95 1
A u s r . N . Z . J . Surg. 1992,62.951-958
THE ROLE OF MANOMETRY, ELECTROMYOGRAPHY AND RADIOLOGY IN THE ASSESSMENT OF FAECAL INCONTINENCE ROY L. FINK,*LESLIEJ. ROBERTS+A N D MARKSCOTT* *University Department of Surgery and Departments of .‘Neurology and ‘Radiology, St Vincent’s Hospital. Fitzroy, Victoria, Australia The results of laboratory investigations in 156 patients presenting with faecal incontinence are reviewed to see if and how these investigations supplement a careful clinical evaluation, and in particular to see if they help in the practical management of the problem. All patients underwent anal manometry, and in addition 52 underwent anal sphincter electromyography and 27 defaecatory proctography. Anal manometry quantified sphincteric weakness, and proved superior to digital assessment in this regard. Resting and squeeze pressures were less in those with complete than those with partial incontinence but the differences were not statistically significant. The measurements of rectal sensation and compliance were not additionally helpful. Single fibre electromyography provided the best measure of denervation with re-innervation and was abnormal in about 85”& of the group studied. Jitter studies were unhelpful. Most patients had some abnormality on defaecatory proctography but clinical significance could not be established. The choice of treatment was made on clinical grounds and was not influenced by these investigations.
Key words: electromyography, faecal incontinence, manometry.
Introduction In recent years there has been a dramatic increase of interest in anorectal functional abnormalities and related pathophysiology of the pelvic floor. While there has been an increase in the number and use of laboratory techniques to assess these abnormalities, the role of these techniques is not yet clearly defined. The results of tests of pelvic floor function can only be used to supplement a careful clinical assessment and a combination of tests allows the best evaluation of the situation. Manometry, electromyography (EMG) and defaecatory proctography are finding their way into routine clinical practice. The purpose of this study was to review experience with these investigations at St Vincent’s Hospital, and to evaluate their contribution to the assessment of faecal incontinence.
Methods During the period 1984-90, 310 patients with a variety of conditions were referred for anorectal manometry. Many also underwent anal sphincter electromyography and/or defaecatory proctography .
Correspondence: Mr R. L. W . Fink, 55 Victoria Parade, Fitrroy. Vic. 306.5, Australia. Accepted for publication 5 August 1992
By far the most common referrals were for patients with a degree of faecal incontinence (156) or constipation (98). Less common referrals were for the assessment of function in association with rectal prolapse (15), ileo anal pouches (13) and patients with anal pain (13). The presumed aetiology of the incontinence based on clinical assessment in the 156 patients studied is shown in Table 1. Most fell into the ‘idiopathic’ group. One hundred and twelve (710/0) were female and most were elderly, especially in the idiopathic group, where the mean age was 57 years. Most had minor incontinence. Ninetyone (58%) had incontinence of flatus and liquid stool only, indicating partial incontinence, whereas 65 (42%) had varying degrees of incontinence to solid stool, but regular whole stool accidents were rare.
Table 1. Patients investigated with incontinence: presumed aetiology ‘Idiopathic’
98
Traumatic
45
Surgical Obstetric Injury DXRT
Neurogenic Prolapse Scleroderma Total
5 5
3 I56
26 14
3 2
FINK E T A L
052
A detailed history was taken to determine the aetiology, nature and severity of the incontinence. Rectal examination allowed a clinical assessment of resting pressure, and squeeze pressure with voluntary contraction, both of which were clinically graded (normal, somewhat reduced, markedly reduced). Perianal appreciation of sensation to light touch and pin prick was also tested, along with the reflex contraction of the external sphincter in response to sharp stimulation. All of the patients underwent anorectal manometry, seven patients having two or more investigations. In addition pelvic floor EMG was carried out in 52 of the patients and defaecatory proctography in 27. The reason for the choice of investigations by the referring practitioner in an individual patient could not always be determined by the authors retrospectively. ANAL MANOMETRY
In order to ensure an empty rectum, the patient was instructed to use a Durolax suppository the night before and the morning of the examination. Anal manometry was performed with the patient in the left lateral position. A water perfused open ended low compliance catheter (internal diameter 2 mm) was linked with pressure transducers (Bell & Howell 4/327/I; range 0-400mmHg) and a chart recorder (Neotrace NT 218). The rate of perfusion was 0.5 mUmin. The catheter, which was marked at 1 cm intervals, was introduced into the mid-rectum through a sigmoidoscope and the sigmoidoscope was then removed. Pressures were recorded by a station pull-through technique at I cm intervals, the height being measured from the anal verge. The catheter remained at each station for sufficient time to allow resting pressure to stabilize and then the pressure with maximum voluntary contraction (squeeze pressure) was recorded, the level being recorded in centimetres from the anal verge. The maximum resting pressure and maximum squeeze pressure were thus determined and the levels at which these were identified were noted (Fig. 1). Rectal sensation was assessed by introducing into the rectal ampulla a soft compliant balloon attached to a similar catheter. The balloon was inflated with air at a rate of 1 mWs. The volume at which first appreciation of the balloon occurred (least perceived volume) and the volume at which severe discomfort occurred (maximum tolerated volume) were both recorded. Rectal compliance was measured by the ratio of the maximum tolerated volume to the pressure within the balloon at that point, as recommended by Pescatori and Ravo. The recto-anal inhibitory reflex was tested by rapid inflation of the rectal balloon (50 mL/s) and noting any drop in resting anal canal pressure as recorded by a second intra-anal manom-
'
eter placed at the level of the previously identified maximum resting pressure (Fig. I). Results were compared to normal values reported in the literature,* especially where exactly similar technique and apparatus had been employed. ELECTROMYOGRAPHY
Electromyography (EMG) recordings were made using a Medelec MS6 EMG machine. A Medelec FCD 37 concentric needle electrode was used for the concentric needle recordings which were made in the external anal sphincter and in most cases the puborectalis muscle as well. Details of the activity seen on needle insertion, at rest, with maximal voluntary contraction, coughing and on voluntary bearing down and relaxing the sphincter muscle were recorded in all patients and in most patients photographic record of this activity was made. Concentric needle electrode recordings were made from the posterior and lateral parts of the external anal sphincter. After digital examination the needle was advanced into the puborectalis muscle where similar recordings were made. A standard Medelec single fibre needle (SF37) with a 25 micron recording surface was used for the single fibre EMG studies. Single fibre EMG recording was performed in the external anal sphincter and measurements of jitter and fibre density were made.4 The single fibre EMG needle was inserted percutaneously in the mid-line posterior to the anal verge. Measurements and recordings were then made from the lateral and posterior parts of the external anal sphincter. The recordings were made during resting activity, although very occasionally voluntary activation was required or a Foley catheter was inserted, the balloon inflated, and a small weight was attached to pull against the anal sphincter. Action potentials of greater than 150yV were accepted and these triggered the sweep of the machine. They were then delayed so they could be visualized. At each recording site one or more action potentials arising from individual muscle fibres were seen. The number of potentials which were linked together was recorded at each site. The fibre density was defined as the mean number of units or potentials seen per insertion from 20 different locations in the external anal sphincter. When simultaneous recording is made of the action potentials of two or more muscle fibres supplied by branches of the same axon, the interval of time between the potentials varies only slightly under normal circumstances. This variability is termed jitter (Fig. 2). When two or more linked potentials were found and when the recording appeared stable, the traces were digitized and transferred on line by a Medelec C16 interface to an IBM compatible PC for storage
INVESTIGATION OF FAECAL INCONTINENCE
Distance from anal verge (cm)
L E
sigmoid colon was outlined. A single spot film was taken at this time to document the direction of the anal canal. The rectal tube was carefully removed. The patient was then positioned on a specially constructed commode for imaging purposes. Both static and dynamic imaging was performed with specific observations being made at rest, with 'gripping' (the patient was instructed how to hold on tightly, thereby attempting to increase the tone of the puborectalis sling), while straining and while evacuating. The patient was viewed in the lateral position. The films were reviewed at the completion of the procedure and if satisfactory, the patient was allowed to leave. The examination usually took approximately 15 min. Measurements were made as follows: (i) pelvic descent (i.e. the distance of the anorectal junction below the line joining the tip of the coccyx and the inferior margin of the pubic symphysis), this is normally < 2.5 cm;5 (ii, the anorectal angle, this is normally 92" (rt 1.5) at rest and 137" (+ 1.5) while strainin& (iii) the distance from the anorectal junction to the lower end of the coccyx (6.4 k 0.5 cm in incontinent patients vs 11.0 k 0.4cm in constipated patients).6 Other observations included evidence of intussusception, evidence of rectal prolapse, and any other abnormal configuration of the rectal wall. The patients were treated by a number of surgeons and by a variety of methods (Table 2), and the patients' records were reviewed to see how the investigations added to the clinical assessment and influenced the choice of treatment.
