Motor Cortex Stimulation

Motor cortex stimulation (MCS) was utilized by Tsubokawa and colleagues, who originally presented this option as a possible safer alternative to DBS surgery in Dejerine syndrome (central thalamic pain syndrome) (Tsubokawa et al., 1991). It has been hypothesized that MCS induces antidromic activation of the sensory cortex, and this results in inhibition of the abnormal burst activity in a damaged and possibly deafferented sensory thalamus. Imaging studies “on” and “off” MCS have shown interesting blood flow changes within the ventrolateral (VL) tier of the thalamus, the medial thalamic regions, and the anterior cingulated cortex, orbitofrontal cortex, anterior insular cortex, and upper brainstem. Perhaps the most interesting aspect of post-MCS imaging has been the finding that blood flow does not seem to increase in the somatosensory cortex. Interpretations of this data have been variable, but the latest suggestions are that descending projections from the cortex to the thalamus can be activated and potentially lead to changes in basal ganglia pathophysiology within a loop responsible for both pain and emotional aspects of pain (Garcia-Larrea et al., 1999). It is thought that MCS may primarily stimulate γ-aminobutyric acid (GABA)-ergic interneurons that are found to be in parallel orientation to other layers of the cerebral cortex (Manola et al., 2007). Researchers have noticed that pain relief (≥40%\xE2\x80’50% improvement on the visual analog scale) from MCS seems to be greatest in cases where the corticospinal tract is found to be relatively intact (Katayama et al., 1998). Interestingly, when MCS is applied, patients with paretic extremities and concomitant central pain problems may have noticeable benefits in motoric weakness, rigidity, and spasticity. These improvements can, in select cases, aid in improving rehabilitation potential (Katayama et al., 1998).

MCS has been applied in cases of pharmacoresistant trigeminal neuralgia, painful peripheral neuropathy, and certain spinal cord pathologies. A greater than 50% pain relief has been reported in various studies of patients with central and neuropathic pain (Katayama et al., 1998; Nguyen et al., 1999). Data from randomized controlled studies for any application of MCS are lacking, however. There appears to be a large placebo effect with the therapy, and it has been noted that the effects may wane over time (Fontaine et al., 2009; Lima et al., 2008). Motor cortex stimulation has been attempted for therapeutic treatment of Parkinson disease (PD) and multiple system atrophy (MSA), but to date, results have been disappointing (Gutierrez et al., 2009; Kleiner-Fisman et al., 2003).

Transcranial Magnetic Stimulation

Information on transcranial magnetic stimulation (TMS) can be found in Chapter 32C as well as in the online version of this chapter at www.ExpertConsult.com.

Transcranial magnetic stimulation (TMS) is a noninvasive device consisting of an electromagnetic coil which when placed on the scalp, converts brief, high-current electrical activity into a rapidly cycling magnetic field. The magnetic field courses unimpeded through the scalp to induce localized intracranial electric current and thereby depolarizes the neurons, resulting in excitation or inhibition of the underlying neuronal circuit (Beric, 1993). Specific stimulation parameters such as current amplitude, duration, and direction can be used to calculate delivery of energy to a particular area in the brain. Repetitive TMS (rTMS) up to 60 Hz can be used to produce incrementally greater clinical effects (Andrews, 2003). The effects in general seem to be excitatory at fast frequencies (>1 Hz) and inhibitory at slow frequencies (≤1 Hz), and this may be similar to the published effects of long-term potentiation and long-term depression (Chen et al., 1997; Pascual-Leone et al., 1994). The state of brain activity, the integrity of the neural networks, and the interactions of the TMS with the targeted brain tissue can all be important in producing positive effects. Specific TMS stimulation parameters may also have positive and negative effects. The biological mechanism of action of TMS is not well understood, although changes at genetic levels and changes in concentrations of various neurotransmitters including monoamines, GABA, and glutamate have been suggested and may lead to corticospinal excitability and/or intracortical inhibition (Albert et al., 2009).

