- Email: [email protected]
Efficacy and Safety of Sacral Nerve Stimulation for Urinary Urge Incontinence: A Systematic Review
Efficacy and Safety of Sacral Nerve Stimulation for Urinary Urge Incontinence: A Systematic Review Miriam Brazzelli,* Alison Murray and Cynthia Fraser From the Department of Clinical Neurosciences, University of Edinburgh, Edinburgh and Health Services Research Unit, University of Aberdeen, Aberdeen, United Kingdom
Purpose: We systematically reviewed the evidence on the efficacy and safety of sacral nerve stimulation (SNS) for severe urge incontinence. Materials and Methods: A systematic review was performed of primary studies of SNS for urge incontinence published in English between 1966 and May 2003, and identified in major electronic databases. Two reviewers independently selected studies, assessed their methodological quality and extracted data. Results: Four randomized controlled trials and 30 case series were identified. Evidence from the randomized controlled trials, involving approximately 120 patients, showed that about 80% achieved continence or greater than 50% improvement in their main incontinence symptoms after SNS compared with about 3% of controls receiving conservative treatments while waiting for an implant. While case series were larger, they were methodologically less reliable. However, they showed similar results with 67% of patients becoming dry or achieving a greater than 50% improvement in symptoms after implantation. Incontinence episodes, leakage severity, voiding frequency and pad use were significantly lower after implantation. Benefits were reported to persist 3 to 5 years after implantation. Adverse events were documented in 27 studies. Overall the reoperation rate in implanted cases was 33%. The most common reason for surgical revision was relocation of the generator because of pain and infection. Common complications were pain at the implant or lead site in 25% of patients, lead related problems such as lead migration in 16%, replacement and repositioning of the implanted pulse generator in 15%, wound problems in 7%, adverse effects on bowel function in 6%, infection in 5% and generator problems in 5%. Permanent removal of the electrodes was reported in 9% of patients. Technical changes with time have been associated with decreased complication rates. Conclusions: There is evidence indicating that SNS is effective for decreasing symptoms in patients with urge incontinence. Adverse events occurred in about half of the implanted cases and surgical revision was performed in 33%. No major irreversible complications were reported in the studies reviewed. Further research is required on the long-term effects of and quality of life in patients with SNS. Key Words: bladder, urinary incontinence, prostheses and implants, electric stimulation, equipment safety
humans since the early 1980s. In 1986 Tanagho and Schmidt reported results in 10 patients who had electrodes implanted on the sacral roots for neuropathic voiding dysfunction.1 Since then, in excess of 8,000 implant procedures have been performed to treat a number of voiding problems refractory to standard conservative treatment.2 The use of SNS is based on the knowledge that electrical stimulation of the pudendal nerves can modulate neural reflexes that influence bladder and pelvic floor behavior. However, the exact physiological mechanism of action by which electrical nerve stimulation works is not yet fully understood.3 In practical terms SNS therapy involves the use of mild electrical pulses to stimulate the sacral nerves. Electrodes are placed next to a sacral nerve, usually S3, by inserting the electrode leads percutaneously into the corresponding foramen of the sacrum with the patient under local anesthesia. Adequate electrode placement is confirmed by obtaining appropriate motor and sensory responses. After this phase of peripheral nerve evaluation patients enter the so-called subchronic phase, which lasts 3 to 7 days and is usually considered successful when there is at least 50% improvement in the main incontinence symptoms. Patients with successful
rge urinary incontinence is one of the most common forms of urinary incontinence. It is typically managed conservatively by behavioral techniques (eg bladder training), physical therapies (eg electrical stimulation using vaginal or anal electrodes) or pharmacotherapy (antimuscarinic and anticholinergic drugs). When these approaches are unsuccessful, more invasive and irreversible surgical procedures are considered, including bladder reconstruction, for example augmentation cystoplasty, and urinary diversion. Implanted SNS has been proposed as an option for severe urge urinary incontinence. SNS has been investigated in
U
Submitted for publication March 16, 2005. The Health Services Research Unit is supported by a core grant from the Chief Scientist Office of the Scottish Executive Health Department and grant support from the United Kingdom National Institute for Clinical Excellence. The views expressed are those of the authors and not necessarily shared by the funders. * Correspondence: Department of Clinical Neurosciences, University of Edinburgh, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, United Kingdom (telephone: 0044 131 5372955; FAX: 0044 131 332 5150).
0022-5347/06/1753-0835/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION
835
Vol. 175, 835-841, March 2006 Printed in U.S.A. DOI:10.1016/S0022-5347(05)00326-5
836
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE
treatment receive a permanent implant, which consists of electrodes inserted subcutaneously and attached to an implantable pulse generator. The procedure is reversible and the implanted pulse generator can be removed at any time. Currently the SNS implanted device is exclusively produced under the name InterStim® therapy. InterStim® therapy was initially marketed for urge incontinence in Europe, Canada and Australia in 1994, and it subsequently received Food and Drug Administration approval in the United States in 1997. We systematically assessed all available evidence on the efficacy and safety of SNS for urinary urge incontinence.
MATERIALS AND METHODS Search strategy. We identified published and unpublished studies evaluating the efficacy and safety of SNS for urinary urge incontinence by searching major electronic databases (see Appendix). In addition, we perused the reference lists of all included studies and contacted experts in the field to inquire about further potentially eligible references. Selected websites were also searched for eligible, evidence based reports. Details of the search strategy and list of selected websites are available on request. Study selection and data extraction. To be included studies had to assess the efficacy and/or safety of implanted SNS in patients with urge urinary incontinence. We excluded studies of transcutaneous electrical nerve stimulation, magnetic sacral nerve stimulation and sacral anterior root stimulation (the Brindley technique). Our primary outcomes of interest for efficacy were the cure/improvement rate, number of leakage episodes daily, number of pads used daily, leakage severity and quality of life. Efficacy measures were derived from patient voiding diaries, physiological measurements, and published questionnaires and checklists. The definitions of cure and improvement were those provided in the reports of the primary studies. Cure was usually defined as no incontinence or greater than 90% clinical improvement, while improvement was defined as a 50% or greater decrease in main incontinence symptoms. The frequency and type of any reported adverse events were tabulated to assess SNS safety. Two of us (MB and AM) assessed each study for inclusion and performed data extraction independently using a specifically designed data extraction form. In case of multiple reports of a study we selected the most up-to-date and complete report. NonEnglish language papers were noted but not included. Any differences of opinion between reviewers about study selection and data extraction were resolved by discussion or arbitration. From each included study we extracted data on study design and quality, participant characteristics, intervention characteristics, and efficacy and safety results. Criteria for assessing the methodological quality of full text trials were the randomization method, allocation concealment, blinding procedures, group comparability at baseline and the number of withdrawals/dropouts. Quality assessment of full text case series included sample selection and representativeness, description of inclusion/exclusion criteria, identification of important prognostic factors, surgeon experience, choice and validity of outcome measures, number lost to followup and result completeness. We were not
able to assess the methodological quality of conference proceedings due to the limited information available.
