SHORT REPORT
ABSTRACT: Severe chronic constipation has been implicated as a cause
of damage to the pelvic floor innervation. The aim of the present study was
to examine the role of mild to moderate chronic constipation, a condition
more relevant for clinical electromyographers, because this complaint is
common in patients sent for evaluation of possible neurogenic dysfunction of
lower sacral myotomes. A group of 59 subjects without major uroneurological dysfunction, proctological disorders, or neurological abnormalities participated in the study, which involved concentric needle electromyography of
the external anal sphincter (EAS). Motor unit potentials (MUPs; sampled
using multi-MUP analysis) and interference pattern (IP, sampled using turn/
amplitude analysis) of chronically constipated and control subjects were
compared. No effect of chronic constipation on MUP/IP parameters compatible with neurogenic injury was found. Our results suggest that mild
chronic constipation does not cause damage to the EAS innervation, and
that no separate reference values are needed for this group of subjects.
© 2000 John Wiley & Sons, Inc. Muscle Nerve 23: 1748–1751, 2000
STANDARDIZATION OF ANAL SPHINCTER
ELECTROMYOGRAPHY: EFFECT OF
CHRONIC CONSTIPATION
SIMON PODNAR, MD, MSc, and DAVID B. VODUŠEK, MD, DSc
Division of Neurology, Institute of Clinical Neurophysiology, University Medical
Center Ljubljana, SI-1525 Ljubljana, Slovenia
Accepted 15 May 2000
The electromyographic (EMG) examination of the
external anal sphincter (EAS) muscle may be helpful
in diagnosing the extent of neurogenic involvement
of the lower sacral myotomes. In establishing the
relevance of mild to moderate EMG abnormalities
in the EAS muscle of patients, a history of vaginal
delivery10 or chronic constipation is commonly
assumed to represent a possible alternative explanation for them, as these conditions have been associated with neuropathic changes in the EAS
muscle.2,5,14 In the case of chronic constipation,
some studies have indeed revealed substantial damage to the pelvic floor innervation,5,14 whereas others failed to provide similar evidence of denervation.18,19 Most of these studies were performed in
groups of severely constipated patients, using techniques prone to examiners’ bias. Whether lesser de-
Abbreviations: DAT, digital audio tape; EAS, external anal sphincter;
EMG, electromyography; IP, interference pattern; MUPs, motor unit potentials
Key words: constipation; external anal sphincter muscle; interference
pattern; motor unit potentials; needle electromyography; pelvic floor; standardization
Correspondence to: S. Podnar; e-mail: simon.podnar@ kclj.si
© 2000 John Wiley & Sons, Inc.
1748
Short Reports
grees of constipation, such as commonly found
within the general population,4 cause such an injury
has not been addressed. This issue is important for
the determination of an adequate population of subjects when compiling reference values for “normal
subjects” (normative data): should the data obtained
in individuals known to suffer from chronic constipation be discarded? We addressed this question by
using two techniques of quantitative concentric
needle EMG, minimally open to bias, and by masking group comparison. The techniques were applied
to the EAS muscle of chronically constipated volunteers and nonconstipated controls.
METHODS
A group of 59 subjects, 41 women of varying parity
(0–4) and 18 men, participated in the study. Their
mean age was 46 years (range, 19–83 years). Examinees were nonselectively recruited from hospital patients without major proctological or uroneurological disorders, and without medical conditions known
to lead to secondary constipation (such as previous
pelvic floor surgery or psychiatric disorders), who
volunteered for the study. Histories regarding their
MUSCLE & NERVE
November 2000
bowel, urinary, and sexual function were evaluated
following our standard questionnaire. The presence
of constipation was established according to the subjects’ affirmative responses to at least two of the following questions on frequency of defecation (less
than three times per week),4 subjective evaluation of
completeness of rectal emptying (incomplete) and
possible constipation (constipated: yes), and description of feces (hard and dry).1,15 The median number
of bowel movements in our group of constipated
patients was two per week. As they did not seek medical attention because of this, no objective evaluation
of their constipation was attempted. All subjects gave
written informed consent. The study was approved
by the National Ethics Committee. Gynecological examination was performed in women, and only those
without genitourinary prolapse were included. Similarly, in all subjects, neurological examination of the
trunk and lower limbs was performed and only those
without neurological abnormalities were included.
A 37-mm-long standard concentric EMG needle
(No. 22583, Vickers Medical, Medelec, Old Woking,
UK) and electromyography system (Keypoint, Dantec Medical, Skovlunde, Denmark) with standard settings (filters: 5 HZ–10 kHZ) were used as described
elsewhere.12 The standardized electrode placement
in the subcutaneous and the deeper parts of EAS
were also as described.11 Several consecutive periods
of EMG activity were taped by a DAT (digital audio
tape) recorder (DCD-D8, Sony, Tokyo, Japan). As no
difference in motor unit potential/interference pattern (MUP/IP) analysis was found between the subcutaneous and the deeper parts of the EAS muscle,13
both parts of the EAS muscle were evaluated together.
