Neurogastroenterol Motil (2008) 20, 1127–1131
doi: 10.1111/j.1365-2982.2008.01154.x
Sacral nerve stimulation and rectal function: results of a prospective study in faecal incontinence S. ROMAN
,* , , à
, , , T. TATAGIBA ,* H. DAMON ,* X. BARTH § & F. MION * à
*Digestive Physiology, Hospices Civils de Lyon, Edouard Herriot Hospital, Lyon, France Universite´ Claude Bernard Lyon I, France àInserm U865, Lyon, France §Digestive Surgery Department, Hospices Civils de Lyon, Edouard Herriot Hospital, Lyon, France
Abstract The mechanisms of action of sacral nerve stimulation (SNS) to treat faecal incontinence remain poorly understood. The aims of our study were: (i) to measure the effect of SNS on rectal function and (ii) to evaluate rectal function as a predictive factor of clinical response to SNS. Rectal function was studied before and 3 months after permanent SNS in 18 patients (17 women, mean age 58.5 years) with faecal incontinence, using an electronic barostat. Rectal sensitivity and volume variations were recorded during isobaric distensions. Three months after SNS, 14 patients had a significant improvement of faecal incontience symptoms and four had not. Baseline Ômaximal tolerated volumeÕ was significantly lower in the positive response group (210 ± 56 vs 286 ± 30 mL, P = 0.02). Baseline rectal compliance was lower in patients with a positive response than those without, although this difference did not reach significance (6.2 ± 3.2 vs 9.2 ± 2.9 mL mmHg)1,P = 0.10). Rectal compliance was not significantly modified by SNS. Our results suggest that an increased rectal capacity as measured by the maximal tolerated volume may be a predictive factor of poor response to SNS in faecal incontinence. SNS does not significantly modify rectal function.
INTRODUCTION Sacral nerve stimulation (SNS) has been used effectively to treat faecal incontinence.1–3 However, the effects of SNS on anorectal physiology, and thus the mechanisms of the efficacy of this therapy, remain poorly understood. Faecal continence depends on various factors such as stool consistency, colorectal motility, rectal sensitivity and compliance and anal sensitivity and sphincter functions. Rectal adaptation to distension may play an important role in faecal incontinence, in relation with the reflex contraction of the external anal sphincter.4 Electrical stimulation of sacral nerves implies stimulation of somatic fibres to the external anal sphincter and pelvic floor, autonomic fibres to the internal anal sphincter and the colon and rectum and afferent sensory fibres from the anus and rectum. It has been shown that SNS may modify not only somatic and autonomic nerves function5,6 but also central nervous system activity.7,8 The effect of SNS on anal sphincter function is inconsistent: some authors have observed that SNS enhances resting and maximum anal squeeze pressure,9–11 while others reported no change in resting 2,3,12 and/or squeeze pressure.1,12,13 Results of SNS on rectal sensitivity and compliance are also unclear. Vaizey et al.1observed a decreased rectal sensitivity to distension with no change in rectal compliance, whereas others noted a decreased threshold of rectal perception.14,15 Finally, recent results indicate that a decrease in the efficacy of chronic SNS over time is observed in a significant proportion of patients, despite a positive SNS test prior to chronic stimulation.3,16,17 The search for predictive factors of long-term response to SNS is thus important in order to optimize the patientsÕ selection. The aims of this study were thus to measure the effect of chronic SNS on rectal function using an
Keywords faecal incontinence, rectal barostat, rectal compliance, sacral nerve stimulation.
Address for correspondence Franc¸ois Mion, Digestive Physiology, Pavillon H, Hopital Edouard Herriot, Place dÕArsonval, 69437 Lyon Cedex 03, France. Tel: +33 4 72 11 01 36; fax: +33 4 72 11 01 47; e-mail:
[email protected] Received: 21 February 2008 Accepted for publication: 30 April 2008 Ó 2008 The Authors Journal compilation Ó 2008 Blackwell Publishing Ltd
