Pediatric Imaging
Tae Il Han, MD
In-One Kim, MD
Woo Sun Kim, MD
Jin Young Chung, MD
Kyuchul Choeh, MD
Mun Kab Song, MD
Yong Gyu Yoon, MD
Index terms:
Anus, US, 757.1298, 757.92
Infants, newborn, gastrointestinal
tract, 757.92
Ultrasound (US), in infants and
children, 757.1298
Radiology 2000; 217:392–394
Abbreviations:
EAS ⫽ external anal sphincter
IAS ⫽ internal anal sphincter
1
From the Departments of Radiology
(T.I.H., J.Y.C., M.K.S., Y.G.Y.) and Pe-
diatrics (K.C.), Eulji University School
of Medicine, 24-14 Mok-Dong, Jung-
Gu, Taejon 301-726, South Korea, and
the Department of Radiology, Seoul
National University College of Medi-
cine, South Korea (I.O.K., W.S.K.).
From the 1999 RSNA scientific assem-
bly. Received December 23, 1999; re-
vision requested January 30, 2000; re-
vision received April 3; accepted April
20. Address correspondence to
T.I.H. (e-mail: tihan@emc.eulji.ac.kr).
©
RSNA, 2000
Author contributions:
Guarantor of integrity of entire study,
T.I.H.; study concepts, T.I.H., I.O.K.;
study design, T.I.H.; definition of intel-
lectual content, T.I.H.; literature re-
search, T.I.H.; clinical studies, T.I.H.;
data acquisition, T.I.H., I.O.K., W.S.K.,
J.Y.C.; data analysis, T.I.H., I.O.K.; sta-
tistical analysis, T.I.H.; manuscript
preparation and editing, T.I.H., I.O.K.;
manuscript review, all authors.
392
US Identification of the Anal
Sphincter Complex and
Levator Ani Muscle in
Neonates: Infracoccygeal
Approach1
PURPOSE: To identify the anal sphincter complex and levator ani muscle at trans-
perineal ultrasonography (US) with the infracoccygeal approach.
MATERIALS AND METHODS: Infracoccygeal US was performed with a 7-MHz
linear-array transducer in 40 healthy neonates. The babies were placed in the supine
position, and both legs were drawn up to the chest. The buttocks were accordingly
lifted up. The approach site was just inferior to the coccyx and posterior to the anus.
Scanning was performed to obtain transverse images of the anorectal area. The thick-
ness of the anal sphincter complex and the puborectalis muscle were measured.
RESULTS: Infracoccygeal US revealed the internal anal sphincter (IAS), the external anal
sphincter (EAS), and the puborectalis muscle in all babies. The IAS and EAS were
depicted as inner and outer hypoechoic ringlike structures, respectively. A round,
hyperechoic space (intersphincteral space) was present between the hypoechoic IAS
and EAS. The puborectalis muscle was identified as a U-shaped hypoechoic structure.
The bulbocavernosus and ischiocavernous muscles were also identified. Mean thick-
nesses were as follows: IAS, 1.3 mm ⫾ 0.3 (SD) (range, 0.8 –1.9 mm); EAS, 1.6 mm ⫾
0.3 (range, 1.2–2.3 mm); and puborectalis muscle, 1.1 mm ⫾ 0.3 (range, 0.6–1.9 mm).
CONCLUSION: Infracoccygeal transperineal US is an excellent diagnostic modality
for demonstrating the anal sphincter complex and levator ani muscle in neonates.
The imaging modalities currently used to examine the anal sphincter complex and levator
ani muscle are endoanal ultrasonography (US) and endoanal magnetic resonance (MR)
imaging (1–3). The disadvantages of these imaging modalities include patient discomfort,
the necessity of using an endoanal probe or coil, and the inability to access an imperforate
anus or stenotic anus.
Transperineal US in children has been used for the evaluation of distal anorectal
anomalies (4). Transperineal US is also useful in diagnosing posterior urethral valves and
in identifying a distal rectal pouch and an internal fistula in an imperforate anus. However,
it does not allow adequate identification of the anal sphincter complex and levator ani
muscle in neonates and infants. Conventional transperineal US has been mainly per-
formed in the sagittal plane, and it cannot be performed in the transverse plane because
of interference by the pubic bone.
