Introduction

Neurogenic patients often present concomitant pelvic functional disturbances and conservative treatments do not always guarantee symptom resolution.1 Recently it has been shown that incomplete spinal cord injury patients (SCIPs) make remarkable clinical improvements with sacral neuromodulation (SNM), either in neurogenic bowel symptoms (NBSs) or in neurogenic lower urinary tract symptoms (NLUTSs).2, 3Moreover, literature reports SNM's positive clinical effects on erectile function in SCIP as well.4

The aim of this study was to conduct medium and long follow-ups of SNM efficacy in SCIPs who had shown significant clinical improvement in more than one pelvic functional disturbance with a permanent SNM implant.

Patients and methods

This is a retrospective study in male patients with incomplete spinal cord lesions who underwent permanent SNM implantation (Medtronic, Inc., Minneapolis, MN, USA) for NBSs and/or NLUTSs refractory to conservative therapies such as anticholinergics for urge urinary incontinence. Patients with neurogenic chronic constipation did not respond to suppositories, whereas antidiarrheal medication failed for subjects with fecal incontinence.

Pre-first stage

The American Spinal Injury Association Impairment Scale was used to classify patients according to injury severity.5

During the 7-day assessment period, patients with NLUTSs suspended their drug therapy and completed a voiding diary. At the end of this phase, urodynamic investigations including uroflowmetry, filling cystometry, and detrusor pressure/flow studies were carried out.

During the 14-day assessment period, patients with NBSs suspended their NBS medication. They kept a bowel diary for 2 weeks and then underwent anorectal manometry and measurement of total gastrointestinal transit times.6

To determine erectile function, patients underwent a 14-day wash-out period and entered a 28-day assessment period during which time they attempted sexual intercourse without any treatment. Subsequently, the International Index of Erectile Function composed of 5 questions (IIEF5) was administered.7

All subjects then completed the Short Form 36 Health Survey questionnaire (SF-36) to determine the impact on health-related quality of life (QoL).8

Table 1 reports the inclusion and exclusion criteria for the first stage of SNM.

Table 1 Patient selection criteria for each pelvic function
Full size table

First stage

During the first stage, which lasted a minimum of 4 weeks (range 28–52 days), the permanent electrode was inserted percutaneously into the monolateral third sacral foramina and was then stimulated by using an external pulse generator (Medtronic Interstim model 3625).

To proceed to the second stage, a minimum clinical improvement of 50% in the following parameters during the first stage was mandatory: residual urine for neurogenic non-obstructive patients, urge urinary incontinence per day, number of fecal incontinence episodes per week, and, finally, the number of evacuations per week for subjects with constipation. The other clinical outcome measures for NBS and NLUTS are reported in Tables 2 and 3.

Table 2 Data on patients with neurogenic urinary tract symptoms
Full size table
Table 3 Data on patients with neurogenic bowel symptoms
Full size table

Second stage

In the second stage, a permanent implantable pulse generator was implanted in the patient's buttock. Follow-ups for patients with NLUTSs and NBSs were scheduled at 1, 3 and 6 months following the second stage, and then every 6 months. Before each follow-up, patients entered a 7- or 14-day assessment period for NLUTSs and NBSs, respectively, during which time they kept their diaries and abstained from symptom-related drug treatment. At the first and final visits, all patients with NLUTSs underwent urodynamics, while anorectal manometric studies were performed on subjects with NBSs. Urodynamics and anorectal manometry were executed every year on average.

The first evaluation for erectile function was done after 3 months.

Regarding erectile function, only a score equal to or higher than 25% compared with baseline indicated remarkable clinical enhancement in patients with neurogenic erectile dysfunction (NED). Only those who benefited significantly in erectile function at the first visit completed the IIEF5 every 6 months. Twenty-eight days before each follow-up, these subjects commenced an assessment period, during which time they attempted sexual intercourse without treatment.

Subsequently, the clinical follow-up data were reported only for SCIPs who demonstrated marked clinical improvement in two pelvic functions at the first second stage until the final visit, which took place by January 2010. The SF-36 was administered to all patients after each follow-up.

The Wilcoxon test was used to compare clinical findings on voiding, and a bowel diary was completed at baseline and during follow-ups post-permanent SNM. Additionally, the Wilcoxon test was applied comparing urodynamic and ano-manometric findings at baseline and during the follow-ups after SNM implantation. The Wilcoxon test was used to compare each domain score of the SF-36 questionnaire at baseline and during follow-ups post-surgery. χ2-test was applied to evaluate which clinical factors were determinant in providing benefits on associated pelvic functions compared with non-responders. The study was conducted after obtaining approval from the local Ethical Committee. All participants provided written informed consent before enrollment in the study.

Results

Seventy-five patients with incomplete SCI were selected for the first stage of SNM. Thirty-seven subjects (49.3%) showed at least two chronic functional disorders. Of those, 44 patients (58.6%) had a lower motoneuron lesion. Sixteen patients (43.2%) with two pelvic dysfunctions evidenced NED with a median IIEF5 score of 15.6 (range 14–18). All patients had satisfactory sexual intercourse; 14 were taking oral phosphodiesterase 5 and 3 were using intracavernous injections of prostaglandin E1.

