Journal of Rehabilitation Research and Development
Vol. 38 No. 6, November/December 2001

Emerging clinical applications of electrical stimulation: opportunities for restoration of function

Warren M. Grill, PhD; Michael D. Craggs, PhD; Robert D. Foreman, PhD; Christy L. Ludlow, PhD; Jerome L. Buller, MD

Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH; Cleveland Louis Stokes VA Medical Center, Cleveland, OH; Cleveland FES Center, Cleveland, OH; Institute of Urology and Nephrology, University College London, London, UK; Spinal Injuries Unit, Royal National Orthopaedic Hospital, Middlesex, UK; Department of Physiology, University of Oklahoma Health Sciences Center College of Medicine, Oklahoma City, OK; National Institute of Neurological Disorders and Stroke, Bethesda, MD; Department of Obstetrics and Gynecology, Walter Reed Army Medical Center, Washington, DC

This material is based upon work supported by a grant from the Whitaker Foundation, with additional support from the Board of Clinical Studies, Royal National Orthopaedic Hospital, Stanmore, Middlesex, UK; Finetech Medical Limited, UK; NeuroControl Corporation, USA; the Intramural Program of NINDS Z01 NS02980-02; and Medtronic, Inc.
Address all correspondence and requests for reprints to: Warren M. Grill, PhD, Department of Biomedical Engineering, Case Western Reserve University, CB Bolton Building, Room 3550, Cleveland, OH 44106-4912; email: wmg@po.cwru.edu.

Abstract--Emerging clinical application of electrical stimulation in three systems is reviewed. In the bladder, stimulation of sacral posterior roots reduces reflex incontinence and significantly improves bladder capacity. By combining anterior and posterior root stimulation, bladder control can be achieved without the need for rhizotomy. Preliminary research demonstrates that bladder contractions may also be generated by stimulation of the urethral sensory branch of the pudendal nerve, even after acute spinal cord transection, while inhibition of the bladder and control of urge incontinence can be achieved by stimulation of the whole pudendal nerve. Spinal cord stimulation can modulate the activity of the intrinsic cardiac nervous system involved in the regulation of regional cardiac function and significantly reduce the pain associated with angina pectoris. Finally in the area of upper airway disorders, functional electrical stimulation has great potential for increasing life support as well as quality of life in chronic ailments, particularly obstructive sleep apnea and dysphagia.

Key words: angina, bladder, functional electrical stimulation, incontinence, neural prostheses, neural stimulation, neuromodulation, spinal cord injury, upper airway.


  The purpose of this paper is to review several emerging applications of electrical stimulation to restore function after neurological impairment (1). Control of three systems will be considered: the bladder, the heart, and the upper airway. The use of neural prostheses for the control of bladder function, and stimulation of the pudendal nerve for treatment of the overactive bladder will be discussed. Potential mechanisms of pain relief from angina pectoris with spinal cord stimulation will be explored. And finally, the role of electrical stimulation in treating four different upper airway disorders--obstructive sleep apnea, dysphagia, bilateral vocal fold paralysis, and laryngeal dystonia, will be considered.


  Physiologic voiding involves bladder contraction and urethral sphincter relaxation. Both voiding and the storage of urine by the bladder require properly coordinated spinal-brainstem-spinal reflex loops (2). In spinal cord injury (SCI) these loops are damaged, resulting in the loss of voluntary sphincter control, sensation of bladder fullness, and proper coordination of the bladder and urinary sphincters. As a consequence, bladder emptying is impaired and the result is reflex incontinence.

  Reflex incontinence is primarily caused by detrusor hyper-reflexia, an aberrant reflex that emerges after a period of spinal shock following SCI. It is often associated with dyssynergenic contractions of the sphincter muscles of the urethra preventing efficient emptying of the bladder. Persons with SCI invariably have large residual bladder volumes, have recurrent urinary tract infections, and, more seriously, can develop a susceptibility to renal impairment if the condition is not treated properly. Conservative treatment is usually a combination of drugs for suppressing detrusor hyper-reflexia and intermittent catheterization for emptying the bladder. However, the anti-muscarinic drugs used to treat incontinence often have debilitating side effects, such as loss of bowel control secondary to constipation, dry mouth, and visual disturbance. Voiding can also be very troublesome, especially for women for whom no simple collection vessels exist, and intermittent catheterization can sometimes introduce bladder infections. These problems, among others, demand that more satisfactory therapies for bladder management in SCI are developed.

  In the 1970s Brindley and his team developed an implantable stimulator to empty the bladder of paralyzed persons (3). The prosthesis uses sacral anterior root stimulation (Finetech-Brindley SARS, Finetech Medical Limited, Wellyn Garden City, UK) to activate bladder motor pathways and produce clinically effective voiding (Figure 1). To overcome reflex incontinence and sphincter dyssynergia in these subjects, the sacral sensory nerve roots from S2-4 have to be cut (sacral de-afferentation). The Finetech-Brindley device has been successfully used in many countries throughout the world (4), and is starting to be more widely used in the United States under the name of Vocare (NeuroControl Corporation, Valley View, Ohio). The implant has been shown to be very effective in increasing bladder volume, promoting complete emptying of the bladder, reducing bladder infections, and significantly improving the quality of life for many subjects. However, the need for sacral de-afferentation (posterior rhizotomy) with the consequent loss of reflex erections and reflex ejaculation can deter many potential young male subjects from accepting SARS. Furthermore, the hope by some subjects of a "cure" for SCI in the future using neural regeneration and repair techniques is a further obstacle to acceptance. However, there may be an alternative solution to sacral de-afferentation using stimulation of these same afferent pathways to suppress reflex incontinence.

