LONG-TERM RESULTS OF SACRAL NERVE STIMULATION (S3) FOR THE TREATMENT OF NEUROGENIC REFRACTORY URGE INCONTINENCE RELATED TO DETRUSOR HYPERREFLEXIA

0022-5347/00/1645-1476/0 THE JOURNAL OF UROLOGY® Copyright © 2000 by AMERICAN UROLOGICAL ASSOCIATION, INC.®

Vol. 164, 1476 –1480, November 2000 Printed in U.S.A.

Original Articles LONG-TERM RESULTS OF SACRAL NERVE STIMULATION (S3) FOR THE TREATMENT OF NEUROGENIC REFRACTORY URGE INCONTINENCE RELATED TO DETRUSOR HYPERREFLEXIA EMMANUEL J. CHARTIER-KASTLER, J. L. H. RUUD BOSCH, MICHEL PERRIGOT, MICHAEL B. CHANCELLOR,* FRANC ¸ OIS RICHARD† AND PIERRE DENYS From the Departments of Urology and Neurologic Rehabilitation, Pitie´-Salpe´trie`re Hospital, University Pierre et Marie Curie (Paris VI), Paris, Department of Neurological Rehabilitation, Raymond Poincare´ Hospital, Universite´ Paris-ouest, Garches, France, Department of Urology, Erasmus Medical Center Rotterdam, Rotterdam, The Netherlands, and Department of Urologic Surgery, Neuro-Urology and Urinary Incontinence Programs, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania

ABSTRACT

Purpose: We assess clinical and urodynamic results of sacral nerve stimulation for patients with neurogenic (spinal cord diseases) urge incontinence and detrusor hyperreflexia resistant to parasympatholytic drugs. Materials and Methods: Since 1992, 9 women with a mean age of 42.6 years (range 26 to 53) were treated for refractory neurogenic urge incontinence with sacral nerve stimulation. Neurological spinal diseases included viral and vascular myelitis in 1 patient each, multiple sclerosis in 5 and traumatic spinal cord injury in 2. Mean time since neurological diagnosis was 12 years. All patients had incontinence with chronic pad use related to detrusor hyperreflexia. Intermittent self-catheterization for external detrusor-sphincter dyssynergia was used by 5 patients. Social life was impaired and these patients were candidates for bladder augmentation. A sacral (S3) lead was surgically implanted and connected to a subcutaneous neurostimulator after a positive test stimulation trial. Results: Mean followup was 43.6 months (range 7 to 72). All patients had clinically significant improvement of incontinence, and 5 were completely dry. Average number of voids per day decreased from 16.1 to 8.2. Urodynamic parameters at 6 months after implant improved significantly from baseline, including maximum bladder capacity from 244 to 377 ml. and volume at first uninhibited contraction from 214 to 340 ml. Maximum detrusor pressure at first uninhibited contraction increased in 3, stabilized in 2 and decreased in 4 patients. Urodynamic results returned to baseline when stimulation was inactivated. All patients subjectively reported improved visual analog scale results by at least 75% at last followup. Conclusions: Sacral nerve stimulation can be used as a reversible treatment option for refractory urge incontinence related to detrusor hyperreflexia in select patients with spinal lesions. KEY WORDS: bladder, neurogenic; reflex, abnormal; sacrum; urinary incontinence

Since the first publication of 2 case reports by Schmidt,1 treatment of neurogenic urge urinary incontinence with chronic sacral nerve stimulation did not gain much popularity. In spinal cord diseases with upper motor neuron lesions incontinence is usually due to detrusor hyperreflexia. Using chronic spinal experimental models it has been demonstrated that a new sacral segmental reflex arc may become functional because of neuroplasticity.2, 3 The afferent limb of this new reflex arc includes unmyelinated C fibers. Spinal cord injuries, multiple sclerosis, spinal vascular lesions, spinal tumors and viral myelitis are the most com-

mon spinal causes of hyperreflexic neurogenic bladder. Voiding disorders, which include various symptoms of incontinence, urgency and irritative voiding complaints, urinary retention due to striated detrusor sphincter dyssynergia and pelvic pain have developed in these patients. The type and location of spinal cord disease may explain individual variation in urinary symptoms and degree of reported inconvenience. Therapeutic options for detrusor hyperreflexia vary and include conservative treatments, such as pelvic floor training, pharmacotherapy with oral anticholinergic drugs and vesical instillation of vanilloid agents4 with or without clean intermittent self-catheterization. Surgical procedures are a last resort and are irreversible, potentially leading to serious side effects, including bladder augmentation, sacral anterior root stimulation5 and urinary diversion. Since the 1960s, electrical stimulation has been used to treat detrusor instability.6 During the last decade, after the publications of Schmidt,1, 7 functional electrical stimulation