RESULTS Search results. The literature search identified 1,562 reports, of which 34 English language reports met inclusion criteria, including 21 full text articles and 13 abstracts. Four of the 34 included studies were RCTs4 –7 and 30 were case series.8 –37 Overall they enrolled 3,146 patients. A total of 26 studies also included patients with urinary retention and/or with urgency/ frequency. Across the studies mean or median patient age was between 34 and 69 years. Overall 2,067 patients were reported to have undergone peripheral nerve evaluation testing and 1,827 received implanted SNS. Sample size was 9 to 581 patients and the proportion of women was 50% to 100% with 1 study including only women.4 Average followup was between 5.330 and 47.110 months (range up to 96).10 Table 1 lists details of the included studies. Efficacy of SNS in RCTs. Table 2 lists cure and improvement rates at 6 months in patients with urge incontinence randomized to a stimulation or a delayed group. Overall 50% of patients in the stimulation group achieved complete continence or greater than 90% improvement in the main incontinence symptoms compared with 1.6% of patients in the delay group.5,7 A 50% improvement in main incontinence symptoms was observed in approximately 37% and 3% of patients in the stimulation and delay groups, respectively.5–7 In addition, the trials of Weil7 and Schmidt5 et al provided further clinical results. However, most outcomes were presented as within-group comparisons (baseline results vs post-intervention results in each group) instead of betweengroup comparisons (stimulation vs control group). Table 3 shows that the number of leakage episodes daily and the number of pads used daily were significantly lower 6 months after implantation in the stimulation group compared with baseline. The mean severity of leakage was also lower at followup, although this was only significant in 1 trial.5 In contrast, patients in the delay group showed no significant improvement or worsening of incontinence symptoms. Weil et al also observed that mean bladder capacity assessed by cystometry significantly increased at 6 months compared with baseline in the stimulation group.7 Changes in urodynamic parameters were not reported in the delay group. Weil et al also noted significantly lower mean leakage episodes (p ⬍0.0005), pad use (p ⬍0.0005) and leakage severity (p ⫽ 0.047) in the implanted group compared to the control group at 6 months. The randomized, controlled trials of Weil7 and Schmidt5 et al also assessed the impact of SNS on patient quality of life using the SF-36, short form. Weil et al found a significant difference between groups in only the emotional role score in the physical health component of the questionnaire (p ⫽ 0.037).7 Similarly Schmidt et al observed a significant between-group difference 6 months after implantation in the physical health component of the questionnaire (p ⫽ 0.0008) but not in the mental health component.5
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE
837
TABLE 1. Included studies References Full text studies: Aboseif et al8 Amundsen and Webster9 Bosch and Graen10 Braun et al11 Cappellano et al13 Cappellano et al12 Edlund et al17 Everaert et al18 Grünewald and Jonas20 Hedlund et al21 Hohentellner et al23 Janknegt et al25 Janknegt et al24 Scheepens et al31 Scheepens et al30 Schmidt et al5 Shaker and Hassouna32 Siegel et al33 Spinelli et al35 Weil et al7 Weil et al37 Totals Abstracts: Caraballo et al14 Das et al15 Dijkema et al16 Everaert et al4 Groenendijk et al19 Heesakkers et al22 Koldewijn et al26 Peters et al27 Ruffion et al28 Ruiz-Cerdá et al29 Spinelli et al34 Thon36 Weil et al6 Totals
Type Study
No. Enrolled (all diagnoses)
Rec No. Peripheral Nerve Evaluation
No. Implant
Mos Followup (range)
47 69 46.2 49 51.1 47 59.8 43 49 54 43.4 Not reported 46 53 51 46.6 42.3 43 43 43 36
160* 25 85 9* 113* 47* 30* 53* 184* 53 11* 96 10* 15* 39* 155 18 581* 32* 123* 36* 1,875*
160* 25 85 Not reported Not reported 47* 30* 177* 184* 53 Not reported Not reported 10* 15* Not reported 155 Not reported 581* 13* 123* Not reported 1,658*
64* 12 45 9* 113* 10 9* 53* 55* 14 10* 96 8* 15* 39* 34 18 219* 22* 44 36* 925*
Mean 24 (6–36) Mean 7.8 (1–16) Mean 47.1 (6–96) Mean 12.5 (7–18) 18 Mean 23.1 (3–47) Mean 19.9 (8–39) Mean 24 (13–39) Mean 44.3 (1–89) Mean 18 (9–32) Mean 13 (9–28) Mean 30.8 (12–60) Mean 16 (4–36) Mean 59 (30–90) Mean 5.3 (1–10) Mean 14.7 (0.9–39.7) Mean 18.8 (3–83) (18–36) Mean 11a (2–25) Median 18 (6–36) Mean 37.8 (12–60)
60.6 47 reported 48 reported reported 40 reported 48.8 47 34 reported reported
17* 256* 25 22* 111 259* 40* 30* 166* 204* 9* 114* 18* 1,271*
Not Not Not Not Not Not Not
17* 256* 25 22* 111 259* 40* 14* 33* 69* 6* 41* 9* 902*
Mean 13.4 (3–22) Mean 26 (15–46) 6 or Greater 12 6 Greater than 12 Mean 29 (5–46) Mean 37 (3–87) Mean 6.8 (2–30) Not reported Mean 4.2 (1–12) 6
Mean Pt Age
Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case RCT Case Case Case RCT Case
series series series series series series series series series series series series series series series
Case Case Case RCT Case Case Case Case Case Case Case Case RCT
series series series
series series series series
series series series series series series series series
Not Not Not Not
Not Not
reported reported reported reported reported reported reported 30* 166* 204* 9* Not reported Not reported 409*
* Including patients with urinary retention and/or urgency-frequency.
Efficacy of SNS in case series. Of the 30 identified case series 22 provided suitable data on the efficacy of SNS in patients with urge urinary incontinence. Table 4 shows the cure, improvement and failure rates reported in 17 case series at last followup. Followup varied among the studies. The cure rate was 7% to 64% and 139 of 361 patients (39%) were reported to be cured. A 50% or greater improvement in incontinence symptoms was observed in 199 of 501 patients (40%). Overall 338 of 501 patients (67%) had a decrease in incontinence symptoms after SNS. In contrast, 163 of 501 cases (32%) did not show a 50% or greater improvement in incontinence symptoms after implantation and they were
TABLE 2. Success rates at 6 months in randomized, controlled trials No./Total No. (%) References Stimulation group: Weil et al7 Schmidt et al5 Weil et al6 Delay group: Weil et al7 Schmidt et al5 Weil et al6
Cured
Improved
9/16 (56) 16/34 (47) Not reported
Not reported (29) 10/34 (29) 5/5 (100)
1/22 (5) 0/42 Not reported
0/22 2/42 (5) Not reported
Cured— greater than 90% and improved— greater than 50% symptom decrease.
classified as failures. Other outcome measures, such as leakage episodes, pad use, leakage severity, voiding frequency and urodynamic parameters, were not recorded consistently. A total of 14 case series indicated the change in the average number of leakage episodes at followup compared to baseline.9⫺13,16,17,22,25,29,31–33,37 Overall leakage frequency was 4.5 to 11.6 episodes daily at baseline and 0.8 to 5.0 at last followup after implantation (decreased by 53% to 92%). The change was reported to be statistically significant in 11 of the 14 studies (p ⬍0.05). A total of 14 studies compared the number of pads used daily at 6 months or last followup after implantation with the number of pads used at baseline.9⫺12,16,17,21–25,31,33,37 The average number of pads decreased from 3.0 to 8.3 to 0.4 to 3.4 (decreased by 49% to 94%) and the change was statistically significant in 10 studies (p ⬍0.05). The severity of incontinence episodes was assessed on a scale of 1 to 3 as 1—mild, 2—moderate and 3—severe in 4 studies.17,25,31,32 Leakage severity was decreased from an average of 1.4 to 2.0 at baseline to 0.8 to 1.6 at 18 months or last followup (decreased by 16% to 40%). The decrease was statistically significant in 3 of the 4 studies (p ⬍0.05). Voiding frequency was assessed in 8 studies.9,10,21,23,25,31,32,37 The mean number of voids daily decreased from 10.0 to 15.0 at baseline to 7.0 to 9.2 at last followup after implantation (decreased by 30% to 50%). The change was statistically significant in 6 studies (p ⬍0.05). Urodynamic measurements were provided in 9 studies at
838
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE TABLE 3. Leakage episodes, pad usage, leakage severity and bladder capacity in patients with urge incontinence in randomized, controlled trials Mean Stimulation Group ⫾ SD References
Leakage episodes/day: Weil et al7 Schmidt et al5 Pad use/day: Weil et al7 Schmidt et al5 Leakage severity:* Weil et al7 Schmidt et al5 ml bladder capacity Weil et al7
Mean Delay Group ⫾ SD
No. Pts
Baseline
6 Mos
p Value
No. Pts
Baseline
6 Mos
p Value
21 34
13.5 ⫾ 7.5 9.7 ⫾ 6.3
1.4 ⫾ 3.3 2.6 ⫾ 5.1
⬍0.0005 ⬍0.0001
22 42
13.5 ⫾ 7.8 9.3 ⫾ 4.8
11.2 ⫾ 5.6 11.3 ⫾ 5.9
Not significant 0.002
21 34
8.7 ⫾ 6.8 6.2 ⫾ 5.0
0.7 ⫾ 1.3 1.1 ⫾ 2.0
⬍0.0005 ⬍0.0001
22 42
8.7 ⫾ 7.1 5.0 ⫾ 3.7
6.8 ⫾ 4.0 6.3 ⫾ 3.6
Not significant 0.003
21 34
2.1 ⫾ 0.6 2.0 ⫾ 0.7
1.6 ⫾ 1.7 0.8 ⫾ 0.9
Not significant ⬍0.0001
22 42
2.1 ⫾ 0.6 1.8 ⫾ 0.6
2.1 ⫾ 0.6 2.0 ⫾ 0.6
Not significant 0.006
21
266 ⫾ 112
370 ⫾ 91
0.013
23
Not reported
Not reported
-
* Assessed on a scale of 0—dry, 1—loss of a few drops of urine, 2—loss of 1 to 2 tablespoons of urine and 3—complete wetting/soaked pad or outer clothing.
baseline and at last followup.10,16,20,21,23,25,31,32,37 Average total voided volume increased from 99 to 195 ml per void to 176 to 402 ml per void (34% to 288% change). The increase was significant in 8 of the 9 included studies (p ⬍0.05). Bladder capacity assessed by cystometry was considered in 9 studies.11,12,16,20,21,23,32,36,37 Total bladder capacity increased from 122 to 400 ml at baseline to 273 to 596 ml at 6 months or last followup (15% to 197% change). The change from baseline was statistically significant in 6 studies. Three case series assessed the impact of SNS on patient quality of life.9,13,32 Capellano et al reported results in 47 patients assessed using a 22 item, domain specific questionnaire developed to detect modifications in self-perceived incontinence severity.13 The questionnaire score was calculated on a scale 0 —poor self-perceived quality of life to 100 —incontinence did not negatively impact quality of life. The average quality of life index score was significantly higher after implantation (83.8 vs 34.4, p ⬍0.01). Amundsen and Webster observed a significantly better score on the Incontinence Impact Questionnaire at last followup compared to baseline (250 at baseline to 62 at last followup, p ⫽ 0.03).9 Shaker and Hassouna used the SF-36 and Beck Depression Inventory to assess the impact of SNS on quality of life in 18 patients with refractory urge incontinence.32 An improvement of 10% to 40% was detected in the Beck Depression
Inventory. No significant differences were observed in the scores of any SF-36 subscales except change of health perception, which was reported to be significantly higher at 6 months compared to baseline (significance not shown). Safety of SNS. Adverse events were documented in a total of 993 patients in 24 studies. In most studies the safety profile of SNS was based on pooling data from all patients under investigation, in some studies including patients in urinary retention, and with urgency and frequency, with safety data not provided separately for each clinical diagnosis. Table 5 lists adverse event rates. Of 855 patients 282 (33%) underwent surgical revision of the SNS implant. The most common reasons for reoperation were relocation of the neurostimulator because of pain at the implant site, revision of the lead system for suspected or detected lead migration and infection. Infection was usually managed by antibiotics but deep infection in some patients required explantation of the pulse generator. Common complications were pain at the implant site in 25% of cases, lead migration in 16%, wound problems in 7%, an adverse effect on bowel function in 6%, infection in 5% and generator problems in 5%. In 42 of 279 patients (15%) the implanted pulse generator was replaced or relocated and 44 of 514 (9%) required permanent explan-
TABLE 4. Cure and improvement rates at last followup in case series References 23
Ruiz-Cerda et al Heesakkers et al22 Amundsen and Webster9 Everaert et al4 Groenendijk et al19 Hedlund et al21 Spinelli et al34 Caraballo et al14 Janknegt et al25 Bosch and Groen10 Grunewald and Jonar20 Siegel et al33 Koldewijn et al26 Hohenfellner et al23 Shaker and Hassouna32 Weil et al37 Dijkema et al16 Total
Abstract Only
Followup (mos)
No. Cured/Total No. (%)
No. Improved/Total No. (%)
No. Failed/Total No. (%)
Yes Yes No Yes Yes No Yes Yes No No No No Yes No No No Yes
Mean 6.8 60 Mean 7.8 12 6 18 Not reported 13.4 Mean 30.8 Mean 47 6 36 Mean 29 Mean 13 Mean 18.8 Mean 37.8 17
14/25 (55) Not reported 2/12 (17) Not reported Not reported 8/14 (57) Not reported 1/15 (7) 25/96 (26) 18/45 (40) 6/26 (23) 19/41 (46) 18/28 (64) Not reported 8/18 (44) 14/24 (58) 6/17 (35) 139/361 (39)
2/25 (8) 27/43 (63) 10/12 (83) 4/5 (80) 55/84 (65) 5/14 (36) 3/3 (100) 11/15 (73) 35/96 (36) 9/45 (20) 13/26 (50) 5/41 (12) 3/28 (11) 5/5 (100) 4/18 (22) 3/24 (13) 5/17 (30) 199/501 (40)
9/25 (36) 16/43 (37) 0/43 1/5 (20) 29/84 (35) 1/14 (7) 0/3 3/15 (20) 36/96 (38) 18/45 (40) 7/26 (27) 17/41 (41) 7/28 (25) 0/5 6/18 (33) 7/24 (29) 6/17 (35) 163/501 (32)
Cured— greater than 90%, improved— greater than 50% and failed—less than 50% symptom decrease.
