Hypothesis / aims of study
Electrical pudendal nerve stimulation (EPNS) combines the advantages of pelvic floor muscle (PFM) training (PFMT), transanal electrical stimulation (TES) and pudendal neuromodulation and incorporates the technique of deep insertion of long needles [1,2]. The aim of this original study was to assess the efficacy of EPNS versus biofeedback (BF)-assisted PFMT plus TES in treating post-radical prostatectomy (PRP) urinary incontinence (PRPUI) and explore its mechanism of action by simultaneous recordings of perineal ultrasonographic PFM movements and pelvic floor surface electromyogram.
Study design, materials and methods
According to the result of a pilot study the required sample size was calculated to be 92 patients for a power of 0.90, an α of 0.05 and a ratio of 2:1. Of 126 PRPUI men assessed for eligibility, 96 were enrolled in study from October 2014 to Jan 2017. They were randomized, through drawing lots at a ratio of 2:1, to group Ⅰ (64 cases) and group Ⅱ (32 cases). Study inclusion criteria were urinary incontinence at 1 month or more after radical prostatectomy (RP), ≥2 incontinence episode a week on baseline 7-day bladder diary and no residual cancer after RP. Exclusion criteria were the presence of preoperative urinary incontinence, treatment with anticholinergics, urinary tract infection, hematuria, postvoid residual volume >100 ml, neurological disorders and urethral stricture. Group Ⅰ was treated with EPNS as described previously [1,2]. Four sacrococcygeal points were selected for deep insertion of long acupuncture needles used as electrodes. The two upper points are located about 1 cm bilateral to the sacrococcygeal joint. A 0.40 Х 100 mm needle was inserted perpendicularly to make the needle tip reach the vicinity of the main trunk of the pudendal nerve (PN).The locations of the two lower points are about 1 cm bilateral to the tip of the coccyx. A 0.40 Х 100 or 125 mm needle was inserted obliquely toward the ischiorectal fossa to make the needle tip reach the perineal nerve (PN branch). Electrical stimulation with a biphasic 2-millisecond pulse duration was provided at a frequency of 2.5 Hz and an intensity of 45~55 mA. Strong rhythmic and cephalad PFM contraction around the root of the penis must be maintained during the entire electrostimulation. EPNS was given 60 minutes 3 times per week for a total of 8 weeks. Group Ⅱ was treated by BF-assisted PFMT and following TES at a current intensity of < 60 mA and alternate frequencies of 15 Hz and 85 Hz, 40 minutes (20 minutes each) 3 times a week for a total of 8 weeks. Group Ⅱ patients also conducted 30 maximal high-intensity PFM contractions for 2-6 seconds, 3 sessions every day at home for a total of 8 weeks. The total time of treatment was basically the same in the two groups. Simultaneous recordings of perineal ultrasonographic PFM contraction (movement) and pelvic floor surface electromyogram were made during EPNS or active PFMT in the first 10 patients of group Ⅰ or group Ⅱ. The primary outcome measure was the International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form (ICIQ-UI SF), which contains three scored items assessing the frequency of leakage, amount of leakage and quality of life. The secondary outcome measure was the number of used incontinence diapers. The ICIQ-UI SF questionnaire was completed and the number of used incontinence diapers was reported by all patients after treatment completion or discontinuation. Outcome assessors were blinded to group assignment. The outcome analysis used an intention-to-treat approach. The Wilcoxon signed rank test and Mann-Whitney U test were used for analysis of the ICIQ-UI SF score and subscores, the incontinence diapers score and the results of the simultaneous recordings.
Results
At baseline there were no significant differences in age (68.5±6.5 vs 67.0±6.7), symptom duration (median 4.5 (range 1.0-65.5) vs 4.5 (range 1.0-30.5) months), the ICIQ-UI SF score (median 18 (IQR 16-19) vs 18 (IQR 15.25-20) ) and subscores, and the incontinence diapers score (median 3 (IQR 3-4) vs 4 (IQR 3-4)) between groups Ⅰ and Ⅱ (all p>0.05). At the end of treatment, the median ICIQ-UI SF score decreased to 11 (IQR 7-14) in group Ⅰ and 15 (IQR 9-17) in group Ⅱ (both p<0.01); the median incontinence diapers score decreased to 3 (IQR 1-3) in group Ⅰ and 3 (IQR 3-4) in group Ⅱ (both p<0.01). The ICIQ-UI SF frequency of leakage, amount of leakage and quality of life subscores also decreased in the two groups at the end of treatment (p<0.05 or p<0.01). Posttreatment ICIQ-UI SF score, incontinence diapers score, amount of leakage subscore and quality of life subscore were significantly lower in group Ⅰ than in group Ⅱ (all p<0.05) (table 1). The ICIQ-UI SF score-based improvement rate was significantly higher in group Ⅰ (a median of 36.8 (IQR 21.1-53.8, range 5.9-100) %) than in group Ⅱ (a median of 16.2 (IQR 10.1-41.9, range 0-71.4) %) (p<0.01). Four group Ⅰ patients and three group Ⅱ patients discontinued treatment ahead of time. Simultaneous recordings showed: (1) B-mode cranio-caudal PFM movements were obviously visible during EPNS and PFMT; (2) M-mode PFM movement amplitude was >0 mm - <3 mm during EPNS and PFMT, and there was no significant difference in the amplitude between EPNS (≥1mm in all 10 patients) and PFMT (≥1mm in 8 of 10 patients) (p=0.69) (fig.1 A and C); (3) pelvic floor electromyogram was sawtooth waves with a median amplitude of 235.3 (IQR 157.0-324.5) mV during EPNS and continuous triangular waves with a median amplitude of 16.9 (IQR 8.6-53.2) mV during PFMT, and there was a significant difference in the amplitude between EPNS and PFMT (p<0.01) (fig. 1 D).
Interpretation of results
Perineal PFM movement amplitude during EPNS was similar to that during PFMT, indicating that EPNS can excite the PN and simulate PFMT in PRPUI patients. Posttreatment ICIQ-UI SF score and incontinence diapers score were significantly lower (both p<0.05) and improvement rate was significantly higher (p<0.01) in group Ⅰ than in group Ⅱ, indicating that EPNS is more effective than BF-assisted PFMT plus TES in treating PRPUI. There is a statistically significant post-treatment differences in the amount of leakage subscore (p<0.05) but not in the frequency of leakage subscore (p=0.68) between the two groups, indicating that a short-term (8 weeks) EPNS is more effective than BF-assisted PFMT plus TES in reducing the amount of leakage and not in decreasing the frequency of leakage in PRPUI patients. The reason why EPNS has a better therapeutic effect may be that EPNS, when performed correctly, can ensure a most effective PFM contraction at every stimulus during the entire electrostimulation in comparison with PFMT done by the patient. On the other hand, EPNS can produce the contractions of all PFMs supplied with the PN and stimulate PN afferents directly by the arrival of the needle tip at the nerve in comparison with TES that mainly produces local perianal PFM contractions and stimulates PN afferents indirectly through the wall of the anal canal because it uses an anal surface electrode.