Accepted for publication May 26, 2000. * Financial interest and/or other relationship with Afferon, Alza Pharmaceuticals, Medtronic, Pharmacia Upjohn and Situs, Inc. † Financial interest and/or other relationship with Lab Fabre, Fournier Debat, Ethicon and Boeringer Ingelheim. Editor’s Note: This article is the first of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with questions on pages 1714 and 1715. 1476

S3 NEUROSTIMULATION AND DETRUSOR HYPERREFLEXIA

has gained popularity for the treatment of chronic lower urinary tract dysfunctions, especially those related to uninhibited bladder contractions. Sacral nerve stimulation has been extensively developed and evaluated during the last 5 years.8 –11 Before surgically implanting the sacral nerve stimulation device, a test stimulation trial is performed to evaluate its effectiveness. We used sacral nerve stimulation to treat patients with a history of refractory neurogenic urge incontinence and detrusor hyperreflexia resistant to first line conservative treatment who were candidates for bladder augmentation. MATERIALS AND METHODS

Since we started percutaneous test stimulation of S3 sacral nerve at our institutions, 23 patients with urge incontinence due to detrusor hyperreflexia who were eligible for bladder augmentation because prolonged pharmacological therapy failed completed testing. Test stimulation is the method of choice to evaluate individual response to sacral nerve stimulation and is comprised of 2 phases. The acute phase of the test involves percutaneous location and identification of the sacral nerve that provides the best neuroanatomical response, which is typically a levator ani contraction and flexion of the great toe. During the chronic phase continuous trial stimulation of the selected sacral nerve is performed for more than 3 days. Stimulation parameters used were pulse width 210 microseconds, frequency 10 Hz. and amplitude ranging 0.5 to 10 V. self-managed by the patient with an external test stimulator device. We selected candidates for surgical implantation based on informed consent, neurogenic pure urge incontinence with confirmed detrusor hyperreflexia, no abnormality of upper urinary tract and kidney function, prolonged failure of conservative treatment, severe alteration of quality of life leading to indication of bladder augmentation, positive neuroanatomical response to acute test stimulation of the S3 nerve, and clinical improvement of more than 75% with a decrease in the number of leakage episodes per day and pad use, decrease in number of voids per day (voiding and intermittent self-catheterization) and/or urgency symptom improvement if available during chronic test stimulation. Criteria also included no observed side effects during test stimulation, return to baseline clinical parameters at conclusion of a successful test, patient desire for surgical implantation after test and level of individual satisfaction was greater than 75% on the visual analog scale, and females with a negative pregnancy test on oral contraception or those who were postmenopausal. Pregnancy risk information was carefully provided at baseline. If the subject became pregnant after implantation the neurostimulator was deactivated until delivery. Preoperative management included patient voiding diaries for 4 continuous days, serum creatinine determination, urine culture, urodynamics, urethrocystography, kidney ultrasound and sacral plane x-ray. Methods, units and definitions used in relation to urodynamics were in accordance with standards recommended by the International Continence Society.12, 13 Filling of the bladder was performed with normal saline at a fill rate of 50 ml. per minute at room temperature. Cystometry during followup was performed with and without sacral nerve stimulation. Patient voiding diaries were used to record leakage episodes, pad use, including type and change, spontaneous voided volume, clean intermittent self-catheterization voided volume if available, date and time. Surgical implantation of the neurostimulator was performed according to procedures described previously.14 The implanted devices included a neurostimulator, quadripolar lead and an extension that connected these 2 devices. Stimulator activation. Stimulation was typically activated 1 day postoperatively. Similar to the test stimulation procedure, the identical pulse width and rate parameters were

