Endoanorectal Ultrasonography Ismail Mihmanli, MD,* Fatih Kantarci, MD,* and Vikram S

REVIEW ARTICLE

Endoanorectal Ultrasonography Ismail Mihmanli, MD,* Fatih Kantarci, MD,* and Vikram S. Dogra, MDÞ

very close to the target structures and thus providing images Abstract: Endoanorectal ultrasonography (EARUS) may be used for of high spatial resolution. The technique has attracted con- diagnosing various anorectal disorders. EARUS is easy to perform, siderable attention because of its ability to demonstrate the has a short learning curve, and causes less discomfort than routine presence and extent of anal sphincter disruption, notably after digital examination. Anal sphincters can be clearly visualized, and vaginal delivery.2 one can easily distinguish between the internal (hypoechoic) and Endoanorectal ultrasonography is also a valuable method external (hyperechoic) anal sphincters. Other pelvic floor structures, for diagnosing and grading perianal inflammation and its like the puborectalis muscle, can also be visualized. The use of con- follow-up.3 EARUS can help in identifying the stage of rectal trast agents can increase the accuracy of EARUS in the assessment and anal tumors by demonstrating adjoining lympadenopathy. of perianal fistulae. In addition, EARUS is an excellent alternative In addition, EARUS can be used in the diagnosis of the pos- to expensive magnetic resonance imaging. Besides its use in incon- sible causes of obstructed defacation (OD), as well as docu- tinence and perianal sepsis, the presence of slight or massive sub- menting dynamic examination. mucosal invasion in early rectal cancer may be imaged in greater detail. This review article will present indications of the With 3-dimensional EARUS, it is possible to diagnose the anorectal EARUS with complementary techniques such as transvaginal diseases, in multiplane, with high spatial resolution, adding important and transperineal US. information about the therapeutic decision. The normal sonographic anatomy of the anorectum, sonographic findings of anorectal diseases, and indications and limitations of endosonography with complemen- TECHNIQUES tary techniques such as transvaginal and transperineal ultrasound are Endoanorectal ultrasonography can be performed in left reviewed in this article. lateral decubitus position, supine lithotomy position, or prone Key Words: ultrasonography, endoanal, endorectal, vaginal, position with patients knees bent at 90 degrees. Some image perineal, fistulas, fistula in ano, tumors, constipation, asymmetry may be induced if the left lateral decubitus position obstructive defecation, incontinence is used, therefore a prone position is preferable for EARUS examination.4 Patients’ sedation or bowel preparation is not (Ultrasound Quarterly 2011;27:87Y104) required before the EARUS procedure. An ultrasound rectal probe is introduced into the rectum after limited rectal clean- ing, aligned in standard orientation (with the anterior end he current diagnostic workup for anal and perianal abnor- uppermost or at the 12-o’clock position; 3-o’clock position T malities comprises a clinical history, physical examination, represents the patient’s left side, 6-o’clock position represents and various physiological tests. During the past 2 decades, the patient’s posterior sideVnatal cleft, and 9-o’clock position imaging has become an increasingly important part of investi- represents the patient’s right side), and then slowly withdrawn gation, and dedicated imaging techniques have been developed down the anal canal until the hyperechoic puborectal muscle, and their use has become more widespread. These techniques used as a landmark. Images are obtained at cranial, middle, have provided new insights into the pathogenesis of pelvic floor and caudal levels of the anal canal (Fig. 1). disorders. Radiologists becoming involved in this field can expect an increasing demand for these examinations, as clin- Sonographic Anatomy of the Anal Canal icians come to rely more heavily on imaging studies for diag- The anal canal is the most distal portion of the gastro- nosis and to select appropriate therapies for patients. intestinal tract, beginning at the anorectal line and ending at Endoanorectal ultrasonography (EARUS) is the tech- the anal verge. A dentate line found 1.5 cm from the anal verge nique that depicts the anal sphincter complex in detail.1 separates the distal squamous (sensory) epithelium and the Simple modification of a rotating rectal endoprobe by cover- proximal columnar epithelium.5 In other words, the dentate ing it with a nondeformable plastic cone allowed the trans- line refers to the mucocutaneous junction and is the location ducer to be withdrawn through the anal canal itself, situating it of the largest number of anal glands, which are implicated in the development of perianal sepsis. Received for publication November 6, 2010; accepted January 9, 2011. Theanalsphinctersform2cylindricallayersbetween *Department of Radiology, Cerrahpasa Medical Faculty, Istanbul University, which hyperechogenic intersphincteric space lies (Figs. 2 and 3). Istanbul, Turkey; and †Department of Imaging Sciences, University of The internal sphincter (IS) forms the innermost muscular layer, Rochester School of Medicine and Dentistry, Rochester, NY. which is an involuntary muscle and is continuous with the cir- Reprints: Ismail Mihmanli, MD, Department of Radiology, Cerrahpasa Medi- cal Faculty,˙ Istanbul University, Istanbul 34098, Turkey (e-mail: mihmanli@ cular muscle fibers of the muscularis propria of the rectum yahoo.com). (Fig. 4). The IS is a hypoechogenic symmetrical ring and Copyright * 2011 by Lippincott Williams & Wilkins is best seen in the middle part of the anal canal where the

Ultrasound Quarterly & Volume 27, Number 2, June 2011 www.ultrasound-quarterly.com 87

FIGURE 1. A, Anal canal cranial (high) level shows concentric IS and puborectalis muscle (*) but anterior part of the ES is normally not seen at this level (arrow). B, Anal canal middle level shows both IS and ES in the same image. The IS is hypoechogenic and the ES is hyperechogenic on US. Note the hyperechogenic intersphincteric area (arrows) between ES and IS. C, Anal canal caudal (distal) level shows only subcutaneous segment of the ES. Sometimes air artifact (arrow) is seen in the natal cleft area, which should not be misdiagnosed as an abnormality.