50k
100.
n -
953
i
4
t
I
1
D
Recto-anal inhibitory reflex
Fig. 1. Anal manometry. Normal tracing showing (a) resting pressures and squeeze pressures resulting from voluntary contractionof sphincters (S) as catheter is withdrawn through the lower 4cm of the anal canal; and (b) normal recto-anal reflex with progressive drop in anal canal pressure with rectal balloon distension (1) and rise with balloon deflation (D).
STATISTICAL METHODS
Fig. 2. Single fibre EMG recording. This shows four traces superimposed.The first unit is the triggering potential. The variation in time between the first and second units constitutesjitter.
and subsequent analysis. Measurement of the jitter (expressed as the mean consecutive difference) was made from at least 10 different potential pairs. Up to 20 recordings were made for each pair. DEFAECATORY PROCTOGRAPHY
The examination was clearly explained to the patient before commencement of the procedure to ensure optimal co-operation. No bowel preparation was used. A standard rectal tube was placed in the rectum under fluoroscopic screening with the patient in the left lateral position. The position of the tip of the tube was confirmed fluoroscopically before a volume of thick barium was syringed into the rectum under screening, which was continued until the
Simple Student's 1-tests were used to compare results between those with partial and complete incontinence, and between groupings separated by digital assessment. The relationship between squeeze pressures and fibre density on electromyography was assessed by using a Pearson correlation coefficient.
Results A N A L MANOMETRY
The results of anal manometry are shown in Table 3 which details results in the group as a whole and then is divided into those with complete and partial incontinence as defined previously. Both resting pressures and squeeze pressures in this group of incontinent patients were lower than normal values quoted in the l i t e r a t ~ r e . While ~ . ~ resting pressures and squeeze pressures were lower in those with complete rather than partial incontinence the differences were not statistically significant. There was also no significant difference in rectal sensation between those with complete and partial incontinence.
FINK E T A L .
954
Table 2. Incontinence: treatment -~
Operative
Conservative
Sphincter repair Post-anal repair Rectopexy Colectomy Colostomy Delorme
70 20 7 5 2 I
Physiotherapy Constipating agents Banding mucosa Enemas Stop medication Biofeedback
8 5 4 2 2 I
Ten patients had no treatment, and the type of treatment was unknown in 33 patients.
The recto-anal reflex was absent in 21.6% of the patients and in those with an absent reflex the mean maximum resting pressure was 20.1 t 7.8 mmHg as compared with 28.2 k 13.2mmHg in those with a reflex ( P 4 0 . 4 p s and/or if there were two or more pairs of 10 which showed jitter in excess of 5 5 p . The correlation between squeeze pressures measured manometrically and fibre density on single fibre studies is shown in Fig. 5 . The correlation was significant ( r = - 0.38; P = 0.019). The data were fitted with a linear regression which showed a drop in squeeze pressure from 48 mmHg to 36 mmHg (i.e. 25%) for a rise in fibre density from 2 to 3.
Table 3. Incontinence: manometry T o t a l group ~
~
~
Pressures (mmHp) MRP MSP Rectal sensation (ccs)
~
__ _ _
( n = 156)
Partial incontinence Complete incontience ( n = 91) ( 1 1 = 65) ~ _ _ _ _ _ __ _ ~~
~~
~
24.7 ? 13 44.9 f 20.4
26.0 I 13.5 47.1 2 20. I
22.6 k 12.4 41.7 5 20.5
LPV 87.1 f 52.4 MTV 210.4 2 79.7 Recto-anal reflex present 109, absent 39 Rectal compliance 2.74 2 1.45 (ccs/mmHg)
85.4 I 45.1 202.9 14.0
89.5 k 61.4 221.3 f 85.9
*
~
~
Results arc expressed as means and standard deviations. MRP = maximum resting pressure; MSP = maximum squeeze pressure: LPV = least perceived volumc; MTV = maximum tolerated volume.
955
1NVESTIGATION OF FAECAL INCONTINENCE
Table 4. Incontinence: electrornyography concentric needle electrode recording
110r
100 -
__ ++
90-
External anal sphincter (n Overall activity Polyphasia Anismus Puborectalis ( n = 34) Overall activity Polyphasia Anismus
80.
- ror
+++
0
r
-m
E E 60-
2
m
Abnormal
Markedly abnormal
31 21 49
13 28 6
2 3 0
27
6 23 3
1
Normal = 52)
10
31
1
0
50-
0
40
Overall activity refers to the assessment of insertional, resting and maximal activity and the response to coughing. Normal polyphasia and anismus refer$ to none being present.
-
Table 5. Incontinence: single fibre electromyography external anal sphincter ( n = 39) No. patients
C
N
R
Resting pressure
P
N
R
P
Squeeze pressure
Fig. 3. Digital assessment of sphincters. Comparison of manometrically assessed pressures with digitally assessed pressure groupings (normal (N), reduced (R) and poor (P)). #Not significant; # # P < 0.001; # # # P < 0.01.
Fibre density < 1.98 > 1.98 Mean Standard deviation Jitter Normal Abnormal
6 33 2.46 0.44 21 18
D EFA ECA TO RY PROCTOGRAPHY
Discussion
The findings on defaecatory proctography are summarized in Table 6. Normal values are those defined and Preston eta1.6It can be seen by Mahieu et al.5*8 that there was wide variation in all of the values looked at and most patients had some abnormality, but these alone may have been insufficient to cause the incontinence.