1477

used after implantation. Pulse amplitude was increased slowly in increments of 0.1 V. until the subject sensed stimulation and/or pelvic floor muscle responses were obtained. While determining threshold voltages for the various electrodes, impedance values for the electrodes were measured. Each of the 4 electrodes was tested, and the electrode providing the best motor/sensory response at the lowest impedance was selected as the cathode and the case was selected as the anode. Followup. Voiding diaries, urodynamics, serum creatinine measurement and ultrasound examination were performed at 6 months and last followup. Each visit included scar examination, neurostimulator parameter review and monitoring for urinary tract infection. At each followup visit patients were provided the opportunity to cease stimulation therapy. All data were reported as individual results and mean values.

RESULTS

Implantation was performed in 9 female patients, including 4 in Paris and 5 in Rotterdam. Patient profiles are summarized in table 1, and all had a neurourological history for more than 3 years. Baseline urological symptoms in patients presenting with multiple sclerosis were documented to be stable for at least 1 year at the time of implantation. Mean followup was 43.6 months (range 7 to 72). Unilateral leads were surgically placed in the right S3 in 5 patients or left S3 in 4. There were no perioperative complications. All patients reported clinically significant improvements immediately after stimulator activation. Of the 9 patients 5 using clean intermittent self-catheterization before implantation have not been able to void. At 6 months after implantation frequency improved from 16.1 to 8 voids per day, and mean volume per void correspondingly increased from 115 to 249 ml. (table 2). Of the 9 patients 6 were completely dry at 6 months, and pad usage clearly decreased in the remaining 3 (table 2). Drug use was no longer required by the patients with implants. The symptomatic results obtained at 6 months remained stable for an average of 43.6 months for all except patients 5 and 8 (table 2). Urodynamic results revealed improvement of maximum bladder capacity and volume at first uninhibited contraction when neurostimulator was activated (table 3). Hyperreflexia disappeared in patient 7. All patients reported an “on-off” effect that was confirmed on urodynamic studies, that is there was no observed difference in urodynamic results between baseline (without stimulation) and at 6 months with the neurostimulator programmed to off (no stimulation). Initial stimulation parameters did not change with time with respect to pulse width (210 microseconds) and frequency (10 Hz.). Amplitude was titrated to avoid painful stimulation with a mean amplitude of 3.9 V. at last followup. Stimulation parameter adjustment and patient adaptation to the stimulation management routine necessitated a mean of 3.5 outpatient visits specifically dedicated to this topic during the first 3 months after device implantation. Patient 1 reported painful stimulation sensations in the lower foot and leg 6 months after implantation, which was treated with pulse width reprogramming from 180 microseconds decreased to 150 microseconds. The improvement was incomplete despite numerous attempts at noninvasive reprogramming. After completing another test stimulation procedure on the contralateral side the lead was surgically relocated 7 months after implantation. Urological efficiency was regained and stimulation therapy was better tolerated. All patients subjectively reported improved visual analog scale results by at least 75% at last followup.

1478

S3 NEUROSTIMULATION AND DETRUSOR HYPERREFLEXIA

No No Clean intermittent selfcatheterization before implant

Detrusor hyperreflexia, detrusorsphincter dyssynergia Yes Detrusor hyperreflexia Urodynamic diagnosis

Detrusor hyperreflexia, detrusorsphincter dyssynergia Yes

Yes

Detrusor hyperreflexia

Detrusor hyperreflexia, detrusorsphincter dyssynergia Yes

Detrusor hyperreflexia, detrusorsphincter dyssynergia No

Detrusor hyperreflexia, detrusorsphincter dyssynergia No

Spinal cord injury Detrusor hyperreflexia, detrusorsphincter dyssynergia Yes Multiple sclerosis Multiple sclerosis Multiple sclerosis Multiple sclerosis

