European Journal of Obstetrics & Gynecology and

Reproductive Biology 103 (2002) 7982

Trans-perineal versus endo-anal ultrasound in the detection

of anal sphincter tears
Lohse Corneliaa, Bretones Stephanb,*, Boulvain Michela, Weil Antoinea, Krauer Felixa
a
Department of Gynaecology, University Hospital of Geneva, Geneva, Switzerland
b
Department of Obstetrics and Gynecology, University Hospital Edouard Herriot, Place dArsonval, 69003 Lyon, France
Received 28 February 2001; accepted 10 December 2001

Abstract

Objective: To compare endo-anal and trans-perineal ultrasonography in the detection of anal sphincter lesions. Study design: Sixty-four
patients sent for a uro-gynaecological consultation were given tests by two ultrasound methods using the same apparatus (Aloka) with
different probes. The endo-anal ultrasound (EAUS) was carried out in a supine position using a rotary circular probe at 7.5 MHz. The trans-
perineal ultrasound (TPUS) was carried out using a curved linear probe at 5 MHz, applied to the perineum in the transverse direction, allowing
a proximal view of the sphincter. For each image the thicknesses of the two components of the sphincter (internal and external) were noted, as
well as the presence of any ultrasonographic tear, its position, and its width on the sphincterian circumference. Results: The average of the
internal and external thickness of the sphincters was of 2.8 and 5.6 mm, respectively for the TPUS, and 2.1 and 6.2 mm for the EAUS. Among
the 64 patients, 15 external lesions of the anal sphincter were found with the TPUS, and 14 with the EAUS. The kappa correlation coefficient
was 48%, with a confidence interval of 95%. Conclusion: The endo-anal technique, which provides a staged analysis of the anal sphincter
muscle, remains the benchmark of sphincterian imaging. TPUS is a useful method, in that it gives good visualisation of anatomical structures,
but its sensitivity in detecting sphincter lesions remain to be demonstrated. # 2002 Elsevier Science Ireland Ltd. All rights reserved.

Keywords: Endosonography; Anal sphincter; Trans-perineal sonography

1. Introduction Endo-anal ultrasonography is a referential method for

exploring the anal sphincter muscle, measuring his thickness
The most common cause of anal sphincter lesions is and integrity, and detecting the presence of scarring or loss
vaginal delivery, during which both sphincters may be of muscle tissue. It is currently considered the best method
damaged [1]. Recent studies using anal endosonographic of post-partum anal sphincter evaluation. It is a simple, non-
techniques have shown that up to a third of women sustain an traumatic method, and is useful in diagnostic terms [3].
occult injury to the anal sphincter during first vaginal Unfortunately, this method needs a specialised material used
delivery [2,3]. only by gastroenterologist and is not performed routinely
During vaginal delivery, muscular injury [4,5] and neu- after childbirth [10]. Trans-perineal ultrasonography is use-
rological injury [6] of the anal sphincter are common ful for every obstetrician using a common ultrasonograph.
problems, and may lead to anal incontinence [7]. The The aim of our study was to compare the two methods and to
frequency of this problem increases with age, though it evaluate the possibility for every obstetrician to carry out an
generally remains asymptomatic until the menopause. ultrasonographic perineal assessment after the childbirth.
The incidence of anal sphincter injury after delivery is
probably underestimated [8,9]. Many young patients are
not symptomatic initially, and therefore are not identified. 2. Materials and methods
Clinical diagnosis is very difficult in the post-delivery
period, but it can be important in cases where perineal All the patients who consulted at the Uro-gynaecological
physiotherapy, or possibly surgical correction is needed. Clinic at the University Hospital in Geneva between January
and June 1999 were included in this prospective study.
*
Corresponding author. Tel.: 33-4-72-11-77-54;
Patients complained of urinary incontinence only. The urin-
fax: 33-4-72-11-77-09. ary incontinence assessment included in our unit, the rea-
E-mail address: [email protected] (B. Stephan). lisation of a perineal ultrasonography in order to measure the

0301-2115/02/$ see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0 3 0 1 - 2 1 1 5 ( 0 2 ) 0 0 0 2 1 - 0
80 L. Cornelia et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 103 (2002) 7982

Fig. 1. Endo-anal ultrasound: External anal sphincter lesion.