TMS has found applications in mapping the motor cortex and in treating post-stroke patients and several neurological disorders (movement disorders, neuropsychiatric disorders, epilepsy, tinnitus, and migraine), with varying degrees of success and failure. Some TMS targets include various regions of motor cortex in Parkinson disease (PD), dorsolateral prefrontal cortex (DLPFC) in depression, and left temporoparietal cortex in schizophrenia (Rossini and Rossi, 2007).

Different stimulation strategies for treatment of drug-resistant major depression have been attempted by randomized trials, suggesting the possibility of short-lived, clinically significant antidepressant effects (stimulation of the left DLPFC). Favorable responses to TMS may be predicted by previous medication treatment failures, lack of comorbid anxiety, and shorter duration of the current depressive episode (Avery et al., 2007; George, Lisanby et al., 2010; Lisanby et al., 2009; O’Reardon et al., 2007), although more data are needed to better understand positive clinical responses to TMS. It is unclear, however, whether the benefits of such treatment strategies will be sustained or whether longer-duration treatments may result in higher response and remission rates (Lam et al., 2008). A recent meta-analysis of rTMS for various neuropsychiatric disorders has supported its role in treating depressive symptoms and auditory hallucinations and addressing the negative symptoms of schizophrenia (Slotema et al., 2010). High-frequency rTMS has been shown to improve short-term motor scores in PD, potentially by enhancing motor cortex excitability (Elahi et al., 2009; Hemond and Fregni, 2007). In post-stroke patients, stimulation of the lesioned hemisphere (or alternatively, inhibition of the contralateral hemisphere) may aid in rehabilitation efforts. Repetitive TMS is also thought to have a potential role in treating neuropathic or visceral pain (Hemond and Fregni, 2007).

Seizure is the major adverse effect raising treating clinicians’ concern when applying TMS. Careful selection of patients may help avoid this complication. In general, patients with increased intracranial pressure, a metallic object(s) embedded in their head, or heart disease should avoid TMS. Other notable and not uncommon side effects from the stimulation include headaches, tinnitus, and facial muscle twitches (Andrews, 2003).

Deep Brain Stimulation

Over the last 2 decades, chronic DBS has become routine for several diagnoses in neurological practice (e.g., PD, dystonia, essential tremor) and has been used experimentally for selected neuropsychiatric indications (e.g., obsessive compulsive disorder [OCD], depression, Tourette syndrome [TS]). Interestingly, as early as the 1950s, temporary DBS electrodes were implanted into the septal region for pain control, and they were reported to have beneficial effects (Hamani et al., 2006). Over the ensuing years, there were various attempts at DBS, with most documented experiences revealing its usefulness in test stimulation prior to ablating brain lesions (Blomstedt and Hariz, 2010). In 1987 when Professor Benabid was operating on a chronic pain patient, he noticed that the patient’s tremor improved during test stimulation, and he decided to chronically stimulate this patient. Over the ensuing 2 decades, multiple DBS placements into multiple brain regions for a variety of clinical indications have occurred (Awan et al., 2009).

High-frequency stimulation (HFS) has been thought to effect a basal ganglia network–wide change, acting as a sort of informational lesion (Birdno and Grill, 2008; McIntyre et al., 2004a; McIntyre et al., 2004b). Thus, PD computer model systems have been developed to attempt to elucidate the mechanism of action of DBS (Birdno and Grill, 2008; McIntyre et al., 2004a; McIntyre et al., 2004b). HFS has been hypothesized to result in a decoupling of the cellular and axonal output within a thalamocortical relay circuit. The firing rates and patterns of the cell body may be suppressed, while the fibers and fibers of passage may be excited. DBS may ultimately effect a corticostriatopallido-thalamocortical (CSPTC) network and result in upstream as well as downstream changes within this complex basal ganglia network (McIntyre and Hahn, 2010). The specific effects of an electrical field are thought to reflect changes relative to the position and orientation of the axon to the actual DBS lead. Also, the electrical field is thought to exert trans-synaptic influences (McIntyre et al., 2004a; McIntyre et al., 2004b). The clinical benefits of DBS have been hypothesized to be due to more than just local neurotransmitter release (Stefani et al., 2005); however, several authors have argued that there is a collective effect and that transmitter release may be very important to the mechanism of action (Dostrovsky and Lozano, 2002; Lee KH et al., 2004; Vitek, 2002). Animal models of DBS have revealed increased extracellular concentrations of glutamate and GABA (Windels et al., 2003) and, most recently, adenosine and dopamine (Chang et al., 2009; Shon et al., 2010). Depolarization blockade, synaptic inhibition, and synaptic depression (McIntyre et al., 2004a; McIntyre et al., 2004b) have also been proposed to play a role in the potential mechanisms of action of DBS.