Laboratory investigations have undoubtedly improved our understanding of the mechanism of faecal continence and causes of incontinence, especially so called idiopathic incontinence. Our aim was to see how these investigations help in the practical management of the condition. The study represents a large experience during several years
Fig. 4. (a) Normal motor unit action potential. (b) Polyphasic unit with multiple changes of phase.
956
FINK ETAL.
80 r
resting pressure within the group. Resting pressures are reported as being very variable in most studies2 and partially relate to age and sex differences. As E 60E might be expected the resting pressure in those with E 503 complete incontinence was lower than in those with 2 40partial incontinence, though the difference was not E =30statistically significant. Because the external sphincN ter is thought to be concerned with maintenance of :2 0 = 10continence to solid stool, and because its contribuIU n tion is largely responsible for squeeze pressure, it 0 : : might be expected that squeeze pressures would be 1 15 2 25 3 35 4 significantly lower in those with complete as opFibre density posed to partial incontinence. Read et al. showed Fig. 5. Correlation between squeeze pressure and fibre such a difference but we were unable to confirm density on single fibre EMB ( r = -0.38; P = 0.019). this." Rectal sensation as measured by the least perceived volume and maximum tolerated volume on Table 6. Incontinence: defaecatory proctography distension with an intra-rectal balloon showed a ~wide range of values in the incontinent patients. Anorectal angle at Similarly other studies have shown a wide range of 91.6 f 12.1 ( n = 9 2 f 1.5) rest (") values for rectal sensation, not only in incontinent Anorectal angle at evacuation (") 115.6 f 25.3 ( n = 137 _+ 1.5) patients but also in normal individuals. "-I4 For this Perineal descent at reason such measurements are not useful in the rest (cm) 2.75 k 2.0 ( n = < 2.0) assessment of patients with faecal incontinence. The Perineal descent at recto-anal inhibitory reflex was absent in 36% of 5.32 t 3.0 ( n = < 4.0) evacuation (cm) the present patients, and resting pressure in these Anococcygeal patients was significantly lower than in those with a distance (cm) 6.6 ? 2.6 reflex. These patients presumably had severe weakness of the internal anal sphincter. Theoretically if Results are expressed as means and standard deviations. the volume required to elicit the reflex was less than the least perceived volume then this could be a factor in incontinence but this was not found in any of the present patients. Rectal compliance was widely variable and aland multiple investigations were frequently perthough a group could be identified in whom lowformed with the expectation that this might give the ered rectal compliance might have contributed to best assessment. the incontinence, sphincter pressures were also low, Clinical evaluation allows, in most cases an adso that a non-compliant rectum could not be regarded equate assessment of the likely cause and degree of as a prime cause. The reason for a non-compliant incontinence. Experienced clinicians often express rectum was not always obvious, but in two it relatthe view that digital assessment of the anal sphinced to previous pelvic irradiation, in one to fibrosis ters is adequate for practical purposes. Indeed, in following trauma and in another to scleroderma. this study, digital assessment did allow satisfactory In summary, manometry alone did not elucidate separation of patients into broad groups with northe cause of incontinence but it was able to quantify mal, reduced and very reduced resting and squeeze the sphincteric component, and this was superior to pressures. It was somewhat easier to assess resting digital assessment. The measurement of rectal sentone than voluntary contraction, especially when sation, the recto-anal inhibitory reflex and rectal the latter was somewhat reduced. It is also noted compliance added little to the assessment. that in individual instances the digital assessment of The role of EMG is to document a neurogenic both resting and squeeze pressures sometimes was basis for incontinence and to quantify this. Concenquite inaccurate. Therefore, while digital assessment tric needle electrode recording showed a normal overis usually accurate there is no doubt that manornetall pattern of activity in the majority of patients. On ric assessment is more reliable. Others have shown the other hand, chronic changes of denervation with a moderate to good correlation between digitally re-innervation (i.e. polyphasia and increased and manometrically measured anal p r e s s ~ r e . ~ ~ partial '~ fibre density) were seen in the vast majority of The present study demonstrated low resting pressures in the group of incontinent patients as compatients. The presence of excess polyphasia and pared with normal values quoted in the literature,*s3 increased fibre density correlated very well. If the fibre density was > 2 then polyphasia was seen in though there was wide variation in the value of 8
r
-
70-
..
.. . .
-
INVESTIGATION OF FAECAL INCONTINENCE
all except one patient. Fibre density recording is favoured because it is a more objective test. Others have shown it to be more sensitive than conventional EMG" and have indicated a lack of overlap between the fibre density of denervated and normal sphincters. Concentric needle EMG therefore adds little to the assessment, although after traumatic damage to the sphincters, it is useful for mapping of sphincter muscle activity. I 5 , l 6 Single fibre EMG has demonstrated evidence of denervation in 85% of the patients investigated, confirming that pudendal nerve neuropathy of varying degrees is a common finding in most cases of incontinence. Jitter was abnormal in about half of the patients but there was no correlation with fibre density or degree of polyphasia. 'Abnormalities' in jitter do not necessarily indicate pathology. The lack of correlation of increased jitter with any other parameters measured suggests that this finding is not significant. Variations of the rates of firing of the action potentials may alter the propagation velocity of action potentials along muscle fibre membranes and thereby affect jitter in the absence of disease.17 This seems a likely explanation of the increased jitter seen and it was concluded that jitter measurement is not useful in this clinical setting. The degree of denervation as measured by fibre density and the maximum squeeze pressures as measured manometrically were compared and a significant correlation between the two was observed. We did not measure the pudendo-anal reflex or anocutaneous reflex as these methods do not appear useful or reliable. 'I Pudendo-anal nerve conduction studies and transcutaneous spinal stimulation at the LI and LA levels were not undertaken although other studies indicate that these may be helpful in locating pudendal nerve and cauda equina lesion^.'^-^' Overall, EMG is a useful research tool in increasing our understanding of the mechanisms of incontinence. The fibre density recording is useful when a neurogenic cause is expected but it is not of great practical value in the routine investigation of incontinence. It may be used to establish whether neuropathy or obstetric sphincter trauma is the main factor causing incontinence in an individual patient. It has been suggested that widening of the anorectal angle at rest, abnormal perineal descent and reduction in the anococcygeal distance may be of importance in the pathogenesis of idiopathic faecal incontinence.22 We found that the mean value for the anorectal angle at rest was in the normal range but that there was extreme variability in the angle. Generally it is said that incontinence is associated with a resting anorectal angle in excess of 130".s.8 None of our patients had a resting angle in excess of 120". Interestingly, many had a reduction in the resting anorectal angle. The importance of the anorectal angle in the maintenance of continence is now
951
being questioned.*' Moreover, the Parks postanal repair for faecal incontinence improves the incontinence but does not change the anorectal angle.24 In this study the mean values for perineal descent at rest and at the point of evacuation were outside the limits of normality as described by Mahieu ef al. ,'.' but again there was extreme variability in the values so that it is difficult to interpret the significance of perineal descent in an individual case. As with the anorectal angle, measurement of perineal descent helps very little in the practical management of the patient. Defaecography is not essential before embarking on sphincter repair or postanal repair. It is of value in excluding rectal intussusception as a precursor of full thickness prolapse when this may be associated with faecal incontinence. The choices of treatment for the incontinent patient are limited, and the results of treatment often unccrtain. In general those with partial incontinence had reduced resting and squeeze pressures and were treated conservatively, but the decision to so treat was a clinical one. Those with total incontinence overall had lower resting and squeeze pressures, and were treated either conservatively or operatively, though this decision again was a clinical one. Sphincter repair was chosen when there was evidence of sphincter injury, though in many there was also evidence of a neurogenic abnormality on electromyographic testing. Those chosen for postanal repair had no history of sphincter trauma, had evidence of neurogenic damage on EMG and had pelvic floor descent and a wide anorectal angle on defaecography . In general, those with better sphincter pressures pre-operatively did better postoperatively whether following post-anal repair or sphincter repair, but the figures were far too small to be meaningful. Yoshioka ef al. have shown that low resting and squeeze pressure pre-operatively predict a poorer outcome following post-anal repair when compared with those with higher pre-operative pressures.24Occasionally anal manometry confirmed a normal sphincter pressure profile in patients claiming incontinence, but suspected clinically of psychological disturbance. Having reviewed this series of patients and their investigations we feel that laboratory investigations supplement a thorough clinical evaluation. They identify the nature of the abnormality rather than demonstrate the aetiology. They are not indicated for minor degrees of incontinence. They should be used selectively to get the best quantification of the defect when active surgical intervention is being contemplated. Anal manometry provides the best assessment of sphincter competence and the other tests should be used selectively in combination. Single fibre EMG with fibre density recording should be used when a neurogenic cause is suspected clinically. Conventional EMG recording is useful for
958
FINK E T A L .