At slight voluntary or reflex activation about 20
different MUPs were obtained from each muscle by
multi-MUP analysis,17 as described before.10,12,13
Amplitude, duration, area, number of phases and
turns, rise time and duration of negative peak, mean
frequency of firing and thickness (thickness = area/
amplitude) were parameters measured or calculated
by the electromyography system.10,12,13
In 53 subjects (40 women, 13 men) at different
levels of voluntary (modest to maximal) and reflex
activation by coughing, 20 different IPs were obtained from each muscle and sampled by “turn/
amplitude” analysis,8,9,16 as previously described.10,13
IP parameters evaluated were: number of turns/
second, amplitude/turn,16 percent activity, number
of short segments, and envelope.8,9 Only samples
with sufficient EAS muscle contraction to obtain
positive values of all IP parameters were included.10,13
Short Reports
MUPs and IP samples were separated into constipated and nonconstipated pools. Because our analyses revealed effect of gender on MUP and IP parameters, and just a few men were constipated (see
results), only pools of females’ MUPs and IPs were
compared by the Mann–Whitney U test.
Association of chronic constipation and MUP/IP
parameters was studied in pools of MUPs and IPs
obtained in both men and women, using the SAS
statistical system (SAS Institute Inc., Cary, North
Carolina) to perform multiple linear regression
analysis. Age and gender were included in this analysis. Additional analysis was performed on pools of
exclusively females’ MUPs/IPs to add parity, childbirth characteristics (weight of the heaviest newborn, history of mediolateral episiotomy or perineal
tear), time elapsed since last delivery, and presence/
absence of slight urinary stress incontinence,10 to
age and presence of chronic constipation as factors
included in the analysis.
RESULTS
Twenty-one subjects (17 of 41 women, and 4 of 18
men) claimed to be chronically constipated according to our criteria. Pools of 784 MUPs/1,242 IPs of
the constipated and 1,069 MUPs/1,321 IPs from the
nonconstipated women, and of 180 MUPs/219 IPs of
the constipated and 533 MUPs/575 IPs from the
nonconstipated men were compiled. Using the Mann–
Whitney U test, only the firing frequency of MUPs
was significantly higher in the control group of
women (P < 0.01). None of the other MUP parameters or any of the IP parameters differed significantly between the two groups. Using multiple linear
regression analysis, no significant effect of constipation on any MUP parameter was found in the pool of
women, as well as the common pool (men and
women combined). Significant effect of constipation
on number of turns/second, percent activity, and
number of short segment parameters (increase, P <
0.01) was found in a pool of women’s IPs (parity and
characteristics of childbirth included into analysis).
In mixed (men and women) IP pool analysis, number of turns/second, and number of short segments
were on the limit of significance (0.01 < P < 0.02).
DISCUSSION
Constipation is a heterogeneous syndrome,1,19 quite
common in the general population if loose diagnostic criteria are applied (17.5%),4 but rarer with more
stringent criteria (2%).6,15
In chronically constipated patients, substantial
damage to the peripheral nerves innervating the pel-
MUSCLE & NERVE
November 2000
1749
vic floor was reported previously in some studies,5,7,14 but not others.18,19 In our study, only the
mean frequency of MUP firing was different
(higher) in the control, as compared to the constipated group of women. This finding— unconnected
as it was to any changes in other MUP parameters—
does not indicate neurogenic injury to the EAS
muscle. It rather reflects different levels of activation
during MUP sampling, possibly due to the difference
in afferent input from a chronically distended bowel.
Constipation affected (increased) values of IP parameters related to density of IP (number of turns/
second, percent activity, and number of short segments). This points to stronger voluntary and reflex
activation by constipated patients. Again no difference in IP parameters indicating neurogenic injury
(amplitude/turn, envelope) were revealed. Differences found would only cause shift of IP dots within
the normative “cloud” upward (increase in number
of short segments) and to the right (increase in
number of turns/second and percent activity), with
no need for separate normative “clouds.”9,16
Differences between the results of our study and
several previous studies may be caused by less severe
constipation in our patients; in previous studies,5,14,18,19 medical attention for constipation was required whereas, in ours, it was an exclusion criterion.
Some previous studies, applying “comparable”
electrophysiologic techniques—single fiber EMG
measurements of fiber density,5,14 and semiquantitative18,19 or “manual-MUP” quantitative3 concentric
needle EMG—have found EAS EMG changes in constipated subjects. One of the unresolved problems in
applying electrophysiologic techniques, however, is
personal bias. In our study, we used automated
methods (multi-MUP and turns/amplitudes) which
are less prone to personal bias than manual EMG
methods.10,17 Using multi-MUP technique, inconsistent editing of sampled MUPs could be a source of
personal bias. To reduce it further, in the present
study, MUPs with unsteady baseline (unclear start/
end of the signal) were discarded. This deviates
from original descriptions of the technique by Stålberg and his group, who permit manual editing of
MUPs with “problematic” cursor settings.17 The personal bias is even smaller in IP analysis by turns/
amplitudes.