A photograph of the Finetech-Brindley Extra-dural Sacral Anterior Root Stimulator Implanta. A series of graphs depicting results of the Finetech-Brindley Extra-dural Sacral Anterior Root Stimulator Implantb.
Figures 1a and b. The Finetech-Brindley Extra-dural Sacral Anterior Root Stimulator Implant. Figure 1a shows the implantable receiver (Internal Stimulator) connected by cables to electrodes placed extra-durally next to the sacral nerves carrying motor fibers to the bladder detrusor muscle. To prevent reflex incontinence, sacral afferent nerves from S2-4 are cut at the conus level through a small laminectomy marked X. To empty the bladder a transmitter coil is aligned with the implanted receiver on the outside of the skin and pulses are transmitted transcutaneously to activate the nerves with short bursts of electrical stimulation. The traces in 1b show rises in bladder and urethral pressures generated by each burst of supra-maximal stimulation at 30 pulses per second, each pulse being up to 500 ms in duration. When the bladder pressure is sufficient to overcome the residual urethral pressure, between the bursts of stimulation, then bladder emptying takes place.

  The development of a sacral posterior and anterior root stimulator implant (SPARSI) is being explored in which stimulation of the anterior roots to empty the bladder is combined with stimulation of the posterior (sensory) roots to prevent reflex incontinence using the same implant to activate the mixed roots (5).

  Preliminary studies using non-invasive multi-pulse magnetic stimulation over the sacrum have shown that suppression of reflex incontinence can be achieved with stimulation of the mixed sacral roots (S2-4) in subjects with detrusor hyper-reflexia (6). The mechanism for this "neuromodulatory" action has yet to be determined in man but one theory, based on experimental work in animals (7) suggests that neuromodulation involves inhibitory action by pudendal afferent (sensory) nerve stimulation on pelvic nerve motor pathways to the bladder through spinal cord circuits (Figure 2)(8). Evidence in support of this theory has been obtained by stimulating pudendal afferent pathways at the level of the dorsal penile (9) (or dorsal clitoral) nerves in subjects. Dorsal penile nerve (DPN) stimulation applied through surface electrodes produces a profound and repeatable suppression of the bladder both when applied conditionally (i.e., when bladder pressure increases during detrusor hyper-reflexia) or continuously (pre-emptively) in subjects with SCI (10) (Figure 3). Furthermore, in addition to decreasing reflex incontinence, DPN stimulation can produce significant increases in bladder volume (11). These effects depend essentially on stimulation, and diminish when stimulation is switched off.

A diagram showing a mechanism of neuromodulation for reflex incontinence in spinal cord injury Figure 2. A mechanism of neuromodulation for reflex incontinence in spinal cord injury. Electrical stimulation of the large pudendal afferents at the level of the dorsal penile nerves or the posterior (sensory) sacral roots suppresses aberrant pelvic reflexes so that activity in the parasympathetic motor pathways to the bladder is inhibited preventing contraction of the bladder detrusor muscle (adapted from (8)).

A diagram showing neuromodulation through pudendal afferentsa. A graph showing the effects of neuromodulation through pudendal afferentsb.
Figure 3. Neuromodulation through pudendal afferents. Electrical stimulation of the dorsal penile nerves with surface electrodes can be used to assess non-invasive neuromodulation of detrusor hyper-reflexia through exclusively afferent branches of the pudendal nerve. In Figure 3b, the upper trace shows suppression of hyper-reflexic contraction shortly after bladder pressure begins to rise (i.e., conditionally when the bladder pressure exceeds about 20cm H2O) with control contractions before and after stimulation. The lower trace shows how the whole contraction can be profoundly suppressed using stimulation pre-emptively (i.e., continuously). (Adapted from (10).)

  Recently, four subjects with a SCI have been implanted with a standard bilateral extra-dural Finetech-Brindley device, but without sacral de-afferentation, to test the concept of SPARSI. Each of these subjects was assessed pre-operatively using DPN stimulation as described above to demonstrate the efficacy of neuromodulation. Preliminary results indicated that subjects were able to achieve both suppression of reflex incontinence and clinically useful increases in bladder volume using the implanted stimulator (12)(Figure 4). Interestingly, intermittent stimulation also appeared to produce good results, although the ideal interval between bursts has yet to be determined. The bladder volumes achieved using neuromodulation were equivalent to those achieved with the anti-muscarinic drug oxybutinin in individual subjects. Hence, this device appears to be an effective therapy for reflex incontinence.

Graphs showing sacral posterior and anterior root stimulator implant a.  An image depicting the method of stimulation b.
Figures 4a and b. Sacral posterior and anterior root stimulator implant (SPARSI). The three urodynamic traces show laboratory results during bladder filling (cystometry) with and without stimulation through SPARSI in one subject with a complete spinal cord injury. The effect of stimulation given either continuously or conditionally (including intermittently) was to considerably increase end fill bladder volume (EFV). The graph below shows the effect in the same subject of using neuromodulation through the implant while at home. Interestingly, compared to either continuous or intermittent neuromodulation, treatment with oxybutinin (an anti-cholinergic drug to reduce reflex incontinence) did not achieve significantly better increases in bladder volume as determined by self-catheterization. During control periods without stimulation bladder volumes were significantly reduced and incontinence episodes rose dramatically. (Adapted from (11)).

  SPARSI does elicit good bladder contractions through anterior root stimulation, but at a much higher stimulation level than is needed for neuromodulation of reflex incontinence. However, simultaneous high-level stimulation of the mixed roots does cause persistent reflex sphincter contractions between the bursts for bladder contraction, and this elevates urethral pressures to a level where bladder emptying is often impaired. To overcome this problem, tests are now being carried out using a variety of techniques including selective nerve blocking ("anode block") of sphincter motor pathways with specially shaped electrical pulses, a method previously shown to be effective in experimental models (13).

  Ultimately it is hoped that by combining neuromodulation for reflex incontinence with neurostimulation for bladder emptying in SPARSI we can completely control the bladder without cutting any sacral sensory nerves, thereby preserving all pelvic floor reflexes in people with a SCI.


  Overactive bladder is a new definition adapted by the International Continent Society and recently by an NIH consensus panel on terminology in urogynecology. This is a symptom-based diagnosis including conditions such as detrusor instability, urge urinary incontinence, detrusor hyperreflexia, unstable bladder, and frequency/urgency syndrome. Traditional diagnoses, based on urodynamic findings, missed 40-60 percent (NOBLE Study Group, 2001, unpublished data) of affected subjects with lifestyle-altering lower urinary tract symptoms. Specifically, there are subjects with incapacitating urinary frequency and urgency, without incontinence, that are undiagnosed and untreated based on traditional diagnoses.