Overall, 14 individuals (37.8%) with two functional pelvic disturbances at baseline were selected. Following the first stage (median duration 40.5 days), 8 out of 21 SCIPs (38.1%) with NLUTSs and NBSs participated in the second stage because their symptoms improved by at least 50%. Moreover, 6 out of 16 patients (37.5%) with non-obstructive retention (NOR) and NED at baseline were included, as their NOR symptoms improved by at least 50% during the first stage, and three months after the second stage they showed at least a 25% increase in the IIEF5 score as well. Figure 1 summarizes the study.

Figure 1
figure 1

Study design: CO+NOR, constipation and non-obstructive retention; DI, double incontinence; NLUTSs, neurogenic lower urinary tract symptoms; NOR+NED, non-obstructive retention and neurogenic erectile dysfunction.

Full size image

The results are reported according to their functional pelvic disturbances.

Neurogenic lower urinary tract symptoms

Non-obstructive retention

Eight subjects with a lower motoneuron lesion suffered from NOR. Six had American Spinal Injury Association Impairment Scale C and two had American Spinal Injury Association Impairment Scale D. Four patients had voluntary anal contraction. Mean age was 44.2 years (range 31–62). Three out of eight patients (37.5%) emptied per void between 90–120 ml using the Valsalva maneuver, but with elevated post-void residual urine utilizing 2–3 catheterizations per day, whereas the others emptied their bladders exclusively with intermittent catheterization.

Pre-first-stage urodynamic findings in the voiding phase showed that five patients had acontractile detrusor and three had detrusor underactivity. During the second-stage follow-ups, only one patient recovered a normal bladder contractility index with a score of more than 100.9

Seven patients used the Valsalva maneuver to empty their bladders, but with a vesical pressure <70 cm H2O evidenced in every videourodynamic control. No vesical ureteral reflux was ever detected.10, 11

At all follow-ups, two patients did not require catheterizations for a balanced micturition, whereas the others required one catheterization per day. Figure 2 reports the median maximum uroflowmetry data at baseline, at the first visit post-second stage, and at the final visit. Table 2 shows the median variation of voiding diary entries in the first visit post-second stage, and at the final visit compared with baseline.

Figure 2
figure 2

Urodynamic findings in patients with neurogenic lower tract symptoms.

Full size image

Each patient improved by 50% compared with baseline for all the parameters investigated. A statistically significant improvement using the Wilcoxon test (P<0.05) was documented at both the first and final visits post-SNM surgery for all voiding entries.

Urge urinary incontinence

Six subjects had urge urinary incontinence. Mean age was 40.5 years (range 26–54). All had upper motoneuron lesion and American Spinal Injury Association Impairment Scale C. All had voluntary anal contraction. No patient used intermittent catheterization and all emptied their bladders with a residual urine less than 50 ml. In pre-surgery urodynamics, all subjects showed neurogenic detrusor hyperactivity of various degrees during the filling phase. After the second stage, each patient improved all parameters by 50% compared with baseline, and a statistically significant improvement using the Wilcoxon test (P<0.05) was documented at both the first and final visits for all voiding entries. The detrusor pressure at maximum flow rate was similar to that post-SNM compared with baseline, with a maximum value of just over 50 cm H2O attained throughout the urodynamic investigation in the patient with grade I detrusor sphincter dyssynergia, according to the Blavais classification.12 Bladder capacity and maximum detrusor pressure are reported in Figure 2. During the assessment period, three patients used anticholinergics, but with lower dosages compared with baseline.

Neurogenic bowel symptoms

Urge and passive fecal incontinence

The same six patients with urge urinary incontinence had both urge and passive fecal incontinence. The mean total gastrointestinal transit time in days was 2.8. All were fully continent post-permanent SNM.

Table 3 reports the main clinical data on bowel function. Each patient improved all parameters by 50% compared with baseline, and a statistically significant improvement using the Wilcoxon test (P<0.05) was detected at both the first and final visits post-SNM surgery for all bowel diary entries recorded. Anorectal manometric findings at baseline are reported in Figure 3. After second stage, all anorectal manometric values compared with baseline shifted no more than ±5%.

Figure 3
figure 3

Anorectal manometric findings in patients with neurogenic bowel symptoms.

Full size image

Constipation

The two patients showed slow global transit concerning bowel function with a total gastrointestinal transit time of 3.9 and 5.2 days, respectively. One patient had voluntary anal contraction. Following the second stage, each patient showed improvement in all parameters by 50% compared with baseline, and a statistically significant improvement using the Wilcoxon test (P<0.05) was detected at both the first and final visits post-SNM surgery for all bowel diary entries recorded. Table 3 reports their bowel diary entries. Figure 3 shows their anorectal manometric findings at baseline. Following the second stage, all anorectal manometric values compared with baseline changed by no more than ±5%.