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE TABLE 5. Adverse event rates Adverse Event
No. Events/No. Pts (%)
Reoperation Implantable pulse generator replaced/relocated Permanent explants Generator problems Electrode ⫹ lead problems Pain Infection Wound problems Adverse bowel function
282/855 (33) 42/279 (15) 44/514 (9) 18/399 (5) 129/785 (16) 162/653 (25) 35/727 (5) 19/273 (7) 20/353 (6)
tation of the pulse generator. No major neurological complications were documented except for a suspected case of nerve injury33 and a case of generalized fasciculation in which the etiology could not be established.18 DISCUSSION SNS for urge urinary incontinence. We systematically reviewed the available evidence on SNS use for urge urinary incontinence. We chose not to consider Brindley sacral anterior root stimulation because the characteristics of this implanted procedure and the categories of patients in whom it is suitable differ from those of SNS. We were not able to combine results from the identified randomized trials in a meta-analysis because of the limited amount of available data. Efficacy data from randomized controlled trials showed that about 80% of patients who received SNS became dry or showed improvement in the main incontinence symptoms. This compared with 3% in the control groups in randomized studies. Fewer episodes of leakage daily, fewer pads used daily and fewer voids daily were usually reported after implantation as well as a small change in volume per void and bladder capacity. On the whole the results of case series were consistent with those of randomized, controlled trials. The overall surgical revision rate in implanted cases was 33%. Pain, lead related complications and pulse generator replacement or relocation appeared to be the most frequently observed adverse events, followed by removal of the pulse generator, wound problems, bowel problems and infection. There were no reports of long lasting neurological adverse events. It is worth noting that safety data were often only reported on a whole population entering a study regardless of the clinical conditions of individuals or patient subgroups. However, studies that did not differentiate patients with urge incontinence symptoms from patients with other clinical diagnoses but reported safety data were deemed suitable for inclusion in this review. The median or mean age of patients considered in the studies included in this review was between 34 and 69 years. Consequently it is uncertain whether the results can apply to an older population. It is noteworthy that the modification of bowel function or adverse effects on bowel function were side effects more often reported in studies that included patients older than 75 years. Limitations and uncertainties. SNS is intended for patients with severe incontinence symptoms refractory to conventional treatments. However, there was great discrepancy in the range of treatments that patients had
839
received before implantation and the severity of incontinence was often not described. In most studies patients had previously been treated with pharmacological therapy and/or surgical operations, while some had received only conservative, nonsurgical treatments, such as behavioral therapy. In all included studies the frequency of incontinent episodes was assessed by voiding/micturition diaries. Although obtaining self-reported information is sometimes problematic, a voiding diary completed for 7 days is considered a reliable and valid method for documenting changes in symptoms in individuals with urge incontinence.38,39 However, the definition of improvement as a greater than 50% decrease in incontinence symptoms was an arbitrary cutoff, of which the clinical importance was not always evident. A 50% or greater decrease in incontinent symptoms is a broad definition of clinical improvement and some patients may just attain the threshold for being classified as improved, ie a 50% decrease, while others may show a much more substantial rate of improvement, eg a 70% to 80% decrease. The clinical impact of potential different levels of improvement was not addressed or discussed in the studies included in this review. The best evidence in this review should come from the 2 full text reports of randomized trials. Nevertheless, neither of these 2 trials was of high quality. In neither was it stated how patient allocation to treatment groups was concealed or whether the outcome assessor was blinded to treatment allocation. Moreover, in each trial intent to treat analysis was not performed. Most outcomes in randomized, controlled trials were also presented as within-group rather than as between-group comparisons. However, the direction and magnitude of differential effects was consistent across trials. Most remaining evidence consisted of case series, which are known to be more prone to biases than randomized, controlled trials. In particular selection bias (patients treated and cases reported chosen by investigators), findings not adjusted for confounding factors (eg age, symptom duration and previous pelvic surgery), the likelihood of some spontaneous improvement because patients were treated at their worst and dropout/withdrawal rates may affect the reliability and magnitude of the treatment effect in case series. The majority of case series were also small, failed to identify important patient prognostic factors and did not provide information on nonresponders and dropouts. The timing of followup varied across case series and was generally of short duration. Furthermore, it was unclear in most studies whether data had been collected prospectively. However, the direction and size of pretreatment and posttreatment differences were consistent across studies and with those of the randomized trials. Patient quality of life was rarely measured in the studies included in this review. Only 3 randomized trials and 3 case series showed some quality of life measures but the results were not consistent across studies. Only the most recently published studies tended to quantify adverse events. Earlier studies discussed them in a more narrative way or did not report them. Most complications observed in included studies were technical problems related to device implantation. The clinical experience and skill of the clinician performing the procedure could have a major impact on the success of the procedure and the subsequent incidence of technical complications. Neither ran-
840
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE
domized trials nor case series provided information about the level of expertise of the clinician(s) performing the procedure. Furthermore, the backup facilities of the hospital/ clinic where the procedures were performed were never described. Evolution of the technique. SNS techniques have evolved with time and reported rates of adverse effects have decreased in parallel. The pulse generator is now positioned in the upper buttock region rather than in the abdominal wall, which has decreased episodes of pain. The test evaluation to select patients for a permanent implant can now be performed as needle test stimulation or as a staged implant using a surgically implantable lead as the test lead. Novel lead systems have been developed for the staged implant evaluation test to decrease infection and prevent lead migration. The feasibility of percutaneous lead placement under local anesthesia and with fascial fixation was recently evaluated in a series of 22 patients.35 The technique is considered to be less invasive, offers the possibility of testing the sensory response during implantation and allows implantation to be performed using local anesthesia. A similar method that has recently gained attention is a complete percutaneous implant in a 2-stage approach. A new, tined permanent lead is used for this technique and no incision or additional fascial fixation is required. The correct placement and identification of lead and electrode positions is confirmed by fluoroscopy. The main disadvantage of the latter 2 procedures is the need for an additional procedure for lead removal in patients who do not respond to stimulation. Currently to our knowledge there is no evidence about the longterm (ie 10-year) efficacy or safety of the procedure and it is likely that further revisions will be required to maintain clinical benefits with time.
CONCLUSIONS The results of randomized, controlled trials provide evidence of a benefit from SNS for decreasing incontinence episodes, pad use and voiding frequency, and in improving bladder capacity and voided volume. Evidence from case series is less reliable because of the risk of potential bias in this type of study design. However, their findings are broadly similar to those of randomized trials, decreasing the likelihood that observed differences were due to bias. The benefits of SNS were reported to persist at followup 3 to 5 years after implantation of the pulse generator. Although the sparse data available suggest improvement, the impact of SNS on the quality of life of patients remains to be established. Surgical revision was necessary in 33% of implanted cases. The most common reasons for reoperation were relocation of the implantable pulse generator because of pain, revision of the lead system and infection. Overall adverse events occurred in almost half of the tested patients. The most common complications were pain at the implant site, lead migration, relocation, replacement or permanent explantation of the implanted pulse generator, and wound problems. Further investigations are needed of the most recent refined SNS techniques, in particular long-term effects and quality of life.