  Approximately 33 million Americans, or 16.6 percent of the adult population have overactive bladder (NOBLE Study Group 2001, unpublished data). Overactive bladder is a serious problem that significantly impacts a person's quality of life. Affected individuals tend to have significant sexual dysfunction, poor self-esteem, and limited physical activities and social interactions (14-15). Some subjects do respond to behavioral therapy, such as bladder and pelvic floor retraining, timed voiding, and biofeedback, but the standard therapy continues to be pharmacological treatment. Anti-muscarinic drugs such as tolteradine and oxybutinin are the mainstay of treatment. The problem with these drugs is compliance because of the undesirable side effects (16).

  For those who fail traditional therapy, there is now sacral neuromodulation. Sacral root stimulation has been demonstrated to be an effective treatment for urge urinary incontinence, frequency/urgency syndrome and non-obstructive urinary retention. However, the implant procedure usually requires general anesthesia and two separate incisions for the stimulating lead and pulse generator. Re-operation rates remain high and are usually for lead migration or pain (17).

  Electrical stimulation of the pudendal nerve has been demonstrated to inhibit detrusor activity and chronic electrical stimulation may provide effective treatment for overactive bladder disorders (18). The hurdle to date has been the technical challenge of placing and maintaining an electrode near the pudendal nerve in humans; however, recent development of the BION has made chronic implantation feasible.

  The BION is a small, self-contained microstimulator that can be injected directly adjacent to the pudendal nerve. The ischial spine is an excellent marker for the pudendal nerve as it re-enters the pelvis through Alcock's canal. This is a very consistent anatomical landmark in both men and women. Also, the implanted electrode is protected in this area by both the sacral tuberous and sacrospinous ligaments. Stimulation in this area activates afferent innervation over up to three sacral segments (Figure 5). Efferent stimulation also provides direct activation of the external urethral sphincter, the external anal sphincter, and the levator ani muscles, which may be of some benefit in bladder control. The external components of this neural prosthesis include a coil that is worn around the subject's hips and a controller that is worn around the shoulder or waist (Figure 6).

A drawing of the posterior view of the BION implant site adjacent to the pudendal nerve Figure 5. Posterior view of the BION implant site adjacent to the pudendal nerve. The BION can be injected directly adjacent to the pudendal nerve. Afferent pudendal pathways are shown with black dots and efferent pudendal pathways are shown with white dots.

A photograph of external components of the BION systema. Another photograph of the external components of the BION systemb.
Figure 6. External components of the BION system. System includes an RF transmitting coil to provide power and commands to the implanted BION and a controller and carrying case.

  The technique chosen to implant the device was that of the transperineal pudendal block. This approach is minimally invasive and is well established. A special implant tool was devised to facilitate placement. The BION implantation technique was developed in cadavers. The optimum insertion location is 1.5 cm medial to the ischeal tuberosity using a vaginal finger to guide the implant toward the ischial spine where electrical stimulation of the pudendal nerve may be confirmed.

  Subjects with overactive bladder symptoms refractory to other therapies were considered potential candidates for chronic pudendal nerve stimulation. The group evaluated included women 18-80 years old with no infection, malignancy, pelvic pain, or pregnancy. All subjects underwent a complete history and physical examination, urine analysis, urine culture, and a screening electrodiagnostic evaluation. The average age of the subjects enrolled in the study was 61 years, average weight was 159 lbs., and the average length of disease was 5.7 years. Each of them had tried at least two medications and had at least one previous urogynecological surgery.

  A percutaneous stimulation test (PST) was developed and proved to be a very effective way to assess acute changes in bladder volumes while stimulating the pudendal nerve. A baseline cystometrogram (CMG) was obtained followed by percutaneous pudendal nerve stimulation for 10 minutes with a repeat CMG. Two stimulation modes were tested, continuous and intermittent. With both types of stimulation, all parameters of bladder function were remarkably improved. The order in which the stimulation was provided was varied and did not appear to make a difference (Table 1).

Table 1.
Percutaneous Stimulation Test (PST) results. These data represent acute cystometrogram changes seen with percutaneous stimulation of the pudendal nerve.

Stimulation mode 1st sensation (cc) 1st urge to void (cc) 1st Max cystometric capacity (cc)

None 132.4 ± 56.1 196.4 ± 88.3 325.0 ± 150.5
Continuous 188.0 ± 88.9 330.4 ± 127.6 501.0 ± 192.3
Intermittent 298.6 ± 40.4 377.4 ± 61.2 609.8 ± 167.8
Significance (Kruskal-Wallis) p = .009 p = .03

  The first implant was done on August 29, 2000. The BION was implanted under local anesthesia with intravenous sedation. Proper placement was verified by palpation and EMG activity. An intermittent stimulation mode of 5 seconds on 5 seconds off was used. Subjects returned 5-7 days later for activation, to distinguish between postoperative pain and potential stimulation pain. Subjects were followed up at 15, 30, and 45 days after activation. At each follow-up visit they underwent another cystometrogram and brought in a 72-hour voiding diary. The results indicated a favorable response to maximum cystometric capacity throughout the study period. Diary entries verified improvement--incontinent episodes decreased by 65 percent, and both daytime and nighttime voids were decreased, as was pad use per day (Table 2).

Table 2.
Voiding diary summary. This table shows voiding diary data prior to implantation of the BION (baseline) and at 15, 30, and 45 days post implant.