IS is the thickest. The IS is discontinuous in the outer third of portion lying close to the anal verge, a superficial part in the the anal canal. The hypoechogenic longitudinal muscle is the middle of the anal canal, and a deep part lying proximally to the continuation of the longitudinal muscle of the rectal wall, with anorectal junction and joining the U-shaped puborectalis mus- striated muscle components from the puboanalis and a con- cle or sling, a part of the levator ani9Y11 (Fig. 7). siderable fibroelastic content derived from the pubocervical fascia. The external sphincter (ES) is the outermost muscle of Three-Dimensional Ultrasound the distal anal canal and usually hyperechogenic on US and is The 3-dimensional (3D) US technique for imaging of a voluntary muscle (Fig. 5) that is composed of several paral- the anal region is an accessible tool that can potentially bring lel bundles of striated muscle. The ES has a circular structure ultrasonographic examination closer to computed tomography (Fig. 6) and surrounds the anal canal in continuity with the (CT) or magnetic resonance imaging (MRI) through its lesser levator ani and puborectalis muscles that are hyperechogenic operator dependence, available postprocessing tools, and poten- on US.6 The anterior part of the ES in the upper level of the tial for better standardization of evaluation and measurement anal canal is shorter in women by approximately 1.5 cm.7 It planes. Also, 3D US has been shown to be useful in the pre- is proposed that the ES consists of 3 components: the sub- operative assessment of anterior trans-sphincteric fistulas by cutaneous, superficial, and deep portions.8 The subcutaneous quantifying the length of muscle to be transected, contributing

FIGURE 2. Coronal view demonstrating the relationship of various anal sphincters.

88 www.ultrasound-quarterly.com * 2011 Lippincott Williams & Wilkins

FIGURE 5. The ES is the outermost muscle of the distal anal canal. In this case, the subcutaneous segment is seen (between arrows). FIGURE 3. Diagrammatic representation of anal sphincter muscles to further illustrate their interrelationship. visualize information in the 3D image depends critically on the rendering technique.14,16 Three basic techniques are used: to the choice of a safe treatment approach and to the reduc- (1) Surface render mode: An operator or algorithm identi- tion of the rate of postoperative continence disorders.12 fies the boundaries of the structures to create a wire-frame As the transition toward total digital image acquisi- representation. Surface-rendering techniques give good tion continues, 3D US, constructed from a synthesis of a high results only when a surface is available for rendering. number of parallel transaxial 2-dimensional (2D) images, has This technique fails when a strong surface cannot be been developed.13Y15 After a 3D data set has been acquired, it found such as in the subtly layered structures within the is immediately possible to select coronal anterior-posterior or anal canal and the rectal wall. posterior-anterior and sagittal right-left views, together with (2) Multiplane-viewing techniques: Three perpendicular any oblique image plane. The 3D image can be rotated, tilted, planes (axial, sagittal, and coronal) are displayed simulta- and sliced to allow the operator to infinitely vary the different neously and can be moved and rotated to allow the opera- section parameters, visualize the lesion at different angles, and tor to infinitely vary the different section parameters and measure accurately distance, area, angle, and volume. How- to visualize the lesion at different angles (Fig. 8). ever, a limitation of 3D reconstruction is that the contents of (3) Volume render mode: This is a special feature that can the volume remain unknown.16 However, render modes are be applied to high-resolution 3D data volume so that the applied high-resolution 3D data volume, and the information inside the cube is reconstructed to some extent. The ability to

FIGURE 4. The IS forms a hypoechogenic ring-like appearance FIGURE 6. Middle anal canal. External sphincter shows mixed (between calipers and arrows). echogenicity (between arrows), forming a circular structure.

* 2011 Lippincott Williams & Wilkins www.ultrasound-quarterly.com 89

FIGURE 7. Sagittal semidiagrammatic representation of anal sphincteric muscles and their relationship with surrounding structures.

information inside the cube is reconstructed to some is good for documentation of the presence, number, and in- extent (Figs. 9 and 10). ternal course of perianal fistulas and useful for planning surgical strategy.17Y20 Some authors declared that this technique Conventional Methods and Hydrogen could be helpful in difficult cases, although no overall diag- Peroxide Administration nostic benefit has been found.21,22 The major problems while investigating primary tracts An important disadvantage inherent to H2O2 injection with EARUS occur because of the structural alterations of the is the very strong reflection that occurs at the gas-tissue anal canal and perianal muscles and tissues or poor definition interface, which blanks out any detail deep in this interface. of the tract when filled with inflammatory tissue. The injection The bubbles produced by H2O2 induce acoustic shadowing of hydrogen peroxide (H2O2) through the external opening of deep in the tract, so all information deep in the inner surface the fistula is expected to improve the diagnostic accuracy of of the tract is lost. Instillation of H2O2 is generally accepted to standard EARUS. When used, hydrogen peroxide produces a be very safe, but several complications have been reported, significant increase in the echogenicity of the fistulous tract, including air embolism.23,24 Such complications have been which then appears as hyperechoic instead of hypoechoic. associated with large infusion volume or forceful instillation. There are several references indicating favorable results However, gentle instillation might negatively influence the 21 about H2O2-enhanced EARUS, pointing that this technique usefulness of the enhancement.

FIGURE 8. Three-dimensional image, multiplane view. A, Three perpendicular planes (axial, left lower; sagittal, right lower; coronal, left upper) are displayed simultaneously. B, Different sections of the hypoechogenic intersphincteric abscess (arrow) with ﬁstulous tract (arrowhead, left lower quadrant) is clearly seen.

90 www.ultrasound-quarterly.com * 2011 Lippincott Williams & Wilkins

tion for anal ultrasound often clearly defines the anal and perianal structures in experienced hands. For endorectal US, a degassed water-filled balloon covering the rotating ultrasound probe is used. The probe, with the empty balloon covered with a condom, is carefully introduced, preferably above the level of the area under investi- gation. The balloon is then filled with degassed water. It is essential that the transducer should not rotate until the balloon is filled with sufficient water. Then the probe is slowly retracted, and images of the entire length of the rectum are obtained. The balloon can be deinflated or reinflated as needed when evaluating rectal wall tumors. For endoanal US, the balloon contact method is not required because the hard anal cone provides sufficient cou- pling with the anal wall, even in patients with lax sphincters. A hard, sonolucent plastic cone covering the transducer is used. Before imaging, the plastic cone is fixed to the tip of the transducer. Ultrasound gel is applied to the transducer before FIGURE 9. Three-dimensional image. Data volume looks and after placement of the condom. As the transducer rotates, like a cube. the probe is carefully introduced. In recent years, there has been growing interest in transperineal25 and transvaginal US approaches in perianal View on Endoanorectal, Transperineal, diseases.26 On transperineal route, the probe is placed in the and Transvaginal Routes area of the fourchette in females, which is most effective for Endoanorectal, transperineal, and transvaginal ap- examination of the anorectum.27 The probe is placed on the proaches, using both 2D and 3D imaging, have made an perineum anterior to the anal canal in males. An empty rectum important impact on the evaluation of the anorectum. Espe- improves evaluation. Pressure is applied to the perineal body, cially, 2D endoanorectal US probes, which are dedicated for and the probe is angled posteriorly and cephalad to image the anorectum, have the capability of imaging the hollow lumen anal canal and perianal region. A rocking motion of the probe of the rectum and anus for the entire 360 degrees in axially. Also, well-known endocavitary probes (mainly used for prostate and gynecologic examinations) can also be used for this purpose, and they can image anorectal region in a longitudinal endorectal orientation as well. Longitudinal endorectal orienta-