APPENDIX Searched Databases MEDLINE (1966 to week 2, May 2003) MEDLINE Extra (May 29, 2003) EMBASE (1980 to week 21, 2003) CINAHL® (1985 to May 2003) BIOSIS® (1985 to May 2003) Science Citation Index (1981 to June 2003) Web of Science Proceedings (1990 to June 2003) Cochrane Controlled Trials Register (Cochrane Library, Issue 2, 2003) Cochrane Database of Systematic Reviews (Cochrane Library, Issue 2, 2003) Database of Abstracts of Reviews of Effectiveness (May 2003) Health Technology Assessment Database (May 2003) National Research Register (Issue 2, 2003) Clinical Trials (May 2003) Current Controlled Trials (May 2003) Research Findings Register (May 2003)
Abbreviations and Acronyms RCT ⫽ randomized controlled trial SF-36 ⫽ RAND 36-Item Health Survey SNS ⫽ sacral nerve stimulation
REFERENCES 1. Tanagho, E. A. and Schmidt, R. A.: Controlled voiding with neurostimulation. J Urol, suppl., 135: 262A, abstract 635, 1986 2. Abrams, P., Blaivas, J. G., Fowler, C. J., Fourcroy, J. L., Macdiarmid, S. A., Siegel, S. W. et al: The role of neuromodulation in the management of urinary urge incontinence. BJU Int, 91: 355, 2003 3. Turnberg, L., Brocklehurst, J., Fowler, C., Borzyskowski, M., Cardozo, L., Chellingsworth, M. et al: Incontinence Causes, Management and Provision of Services. London: Royal College of Physicians, 1995 4. Everaert, K., Kerckhaert, W. and Caluwaerts, H.: The staged implant does not increase subjective or objective improvement in overactive bladder symptoms in patients selected for sacral nerve stimulation. Neurourol Urodyn, 21: 402, 2002 5. Schmidt, R. A., Jonas, U., Oleson, K. A., Janknegt, R. A., Hassouna, M. M., Siegel, S. W. et al: Sacral nerve stimulation for treatment of refractory urinary urge incontinence. J Urol, 162: 352, 1999 6. Weil, E. H., Eerdmans, P. H. and Janknegt, R.: A new randomized study of neuromodulation versus conventional treatment for incontinence or dysfunctional voiding patterns a preliminary report. Neurourol Urodyn, 15: 284, 1996 7. Weil, E. H., Ruiz-Cerda, J. L., Eerdmans, P. H., Janknegt, R. A., Bemelmans, B. L. and van Kerrebroeck, P. E.: Sacral root neuromodulation in the treatment of refractory urinary urge incontinence: a prospective randomized clinical trial. Eur Urol, 37: 161, 2000 8. Aboseif, S., Tamaddon, K., Chalfin, S., Freedman, S. and Kaptein, J.: Sacral neuromodulation as an effective treatment for refractory pelvic floor dysfunction. Urology, 60: 52, 2002 9. Amundsen, C. L. and Webster, G. D.: Sacral neuromodulation in an older, urge-incontinent population. Am J Obstet Gynecol, 187: 1462, 2002 10. Bosch, J. L. H. E. and Groen, J.: Sacral nerve neuromodulation in the treatment of patients with refractory motor urge incontinence: long-term results of a prospective longitudinal study. J Urol, 163: 1219, 2000 11. Braun, P. M., Boschert, J., Bross, S., Scheepe, J. R., Alken, P. and Juenemann, P.: Tailored laminectomy: a new technique for neuromodulator implantation. J Urol, 162: 1607, 1999
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE 12. Cappellano, F., Ciotti, M. G., Pizzoccaro, M., Catanzaro, M., Santambrogio, S. and Catanzaro, F.: Sacral root neuromodulation in the treatment of female urge and mixed urinary incontinence. Urogynaecol Int J, 12: 111, 1998 13. Cappellano, F., Bertapelle, P., Spinelli, M., Catanzaro, F., Carone, R., Zanollo, A. et al: Quality of life assessment in patients who undergo sacral neuromodulation implantation for urge incontinence: an additional tool for evaluating outcome. J Urol, 166: 2277, 2001 14. Caraballo, R., Bologna, R. A., Lukban, J. and Whitmore, K. E.: Sacral nerve stimulation as a treatment for urge incontinence and associated pelvic floor disorders at a pelvic floor center: a follow-up study. Urology, suppl., 57: 121, 2001 15. Das, A. K., Siegel, S., Rivas, D. A. and Schmidt, R. A.: Upper buttock placement of sacral neurostimulator results in decreased adverse events and reoperation rates. Presented at International Continence Society 32nd Annual Meeting, Heidelberg, Germany, August 28-30, 2002 16. Dijkema, H., Ijzerman, W., Mijs, P. T., Weil, E. H. J. and Janknegt, R. A.: Electrical modulation of sacral nerves for treatment of urge-incontinence. J Urol, suppl., 151: 512A, abstract 1137, 1994 17. Edlund, C., Hellstrom, M., Peeker, R. and Fall, M.: First Scandinavian experience of electrical sacral nerve stimulation in the treatment of the overactive bladder. Scand J Urol Nephrol, 34: 366, 2000 18. Everaert, K., De Ridder, D., Baert, L., Oosterlinck, W. and Wyndaele, J. J.: Patient satisfaction and complications following sacral nerve stimulation for urinary retention, urge incontinence and perineal pain: a multicenter evaluation. Int Urogynecol J Pelvic Floor Dysfunct, 11: 231, 2000 19. Groenendijk, P. M., Lycklama a Nyeholt, A. A. B. and van den Hombergh, U.: Urodynamic evaluation of sacral nerve stimulation treatment for patients with urge incontinence. Presented at International Continence Society 32nd Annual Meeting, Heidelberg, Germany, August 28-30, 2002 20. Grunewald, V. and Jonas, U.: Neurostimulation for lower urinary tract voiding problems. Curr Urol Rep, 1: 199, 2000 21. Hedlund, H., Schultz, A., Talseth, T., Tonseth, K. and van der Hagen, A.: Sacral neuromodulation in Norway: clinical experience of the first three years. Scand J Urol Nephrol Suppl, 210: 87, 2002 22. Heesakkers, J., Bemelmans, B. L., Van Kerrebroeck, E. V. and Debruyne, F. M.: Long term effects of Interstim in patients suffering form urinary incontinence, urgency/frequency syndrome and urinary retention: a prospective study. Eur Urol Suppl, 2: 143, 2003 23. Hohenfellner, M., Schultz-Lampel, D., Dahms, S., Matzel, K. and Thuroff, J. W.: Bilateral chronic sacral neuromodulation for treatment for lower urinary tract dysfunction. J Urol, 160: 821, 1998 24. Janknegt, R. A., Weil, E. H. and Eerdmans, P. H.: Improving neuromodulation technique for refractory voiding dysfunctions: two-stage implant. Urology, 49: 358, 1997 25. Janknegt, R. A., Hassouna, M. M., Siegel, S. W., Schmidt, R. A., Gajewski, J. B., Rivas, D. A. et al: Long-term effectiveness of sacral nerve stimulation for refractory urge incontinence. Eur Urol, 39: 101, 2001
841
26. Koldewijn, E., Meuleman, E., Bemelmans, B., van Kerrebroeck, P. and Debruyne, F.: Neuromodulation effective in voiding dysfunction despite high reoperation rate. J Urol, suppl., 161: 255, abstract 984, 1999 27. Peters, K. M., Carey, J. M. and Konstandt, D. B.: Does technique used for sacral nerve modulation impact outcomes? J Urol, suppl., 167: 249, abstract 978, 2002 28. Ruffion, A., N’Goi, C., Dembele, D., Morel-Journel, N., Azam, P. and Leriche, A.: Sacral root neuromodulation: prospective evaluation in 166 cases. Eur Urol Suppl, 2: 143, 2003 29. Ruiz-Cerda, J. L., Arlandis, S., Gonzalez-Chamorro, F., Fernandez, E., Jimenez, M. A., Castro, D. et al: Spanish experience in sacral nerve stimulation: case register of the Spanish Sacral Neuromodulation Group (GENS). Eur Urol Suppl, 2: 142, 2003 30. Scheepens, W. A., Weil, E. H., van Koeveringe, G. A., Rohrmann, D., Hedlund, H. E., Schurch, B. et al: Buttock placement of the implantable pulse generator: a new implantation technique for sacral neuromodulation—a multicenter study. Eur Urol, 40: 434, 2001 31. Scheepens, W. A., van Koeveringe, G. A., De Bie, R. A., Weil, E. H. and van Kerrebroeck, P. E.: Long-term efficacy and safety results of the two-stage implantation technique in sacral neuromodulation. BJU Int, 90: 840, 2002 32. Shaker, H. S. and Hassouna, M.: Sacral nerve root neuromodulation: an effective treatment for refractory urge incontinence. J Urol, 159: 1516, 1998 33. Siegel, S. W., Catanzaro, F., Dijkema, H. E., Elhilali, M. M., Fowler, C. J., Gajewski, J. B. et al: Long-term results of a multicenter study on sacral nerve stimulation for treatment of urinary urge incontinence, urgency-frequency, and retention. Urology, suppl., 56: 87, 2000 34. Spinelli, M., Gerber, M., Arduini, A. and Giardiello, G.: Improving neuromodulation technique: preliminary results of minimally invasive implant with tined lead. Neurourol Urodyn, 21: 388, 2002 35. Spinelli, M., Giardiello, G., Arduini, A. and van den Hambergh, U.: New percutaneous technique of sacral nerve stimulation has high initial success rate: preliminary results. Eur Urol, 43: 70, 2003 36. Thon, W. F.: Functional electrical neuromodulation of voiding dysfunctions using foramen electrodes: results of the European Study Group. Neurourol Urodyn, 11: 325, 1992 37. Weil, E. H., Ruiz-Cerda, J. L., Eerdmans, P. H., Janknegt, R. A. and Van Kerrebroeck, P. E.: Clinical results of sacral neuromodulation for chronic voiding dysfunction using unilateral sacral foramen electrodes. World J Urol, 16: 313, 1998 38. Brown, J. S., McNaughton, K. S., Wyman, J. F., Burgio, K. L., Harkaway, R. Bergner, D. et al: Measurement characteristics of a voiding diary for use by men and women with overactive bladder. Urology 61: 802, 2003 39. Groutz, A., Blaivas, J. G., Chaikin, D. C., Resnick, N. M., Engleman, K., Anzalone, D. et al: Noninvasive outcome measures of urinary incontinence and lower urinary tract symptoms: a multicenter study of micturition diary and pad tests. J Urol, 164: 698, 2000
Purpose: We systematically reviewed the evidence on the efficacy and safety of sacral nerve stimulation (SNS) for severe urge incontinence. Materials and Methods: A systematic review was performed of primary studies of SNS for urge incontinence published in English between 1966 and May 2003, and identified in major electronic databases. Two reviewers independently selected studies, assessed their methodological quality and extracted data. Results: Four randomized controlled trials and 30 case series were identified. Evidence from the randomized controlled trials, involving approximately 120 patients, showed that about 80% achieved continence or greater than 50% improvement in their main incontinence symptoms after SNS compared with about 3% of controls receiving conservative treatments while waiting for an implant. While case series were larger, they were methodologically less reliable. However, they showed similar results with 67% of patients becoming dry or achieving a greater than 50% improvement in symptoms after implantation. Incontinence episodes, leakage severity, voiding frequency and pad use were significantly lower after implantation. Benefits were reported to persist 3 to 5 years after implantation. Adverse events were documented in 27 studies. Overall the reoperation rate in implanted cases was 33%. The most common reason for surgical revision was relocation of the generator because of pain and infection. Common complications were pain at the implant or lead site in 25% of patients, lead related problems such as lead migration in 16%, replacement and repositioning of the implanted pulse generator in 15%, wound problems in 7%, adverse effects on bowel function in 6%, infection in 5% and generator problems in 5%. Permanent removal of the electrodes was reported in 9% of patients. Technical changes with time have been associated with decreased complication rates. Conclusions: There is evidence indicating that SNS is effective for decreasing symptoms in patients with urge incontinence. Adverse events occurred in about half of the implanted cases and surgical revision was performed in 33%. No major irreversible complications were reported in the studies reviewed. Further research is required on the long-term effects of and quality of life in patients with SNS. Key Words: bladder, urinary incontinence, prostheses and implants, electric stimulation, equipment safety