bladder neck mobility. For those reasons we proposed our by two independent operators. On transverse sonogram, the
patients to be included in a clinical research to compare two internal anal sphincter (IAS) is a hypoechoic circular band
different sonographic methods for diagnosing occult anal and the external anal sphincter (EAS), an echogenic circular
sphincter lesions. It is important to know that no patient had band (Fig. 2). An EAS defect appears as a break in the
fecal incontinence. They were given endo-anal (EAUS) and normal echogenic ring (Fig. 3), an internal anal defect as a
trans-perineal (TPUS) ultrasound tests, using an Aloka SSD- gap in the hypoechoic ring.
2000. The gold standard method for detecting anal sphincter Informed written consent was obtained from all 64
lesion remains the endo anal sonography. In our study, the patients, who were examined in the supine position.
rectal endoprobe had a 7.5 MHz high-frequency transducer The data were analysed by the Center for Disease Control
(Fig. 1). For the TPUS, the rectal probe was replaced by a Epi Info Statistics software. The correlation between the
linear sonographic probe at 5 MHz. Each test was performed two methods was calculated by the kappa coefficient. This

Fig. 2. Trans-perineal ultrasound. Normal anal sphincter (1) EAS and (2) IAS.
 L. Cornelia et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 103 (2002) 7982 81

Fig. 3. Trans-perineal ultrasound. External anal sphincter lesion.

prospective study was accepted by the maternity ethics lesions. The TPUS revealed one IAS lesion and 15 isolated
committee of the University Hospital in Geneva. EAS lesion. The kappa correlation coefficient was 48%,
with a confidence interval of 95%. There was no difference
among the lesions diagnosed by both methods on the level of
3. Results the seat of the lesions (100% of correlation) and the size of
those (P 0:18). If one takes anal endosonography as a
Sixty-four patients without any previous history of anor- reference method, the sensitivity of the TPUS for the
ectal surgery were included in our study. The mean thickness diagnosis of sphincter lesions is 50%, only seven of the
of the IAS as measured by anal endosonography (Table 1) 14 EAS lesions were detected by TPUS. The specificity of
was 2.1 mm (S:D: 0:8), and that of the EAS 6.2 mm the diagnosis of sphincter lesions with ETP is 84% because
(S:D: 1:5). Using TPUS, the mean thickness of the IAS of the 50 sphincters considered to be normal, eight only had
was found to be 2.8 mm (S:D: 0:7), and that of the EAS a lesion with the TPUS. Both techniques were well tolerated
5.6 mm (S:D: 1:8). Statistical analysis using the t-test by the patients. No complaints were reported.
showed statistically significant differences between the
EAUS and TPUS values for both the IAS (P < 0:0001)
and the EAS (P < 0:039). 4. Discussion
No IAS lesions were found in any of the 64 patients using
the EAUS (Table 2), but 14 patients had isolated EAS Since Sultans publication in 1993 [2], ultrasonographic
imaging of the anal sphincter in the post-partum period has
generated considerable interest. The EAUS technique using
Table 1
Mean values of sphincter thickness in mm (S.D.), using two methods: trans- a rotating endoprobe seems to be the best method for the
perineal ultrasound (TPUS) and endo-anal ultrasound (EAUS) (n 64) diagnosis of an anal sphincter lesion.
This method requires expensive equipment and practical
EAUS TPUS P value (t-test)
experience in using it. Although, EAUS still remains the
Internal anal sphincter 2.1(0.8) 2.8(0.7) 0.001 standard technique for imaging the anal sphincter, its avail-
External anal sphincter 6.2(1.5) 5.6(1.8) 0.04 ability remains limited. Ten percent of primiparous women
complain of persistent anal incontinence 3 months after
Table 2
delivery, and 30% of primiparous women, with or without
Anal sphincter lesions related to the mode of screening, trans-perineal incontinence, who are given an EAUS test, are found to have
ultrasound (TPUS) and endo-anal ultrasound (EAUS) (n 64) EAS lesions [11].
Several attempts have already been made to use ultra-
EAUS TPUS TPUS and EAUS
sound technique in the investigation of the anal sphincter. In
External anal sphincter 14 15 7 1997, Peschers [12] used TPUS to evaluate 55 patients who
Internal anal sphincter 00 01 0
complained of anal incontinence, and 43 who had no bowel
Total 14 16 7
symptoms. Intra-observer agreement among researchers on
82 L. Cornelia et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 103 (2002) 7982

