using the same CGM for the duplicate meals using linear correlations, intra-class correlation coefficients (ICC), Bland-Altman analyses, and compared individual variability of glycemic responses to duplicate meals versus different meals using standard deviations (SDs). ResultsThere were weak to moderate positive linear correlations between within-subject iAUCs for duplicate meals (Abbott r=0.47, p<0.0001, Dexcom r=0.43, p<0.0001), with low within-participant reliability indicated by ICC (Abbott 0.31, Dexcom 0.14). Bland-Altman analyses indicated wide limits of agreement (Abbott -31.3 to 31.5 mg/dL, Dexcom -30.8 to 30.4 mg/dL) but no significant bias of mean iAUCs for duplicate meals (Abbott 0.1 mg/dL, Dexcom -0.2 mg/dL). Individual variability of glycemic responses to duplicate meals was similar to that of different meals evaluated each diet week for both Abbott (SDduplicate = 10.7 mg/dL, SDweek 1 =12.4 mg/dL, SDweek 2 =11.6 mg/dL, p=0.38) and Dexcom (SDduplicate = 11.1 mg/dL, SDweek 1 = 11.5 mg/dL, SDweek 2 =11.9 mg/dL, p=0.60). ConclusionsIndividual postprandial CGM responses to duplicate meals were unreliable in adults without diabetes. Personalized diet advice based on CGM measurements in adults without diabetes requires more reliable methods involving aggregated repeated measurements. This secondary analysis contains data from two trials registered at clinicaltrials.gov (NCT03407053 and NCT03878108).