mapping of the sphincter complex after obstetric and other sphincter trauma. Defaecography should be added when occult rectal prolapse is suspected. A combination of investigationsgives the best assessment of the factors involved but the choice of treatment is made on clinical grounds.
Acknowledgements The authors thank the members of the Colorectal Unit at St Vincent’s Hospital (Mr Peter Ryan, Mr Brian Collopy, Mr John Mackay and Mr Rodney Woods) for providing the vast majority of the patients studied. They also wish to thank Professor Richard Bennett for useful criticism in preparation of this paper.
References 1. PESCATORI M. & RAVO B. (1988) Diagnostic anorectal
2.
3.
4. 5.
6. 7.
8. 9. 10.
functional studies: Manometry, sphincter electromyography and defaecography . Surg. Clin. North Am. 68, 1231-48. FELT-BERSMA R. J . F. & MEUWISSEN S . G. M. (1990) Anal manometry. Int. J . Colorectal Dis. 5, 170-3. DUTHIE H. C. & BENNETT R. C. (1963) The relation of sensation in the anal canal to the functional anal sphincter: A possible factor in anal continence. Gut 4, 179-82. E. & TRONTEU J. V. (1979) Single Fibre STALBERG Electromyogruphy. Mirvalle Press, Old Woking, Surrey. MAHIEU P., PRINGOT J. & BODART P. (1984) Defaecography 1: Description of a new procedure and results in normal patients. Gastrointest. Radiol. 9, 247-5 1. PRESTON D.M., LENNARD-JONES J. E. & THOMAS B. M. (1984) The balloon proctogram. Br. J . Surg. 71, 2932. M. (1980) Increased motor NEILLM. E. & SWASH unit fibre density in the external anal sphincter muscle in ano-rectal incontinence: A single fibre EMG study. J . Neurol. Neurosurg. Psychiatry 43, 343-1. P. (1984) DefaeMAIIIEU P . , PRINGOT J. & BODART cography 11: Contribution to the diagnosis of defaecation disorders. Gastrointest. Radiol. 9, 253-61. FtLT-BERSMA R. J . F . , KLINKENBERG-KNOL E. c. & M~U W IS SSE. G. N M. (1988) Investigation of anorectal function. Br. J . Surg. 75, 53-5. HALLAN R. I . , MARZOUK D. E. M. M., WALDRON D. J . , WOMACK N. R. & W~LLIAMS N. S . (1989) Comparison of digital and manometric assessment of anal sphincter function. Br. J . Surg. 76, 973-5.
11. READN. W., BARTOLO D. C. C. & READM. G . (1984) Differences in anal function in patients with incontinence to solids and in patients with incontinence to liquids. Br. J . Surg. 71, 39-42. 12. FELT-BERSMA R. J. F., JANSSEN J . J . W. M ., KLINKENE. C., HOITSMA H. F. W. & MEUWISSEN S. BERG-KNOL G. M. (1989) Soiling: Anorectal function and results of treatment. Int. J. Colorectal Dis. 4, 37-40. 13. FERGUSON G. H ., REDFORD J., BARRET J. A. & KIFFE. S. (1989) The appreciation of rectal distension in fecal incontinence. Dis. Colon Rectum 32, 964-7. R. J. F., KLINKERNBERG-KNOL E. C . & 14. FELT-BERSMA MEUW~SSEN S . G . M. (1990) Anorectal function investigations in incontinent and continent patients. Dis. Colon Rectum 33, 479-86. S. J., BARNES P. R. H. & 15. H ~ N RM. Y M., SNOOKS SWASH M. (1985) Investigation of disorders of the anorectum and colon. Ann. R . Coll. Surg. EngI. 67, 355-60. M. (1985) S . J . , HENRY M . M. & SWASH 16. SNOOKS Faecal incontinence due to external anal sphincter divisions in childbirth is associated with damage to the innervation of the pelvic floor musculature: a double pathology. Br. J. Obster. Gynaecol. 92, 824-8. E. & MIHELIN M. (1990) 17. TRONTEU J. V . , STALBERG Jitter in the muscle fibre. J . Neurol. Neurosurg. Psychiatry 53, 49-54. 18. KEIGHLEYM.R.B.,HENRYM.M.,BARTOLOD.C.C. & MORTENSEN N. J. Mc. C. (1989) Anorectal physiology measurement. Report of a working party. Br. J . Surg. 76, 356-7. 19. KIFFE. S. & SWASH M. (1984) Normal proximal and delayed distal conduction in the pudendal nerve of patients with idiopathic (neurogenic) faecal incontinence. J . Neurol. Neurosurg. Psychiatry 47, 820-3. 20. SNOOKS S. J., SWASH M. & HENRY M. M. (1985) Abnormalities in central and peripheral conduction in patients with ano-rectal incontinence. J . R . Soc. Med. 78, 294-300. 21. SWASH M. & SNCOKS S. J. (1986) Slowed motor conduction in lumbosacral nerve roots in cauda equina lesions: a new diagnostic technique. J . Neurol. Neurosurg. Psychiatry 49, 808- 16. 22. HARDCASTLE J. D. & PARKS A. G. (1970) A study of anal incontinence and some principles of surgical treatment. Proc. R . Soc. Med. 63, 1 16- 18. J. J., GIBBONS C . & READN. W. (1987) 23. BANNISTER Preservation of faecal continence during rises in intra-abdominal pressure: Is there a role for the flap valve? Gut 28, 1242-5. 24. YOSHIOKA K., HYLAND G. & KEICHLEY M. R. B. ( 1988) Physiological changes after postanal repair and parameters predicting outcome. Br. J . Surg. 75, 1220-4.