Neurogenic erectile dysfunction

Six patients suffering from NOR also had NED. Three patients had voluntary anal contraction. Their median IIEF5 score was 15.2 (range 14–18). After the second stage, their median score was 22. They stopped taking oral phosphodiesterase 5 inhibitors. Two patients temporarily returned to baseline clinical status with regard to voiding and erectile function after a follow-up of 38.5 months (range 29–48). Diagnosable reasons for efficacy loss were not identified. During the first-stage implant in the opposite S3 sacral root, the two subjects using the Valsalva maneuver had a residual urine volume of less than 100 ml. The median time from the new definite implant to the final visit was 40 months (range 23–57). After the new second stage, these individuals achieved and maintained an IIEF5 score of 22 and engaged in sexual intercourse without needing oral phosphodiesterase 5 inhibitors (see Table 4).

Table 4 Results of the six patients with erectile dysfunction at baseline
Full size table

During follow-ups, none of the other eight subjects reported sexual impairments.

Quality of life

Patients were divided into three groups according to their combined symptoms: double incontinence, NOR+NED, and constipation+NOR. In all second-stage follow-ups, a significant statistical improvement in two subscales (mental and general health) of the SF-36 QoL questionnaire was observed in each group compared with baseline. Patients with double incontinence maintained a significant statistical improvement at all post-SMN follow-ups according to the Wilcoxon test (P<0.05) regarding the emotional role and social functioning domains of the SF-36 (see Table 5).

Table 5 The SF-36 domains where a significant statistical improvement was reached throughout the entire study
Full size table

Complications

Eleven drawbacks were resolved by telemetrically modifying the previous stimulation parameters. Complications were reported by dividing the patients into three groups, as was done for QoL (see Table 6).

Table 6 Number and type of complications for each group
Full size table

Stimulation

The temporary external patient stimulator and permanent implantable pulse generator were both set at a frequency between 5 and 20 Hz and pulse width (210 μs).

Analysis of the failure group

Patients were divided into two groups according to the length of time from diagnosis to SNM surgery (more or less than 3 years); a trend in non-responsiveness was detected in group 1 (duration longer than 3 years). In group 2, 11 out of 14 patients continued to the second stage versus 11 out of 23 from group 1 (P=0.133 using χ2-test).

Discussion

Our experience shows that 37.8% of patients with two pelvic dysfunctions pre-SNM maintained a remarkable concomitant clinical enhancement on two pelvic functions, leading us to believe that males with incomplete SCI may be good candidates for SNM. Side effects were scarce and moderate.

One possible explanation of SNM's combined clinical benefit for NOR and NED is its action on the parasympathetic nervous system that influences both detrusor contraction and erectile function because of the fact that both reflex erection and micturition are dependent on intact sacral conus and its reflex loops. This evidence is reinforced by two patients who simultaneously lost the clinical improvement on both functions during follow-ups, and subsequently recovered after a new implant in the contralateral sacral S3 root.13 The hypothesis that increased erectile function is simply due to the improvement in urinary symptoms is difficult to support; in fact, other individuals with NED and NOR showed remarkable clinical improvement only in voiding function.

Considering NLUTS, seven out of eight patients with NOR used Valsalva maneuvers to empty their bladders; this has been urodynamically documented as safe.10, 11 However, Wyndaele et al.10 also report that with increasing time, more than 40% of patients using the Valsalva maneuver showed influx into the prostate and seminal vesicles, and the bladder expression may generate reflux to the upper urinary tract.

In follow-ups longer than 4 years post-second stage, our patients never encountered these complications. Moreover, the significant reduction or abandoned use of catheterization progressively reduced the cost of the neuromodulator over time, and SNM increased their QoL.

The occurrence of late failures in NOR (25%) due to the loss of SNM efficacy is not relevant to the selection criteria with regard to the difference between upper motoneuron versus lower motoneuron lesion patients on NLUTSs. However, these results seem to suggest that in medium- and long-term follow-ups, lower motoneuron lesion patients may have a higher risk of losing their clinical benefits because of the significant reduction or total disappearance of sensation in their perineal area, leading to the absence of perineal relaxation during their bladder expression.

Our study confirms the encouraging clinical results reported in the literature, which describe a significant improvement in bowel symptom severity of varied origins and associated with urinary disturbances.14, 15, 16

Double incontinence represents a much more severe and stressful condition of pelvic floor dysfunction.17 This explains why these patients always attained a marked statistical improvement in four domains of the SF-36 following permanent SNM. Though this is a clinical study, it is still possible to comment on ano-manometric and urodynamic findings. Some post-surgery urodynamic findings showed impressive statistical improvement.

As regards NBS, no variation was detected in ano-manometric measurements after second stage.

Anorectal findings following permanent SNM are varying and discordant in the literature.18, 19 At this time, ano-manometric study is not recommended for SCI patients before NBS surgeries.20 However, the utility of ano-manometric investigation either to better select SCI patients for SNM or to better comprehend SNM's mechanism of action in NBS is not yet well known.

Future research should investigate any predictable clinical factors that are correlated with the dual advantages of SNM on pelvic functions. The duration of time from neurological diagnosis to SNM therapy should be a parameter. Considering the small number of patients, larger, high-quality randomized crossover trials are needed to allow the effects of SNM on these conditions to be assessed with more certainty.