FIGURE 10. Three-dimensional image, multiplane view. FIGURE 11. Perineal ultrasound with the image aligned Information inside the cube is reconstructed. The hypoechogenic vertically. The anterior and posterior aspects of the IS display abscess is clearly seen on this image (asterisks and arrows). 2 well-deﬁned, almost parallel hypoechoic structures (arrows).

* 2011 Lippincott Williams & Wilkins www.ultrasound-quarterly.com 91

FIGURE 12. Perineal ultrasound, the image aligned vertically. The ES shows mixed (hyperechogenity is dominant) echogenicity and the last part (subcutaneous segment, arrow) is lower than the internal sphincter. FIGURE 14. Internal sphincter defect (endovaginal route). Defect between 10 and 3 o’clock on lithotomy position from the site of optimal visualization is performed to allow at this level of the anal canal (long arrows). Note that the visualization of the full length of the anal canal. The area is posterior half of the ring is normal (short arrows). scanned in the transverse and sagittal planes. When the probe is appropriately placed in the midsagittal plane, the anorectum In female patients, the transvaginal probe is placed low will be fully visualized. The anterior and posterior aspects of in the vagina and angled posteriorly to image the anal canal. the IS are seen as 2 well-defined, almost parallel hypoechoic Transvaginal sonography is a reliable method for evaluating structures (Fig. 11). The ES appears as a structure of mixed the anal sphincter (Fig. 14), with an accuracy equivalent to that echogenicity (Fig. 12). Posteriorly, the ES continues to the of the endoanal technique.26 Also, transvaginal sonography is anorectal junction, with the puborectalis muscle (‘‘sling’’) preferable for the evaluation of the perineal body and perianal representing the most proximal part of the ES (Fig. 13). The inflammatory processes. It can be used to detect rectal lesions anterior ES is 0.5 to 1.5 cm shorter than the posterior, owing and to stage rectal tumor. The accuracy of this technique is to the paucity of anterior fibers in the puborectalis sling.7,28 It equivalent to that of conventional transrectal sonography for is very important to take this into account, when the integrity staging of rectal cancer.29 Endoanorectal, transvaginal, and of the ES is assessed in the most proximal area near the transperineal US can diagnose or confirm many of the anorectal junction, so that normal variant anatomy is not misconstrued as an ES defect.

FIGURE 13. Perineal ultrasound, midsagittal plane, the image FIGURE 15. External sphincter shows focal hypoechogenic is aligned vertically. The anorectum is seen. Note that ES areas (arrows), consistent with ﬁbrosis where sphincter is continuous with the puborectalis muscle (PR, arrow), reparation was performed 15 years ago in this still representing the most proximal part of ES. incontinent patient.

92 www.ultrasound-quarterly.com * 2011 Lippincott Williams & Wilkins

impact on quality of life.32 Incontinence can be divided into passive (when there is leakage without awareness, which is often related to IS damage) or urge (when there is inability to defer defacation, which is more likely to be due to ES damage).33 Although it can affect all ages and both sexes and has many causes, fecal incontinence is most common in women after injury to the anal sphincter, usually its anterior portion, during vaginal delivery.34,35 In the elderly, incontinence is common and neurogenic in origin. In younger age groups, diminished rectal compliance, anal surgery, diarrhea, and neu- rologic diseases may be specific causes. Endoanorectal ultrasonography represents a powerful tool in distinguishing incontinent patients with intact anal sphincters and those with sphincter lesions. It may confirm the presence or absence of sphincter defects and residual defects after surgery (Fig. 15). Defects are documented using a clock- face orientation with anterior at 12 o’clock; also, number, site, longitudinal extension, and radial angle of the defect should FIGURE 16. Internal sphincter defect (endoanal route). Defect be recorded. An IS defect appears as an interruption in the appears as an interruption in the hypoechoic ring at this level of hypoechoic ring (Fig. 16), and ES defect appears as a break, the anal canal. Note that the anterior half of the ring is missing. usually hypoechoic in the normal texture of the echogenic muscle ring (Fig. 17). The thickness of the IS on EARUS can abnormalities seen in patients with OD. However, the presence 30 indirectly show whether the neuropathic anal sphincter is of enterocele is detected by routes other than EARUS. At our present.36 Differences in echogenicity (Fig. 18) and thickness institution, we use endoanorectal and transperineal routes in of the sphincters and other local alterations should be carefully male patients and endoanorectal, transperineal, and transvaginal assessed and should always be reported. routes in female patients regardless of the abnormality in the A limitation of EARUS continues to be the identification pelvic floor. of ES atrophy in patients with idiopathic fecal incontinence because of the vague contours of the muscle ring.16 Which Probe: Thin or Thick? Perineal body thickness in female patients can be mea- The main limitations of EARUS are stricture, painful sured to determine whether an anterior sphincter defect is conditions, or stenotic anus. These caveats are largely caused present. A gloved lubricated finger is inserted into the vagina by the size of regular rectal probes with an average diameter of and held with gentle pressure against the posterior vaginal 17 to 20 mm.31 These limitations could be overcome by using V V wall. At the level of the midanal canal, the distance between thin diameter below 1 cm probes instead of thick ones. the sonographic reflection of the fingertip and the inner border of the IS and then the outer border of the subepithelial layer 37 SONOGRAPHIC FEATURES is measured on the frozen US image. The thickness of more than 12 mm is mostly consistent with a normal anterior FecalVAnalVIncontinence sphincter complex (Fig. 19). Detection of an anterior ES Anal incontinence, the involuntary loss of gas, liquid, defect may even be improved by the demarcation and gentle or solid stool, is a distressing condition with a substantial pressure on the posterior vaginal wall.38

FIGURE 17. A, External sphincter is absent (defective) between 10 and 1 o’clock on lithotomy in the whole segment (arrows) (between calipers, IS shows ovoid conﬁguration). B, Subcutaneous portion of the ES shows a focal hypoechogenic defect at 9 o’clock on lithotomy (between arrows) due to complication of LIS operation.

