Intolerance to shift work may result from individual susceptibility to an internal desynchronization. Some shift workers (SW) who show desynchronization of their circadian rhythms (e.g., sleep-wake, body temperature, and grip strength of both hands) exhibit symptoms of SW intolerance, such as sleep alteration, persistent fatigue, sleep medication dependence, and mood disturbances, including depression. Existing time series data previously collected from 48 male Caucasian French SW were reanalyzed specifically to test the hypothesis that internal synchronization of circadian rhythms is associated with SW intolerance and symptoms. The entry of the subjects into the study was randomized. Three groups were formed thereafter: SW with good tolerance (n=14); SW with poor tolerance, as evident by medical complaints for at least one year (n=19); and former SW (n=15) with very poor tolerance and who had been discharged from night work for at 1.5 yr span but who were symptom-free at the time of the study. Individual and longitudinal time series of selected variables (self-recorded sleep-wake data using a sleep log, self-measured grip strength of both hands using a Colin Gentile dynamometer, and oral temperature using a clinical thermometer) were gathered for at least 15 days, including during one or two night shifts. Measurements were performed 4-5 times/24 h. Power spectra used to quantify the prominent period (tau) and t-test, chi square, and correlation coefficient were used as statistical tools. The mean (+/-SEM) age of SW with good tolerance was greater than that of SW with poor tolerance (44.9+/-2.1 yrs vs. 40.1+/-2.6 yrs, p<.001) and of former SW discharged from night work (very poor tolerance; 33.4+/-1.7, p<.001). The shift-work duration (yrs) was longer in SW with good than poor tolerance (19.9+/-2.2 yrs vs. 15.7+/-2.2; p<0.002) and former SW (10.7+/-1.2; p<.0001). The correlation between subject age and shift-work duration was stronger in tolerant SW (r=0.97, p<.0001) than in non-tolerant SW (r=0.80, p<0.001) and greater than that of former SW (r=0.72, p<.01). The mean sleep-wake rhythm tau was 24 h for all 48 subjects. The number of desynchronized circadian rhythms (tau differing from 24 h) was greater in non-tolerant than in tolerant SW (chi square=38.9, p<.0001). In Former SW (i.e., 15 individuals assessed in follow-up studies done 1.5 to 20 yrs after return to day work), both symptoms of intolerance and internal desynchronization were reduced or absent. The results suggest that non-tolerant SW are particularly sensitive to the internal desynchronization of their circadian time organization.