Selective placement of the DBS leads within different anatomical regions and different somatotopies may affect the neuronal network and, in the best possible cases, lead to improvement in clinical symptoms. For example, the placement of DBS electrodes in the ventralis intermedius nucleus of the thalamus for essential tremor may have an effect on an abnormal thalamocortical oscillatory loop and ultimately suppress tremor (Birdno et al., 2007). Similarly in PD, placement of a DBS lead in the ventralis intermedius may improve tremor; however, to affect bradykinesia and rigidity, one must either implant the ventralis oralis thalamic nucleus or, alternatively, either the subthalamic nucleus or globus pallidus internus (Mann et al., 2009; Oh et al., 2002).

DBS is a relatively simple technology. In its current form, it involves placement of a quadripolar (four contacts) lead into a specific and predetermined brain target. The lead is usually connected to a neurostimulator placed subcutaneously under the clavicle, although the battery can be placed in a multitude of regions. The neurostimulator can then be programmed or adjusted to tailor a setting to an individual patient. There are thousands of different combinations that may be chosen, and the voltage, frequency, and pulse width may be liberally changed. The optimal settings are patient and symptom specific, and they usually require that patients be reprogrammed frequently for the first 4 to 6 months and medications as well as stimulation monitored (Ondo et al., 2005; Rodriguez et al., 2007).

Each disorder or symptom considered for treatment with DBS should be carefully evaluated. Only a small fraction of any neurological or neuropsychiatric disorder may be eligible for this type of therapy. Most patients receiving DBS should be medication resistant, and they should undergo a complete multi-/interdisciplinary screening with a neurologist, psychiatrist, neuropsychologist, and neurosurgeon. Following screening, there should be a detailed interdisciplinary discussion about the goals of therapy (symptoms targeted, symptoms that will likely respond, symptoms not likely to respond). In cases of PD, patients should undergo an “off/on” levodopa medication challenge to determine which symptoms respond best to medication—these usually are the ones that respond best to stimulation (with the exceptions of tremor and dyskinesia). Risks and benefits of a potential DBS surgery, as well as the potential brain target(s), and unilateral versus bilateral DBS should all be carefully addressed in preoperative conversations with patients and families (Alberts et al., 2008; Kluger et al., 2009; Okun et al., 2004; Okun et al., 2007; Okun et al., 2009; Okun and Foote, 2004; Okun and Foote, 2009; Rodriguez et al., 2007; Skidmore et al., 2006; Ward et al., 2010). There are many potential adverse events that may occur as a result of DBS, some of which may constitute emergencies (Morishita et al., 2010).

Depending on the region of the world and the preference of individual surgical teams, leads and batteries may be placed in a single setting or may be staged (separate operating room procedures). One lead, two leads, or in exceptional circumstances, more than two leads may be implanted in a single session. One recent review of DBS hardware-related complications cited lead migration, lead fracture, lead erosion/infection, and lead malfunction as common occurrences (Lyons et al., 2004; Oh et al., 2002). DBS surgically related and stimulation-related complications can occur and may include (but are not limited to) hemorrhage, infections, strokes, seizures, paresthesias, dysarthria, hypophonia, dystonia, mood worsening, suicide, and worsening of comorbidities. Difficulty with verbal fluency and anger seem to be common sequelae in PD patients (Blomstedt and Hariz, 2005; Blomstedt and Hariz, 2006; Hariz et al., 2008a; Okun et al., 2008a; Saint-Cyr and Albanese, 2006). DBS teams must differentiate between lesion effects, stimulation-induced effects, and transient versus permanent neurological dysfunction. More DBS and device-related studies are needed to prospectively document all adverse effects; this constitutes a critical weakness of many published series (Hariz et al., 2008b).