Spinal cord injury Detrusor hyperreflexia Rubeolic myelitis Vascular myelitis

Multiple sclerosis

26 3 53 5 39 16 42 8 53 9 35 14 48 16

43 18

60

45 15

Pt. 8

60 60

Pt. 7 Pt. 6

72 72

Pt. 5 Pt. 4

TABLE 1. Patient profiles

7 11 19 31

Pt. 3 Pt. 2 Pt. 1

Mean mos. followup Pt. age at implant Mean yrs. since diagnosis Diagnosis

Pt. 9

DISCUSSION

There is an effect of sacral nerve stimulation on neurogenic urge incontinence. Conservative treatment of incontinence related to detrusor hyperreflexia by chronic S3 sacral nerve stimulation is feasible for select patients. Including Ruud Bosch and Groen,15 to our knowledge we are the first to report on clinical efficacy of neuromodulation for the treatment of incontinence in patients who have detrusor hyperreflexia. In confirmation of early reports made by Schmidt,1 our studies document the potential of this therapy for neurogenic bladder and spinal cord disease. It is thought that treatment of detrusor instability in patients presenting with nonneurogenic incontinence could be partly subjective. To our knowledge correlation between urodynamics and clinical results has not been previously demonstrated for nonneurogenic irritative symptoms.8, 9 Our series confirms the efficiency of sacral nerve stimulation in patients with urodynamically demonstrable detrusor hyperreflexia. Beneficial effects on incontinence with transcutaneous stimulation have been reported based on subsequent experiments with stimulation of the pudendal nerve via electrodes on the anal mucosa, vaginal mucosa, perineal skin and penis. Suppression of unstable bladder contractions has been effective with electrical stimulation of pudendal afferent branches in the penis. External magnetic stimulation has also been reported to suppress bladder contractions.16 We hypothesize that sacral nerve stimulation depends on electrical stimulation of afferent axons in the spinal roots that in turn modulate voiding and continence reflex pathways in the central nervous system. The afferent system is the most likely target because beneficial effects can be elicited at stimulation intensities that do not activate striated muscle contractions.2, 17, 18 A key question that remains to be answered is what type of afferents contribute to the effects of sacral nerve stimulation. It is likely that effects are mediated by somatic afferents because visceral afferents even myelinated A-delta fibers would not be activated by stimulation amplitudes typically used in sacral neuromodulation to produce an effective response.19 The principles behind sacral nerve stimulation can be summarized as somatic afferent inhibition of sensory processes. All 9 of our patients had dramatic symptomatic improvement in incontinence. In regard to the visual analog scale patients reported significant, positive changes in everyday life. Urodynamic studies showed an increase of maximum bladder capacity and bladder volume at first uninhibited contraction. Although statistical comparison was not possible due to small sample size, the urodynamic effect on bladder volume appears to correlate with clinical improvement and voiding diary results. The only confounding parameter was detrusor pressure at maximum uninhibited contraction, which was higher during stimulation therapy. The provocative nature of the urodynamic technique could be 1 reason why only 1 patient exhibited total suppression of detrusor hyperreflexia during the test. Individual neurological variations could also explain these results. Ruud Bosch and Groen also reported opposing urodynamic and clinical results in 1 patient.9, 20 Another plausible explanation may be that by inhibition of early premature bladder contractions, functional bladder volume is increased and contraction might reach a higher amplitude because of larger volume at contraction initiation. Currently the criteria for surgical implantation are based on clinical and subjective improvements during test stimulation. Further research to define additional predictive factors that correlate with favorable clinical outcome should be performed and include assessment of urodynamic and stimulation parameters. According to Fall and Lindstro¨m, use of

1479

S3 NEUROSTIMULATION AND DETRUSOR HYPERREFLEXIA TABLE 2. Voiding diary results with comparison of data before and after implantation Voiding frequency/day (24-hr. clean intermittent selfcatheterization, natural voiding): Before implant 6 Mos. after implant Last followup after implant Mean ml. vol./voiding: Before implant 6 Mos. after implant Last followup after implant Mean No. leakages/24 hrs.: Before implant 6 Mos. after implant Last followup after implant Pad use: Before implant After implant

