Shift workers are known to have an increased risk of developing cardiovascular disease (CVD) compared with day workers. An important factor contributing to this increased risk could be the increased incidence of postprandial metabolic risk factors for CVD among shift workers, as a consequence of the maladaptation of endogenous circadian rhythms to abrupt changes in shift times. We have previously shown that both simulated and real shift workers showed relatively impaired glucose and lipid tolerance if a single test meal was consumed between 00:00-02:00 h (night shift) compared with 12:00-14:00 h (day shift). The objective of the present study was to extend these observations to compare the cumulative metabolic effect of consecutive snacks/meals, as might normally be consumed throughout a period of night or day shift work. In a randomized crossover study, eight healthy nonobese men (20-33 yrs, BMI 20-25kg/m2) consumed a combination of two meals and a snack on two occasions following a standardized prestudy meal, simulating night and day shift working (total energy 2500 kcal: 40% fat, 50% carbohydrate, 10% protein). Meals were consumed at 01:00/ 13:00 h and 07:00/19:00h, and the snack at 04:00/16:00 h. Blood was taken after an overnight fast, and for 8 h following the first meal on each occasion, for the measurement of glucose, insulin, triacylglycerol (TAG), and nonesterified fatty acids (NEFA). RM-ANOVA (factors time and shift) showed a significant effect of shift for plasma TAG, with higher levels on simulated night compared to day shift (p < 0.05). There was a trend toward an effect of shift for plasma glucose, with higher plasma glucose at night (p = 0.08), and there was a time-shift interaction for plasma insulin levels (p < 0.01). NEFA levels were unaffected by shift. Inspection of the area under the plasma response curve (AUC) following each meal and snack revealed that the differences in lipid tolerance occurred throughout the study, with greatest differences occurring following the mid-shift snack. In contrast, glucose tolerance was relatively impaired following the first night-time meal, with no differences observed following the second meal. Plasma insulin levels were significantly lower following the first meal (p < 0.05), but significantly higher following the second meal (p < 0.01) on the simulated night shift. These findings confirm our previous observations of raised postprandial TAG and glucose at night, and show that sequential meal ingestion has a more pronounced effect on subsequent lipid than carbohydrate tolerance.