Parkinson Disease

Parkinson Disease is a complex disorder thought to be the result of extensive loss of neurons and their projections within motor and nonmotor basal ganglia circuitry (Alexander et al., 1986). A rationale for neuromodulatory therapy has been developed as a result of models of basal ganglia physiology. Perhaps the most famous model reveals loss of dopaminergic neurons in the substantia nigra pars compacta, with a resultant abnormal neuronal activity in both the direct and indirect basal ganglia circuitry. These changes are thought to result in the genesis of many of the motor symptoms of PD. Initial treatment of PD is usually with dopaminergic therapy, though over time, disease progression may lead to limitations in medical therapy including such symptoms as wearing off between doses, on/off fluctuations, and hyperkinetic dyskinesia. Subthalamic nucleus (STN) or globus pallidus interna (GPi) DBS have been used to neuromodulate basal ganglia pathways and restore important functions in selected patients (Pahwa et al., 2005; Weaver et al., 2009). Studies are underway to define the selection criteria and help tailor the procedure for individual patients. To date, STN and GPi DBS have shown similar motor outcomes, but STN DBS may allow for larger dopaminergic medication reductions, and GPi DBS may provide better dyskinesia suppression and a relatively safer risk/benefit profile (Anderson et al., 2005; Follett et al., 2010; Mikos et al., 2010; Okun and Foote, 2009; Zahodne et al., 2009). These conclusions, however, will need to be bolstered by randomized clinical studies.

Long-term efficacy of DBS in PD has been good overall, but in all cases, the disease progresses. An important guideline is that symptoms that respond to levodopa continue to respond to DBS, with the exceptions of tremor and dyskinesia, which may have persistent benefits (Krack et al., 2003; Schüpbach et al., 2005; Wider et al., 2008).

Dystonia

Dystonia results from co-contraction of agonist and antagonist muscles, and patients may experience involuntary repetitive movements that result in twisted and sometimes painful postures. Dystonia may be focal, segmental, or generalized based on the body region affected. Other classification systems use age of onset or etiology. The most studied dystonia syndrome has been DYT-1 genetic dystonia; when the DYT-1 gene is expressed prior to age 26, the disease manifests. In those who live asymptomatic past age 26, the mutation (which occurs in the torsin A gene and is due to deletion of the GAG code) remains clinically silent (a carrier state) (Bressman, 2003). This is a phenomenon referred to as incomplete penetrance.

Lesion surgery (pallidotomy and thalamotomy) have both been successfully employed for various primary and secondary dystonias (Lozano et al., 1995; Yoshor et al., 2001), though most centers prefer DBS because bilateral lesions may result in speech or cognitive issues (Hua et al., 2003; Ondo et al., 1998). DBS therapy is mainly performed in the pallidal (GPi) target, because stimulation in this region has provided a reasonable alternative to lesion therapy. Most DBS cases have responded best if the dystonia has been of primary origin, although selected secondary dystonias as well as tardive dystonia have had meaningful improvements in small series (Coubes et al., 1999; Kumar R et al., 1999; Kupsch et al., 2003; Tronnier and Fogel, 2000; Vercueil et al., 2001). There have been two large randomized trials to date that addressed primary generalized dystonia, and both short- and long-term outcomes have been promising (Coubes et al., 2004; Vidailhet et al., 2005; Vidailhet et al., 2007). Additionally, the number of indications has been expanding within dystonia (e.g., cerebral palsy), and the number of brain targets also continues to expand (e.g., STN).

One interesting and unique aspect of DBS for dystonia has been the phenomenon that in many cases, the effects seem to be delayed and appear gradually after stimulation initiation (weeks to months). It has been hypothesized that this phenomenon may be the result of neuroplasticity, but its true mechanism remains a mystery. The other evolving story in dystonia DBS has been the use of lower stimulation frequencies for selected cases (Alterman et al., 2007). Which cases may respond to lower frequencies remains an area of investigation.

