To describe patient selection for sacral neuromodulation, also known as Interstim therapy, and the results of tined-lead implantation in the medium term.
To describe patient selection for sacral neuromodulation, also known as Interstim therapy, and the results of tined-lead implantation in the medium term.
In all, 49 patients, 39 with refractory overactive bladder symptoms and 10 with urinary retention, were implanted with the tined lead under local anaesthesia. The mean (sd) test period was 12.4 (5.8) days. Patients were implanted when they had a ≥ 50% improvement in voiding diary variables during the test period. The mean follow-up for implanted patients was 15.5 (7.9) months. Changes in voiding variables were compared using a t-test.
Ten patients had a one-stage and 39 a two-stage implant; of the latter group, 31 (80%) had a positive response and eight (21%) did not. In all, 31 patients were included in the follow-up. At the last follow-up, 28 (90%) patients had a >50% improvement in diary variables and three (10%) did not. In 21 patients with urgency symptoms the mean (sd) number of voids decreased from 11.7 (8.9)/day at baseline to 7.3 (3.4)/day (P = 0.1); the voided volume increased from 160.2 (70.7) mL to 231.1 (119.5) mL (P = 0.001); and the number of leakages decreased from 9.5 (8.7) to 3.3 (2.2)/day (P = 0.17). In the 10 patients with retention, the number of catheterizations decreased from 5.44 (1.6)/day with a volume of 297.6 (76.8) mL, to 1.2 (1.7)/day and 111.6 (158.1) mL; the mean number of voids increased from 3.7 (3.8)/day with a volume of 123.3 (141.7) mL, to 4.2 (2.4)/day and 248.3 (146.0)mL. There were no significant differences in the variables in the patients with retention. Seven patients had an adverse event. There was one incomplete electrode migration that was treated conservatively.
This new minimally invasive approach gives positive results in the medium term. Two-stage testing with the tined lead seems more reliable than the classic percutaneous nerve evaluation. The lead anchoring method seems sufficient for fixing the electrode in the medium term.
percutaneous nerve evaluation
implantable pulse generator
Since its introduction about a decade ago, sacral neuromodulation (SN) is applied in patients with refractory urge incontinence, urgency-frequency and unobstructive urinary retention [1–5]. Typically, potential candidates are tested with a temporary lead before the implantation of a definitive neuromodulation system is considered. Clinical experience with the test procedure for neuromodulation showed that the temporary electrode is prone to migration during the subchronic test phase. Consequently, and due to the probability of bacterial infections in temporary electrodes that are implanted for a prolonged period, the duration of the subchronic test phase is limited to a maximum of ≈ 1 week. Moreover, there are some implanted patients who experience inconsistencies between the results of the test period and the results of the definitive implant.
In 2003, Spinelli et al. [6,7] introduced a new, minimally invasive technique for implanting a definitive electrode. Using this electrode, the duration of the subchronic test phase can be prolonged, and if the test results are insufficient, the electrode can easily be removed. In this article we describe our clinical results with this procedure.
In all, 49 patients (nine men and 40 women, mean age 50.0 years, sd 13.2, range 18–73) with symptoms of overactive bladder (OAB) or urinary retention refractory to conservative therapy, had the new electrode implanted. All patients had a percutaneous nerve evaluation (PNE) with the conventional temporary electrode before the tined-lead implant. Under local anaesthesia, in a minimally invasive procedure, a lead (Model 3889, Medtronic, Minneapolis, MN, USA) was placed in the left or right S3 foramen . The lead is equipped with silicone tines that allow it to be fixed in the tissue above the sacrum. Local anaesthesia allowed both patients’ sensory and motor responses to be used as a guide. The depth of the electrode was controlled by lateral fluoroscopy.
The patients with a positive result from the classic PNE had a one-stage implant. In these patients, the electrode was tunnelled to an incision made at the ipsilateral gluteal area and connected to an extension cable. This extension cable is connected to an implantable pulse generator (IPG, model 3023, Medtronic) inserted into a s.c. pocket in the gluteal area. The one-stage procedure took a mean of 40 min.
The patients with inconclusive results of the classic PNE had a two-stage procedure. In these, the lead was tunnelled to a small ipsilateral gluteal incision. After connecting it to an extension cable, the lead was tunnelled to a contralateral exit point and connected to an external stimulator (model 3625, Medtronic). The subchronic phase of the two-stage test took a mean (sd, range) of 12.4 (5.8, 5–31) days. During this phase, patients were asked to keep voiding diaries to record changes in their urinary symptoms. These data were compared to baseline voiding diaries.