A u s r . N . Z . J . Surg. 1992,62.951-958
THE ROLE OF MANOMETRY, ELECTROMYOGRAPHY AND RADIOLOGY IN THE ASSESSMENT OF FAECAL INCONTINENCE ROY L. FINK,*LESLIEJ. ROBERTS+A N D MARKSCOTT* *University Department of Surgery and Departments of .‘Neurology and ‘Radiology, St Vincent’s Hospital. Fitzroy, Victoria, Australia The results of laboratory investigations in 156 patients presenting with faecal incontinence are reviewed to see if and how these investigations supplement a careful clinical evaluation, and in particular to see if they help in the practical management of the problem. All patients underwent anal manometry, and in addition 52 underwent anal sphincter electromyography and 27 defaecatory proctography. Anal manometry quantified sphincteric weakness, and proved superior to digital assessment in this regard. Resting and squeeze pressures were less in those with complete than those with partial incontinence but the differences were not statistically significant. The measurements of rectal sensation and compliance were not additionally helpful. Single fibre electromyography provided the best measure of denervation with re-innervation and was abnormal in about 85”& of the group studied. Jitter studies were unhelpful. Most patients had some abnormality on defaecatory proctography but clinical significance could not be established. The choice of treatment was made on clinical grounds and was not influenced by these investigations.
Key words: electromyography, faecal incontinence, manometry.
Introduction In recent years there has been a dramatic increase of interest in anorectal functional abnormalities and related pathophysiology of the pelvic floor. While there has been an increase in the number and use of laboratory techniques to assess these abnormalities, the role of these techniques is not yet clearly defined. The results of tests of pelvic floor function can only be used to supplement a careful clinical assessment and a combination of tests allows the best evaluation of the situation. Manometry, electromyography (EMG) and defaecatory proctography are finding their way into routine clinical practice. The purpose of this study was to review experience with these investigations at St Vincent’s Hospital, and to evaluate their contribution to the assessment of faecal incontinence.
Methods During the period 1984-90, 310 patients with a variety of conditions were referred for anorectal manometry. Many also underwent anal sphincter electromyography and/or defaecatory proctography .
Correspondence: Mr R. L. W . Fink, 55 Victoria Parade, Fitrroy. Vic. 306.5, Australia. Accepted for publication 5 August 1992
By far the most common referrals were for patients with a degree of faecal incontinence (156) or constipation (98). Less common referrals were for the assessment of function in association with rectal prolapse (15), ileo anal pouches (13) and patients with anal pain (13). The presumed aetiology of the incontinence based on clinical assessment in the 156 patients studied is shown in Table 1. Most fell into the ‘idiopathic’ group. One hundred and twelve (710/0) were female and most were elderly, especially in the idiopathic group, where the mean age was 57 years. Most had minor incontinence. Ninetyone (58%) had incontinence of flatus and liquid stool only, indicating partial incontinence, whereas 65 (42%) had varying degrees of incontinence to solid stool, but regular whole stool accidents were rare.
Table 1. Patients investigated with incontinence: presumed aetiology ‘Idiopathic’
98
Traumatic
45
Surgical Obstetric Injury DXRT
Neurogenic Prolapse Scleroderma Total
5 5
3 I56
26 14
3 2
FINK E T A L
052
A detailed history was taken to determine the aetiology, nature and severity of the incontinence. Rectal examination allowed a clinical assessment of resting pressure, and squeeze pressure with voluntary contraction, both of which were clinically graded (normal, somewhat reduced, markedly reduced). Perianal appreciation of sensation to light touch and pin prick was also tested, along with the reflex contraction of the external sphincter in response to sharp stimulation. All of the patients underwent anorectal manometry, seven patients having two or more investigations. In addition pelvic floor EMG was carried out in 52 of the patients and defaecatory proctography in 27. The reason for the choice of investigations by the referring practitioner in an individual patient could not always be determined by the authors retrospectively. ANAL MANOMETRY
In order to ensure an empty rectum, the patient was instructed to use a Durolax suppository the night before and the morning of the examination. Anal manometry was performed with the patient in the left lateral position. A water perfused open ended low compliance catheter (internal diameter 2 mm) was linked with pressure transducers (Bell & Howell 4/327/I; range 0-400mmHg) and a chart recorder (Neotrace NT 218). The rate of perfusion was 0.5 mUmin. The catheter, which was marked at 1 cm intervals, was introduced into the mid-rectum through a sigmoidoscope and the sigmoidoscope was then removed. Pressures were recorded by a station pull-through technique at I cm intervals, the height being measured from the anal verge. The catheter remained at each station for sufficient time to allow resting pressure to stabilize and then the pressure with maximum voluntary contraction (squeeze pressure) was recorded, the level being recorded in centimetres from the anal verge. The maximum resting pressure and maximum squeeze pressure were thus determined and the levels at which these were identified were noted (Fig. 1). Rectal sensation was assessed by introducing into the rectal ampulla a soft compliant balloon attached to a similar catheter. The balloon was inflated with air at a rate of 1 mWs. The volume at which first appreciation of the balloon occurred (least perceived volume) and the volume at which severe discomfort occurred (maximum tolerated volume) were both recorded. Rectal compliance was measured by the ratio of the maximum tolerated volume to the pressure within the balloon at that point, as recommended by Pescatori and Ravo. The recto-anal inhibitory reflex was tested by rapid inflation of the rectal balloon (50 mL/s) and noting any drop in resting anal canal pressure as recorded by a second intra-anal manom-
'
eter placed at the level of the previously identified maximum resting pressure (Fig. I). Results were compared to normal values reported in the literature,* especially where exactly similar technique and apparatus had been employed. ELECTROMYOGRAPHY
Electromyography (EMG) recordings were made using a Medelec MS6 EMG machine. A Medelec FCD 37 concentric needle electrode was used for the concentric needle recordings which were made in the external anal sphincter and in most cases the puborectalis muscle as well. Details of the activity seen on needle insertion, at rest, with maximal voluntary contraction, coughing and on voluntary bearing down and relaxing the sphincter muscle were recorded in all patients and in most patients photographic record of this activity was made. Concentric needle electrode recordings were made from the posterior and lateral parts of the external anal sphincter. After digital examination the needle was advanced into the puborectalis muscle where similar recordings were made. A standard Medelec single fibre needle (SF37) with a 25 micron recording surface was used for the single fibre EMG studies. Single fibre EMG recording was performed in the external anal sphincter and measurements of jitter and fibre density were made.4 The single fibre EMG needle was inserted percutaneously in the mid-line posterior to the anal verge. Measurements and recordings were then made from the lateral and posterior parts of the external anal sphincter. The recordings were made during resting activity, although very occasionally voluntary activation was required or a Foley catheter was inserted, the balloon inflated, and a small weight was attached to pull against the anal sphincter. Action potentials of greater than 150yV were accepted and these triggered the sweep of the machine. They were then delayed so they could be visualized. At each recording site one or more action potentials arising from individual muscle fibres were seen. The number of potentials which were linked together was recorded at each site. The fibre density was defined as the mean number of units or potentials seen per insertion from 20 different locations in the external anal sphincter. When simultaneous recording is made of the action potentials of two or more muscle fibres supplied by branches of the same axon, the interval of time between the potentials varies only slightly under normal circumstances. This variability is termed jitter (Fig. 2). When two or more linked potentials were found and when the recording appeared stable, the traces were digitized and transferred on line by a Medelec C16 interface to an IBM compatible PC for storage
INVESTIGATION OF FAECAL INCONTINENCE
Distance from anal verge (cm)
L E
sigmoid colon was outlined. A single spot film was taken at this time to document the direction of the anal canal. The rectal tube was carefully removed. The patient was then positioned on a specially constructed commode for imaging purposes. Both static and dynamic imaging was performed with specific observations being made at rest, with 'gripping' (the patient was instructed how to hold on tightly, thereby attempting to increase the tone of the puborectalis sling), while straining and while evacuating. The patient was viewed in the lateral position. The films were reviewed at the completion of the procedure and if satisfactory, the patient was allowed to leave. The examination usually took approximately 15 min. Measurements were made as follows: (i) pelvic descent (i.e. the distance of the anorectal junction below the line joining the tip of the coccyx and the inferior margin of the pubic symphysis), this is normally < 2.5 cm;5 (ii, the anorectal angle, this is normally 92" (rt 1.5) at rest and 137" (+ 1.5) while strainin& (iii) the distance from the anorectal junction to the lower end of the coccyx (6.4 k 0.5 cm in incontinent patients vs 11.0 k 0.4cm in constipated patients).6 Other observations included evidence of intussusception, evidence of rectal prolapse, and any other abnormal configuration of the rectal wall. The patients were treated by a number of surgeons and by a variety of methods (Table 2), and the patients' records were reviewed to see how the investigations added to the clinical assessment and influenced the choice of treatment.
