4.1.3. Remifentanil

The last PSA agen<sup>t</sup> we investigated was remifentanil. Remifentanil is an ultrashort-acting μ-receptor opioid agonist with a rapid time to peak activity after a bolus dose and a short context-sensitive half-time of less than 4 min without regard to infusion duration [30]. In rats, opioids have been reported to exert an inhibitory modulation of GABAergic and glutamatergic synaptic transmission in the STN via presynaptic μ- and δ-receptors [31,32]. Only a limited number of studies have addressed e ffects of opioids during STN-DBS surgery [9,22,23,33]. In these studies, neurophysiological data were obtained while patients received propofol in combination with opioids. Therefore, the opioid e ffects on neuronal activity could not be well characterized. To our knowledge only one other study has assessed the e ffect of REMI on STN neurons in PD patients [24]. In that study, single cell activity of 4 neurons were analyzed before and after a bolus of 0.5 μg kg−1. REMI did not significantly alter short interval discharge activity, a measure related to firing rate. Our results also showed no significant e ffect of REMI on firing rates or MUA. The use of REMI was only associated with a significant increase in CV compared to the control group.

### *4.2. E*ff*ect of Clinical and Demographic Variables on Microelectrode Recordings*

In addition to the e ffects of the PSA agents, clinical and demographic variables might affect neuronal firing properties of the STN. Therefore, we tested the e ffect of these variables on electrophysiological recordings.

One interesting finding was that the CV increased in patients with a longer disease duration. An increase in burst activity is generally observed in animal models of PD [34,35]. One potential explanation of this phenomenon is that neuronal firing becomes more strongly locked to a more powerful beta rhythm, seen with advanced disease, thereby reducing variability in the interspike-interval. The underlying mechanism of our finding therefore remains elusive. Finally, while CV was related to disease duration, no relation with the UPDRS III score, side of onset or age of the patients was observed.

The hemisphere of the recordings significantly influenced MUA with the right STN having higher power than the left. Currently no study to our knowledge has reported lateralization of MUA in the STN during rest. Interestingly, several studies in humans and animals have shown that the left and right cortex are related to di fferent functions. In line with functional lateralization in the cortex, emotions seems to be processed in the right STN [36–39]. We therefore speculate that the di fference we observed may be related to functional lateralization of the STN; a hypothesis that needs further validation.

Finally, the age of the patient significantly influenced MUA. Within the STN, older patients showed lower power compared to younger patients. An e ffect of aging on neuronal power spectrum in the cortex has been reported previously [40–42]. In contrast to our findings, these studies all reported an increase in high frequency power with age, which they discuss in terms of the neuronal noise hypothesis of aging [43]. A number of interesting hypothesis may account for the reversal of this pattern that we observed in STN. First, our population was generally older than the older adults in these papers, to the extent that our younger patients would be considered to belong to their 'older' groups. One interpretation of these data could be that the relationship between age and neural population power is non-linear such that adults in late middle age show the highest power. Second, there may be task e ffects, since our data were recorded with the patient not performing a cognitive task, while previous reports have all been in the context of active cognitive tasks. Third, our data may show a genuine di fference in the e ffect of aging in the STN compared to the cortex. Finally, the possibility exists that the change in power we observed reflects age-related changes in shape and location of the STN [44]. According to this hypothesis, recording sites at the same stereotaxic coordinates may represent di fferent functional domains, with di fferent neuronal properties, within the STN in di fferent age groups. Future work may elucidate these possibilities.