To overcome this problem, we developed an infracoccygeal approach with transperineal
US to identify the anal sphincter complex and levator ani muscle. This study was con-
ducted to determine the normal appearance of the anal sphincter complex and levator ani
muscle on US images obtained by using an infracoccygeal approach.
MATERIALS AND METHODS
From January 1999 through March 1999, 40 consecutive neonates (20 male and 20 female;
age range, 1–21 days; mean age, 6 days) with normal meconium passage were examined

Figure 1. (a) Photograph demonstrates placement of the linear transducer (T [7-MHz; Acuson,
Mountain View]) between the anus and coccyx at infracoccygeal US. (b) Lateral schematic of the
pelvis shows the transverse scanning levels: I, lower, to evaluate the anal sphincter; II, upper, to
evaluate the puborectalis muscle; and III, anterior, to evaluate the bulbocavernosus and ischio-
cavernous muscles. Scanning sites (*) are just inferior to the coccyx and posterior to the anus.
Figure 2. Transverse infracoccygeal sono-
gram of the anal sphincter complex obtained
at level I (Fig 1b) shows the inner hypoechoic
IAS (arrows), outer hypoechoic EAS (arrow-
heads), and hyperechoic intersphincteral space.
Figure 3. Transverse sonogram obtained at
level II (Fig 1b) shows the urethra (U) and rec-
tum (R) between two slings of the hypoechoic
U-shaped puborectalis muscle (arrows).
with infracoccygeal US to identify the
anal sphincter complex, levator ani mus-
cle, and adjacent pelvic floor muscles.
Our institutional review board approved
the study protocol, and informed con-
sent was obtained from the parents of the
neonates. Gestational ages ranged from
Volume 217 䡠 Number 2
33 to 40 weeks, with a mean of 38 weeks.
Body weights ranged from 2,520 to 4,200
g (mean, 3,220 g). We excluded prema-
ture neonates (⬍37 weeks gestational
age) with a low birth weight (⬍2,500 g).
One examiner (T.I.H.) performed all
US examinations in this prospective
study by using a 7-MHz linear-array
transducer (model 128; Acuson, Moun-
tain View, Calif). We modified the trans-
perineal US technique to use an infracoc-
cygeal approach to obtain transverse
images. The babies were placed in the
supine position, both legs were drawn up
to the chest, and the buttocks were ac-
cordingly lifted up (Fig 1a). The approach
site for infracoccygeal US was just inferior
to the coccyx and posterior to the anus.
Scanning was performed to obtain
transverse images of the anorectal area.
Standard gray-scale settings for evalua-
tion of small anatomic parts were used.
Infracoccygeal US was easily performed
in all babies without sedation, and each
examination required 2–5 minutes. A
thick layer of gel was applied between the
transducer and the perineum to prevent
artifacts from intervening air. US images
were rotated vertically to correspond to
the usual orientation on computed to-
mographic or MR images.
The echogenicity of the structures was
evaluated as high (high-level echoes sim-
ilar to those of fat) or low (low-level ech-
oes similar to those of muscle) on the
basis of US descriptions provided by pre-
vious investigators (5,6). If the echoge-
nicity was between high and low, the
structure was described as having inter-
mediate echogenicity.
The thickness of the internal anal
sphincter (IAS), external anal sphincter
(EAS), and puborectalis muscle was mea-
sured in the resting state.
Anal Spincter and Levator Ani Muscle: Infracoccygeal US
RESULTS
Infracoccygeal US revealed the IAS and
EAS in the lower part of the anal canal
and the puborectalis muscle in the upper
part of the anorectal region (Fig 1b). On
the transverse image obtained through
the lowest part of the anal canal, two
hypoechoic ring-shaped structures were
identified in all babies (Fig 2). The inner,
hypoechoic, ringlike structure was the
IAS, and the outer structure was the EAS.
The IAS had low echogenicity in all ba-
bies, whereas the EAS had low (n ⫽ 30) or
intermediate (n ⫽ 10) echogenicity. A
round hyperechoic space between the
hypoechoic IAS and EAS was the inter-
sphincteral space (Fig 2), which con-
tained longitudinal muscle. On the up-
per transverse scan, the puborectalis
muscle was identified as a U-shaped hy-
poechoic structure in all babies (Fig 3).