Study period Incontinent episodes per day Daytime voids Nighttime voids Pads per day

Baseline 3.5 ± 2.6 7.8 ± 3.0 1.8 ± 1.7 4.3 ± 4.2
Day 15 1.7 ± 3.0 4.8 ± 2.4 0.4 ± 0.8 1.0 ± 1.0
Day 30 1.6 ± 2.7 6.0 ± 1.0 0.7 ± 0.5 1.0 ± 1.0
Day 45 1.5 ± 1.5 6.5 ± 1.2 1.0 ± 1.0 1.4 ± 1.0
"Percent change" 64.6 ± 44.4% 17.9 ± 30.9% 38.6 ± 43.3% 51.0 ± 34.8%

  In general, there were no peri-operative complications with either the percutaneous study or the implant. The mean duration of stimulation was 8.6 hours per day. With a minimal amount of stimulation, normal voiding was facilitated. The feasibility of direct pudendal nerve stimulation in a minimally invasive manner has been established. The percutaneous stimulation test may indeed be a useful test to predict candidates for future BION implantation. A favorable response of the lower urinary tract to chronic pudendal nerve stimulation was also observed. Most importantly, the subjects were satisfied with the device and indicated that it is having a profound effect on their quality of life.


  Another emerging application of pudendal nerve stimulation is electrical activation of urethral sensory nerve fibers to elicit bladder contraction and voiding (19). This approach is based on the augmenting reflex whereby fluid flow in the urethra initiates bladder contractions in the quiescent bladder and augments ongoing contractions in the active bladder (20,21). Pre-clinical studies in animals demonstrated that bladder contractions may be generated by electrical stimulation of the urethral sensory branch of the pudendal nerve (19,22), and that the urethro-bladder reflex was preserved following acute spinal transection (19-22). The excitatory urethro-bladder reflex was found to be strongly state-dependent (19,23). When the bladder is at low volumes, electrical stimulation of the urethral sensory branch does not cause excitation of the bladder but evokes an increase in urethral sphincter activity. When the bladder is at higher volumes, electrical stimulation of the urethral sensory branch of the pudendal nerve leads to excitation of the bladder (24) and micturition-like increases in bladder pressure (19,22), and prolonged stimulation leads to long-term augmentation of the urethro-bladder reflex (25). These results challenge the traditional view that coordinated voiding requires intact spinal-brainstem-spinal reflex loops (2) and suggest that the inter-neuronal circuitry required to elicit coordinated bladder evacuation exists within the spinal cord. This may provide a new approach to restoration of bladder evacuation following SCI.


  Angina pectoris associated with ischemic heart disease is referred pain that subjects generally feel as crushing, burning, and squeezing. The pain is usually sensed in the chest although it can sometimes radiate to the left arm and left jaw (26,27). The mechanism is believed to be dependent on convergence of cardiac afferent information on spinothalamic tract cells that also receive cutaneous and muscle afferent input (28,29). Thus the same pathway is transmitting different information but the areas of the brain involved with nociceptive processing have "learned" to interpret the cardiac information as coming from somatic structures. Angina pectoris can be a deceptive symptom. It may present itself in subjects who have normal coronary arteries; angina may be absent in subjects whose ECG indicates significant ischemia; and this pain can be unrelated to origin because the esophagus can readily mimic angina pectoris (30).

  Under normal conditions, exercise produces an increase in a value called the rate pressure product, an increase in heart rate and blood pressure producing a greater demand by the heart for oxygen. However, in ischemic heart disease, coronary blood flow is impaired. In these subjects, exercise increases oxygen demand that results in an imbalance between coronary blood flow reserve and myocardial oxygen consumption (31). The symptomatic result of ischemic heart disease under these conditions is angina pectoris. Typically, chronic stable angina can be treated with medications, such as ACE inhibitors, beta-blockers, and calcium channel inhibitors (32). Other common treatments include re-vascularization procedures such as percutaneous transluminal angioplasty or coronary artery bypass surgery.

  Unfortunately there are subjects who have chronic refractory angina pectoris but do not get sufficient pain relief or restoration of function even from optimal medical treatment (32). These are subjects with reversible myocardial ischemia resulting from significant coronary artery disease and they cannot be treated with additional re-vascularization. Between Europe and the United States, there are approximately 70,000-80,000 people with this problem who are not being treated.

  Adjuvant therapies for chronic refractory angina pectoris include restoration of function using rehabilitation and angiogenesis, trans-epidural anesthetics, external counter pulsation, and destructive therapies such as sympathectomy and laser ablation (33,34). Another therapy that has shown promise in relieving chronic refractory angina is neuromodulation via spinal cord stimulation (SCS). In Europe, this technique has been beneficial in approximately 80 percent of the cases (35). In addition, about 60 percent of the subjects experienced improved exercise capacity and quality of life for up to five years (36,37). It appears to be a safe, cost effective and reversible therapy, but its mechanism of action is still not understood (34). Subjects who receive SCS still experience the pain of myocardial infarction even when the stimulator is on, and thus the warning system is still intact (32,33,38,39).

  The procedure for spinal cord stimulation to alleviate angina involves the insertion, under local anesthesia, of an electrode between the T9 and T11 vertebrae (35). The electrode is then moved epidurally up to the T1-T2 level and placed just to the left of midline. The stimulation is turned on and tests are run to confirm that paresthesia evoked by activation of the stimulator is sensed in the somatic regions where the subject experienced angina pectoris. A typical subject regimen would be stimulation three times a day for one hour in addition to stimulation during an angina attack (35). The subject can activate or deactivate the stimulation by either using a magnet or subject programmer. If exercise is anticipated, the stimulus intensity can be increased prior to the onset of exercise.

  The challenge is to determine the underlying mechanisms that produce the anginal pain relief experienced with SCS. Spinal cord stimulation influences the processing of information within the central nervous system (40,41). Furthermore, SCS increases peripheral blood flow (42-47). Recent studies have shown that the intrinsic cardiac nervous system of the heart serves as a final common neuronal regulator of cardiac function (48). Because the intrinsic nervous system is interposed between the information processing of the central nervous system and cardiac function, it was proposed that SCS might modulate the processing of the final common neuronal regulator (49).