Pt. 1

Pt. 2

Pt. 3

Pt. 4

Pt. 5

Pt. 6

Pt. 7

Pt. 8

Pt. 9

Mean

35 8.2 7.7

11.4 5.3 5.5

16.2 6.5 7.1

10.8 5.7 5.7

11.8 11.4 13.7

18.8 11.3 9.0

13.2 6.3 7.6

15.0 10.3 10.0

12.9 7.6 5.7

16.1 8.1 8

150 378 345 12.5 0 1 Yes Yes

125 250 250

90 235 219

8 1 0–1

7 1 0–1

Yes No

Yes No

110 330 335

140 189 125

9.1 0 0 Yes No

82 126 158

3.9 0 2.0 Yes Yes

3.4 0.5 0 Yes No

119 379 546

193 113 139

6.5 0 0 Yes No

24 102 209

4.7 0 0

11.3 3.0 0

Yes Yes

115 233 249 7.3 0.6 0.3

Yes No

TABLE 3. Results of urodynamic studies before and 6 months after implantation Max. bladder capacity before leakage (ml.): Baseline before implant S3 off at 6 mos. S3 on at 6 mos. Vol. at first uninhibited contraction (ml.): Baseline before implant S3 off at 6 mos. S3 on at 6 mos. Detrusor pressure at max. unstable contraction (cm. H2O): Baseline before implant S3 off at 6 mos. S3 on at 6 mos.

Pt. 1

Pt. 2

Pt. 3

Pt. 4

Pt. 5

Pt. 6

Pt. 7

Pt. 8

Pt. 9

Mean

190 153 600

169 160 250

135 145 340

235 250 460

478 487 452

201 311 331

177 291 490

267 93 135

344 350 343

244 249 377

100 145 600

145 110 160

135 125 320

200 220 455

440 487 405

195 307 328

175 Not applicable Not applicable

210 72 134

326 350 319

214 227 340

45 35 45

80 80 70

90 96 64

48 44 100

56 24 25

35 36 46

48 Not applicable Not applicable

24 33 41

110 116 108

59 58 62

increased stimulation frequency parameter may be considered to obtain better efficiency of bladder inhibition.21 As previously reported15 the majority of our patient population is female. It has been advocated that sacral nerve stimulation therapy may be more effective in females.8 Females presenting with neurogenic bladder dysfunction have more to gain from sacral neuromodulation than males because there are no acceptable urine collection devices for the incontinent female population. Candidates selected for neuromodulation should be able to make basic transfers independently. Postponing the initiation of voiding reflex in patients who are not capable of independent voiding is not helpful from a clinical perspective except in those who perform clean intermittent self-catheterization. Consistent with the previous report,15 our favorable results were also obtained in a female subpopulation. Given the fact that none of these women presented with stress incontinence, assessment of bladder neck behavior would have been of scientific interest. We found that modulation of urge sensation was an important success factor for our patients and confirmed previous findings reported for the treatment of nonneurogenic irritative voiding disorders.22 From our series we believe that suppression of detrusor hyperreflexia is not likely to be the primary mechanism of action for neuromodulation. The indication for long-term sacral neuromodulation therapy is based on the results of acute and chronic test stimulation. Most authors now agree with this diagnostic strategy, and Bemelmans et al recently summarized the typical schematic overview of assessment before surgical implantation.22 We hope that urodynamic study results will become a key decision factor for implantation to improve the long-term results and facilitate expansion of this treatment modality to neurogenic cases.

CONCLUSIONS

Because sacral neuromodulation therapy is nondestructive and does not compromise the potential for future treatment, it appears to be a promising therapy for neurogenic incontinence. The long-term results will have to be carefully studied recognizing that evolution of neurological disease can sometimes be unpredictable (multiple sclerosis). Simplicity of the surgical implantation technique and low complication rate are encouraging. To date, bladder augmentation has been successfully avoided in our series of neurogenic cases. The main advantages of sacral neuromodulation include a potential curative effect, few side effects and relatively low cost, assuming the life of the neurostimulator battery exceeds 5 years.21 In an era of minimally invasive therapy development sacral nerve stimulation could be a viable treatment modality for patients who are disabled. REFERENCES

