We have studied 30 patients with idiopathic (neurogenic) faecal incontinence using anorectal manometry and concentric needle and single fibre electromyographic methods. We have measured the terminal motor latency in the pudendal nerves of these patients using a new digitally directed transrectal stimulation and recording technique (right mean (+/- s.d.) 3.2 +/- 0.9 ms, left mean (+/- s.d.) 3.0 +/- 0.9 ms) and compared the results with those obtained from 28 normal subjects (right mean (+/- s.d.) 2.0 +/- 0.5 ms, left mean (+/- s.d.) 1.9 +/- 0.3 ms). These differences between normal and incontinent patients were significant (P = 0.01) using the Wilcoxon Rank Sum Test. The findings support the hypothesis that idiopathic (neurogenic) faecal incontinence is due to damage to the nerve supply of the pelvic floor musculature.

SNS produces symptom benefit in patients with faecal incontinence that is associated with a reversible reduction in corticoanal excitability. SNS therefore drives dynamic brain changes that may play a functional role in influencing anal continence.

Objective To design and validate a condition-specific health-related quality of life questionnaire for the assessment of women with anal incontinence. Design A psychometric study by postal survey.Setting South Manchester University Hospital, UK.Sample Two hundred and twenty women with known anal incontinence. Method The questionnaire was adapted from the King's Health Questionnaire, a condition-specific healthrelated quality of life questionnaire for the assessment of women with urinary incontinence. The questionnaire was then tested for acceptability, reliability and validity by postal survey. Results The Manchester Health Questionnaire was found to be highly acceptable to women and showed excellent internal consistency, test-retest reliability, criterion and construct validity. Conclusion The questionnaire is both a valid and reliable instrument for the assessment of health-related quality of life among women with anal incontinence. It will be useful in many different clinical settings and be of practical use in the evaluation of women after childbirth. As the good response rates show it could be a successful part of a postal survey.

SUMMARY The latency of the response in the external anal sphincter muscle following transcutaneous stimulation of the cauda equina at the Li vertebral level was measured in nine women with neurogenic faecal incontinence (mean 7-3 SD 0-7 ms) and 11 normal subjects (mean 5-6 SD 0-6 ms) (p = 0.01). There was no difference in conduction velocity between the Li and L4 vertebral levels thus supporting the suggestion that conduction delay in faecal incontinence occurs distally.In patients with idiopathic faecal incontinence there is evidence of neurogenic damage to the puborectalis and external anal sphincter muscles.'-3 We have recently demonstrated an increased latency in the response of the external anal sphincter muscle to trans-rectal stimulation of the pudendal nerve at the level of the ischial spine in these patients.4 Marsden, Merton and Morton' measured the latency of the response in this muscle in two normal subjects following transcutaneous spinal stimulation and we have now applied this method to investigate our patients with faecal incontinence.Parks, Swash and Urich suggested in 1977 that because the pudendal nerves are tightly bound by connective tissue on leaving the pelvis and entering the pudendal canal, descent of the pelvic floor during childbirth or defaecation straining might result in stretch injury to these nerves.' Henry, Parks and Swash6 have shown that the amount of descent of the pelvic floor measured in incontinent patients with the descending perineum syndrome would result in a 20% increase in the length of the pudendal nerves-more than sufficient to cause injury.7 We wished to know if the increased latency found on transrectal stimulation of the pudendal nerves in incontinent patients was limited to the distal portion of the pudendal nerves, as would be expected with this entrapment/stretch hypothesis, or whether conAddress for reprint requests: Dr M Swash, The London Hospital, Whitechapel, London El 1BB, UK.Received 1 1 November 1983 and in revised form 17 January 1984. Accepted 30 January 1984 duction was delayed in the whole length of these nerves suggesting a more proximal lesion. Patients and methodsNine women aged 28-71 years (mean 50 9, SD 16-9 years) presenting with neurogenic faecal incontinence were studied, and the results compared with those obtained from 11 normal volunteers or patients attending for follow-up of benign colonic pathology, or minor ano-rectal conditions, aged 18-58 years (mean 42-9, SD 13-2 years). Manometry and EMGThe incontinent patients were all examined by ano-rectal manometry using methods previously standardised in our laboratory.

SUMMARY In 17 women with chronic constipation, and abnormal perineal descent on straining at stool, there was more severe neurogenic damage to the external anal sphincter muscle and to its pudendal innervation in those patients with a long history than in those with a short history. These results suggest that recurrent trauma to the pudendal nerves can occur during perineal descent, and that this can lead to denervation and weakness of the external anal sphincter muscle.Between 30% and 66% of patients with faecal incontinence give a long history of excessive straining at stool.' 2 It has been suggested that prolonged straininf during defaecation may cause perineal descent; this in turn may result in recurrent injury to the pudendal nerves and so to progressive denervation atrophy of the external anal sphincter muscle supplied by these nerves. The activity of this muscle is reduced in incontinent patients as has been shown by anorectal manometry and electromyography (EMG).2 4Using single fibre EMG, Neill and Swash5 found an increased motor unit fibre density in the external anal sphincter muscle of incontinent patients. This increased fibre density was consistent with the histological evidence of reinnervation in biopsies of this muscle taken from similar patients.3 6 7 Henry, Parks, and Swash7 found muscle fibre hypertrophy in the external anal sphincter muscle of both continent and incontinent patients with abnormal degrees of perineal descent. They suggested that this hypertrophy represents compensation for the loss of muscle fibres caused by pudendal nerve damage induced by the perineal descent.We have reported an increased terminal motor latency in the pudendal nerves after stimulation of these nerves at the level of the ischial spine in patients with neurogenic faecal incontinence.' This work supports our suggestion that the site of damage to this nerve may be at the point where it passes beneath the sacrospinous ligament.3 We have now tested this hypothesis that the pudendal nerves may be damaged in patients with perineal descent and Address for correspondence: Dr M Swash, Neurological Department, The London Hospital, London El 1BB. Received for publication 29 January 1984 1279 chronic constipation by direct measurement of pudendal nerve conduction. Methods PATIENTSSeventeen women with a history of chronic constipation were studied. All had perineal descent on straining. Eleven patients, aged 34-70 years (mean 48-1±9.7 years) had been intermittently straining at stool for between 20 and 50 years (mean 26 years), whereas in the other six patients aged 22-25 years (mean 23*8±1-5 years) the history of straining was of 10 years or less (mean 6-8 years). The mean frequency of straining per week was similar in both groups; 6-0 and 6-3 respectively. None of the six young patients was parous and none had had pelvic surgery, whereas nine of the group with a long history of straining were parous and two of these women gave a history of difficult labour. In addition three patients in this group had had a hysterect...