humans since the early 1980s. In 1986 Tanagho and Schmidt reported results in 10 patients who had electrodes implanted on the sacral roots for neuropathic voiding dysfunction.1 Since then, in excess of 8,000 implant procedures have been performed to treat a number of voiding problems refractory to standard conservative treatment.2 The use of SNS is based on the knowledge that electrical stimulation of the pudendal nerves can modulate neural reflexes that influence bladder and pelvic floor behavior. However, the exact physiological mechanism of action by which electrical nerve stimulation works is not yet fully understood.3 In practical terms SNS therapy involves the use of mild electrical pulses to stimulate the sacral nerves. Electrodes are placed next to a sacral nerve, usually S3, by inserting the electrode leads percutaneously into the corresponding foramen of the sacrum with the patient under local anesthesia. Adequate electrode placement is confirmed by obtaining appropriate motor and sensory responses. After this phase of peripheral nerve evaluation patients enter the so-called subchronic phase, which lasts 3 to 7 days and is usually considered successful when there is at least 50% improvement in the main incontinence symptoms. Patients with successful
rge urinary incontinence is one of the most common forms of urinary incontinence. It is typically managed conservatively by behavioral techniques (eg bladder training), physical therapies (eg electrical stimulation using vaginal or anal electrodes) or pharmacotherapy (antimuscarinic and anticholinergic drugs). When these approaches are unsuccessful, more invasive and irreversible surgical procedures are considered, including bladder reconstruction, for example augmentation cystoplasty, and urinary diversion. Implanted SNS has been proposed as an option for severe urge urinary incontinence. SNS has been investigated in
U
Submitted for publication March 16, 2005. The Health Services Research Unit is supported by a core grant from the Chief Scientist Office of the Scottish Executive Health Department and grant support from the United Kingdom National Institute for Clinical Excellence. The views expressed are those of the authors and not necessarily shared by the funders. * Correspondence: Department of Clinical Neurosciences, University of Edinburgh, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, United Kingdom (telephone: 0044 131 5372955; FAX: 0044 131 332 5150).
0022-5347/06/1753-0835/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION
835
Vol. 175, 835-841, March 2006 Printed in U.S.A. DOI:10.1016/S0022-5347(05)00326-5
836
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE
treatment receive a permanent implant, which consists of electrodes inserted subcutaneously and attached to an implantable pulse generator. The procedure is reversible and the implanted pulse generator can be removed at any time. Currently the SNS implanted device is exclusively produced under the name InterStim® therapy. InterStim® therapy was initially marketed for urge incontinence in Europe, Canada and Australia in 1994, and it subsequently received Food and Drug Administration approval in the United States in 1997. We systematically assessed all available evidence on the efficacy and safety of SNS for urinary urge incontinence.
MATERIALS AND METHODS Search strategy. We identified published and unpublished studies evaluating the efficacy and safety of SNS for urinary urge incontinence by searching major electronic databases (see Appendix). In addition, we perused the reference lists of all included studies and contacted experts in the field to inquire about further potentially eligible references. Selected websites were also searched for eligible, evidence based reports. Details of the search strategy and list of selected websites are available on request. Study selection and data extraction. To be included studies had to assess the efficacy and/or safety of implanted SNS in patients with urge urinary incontinence. We excluded studies of transcutaneous electrical nerve stimulation, magnetic sacral nerve stimulation and sacral anterior root stimulation (the Brindley technique). Our primary outcomes of interest for efficacy were the cure/improvement rate, number of leakage episodes daily, number of pads used daily, leakage severity and quality of life. Efficacy measures were derived from patient voiding diaries, physiological measurements, and published questionnaires and checklists. The definitions of cure and improvement were those provided in the reports of the primary studies. Cure was usually defined as no incontinence or greater than 90% clinical improvement, while improvement was defined as a 50% or greater decrease in main incontinence symptoms. The frequency and type of any reported adverse events were tabulated to assess SNS safety. Two of us (MB and AM) assessed each study for inclusion and performed data extraction independently using a specifically designed data extraction form. In case of multiple reports of a study we selected the most up-to-date and complete report. NonEnglish language papers were noted but not included. Any differences of opinion between reviewers about study selection and data extraction were resolved by discussion or arbitration. From each included study we extracted data on study design and quality, participant characteristics, intervention characteristics, and efficacy and safety results. Criteria for assessing the methodological quality of full text trials were the randomization method, allocation concealment, blinding procedures, group comparability at baseline and the number of withdrawals/dropouts. Quality assessment of full text case series included sample selection and representativeness, description of inclusion/exclusion criteria, identification of important prognostic factors, surgeon experience, choice and validity of outcome measures, number lost to followup and result completeness. We were not
able to assess the methodological quality of conference proceedings due to the limited information available.
RESULTS Search results. The literature search identified 1,562 reports, of which 34 English language reports met inclusion criteria, including 21 full text articles and 13 abstracts. Four of the 34 included studies were RCTs4 –7 and 30 were case series.8 –37 Overall they enrolled 3,146 patients. A total of 26 studies also included patients with urinary retention and/or with urgency/ frequency. Across the studies mean or median patient age was between 34 and 69 years. Overall 2,067 patients were reported to have undergone peripheral nerve evaluation testing and 1,827 received implanted SNS. Sample size was 9 to 581 patients and the proportion of women was 50% to 100% with 1 study including only women.4 Average followup was between 5.330 and 47.110 months (range up to 96).10 Table 1 lists details of the included studies. Efficacy of SNS in RCTs. Table 2 lists cure and improvement rates at 6 months in patients with urge incontinence randomized to a stimulation or a delayed group. Overall 50% of patients in the stimulation group achieved complete continence or greater than 90% improvement in the main incontinence symptoms compared with 1.6% of patients in the delay group.5,7 A 50% improvement in main incontinence symptoms was observed in approximately 37% and 3% of patients in the stimulation and delay groups, respectively.5–7 In addition, the trials of Weil7 and Schmidt5 et al provided further clinical results. However, most outcomes were presented as within-group comparisons (baseline results vs post-intervention results in each group) instead of betweengroup comparisons (stimulation vs control group). Table 3 shows that the number of leakage episodes daily and the number of pads used daily were significantly lower 6 months after implantation in the stimulation group compared with baseline. The mean severity of leakage was also lower at followup, although this was only significant in 1 trial.5 In contrast, patients in the delay group showed no significant improvement or worsening of incontinence symptoms. Weil et al also observed that mean bladder capacity assessed by cystometry significantly increased at 6 months compared with baseline in the stimulation group.7 Changes in urodynamic parameters were not reported in the delay group. Weil et al also noted significantly lower mean leakage episodes (p ⬍0.0005), pad use (p ⬍0.0005) and leakage severity (p ⫽ 0.047) in the implanted group compared to the control group at 6 months. The randomized, controlled trials of Weil7 and Schmidt5 et al also assessed the impact of SNS on patient quality of life using the SF-36, short form. Weil et al found a significant difference between groups in only the emotional role score in the physical health component of the questionnaire (p ⫽ 0.037).7 Similarly Schmidt et al observed a significant between-group difference 6 months after implantation in the physical health component of the questionnaire (p ⫽ 0.0008) but not in the mental health component.5
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE
837
TABLE 1. Included studies References Full text studies: Aboseif et al8 Amundsen and Webster9 Bosch and Graen10 Braun et al11 Cappellano et al13 Cappellano et al12 Edlund et al17 Everaert et al18 Grünewald and Jonas20 Hedlund et al21 Hohentellner et al23 Janknegt et al25 Janknegt et al24 Scheepens et al31 Scheepens et al30 Schmidt et al5 Shaker and Hassouna32 Siegel et al33 Spinelli et al35 Weil et al7 Weil et al37 Totals Abstracts: Caraballo et al14 Das et al15 Dijkema et al16 Everaert et al4 Groenendijk et al19 Heesakkers et al22 Koldewijn et al26 Peters et al27 Ruffion et al28 Ruiz-Cerdá et al29 Spinelli et al34 Thon36 Weil et al6 Totals
Type Study
No. Enrolled (all diagnoses)
Rec No. Peripheral Nerve Evaluation
No. Implant
Mos Followup (range)
47 69 46.2 49 51.1 47 59.8 43 49 54 43.4 Not reported 46 53 51 46.6 42.3 43 43 43 36
160* 25 85 9* 113* 47* 30* 53* 184* 53 11* 96 10* 15* 39* 155 18 581* 32* 123* 36* 1,875*
160* 25 85 Not reported Not reported 47* 30* 177* 184* 53 Not reported Not reported 10* 15* Not reported 155 Not reported 581* 13* 123* Not reported 1,658*
64* 12 45 9* 113* 10 9* 53* 55* 14 10* 96 8* 15* 39* 34 18 219* 22* 44 36* 925*
Mean 24 (6–36) Mean 7.8 (1–16) Mean 47.1 (6–96) Mean 12.5 (7–18) 18 Mean 23.1 (3–47) Mean 19.9 (8–39) Mean 24 (13–39) Mean 44.3 (1–89) Mean 18 (9–32) Mean 13 (9–28) Mean 30.8 (12–60) Mean 16 (4–36) Mean 59 (30–90) Mean 5.3 (1–10) Mean 14.7 (0.9–39.7) Mean 18.8 (3–83) (18–36) Mean 11a (2–25) Median 18 (6–36) Mean 37.8 (12–60)
60.6 47 reported 48 reported reported 40 reported 48.8 47 34 reported reported
17* 256* 25 22* 111 259* 40* 30* 166* 204* 9* 114* 18* 1,271*
Not Not Not Not Not Not Not
17* 256* 25 22* 111 259* 40* 14* 33* 69* 6* 41* 9* 902*
Mean 13.4 (3–22) Mean 26 (15–46) 6 or Greater 12 6 Greater than 12 Mean 29 (5–46) Mean 37 (3–87) Mean 6.8 (2–30) Not reported Mean 4.2 (1–12) 6
Mean Pt Age
Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case RCT Case Case Case RCT Case
series series series series series series series series series series series series series series series
Case Case Case RCT Case Case Case Case Case Case Case Case RCT
series series series
series series series series
series series series series series series series series
Not Not Not Not
Not Not
reported reported reported reported reported reported reported 30* 166* 204* 9* Not reported Not reported 409*
* Including patients with urinary retention and/or urgency-frequency.