  Recent studies have shown that the intrinsic cardiac nervous system strongly influences heart function. The intrinsic cardiac nervous system is generally located in 7 to 8 areas of the heart that contain fat pads with aggregates of cardiac neurons and neural interconnections embedded within them that modulate cardiac function (48). These intrinsic cardiac neurons receive a stream of inputs from the neurons of the spinal cord to regulate regional cardiac function with each beat (49-51). Changes in the activity of the intrinsic nervous system can alter ventricular regional blood flow (52). Transient ischemic episodes of ventricular regions can markedly increase the activity of the intrinsic cardiac neurons (53). It is also important to note that chemical activation of intrinsic cardiac neurons can lead to restricted cardiac dysrhythmias (54). An important feature of the neural organization is that it is a hierarchy of control from higher centers down through the spinal cord and into the intrinsic nervous system to determine how the heart is modulated (Figure 7).

A diagram showing he hierarchical levels of integration important in neural control of the heart
Figure 7. Schematic diagram of the hierarchical levels of integration important in neural control of the heart. The hierarchy is comprised of descending projections from the spinal cord and medulla and, as a consequence, Higher Centers from the brainstem and cortex. These descending projections continuously influence the Intrinsic Cardiac Nervous System in ganglionated plexi via the input and modulation that occurs in the Intrathoracic Extracardiac Ganglia (IEG). The dashed lines and stippled boxes represent the afferent feedback originating from the heart. The black lines with arrows illustrate direct and successive synaptic relays from the medulla and spinal cord, within this autonomic neuronal hierarchy. Please note that the black arrows within the Intrinsic Cardiac Nervous System and IEG represent afferent and efferent connections, and Local Circuit Neurons (LCN). The hierarchy of the peripheral autonomic nested feedback loops functions in an interdependent manner to regulate regional cardiac function on a beat-to-beat basis. NODOSE, ganglion for vagal afferent fibers; DRG, Dorsal Root Ganglia for sympathetic afferent fibers; Symp, Sympathetic; Parasymp, Parasympathetic; b1, Beta-adrenergic receptors; M2, Muscarinic receptors.)

  Experiments were performed in animal models to determine if the intrinsic cardiac nervous system could be modulated with SCS (Figure 8) (49). Results indicated that both transient coronary occlusion and subsequent reperfusion increase intrinsic cardiac neuronal activity. Spinal cord stimulation is able to suppress this activity, even during transient myocardial ischemia (Figure 9). An interesting observation was that during the 15 minutes of stimulation, the intrinsic cardiac activity was suppressed, however, this suppression continued for another 15-30 minutes after the stimulation was turned off (55). In some cases these activities remained suppressed for 2-3 hours. Thus, there is a marked ability to change that nervous system such that it becomes less responsive once SCS has been applied.

A diagram showing the experimental setup for examining effects of spinal cord stimulation on the activity generated by the intrinsic cardiac neurons in the right atrial ganglionated plexus
Figure 8. Schematic diagram of the experimental setup for examining effects of spinal cord stimulation on the activity generated by the intrinsic cardiac neurons in the right atrial ganglionated plexus. (A.) Placement of a quadrapole electrode on the dorsal surface of the spinal cord (C, Cervical; T, Thoracic). The electrode was stimulated with a stimulator. (B.) Penetration of a microelectrode in the right atrial ganglionated plexus (gray ellipse, intrinsic cardiac nervous system) of the heart. (C.) Recordings of intrinsic nerve activity (ICN) on chart and tape recorders, an oscilloscope (CRO) and computer. Cardiovascular recordings were also stored on the recorders and the computer (ECG, electrocardiogram; BP, blood pressure; LV IMP, left ventricular intramyocardial pressure).

A graph showing responses of the averaged activity generated by the intrinsic cardiac nervous system during occlusion of the left coronary artery and spinal cord stimulation
Figure 9. Responses of the averaged activity (impulses per minute, imp/min) generated by the intrinsic cardiac nervous system (right atrial ganglionated plexus) during occlusion of the left coronary artery and spinal cord stimulation. The hatched horizontal bar represents the period of coronary artery occlusion (CAO) and the stippled bar is the period of spinal cord stimulation (SCS). The first bar is control intrinsic activity and the last bar is recovery of activity. These results show that SCS significantly reduces activity generated by intrinsic cardiac neurons even during CAO. * Represents data that were significantly different from control values (P<0.05).

  In summary, spinal cord stimulation of the upper thoracic segments suppresses the function of the intrinsic cardiac nervous system that serves as the final neural regulator of cardiac function, even in the face of a myocardial ischemic challenge. Thus, spinal cord stimulation may alter the function of the intrinsic cardiac nervous system to protect the heart from some of the detrimental consequences resulting from myocardial ischemia that lead to angina pectoris.


  Another emerging area of clinical application of electrical stimulation is the treatment of upper airway disorders, and four disorders are discussed. The first is obstructive sleep apnea, which is a prevalent disorder that can have severe negative consequences on an individual's quality of life and even be life threatening, for example when it causes people to fall asleep behind the wheel of a car. The second is dysphagia; swallowing disorders are also prevalent, life threatening, and induce significant health care costs. Third is bilateral vocal fold paralysis, a rare but life-threatening problem. Finally, laryngeal dystonia will be discussed, a condition that results in a disabling speech disorder.

  The current treatment options available for all four disorders are less than ideal. Sleep apnea is treated with tissue excision, which is painful and permanent and not always beneficial. Dysphagia subjects with chronic aspiration have to be put on nothing by mouth and enteric feeding with permanent gastrostomy PEG, which requires nursing care. Bilateral vocal fold subjects have tracheostomy, which does provide a permanent airway, but prohibits subjects from showering and swimming, and increases their risk of infection. In the case of laryngeal dystonia, botulinum toxin can provide temporary benefit in adductor spasmodic dystonia, but is not effective in abductor spasmodic dystonia.

Obstructive Sleep Apnea
  In obstructive sleep apnea, the tongue muscle relaxes during sleep and the tongue falls into the pharyngeal area, narrowing the hypopharynx. With inspiration, airflow through this area produces a negative pressure, sucking the pharynx closed to produce a further obstruction in the hypopharynx. There has already been some attention given to using hypoglossal nerve stimulation to dilate the hypopharynx during sleep (56). However, there have been problems with the nerve cuff associated with this technique, probably because there are a lot of strong movements of the tongue in the floor of the mouth during swallowing and chewing that could injure the nerve.