On the anterior transverse scan of the
anus, the bulbocavernosus muscle was
well delineated in all female babies (Fig
4a). On more cranial images, the ischio-
cavernous muscle was identified as an in-
verted V-shaped hypoechoic structure in
all babies (Fig 4b).
The mean thickness of the IAS was 1.3
mm ⫾ 0.3 (SD) (range, 0.8 –1.9 mm). The
mean thickness of the EAS was 1.6 mm ⫾
0.3 (range, 1.2–2.3 mm). The mean thick-
ness of the puborectalis muscle was 1.1
mm ⫾ 0.3 (range, 0.6 –1.9 mm).
DISCUSSION
US evaluation of the anus and distal rec-
tum has been performed with transperi-
neal or endoanal US. Endoanal US in
adults has evolved rapidly in recent years
and has become an established procedure
to evaluate the sphincteral defects in fe-
cal incontinence or perianal abscesses in-
cluding fistulous tracts (6 – 8). However,
in children or neonates, endoanal US
cannot be used in cases of anal stenosis or
imperforate anus.
Infracoccygeal US provides a transverse
image that enables us to identify the anal
sphincter complex and levator ani mus-
cle. On infracoccygeal US images, the IAS
was seen as a clearly defined inner hypo-
echoic ring, which has a similar appear-
ance on endoanal US images (5,9,10).
The IAS is an intrinsic muscle surround-
ing the anal canal just beneath the mu-
cosal layer. Anatomically, the circular
smooth muscle layer of the rectum con-
tinues to the anal canal, where it thickens
to become the IAS. The echogenicity of
the IAS increases with age, with the re-
䡠 393
placement of smooth muscle by the con-
nective tissues (7,11). In our study, the
IAS in all neonates was hypoechoic.
The EAS was identified as an outer hy-
poechoic ring in our study. The EAS has
three components: deep, superficial, and
subcutaneous. The deep external sphinc-
ter blends imperceptibly into the inferior
portion of the puborectalis muscle. The
echogenicity of the EAS is variable with
endoanal US in adults, and it was seen as a
circular ring with low or intermediate
echogenicity on infracoccygeal US images.
The measured thicknesses of the IAS
(range, 0.8 –1.9 mm) and the EAS (range,
1.2–2.3 mm) were less than the ranges
reported (1,2,12) with endoanal US in
adults (IAS range, 0.5–3.7 mm; EAS
range, 5.8 –9.8 mm).
The intersphincteral space was re-
vealed as a round hyperechoic ringlike
structure between the hypoechoic IAS
and EAS. This space contains fat and lon-
gitudinal muscle, which is the terminal
portion of longitudinal smooth muscle of
the rectum (13). The longitudinal muscle is
the least understood anorectal structure,
but it is thought to play a role in binding
and bracing the components of the anal
canal together and in fixing the perineum
to the pelvis (13,14).
Sultan et al (15,16) have suggested that
the longitudinal muscle is hyperechoic
partly because the direction of the fibers
affects the reflectivity and also because
the longitudinal muscle may be mixed
with fibrous tissue. They described that
while the longitudinal muscle could be
identified in adult males, it was identified
in less than half of the adult females; this
finding can be explained by the differ-
ences in the echogenicity of the adjacent
EAS (15,16). However, we could find the
intersphincteral space as a hyperechoic
structure regardless of sex. Identification
of the IAS and EAS may be helpful in the
postoperative evaluation of patients with
imperforate anus and in the determina-
tion of the anatomic relationship of the
pulled-through intestine with the anal
sphincter complex.
The puborectalis muscle was also iden-
tified as a hypoechoic U-shaped band at
the level of the anorectal flexure, which
also merged with the hypoechoic EAS.
The puborectalis muscle is the innermost
portion of the levator ani muscle, and it
is considered to have an important role
in the control of bowel function after
surgical repair of an anorectal anomaly.