By contrast, the use of the “trigger and delay
unit” (on concentric needle EMG), and determination of fiber density (on single fiber EMG) is quite
prone to personal bias.10 Personal bias in our analysis was reduced even further by consecutive analysis
of about 50 subjects from DAT recorded tape, with
only the subjects’ initials available. To our knowl-
1750
Short Reports
edge, in none of the previous studies have such measures been undertaken.
In conclusion, our results imply that abnormalities in a patient examined because of suspected
cauda equina lesion, multiple system atrophy, or
some other cause for damage to the EAS innervation, cannot be dismissed as possibly due to a history
of mild constipation. Our study does not refute previous reports of some neurogenic EAS involvement
in severely constipated patients, seeking medical attention for their complaint.5,14
The authors thank Mičo Mrkaič, PhD, and Nacek Zidar for statistical analysis, and Dr. Dianne Jones for language review. The study
was supported by the Ministry of Science and Technology of the
Republic of Slovenia, grant J3 7899.
REFERENCES
1. Agachan F, Chen T, Pfeifer J, Reissman P, Wexner S. A constipation scoring system to simplify evaluation and management of constipated patients. Dis Colon Rectum 1996;39:
681–685.
2. Allen R, Hosker G, Smith A, Warrell D. Pelvic floor damage
and childbirth: a neurophysiological study. Br J Obstet Gynaecol 1990;97:770–779.
3. Bartolo DC, Read NW, Jarratt JA, Read MG, Donnelly TC,
Johnson AG. Differences in anal sphincter function and clinical presentation in patients with pelvic floor descent. Gastroenterology 1983;85:68–75.
4. Drossman DA, Sandler RS, McKee DC, Lovitz AJ. Bowel patterns among subjects not seeking health care. Use of a questionnaire to identify a population with bowel dysfunction.
Gastroenterology 1982;83:529–534.
5. Fink RL, Roberts LJ, Scott M. The role of manometry, electromyography and radiology in the assessment of intractable
constipation. Aust N Z J Surg 1992;62:959–964.
6. Johanson JF, Sonnenberg A, Koch TR. Clinical epidemiology
of chronic constipation. J Clin Gastroenterol 1989;11:525–
536.
7. Lubowski DZ, Swash M, Nicholls RJ, Henry MM. Increase in
pudendal nerve terminal motor latency with defaecation
straining. Br J Surg 1988;75:1095–1097.
8. Nandedkar SD, Sanders DB, Stålberg EV. Simulation and
analysis of the electromyographic interference pattern in normal muscle. Part II: activity, upper centile amplitude, and
number of small segments. Muscle Nerve 1986;9:486–490.
9. Nandedkar SD, Sanders DB, Stålberg EV. Automatic analysis
of the electromyographic interference pattern. Part II: findings in control subjects and in some neuromuscular diseases.
Muscle Nerve 1986;9:491–500.
10. Podnar S, Lukanovič A, Vodušek DB. Anal sphincter electromyography after vaginal delivery: neuropathic insufficiency or
normal wear and tear? Neurourol Urodyn 2000;19:249–257.
11. Podnar S, Rodi Z, Lukanovič A, Tršinar B, Vodušek DB. Standardization of anal sphincter EMG: technique of needle examination. Muscle Nerve 1999;22:400–403.
12. Podnar S, Vodušek DB. Standardisation of anal sphincter
EMG: low and high threshold motor units. Clin Neurophysiol
1999;110:1488–1491.
13. Podnar S, Vodušek DB. Standardization of anal sphincter
electromyography: uniformity of the muscle. Muscle Nerve
2000;23:122–125.
14. Snooks SJ, Barnes PR, Swash M, Henry MM. Damage to the
MUSCLE & NERVE
November 2000
innervation of the pelvic floor musculature in chronic constipation. Gastroenterology 1985;89:977–981.
15. Sonnenberg A, Koch TR. Epidemiology of constipation in the
United States. Dis Colon Rectum 1989;32:1–8.
16. Stålberg E, Chu J, Bril V, Nandedkar SD, Stålberg S, Ericsson
M. Automatic analysis of the EMG interference pattern. Electromyogr Clin Neurophysiol 1983;56:672–681.
17. Stålberg E, Falck B, Sonoo M, Stålberg S, Åström M. MultiMUP EMG analysis—a two year experience in daily clinical
Short Reports
work. Electroencephalogr Clin Neurophysiol 1995;97:
145–154.
18. Vaccaro CA, Cheong DM, Wexner SD, Nogueras JJ, Salanga
VD, Hanson MR, Phillips RC. Pudendal neuropathy in evacuatory disorders. Dis Colon Rectum 1995;38:166–171.
19. Vaccaro CA, Wexner SD, Teoh TA, Choi SK, Cheong DM,
Salanga VD. Pudendal neuropathy is not related to physiologic pelvic outlet obstruction. Dis Colon Rectum 1995;38:
630–634.
MUSCLE & NERVE
November 2000
1751