Efficacy of SNS in case series. Of the 30 identified case series 22 provided suitable data on the efficacy of SNS in patients with urge urinary incontinence. Table 4 shows the cure, improvement and failure rates reported in 17 case series at last followup. Followup varied among the studies. The cure rate was 7% to 64% and 139 of 361 patients (39%) were reported to be cured. A 50% or greater improvement in incontinence symptoms was observed in 199 of 501 patients (40%). Overall 338 of 501 patients (67%) had a decrease in incontinence symptoms after SNS. In contrast, 163 of 501 cases (32%) did not show a 50% or greater improvement in incontinence symptoms after implantation and they were
TABLE 2. Success rates at 6 months in randomized, controlled trials No./Total No. (%) References Stimulation group: Weil et al7 Schmidt et al5 Weil et al6 Delay group: Weil et al7 Schmidt et al5 Weil et al6
Cured
Improved
9/16 (56) 16/34 (47) Not reported
Not reported (29) 10/34 (29) 5/5 (100)
1/22 (5) 0/42 Not reported
0/22 2/42 (5) Not reported
Cured— greater than 90% and improved— greater than 50% symptom decrease.
classified as failures. Other outcome measures, such as leakage episodes, pad use, leakage severity, voiding frequency and urodynamic parameters, were not recorded consistently. A total of 14 case series indicated the change in the average number of leakage episodes at followup compared to baseline.9⫺13,16,17,22,25,29,31–33,37 Overall leakage frequency was 4.5 to 11.6 episodes daily at baseline and 0.8 to 5.0 at last followup after implantation (decreased by 53% to 92%). The change was reported to be statistically significant in 11 of the 14 studies (p ⬍0.05). A total of 14 studies compared the number of pads used daily at 6 months or last followup after implantation with the number of pads used at baseline.9⫺12,16,17,21–25,31,33,37 The average number of pads decreased from 3.0 to 8.3 to 0.4 to 3.4 (decreased by 49% to 94%) and the change was statistically significant in 10 studies (p ⬍0.05). The severity of incontinence episodes was assessed on a scale of 1 to 3 as 1—mild, 2—moderate and 3—severe in 4 studies.17,25,31,32 Leakage severity was decreased from an average of 1.4 to 2.0 at baseline to 0.8 to 1.6 at 18 months or last followup (decreased by 16% to 40%). The decrease was statistically significant in 3 of the 4 studies (p ⬍0.05). Voiding frequency was assessed in 8 studies.9,10,21,23,25,31,32,37 The mean number of voids daily decreased from 10.0 to 15.0 at baseline to 7.0 to 9.2 at last followup after implantation (decreased by 30% to 50%). The change was statistically significant in 6 studies (p ⬍0.05). Urodynamic measurements were provided in 9 studies at
838
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE TABLE 3. Leakage episodes, pad usage, leakage severity and bladder capacity in patients with urge incontinence in randomized, controlled trials Mean Stimulation Group ⫾ SD References
Leakage episodes/day: Weil et al7 Schmidt et al5 Pad use/day: Weil et al7 Schmidt et al5 Leakage severity:* Weil et al7 Schmidt et al5 ml bladder capacity Weil et al7
Mean Delay Group ⫾ SD
No. Pts
Baseline
6 Mos
p Value
No. Pts
Baseline
6 Mos
p Value
21 34
13.5 ⫾ 7.5 9.7 ⫾ 6.3
1.4 ⫾ 3.3 2.6 ⫾ 5.1
⬍0.0005 ⬍0.0001
22 42
13.5 ⫾ 7.8 9.3 ⫾ 4.8
11.2 ⫾ 5.6 11.3 ⫾ 5.9
Not significant 0.002
21 34
8.7 ⫾ 6.8 6.2 ⫾ 5.0
0.7 ⫾ 1.3 1.1 ⫾ 2.0
⬍0.0005 ⬍0.0001
22 42
8.7 ⫾ 7.1 5.0 ⫾ 3.7
6.8 ⫾ 4.0 6.3 ⫾ 3.6
Not significant 0.003
21 34
2.1 ⫾ 0.6 2.0 ⫾ 0.7
1.6 ⫾ 1.7 0.8 ⫾ 0.9
Not significant ⬍0.0001
22 42
2.1 ⫾ 0.6 1.8 ⫾ 0.6
2.1 ⫾ 0.6 2.0 ⫾ 0.6
Not significant 0.006
21
266 ⫾ 112
370 ⫾ 91
0.013
23
Not reported
Not reported
-
* Assessed on a scale of 0—dry, 1—loss of a few drops of urine, 2—loss of 1 to 2 tablespoons of urine and 3—complete wetting/soaked pad or outer clothing.
baseline and at last followup.10,16,20,21,23,25,31,32,37 Average total voided volume increased from 99 to 195 ml per void to 176 to 402 ml per void (34% to 288% change). The increase was significant in 8 of the 9 included studies (p ⬍0.05). Bladder capacity assessed by cystometry was considered in 9 studies.11,12,16,20,21,23,32,36,37 Total bladder capacity increased from 122 to 400 ml at baseline to 273 to 596 ml at 6 months or last followup (15% to 197% change). The change from baseline was statistically significant in 6 studies. Three case series assessed the impact of SNS on patient quality of life.9,13,32 Capellano et al reported results in 47 patients assessed using a 22 item, domain specific questionnaire developed to detect modifications in self-perceived incontinence severity.13 The questionnaire score was calculated on a scale 0 —poor self-perceived quality of life to 100 —incontinence did not negatively impact quality of life. The average quality of life index score was significantly higher after implantation (83.8 vs 34.4, p ⬍0.01). Amundsen and Webster observed a significantly better score on the Incontinence Impact Questionnaire at last followup compared to baseline (250 at baseline to 62 at last followup, p ⫽ 0.03).9 Shaker and Hassouna used the SF-36 and Beck Depression Inventory to assess the impact of SNS on quality of life in 18 patients with refractory urge incontinence.32 An improvement of 10% to 40% was detected in the Beck Depression
Inventory. No significant differences were observed in the scores of any SF-36 subscales except change of health perception, which was reported to be significantly higher at 6 months compared to baseline (significance not shown). Safety of SNS. Adverse events were documented in a total of 993 patients in 24 studies. In most studies the safety profile of SNS was based on pooling data from all patients under investigation, in some studies including patients in urinary retention, and with urgency and frequency, with safety data not provided separately for each clinical diagnosis. Table 5 lists adverse event rates. Of 855 patients 282 (33%) underwent surgical revision of the SNS implant. The most common reasons for reoperation were relocation of the neurostimulator because of pain at the implant site, revision of the lead system for suspected or detected lead migration and infection. Infection was usually managed by antibiotics but deep infection in some patients required explantation of the pulse generator. Common complications were pain at the implant site in 25% of cases, lead migration in 16%, wound problems in 7%, an adverse effect on bowel function in 6%, infection in 5% and generator problems in 5%. In 42 of 279 patients (15%) the implanted pulse generator was replaced or relocated and 44 of 514 (9%) required permanent explan-
TABLE 4. Cure and improvement rates at last followup in case series References 23
Ruiz-Cerda et al Heesakkers et al22 Amundsen and Webster9 Everaert et al4 Groenendijk et al19 Hedlund et al21 Spinelli et al34 Caraballo et al14 Janknegt et al25 Bosch and Groen10 Grunewald and Jonar20 Siegel et al33 Koldewijn et al26 Hohenfellner et al23 Shaker and Hassouna32 Weil et al37 Dijkema et al16 Total
Abstract Only
Followup (mos)
No. Cured/Total No. (%)
No. Improved/Total No. (%)
No. Failed/Total No. (%)
Yes Yes No Yes Yes No Yes Yes No No No No Yes No No No Yes
Mean 6.8 60 Mean 7.8 12 6 18 Not reported 13.4 Mean 30.8 Mean 47 6 36 Mean 29 Mean 13 Mean 18.8 Mean 37.8 17
14/25 (55) Not reported 2/12 (17) Not reported Not reported 8/14 (57) Not reported 1/15 (7) 25/96 (26) 18/45 (40) 6/26 (23) 19/41 (46) 18/28 (64) Not reported 8/18 (44) 14/24 (58) 6/17 (35) 139/361 (39)
2/25 (8) 27/43 (63) 10/12 (83) 4/5 (80) 55/84 (65) 5/14 (36) 3/3 (100) 11/15 (73) 35/96 (36) 9/45 (20) 13/26 (50) 5/41 (12) 3/28 (11) 5/5 (100) 4/18 (22) 3/24 (13) 5/17 (30) 199/501 (40)
9/25 (36) 16/43 (37) 0/43 1/5 (20) 29/84 (35) 1/14 (7) 0/3 3/15 (20) 36/96 (38) 18/45 (40) 7/26 (27) 17/41 (41) 7/28 (25) 0/5 6/18 (33) 7/24 (29) 6/17 (35) 163/501 (32)
Cured— greater than 90%, improved— greater than 50% and failed—less than 50% symptom decrease.