IAS lesions was 100%, as confirmed by surgery. And in the The utilization of TPUS by obstetricians and gynaecol-
EAS group, there was just one discordant result. ogists in their every-day practice to detect anal sphincter
In 1994, Sultan [13] compared vaginal and anal endoso- lesions in post-partum patients could provide useful infor-
nography in the diagnosis of anal sphincter lesions. He found mation on the practicality and sensitivity of this technique.
a high level of correlation between the two techniques in the
evaluation of IAS lesion size. The endosonographic results
were confirmed by vaginal ultrasound test. Acknowledgements
In a prospective study in 1995, Sandridge [14] used endo-
vaginal ultrasound to investigate 70 patients for anal sphinc- The authors would like to thank Alain Watier (University
ter lesions. He showed that 36% of all had lesions of the IAS, of Sherbrook, Canada) for his final lecture of this paper.
and 29% of the EAS.
In 1998, Poen [15] compared endo-vaginal and EAUS test
results for 36 patients with fecal incontinence and 20 patients References
with perianal sepsis. Endo-vaginal ultrasound seems to give
better results for all the structures except the EAS. As regards [1] Swash M. Faecal incontinence: childbirth is responsible for most
the diagnosis of sphincter lesions, the correlation between the cases. BMJ 1993;307:6367.
two methods was poor. The EAUS technique overestimated [2] Sultan AH, Kamm MA, Hudson CN, Bartram C. Anal sphincter
anal sphincter lesions in patients with fecal incontinence. disruption during vaginal delivery. New Engl J Med 1993;329:1905
11.
In our study, the thickness of the internal sphincter was
[3] Kamm MA. Obstetric damage and fecal incontinence. Lancet
found to be higher with trans-perineal than endo-anal ultra- 1994;334:7303.
sonography, probably due to the absence of endoluminal [4] Duthie HL, Watts JM. Contribution of the external anal sphincter to
distention of the muscle in the former case. This difference the pressure zone on the anal canal. Gut 1965;6:648.
is also found in comparisons between endo-anal and endo- [5] Gibbons CP, Trowbridge EA, Bannister JJ, Read NW. Role of anal
cushions in maintaining continence. Lancet 1986;i:88688.
vaginal sonographic techniques [13], with measurements of
[6] Gee AS, Durdey P. Urge incontinence of feces is a marker of severe
the internal sphincter ranging between 2.3 mm, for endo-anal external anal sphincter dysfunction. Br J Surg 1995;82:117982.
ultrasonography (S:D: 0:5), and 3.2 mm (S:D: 1:2) for [7] Parks AG, Swash M, Urich H. Sphincter denervation in anorectal
endo-vaginal sonography (P 0:01). Similar values were incontinence and rectal prolapse. Gut 1977;18:65665.
reported by Poen [15], who also found that all the muscular [8] Leigh RJ, Trumberg LA. Fecal incontinence: the unvoiced symptom.
Lancet 1982;1:134951.
structures in the EEA were thicker, except for the external
[9] Glazener CM, Abdulla M, Stroud P, Nagi S, Tempelton A, Russell I.
sphincter, which was found to be significantly thin. As in our Postnatal maternal morbidity: extent, causes, prevention and treat-
study, the difference was more marked in the case of the ment. Br J Obstet Gynaecol 1995;102:2827.
internal sphincter. The kappa coefficient values suggest that [10] Faltin DL, Boulvain M, Irion O, Bretones S, Stan C, Weil A.
this is a promising technique, but this needs further confirma- Diagnosis of clinically occult anal sphincter tears by endosonography
performed immediately after delivery. Obstet Gynecol 2000;95(5):
tion. The TPUS method still needs anatomical corroboration.
6437.
Furthermore, the indications, the availability and the feasi- [11] Damon H, Henry L, Bretones S, Mellier G, Minaire Y, Mion F.
bility of ultrasound imaging of the anal canal need to be Postdelivery anal function in primiparous females: ultrasound and
improved so that diagnosis can take place as early as possible, manometric study. Dis Colon Rectum 2000;43:4727.
thus optimising the possibilities of treatment. [12] Peschers UM, DeLancey JOL, Schaer GN, Schuessler B. Exoanal
ultrasound of the anal sphincter: normal anatomy and sphincter
The interest of a simple method of diagnosis of the
defects. Br J Obstet Gynaecol 1997;104:9991003.
sphincterian lesions after a childbirth is to allow the surgical [13] Sultan AH, Loder PB, Bartram I, Kamm MA, Hudson CN. Vaginal
repair in the early post-partum period where the cicatrization endosonography: new approach to image the undisturbed anal
is optimal. Apart from this, particular period of the post- sphincter. Dis Colon Rectum 1994;37:12969.
partum, period the diagnosis of an asymptomatic lesion of [14] Sandridge DA, Thorp JM. Vaginal endosonography in the assessment
of the anorectum. Obstet Gynecol 1995;86:10019.
the sphincter must push the gynaecologist surgeon to carry
[15] Poen AC, Felt-Bersma RJF, Cuesta MA, Meuwissen SGM. Vaginal
out a complete assessment (EEA-Manometry) in order to endosonography of the anal sphincter complex is important in the
direct the gesture towards a sphincterian repair associated assessment of fecal incontinence and perianal sepsis. Br J Surg 1998;
with a cure of prolapse. 58:35963.