* 2011 Lippincott Williams & Wilkins www.ultrasound-quarterly.com 93

FIGURE 18. Internal sphincter is thin and shows focal hyperechogenic areas (arrows) but no interruptions; this is consistent with degenerative changes. FIGURE 20. Rectoanal intussusception. Lower rectum (arrows) forms a ring around the anal canal (AC) on straining. Obstructed Defecation Nearly half of constipated patients experience OD with characterized mainly by difﬁcult evacuation of stools, forced fractioned bowel motion, straining at stool, and sense of in- straining during defecation, and a sensation of incomplete complete evacuation.39 Obstructed defecation could result evacuation after defecation, or to an elliptical distortion 40Y42 from rectal intussusception, rectocele, enterocele, rectal inter- (Fig. 21) of the sphincter. A rectocele is deﬁned as a nal mucosal prolapse, solitary rectal ulcer, or anismus. These bulgeoftheanteriorrectalwallmorethan2cminsizeon abnormalities can easily be diagnosed by ultrasound (endoanal, straining. This abnormality can be diagnosed on transperineal endorectal, transvaginal, transperineal routes).30 Intussuscep- or transvaginal routes as a semicircumferential anterior hypo- tion is diagnosed by direct sign, such as double hyperechoic echoic area between the rectum and the vagina on straining, circle in the lower rectum representing the muscular layer of possibly due to an interruption of the rectal muscle layer both intussuscipiens and intussusceptum (Fig. 20), and/or by (Fig. 22), or as an endoanal bulk of mixed echogenicity due indirect signs, such as hypertrophy of the internal anal sphincter, possibly due to hypercontractility of the smooth muscle aimed at contrasting dyschezia, which is an entity

FIGURE 19. At the level of the mid anal canal, the distance between the sonographic reflection of the finger tip and the FIGURE 21. Indirect sign of the intussusception. Internal inner border of the IS and the outer border of the subepithelial sphincter (IS) shows ovoid configuration (elliptical distortion). layer (between calipers) displayed 13.8 mm, showing normal Also, note that 2 separate focal hyperechoic areas perineal body thickness (arrow, finger tip echo). (degenerative changes) are present in the sphincter (arrows).

94 www.ultrasound-quarterly.com * 2011 Lippincott Williams & Wilkins

FIGURE 22. Rectocele. A bulk of mixed echogenicity (between calipers) on straining is seen in the region of rectoanogenital septum at the cranial level of anal canal. FIGURE 24. The immediate postanal area of the distal rectum is seen. This is technically challenging area to achieve a to a concomitant rectal internal mucosal prolapse. Enterocele perpendicular image. P indicates prostate. can be diagnosed on transperineal or transvaginal routes as an oval hypoechoic area (intestinal fluid) surrounded by a Tumors hyperechoic layer (intestinal wall) between the anorectum and the vagina, which is more evident on straining (Fig. 22). A Rectal Tumors solitary rectal ulcer can be diagnosed on EARUS by detecting Endoanorectal ultrasonography is the method of choice an area of mixed echogenicity mimicking a polypoid lesion at for the preoperative evaluation and staging of rectal cancer.44 the level of the distal rectum.30 Anismus is also called a para- But 4 areas where it may be technically challenging to achieve doxical puborectalis contraction or spastic pelvic floor syn- a perpendicular imaging plane are (1) the immediately post- drome. The diagnosis of anismus has been based historically on anal area of the distal rectum, especially posteriorly (Fig. 24); finding the inappropriate puborectal contraction. It was assumed (2) the rectosigmoid junction, where flexible transducers are that this would be visible during proctography as a persistent likely to provide more accurate results; (3) the transverse rectal indentation posteriorly just above the anorectal junction. In folds (valves of Houston); and (4) the bowel that has been practice, this is a poor predictor.43 Under ultrasound, anismus is plicated surgically or by scarring. Mostly, this situation causes defined as the difference in millimeters between the distance no problems with the newer US machines. of the inner edge of the puborectalis muscle posteriorly and Classically, rectal tumors are staged according to the probe at rest and on straining being less than 0.5 mm TNM classification.45 Sarcoma, lymphoma, carcinoids, (Fig. 23).30 and melanoma cannot be applied to this staging system.

FIGURE 23. The measurement of the difference between the distance of the inner edge of the puborectalis muscle (PR) posteriorly and the probe anteriorly (between calipers) are 10 mm at rest (A) and 14.2 mm on straining (B) show normality.

* 2011 Lippincott Williams & Wilkins www.ultrasound-quarterly.com 95

FIGURE 25. T1 rectal tumor on grayscale (A) and color Doppler (B) US images. A hypoechogenic polypoid mass, which is conﬁned to the submucosa with no extension of the muscularis propria is clearly seen (between calipers). Also, the mass shows moderate vascularity especially near the submucosal portion (arrow).

Endoanorectal ultrasonography is used for detecting T (tumor) rectal fat. Approximately 10% to 35% of stage T2 lesions are and N (nodal condition). likely to demonstrate regional lymph node involvement.48Y50 A stage T1 tumor is diagnosed when it is conﬁned to A stage T3 tumor is diagnosed by extension beyond the submucosa with no extension to the muscularis propria the muscularis propria into the perirectal fat (Fig. 27). These (Fig. 25). Only 6% to 11% of stage T1 lesions are likely to patients should probably receive adjuvant preoperative demonstrate lymphatic spread.46Y49 therapy.46 A stage T2 tumor is diagnosed by extension through A stage T4 tumor is diagnosed by invasion of adjacent the submucosa into the muscularis propria (Fig. 26) but is organs or the pelvic side wall (Fig. 28). Endoanorectal US conﬁned to the bowel wall. The muscularis propria is expan- may be able to demonstrate the vagina, uterus, and bladder ded and thickened with preservation of the hyperechoic peri- in females and the Denonvilliers fascia, prostate gland, bladder, and seminal vesicles in males. Lymph node metastases increase with T stage. T3 and T4 lesions are likely to demonstrate approximately 50% prevalence rates of lymph node involvement.48,49

FIGURE 26. T2 rectal tumor. A hypoechogenic lobulated mass is detected in the right and posterior side of the rectal wall FIGURE 27. T3 rectal tumor. A hypoechogenic lobulated (between calipers). Tumor is extended through the submucosa mass (M) is detected in the right side of the rectal wall. into the muscularis propria and has effaced this muscular Extension beyond the muscularis propria into the perirectal layer (arrows), but it is still conﬁned to the bowel wall. fat is clearly seen (arrows).

