We thank Drs. Eikermann and Chamberlin for their comments about our article.1We agree entirely that the key issue is whether anesthetics themselves can directly influence the functioning of the brain circadian clock. They contend that it is unacceptable for us to conclude that propofol anesthesia acts directly on the circadian rhythm of circadian melatonin as well as the circadian rhythm of rest-activity and temperature in rodents.2
First, they make the point that the effects of intraperitoneal injection of propofol cannot be linked with propofol-induced anesthesia, arguing that the study design was not appropriate. We concur with them that we did not assess the depth of anesthesia; this was not the aim of our study. Because it is unclear in the first place from any clinical data available in the literature whether propofol injection could modify per se the plasma melatonin within the following 24 h, our study was designed to clarify this point. To the best of our knowledge, the loss of righting reflex in rats is an agreed upon method for assessing clinical anesthesia in rats in these circumstances.
Likewise, they use unusual logic to conclude that propofol has an opioid effect on melatonin secretion: (1) propofol has a pleasant effect that might be linked to an opioid effect; and (2) opiates indirectly affect melatonin secretion. As we know, the pleasant effect could be due to other factors, such as a dopaminergic effect.3Such tautology does not permit us to concur with them on this point.
Second, we understand that the suggestion in the single sentence in the “what this article tells us that is new” may appear provocative. It is always challenging to summarize the innovative aspects of data in one brief sentence. However, as an in-depth reading of the results and discussion sections clearly show, there is an evident visual phase advance of melatonin secretion with significant differences between propofol injection and control at early (decrease) and late (increase) periods of melatonin collection. Cosinor analysis of the raw data supports this observation with a statistical trend (P = 0.06). Moreover, we have very clearly pointed out the limitations of our study and have stated that “from our data obtained in rats, we cannot demonstrate that the fatigue, drowsiness, and sleep disorders observed in patients are related to a disturbed circadian pattern of human melatonin.” We also suggest that our data provide an opportunity to open new lines of research to better understand these symptoms. Indeed, there is no clear explanation yet for these undesirable symptoms that could occur even after a short duration of anesthesia for small medical procedures.
Third, using a similar approach, Drs. Eikermann and Chamberlin do not accept our statements of a previously described desynchronization of the rest-activity rhythm induced by propofol because, as they claim, the data were not obtained in constant darkness. To support their statements, they cite one of our previous articles where experiments were performed in dark/light conditions.2However, we are fully aware that it is necessary to have data in constant darkness to establish a direct linkage between anesthetic administration and circadian clock disruption. To that end, we published a study4in Neuropsychopharmacology in 2007 (cited in the article) in which the same experiments were performed in constant darkness. This study was unfortunately overlooked by Drs. Eikermann and Chamberlin. Indeed, we provided evidence that propofol anesthesia desynchronized the circadian rhythms of rest-activity and body temperature in rodents in the experimental condition of either alternation of light/dark2or constant darkness.4
By citing the study of Sessler et al. of 1991,5they create the impression that anesthetic exposure does not affect the circadian rhythms in humans. It must be pointed out that this first study was unable to demonstrate any effect in five human volunteers. Sessler et al. 5acknowledged that such data did not exclude an effect that could be missed. Indeed, the shifts in acrophase were +1.2, +2.1, −0.7, −1.6, and −0.7 h in the five subjects. Drs. Eikermann and Chamberlin have once again overlooked our previous study (cited in the article) that demonstrated a desynchronization of the circadian rest-activity rhythm after propofol anesthesia in patients.6When dealing with clinical studies, as we clearly state in the discussion of our article, one has to be cautious in drawing conclusion from merely one or two studies. Further studies are necessary to specify the magnitude of anesthetic effects on human circadian rhythms.
In contrast to the concerns of Drs. Eikermann and Chamberlin, we find that data in this field support the concept that either the “gears” and/or the “hands” of the clock might be influenced by propofol administration. It is premature to eliminate the importance of the concept of gears or hands at this early stage, as they suggest. However, we agree with them that a more profound understanding of the mechanism is an important question. Future studies should rigorously examine the effects of anesthesia on the complex pathways involved in the regulation of the clock. To this end, we are currently pursuing further experiments on the effect of anesthesia on some of these pathways (i.e. , the expression of clock genes within the suprachiasmatic nucleus, the melatonin release by the pineal gland).
*Fondation A. de Rothschild, Paris, France. firstname.lastname@example.org