1127
S. Roman et al.
Neurogastroenterology and Motility
related disease Quality of Life Index (GIQLI) score. This questionnaire was used again at 3, 6, 12 and 24 months after SNS and the efficacy of the therapy evaluated by a decrease of more than 40% of the JWS. For SNS, a quadripolar lead (InterstimÔ, Medtronic, Minneapolis, MN, USA) was placed percutaneously, under local anaesthesia, adjacent to the sacral nerves through the third right or left sacral foramen and connected to an external pulse generator. A test was then performed for 2–4 weeks, with continuous stimulation, a pulse width of 210 ls, a frequency of 10 Hz and the minimal current voltage to obtain a painless anal or perineal sensation (from 1 to 6.5 Volts). The positivity of the test was assessed based on the selfassessment of improvement by the patients and a decrease of more than 40% of the JWS. The test was positive in all 18 patients and a pulse stimulator (Medtronic InterstimÔ) was implanted subcutaneously, below the superficial fascia, in the upper part of the buttocks ipsilateral to the electrode and connected to the lead. The stimulation parameters used during the test were applied for chronic stimulation. Rectal barostat measurements were performed before the SNS test (without any SNS) and 3 months after permanent implantation (with the stimulator turned on). All examinations were performed after an 8-h fast and a small water enema, in the left lateral recumbent position. An infinitely compliant plastic bag with a maximal volume of 600 mL was secured to the tip of polyvinyl catheter connected to a computer-controlled barostat (G&J Electronics, Toronto, ON, Canada). Before placement in the rectum, the bag was distended and deflated. After a 20-min period of adaptation, a phasic isobaric distension protocol was performed by rapidly inflating the bag to successive predetermined ascending levels of pressure (increment 4 mmHg, maximal pressure 48 mmHg), each level being maintained for 120 s and separated from the next one by a 30s rest period at 0 mmHg. For each level of distension, the mean volume was calculated within the last 60 s of the step. Patients were asked to report the rectal sensation during the last 60 s of each step on a 0–6 scale (0 = no sensation, 1 = vague sensation, 2 = moderate sensation, 3 = clear sensation, 4 = pronounced sensation, 5 = uncomfortable sensation, 6 = painful sensation) and especially to neglect the sensation felt during the first 15 s of each distension step. Distension was stopped as soon as the patient reported painful sensation or when the last distension step was reached (48 mmHg). Maximal tolerated volume (MTV) and pressure were recorded. Rectal compliance was defined as the slope of the volume/pressure curve for the 4–20 mmHg distension steps.18
electronic barostat and to evaluate if rectal function could serve as a predictive factor of SNS efficacy in faecal incontinence.
PATIENTS AND METHODS Patients From September 2002 to May 2007, 68 patients underwent chronic SNS implantation for severe faecal incontinence at our institution. Among those, 18 patients (17 women, one man, mean age 58.5 years) were included in a prospective study to evaluate specifically the effect of SNS on rectal function. This study was performed in accordance with the French Public Health Law. The aetiology of faecal incontinence was idiopathic (12 patients), post-obstetrical injury (three patients), neurologic (one patient), post-recto-anal fistula (one patient) and post-hysterectomy for cervix cancer (one patient). Six patients had urge incontinence, two passive incontinence and 10 mixed incontinence. The duration of symptoms ranged from 1 to 27 years with a mean of 9 years. Anal endosonography revealed an internal sphincter defect in three patients and an external one in three others. No patient had a combined internal and external sphincter defect. Results of ano-rectal manometry were available in 15 patients. Mean anal resting pressure was 4.8 ± 1.4 kPa. The mean squeeze pressure increase was 3.3 ± 2.3 kPa. Rectal volume for the first need to defecate as assessed by manometry was 73 ± 48 mL (range: 10–200 mL). Baseline characteristics of the 18 patients are summarized in Table 1.
Methods Prior to SNS, all patients filled in a self-administered questionnaire including the faecal incontinence Jorge and Wexner score (JWS) and the GastroIntestinal Table 1 Baseline patientsÕ characterisitics Study group (n = 18) Age (years) Sex ratio M/F Body mass index (kg m)2) Previous recto-anal surgery Rectopexy Surgical sphincter repair Diabetes Dyschezia Urinary incontinence Jorge and Wexner score before SNS
58.5 ± 11.5 1/17 24.9 ± 3.8 6 2 4 3 9 9 14.8 ± 2.4
SNS, sacral nerve stimulation.
1128
Ó 2008 The Authors Journal compilation Ó 2008 Blackwell Publishing Ltd
Volume 20, Number 10, October 2008
Sacral nerve stimulation and rectal function
Table 2 Rectal function as evaluated by electronic barostat All patients Number of patients 18 First sensation volume (mL) Before SNS 44 ± 24 3 months after SNS 71 ± 43 P 0.24 Maximal tolerated volume (mL) Before SNS 227 ± 60 3 months after SNS 201 ± 60 P 0.14 Maximal tolerated pressure (mmHg) Before SNS 34 ± 9 3 months after SNS 29 ± 9 P 0.09 Compliance (mL mmHg)1) Before SNS 6.89 ± 3.17 3 months after SNS 6.97 ± 2.94 P 0.88
Patients with improvement
Patients without improvement
P
14
4
44 ± 24 71 ± 47 0.17
45 ± 31 68 ND
0.22 0.95
210 ± 56 202 ± 67 0.64
286 ± 30 199 ± 23 0.07
0.02 0.95
34 ± 9 31 ± 10 0.33
35 ± 8 25 ± 2 0.10
0.85 0.30
6.23 ± 3.02 6.36 ± 3.01 0.93
9.20 ± 2.89 9.10 ± 1.33 0.72
0.10 0.10
SNS, sacral nerve stimulation.
Statistical analysis
Effect of SNS on rectal function
Quantitative variables were expressed as mean ± standard deviation. Qualitative variables were expressed as positive values. Distribution of quantitative variables was compared between groups by ANOVA and those of qualitative variables by the chi-squared test. The Wilcoxon paired rank test was used to compare quantitative parameters before and after SNS. For each analysis, P-values of