Patients with a ≥ 50% improvement in their relevant urinary symptoms were implanted with an IPG as a second stage. In these, the ipsilateral incision was re-opened and the extension cable was disconnected and removed. A short extension cable was connected to the lead and an IPG was implanted. When the subchronic phase was unsuccessful the lead and extension cable were disconnected and removed. The lead could easily be removed up to a month after implantation by applying moderate traction. The first stage of the two-stage procedure took a mean of 30 min, while the second stage took 15 min. All procedures were done under local anaesthesia on an outpatient basis. After implanting the IPG, patients were followed at regular intervals of 6 weeks, 6 months, and yearly thereafter.
Patients were told to contact the clinic when they had problems related to the implanted system. All patients were told to avoid stretching, bowing and bicycling for at least 6 weeks after implantation, after which they were encouraged to resume all activities that they did before the implant. Each patient with loss of efficacy during follow-up had a radiograph taken to check whether there was electrode migration. In June 2005, all implanted patients were asked to complete another set of voiding diaries; these were compared with the baseline diaries. The variables at baseline and at the last follow-up were compared using a t-test.
Of the 49 patients, 39 (80%) were implanted for OAB symptoms, and 10 (20%) had urinary retention. Ten patients had a positive classic PNE and were implanted in a one-stage procedure. The other 39 patients had an inconclusive PNE, and were re-tested in a two-stage procedure. Reasons for the inconclusive results were: discrepancy in recorded 24-h output between the baseline and test period (13 patients), suspected electrode migration (seven) and technical failure of the PNE (two). One patient with a technical failure had a different anatomy due to spina bifida; in another patient the external stimulator switched off by itself. The other 17 patients with an inconclusive PNE had subjective improvement, while analysis of the voiding diary showed a <50% improvement. In the 39 patients who had two-stage testing, 31 (80%) were eligible for the definitive implant and eight (21%) were not.
Of all 49 patients, 41 had a definitive implant either in one stage or after two-stage testing, and 31 of them were included in the further follow-up. Four of the 10 patients who were not included had a follow-up of <6 weeks. A fifth patient with a follow-up of 18.3 months had loss of efficacy after a vaginal wall correction. A sixth patient died 8.9 months after implantation from causes unrelated to neuromodulation, and the seventh patient was explanted for to psychiatric reasons after 13.3 months of follow-up, but had had good results. This patient had hallucinations after the implant and blamed adverse life-events on having the nerve stimulation system. Three patients were lost to follow-up and did not respond to repeated requests to complete an additional diary. These 10 patients were excluded from the analysis.
The mean (sd, range) follow-up of the 31 analysed patients was 15.5 (7.9, 3.2–32.4) months. At the last follow-up, 28 (90%) of the patients still had a >50% improvement in at least one of the relevant voiding diary variables and three (10%) patients had a <50% improvement. In all implanted patients, the voided volume was increased from 153.0 (86.9) to 234.5 (122.5) mL at the last follow-up (P < 0.001). Table 1 shows the overall success rates in each patient category.
|Patients, n (%)||Success||Failure||Total|
|OAB symptoms||19 (90)||2 (10)||21|
|Total||28 (90)||3 (10)||31|
In the patients with urge incontinence or urgency-frequency the mean number of voids at baseline was 11.7 (8.9)/day. At last follow-up this decreased to 7.3 (3.4)/day (P = 0.1). The voided volume increased from 160.2 (70.7) to 231.1 (119.5) mL at the last follow-up (P = 0.001).
The patients who had urge incontinence had a mean of 9.5 (8.7) incontinent episodes/day at baseline; at the last follow-up this decreased to 3.3 (2.2)/day (P = 0.17). In the 10 patients with retention, the mean number of catheterizations was 5.44 (1.6)/day and the catheterized volume was 297.6 (76.8) mL at baseline. At last the follow-up the mean number of catheterizations was 1.2 (1.7)/day and the catheterized volume was 111.6 (158.1) mL. Three of these 10 patients had complete retention at baseline. At the last follow-up, two of these three patients had a normal voiding pattern. Six of the patients with retention no longer catheterize, and the other four catheterize up to a maximum of twice a day. The mean number of voids at baseline was 3.7 (3.8)/day with a volume of 123.3 (141.7) mL. At the last follow-up the number of voids was 4.2 (2.4)/day with a volume of 248.3 (146.0) mL. There were no significant differences in the variables in patients with retention, when analysed by a t-test.
Of 31 patients, seven had a significant adverse event. Two patients had loss of efficacy after non-urological surgery; both patients had positive results after up to two re-programming sessions. Another patient with retention due to an incomplete spinal cord lesion still has no benefit from his system. Two patients complained of pain at the IPG implant site; one had a re-operation under local anaesthesia to reposition the IPG, and the other had a buttock implant of the IPG, while she was dependent on a wheelchair. In this patient the IPG was re-positioned from the gluteal region to the abdominal wall. Both patients are now free of pain and have positive results.
A patient with positive results had an open revision because of troublesome leg stimulation. Investigation showed that the tined lead was placed too deep and stimulated the S2 nerve that runs below the level of the S3 root.