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE TABLE 5. Adverse event rates Adverse Event
No. Events/No. Pts (%)
Reoperation Implantable pulse generator replaced/relocated Permanent explants Generator problems Electrode ⫹ lead problems Pain Infection Wound problems Adverse bowel function
282/855 (33) 42/279 (15) 44/514 (9) 18/399 (5) 129/785 (16) 162/653 (25) 35/727 (5) 19/273 (7) 20/353 (6)
tation of the pulse generator. No major neurological complications were documented except for a suspected case of nerve injury33 and a case of generalized fasciculation in which the etiology could not be established.18 DISCUSSION SNS for urge urinary incontinence. We systematically reviewed the available evidence on SNS use for urge urinary incontinence. We chose not to consider Brindley sacral anterior root stimulation because the characteristics of this implanted procedure and the categories of patients in whom it is suitable differ from those of SNS. We were not able to combine results from the identified randomized trials in a meta-analysis because of the limited amount of available data. Efficacy data from randomized controlled trials showed that about 80% of patients who received SNS became dry or showed improvement in the main incontinence symptoms. This compared with 3% in the control groups in randomized studies. Fewer episodes of leakage daily, fewer pads used daily and fewer voids daily were usually reported after implantation as well as a small change in volume per void and bladder capacity. On the whole the results of case series were consistent with those of randomized, controlled trials. The overall surgical revision rate in implanted cases was 33%. Pain, lead related complications and pulse generator replacement or relocation appeared to be the most frequently observed adverse events, followed by removal of the pulse generator, wound problems, bowel problems and infection. There were no reports of long lasting neurological adverse events. It is worth noting that safety data were often only reported on a whole population entering a study regardless of the clinical conditions of individuals or patient subgroups. However, studies that did not differentiate patients with urge incontinence symptoms from patients with other clinical diagnoses but reported safety data were deemed suitable for inclusion in this review. The median or mean age of patients considered in the studies included in this review was between 34 and 69 years. Consequently it is uncertain whether the results can apply to an older population. It is noteworthy that the modification of bowel function or adverse effects on bowel function were side effects more often reported in studies that included patients older than 75 years. Limitations and uncertainties. SNS is intended for patients with severe incontinence symptoms refractory to conventional treatments. However, there was great discrepancy in the range of treatments that patients had
839
received before implantation and the severity of incontinence was often not described. In most studies patients had previously been treated with pharmacological therapy and/or surgical operations, while some had received only conservative, nonsurgical treatments, such as behavioral therapy. In all included studies the frequency of incontinent episodes was assessed by voiding/micturition diaries. Although obtaining self-reported information is sometimes problematic, a voiding diary completed for 7 days is considered a reliable and valid method for documenting changes in symptoms in individuals with urge incontinence.38,39 However, the definition of improvement as a greater than 50% decrease in incontinence symptoms was an arbitrary cutoff, of which the clinical importance was not always evident. A 50% or greater decrease in incontinent symptoms is a broad definition of clinical improvement and some patients may just attain the threshold for being classified as improved, ie a 50% decrease, while others may show a much more substantial rate of improvement, eg a 70% to 80% decrease. The clinical impact of potential different levels of improvement was not addressed or discussed in the studies included in this review. The best evidence in this review should come from the 2 full text reports of randomized trials. Nevertheless, neither of these 2 trials was of high quality. In neither was it stated how patient allocation to treatment groups was concealed or whether the outcome assessor was blinded to treatment allocation. Moreover, in each trial intent to treat analysis was not performed. Most outcomes in randomized, controlled trials were also presented as within-group rather than as between-group comparisons. However, the direction and magnitude of differential effects was consistent across trials. Most remaining evidence consisted of case series, which are known to be more prone to biases than randomized, controlled trials. In particular selection bias (patients treated and cases reported chosen by investigators), findings not adjusted for confounding factors (eg age, symptom duration and previous pelvic surgery), the likelihood of some spontaneous improvement because patients were treated at their worst and dropout/withdrawal rates may affect the reliability and magnitude of the treatment effect in case series. The majority of case series were also small, failed to identify important patient prognostic factors and did not provide information on nonresponders and dropouts. The timing of followup varied across case series and was generally of short duration. Furthermore, it was unclear in most studies whether data had been collected prospectively. However, the direction and size of pretreatment and posttreatment differences were consistent across studies and with those of the randomized trials. Patient quality of life was rarely measured in the studies included in this review. Only 3 randomized trials and 3 case series showed some quality of life measures but the results were not consistent across studies. Only the most recently published studies tended to quantify adverse events. Earlier studies discussed them in a more narrative way or did not report them. Most complications observed in included studies were technical problems related to device implantation. The clinical experience and skill of the clinician performing the procedure could have a major impact on the success of the procedure and the subsequent incidence of technical complications. Neither ran-
840
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE
domized trials nor case series provided information about the level of expertise of the clinician(s) performing the procedure. Furthermore, the backup facilities of the hospital/ clinic where the procedures were performed were never described. Evolution of the technique. SNS techniques have evolved with time and reported rates of adverse effects have decreased in parallel. The pulse generator is now positioned in the upper buttock region rather than in the abdominal wall, which has decreased episodes of pain. The test evaluation to select patients for a permanent implant can now be performed as needle test stimulation or as a staged implant using a surgically implantable lead as the test lead. Novel lead systems have been developed for the staged implant evaluation test to decrease infection and prevent lead migration. The feasibility of percutaneous lead placement under local anesthesia and with fascial fixation was recently evaluated in a series of 22 patients.35 The technique is considered to be less invasive, offers the possibility of testing the sensory response during implantation and allows implantation to be performed using local anesthesia. A similar method that has recently gained attention is a complete percutaneous implant in a 2-stage approach. A new, tined permanent lead is used for this technique and no incision or additional fascial fixation is required. The correct placement and identification of lead and electrode positions is confirmed by fluoroscopy. The main disadvantage of the latter 2 procedures is the need for an additional procedure for lead removal in patients who do not respond to stimulation. Currently to our knowledge there is no evidence about the longterm (ie 10-year) efficacy or safety of the procedure and it is likely that further revisions will be required to maintain clinical benefits with time.
CONCLUSIONS The results of randomized, controlled trials provide evidence of a benefit from SNS for decreasing incontinence episodes, pad use and voiding frequency, and in improving bladder capacity and voided volume. Evidence from case series is less reliable because of the risk of potential bias in this type of study design. However, their findings are broadly similar to those of randomized trials, decreasing the likelihood that observed differences were due to bias. The benefits of SNS were reported to persist at followup 3 to 5 years after implantation of the pulse generator. Although the sparse data available suggest improvement, the impact of SNS on the quality of life of patients remains to be established. Surgical revision was necessary in 33% of implanted cases. The most common reasons for reoperation were relocation of the implantable pulse generator because of pain, revision of the lead system and infection. Overall adverse events occurred in almost half of the tested patients. The most common complications were pain at the implant site, lead migration, relocation, replacement or permanent explantation of the implanted pulse generator, and wound problems. Further investigations are needed of the most recent refined SNS techniques, in particular long-term effects and quality of life.