Tremor

Tremor has been broadly defined as an involuntary and rhythmic oscillation of a body part and has been classified according to its etiology and/or characteristics (e.g., phenomenology, physiology, etc.). It has been hypothesized that physiological disturbances in the cerebellothalamic and pallidothalamic pathways may be the genesis of some but not all tremor subtypes. According to Deuschl et al. (2001), four pathophysiological mechanisms are proposed to result in tremor: mechanical oscillation of the extremity, reflex activation of the oscillation network, central oscillation, and a disturbance in the regulatory feed-forward or feedback loops of an oscillation network. The ventralis intermedius (VIM) nucleus of the thalamus, which takes its input from the cerebellum, forms a vital piece of this regulatory network and has been frequently targeted for high-frequency (≥100 Hz) DBS to address various medication refractory tremors, with the most common being essential tremor (Benabid et al., 1996). DBS therapy has been reported to have similar efficacy as thalamotomy (Schuurman et al., 2000), and it may have fewer short-term side effects but more long-term (device related) side effects when compared to lesion therapy. Although VIM DBS is preferred for pure essential tremor and certain cases of PD tremor, other cerebellothalamic and pallidothalamic pathways may be altered with DBS. For example, outflow (cerebellar/midbrain) tremor, posttraumatic tremor, and multiple sclerosis (MS) tremor have all been treated in small case series by either single or multiple leads in VIM, ventralis oralis posterior, or zona incerta (Foote and Okun, 2005; Foote et al., 2006; Papavassiliou et al., 2008). The exact target(s) for these more complex tremor disorders remain to be investigated.

Typically, unilateral VIM DBS has been employed to control medication-refractory tremor in a contralateral extremity. Unilateral DBS commonly results in side effects of ataxia and speech problems, and these issues may be more commonly encountered when bilateral DBS is used (Pahwa et al., 2006). Midline tremor, head tremor, and voice tremor seem to respond less consistently to DBS (Ondo et al., 2001). Longitudinal follow-up studies have revealed good long-term benefits, although there has been an emerging concern in the field about tolerance and disease progression (Blomstedt et al., 2007; Pahwa et al., 2006; Sydow et al., 2003; Zhang et al., 2010). Other indications such as cerebellar tremor, Holmes tremor, and MS tremor have had worse efficacy when compared to essential tremor, but these outcomes may change with the emergence of better targets, better technology, and multiple lead approaches.

Neuropsychiatric Disorders

Pain

Neuromodulatory approaches have been employed for persistent pain syndromes for more than half a century. Many initial studies focused on the hypothalamus as the treatment target, but in later studies, many regions emerged as potential targets for therapy, including various areas of the thalamus as well as periventricular (PV) and periaqueductal (PA) gray regions. PV/PA gray matter regions have been hypothesized to respond better to nociceptive pain, whereas targeting the sensory thalamus has been thought to be better for deafferentation pain (Levy et al., 1987). Neuromodulation has been specifically employed for pain due to phantom-limb, stroke, and anesthesia dolorosa (Bittar et al., 2005). Additionally, after it was appreciated by functional imaging that cluster headache and facial pain syndromes may be related to hypothalamic circuitry, neuromodulation of this region was attempted and has been reported successful in multiple cases (Leone et al., 2006). Follow-up performed for up to 2 years revealed improvement in 50% of patients with cluster headache (Bartsch et al., 2008; Broggi et al., 2007; Schoenen et al., 2005; Starr et al., 2007). A recent study, however, did not show as robust a benefit (Fontaine et al., 2010). Greater occipital nerve stimulation is a less invasive alternative being explored for refractory headaches (Burns et al., 2009).

Epilepsy

Epilepsy affects approximately 50 million people worldwide (Kale, 2002). Despite active antiepileptic drug (AED) development, up to 20% of patients suffer from poor seizure control even with optimal medical therapy (Devinsky, 1999). A subset of these patients may be candidates for anterior temporal lobectomy (ATL), which may result in 80% to 90% seizure freedom (Yoon et al., 2003). For the remaining patients, alternative therapies such as vagal nerve stimulation (VNS) have proven limited in efficacy, although some studies report remarkable improvements (see Table 32D.3). Given the tremendous success of DBS for the treatment of movement and neuropsychiatric disorders, clinicians have begun to explore the potential of electrical stimulation for the treatment of a select group of patients with medication-refractory epilepsy.

