The glycemic index (GI) and sleep: Efficacy of a high GI mixed macronutrient meal to improve sleep quality

Abstract

Carbohydrate (CHO) ingestion, especially a high glycemic index (GI) CHO meal can improve sleep in healthy individuals, but evidence is lacking regarding its efficacy to improve the poor sleep quality inherent in insomnia. The efficacy of a high GI meal to improve sleep lies in its capacity to increase the availability of tryptophan (TRP), the amino acid precursor to serotonin (5HT), which is an intermediary product to melatonin, a primary neurohormone that influences sleep initiation. The addition of fat or protein to a CHO meal lowers the glycemic response. Thus, the effects of the GI of mixed macronutrient meals on sleep remain unknown. This thesis: (1) investigated the effects of a mixed macronutrient high GI (MHGI) and low GI (MLGI) meal compared to a carbohydrate only high GI (CHGI) meal on the postprandial availability of TRP to the brain, measured by it concentration relative to other large neutral amino acids (TRP/LNAA); and (2) evaluated the efficacy of a high GI mixed macronutrient meal to improve subjective sleep and objective sleep measures using polysomnography (PSG) in a group of men and women with insomnia. A meta-analysis of the mean data on 9 PSG sleep variables commonly affected by the first night effect (FNE) was completed (Chapter 4). The results of Chapter 4 confirm the FNE in healthy sleepers and further describe the FNE by a prolonged SOL (4.46 min), delayed REML (20.55 min), an increase in xiv WASO (3.88%), and reductions in REMS (-2.17%), TST (-18.81 min), and SE (- 4.66%; all p < 0.01). In Chapter 5, there was a significant postprandial rise in plasma TRP/LNAA levels after all meals with a peak between 2-3 h (p < 0.05); the percentage rise from baseline was greatest after the CHGI meal (23%), followed by the MHGI (17%), and the MLGI meals (8%); the TRP/LNAA response was greater after CHGI compared to MLGI (p = 0.04) indicated by their respective total area under the curve (AUC). Finally, in Chapter 6, the efficacy of the MHGI meal compared to the MLGI meal to improve subjective and PSG measures of sleep quality in men and women with insomnia was investigated in the final study. The study was a repeated measure, cross-over design (n = 8, 4 men, 4 women) where participants underwent testing on 2 consecutive nights. Meals were given 3 h prior to individual habitual bedtimes. Subjective ratings of palatability, satiety, and sleepiness were assessed after each meal by a 10 cm visual analogue scale. Subjective measures of post-sleep quality were assessed using a 5- point Likert scale and the level of sleepiness post-study was assessed by a 10 cm visual analogue scale. Postprandial measures of TRP/LNAA and 5HT concentrations were assessed at baseline prior to meal consumption and for 3 h until bedtime. The average energetic load (kJ/kg) was greater for women (33.0 ± 4.1) than men (25.4 ± 3.8, p = 0.03). Participants were self-reportedly more rested after the MHGI (2.8 cm) compared to the MLGI meal (2.2 cm; p = 0.01) corresponding to “average” and “somewhat”, respectively; this was clearly evident in women but not men. There were no significant differences between the MHGI or MLGI meals on averaged 2-night PSG sleep variables. Sex differences were present. The SOL tended to be shorter in women compared to men on all testing nights; PSG SOL values were lower than subjective SOL values. Men had more EEG arousals (AI) than women after the MHGI (p < 0.001) and MLGI meal (p < 0.05). Light sleep (NREM1) tended to be greater in men than women, especially after the MLGI meal. Both meals elicited a postprandial rise plasma TRP/LNAA levels; the percentage peak rise from baseline after the MHGI meal (17%) was 3.5 fold greater than after the MLGI (8%) meal despite marginal statistical significance; the AUC after the MHGI (1.73 ± 1.3) meal was also not statistically larger than after the MLGI (0.50 ± 0.46; p = 0.12) meal. There was no significant postprandial change in PPP-5HT concentrations. The findings in the final study indicate the MHGI meal improves subjective sleep quality when provided as a regular evening meal 3 h prior to bedtime but only to women with insomnia. The PSG data demonstrated significant night-to-night variability in sleep quality in both men and women. The observed changes in blood glucose and insulin indicate the postprandial plasma TRP/LNAA response is likely mediated by the GL of a mixed macronutrient meal. However, despite an elevated TRP/LNAA response, especially after the MHGI meal, platelet poor plasma (PPP) 5HT concentration was not changed. Further studies are needed to evaluate a range of GL values from mixed macronutrient meals on the postprandial plasma TRP/LNAA concentration. Further studies are required to confirm the mechanisms underpinning related to improvements to sleep after high GI mixed macronutrient meals.