At the height of the psychosurgery debate the editors of The Lancet referenced the popular hesitancy to intervene surgically with the workings of the brain. They wrote that to do so `carries a peculiar penumbra of sacrilege' [1, 2]. Penumbra of sacrilege is a memorable, even strange phrase. It is one worth unpacking as we embark on an era which will see ever more border crossings at the blood-brain barrier. By invoking popular beliefs about a penumbra of sacrilege, the editors were suggesting that psychosurgery represented a gray zone in the shadows, a desecration or violation of a sacred space, the seat of the soul, the self. And as such, they were reflecting a cultural hesitancy, a lay reluctance to pursue this work. Fortunately for those with intractable neuropsychiatric disorders, neuromodulation has evolved beyond the primitive---and barbaric---sweep of the lobotomy, and with this advance, categorical resistance to this work has dissipated. And that is all to the good. But as the field progresses, and we implant more and more devices for therapeutic and investigational purposes, we must not let the placement of electrodes become too easy. It is a decision that should be made with full awareness of its implications for patients and families. Recently, I spoke to a friend who appeared to have early essential tremor. (Some non-essential aspects of this story have been changed to protect confidentiality.) He asked me, `Do you know anything about deep brain stimulation?' I indicated that I did. `Really?', he asked, seemingly unaware of my scholarly interest in the topic [3, 4, 5, 6] and involvement as a co-investigator in the use of deep brain stimulation in the minimally conscious state [7, 8, 9]. Satisfied that I might be a credible source for some free advice, he told me that his neurologist had sent him to a neurosurgeon to see about a stimulator. I asked him how disabling his condition was. As he competently drank a cup of soda, he told me it was a bit of a nuisance but nothing worse. As I recall the conversation, he was not receiving much more than a low-dose beta-blocker by way of medical management. Although I am an internist, and neither a neurologist nor neurosurgeon, it struck me as a bit premature to shuttle my friend off for an implant. It just should not be so easy. I asked myself: where had that penumbra of sacrilege gone? At some level, has this gotten too easy, too routine? Although this is but an anecdote, it is a disturbing one. My friend's referral was outside coverage norms established by the Centers for Medicare & Medicaid Services (CMS). Their 2003 national coverage determination for thalamic ventralis intemedius nucleus (VIM) deep brain stimulation (DBS) in essential tremor requires `marked disabling tremor of at least level 3 or 4 on the Fahn-Tolosa-Marin tremor rating scale (or equivalent scale) in the extremity intended for treatment, causing significant limitation in daily activities despite optimal medical therapy' [10, 11]. As best as I could tell, my friend met neither criteria for symptom severity nor adequate medical treatment. Even more striking was the casualness with which he told me about his neurologist's referral. One would think that he was being sent for the simplest of procedures, without any risks or long-term sequelae, notwithstanding specific complications associated with thalamic DBS for essential tremor [12]. It is a tribute to the nascent field of neuromodulation that, in the twenty years since Professor Alim Benabid's pioneering work heralded these new treatment modalities [13], stimulator placement has been analogized to the insertion of a heart pacemaker. But is the insertion of a cerebral pacemaker as routine as its cardiac counterpart? At this juncture I would venture to say it is not. While the acute surgical risks are slim, the longitudinal challenge for competent on-going care is high. Simply put, the community-based infrastructure to follow and support the growing number of patients with deep brain stimulators does not exist. Most patients go to highly specialized centers that have interdisciplinary teams able to assess, implant and support patients. But after surgery they return to their communities only to find a paucity of qualified neurologists and neurosurgeons able to provide on-going care. Even for rather routine matters like battery replacement or the adjustment of stimulation parameters, they need to return to the centers that performed the surgery. Follow-up there is all the more necessary for hardware failures, which still occur at non-trivial rates [14, 15]. This dependence on the mother ship is not the same for cardiac pacemakers. Any community hospital with a cardiology service can handle most complications and provide routine maintenance. Until a comparable neuromodulation infrastructure is in place, we need to be more prudent in determining who gets a stimulator. The prospects are even worse for those who are enrolled in clinical trials for new indications or have an innovative investigational device. What is their fate? What happens to these patients when the trial ends? Who provides on-going care? Who pays for battery replacement? Who removes a broken device? Who adjusts stimulation parameters ... in perpetuity? Because there is still virtually no group to take on these tasks, it is critically important that the neuromodulation community collectively affirm our on-going ethical obligation to these subjects once they leave trials and become patients. Our professional norms should reflect adherence to the ethical principle of non-abandonment. This duty is grounded in a deontological respect for persons. In the context of a neuromodulation trial, this means that once a subject is enrolled in a trial or under our care, we have a longitudinal fiduciary obligation to provide them with support. After a subject is implanted, the investigative team---and its sponsors---incur a clinical responsibility to provide on-going care and a fiscal responsibility for any associated costs. It is a breach of professional ethics to do otherwise. Such enduring covenants must be articulated in clinical protocols and be determinative in regulatory decisions by local Institutional Review Boards (IRBs) to reject or approve investigative protocols. The articulation of such `after-care' provisions should also be expected in any new IDE application to the Food and Drug Administration. To sustain