Technical advances in deep brain stimulation: how far is enough?
The introduction of deep brain stimulation (DBS) represented a major advance in the management of Parkinson's disease (PD). In this procedure, stimulating electrodes are stereotactically placed in the desired target with the aid of navigation software and neuroimaging. Many teams worldwide also rely on the use of intraoperative microrecording and microstimulation to reduce the risk of suboptimal placement of the electrodes with reduced antiparkinsonian efficacy and/or side effects. However, microrecording requires extensive neurophysiological training and experience. Alternatively, local field potential activity can be used to aid in target localization and stimulation through the implanted lead being used to provide information about the location of neighboring structures. However, none of these approaches guarantees ideal electrode placement with 100% accuracy.
In this issue of the journal, Janssen et al1 describe a new approach to localizing the motor/sensory region of the subthalamic nucleus (STN), with the aim of providing more accurate placement of DBS electrodes. Their work is based on findings in the monkey by Nambu et al,2 who have shown that microstimulation of the motor cortex induces a triphasic response in the motor regions of the STN. This response comprises an early excitation phase, thought to represent transmission via the hyperdirect cortico-STN pathway, followed by late excitatory and inhibitory responses. A recent study described similar cortical induced responses in the human GPi.3 In the current report, Janssen et al propose that these responses could be used intraoperatively to more accurately and more simply locate the motor portion of the STN in patients with PD. Although this approach still requires the use of microelectrode recording, the expertise required to identify the motor region of the STN based on increases or decreases in firing rate should be less than that required to identify the different subregions of the nucleus based on identification of specific discharge patterns. However, the procedure requires a second burr hole and placement of subdural electrodes for stimulation of the motor cortex. In this study, 3 of 5 patients experienced seizures. In 1 case this occurred even with prophylactic phenytoin treatment. Thus, although this new procedure has theoretical advantages, it is clearly associated with increased risk, and it remains uncertain if there are any benefits to justify this risk.
The article raises another set of important issues—how far is far enough? How much risk are we prepared to accept for small incremental advances? At what point does scientific investigation constitute human experimentation? And, what are the responsibilities of the editors in accepting such articles for publication?
In fairness to the authors, they consider that the benefits of achieving correct placement of the DBS lead outweighs the inconvenience of making an additional burr hole and the risk of inducing seizures by cortical stimulation. The intensity used for cortical stimulation was within the usual safety parameters.3, 4 Their institutional review board approved the protocol. And, the patients were informed about the risks and provided informed consent. After the first 2 patients had seizures, the protocol was revised to include the use of prophylactic anticonvulsants, and the ethics committee was advised and approved continuation of the study. We believe, therefore, that the authors followed established rules in obtaining the appropriate approval of the ethics committee and obtaining proper informed consent from the patients. The risk of seizure with the stimulation parameters employed was considered low and the potential benefits important. In our view, the proper sequence was followed, and now the procedure and the side effects observed are openly described and discussed.1
The question is what to do now. We conclude that although STN responses to cortical stimulation can be obtained in human PD patients, the risk of inducing seizures is high and outweighs the potential benefits of possibly better locating the motor portion of the STN. In this regard it should be appreciated that with current approaches, it is already possible to do very well in locating the motor part of the STN without this increased risk, and the optimal site for stimulation has not been completely established.
Our interest in publishing this article is twofold. On the one hand, the description of the STN response to cortical stimulation in humans confirms previous works in primates2 and has scientific interest. On the other, the appearance of seizures related to cortical stimulation in such a large proportion of patients indicates that in its current form this is not a safe procedure, and publication of these results may help to avoid such potentially harmful procedures in the future.
Finally, it is reasonable to question when such research is warranted if it involves exposing patients to a significant risk. In the specific case of DBS-STN for the treatment of PD, we believe the procedure has reached a plateau, and better identification of STN motor region is not a high priority.5 Currently, the major indication for surgery in PD patients is for the treatment of motor complications, and with advances in medical therapy and increasing awareness of DBS side effects, there appears to be a decline in the need for this procedure. It remains to be seen if DBS will have benefits in other areas such as gait, depression, and cognition, but for the present these potential indications remain unproven.6 In our view, invasive procedures with accompanying risks aimed solely at refining STN placement for its current indications do not seem warranted at this time.
1 , , , et al. Subthalamic neuronal responses to cortical stimulation. Mov Disord. 2012; 27: 435–438.
2 , , , et al. Excitatory cortical inputs to pallidal neurons via the subthalamic nucleus in the monkey. J Neurophysiol. 2000; 84: 289–300.
3 , , , et al. Cortically evoked responses of human pallidal neurons recorded during stereotactic neurosurgery. Mov Disord. 2011; 26: 469–476.
4 , . Intraoperative neurophysiological mapping and monitoring for supratentorial procedures. In: Deletis V, Shils JL. Neurophysiology in Neurosurgery. A Modern Intraoperative Approach. San Diego, CA: Elsevier Science; 2002: 339–404.
5 , . Long-term disability and DBS in Parkinson's disease: thinking about the long-term in the short-term. Mov Disord. 2011 (in press).
6 , , , , . Deep brain stimulation: from neurology to psychiatry? Trends Neurosci. 2011; 33: 474–484.