96 www.ultrasound-quarterly.com * 2011 Lippincott Williams & Wilkins

It is important for the therapeutic management of early rectal cancers to classify the depth of invasion of the submucosa as slight or massive.51 This depth is expressed as ‘‘sm’’ levels according to the Japanese classiﬁcation. It is possible to distinguish an early invasive tumor and to differentiate slight (tumor limited to the upper third of the hyperechoic layer representing the submucosa) from massive submucosal invasion (tumor extended to the middle or lower third of the submucosa). When the depth of submucosal invasion is slight (sm-s), there is no risk of lymph node involvement, and cure can be achieved by local resection. However, if the depth of submucosal invasion is massive (sm-m), risk of lymphatic metastasis is substantial and an oncologic resection is necessary.51Y53 Although criteria for lymph node invasion vary among investigators, classic nodal involvement is diagnosed by identifying well-deﬁned hypoechoic lesions in the perirectal area,14 FIGURE 28. T4 rectal tumor. A hypoechogenic lobulated mass larger round nodes (91.0 cm) rather than oval nodes with loss extending into the pelvic side wall (arrows) is seen. Bladder of the central hyperechoic fat (Fig. 29). But the operator was also invaded (not shown). M indicates mass. should keep in mind that the tumor may involve small nodes (G2mmindiameter)54,55 and enlarged nodes may result from

FIGURE 29. A, Hypoechogenic lymphadenopathy (between calipers) in the perirectal area is detected in a patient with anastomotic recurrence (the same case as in Fig. 31). Classic nodal involvement is seen by identifying well-defined but lobulated contour with circular shape and loss of the hilar hyperechogenicity. Also, note that node is larger than 1.0 cm. B, Tru-cut biopsy of the lymphadenopathy is seen (arrows, echogenic needle in the target). C, Histopathologic examination of lymphadenopathy (Â200 high-power field). Invasive malignant glands of colonic type are seen in a desmoplastic stroma devoid of any histologic landmarks suggestive of the origin of the tissue (pathologic figure is from Dr M. Kemal Sarman, pathologist). D, Another patient with rectum tumor showed a small lymphadenopathy (arrow) in the perirectal area, which is 15 cm cranial from the anal verge, with loss of the hilar hyperechogenicity. Color Doppler ultrasound (CDUS) examination displayed no color coding in the node. Sometimes, CDUS is preferred to differentiate a lymph node from a vascular structure, which can be misdiagnosed as lymphadenopathy.

* 2011 Lippincott Williams & Wilkins www.ultrasound-quarterly.com 97

FIGURE 32. High level of the anal canal. Concentrically FIGURE 30. After neoadjuvant therapy (the same case as in spreading, hypoechogenic, inﬁltrative tumor is penetrating Fig. 27), the tumor has disappeared, but the thickening the IS. Note that a small lymphadenopathy is also present of the wall, especially in the muscular layer, is seen on the (between calipers). right side (arrows).

matory and fibrotic changes in the rectal wall. Therefore, some inflammatory change.56 For these reasons, biopsy of suspicious authors do not apply TNM staging when inspecting lesions nodes may need to be performed. for their response to preoperative chemoradiotherapy.14 In- Limitations of EARUS imaging include overstaging stead, they assess for evidence of tumor regression from the caused by desmoplastic reaction and understaging in cases surrounding organs, in particular the anal sphincters, vagina, of microscopic infiltration in or through the wall.57 The results and prostate. Also, the location of the tumor, stenotic tumors, of tumor staging with EARUS are largely dependent on the postsurgical changes, postbiopsy changes, or hemorrhage can experience of the operator. Awareness of the technical and contribute to false staging of the tumor. anatomic factors that produce overestimation or underesti- A local recurrence (Fig. 31) of rectal cancer after pre- mation of depth of tumor invasion will allow more accurate sumed curative resection occurs in 10% to 15% of cases, tumor staging and thus facilitate clinical management. Most usually within the first 2 years after surgery.14 Endoanorectal postirradiation (Fig. 30) and inflammatory changes appear as US may be useful in the detection of suspected local recur- hypoechoic areas in the rectal wall and can easily be confused rence, when no mucosal lesions are seen during surveillance with carcinoma.44 In other words, neoadjuvant therapy for sigmoidoscopy. Endoanorectal US can detect local tumor rectal cancer results in tumor regression/necrosis and inflam- recurrence with a high diagnostic accuracy (990%).58

FIGURE 31. A, Tumor recurrence at the anastomotic region, 12 cm cranial to the anal verge. A hypoechogenic lobulated mass (M), extending into the pelvic side is detected (between calipers). B, Tru-cut biopsy of the mass is seen (arrows, echogenic needle in the target). C, Histopathologic examination of anastomotic mass (Â200 high-power ﬁeld). Invasive malignant glands of colonic type are evident among the smooth muscle bundles (pathologic ﬁgure is from Dr M. Kemal Sarman, pathologist).

98 www.ultrasound-quarterly.com * 2011 Lippincott Williams & Wilkins

FIGURE 33. Middle level of the anal canal. A, A hypoechogenic tumor is seen in the right side of the canal (arrows), which is invading IS and sphincter contour can not be detected at this level. B, After radiotherapy, sphincter contour is detected (arrow), but some hypoechogenicity is seen (scar tissue) near the sphincter (between calipers). The patient is still on follow-up.

