Delta Homeostasis, Stress, and Sleep Deprivation in the Rat: A Comment on Rechtschaffen et al

Irwin Feinberg
1999 Sleep  
RAT. They concluded that their observations disprove the delta homeostatic model. They particularly criticize 1 my statement that: " . . .it is now generally accepted that it is primarily slow-wave or non-rapid eye movement sleep that is restorative for the brain and the process is most intense during high amplitude delta sleep." Rechtschaffen et al. note that "The major purpose of this [their] paper is to counter the reification of this kind of thinking." Although many investigators currently
more » ... ubscribe to this view of delta homeostasis, Rechtschaffen et al. appropriately selected me for this criticism since I proposed this model. 2 Here I argue that the conclusions of Rechtschaffen et al. cannot yet be accepted. Some are based on a faulty reading of the literature and others can be interpreted as direct and indirect effects of stress. I present evidence that the massive delta rebound and the negative delta rebound are direct and indirect pathological effects of stress that can be interpreted in the framework of the one-stimulus model of NREM-REM alternation. The hypotheses derived from this model are experimentally testable. Resolution of these issues is essential if sleep deprivation responses in the rat are to be used as a model for human sleep. THE RESPONSE OF THE RAT TO SLEEP DEPRIVATION There is good agreement among several laboratories on the rat response to TSD. With 12-24 hours of total sleep deprivation (TSD) the rat shows a small and transient delta rebound accompanied by an immediate (first 12 hours of recovery) and massive REM rebound. The REM rebound makes up for all of the REM sleep lost during deprivation, whereas the delta rebound makes up only a small fraction of the lost delta. The massive REM rebound is typically followed by a period (from about hour 12 to hour 36), of markedly subnormal delta production (the negative delta rebound) which exacerbates the existing delta deficit. In rats on a 12 hour light-dark cycle, total NREM sleep does not increase in either of the 12 hour light periods (LP) during the two days of recovery following deprivation. However, the overall delta deficit (small delta rebound followed by negative delta rebound) is somewhat lessened by an increase of about 50% in NREM sleep during the two dark periods. Rechtschaffen et al. point out that the small delta increase coupled with the negative delta rebound in the rat pose severe challenges to the homeostatic model based on human sleep data. Campbell and I fully agree. In discussing our TSD results in 1992, we noted that, if these were normal physiological responses, they would be "squarely inconsistent" with the homeostatic model of human NREM delta 3 . However, we proposed that these responses might not be normal because we detected abnormal EEG amplitude patterns following TSD. Normally, delta amplitude and incidence decline in parallel across NREM sleep in the LP. 4 Their trajectories then dissociate in dark period (DP) sleep; at DP onset, delta amplitude rises to its maximal level in the very first bout of NREM sleep and fluctuates around this level throughout the remaining episodes of DP sleep. In contrast, delta incidence is initially low and increases gradually across DP sleep. (This dissociation does not occur in the human; both delta amplitude and incidence increase in parallel across naps taken throughout the day). 5, 6 In the negative delta rebound the rat's normal amplitude response to darkness is altered. NREM delta amplitude does not increase at the start of the DP and remains suppressed through the subsequent LP. 3 In addition, NREM 10-30 Hz amplitude, which normally increases at DP onset, fails to increase during the negative delta rebound (Campbell and Feinberg, unpublished observations, cf). 7 This latter finding gains interest because 10-30 Hz NREM frequencies do not behave homeostatically; 8 the abnormal amplitude response in these frequencies, therefore, cannot be considered part of a compensatory response to sleep loss. Campbell and I suggested that these findings indicated that electrocortical function is impaired. Rechtschaffen et al. seem to misunderstand our interpretation of the negative delta rebound when they state that it depends on "par-
doi:10.1093/sleep/22.8.1021 fatcat:cv6kozquxzgyneobqd5ghbmnme