The process of empirical trials for epilepsy has to date resulted in the discovery of unlikely DBS targets, including the cerebellum and thalamic nuclei. Additionally, DBS for epilepsy has reinvigorated interest in the possibility of long-term, potentially positive neuronal changes that may occur secondary to chronic stimulation and may have benefits even when stimulation has ceased (e.g., batteries of devices burn out, closed-loop devices are employed). This idea of neuronal-level changes has been based in part on the phenomenon of secondary epileptogenesis in patients with long-term poorly controlled seizures. A challenge that DBS may be utilized to address is in repeated ictal insults that may be initiated from a single site and may eventually induce remote and independent ictal activity (Khalilov et al., 2003; Morrell, 1985; Wilder et al., 1969).

Despite recent FDA hearings for approval of DBS for epilepsy, the treatment remains investigational. The mechanisms by which DBS affects seizures or movement disorders are not completely understood. Some theorize that electrical stimulation results in the release of inhibitory neurotransmitters. Others posit that stimulation inactivates neurons via depolarization blockade, although most movement disorders experts agree this is likely not the mechanism of action. There are likely synaptic-level changes, neurochemical changes, and also neurophysiological changes underpinning the benefits of DBS (Dostrovsky and Lozano, 2002; McIntyre et al., 2004a; Vitek, 2002). Recently it has been discovered that DBS may actually inhibit cells close to an electrical field and excite those farther from it. The timing of delivery of electrical stimulation is also an area of active research. Chronic stimulation with an “open-loop” system that responds to a cue (i.e., seizure) has been recently developed and has been referred to as “closed-loop” treatment (Fisher et al., 2010; Skarpaas and Morrell, 2009). Also, scheduled rather than chronic stimulation has been proposed. Multiple targets have been evaluated for DBS in epilepsy, with variable results. One of the reasons studies have reported variable results with the same DBS target may be that certain seizure types may respond differently to stimulation of a particular target. Multiple targets have been proposed and tried for epilepsy DBS (hippocampus, anterior nucleus, thalamus, CM nucleus, cerebellum, STN, etc.). Despite all the uncertainty, several trials have empirically demonstrated the efficacy of DBS for seizures, even in patients who have failed other therapies. These exciting results have fueled a number of studies designed to firmly establish DBS as an effective treatment for intractable epilepsy. In the largest controlled study of DBS for epilepsy (anterior nucleus stimulation) (Fisher et al., 2010), even though patients with a temporal lobe focus or foci had a significant reduction in seizure frequency, those with diffuse frontal, occipital, or parietal seizure foci failed to benefit. In contrast, STN DBS seems to be more effective in patients with seizures having a frontal focus or spread (Shon et al., 2005).

Spinal Cord Stimulation

The rationale for spinal cord stimulation (SCS) for pain syndromes can be linked to the gate control theory of pain proposed by Melzack and Wall. In this theory, the wide dynamic range (WDR) neurons of the dorsal horn are thought to be abnormally excitable in neuropathic pain, and this may then be neuromodulated and inhibited by SCS. SCS may result in pre- and postsynaptic inhibition of the abnormal peripheral neurons responsible for maintaining excitability of the WDR cells. GABA-ergic function may be similarly augmented. SCS has been shown to result in the release of vasodilatory neurotransmitters and to have resultant reductions in sympathetic tone, especially when utilized for ischemic pain syndromes such as those resulting from angina and peripheral vascular disease (Kunnumpurath et al., 2009).