Anal Tumors biopsy, is usually necessary to distinguish recurrent tumor Endoanorectal US may provide helpful information such from scar tissue.59,60 as tumor size, location, depth of mural penetration, and extent beyond the anal canal into adjacent structures (Fig. 32). The Fistula In Ano current TNM staging for anal tumor is based on tumor size Perianal abscess and fistula in ano are thought to arise (T1, tumor e2 cm; T2, tumor 92 cm and e5 cm; T3, tumor from infection in small intersphincteric anal glands with abscess 95 cm; T4, invasion of adjacent organ, regardless of the size representing acute manifestation. These intersphincteric anal of the tumor).45 Endoanorectal US can provide an alterna- glands are predominantly located at the dentate line, in other tive tumor staging based on depth of tumor invasion.59 In words, at the mid anal canal at a level at which intersphincteric this staging; T1 is confined to the subepithelium, T2 is limited glands are most densely populated.61 Perianal inflammatory to the sphincter muscles, T3 penetrates through the ES, and disease is particularly common in Crohn disease, with as many T4 involves adjacent structures. This classification may be of as one third of all patients presenting with perianal fissures, value in determining the response to radiotherapy.14 After fistulas, or perianal abscesses at some stage.62 In patients who treatment, EARUS may have a role in monitoring the patient do not have Crohn disease, perianal inflammatory disease is for local recurrence as early tumor growth may be difficult to still largely cryptoglandular in origin, likely to be caused by palpate within the scar tissue (Fig. 33). In the latter situation, obstruction of intersphincteric anal glands by fecal material, by repeat studies to document an increase in the size of a lesion foreign bodies, or by trauma, with subsequent stasis and in- raising suspicions of recurrent disease, or ultrasound-guided fection. Infection may spread caudally to the skin at the anal

FIGURE 34. Intersphincteric fistula. A, Two separate hypoechogenic fistulas (arrows) are seen in the intersphincteric areas at 9 and 10 o’clock on lithotomy position. B, Another case shows a hypoechogenic large intersphincteric (between long arrows) and thin sphincteric (between short arrows) (through which it is opening into the anal canal) components of the fistula at 6 o’clock on lithotomy position.

* 2011 Lippincott Williams & Wilkins www.ultrasound-quarterly.com 99

FIGURE 35. Trans-sphincteric fistula. Three sequential images. A, A hypoechogenic tract (arrow) is crossing (B) ES (arrow) at the 3-o’clock position forming an abscess formation (C) in the left ischioanal fossa (arrows). Note also a glandular crypt at the 10-o’clock position on A (arrowhead). margin, cranially into the rectal wall, or laterally through the ES cation system, there are 4 main subtypes: intersphincteric into the ischioanal space.63 (between the IS and ES) (Fig. 34), trans-sphincteric (crossing Documentation of fluid collections and the relationship both the IS and ES into the ischiorectal or ischioanal fossa) of inflammatory tracts to the sphincter mechanism is impor- (Fig. 35), suprasphincteric (a tract passing upward in the tant for surgical treatment. If the ischioanal and ischiorectal intersphincteric plane to a point above the puborectalis muscle fossae are unaffected, the disease is likely to be confined to the where it tracks laterally and caudally into the ischioanal fossa) sphincter complex. The outcome after simple surgical man- (Fig. 36), and extrasphincteric (a tract passing directly from the agement is favorable. If there is any track or abscess within perineal skin to the rectum outside both sphincters) (Fig. 37). the ischioanal and the ischiorectal fossa, it is usually related Secondary tracts, location of the internal opening, and fluid- to a complex perianal fistula. Correspondingly, more complex abscess collections are also documented. An internal opening surgery may be required that may threaten continence or may is defined as an opening into the rectum or anus, whereas an require fecal diversion (colostomy) to allow healing. When the external opening is defined as an opening into the skin. Because track traverses the levator plate, a source of pelvic sepsis should by definition, a fistula is an abnormal communication between be sought. any 2 epithelial lined surfaces, any tract which is blind-ending For this purpose, primary fistula tract is traditionally and does not have both external and internal openings is classified according to Parks et al.64 This classification system labeled as a sinus tract. Complex horseshoe communications provides an anatomic description of fistulous tracts, which acts (Fig. 38) should also be reported, which can be easily detected as a guide to operative treatment. According to this classifi- by EARUS.65 A transperineal or transvaginal scan is added if an inflammatory process extends beyond the field of view allowed by the endoanorectal probe. The transducer is placed over any

FIGURE 36. Suprasphincteric ﬁstula. A hypoechogenic tract (arrows) passing upward in the intersphincteric plane to a point FIGURE 37. Extrasphincteric ﬁstula. A hypoechogenic tract above the puborectalis muscle for which it will track laterally (arrows) extending between rectum and perineal skin is seen and caudally into the ischioanal fossa. outside both sphincters (IS and ES).

100 www.ultrasound-quarterly.com * 2011 Lippincott Williams & Wilkins

FIGURE 38. Hypoechogenic horseshoe fistula is seen in posterior perianal region in 2 separate patients (between arrows) (A and B). The fistula in B includes lots of echogenic air bubbles in the lumen. external opening, and oblique images are obtained as needed EARUS with multiplanar viewing makes sphincter assess- to show any collections and to follow the course of fistulous ment easier and results in improved diagnostic confidence.74 and sinus tracts in relation to the sphincter mechanism. Rou- It can also be used for postsurgical evaluation to assess resi- tine pelvic sonography should be performed to assess any high dual sphincter defects.73 pelvic abscess beyond the range of the transperineal or trans- Endoanorectal ultrasonography can be used in the vaginal probe. evaluation of OD, as an alternative to defacography and for screening of evacuation dysfunctions.75 In general, all imaging Perianal Inflammations Caused by Inflammatory methods are complementary to each other.76 Bowel Disease Endoanorectal ultrasonography is said to be superior, Perianal lesions are seen in both Crohn disease and when compared with MRI, in assessing rectal tumors. MRI ulcerative colitis,66 mostly in the former. Perianal lesions may cannot reliably distinguish T1 from T2 tumors and sessile or 77,78 be either primary or secondary in Crohn disease.67 The former polypoid adenomas from T1 adenocarcinomas as some- include fissures and ulcers; strictures, abscesses, fistulas, and times do EARUS. Magnetic resonance imaging has an inherent skin tags are secondary lesions. This article will only focus advantage in the preoperative staging of rectal cancer because 79 on fistulas and abscesses. of its superiority in detecting lymph node metastases (N stage). The more complex the fistula tract, the higher the inci- We think that EARUS and MRI are complementary to each dence of abscess formation. Fistulas are commonly described using the classic nomenclature as described above; the trans- sphincteric is the commonest type of fistula with an incidence of 29% to 47%.68,69 Rectovaginal fistulas are mostly low, even anovaginal in location.70 Rectovaginal fistulas may also be described as intersphincteric, trans-sphincteric, suprasphincteric, or extrasphincteric.71 Alternatively, these fistulas can be classified according to size: small (G0.5 cm), medium (0.5Y2.5 cm), and large (92.5 cm).70 In most cases, the opening of the fistula is located anteriorly in the rectum or anus (Fig. 39). The higher the fistulas, the worse the symptoms. Rare cases of high fistula are often a result of Crohn disease involving the sigmoid colon or small bowel, with a fistula tract to the apex of the vagina.

