History of the Problem

Bladder control is profoundly disrupted by spinal cord injury. Bladder and kidney complications are responsible for much morbidity and frequent outpatient and inpatient care, with substantial economic and social costs.

Lesions of the cord above the sacral segments that innervate the lower urinary tract interrupt communication between those segments and the brain and pontine micturition center, while leaving segmental innervation intact. As a result, the bladder and sphincter muscles commonly become hyperreflexic, and their contractions are no longer coordinated by the pons to alternate between storage and emptying of urine. Hyperreflexia of the sphincter hinders emptying, and hyperreflexia of the bladder hinders storage. Concurrent hyperreflexia of these muscles can result in high bladder pressures, causing hypertrophy and diverticula of the bladder, and high pressure in the kidneys causing renal failure; reflux of urine up the ureters to the kidneys can result in transmission of infections from the bladder to the kidneys. Bladder and kidney stones often follow these infections and in turn perpetuate chronic and recurrent infection.

Lesions of the cord in the sacral segments or cauda equina typically damage the cell bodies and axons of somatic, parasympathetic, and sensory nerves innervating the lower urinary tract. As a result, the bladder and/or sphincters may suffer from flaccid paralysis. Weakness of the sphincter muscles impairs storage and can result in stress incontinence, and impaired contraction of the bladder can impair emptying. Damage to the sacral nerve cells or axons renders them unresponsive to electrical stimulation.

For people with suprasacral spinal cord injuries, however, significant control can be restored by electrical techniques. These have focused mainly on replacing the function of damaged neural tissue by neural prostheses, but attention is also being given to modifying the function of remaining neural tissue by neuromodulation.

Similar concepts apply to control of the lower bowel after spinal cord injury. The bowel proximal to the splenic flexure of the colon obtains its parasympathetic supply from the vagus nerve which is typically intact after spinal cord injury, but distal to this point the remainder of the bowel obtains its innervation from the sacral segments of the cord. Suprasacral cord lesions can therefore result in a reflexic or hyperreflexic lower bowel and sphincters, with similar problems of co-contraction impairing emptying and continence. Impaired emptying can also hinder the peristaltic transport of stool, resulting in constipation. Lesions of the sacral cord or cauda equina abolish spinal reflexes, resulting in a hyporeflexic bowel and severe constipation, although intramural reflexes maintain some peristalsis. Hyporeflexia of the anal sphincter can result in stress incontinence of feces, and fecal incontinence can be much more socially disruptive than urinary incontinence. These bowel problems greatly impair quality of life.

History of Current Solutions

The most effective method to date of restoring bladder and bowel function by electrical stimulation after spinal cord injury consists of stimulating the sacral parasympathetic preganglionic efferent axons to produce contraction of the bladder and bowel and improve emptying of these organs.

Brindley in England developed a system in nonhuman primates that applied stimulation via electrodes implanted intradurally on the sacral anterior roots containing these parasympathetic efferent axons. These roots also contain somatic efferent axons to the external urethral sphincter, which have a lower threshold for stimulation and are therefore usually activated when the nerves to the bladder are activated. Although this can produce co-contraction of the bladder and sphincter, Brindley showed that intermittent bursts of stimulation, which produced sustained contraction of the smooth muscle of the bladder while allowing intermittent relaxation of the sphincter, resulted in effective bladder emptying in a pattern known as poststimulus voiding. Defecation was produced in many of these patients by a similar mechanism, and continuous stimulation of the S2 roots produced penile erection in many males. This technique was first used successfully in humans in 1978 and the system was commercialized in Europe in 1982 by Finetech Medical Ltd. as the Finetech Brindley Bladder Controller. The history of the development of the device from 1969 to 1982 was summarized in 1993 ( ). Between 1998 and 2002 it was distributed in the USA by NeuroControl Corporation as the Vocare system, and it continues to be distributed in other countries by Finetech Medical Ltd. It first received US Food and Drug Administration (FDA) approval in 1998 for providing urination on demand and reducing postvoid residual volumes of urine, and in 1999 for bowel evacuation, in patients with clinically complete spinal cord injury.

Tanagho and Schmidt in the United States described poststimulus voiding in dogs and concluded that the most successful combination to achieve continence and promote bladder emptying was to implant electrodes on the ventral component of S3 or S4 with extensive dorsal rhizotomy and selective peripheral neurotomy of somatic efferent axons ( ). Between 1983 and 1989 they used these techniques in 22 patients with various serious neuropathic voiding disorders, and reported restoration of emptying with electrical stimulation and low-pressure storage in 8; restoration of storage was achieved in a further 10 of these patients ( ). Subsequently they focused on patients with a variety of nonneurogenic lower urinary tract symptoms, and on modifying these symptoms by stimulating S3 nerves via an electrode implanted in the sacral foramen. The mechanism for modifying these symptoms is probably stimulation of sacral afferents to alter sacral and pelvic reflexes, a form of electrical neuromodulation. This system for sacral neuromodulation was commercialized by Medtronic as the Interstim system. It received the CE Mark in Europe in 1994 for chronic intractable (functional) disorders of the pelvis and lower urinary or intestinal tract. It received FDA approval in 1997 for urinary urge incontinence, in 1999 for urinary urgency-frequency and urinary retention, and in 2011 for chronic fecal incontinence. It is generally used in patients without spinal cord injury.

For patients with spinal cord injury, the main benefit of sacral anterior root stimulation is restoration of bladder emptying on demand with low postvoid residual volumes of urine. This usually allows patients to avoid catheterization. The low residual volumes and the absence of a foreign body such as a catheter in the bladder contribute to greatly reduced urinary tract infection.