394 䡠 Radiology 䡠 November 2000
Figure 4. (a) Transverse sonogram obtained at level III (Fig 1b) shows a hypoechoic bulbocav-
ernosus muscle (arrowheads). The inner mixed hyperechoic structure is the vagina (V). (b) More
cranial transverse sonogram shows the ischiocavernous muscle (arrows) as an inverted V-shaped
structure. The urethra (U) and the vagina (V) are seen between two slings of the ischiocavernous
muscle.
Both the EAS and the puborectalis muscle
are composed of striated muscle fibers,
and they are also in close contact with
each other. The puborectalis muscle is a
landmark used in distinguishing patients
with low-type imperforate anus and those
with high-type imperforate anus. Because
infracoccygeal US can demonstrate the pu-
borectalis muscle, it may be useful for the
evaluation of the relationship between the
puborectalis muscle and the distal rectal
pouch; this evaluation is important in the
determination of the type of imperforate
anus.
In conclusion, infracoccygeal US in the
neonate shows the anatomy of the anal
sphincter complex and levator ani mus-
cle and will be helpful in the detection of
disease in this area.
Acknowledgment: The authors thank Seo-
Yoon Choi for drawing the schematic repre-
sentation.
References
1. Nielsen MB, Pedersen JF, Hauge C, Ras-
mussen OO, Christiansen J. Endosonog-
raphy of the anal sphincter: findings in
healthy volunteers. AJR Am J Roentgenol
1991; 157:1199 –1202.
2. Hussain SM, Stoker J, Lameris JS. Anal
sphincter complex: endoanal MR imag-
ing of normal anatomy. Radiology 1995;
197:671– 677.
3. deSouza NM, Gilderdale DJ, MacIver DK,
Ward HC. High-resolution MR imaging of
the anal sphincter in children: a pilot
study using endoanal receiver coils. AJR
Am J Roentgenol 1997; 169:201–206.
4. Teele RL, Share JC. Transperineal sonog-
raphy in children. AJR Am J Roentgenol
1997; 168:1263–1267.
5. Alexander AA, Miller LS, Liu JB, Feld RI,
Goldberg BB. High-resolution endolumi-
nal sonography of the anal sphincter com-
plex. J Ultrasound Med 1994; 13:281–284.
6. Alexander AA, Liu JB, Merton DA, Nagle
DA. Fecal incontinence: transvaginal US
evaluation of anatomic causes. Radiology
1996; 199:529 –532.
7. Bartram CI, Sultan AH. Anal endosonog-
raphy in faecal incontinence. Gut 1995;
37:4 – 6.
8. Nielsen MB, Hauge C, Pedersen JF, Chris-
tiansen J. Endosonographic evaluation of
patients with anal incontinence: findings
and influence on surgical management.
AJR Am J Roentgenol 1993; 160:771–775.
9. Schafer A, Enck P, Furst G, Kahn T, Frieling
T, Lubke HJ. Anatomy of the anal sphinc-
ters: comparison of anal endosonography
to magnetic resonance imaging. Dis Colon
Rectum 1994; 37:777–781.
10. Law PJ, Bartram CI. Anal endosonogra-
phy: technique and normal anatomy.
Gastrointest Radiol 1989; 14:349 –353.
11. Nielsen MB, Pedersen JF. Changes in the
anal sphincter with age: an endosono-
graphic study. Acta Radiol 1996; 37:357–361.
12. Sangwan YP, Solla JA. Internal anal
sphincter: advances and insights. Dis Co-
lon Rectum 1998; 41:1297–1311.
13. Lunniss PJ, Phillips RK. Anatomy and
function of the anal longitudinal muscle.
Br J Surg 1992; 79:882– 884.
14. Halligan S. Imaging fistula-in-ano. Clin
Radiol 1998; 53:85–95.
15. Sultan AH, Kamm MA, Hudson CN, Ni-
cholls JR, Bartram CI. Endosonography of
the anal sphincters: normal anatomy and
comparison with manometry. Clin Ra-
diol 1994; 49:368 –374.
16. Sultan AH, Nicholls RJ, Kamm MA, Hud-
son CN, Beynon J, Bartram CI. Anal en-
dosonography and correlation with in
vitro and in vivo anatomy. Br J Surg 1993;
80:508 –511.
Han et al