  Recently, genioglossus stimulation using percutaneous electrode insertion has been considered. In this approach, a 30Hz, 3mA pulse is delivered for 1 second to result in the epiglottis being pulled forward to prevent the collapse of the pharynx. In preliminary studies done with control subjects, this stimulation produced rather dramatic opening of the hypopharynx. Particularly impressive was the rapid rate of opening that occurs and that this opening remains sustained during stimulation. Therefore, it is believed that genioglossus stimulation does have potential particularly if a selected region of the genioglossus is targeted. This approach carries less risk of hypoglossal nerve injury that could cause swallowing problems. The next phase is subject evaluation.

  It has been shown in previous clinical studies that the usual cause of aspiration in dysphagia is a problem with laryngeal elevation. Reduced extent of elevation and delayed onset of elevation often can allow the bolus of food to enter the airway before the onset of swallowing. Lack of opening of the upper esophageal sphincter also results in the food remaining in the pharynx. Reduced speed and extent of laryngeal elevation or closure are the most prevalent in disorders such as post medullary CVA, where the brainstem central pattern generator is thought to be involved in swallowing, and also in neurodegenerative diseases (57).

  In pre-clinical studies that examined the potential use of neural stimulation to treat swallowing problems, control of laryngeal closure was investigated (58). A relatively non-invasive technique was used, percutaneous insertion through the thyroid cartilage, to implant an electrode directly in the thyroarytenoid muscle, which is the major muscle in the vocal fold. Results indicated that the anterior part of this muscle appeared to be most effective in closing.

  To examine the effects of using this instrumentation over time, 6 dogs were evaluated for 4-8 months with implants in the 2 superior and the 2 inferior thyroarytenoid muscles. Unilateral intermittent stimulation was used, 5 seconds on, 5 seconds off, for 8 hours per day. Intermittent stimulation was chosen because the laryngeal muscles are very fast muscles, 90 percent of the fibers are type II myosin and at risk for conversion to slow fibers under continuous stimulation. Monthly evaluations comparing stimulated with nonstimulated sides were performed. With stimulation the velocity of movement in the larynx was improved significantly. Intermittent stimulation did not alter muscle histochemistry; there was no conversion to slow fiber type in chronically stimulated dogs. Unilateral thyroarytenoid stimulation was found to produce bilateral closure, suggesting that this muscle can be thought of as a sphincter.

  Based on pre-clinical studies in which mylohyoid stimulation produced elevation of the larynx, a clinical study is ongoing to determine if stimulation of the mylohyoid and thyrohyoid muscles could assist in elevating the larynx in the human. Fifty percent of swallowing elevation was achieved with stimulation of just one of the muscles, and greater than 50 percent elevation could sometimes be achieved when both muscles were stimulated. It was also found that the two muscles elevate faster than either one muscle alone. This approach might be used in treating stroke subjects during early rehabilitation when they are having difficulty swallowing.

Bilateral Vocal Fold Paralysis
  In bilateral vocal fold paralysis, the reduced opening of the folds becomes closed when airflow during inspiration produces a negative air pressure, sucking the folds closed. In recent clinical studies of chronically paralyzed subjects, stimulation of the denervated posterior cricoarytenoid muscle was able to produce opening of the vocal folds (59).

  Many subjects who have vocal fold paralysis do eventually re-innervate, however, by the time they do, the muscle is often fibrotic and the joint is not flexible. FES was evaluated pre-clinically to determine if stimulation could be used to condition the muscle to remain healthy until re-innervation occurs. The results were very good; the movement threshold in the sectioned side of an animal that had received stimulation decreased over time and the velocity over time increased. However, in examining the animals in the final end of the year stimulation, the sectioned side still remained immobile during spontaneous breathing. Thus, this application would require chronic implant and would not be able to prevent the long-term effects of synkinesis interfering with voluntary movement.

Laryngeal Dystonia
  The most common form of laryngeal dystonia is a spasmodic dystonia, a voice disorder that occurs in middle life, which can cause disabilities prohibiting an individual from being able to work due to an ineffective voice for communication. The most frequent type of this disorder, adductor spasmodic dysphonia, is well managed with the use of botulinum toxin, however, the more rare abductor spasmodic dysphonia is not.

  Subjects with this latter condition experience a reflexive opening of the glottis, which is thought to be caused by over activity of the posterior cricoarytenoid muscle. Continued stimulation of the thyroarytenoid muscle during speech has been shown to reduce symptoms, particularly in more severe subjects (60), suggesting that there might be some role of neuromuscular stimulation for use in speech communication as well.

  Application of electrical stimulation in upper airway disorders is clearly an emerging area. Speech, breathing, and swallowing are important functions with few treatment options available for persons chronically impaired by neurological disorders affecting these functions. The challenge is that the muscles in this region are very small and the muscles are used differently for speech, chewing, swallowing and coughing. Thus ability to stimulate one specific muscle without dislodging the hardware is a technical concern that must be addressed. In addition, the head and neck area is very dense with vascularization and nerves, so surgical implant techniques are going to require skill and experience.


  There appear to be many potential applications of electrical stimulation--beyond the most familiar uses to restore upper- and lower-limb movement. Although there is a long history of electrical stimulation for restoration of bladder emptying, new applications in the area of bladder management are emerging. The primary focus of these new applications is on restoration of continence, and will result in the more widespread application of this technology. Similarly, spinal cord stimulation has been applied to neuropathic pain since the late 1960s. However, new applications for control of angina pain and other aspects of cardiac function are emerging. Success in new applications will require a greater understanding of mechanisms by which spinal cord stimulation exerts its effects. Clearly there are a number of applications of electrical stimulation for restoration of function in the upper airways. These are quite advanced and require multi-center clinical trials of safety and efficacy to move forward.


  This paper arose from a special session at the 6th Annual Meeting of the International Functional Electrical Stimulation Society, held June 16-19, 2001 in Cleveland, Ohio.