ADVANTAGES AND DISADVANTAGES OF EARUS COMPARED TO OTHER IMAGING METHODS With respect to anal sphincters, EARUS is superior to MRI in diagnosing IS defects.72 For the depiction of ES defects, EARUS and MRI are both useful.73 In other words, EARUS and endoanal MRI are equivalent in diagnosing ES 68 defects. Generalized atrophy of the sphincter muscles can FIGURE 39. Anovaginal ﬁstula (AC, anal canal; V, vagina) be assessed with MRI and also with EARUS. In addition, 3D with echogenic air bubbles in track (arrows).

* 2011 Lippincott Williams & Wilkins www.ultrasound-quarterly.com 101

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Mihmanli et al Ultrasound Quarterly & Volume 27, Number 2, June 2011 other, in that EARUS is more accurate in determining bowel EARUS is also of value in the assessment of perianal sepsis. wall penetration of the tumor, whereas MRI is more accurate in It can help the surgeon in planning the surgical strategy by the evaluation of lymph node involvement. On the other hand, depicting the anatomy of the fistula tract. preoperative 3-T MRI in rectal cancer patients for staging with conventional 2D and multiplanar reconstruction 3D T2- weighted imaging protocols has shown no significant differences in accuracy of T and N category staging and overall REFERENCES 1. Law PJ, Bartram CI. Anal endosonography: technique and normal image quality, as determined by degree of artifact. In addition, anatomy. Gastrointest Radiol. 1989;14:349Y535. 3D T2-weighted imaging protocol has had limitations about 2. Sultan AH, Kamm MA, Hudson CN, et al. Anal sphincter disruption lesion conspicuity.80 After neoadjuvant therapy, although the during vaginal delivery. N Engl J Med. 1993;329:1905Y1911. role of both methods in the assessment of residual rectal 3. Ziech M, Felt-Bersma R, Stoker J. Imaging of perianal fistulas. 81 Clin Gastroenterol Hepatol. 2009;7:1037Y1045. tumors remains uncertain, EARUS may be useful in the 4. Stoker J, Halligan S, Bartram C. Pelvic floor imaging: state of the art. detection of suspected local recurrence when no mucosal Radiology. 2001;218:621Y641. lesions are seen during surveillance sigmoidoscopy. In addi- 5. Godlewski G, Prudhomme M. Embryology and anatomy of the tion, EARUS can detect local tumor recurrence with a high anorectum: basis of surgery. Surg Clin North Am. 2000;80:319Y343. diagnostic accuracy (990%).58 Positron emission tomography 6. Al-Ali S, Blyth P, Beatty S, et al. Correlation between gross anatomical topography, sectional sheet plastination, microscopic anatomy and is a useful adjunct to systemic and regional staging, especially endoanal sonography of the anal sphincter complex in human males. of a recurrent rectal cancer, but is rarely used as part of loco- J Anat. 2009;215:212Y220. regional staging of rectal cancer preoperatively.82,83 Positron 7. Rociu E, Stoker J, Eijkemans MJ, et al. Normal anal sphincter anatomy emission tomography and CT have limitations, such as low and age- and sex-related variations at high-spatial-resolution endoanal 84 MR imaging. Radiology. 2000;217:395Y401. spatial resolution and high cost. 8. Milligan ETC, Morgan CN. Surgical anatomy of the anal canal, with When EARUS is compared with MRI for staging anal special references to anorectal fistula. Lancet. 1934;2:1150Y1156. tumors, EARUS is found to be superior to MRI for detect- 9. Goligher JC, Leacock AG, Brossy JJ. The surgical anatomy of the anal ing superficial tumors, but MRI is better for diagnosing N canal. Br J Surg. 1955;43:51Y61. staging.85 10. Dalley AF II. The riddle of the sphincters: the morphophysiology of the anorectal mechanism reviewed. Am Surg. 1987;53:298Y306. When EARUS is compared with physical examina- 11. Oh C, Kark AE. Anatomy of the external anal sphincter. Br J Surg. tion, the accuracy of EARUS is significantly higher than that 1972;59:717Y723. of physical examination in detecting the primary tract and 12. Murad-Regadas SM, Regadas FS, Rodrigues LV, et al. The role of horseshoe extension and in localizing the internal opening.86 3-dimensional anorectal ultrasonography in the assessment of anterior transsphincteric fistula. Dis Colon Rectum. 2010;53:1035Y1040. When EARUS is compared with clinical examination and 13. Santoro GA, Fortling B. New technical developments in endoanal and fistulography in preoperative classification of the fistula in ano, endorectal ultrasonography. In: Santoro GA, Di Falco G, eds. Benign EARUS is found to be the most accurate diagnostic modality Anorectal Diseases. Diagnosis With Endoanal and Endorectal in the preoperative assessment of the fistula in ano.87 Endo- Ultrasonography and New Treatment Options. Milan, Italy: Springer; anorectal ultrasonography is well comparable to MRI both in 2006:13. 88 14. Giovannini M, Ardizzone S. Anorectal ultrasound for neoplastic classifying and in describing the topography of an anal fistula. and inflammatory lesions. Best Pract Res Clin Gastroenterol. Computed tomography is of limited usefulness and offers little 2006;20:113Y135. Y additional clinical information.89 92 15. Prager RW, Ijaz Uz, Gee AH, et al. Three-dimensional ultrasound imaging. Proc Inst Mech Eng H. 2010;224:193Y223. 16. Santoro GA, Fortling B. The advantages of volume rendering in three-dimensional endosonography of the anorectum. Dis Colon SUMMARY AND CONCLUSIONS Rectum. 2006;50:359Y368. Endoanorectal, transvaginal, and transperineal US exam- 17. Cheong DM, Nogueras JJ, Wexner SD, et al. Anal endosonography inations are components of routine sonographic procedures for for recurrent anal fistulas: image enhancement with hydrogen peroxide. Y the evaluation of patients with disease in the perianal region. Dis Colon Rectum. 1993;36:1158 1160. 18. Poen AC, Felt-Bersma RJ, Eijsbouts QA, et al. Hydrogen The diagnostic accuracy of EARUS is high, and the procedure is peroxideYenhanced transanal ultrasound in the assessment of well tolerated. fistula-in-ano. Dis Colon Rectum. 1998;41:1147Y1152. For the local staging of the rectal carcinoma, EARUS 19. Ratto C, Gentile E, Merico M, et al. How can the assessment of is probably the most accurate tool. The optimal method of fistula-inano be improved? Dis Colon Rectum. 2000;43:1375Y1382. 20. Kruskal JB, Kane RA, Morrin MM. Peroxide-enhanced anal management of rectal carcinoma, type of resection, and can- endosonography: technique, image interpretation, and clinical didacy for neoadjuvant therapy are accurately directed by applications. Radiographics. 2001;21:173Y189. EARUS. Also, EARUS-guided FNA can be used to detect sus- 21. Tsai SK, Lee TY, Mok MS. Gas embolism produced by hydrogen pected local recurrence. In addition to providing accurate T and peroxide irrigation of an anal fistula during anesthesia. Anesthesiology. Y N staging, EARUS allows an assessment of the IS and ES. 1985;63:316 317. 22. Schwab C, Dilworth K. Gas embolism produced by hydrogen Endoanorectal ultrasonography examination is easy to peroxide abscess irrigation in an infant. Anaesth Intensive Care. perform and is a reliable tool in the assessment of anorectal 1999;27:418Y420. disease. When compared with the routine digital examination, 23. Kim Y, Park YJ. Three-dimensional endoanal ultrasonographic it causes less discomfort. It provides an alternative tumor assessment of an anal fistula with and without H2O2 enhancement. World J Gastroenterol. 2009;15:4810Y4815. staging based on depth of tumor invasion and may have a role 24. Buchanan GN, Bartram CI, Williams AB, et al. Value of hydrogen in monitoring the patient for local recurrence as early tumor peroxide enhancement of three-dimensional endoanal ultrasound in growth may be difficult to palpate within the scar tissue. fistula-in-ano. Dis Colon Rectum. 2005;48:141Y147.