1. Schmidt, R. A.: Treatment of unstable bladder. Urology, 37: 28, 1991 2. de Groat, W. C., Kawatani, M., Hisamitsu, T. et al: Mechanism underlying the recovery of urinary bladder function following spinal cord injury. J Auton Nerv Syst, 30: S71, 1990 3. Yoshimura, N.: Bladder afferent pathway and spinal cord injury: possible mechanisms inducing hyperreflexia of the urinary bladder. Prog Neurobiol, 57: 583, 1999 4. Chancellor, M. B. and de Groat, W. C.: Intravesical capsaicin and resiniferatoxin therapy: spicing up the ways to treat the overactive bladder. J Urol, 162: 3, 1999 5. Brindley, G. S.: The first 500 sacral anterior root stimulator implants: general description. Paraplegia, 32: 795, 1994 6. Caldwell, K. P.: The treatment of incontinence by electronic implants: Hunterian Lecture delivered at the Royal College of Surgeons of England on 8th December 1966. Ann R Coll Surg Engl, 41: 447, 1967

1480

S3 NEUROSTIMULATION AND DETRUSOR HYPERREFLEXIA

7. Schmidt, R. A.: Applications of neurostimulation in urology. Neurourol Urodyn, 7: 585, 1988 8. Koldewijn, E. L., Rosier, P. F. W. M., Meuleman, E. J. H. et al: Predictors of success with neuromodulation in lower urinary tract dysfunction: results of trial stimulation in 100 patients. J Urol, 152: 2071, 1994 9. Bosch, J. L. H. R. and Groen, J.: Sacral (S3) segmental nerve stimulation as a treatment for urge incontinence in patients with detrusor instability: results of chronic electrical stimulation using an implantable neural prosthesis. J Urol, 154: 504, 1995 10. Chartier-Kastler, E., Richard, F., Denys, P. et al: S3 sacral neuromodulation in patients with chronic refractory miction disorders. Presse Med, 26: 466, 1997 11. Schmidt, R. A., Jonas, U., Oleson, K. A. et al: Sacral nerve stimulation for treatment of refractory urinary urge incontinence. J Urol, 162: 352, 1999 12. Abrams, P., Blaivas, J. G., Stanton, S. et al: Standardization of terminology of lower urinary tract function. In: Clinical NeuroUrology, 2nd ed. Edited by R. J. Krane and M. B. Siroky. Boston: Little, Brown and Co., Appendix, pp. 649 – 670, 1991 13. Abrams, P.: Urodynamics, 2nd ed. London: Springer-Verlag, chapt. 3, pp. 18-117, 1997 14. Siegel, S. W.: Management of voiding dysfunction with an implantable neuroprosthesis. Urol Clin North Am, 19: 163, 1992 15. Ruud Bosch, J. L. and Groen, J.: Treatment of refractory urge

16.

17.

18.

19.

20.

21.

22.

incontinence with sacral spinal nerve stimulation in multiple sclerosis patients. Lancet, 348: 717, 1996 Foley, S. J., McFarlane, J. P., de Winter, P. et al: The effect of detrusor sphincter dyssynergia on neuromodulation in detrusor hyper-reflexia. Br J Urol, 79: 47, 1997 Vodusˇek, D. B., Light, J. K. and Libby, J. M.: Detrusor inhibition induced by stimulation of pudendal nerve afferents. Neurourol Urodyn, 5: 381, 1986 Thon, W. F., Baskin, L. S., Jonas, U. et al: Surgical principles of sacral foramen electrode implantation. World J Urol, 9: 133, 1991 Kruse, M. N. and de Groat, W. C.: Spinal pathways mediate coordinated bladder/urethral sphincter activity during reflex micturition in decerebrate and spinalized neonatal rats. Neurosci Lett, 152: 141, 1993 Bosch, J. L. and Groen, J.: Neuromodulation: urodynamic effects of sacral (S3) spinal nerve stimulation in patients with detrusor instability or detrusor hyperflexia. Behav Brain Res, 92: 141, 1998 Fall, M. and Lindstro¨m, S.: Electrical stimulation. A physiologic approach to the treatment of urinary incontinence. Urol Clin North Am, 18: 393, 1991 Bemelmans, B. L., Mundy, A. R. and Craggs, M. D.: Neuromodulation by implant for treating lower urinary tract symptoms and dysfunction. Eur Urol, 36: 81, 1999