  1. Grill WM, Kirsch RF. Neuroprosthetic applications of electrical stimulation. Assistive Technology 2000;12(1):6-20.
  2. de Groat WC. Central nervous system control of micturition. In: O'Donnell P. Urinary incontinence. St Louis: Mosby; 1997. p. 33-47.
  3. Brindley GS, Polkey CE, Rushton DN. Sacral anterior root stimulators for bladder control in paraplegia. Paraplegia 1982;20:365-81.
  4. Brindley GS. The first 500 patients with sacral anterior root stimulator implants: general description. Paraplegia 1994;32:795-805.
  5. Craggs MD, Casey A, Shah PJR, Creasey G, Kirkham A, Khastgir J, Knight SL. SPARSI: an implant to empty the bladder and control incontinence without posterior rhizotomy in spinal cord injury. Brit J Urol Internat 2000;85(suppl 5):2.
  6. Sheriff MKM, Shah PJR, Fowler C, Mundy AR, Craggs MD. Neuromodulation of detrusor hyper-reflexia by functional magnetic stimulation of the sacral roots. Brit J Urol 1996;78:39-46.
  7. Lindstrom S, Fall M, Carlsson CA, Erlandson BE. The neurophysiological basis of bladder inhibition in response to intravaginal electrical stimulation. J Urol 1983;129:405-10.
  8. Craggs MD, McFarlane JP. Neuromodulation of the lower urinary tract. Exp Physiol 1999;84:149-60.
  9. Nakamura M, Sakurai T. Bladder inhibition by penile electrical stimulation. Br J Urol 1984;56:413-5.
  10. Shah N, Edhem I, Knight SL, Craggs MD. Acute suppression of provoked detrusor hyperreflexia with detrusor sphincter dyssynergia by electrical stimulation of the dorsal penile nerves in patients with a spinal injury. Eur Urol 1998;3(suppl):60.
  11. Kirkham APS, Shah NC, Knight SL, Shah PJR, Craggs MD. The acute effects of continuous and conditional neuromodulation on the bladder in spinal cord injury. Spinal Cord 2001;39:420-8.
  12. Kirkham APS, Knight SL, Casey ATM, Avenell S, Creasey G, Shah PJR, Craggs MD. Acute and chronic use of a sacral posterior and anterior nerve root stimulator to increase bladder capacity in spinal cord injury. Proc 6th Annual Conference of the International Functional Electrical Stimulation Society, 2001. pp. 172-4. Cleveland: IFESS.
  13. Brindley GS, Craggs, MD. A technique for anodally blocking large nerve fibres through chronically implanted electrodes. J Neurol Neurosurg Psychiatry 1980;43:1083-90.
  14. Herzog AR, Fultz NH, Brock BM, Brown MB, Diokno AC. Urinary incontinence and psychological distress among older adults. Psychol Aging 1988;3:115-21.
  15. Wyman JF, Harkins SW, Choi SC, Taylor JR, Fantl JA. Psychosocial impact of urinary incontinence in women. Obstet Gynecol 1987;70:378-81.
  16. Yarker YE, Goa KL, Fitton A. Oxybutinin: a review of its pharmacodynamic and pharmacokinetic properties, and its therapeutic use in detrusor instability. Drugs Aging 1995;6:234-62.
  17. Dijkema H, Weil E, Mijs P, Janknegt R. Neuromodulation of sacral nerves for incontinence and voiding dysfunctions: Clinical results and complications. Eur Urol 1993;24:72-6.
  18. Vodusek DB, Light JK, Libby JM. Detrusor inhibition induced by stimulation of pudendal nerve afferents. Neurourol Urodyn 1986;5:2381-9.
  19. Gustafson KJ, Grill WM. Bladder contractions evoked by electrical stimulation of pudendal afferents in the cat. Proceedings of the 6th Annual Meeting of the International FES Society, 2001. pp. 16-8. Cleveland: IFESS.
  20. Barrington FJF. The component reflexes of micturition in the cat. I and II. Brain 1931;54:177-88.
  21. Barrington, FJF. The component reflexes of micturition in the cat. III. Brain 1941;64:239-43.
  22. Shefchyk, SJ, Buss RR. Urethral pudendal afferent-evoked bladder and sphincter reflexes in decerebrate and acute spinal cats. Neurosci Lett 1998;244(3):137-40.
  23. Garry RC, Roberts TDM, Todd JK. Reflexes involving the external urethral sphincter in the cat. J Physiol 1959:149:653-65
  24. Mazieres L, Jiang, C, Lindstrom S. Bladder parasympathetic response to electrical stimulation of urethral afferents in the cat. Neurol Urodynam 1997;16(5):471-2.
  25. Jiang CH, Lindstrom S. Prolonged enhancement of the micturition reflex in the cat by repetitive stimulation of bladder afferents. J Physiol (Lond) 1999;517:599-605.
  26. Harrison TR, Reeves RJ. Principles and problems of ischemic heart disease. Chicago: Medical Publishers; 1968. pp. 197-204.
  27. Sampson JJ, Cheitlin MD. Pathophysiology and differential diagnosis of cardiac pain. Prog Cardiovasc Dis 1971;23:507-31.
  28. Hobbs SF, Chandler MJ, Bolser DC, Foreman RD. Segmental organization of visceral and somatic input onto C3-T6 spinothalamic tract cells of the monkey. J Neurophysiol 1992;68:1575-88.
  29. Foreman RD. Mechanisms of cardiac pain. Ann Rev Physiol 1999;61:143-67.
  30. Crea F. The management of refractory angina: a multidisciplinary international workshop. Bristol, UK; November 1998.
  31. Maseri A, ed. Ischemic heart disease. New York: Churchill Livingston; 1995.
  32. Jessurun GAJ, Ten Vaarwerk IAM, deJongste MJL, Tio RA, Staal MJ. Sequelae of spinal cord stimulation for refractory angina pectoris. Reliability and safety profile of long-term clinical application. Cor Artery Dis 1997;8:33-7.