50k
100.
n -
953
i
4
t
I
1
D
Recto-anal inhibitory reflex
Fig. 1. Anal manometry. Normal tracing showing (a) resting pressures and squeeze pressures resulting from voluntary contractionof sphincters (S) as catheter is withdrawn through the lower 4cm of the anal canal; and (b) normal recto-anal reflex with progressive drop in anal canal pressure with rectal balloon distension (1) and rise with balloon deflation (D).
STATISTICAL METHODS
Fig. 2. Single fibre EMG recording. This shows four traces superimposed.The first unit is the triggering potential. The variation in time between the first and second units constitutesjitter.
and subsequent analysis. Measurement of the jitter (expressed as the mean consecutive difference) was made from at least 10 different potential pairs. Up to 20 recordings were made for each pair. DEFAECATORY PROCTOGRAPHY
The examination was clearly explained to the patient before commencement of the procedure to ensure optimal co-operation. No bowel preparation was used. A standard rectal tube was placed in the rectum under fluoroscopic screening with the patient in the left lateral position. The position of the tip of the tube was confirmed fluoroscopically before a volume of thick barium was syringed into the rectum under screening, which was continued until the
Simple Student's 1-tests were used to compare results between those with partial and complete incontinence, and between groupings separated by digital assessment. The relationship between squeeze pressures and fibre density on electromyography was assessed by using a Pearson correlation coefficient.
Results A N A L MANOMETRY
The results of anal manometry are shown in Table 3 which details results in the group as a whole and then is divided into those with complete and partial incontinence as defined previously. Both resting pressures and squeeze pressures in this group of incontinent patients were lower than normal values quoted in the l i t e r a t ~ r e . While ~ . ~ resting pressures and squeeze pressures were lower in those with complete rather than partial incontinence the differences were not statistically significant. There was also no significant difference in rectal sensation between those with complete and partial incontinence.
FINK E T A L .
954
Table 2. Incontinence: treatment -~
Operative
Conservative
Sphincter repair Post-anal repair Rectopexy Colectomy Colostomy Delorme
70 20 7 5 2 I
Physiotherapy Constipating agents Banding mucosa Enemas Stop medication Biofeedback
8 5 4 2 2 I
Ten patients had no treatment, and the type of treatment was unknown in 33 patients.
The recto-anal reflex was absent in 21.6% of the patients and in those with an absent reflex the mean maximum resting pressure was 20.1 t 7.8 mmHg as compared with 28.2 k 13.2mmHg in those with a reflex ( P 4 0 . 4 p s and/or if there were two or more pairs of 10 which showed jitter in excess of 5 5 p . The correlation between squeeze pressures measured manometrically and fibre density on single fibre studies is shown in Fig. 5 . The correlation was significant ( r = - 0.38; P = 0.019). The data were fitted with a linear regression which showed a drop in squeeze pressure from 48 mmHg to 36 mmHg (i.e. 25%) for a rise in fibre density from 2 to 3.
Table 3. Incontinence: manometry T o t a l group ~
~
~
Pressures (mmHp) MRP MSP Rectal sensation (ccs)
~
__ _ _
( n = 156)
Partial incontinence Complete incontience ( n = 91) ( 1 1 = 65) ~ _ _ _ _ _ __ _ ~~
~~
~
24.7 ? 13 44.9 f 20.4
26.0 I 13.5 47.1 2 20. I
22.6 k 12.4 41.7 5 20.5
LPV 87.1 f 52.4 MTV 210.4 2 79.7 Recto-anal reflex present 109, absent 39 Rectal compliance 2.74 2 1.45 (ccs/mmHg)
85.4 I 45.1 202.9 14.0
89.5 k 61.4 221.3 f 85.9
*
~
~
Results arc expressed as means and standard deviations. MRP = maximum resting pressure; MSP = maximum squeeze pressure: LPV = least perceived volumc; MTV = maximum tolerated volume.
955
1NVESTIGATION OF FAECAL INCONTINENCE
Table 4. Incontinence: electrornyography concentric needle electrode recording
110r
100 -
__ ++
90-
External anal sphincter (n Overall activity Polyphasia Anismus Puborectalis ( n = 34) Overall activity Polyphasia Anismus
80.
- ror
+++
0
r
-m
E E 60-
2
m
Abnormal
Markedly abnormal
31 21 49
13 28 6
2 3 0
27
6 23 3
1
Normal = 52)
10
31
1
0
50-
0
40
Overall activity refers to the assessment of insertional, resting and maximal activity and the response to coughing. Normal polyphasia and anismus refer$ to none being present.
-
Table 5. Incontinence: single fibre electromyography external anal sphincter ( n = 39) No. patients
C
N
R
Resting pressure
P
N
R
P
Squeeze pressure
Fig. 3. Digital assessment of sphincters. Comparison of manometrically assessed pressures with digitally assessed pressure groupings (normal (N), reduced (R) and poor (P)). #Not significant; # # P < 0.001; # # # P < 0.01.
Fibre density < 1.98 > 1.98 Mean Standard deviation Jitter Normal Abnormal
6 33 2.46 0.44 21 18
D EFA ECA TO RY PROCTOGRAPHY
Discussion
The findings on defaecatory proctography are summarized in Table 6. Normal values are those defined and Preston eta1.6It can be seen by Mahieu et al.5*8 that there was wide variation in all of the values looked at and most patients had some abnormality, but these alone may have been insufficient to cause the incontinence.
