Peak Anal Pressure 0 25 50 75 100 125 150 175 200 RPT SPT RPT SPT Rest Squeeze Pressure (mmHg) Before PNB After PNB * 0 200 400 600 800 1000 1200 EAS PRM Before PNB EMG (au) * * * * EAS PRM After PNB Baseline RPT Squeeze * * EMG Activities S1837 Puborectalis Muscle Function Assessed By Dynamic 3d Ultrasound Imaging and Vaginal Pressures Noelani Guaderrama, Jianmin Liu, Charles W. Nager, Dolores H. Pretorius, Geoffrey Sheean, Ravinder K. Mittal Our previous studies suggest that the anal canal is made of two zones. Proximal zone is made of overlapping internal anal sphincter (IAS) and puborectalis muscle (PRM) and the distal zone of IAS and external anal sphincter (EAS). Voluntary squeeze increases pressure in both zones. The mechanism by which PRM contraction increases anal canal pressure is not known. We hypothesize that contraction of the PRM compresses anal canal against vagina. METHODS: Vaginal pressure (VP) was measured in 11 normal nulliparous women using rapid pull through, side-hole manometry technique. The PRM-EMG was measured using bipolar needle electrodes in arbitrary units (au). 3D transperineal ultrasound (US) imaging of the pelvis was performed. Pudendal nerve block (PNB) was performed trans- vaginally with 10 cc of 1% lidocaine on each side. Pressures are measured in reference to atmospheric pressure. PRM length is measured as the antero-posterior distance between the lower end of pubic bone and the apex of the ano-rectal angle identified in the 3D-US images. RESULTS: VP recording show a high pressure zone at rest, which is significantly increased by squeeze. The PNB decreases rest and squeeze VP and PRM-EMG. PRM length decreases with squeeze and it increases following PNB, both at rest and squeeze. CONCLUSIONS: Contraction-related shortening of the PRM compresses anal canal against vagina, which is the likely mechanism for the increase in proximal anal canal pressure during voluntary squeeze. The PRM function in women may be easily assessed by measuring rest and squeeze pressure in the vagina. Vaginal Pressure 0 20 40 60 80 100 120 Before PNB After PNB Pressure (mmHg) Rest Squeeze * * * PRM EMG 0 200 400 600 800 1000 1200 Before PNB After PNB EMG (au) Rest Squeeze * * * PRM Length 0 20 40 60 80 Before PNB After PNB Length (mm) Rest Squeeze * * * S1838 Randomized Controlled Trial Shows Biofeedback To Be Superior To Alternative Treatments for Patients With Pelvic Floor Dyssenergia-Type Constipation Steve Heymen, Yolanda Scarlett, Kenneth Jones, Douglas Drossman, Yehuda Ringel, William E. Whitehead Introduction A meta-analysis of uncontrolled trials showed that 75% of subjects reported improvement in their constipation symptoms after biofeedback (Heymen et al., 2003). However, randomized controlled trials in adults are lacking. Aim To compare electromyographic (EMG) biofeedback to two alternative treatments for patients with pelvic floor dyssynergia-type constipation (PFD). Methods All subjects participated in a 4-week run-in (16% had adequate relief). The remaining 75 (63 females) were randomly assigned to one of three treatment groups: 1) EMG biofeedback, n28; 2) 5mg diaze- pam, n26; 3) placebo pill, n21. Diazepam/placebo pills were taken one hour prior to dinner. All patients received 6 biweekly 1-hour sessions and were trained to do pelvic floor muscle (PFM) exercises with instructions for proper defecation attempts after dinner. Diary data (cathartic use, straining effort, incomplete bowel movement (BM) sensations, Bristol stool scores, and compliance with PFM homework) was reviewed in each session. The primary dependent measure was a report of adequate relief (yes or no) three months post-treatment. Results As an interim analysis, alpha was set at .01. Biofeedback for PFD was superior to a placebo pill group. 71% of biofeedback vs. 33% of placebo patients were successful (X2  7.04, p.008). Biofeed- back was also superior to the diazepam group. 71% vs. 20% were successful (X2  14.02, p.001). In addition, there were significantly more unassisted BMs following biofeedback (5.6/week) compared to placebo (2.5/week, t(47) 3.3, p.002), and a trend favoring biofeedback over diazepam (3.2/week, t(51)2.5, p.017). Prior to treatment, the groups did not differ (p.05) on demographic variables [gender, race, age (mean 48.6 years), symptom duration (mean 13.4 years), or number of physician visits in the previous 6 months (mean 5.0 visits)], physiological variables [rest or push anal EMG values, first sensation threshold, or colonic transit time], or psychological variables [depression, anxiety, sexual abuse history]. In addition, there were no differ- ences between groups in the expectation of benefit after the first session. Conclusion This interim analysis provides definitive support for the efficacy of biofeedback for PFD, and enrollment in the trial has been closed. Supported by grants R01 DK57048, R24 DK67674, and NCRR grant M01 