Stimulation of these roots produces defecation on demand in many users by a similar mechanism of poststimulation voiding. The muscle of the lower bowel contracts more slowly than that of the bladder, and the contents of the bowel are more viscous than urine, so the intermittent bursts of stimulation and the gaps between these bursts are generally 2–3 times longer than those used for poststimulus voiding of the bladder.

Regular stimulation of sacral anterior roots for bladder emptying also improves peristalsis and reduces constipation. The majority of male patients are also able to produce and maintain penile erection by sustained stimulation of the S2 anterior roots that supply the nervi erigentes.

Continence is not directly targeted by sacral anterior root stimulation, but it was found to be improved in most of the initial patients using the stimulator. This may have been partly due to the fact that voiding with a low residual volume of urine increases the volume of urine that can enter the bladder from the kidneys before the threshold for a reflex bladder contraction is reached; reduction of bladder infection can also reduce reflex bladder contractions. Starting in 1981, some patients had some of the S2 and S3 posterior roots cut at the time of implantation to reduce hyperreflexia of the bladder and improve continence. Sauerwein in Germany advocated a full posterior sacral rhizotomy of S2-S4, on the grounds that this not only guaranteed an areflexic bladder and sphincter and low-pressure storage of urine but also protected the kidneys from reflux and hydronephrosis. It also abolished autonomic dysreflexia associated with the bladder or lower bowel. This gradually became a common practice. Of the first 50 patients who received an implantable stimulator, 34% had posterior rhizotomy ( ). Of the first 500 patients to receive the implant, 74% had posterior rhizotomy ( ) and Sauerwein’s group in Germany reported in 2005 that in their single-center experience of 464 patients implanted between 1986 and 2002, all had posterior rhizotomy ( ).

Many patients valued the reduction of autonomic dysreflexia and women in particular valued freedom from reflex incontinence, catheters, and urine collection devices. Most patients were able to discontinue anticholinergic medication that had been needed to control bladder hyperreflexia but that often had undesirable side effects such as dry mouth and constipation. These advantages of posterior rhizotomy were seen by such patients as outweighing its disadvantages of abolishing reflex erection and reflex ejaculation, particularly since better erection could often be produced by electrical stimulation or intracavernosal self-injection of vasodilators, and seminal emission could be produced by electrical stimulation of the seminal vesicles using a rectal electrode.

The addition of posterior sacral rhizotomy meant that it was not necessary to limit stimulation to the sacral anterior roots, but possible to apply stimulation to the mixed sacral nerves, since their afferent axons were divided proximally at the sacral afferent roots. This allowed the electrodes to be applied extradurally in the spinal canal, reducing the complexity of implantation. It was shown by Sauerwein that the sacral rhizotomy was still best performed intradurally ( ).

This combination of extradural electrodes and intradural rhizotomy was adopted in a clinical trial in the United States beginning in 1996 that resulted in 1998 in approval by the FDA of the implant as a Humanitarian Use Device. It was distributed in North America by NeuroControl Corporation until 2002 when the company went out of business.

The implant and an improved external controller continue to be marketed by the original manufacturer in England and used in Europe and other countries. However, with an increasing proportion of patients whose spinal cord is incomplete, and with their hopes for spinal cord regeneration in the future, posterior rhizotomy is less acceptable because of the desire to preserve remaining sensory and reflex function now and for the future. While it is still possible to use the implant in its original form without rhizotomy, there is increasing interest in finding ways to achieve the benefits of rhizotomy without its disadvantages. These trends and pathways for the future are discussed later.

Clinical Effects

Reliability

In , Brindley reported on implant failures and their repair in the first 500 patients to receive the stimulator he developed. Over 1922.9 implant-years, there were 98 implant failures in 72 patients, giving a mean time between failures of 19.6 years.

Forty-two were breaks in cables and 18 were in the implanted plug and socket connectors, which can usually be repaired or replaced. Eighteen were failures of receivers made before March 1988; at that date the way the receivers were manufactured was changed and there had not been further receiver failures at the time of the report.

In 8 of the 500 patients, there was late deterioration of root function; in one it was sudden, due to a fracture-dislocation of the lumbar vertebra; in the other seven it was gradual. In one case, it was associated with implant infection, in two with a descending syrinx, and in one with multiple sclerosis and possible progression of disease. In the other three the reason was not clear, but the patients remained continent on intermittent self-catheterization.

In a subsequent single-center series of 440 patients with a mean follow up of 8.6 years (range 1.5–18 years), receiver or cable failures required surgical repair in 44 patients over the follow-up period of 3784 implant-years ( ).

Quality of Life, Cost-Effectiveness, and Economic Consequences

In 1997 the Dutch Study Group on Sacral Anterior Root Stimulation compared costs and general quality of life measures in a group of 52 patients before and for 2 years after the procedure ( ). Although general measures of quality of life did not show significant differences, overall well-being as assessed with the Affect Balance Scale and disease-specific measures of satisfaction with bladder emptying improved. The average costs of bladder care per patient per year were reduced to less than one-third of the costs of conventional care, and the procedure was projected to pay for itself in 8 years.

In , Creasey and Dahlberg reported on economic consequences of the use of the implanted neuroprosthesis in the US health system, and calculated that the procedure reduced median costs of bladder and bowel management in the 12 patients studied from $8152 to $948 per year, and was projected to pay for itself within 5 years.

In 2003 Vastenholt in the Netherlands compared the specific impact of urinary problems and the quality of life of 37 patients who had undergone the procedure with a reference group of 400 SCI patients with neurogenic bladder who had not undergone the procedure, using the Qualiveen bladder-specific quality of life instrument. They found that those who had undergone the procedure reported a reduced specific impact of urinary problems on quality of life and a significantly greater quality of life ( ).

In Martens et al. used the Qualiveen and SF-36 questionnaires to compare quality of life in 93 patients who had had the procedure with 70 matched patients who had not, and found similar results.