SCS has been the most commonly applied neuromodulatory approach in the treatment of spinal pain syndromes and especially in the failed back surgery syndrome (FBSS). FBSS can be diagnosed after recurrence of lower back/leg pain and follows a previously successful surgery. Although randomized control trials have revealed that SCS is more effective than conventional medical treatment for FBSS, with sustained pain relief over prolonged periods, the level of evidence remains moderate, and the experts remain somewhat skeptical, since the efficacy is in comparison to a failed treatment (Kumar K et al., 2007; Kumar K et al., 2008; Taylor, 2006). Treatment effects are shown to diminish in other patient populations (i.e., those with chronic complex regional pain syndrome) when followed over a 5-year period (Kemler et al., 2008). An interesting trial of thoracic SCS showed improvement in coughing and better breathing control in cervical spinal cord injury survivors, which would be helpful in improving their quality of life (DiMarco et al., 2009).

SCS can be a useful option when conventional medical therapies have failed. SCS requires a highly experienced team in both the operation and the follow-up programming and medical management. Patients and clinicians should carefully consider the risk/benefit ratio when considering SCS. There is a potential for adverse events including infection, bleeding, cerebrospinal fluid leak from dural puncture, and hardware-related failures (Kunnumpurath et al., 2009).

Transcutaneous Electrical Nerve Stimulation

Transcutaneous electrical nerve stimulation (TENS) has long been used as a means of analgesia in chronic pain, including neuropathic conditions. It has also been used in acute low back pain, labor pain, and postoperative pain. Electrodes can be applied to the skin to deliver modulated current output that can be varied according to different stimulus parameters (frequency, duration, amplitude) and by connecting to a battery-operated device. TENS inhibits the nociceptive C-fiber evoked responses within the dorsal horn by activating the large-diameter afferents (Fields and Levine, 1984). The effects of TENS are mediated by endorphins, and they also cause decreased levels of glutamate and increased release of GABA and serotonin (DeSantana et al., 2008).

While TENS has been popularly used in pain clinics, its application has not been substantiated by careful studies. A recent analysis by the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology did not recommend TENS for treatment of chronic low back pain. There is, however, level B evidence supporting TENS for painful diabetic neuropathy (Dubinsky and Miyasaki, 2010).

Sacral Neuromodulation and Incontinence of Bowel and Bladder

Urinary urge incontinence results from the loss of voluntary control of voiding that may occur with aging and in neurological or inflammatory diseases that result in formation of reflex circuits mediated by C-fiber afferents. These afferents may be sensitized by bladder distension and trigger micturition reflexes. Bladder retention, on the other hand, usually results from persistent guarding reflexes that may be due to absent suprasacral pathways in spinal cord injury. There may be urethral-sphincteric relaxation blocks and detrusor contraction.

Sacral neuromodulation targets the somatic afferent axons in the spinal cord that modulate the voiding and continence reflex pathways (Leng and Chancellor, 2005). Direct inhibition of bladder preganglionic neurons, as well as inhibition of interneuronal transmission in the bladder reflex pathway, may suppress detrusor overactivity while inhibiting the guarding reflex. This may help with urinary retention and dysfunctional voiding (Leng and Chancellor, 2005).

Selection of patients should involve a careful history, physical examination, and review of a voiding diary (with evidence of having failed traditional measures) (Koldewijn et al., 1994; Scheepens et al., 2002). The surgical procedure usually involves a test stimulation staged percutaneous nerve evaluation prior to placement of a quadripolar electrode in a sacral foramen. When a DBS lead is placed, it is most commonly attached to S3.

Sacral neuromodulation is an FDA-approved treatment to improve urinary control. Long-term (≥2 years) follow-up of patients has revealed a reasonable success rate, with more than 50% improvement reported in 40% to 80% of patients (Oerlemans and van Kerrebroeck, 2008). With improvements in technology and expertise and low rates of complications, this approach appears to be an attractive option for patients, even for those with complete spinal cord injury (Sievert et al., 2010).

Additionally, outcome studies in patients with another disorder (fecal incontinence) followed for 2 years or more have been encouraging, with better than 80% reduction in incontinence episodes in 80% of patients (Kenefick, 2006; Melenhorst et al., 2007). Recent results showing improved bowel frequency in constipated patients have also raised this approach as an option in the most severe cases (Holzer et al., 2008; Kamm et al., 2010).