  33. Ten Vaarwerk IAM, Jessurun GA, deJongste MJL, Andersen C, Mannheimer C, Elisson T, et al: for the working group on neurocardiology. Clinical outcome of patients treated with spinal cord stimulation for therapeutically refractory angina pectoris. Heart 1999;82:82-8.
  34. deJongste MJL. Spinal cord stimulation for ischemic heart disease. Neuro Res 2000 Apr;22(3):293-8.
  35. deJongste MJL, Hautvast RWM, Hillege H, Lie KI. Efficacy of spinal cord stimulation as an adjuvant therapy for intractable angina pectoris: A prospective randomized clinical study. J Am Coll Cardiol 1994;23:1592-7.
  36. Sanderson JE, Brooksby P, Waterhouse D, Palmer RBG, Neuhauser K. Epidural spinal electrical stimulation for severe angina. A study of effects on symptoms, exercise tolerance and degree of ischemia. Eur Heart J 1992;13:628-33.
  37. Bagger JP, Jensen BS, Johannsen G. Long-term outcome of spinal electrical stimulation in patients with refractory chest pain. Clin Cardiol 1998;21:286-8.
  38. Andersen C, Hole P, Oxhoj H. Does pain relief with spinal cord stimulation for angina conceal myocardial infarction? Br Heart J 1994;71:419-21.
  39. Sanderson JE, Ibrahim B, Waterhouse D, Palmer RB. Spinal cord stimulation for intractable angina--long-term clinical outcome and safety. Eur Heart J 1994;15:810-4.
  40. Chandler MJ, Brennan TJ, Garrison DW, Kim KS, Schwartz PJ, Foreman RD. A mechanism of cardiac pain suppression by spinal cord stimulation: implications for patients with angina pectoris. Eur Heart J 1993;14:96-105.
  41. Yakhnitsa V, Linderoth B, Meyerson BA. Spinal cord stimulation attenuates dorsal horn neuronal hyperexcitability in a rat model of mononeuropathy. Pain 1999;79:223-33.
  42. Linderoth B, Gunasekera L, Meyerson BA. Effects of sympathectomy on skin and muscle microcirculation during dorsal column stimulation: animal studies. Neurosurgery 1991;29:874-9.
  43. Linderoth B, Herregodts P, Meyerson BA. Sympathetic mediation of peripheral vasodilation induced by spinal cord stimulation: animal studies of the role of cholinergic and adrenergic receptor subtypes. Neurosurgery 1994;35:711-9.
  44. Augustinsson LE, Linderoth B, Mannheimer C, Elisasson T. Spinal cord stimulation in cardiovascular disease. In: Gildenberg P, editor. Neurosurgery Clinics of North America; 1995. p. 157-65.
  45. Augustinsson LE, Linderoth B, Eliasson T, Mannheimer C. Spinal cord stimulation in peripheral vascular disease and angina pectoris. In: Gildenberg P, Tasker R, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill. 1997. p. 1973-9.
  46. Croom JE, Foreman RD, Chandler MJ, Barron KW. Cutaneous vasodilation during dorsal column stimulation is mediated by dorsal roots and C.G.R.P. Am J Physiol 272 (Heart Circ Physiol) 1997;272:H950-7.
  47. Linderoth B, Foreman RD. Physiology of spinal cord stimulation: review and update. Neuromodulation 1999;2:150-64.
  48. Armour JA. Myocardial ischaemia and the cardiac nervous system. Cardiovasc Res 1999;41:41-54.
  49. Foreman RD, Linderoth B, Ardell JL, Barron K, Chandler MJ, Hull SS, et al. Modulation of intrinsic cardiac neurons by spinal cord stimulation: implications for its therapeutic use in angina pectoris. Cardiovasc Res 2000;47:367-75.
  50. Armour JA, Hopkins DA. Activity of canine in situ left atrial ganglion neurons. Am J Physiol 1990;259:H1207-15.
  51. Gagliardi M, Randall WC, Bieger D, Wurster RD, Hopkins DA, Armour JA. Activity of in vivo canine cardiac plexus neurons. Am J Physiol 1988;255:H879-900.
  52. Kingma JG, Armour JA, Rouleau JR. Chemical modulation of in situ intrinsic cardiac neurons influences myocardial blood flow in the anesthetized dog. Cardiovasc Res 1994;28:1403-6.
  53. Huang MH, Ardell JL, Hanna BD, Wolf SG, Armour JA. Effects of transient coronary artery occlusion on canine intrinsic cardiac neuronal activity. Integr Physiol Behav Sci 1993;28:5-21.
  54. Huang MH, Wolf SG, Armour JA. Ventricular arrhythmias induced by chemically modified intrinsic cardiac neurons. Cardiovasc Res 1994;28:363-642.
  55. Armour JA, Linderoth B, Arora RC, deJongste MJL, Ardell JL, Kingma JL, et al. Long-term modulation of the intrinsic nervous system by spinal cord neurons in normal and ischemic hearts. Autonomic Neurosci: Basic and Clinical (In press) 2001.
  56. Eisele DW, Smith PL, Alam DS, Schwartz AR. Direct hypoglossal nerve stimulation in obstructive sleep apnea. Arch Otolaryngol Head Neck Surg 1997;123(1):57-61.
  57. Lundy DS, Smith C, Colangelo L, Sullivan PA, Logemann JA, Lazarus CL, et al. (1999). Aspiration: etiology and implications. Otolaryngology Head Neck Surg 1999;120:474-8.
  58. Ludlow CL, Bielamowicz S, Rosenberg MD, Rossini K, Gillespie M, Hampshire V, et al. (2000). Chronic intermittent stimulation of the thyroarytenoid muscle maintains dynamic control of glottal closure. Muscle Nerve 2000;23:44-57.
  59. Zealear DL, Rainey CL, Herzon GD, Netterville JL, Ossoff RH. Electrical pacing of the paralyzed human larynx. Ann Otol Rhinol Laryngol 1996;105:689-93.
  60. Bidus KA, Thomas GR, Ludlow CL. Effects of adductor muscle stimulation on speech in abductor spasmodic dysphonia. Laryngoscope 2000;110(11):1943-9.

View Contents

  Last Updated Thursday, December 11, 2008 11:01 AM

Back to Top

Last revised Thu 10/25/2001; comments, problems, etc., to WM.