APPENDIX Searched Databases MEDLINE (1966 to week 2, May 2003) MEDLINE Extra (May 29, 2003) EMBASE (1980 to week 21, 2003) CINAHL® (1985 to May 2003) BIOSIS® (1985 to May 2003) Science Citation Index (1981 to June 2003) Web of Science Proceedings (1990 to June 2003) Cochrane Controlled Trials Register (Cochrane Library, Issue 2, 2003) Cochrane Database of Systematic Reviews (Cochrane Library, Issue 2, 2003) Database of Abstracts of Reviews of Effectiveness (May 2003) Health Technology Assessment Database (May 2003) National Research Register (Issue 2, 2003) Clinical Trials (May 2003) Current Controlled Trials (May 2003) Research Findings Register (May 2003)
Abbreviations and Acronyms RCT ⫽ randomized controlled trial SF-36 ⫽ RAND 36-Item Health Survey SNS ⫽ sacral nerve stimulation
REFERENCES 1. Tanagho, E. A. and Schmidt, R. A.: Controlled voiding with neurostimulation. J Urol, suppl., 135: 262A, abstract 635, 1986 2. Abrams, P., Blaivas, J. G., Fowler, C. J., Fourcroy, J. L., Macdiarmid, S. A., Siegel, S. W. et al: The role of neuromodulation in the management of urinary urge incontinence. BJU Int, 91: 355, 2003 3. Turnberg, L., Brocklehurst, J., Fowler, C., Borzyskowski, M., Cardozo, L., Chellingsworth, M. et al: Incontinence Causes, Management and Provision of Services. London: Royal College of Physicians, 1995 4. Everaert, K., Kerckhaert, W. and Caluwaerts, H.: The staged implant does not increase subjective or objective improvement in overactive bladder symptoms in patients selected for sacral nerve stimulation. Neurourol Urodyn, 21: 402, 2002 5. Schmidt, R. A., Jonas, U., Oleson, K. A., Janknegt, R. A., Hassouna, M. M., Siegel, S. W. et al: Sacral nerve stimulation for treatment of refractory urinary urge incontinence. J Urol, 162: 352, 1999 6. Weil, E. H., Eerdmans, P. H. and Janknegt, R.: A new randomized study of neuromodulation versus conventional treatment for incontinence or dysfunctional voiding patterns a preliminary report. Neurourol Urodyn, 15: 284, 1996 7. Weil, E. H., Ruiz-Cerda, J. L., Eerdmans, P. H., Janknegt, R. A., Bemelmans, B. L. and van Kerrebroeck, P. E.: Sacral root neuromodulation in the treatment of refractory urinary urge incontinence: a prospective randomized clinical trial. Eur Urol, 37: 161, 2000 8. Aboseif, S., Tamaddon, K., Chalfin, S., Freedman, S. and Kaptein, J.: Sacral neuromodulation as an effective treatment for refractory pelvic floor dysfunction. Urology, 60: 52, 2002 9. Amundsen, C. L. and Webster, G. D.: Sacral neuromodulation in an older, urge-incontinent population. Am J Obstet Gynecol, 187: 1462, 2002 10. Bosch, J. L. H. E. and Groen, J.: Sacral nerve neuromodulation in the treatment of patients with refractory motor urge incontinence: long-term results of a prospective longitudinal study. J Urol, 163: 1219, 2000 11. Braun, P. M., Boschert, J., Bross, S., Scheepe, J. R., Alken, P. and Juenemann, P.: Tailored laminectomy: a new technique for neuromodulator implantation. J Urol, 162: 1607, 1999
SACRAL NERVE STIMULATION FOR URGE INCONTINENCE 12. Cappellano, F., Ciotti, M. G., Pizzoccaro, M., Catanzaro, M., Santambrogio, S. and Catanzaro, F.: Sacral root neuromodulation in the treatment of female urge and mixed urinary incontinence. Urogynaecol Int J, 12: 111, 1998 13. Cappellano, F., Bertapelle, P., Spinelli, M., Catanzaro, F., Carone, R., Zanollo, A. et al: Quality of life assessment in patients who undergo sacral neuromodulation implantation for urge incontinence: an additional tool for evaluating outcome. J Urol, 166: 2277, 2001 14. Caraballo, R., Bologna, R. A., Lukban, J. and Whitmore, K. E.: Sacral nerve stimulation as a treatment for urge incontinence and associated pelvic floor disorders at a pelvic floor center: a follow-up study. Urology, suppl., 57: 121, 2001 15. Das, A. K., Siegel, S., Rivas, D. A. and Schmidt, R. A.: Upper buttock placement of sacral neurostimulator results in decreased adverse events and reoperation rates. Presented at International Continence Society 32nd Annual Meeting, Heidelberg, Germany, August 28-30, 2002 16. Dijkema, H., Ijzerman, W., Mijs, P. T., Weil, E. H. J. and Janknegt, R. A.: Electrical modulation of sacral nerves for treatment of urge-incontinence. J Urol, suppl., 151: 512A, abstract 1137, 1994 17. Edlund, C., Hellstrom, M., Peeker, R. and Fall, M.: First Scandinavian experience of electrical sacral nerve stimulation in the treatment of the overactive bladder. Scand J Urol Nephrol, 34: 366, 2000 18. Everaert, K., De Ridder, D., Baert, L., Oosterlinck, W. and Wyndaele, J. J.: Patient satisfaction and complications following sacral nerve stimulation for urinary retention, urge incontinence and perineal pain: a multicenter evaluation. Int Urogynecol J Pelvic Floor Dysfunct, 11: 231, 2000 19. Groenendijk, P. M., Lycklama a Nyeholt, A. A. B. and van den Hombergh, U.: Urodynamic evaluation of sacral nerve stimulation treatment for patients with urge incontinence. Presented at International Continence Society 32nd Annual Meeting, Heidelberg, Germany, August 28-30, 2002 20. Grunewald, V. and Jonas, U.: Neurostimulation for lower urinary tract voiding problems. Curr Urol Rep, 1: 199, 2000 21. Hedlund, H., Schultz, A., Talseth, T., Tonseth, K. and van der Hagen, A.: Sacral neuromodulation in Norway: clinical experience of the first three years. Scand J Urol Nephrol Suppl, 210: 87, 2002 22. Heesakkers, J., Bemelmans, B. L., Van Kerrebroeck, E. V. and Debruyne, F. M.: Long term effects of Interstim in patients suffering form urinary incontinence, urgency/frequency syndrome and urinary retention: a prospective study. Eur Urol Suppl, 2: 143, 2003 23. Hohenfellner, M., Schultz-Lampel, D., Dahms, S., Matzel, K. and Thuroff, J. W.: Bilateral chronic sacral neuromodulation for treatment for lower urinary tract dysfunction. J Urol, 160: 821, 1998 24. Janknegt, R. A., Weil, E. H. and Eerdmans, P. H.: Improving neuromodulation technique for refractory voiding dysfunctions: two-stage implant. Urology, 49: 358, 1997 25. Janknegt, R. A., Hassouna, M. M., Siegel, S. W., Schmidt, R. A., Gajewski, J. B., Rivas, D. A. et al: Long-term effectiveness of sacral nerve stimulation for refractory urge incontinence. Eur Urol, 39: 101, 2001
841
26. Koldewijn, E., Meuleman, E., Bemelmans, B., van Kerrebroeck, P. and Debruyne, F.: Neuromodulation effective in voiding dysfunction despite high reoperation rate. J Urol, suppl., 161: 255, abstract 984, 1999 27. Peters, K. M., Carey, J. M. and Konstandt, D. B.: Does technique used for sacral nerve modulation impact outcomes? J Urol, suppl., 167: 249, abstract 978, 2002 28. Ruffion, A., N’Goi, C., Dembele, D., Morel-Journel, N., Azam, P. and Leriche, A.: Sacral root neuromodulation: prospective evaluation in 166 cases. Eur Urol Suppl, 2: 143, 2003 29. Ruiz-Cerda, J. L., Arlandis, S., Gonzalez-Chamorro, F., Fernandez, E., Jimenez, M. A., Castro, D. et al: Spanish experience in sacral nerve stimulation: case register of the Spanish Sacral Neuromodulation Group (GENS). Eur Urol Suppl, 2: 142, 2003 30. Scheepens, W. A., Weil, E. H., van Koeveringe, G. A., Rohrmann, D., Hedlund, H. E., Schurch, B. et al: Buttock placement of the implantable pulse generator: a new implantation technique for sacral neuromodulation—a multicenter study. Eur Urol, 40: 434, 2001 31. Scheepens, W. A., van Koeveringe, G. A., De Bie, R. A., Weil, E. H. and van Kerrebroeck, P. E.: Long-term efficacy and safety results of the two-stage implantation technique in sacral neuromodulation. BJU Int, 90: 840, 2002 32. Shaker, H. S. and Hassouna, M.: Sacral nerve root neuromodulation: an effective treatment for refractory urge incontinence. J Urol, 159: 1516, 1998 33. Siegel, S. W., Catanzaro, F., Dijkema, H. E., Elhilali, M. M., Fowler, C. J., Gajewski, J. B. et al: Long-term results of a multicenter study on sacral nerve stimulation for treatment of urinary urge incontinence, urgency-frequency, and retention. Urology, suppl., 56: 87, 2000 34. Spinelli, M., Gerber, M., Arduini, A. and Giardiello, G.: Improving neuromodulation technique: preliminary results of minimally invasive implant with tined lead. Neurourol Urodyn, 21: 388, 2002 35. Spinelli, M., Giardiello, G., Arduini, A. and van den Hambergh, U.: New percutaneous technique of sacral nerve stimulation has high initial success rate: preliminary results. Eur Urol, 43: 70, 2003 36. Thon, W. F.: Functional electrical neuromodulation of voiding dysfunctions using foramen electrodes: results of the European Study Group. Neurourol Urodyn, 11: 325, 1992 37. Weil, E. H., Ruiz-Cerda, J. L., Eerdmans, P. H., Janknegt, R. A. and Van Kerrebroeck, P. E.: Clinical results of sacral neuromodulation for chronic voiding dysfunction using unilateral sacral foramen electrodes. World J Urol, 16: 313, 1998 38. Brown, J. S., McNaughton, K. S., Wyman, J. F., Burgio, K. L., Harkaway, R. Bergner, D. et al: Measurement characteristics of a voiding diary for use by men and women with overactive bladder. Urology 61: 802, 2003 39. Groutz, A., Blaivas, J. G., Chaikin, D. C., Resnick, N. M., Engleman, K., Anzalone, D. et al: Noninvasive outcome measures of urinary incontinence and lower urinary tract symptoms: a multicenter study of micturition diary and pad tests. J Urol, 164: 698, 2000