Laboratory investigations have undoubtedly improved our understanding of the mechanism of faecal continence and causes of incontinence, especially so called idiopathic incontinence. Our aim was to see how these investigations help in the practical management of the condition. The study represents a large experience during several years
Fig. 4. (a) Normal motor unit action potential. (b) Polyphasic unit with multiple changes of phase.
956
FINK ETAL.
80 r
resting pressure within the group. Resting pressures are reported as being very variable in most studies2 and partially relate to age and sex differences. As E 60E might be expected the resting pressure in those with E 503 complete incontinence was lower than in those with 2 40partial incontinence, though the difference was not E =30statistically significant. Because the external sphincN ter is thought to be concerned with maintenance of :2 0 = 10continence to solid stool, and because its contribuIU n tion is largely responsible for squeeze pressure, it 0 : : might be expected that squeeze pressures would be 1 15 2 25 3 35 4 significantly lower in those with complete as opFibre density posed to partial incontinence. Read et al. showed Fig. 5. Correlation between squeeze pressure and fibre such a difference but we were unable to confirm density on single fibre EMB ( r = -0.38; P = 0.019). this." Rectal sensation as measured by the least perceived volume and maximum tolerated volume on Table 6. Incontinence: defaecatory proctography distension with an intra-rectal balloon showed a ~wide range of values in the incontinent patients. Anorectal angle at Similarly other studies have shown a wide range of 91.6 f 12.1 ( n = 9 2 f 1.5) rest (") values for rectal sensation, not only in incontinent Anorectal angle at evacuation (") 115.6 f 25.3 ( n = 137 _+ 1.5) patients but also in normal individuals. "-I4 For this Perineal descent at reason such measurements are not useful in the rest (cm) 2.75 k 2.0 ( n = < 2.0) assessment of patients with faecal incontinence. The Perineal descent at recto-anal inhibitory reflex was absent in 36% of 5.32 t 3.0 ( n = < 4.0) evacuation (cm) the present patients, and resting pressure in these Anococcygeal patients was significantly lower than in those with a distance (cm) 6.6 ? 2.6 reflex. These patients presumably had severe weakness of the internal anal sphincter. Theoretically if Results are expressed as means and standard deviations. the volume required to elicit the reflex was less than the least perceived volume then this could be a factor in incontinence but this was not found in any of the present patients. Rectal compliance was widely variable and aland multiple investigations were frequently perthough a group could be identified in whom lowformed with the expectation that this might give the ered rectal compliance might have contributed to best assessment. the incontinence, sphincter pressures were also low, Clinical evaluation allows, in most cases an adso that a non-compliant rectum could not be regarded equate assessment of the likely cause and degree of as a prime cause. The reason for a non-compliant incontinence. Experienced clinicians often express rectum was not always obvious, but in two it relatthe view that digital assessment of the anal sphinced to previous pelvic irradiation, in one to fibrosis ters is adequate for practical purposes. Indeed, in following trauma and in another to scleroderma. this study, digital assessment did allow satisfactory In summary, manometry alone did not elucidate separation of patients into broad groups with northe cause of incontinence but it was able to quantify mal, reduced and very reduced resting and squeeze the sphincteric component, and this was superior to pressures. It was somewhat easier to assess resting digital assessment. The measurement of rectal sentone than voluntary contraction, especially when sation, the recto-anal inhibitory reflex and rectal the latter was somewhat reduced. It is also noted compliance added little to the assessment. that in individual instances the digital assessment of The role of EMG is to document a neurogenic both resting and squeeze pressures sometimes was basis for incontinence and to quantify this. Concenquite inaccurate. Therefore, while digital assessment tric needle electrode recording showed a normal overis usually accurate there is no doubt that manornetall pattern of activity in the majority of patients. On ric assessment is more reliable. Others have shown the other hand, chronic changes of denervation with a moderate to good correlation between digitally re-innervation (i.e. polyphasia and increased and manometrically measured anal p r e s s ~ r e . ~ ~ partial '~ fibre density) were seen in the vast majority of The present study demonstrated low resting pressures in the group of incontinent patients as compatients. The presence of excess polyphasia and pared with normal values quoted in the literature,*s3 increased fibre density correlated very well. If the fibre density was > 2 then polyphasia was seen in though there was wide variation in the value of 8
r
-
70-
..
.. . .
-
INVESTIGATION OF FAECAL INCONTINENCE
all except one patient. Fibre density recording is favoured because it is a more objective test. Others have shown it to be more sensitive than conventional EMG" and have indicated a lack of overlap between the fibre density of denervated and normal sphincters. Concentric needle EMG therefore adds little to the assessment, although after traumatic damage to the sphincters, it is useful for mapping of sphincter muscle activity. I 5 , l 6 Single fibre EMG has demonstrated evidence of denervation in 85% of the patients investigated, confirming that pudendal nerve neuropathy of varying degrees is a common finding in most cases of incontinence. Jitter was abnormal in about half of the patients but there was no correlation with fibre density or degree of polyphasia. 'Abnormalities' in jitter do not necessarily indicate pathology. The lack of correlation of increased jitter with any other parameters measured suggests that this finding is not significant. Variations of the rates of firing of the action potentials may alter the propagation velocity of action potentials along muscle fibre membranes and thereby affect jitter in the absence of disease.17 This seems a likely explanation of the increased jitter seen and it was concluded that jitter measurement is not useful in this clinical setting. The degree of denervation as measured by fibre density and the maximum squeeze pressures as measured manometrically were compared and a significant correlation between the two was observed. We did not measure the pudendo-anal reflex or anocutaneous reflex as these methods do not appear useful or reliable. 'I Pudendo-anal nerve conduction studies and transcutaneous spinal stimulation at the LI and LA levels were not undertaken although other studies indicate that these may be helpful in locating pudendal nerve and cauda equina lesion^.'^-^' Overall, EMG is a useful research tool in increasing our understanding of the mechanisms of incontinence. The fibre density recording is useful when a neurogenic cause is expected but it is not of great practical value in the routine investigation of incontinence. It may be used to establish whether neuropathy or obstetric sphincter trauma is the main factor causing incontinence in an individual patient. It has been suggested that widening of the anorectal angle at rest, abnormal perineal descent and reduction in the anococcygeal distance may be of importance in the pathogenesis of idiopathic faecal incontinence.22 We found that the mean value for the anorectal angle at rest was in the normal range but that there was extreme variability in the angle. Generally it is said that incontinence is associated with a resting anorectal angle in excess of 130".s.8 None of our patients had a resting angle in excess of 120". Interestingly, many had a reduction in the resting anorectal angle. The importance of the anorectal angle in the maintenance of continence is now
951
being questioned.*' Moreover, the Parks postanal repair for faecal incontinence improves the incontinence but does not change the anorectal angle.24 In this study the mean values for perineal descent at rest and at the point of evacuation were outside the limits of normality as described by Mahieu ef al. ,'.' but again there was extreme variability in the values so that it is difficult to interpret the significance of perineal descent in an individual case. As with the anorectal angle, measurement of perineal descent helps very little in the practical management of the patient. Defaecography is not essential before embarking on sphincter repair or postanal repair. It is of value in excluding rectal intussusception as a precursor of full thickness prolapse when this may be associated with faecal incontinence. The choices of treatment for the incontinent patient are limited, and the results of treatment often unccrtain. In general those with partial incontinence had reduced resting and squeeze pressures and were treated conservatively, but the decision to so treat was a clinical one. Those with total incontinence overall had lower resting and squeeze pressures, and were treated either conservatively or operatively, though this decision again was a clinical one. Sphincter repair was chosen when there was evidence of sphincter injury, though in many there was also evidence of a neurogenic abnormality on electromyographic testing. Those chosen for postanal repair had no history of sphincter trauma, had evidence of neurogenic damage on EMG and had pelvic floor descent and a wide anorectal angle on defaecography . In general, those with better sphincter pressures pre-operatively did better postoperatively whether following post-anal repair or sphincter repair, but the figures were far too small to be meaningful. Yoshioka ef al. have shown that low resting and squeeze pressure pre-operatively predict a poorer outcome following post-anal repair when compared with those with higher pre-operative pressures.24Occasionally anal manometry confirmed a normal sphincter pressure profile in patients claiming incontinence, but suspected clinically of psychological disturbance. Having reviewed this series of patients and their investigations we feel that laboratory investigations supplement a thorough clinical evaluation. They identify the nature of the abnormality rather than demonstrate the aetiology. They are not indicated for minor degrees of incontinence. They should be used selectively to get the best quantification of the defect when active surgical intervention is being contemplated. Anal manometry provides the best assessment of sphincter competence and the other tests should be used selectively in combination. Single fibre EMG with fibre density recording should be used when a neurogenic cause is suspected clinically. Conventional EMG recording is useful for
958
FINK E T A L .