During the follow-up there was one incomplete electrode migration. The patient presented at 5.3 months of follow-up with different sensations and decreased efficacy after a fall. A radiograph image of the sacrum showed that the electrode had migrated by 2–3 mm. The patient was successfully treated by re-programming the stimulator.
Successful neuromodulation largely depends on the careful selection of candidates. To date the only means of selecting patients likely to benefit was the classic PNE . However, the PNE has its limitations. The duration of the subchronic phase of the PNE is limited to a maximum of ≈ 1 week because of the chance of infection. Furthermore. there is a chance of electrode migration. Edlund et al.  reported that eight of 20 patients had a loss of efficacy during the subchronic phase of the PNE; this was probably due to electrode migration, but not in all patients with an insufficient result, as electrode migration could be shown by a sacral radiograph.
In 1997, the two-stage approach was introduced [9,10]. Testing with a definitive electrode enabled longer testing periods and decreased the chance of a negative test due to electrode migration. Recently Everaert et al.  proved in a randomized trial that testing with the two-stage procedure gave more positive subjective results and a greater improvement in diary variables. As the tined lead can be implanted using a minimally invasive procedure under local anaesthesia, and can easily be removed, testing through the two-stage procedure is easier and applied more often.
The proportion of patients in the present study who had a positive two-stage procedure (80%) is higher than the published positive results with the classic PNE, of 30–62%[8,11–14]. This comparison is not entirely fair, because some nonresponders were filtered out by the classic PNE, but shows that ≈ 20% of the negative classic PNEs are false-negative, possibly because the testing period was too short to show clear results, or perhaps because of undetected electrode migration. In a recent study, Kessler et al.  reported a significant difference between 50% positive responders after 4–7 days vs 80% after ≥ 14 days. These results indicate that the duration of the test period is important in selecting candidates for SN. It would be interesting to compare the results of the PNE and the two-stage tined lead in a randomized prospective trial.
However, in our experience the PNE still has value for initial screening, as it is a cost-effective way of establishing whether a patient is a suitable candidate for SN. Until there are better predictors for the success of SN, the only tool for patient selection is a temporary test procedure. The PNE keeps the threshold for testing potential candidates low, while filtering unsuitable candidates at an early stage of the assessment. Examples of unsuitable candidates are patients who cannot cope with the testing equipment or the diaries, or those who do not tolerate SN.
Indications for a two-stage tined lead procedure are a <50% symptom reduction during the PNE and/or technical difficulties of the PNE, including suspected lead migration. The patients with urgency had significant improvements at the last follow-up vs baseline. Although patients with retention improved, the variables did not change significantly, but the present group of patients with retention is small and most patients had no complete retention at baseline. The differences in number of voids and voided volume are relatively small, accounting for high P values. However, the number of catheterizations decreased to a mean of 1/day, and the need to catheterize is bothersome for patients. Of the implanted patients, 90% maintained a favourable result during the follow-up. This suggests that, in addition to better selection, placing the definitive lead implant under local anaesthesia gives a better chance of lasting positive results than the open implant.
Minimally invasive lead placement is guided by both sensory and motor responses, rather than only motor responses in the open procedure. We suspect that the use of sensory responses in addition to motor responses provides better electrode placement and less chance of complications and therapy failure. This has yet to be confirmed in a study with more patients and a longer follow-up.
One patient had an incomplete electrode migration after a fall. We took sacral radiographs in each patient who had loss of efficacy. Despite this rigorous screening policy, there was one migration of 2–3 mm, that was treated conservatively. Kessler et al.  described a patient with bilateral electrode migration, who was successfully treated by re-programming the modulator.
In the present study, two patients needed a re-operation due to pain at the implant site. One patient uses a wheelchair, and the IPG implanted in the gluteal area caused discomfort due to the pressure of the wheelchair on the region of the implanted IPG. After repositioning the IPG in the abdominal wall, the patient is free of complaints. Therefore abdominal placement of the IPG should be considered in wheelchair-bound patients. In the other patient with pain at the position of the IPG, we found during re-operation a subcutaneous nerve running through the pocket wall. After re-positioning the IPG away from the nerve the patient was free of complaints.
In conclusion, the new minimally invasive approach for SN is easier to perform than the classic open method. It gives positive results in both the short- and medium-term. Two-stage testing with the tined lead appears to be a more reliable method than the classic PNE because of a prolonged testing period and perhaps because of the lower risk of electrode migration. Therefore testing with a tined lead is a useful alternative in patients with an inconclusive PNE result. As there was only one lead migration during the medium-term follow-up of up to 3 years, it is obvious that that the lead anchoring method is sufficient for fixing the electrode. However, long-term follow-up data on the performance and possible migration of the tined lead are not yet available.
P. van Kerrebroeck and E. Weil are study investigators for Medtronic; U. van den Hombergh is an employee of Medtronic Switzerland. Source of funding: WAMU Foundation, Maastricht, the Netherlands.