RR00046. Diazepam provided by Milan Pharma- ceuticals, PAC-QOL questionnaire provided by Jansen Pharmaceuticals. A-266 AGA Abstracts S1839 Anorectal Physiologic Studies in Asymptomatic, Healthy Volunteers Shows Variety of Abnormalities Hyun Jeong Kim, Joon Seong Lee, Hee Hyuk Im, Kyung Ran Hwang, Bong Min Ko, Su Jin Hong, Chang Beom Ryu, Jin Oh Kim, Joo Young Cho, Moon Sung Lee, Chan Sup Shim, Boo Sung Kim BACKGROUND: Anorectum plays an important role in the regulation and in the maintenance of continence. A wide range of tests were developed to assess anorectal function, but there was a lack of consensus regarding which test to use and when. It is been argued that anorectal physiologic studies fail to meet the criteria of a useful clinical test. AIM: To investigate the incidence of abnormal findings from the anorectal physiologic studies and agreement with other tests in the healthy volunteers without any symptoms including defecatory problems.. METHOD: Our prospective study included 25 asymptomatic healthy volunteers (M : F  15 : 10, mean age : 26.8  8.3 yr.). To all patients, questionnaires were given to check for bowel movement problems. Anorectal manometry (ARM), balloon expulsion test (BET), electromyography (EMG), colon transit time (CTT) and video defecography (DFG) were performed. RESULT: See table. In video defecography, the difference of anorectal angle between straining and squeezing was 9.6  13.5° (mean  SD) and in 13 subjects (52%), the angle was less than 10°. Eight subjects had an anterior rectocele over 1 cm, 5 intussuscep- tion, 11 spastic pelvic floor and 3 showed 50% remnant barium after defecation. In agreement ARM BET EMG CTT DFG ARM Out Obst.(13) 8% 54% 0% 54% Normal (12) 92% 83% 83% 67% BET Out Obst. (2) 0% 0% 50% 0% Normal (23) 91% 61% 96% 43% EMG Out Obst. (9) 22% 78% 0% 44% Normal (16) 100% 63% 88% 56% CTT Out Obst. (2) 50% 0% 100% 50% Normal (23) 52% 61% 70% 50% DFG Out Obst.(11) 9% 36% 64% 9% Normal (14) 93% 64% 50% 93% CONCLUSIONS: Using solitary anorectal physiologic study, interpretation should be per- formed cautiously because the abnormal findings of anorectal physiologic studies can be frequently observed in asymptomatic subjects. Key words : Anorectal Physiologic Studies, Healthy Volunteers S1840 Digital Rectal Stimulation Causes Increased Left Sided Colonic Motility in Persons With Sci Mark A. Korsten, Amit Monga, Geeta Chaparala, Amir Khan, Ron Palmon, Reagan Mendoza, Alan S. Rosman, Ann Spungen, William A. Bauman Background: Difficulty with evacuation (DWE) is common after spinal cord injury (SCI) and adversely affects the quality of life. In addition to laxatives and enemas, digital rectal stimulation (DRS) is often employed to facilitate bowel evacuation. However, the basis for the efficacy of DRS is not known. In the present study, we assessed whether DRS increases colonic motility because normal defecation results from such bowel action. Methods: Five subjects with SCI (3 with paraplegia, 2 with tetraplegia) were studied using methods pre- viously reported (Korsten et al, Gastro 2003;124A115). Subjects were studied several hours after a typical bowel care session. After a motility catheter with four solid-state transducers was endoscopically affixed to the left side of the colon, a barium oatmeal paste (with the consistency of stool) was introduced into the rectum and descending colon. After a 10 min. baseline period, subjects underwent DRS for 1 min. repeated for a total of five times, with 2 min. intervals between each manipulation. Manometric and fluoroscopic data were recorded using a videofluoromanometric apparatus (Kay Elemetrics). Results: Spontaneous colonic motility were not observed during the baseline period prior to DRS. During and after each DRS, evidence for peristaltic activity was observed both manometrically and fluoroscopically. Compared to the baseline (0 waves/minute), the mean number of peristaltic waves ( SEM) increased to 1.75  0.43 waves per minute during DRS and 1.55  0.32 waves per minute during the period immediately after cessation of DRS (p  0.05, compared to baseline by repeated measures ANOVA and Newman-Keuls test). The frequency of contractions during or immediately after DRS was not significantly different. The barium paste was eliminated in each of the five subjects by the fifth DRS. The mean amplitude to the peristaltic contractions was 44 mm Hg (range 3.3-120). There was no difference in this parameter when comparing contractions during and after DRS. Conclusions: DRS causes left-sided colonic activity in subjects with SCI. At least in part, enhanced contractions of the descending colon and rectum may contribute to bowel evacuation in these individuals. The neurophysiological mechanism for the observed effect of DRS on colonic motility is not known but may involve a reflex arc between anorectal sensory receptors and prokinetic, possibly parasympathetic, efferent neurons.