mapping of the sphincter complex after obstetric and other sphincter trauma. Defaecography should be added when occult rectal prolapse is suspected. A combination of investigationsgives the best assessment of the factors involved but the choice of treatment is made on clinical grounds.
Acknowledgements The authors thank the members of the Colorectal Unit at St Vincent’s Hospital (Mr Peter Ryan, Mr Brian Collopy, Mr John Mackay and Mr Rodney Woods) for providing the vast majority of the patients studied. They also wish to thank Professor Richard Bennett for useful criticism in preparation of this paper.
References 1. PESCATORI M. & RAVO B. (1988) Diagnostic anorectal
2.
3.
4. 5.
6. 7.
8. 9. 10.
functional studies: Manometry, sphincter electromyography and defaecography . Surg. Clin. North Am. 68, 1231-48. FELT-BERSMA R. J . F. & MEUWISSEN S . G. M. (1990) Anal manometry. Int. J . Colorectal Dis. 5, 170-3. DUTHIE H. C. & BENNETT R. C. (1963) The relation of sensation in the anal canal to the functional anal sphincter: A possible factor in anal continence. Gut 4, 179-82. E. & TRONTEU J. V. (1979) Single Fibre STALBERG Electromyogruphy. Mirvalle Press, Old Woking, Surrey. MAHIEU P., PRINGOT J. & BODART P. (1984) Defaecography 1: Description of a new procedure and results in normal patients. Gastrointest. Radiol. 9, 247-5 1. PRESTON D.M., LENNARD-JONES J. E. & THOMAS B. M. (1984) The balloon proctogram. Br. J . Surg. 71, 2932. M. (1980) Increased motor NEILLM. E. & SWASH unit fibre density in the external anal sphincter muscle in ano-rectal incontinence: A single fibre EMG study. J . Neurol. Neurosurg. Psychiatry 43, 343-1. P. (1984) DefaeMAIIIEU P . , PRINGOT J. & BODART cography 11: Contribution to the diagnosis of defaecation disorders. Gastrointest. Radiol. 9, 253-61. FtLT-BERSMA R. J . F . , KLINKENBERG-KNOL E. c. & M~U W IS SSE. G. N M. (1988) Investigation of anorectal function. Br. J . Surg. 75, 53-5. HALLAN R. I . , MARZOUK D. E. M. M., WALDRON D. J . , WOMACK N. R. & W~LLIAMS N. S . (1989) Comparison of digital and manometric assessment of anal sphincter function. Br. J . Surg. 76, 973-5.
11. READN. W., BARTOLO D. C. C. & READM. G . (1984) Differences in anal function in patients with incontinence to solids and in patients with incontinence to liquids. Br. J . Surg. 71, 39-42. 12. FELT-BERSMA R. J. F., JANSSEN J . J . W. M ., KLINKENE. C., HOITSMA H. F. W. & MEUWISSEN S. BERG-KNOL G. M. (1989) Soiling: Anorectal function and results of treatment. Int. J. Colorectal Dis. 4, 37-40. 13. FERGUSON G. H ., REDFORD J., BARRET J. A. & KIFFE. S. (1989) The appreciation of rectal distension in fecal incontinence. Dis. Colon Rectum 32, 964-7. R. J. F., KLINKERNBERG-KNOL E. C . & 14. FELT-BERSMA MEUW~SSEN S . G . M. (1990) Anorectal function investigations in incontinent and continent patients. Dis. Colon Rectum 33, 479-86. S. J., BARNES P. R. H. & 15. H ~ N RM. Y M., SNOOKS SWASH M. (1985) Investigation of disorders of the anorectum and colon. Ann. R . Coll. Surg. EngI. 67, 355-60. M. (1985) S . J . , HENRY M . M. & SWASH 16. SNOOKS Faecal incontinence due to external anal sphincter divisions in childbirth is associated with damage to the innervation of the pelvic floor musculature: a double pathology. Br. J. Obster. Gynaecol. 92, 824-8. E. & MIHELIN M. (1990) 17. TRONTEU J. V . , STALBERG Jitter in the muscle fibre. J . Neurol. Neurosurg. Psychiatry 53, 49-54. 18. KEIGHLEYM.R.B.,HENRYM.M.,BARTOLOD.C.C. & MORTENSEN N. J. Mc. C. (1989) Anorectal physiology measurement. Report of a working party. Br. J . Surg. 76, 356-7. 19. KIFFE. S. & SWASH M. (1984) Normal proximal and delayed distal conduction in the pudendal nerve of patients with idiopathic (neurogenic) faecal incontinence. J . Neurol. Neurosurg. Psychiatry 47, 820-3. 20. SNOOKS S. J., SWASH M. & HENRY M. M. (1985) Abnormalities in central and peripheral conduction in patients with ano-rectal incontinence. J . R . Soc. Med. 78, 294-300. 21. SWASH M. & SNCOKS S. J. (1986) Slowed motor conduction in lumbosacral nerve roots in cauda equina lesions: a new diagnostic technique. J . Neurol. Neurosurg. Psychiatry 49, 808- 16. 22. HARDCASTLE J. D. & PARKS A. G. (1970) A study of anal incontinence and some principles of surgical treatment. Proc. R . Soc. Med. 63, 1 16- 18. J. J., GIBBONS C . & READN. W. (1987) 23. BANNISTER Preservation of faecal continence during rises in intra-abdominal pressure: Is there a role for the flap valve? Gut 28, 1242-5. 24. YOSHIOKA K., HYLAND G. & KEICHLEY M. R. B. ( 1988) Physiological changes after postanal repair and parameters predicting outcome. Br. J . Surg. 75, 1220-4.