clinical progress and investigative momentum, the neuromodulation community must embrace its ethical responsibility for comprehensive and on-going follow-up care. We need to populate a clinical infrastructure that can support patients and research subjects in their communities, especially as their conditions deteriorate and travel becomes more difficult. We need to disseminate our sequestered expertise more widely to primary care specialties. This ethical mandate transcends the clinical assessors, operators, and sponsors of clinical trials. It also applies to engineers who are well-positioned to help lessen the burden for patients and subjects. On the engineering side of this equation, innovation is the key. The development of better batteries with longer shelf lives or power management systems that optimize utilization of available capacitance will be a tremendous boon in streamlining follow-up care. So too will be the development of simplified device control systems to manipulate basic functions that would be operable by generalist physicians. Imagine a universal wand that could deactivate a device, and perform some rudimentary functions, that every Emergency Department could stock. Making such a basic parallel low-tech system universal amongst many device manufacturers would provide additional access to care in the community and a degree of safety in an emergency. Engineers might also work towards the development of mechanisms to give patients and subjects greater control over their devices, which are after all extrinsic intrusions on their bodies and their selves. Previously, I suggested that stimulation parameters for the treatment of neuropsychiatric disorders might be manipulated by patients one day. I envisioned a degree of patient discretion, within a pre-set safe range determined by physicians, much like patient-controlled analgesia (PCA) pumps give patients control over the dosing of opioid analgesia [3]. I am glad that such an advance is evolving as a means to preserve batteries in the treatment of motor disorders [16]. I would encourage the neural engineers to embrace the ethical mandate to develop additional platforms that might enhance patient self-determination and foster a greater degree of functional independence. While the neuromodulation community has every reason to celebrate its accomplishments, it would be better served by appreciating that the insertion of a device into the human brain comes with, if not the penumbra of sacrilege, a moral obligation to step out of the shadows and remain clearly available to patients and families over the long haul. Although neuromodulation has liberated many patients from the shackles of disease, we need to appreciate that the hardware that has made this possible can remain tethering. The challenge for the next generation of innovators is to minimize these burdens at this neural interface. By reducing barriers to care that exist in an unprepared health care system and developing more user-friendly technology, the neuromodulation community can expand its reach and broaden the relief provided by these neuro-palliative interventions [17]. Acknowledgements and Disclosures Dr Fins is the recipient of an Investigator Award in Health Policy Research (Minds Apart: Severe Brain Injury and Health Policy) from The Robert Wood Johnson Foundation. He also gratefully acknowledges grant support from the Buster Foundation (Neuroethics and Disorders of Consciousness). He is an unfunded co-investigator of a study of deep brain stimulation in the minimally conscious state, funded by Intelect Medical Inc. References [1] 1972 Editorial: Psychosurgery Lancet 7767 69-70 [2] Fins J J 2002 The ethical limits of neuroscience The Lancet Neurology 1 213 [3] Fins J J 2003 From psychosurgery to neuromodulation and palliation: history's lessons for the ethical conduct and regulation of neuropsychiatric research Neurosurgery Clinics of North America 14 303-19 [4] Fins J J 2004 Deep brain stimulation Encyclopedia of Bioethics, Vol 2 3rd edn, ed S G Post (New York: MacMillan Reference) pp 629-34 [5] Fins J J 2004 Neuromodulation, free will and determinism: lessons from the psychosurgery debate Clinical Neuroscience Research 4 113-18 [6] Fins J J 2009 Deep brain stimulation: ethical issues in clinical practice and neurosurgical research Neuromodulation eds E Krames, P H Peckham and A Rezai (London: Elsevier) pp 81-91 [7] Schiff N D, Giacino J T, Kalmar K, Victor J D, Baker K, Gerber M, Fritz B, Eisenberg B, O'Connor J, Kobylarz E J, Farris S, Machado A, McCagg C, Plum F, Fins J J, Rezai A R 2007 Behavioral improvements with thalamic stimulation after severe traumatic brain injury Nature 448 600-3 [8] Schiff N D and Fins J J 2007 Deep brain stimulation and cognition: moving from animal to patient Current Opinion in Neurology 20 638-42 [9] Schiff N D, Giacino J T and Fins J J 2009 Deep brain stimulation, neuroethics and the minimally conscious state: moving beyond proof of principle Arch. Neurology 66 697-702 [10] CMS 160.24 NCD for deep brain stimulation for essential tremor and Parkinson's disease, 1~April~2003 [11] CMS Manual System 100-04 Medicare claims processing, transmittal 128, 26 March 2004 http://www.cms.hhs.gov/Transmittals/Downloads/R128CP.pdf [12] Schwalb J M, Riina H A, Skolnick B, Jaggi J L, Simuni T and Baltuch G H 2001 Revision of deep brain stimulator for tremor: technical note J. Neurosurg. 94 1010-12 [13] Speelman J D and Bosch D A 1998 Resurgence of functional neurosurgery for Parkinson's disease: a historical perspective Movement Disorders 13 582-8 [14] Kondziolka D, Whiting D, Germanwala A and Oh M 2002 Hardware-related complications after placemen of thalamic deep brain stimulator systems Stereotact. Funct. Neurosurg. 79 228-33 [15] Okun M S, Tagliati M, Pourfar M, Fernandez H H, Rodriguez R L, Alterman R L and Foote K~D 2005 Management of referred deep brain stimulation failures: a retrospective analysis from movement disorders centers Arch. Neurology 62 1250-5 [16] Kronenbuerger M, Fromm C, Block F, Coenen V A, Rohde I, Rohde V and Noth J 2006 On-demand deep brain stimulation for essential tremor: a report on four cases Movement Disorders 21 401-5 [17] Fins J J 2008 Neuroethics and disorders of consciousness: a pragmatic approach to neuro-palliative care The Neurology of Consciousness, Cognitive Neuroscience and Neuropathology eds S Laureys and G Tononi (New York: Academic-Elsevier) pp 234-44