102 www.ultrasound-quarterly.com * 2011 Lippincott Williams & Wilkins

25. Barreiros AP, Hirche TO, Ignee A, et al. Indications and limitations 52. Nascimbeni R, Nivatvongs S, Larson DR, et al. Long-term survival after of perineal ultrasound examination. Scand J Gastroenterol. local excision for T1 carcinoma of the rectum. Dis Colon Rectum. 2010;45:764Y765. 2004;47:1773Y1779. 26. Stewart LK, Wilson SR. Transvaginal sonography of the anal sphincter: 53. Kikuchi R, Takano M, Takagi K, et al. Management of early invasive reliable or not? AJR Am J Roentgenol. 1999;173:179Y185. colorectal cancer: risk of recurrence and clinical guidelines. 27. Valsky DV, Yagel S. Three-dimensional transperineal ultrasonography Dis Colon Rectum. 1995;38:710Y717. of the pelvic floor: improving visualization for new clinical applications 54. Hinder JM, Chu J, Bokey EL. Use of transrectal ultrasound to evaluate and better functional assessment. JUltrasoundMed. 2007;26: direct tumor spread and lymph node status in patients with rectal 1373Y1387. cancer. Aust N ZJ Surg. 1990;60:19Y23. 28. Nielsen MB, Pedersen JF, Hauge C, et al. Endosonography of the anal 55. Spinelli P, Schiavo M, Meroni E, et al. Results of EUS in detecting sphincter findings in healthy volunteers. AJR Am J Roentgenol. perirectal lymph node metastases of rectal cancer: the pathologist makes 1991;157:1199Y1202. the difference. Gastrointest Endosc. 1999;49:754Y758. 29. Berton F, Gola G, Wilson SR. Perspective on the role of transrectal 56. Beynon J, Mortensen NJM, Foy DMA, et al. Preoperative assessment and transvaginal sonography of tumors of the rectum and anal canal. of mesorectal lymph node involvement in rectal cancer. Br J Surg. AJR Am J Roentgenol. 2008;190:1495Y1504. 1989;76:276Y279. 30. Brusciano L, Limongelli P, Pescatori M, et al. Ultrasonographic patterns 57. Kim J, Yu CS, Jung HY, et al. Source of errors in the evaluation of early in patients with obstructed defaecation. Int J Colorectal Dis. 2007; rectal cancer by endoluminal ultrasonography. Dis Colon Rectum. 22:969Y977. 2001;44:1302Y1309. 31. Weyandt GH, Eggert AO, Brocker EB, et al. The sonographic appearance 58. Muller C, Kahler G, Scheele J. Endosonographic examination of of the anal canal by endoanal sonography with a 7 mm probe. gastrointestinal anastamoses with suspected locoregional tumor Coloproctology. 2004;26:15Y20. recurrence. Surg Endosc. 2000;14:45Y50. 32. Fuchsjager MH, Maier AG. Imaging fecal incontinence. Eur J Radiol. 59. Goldman S, Norming U, Svensson C, et al. Transanorectal ultrasonography 2003;47:108Y116. in the staging of anal epidermoid carcinoma. Int J Colorect. 1991;6: 33. Engel AG, Kamm MA. Relationship of symptoms in fecal incontinence 152Y157. to specific sphincter abnormalities. Int J Colorectal Dis. 1995;10: 60. Finne CO. The role of endorectal sonography in the follow-up of 152Y155. malignant disease of the anorectum. Semin Colon Rectal Surg. 34. Christensen AF, Nyhuus B, Nielsen MB, et al. Three-dimensional anal 1995;6:86Y93. endosonography may improve diagnostic confidence of detecting damage 61. Lori K, Stewart LK, McGee J, et al. Transperineal and transvaginal to the anal sphincter complex. Br J Radiol. 2005;78:308Y311. sonography of perianal inflammatory disease. AJR Am J Roentgenol. 35. Nichols CM, Gill EJ, Nguyen T, et al. Anal sphincter injury in women 2001;177:627Y632. with pelvic floor disorders. Obstet Gynecol. 2004;104:690Y696. 62. Kornbluth A, Salomen P, Sachar DB. Crohn’s disease. In: Sleisenger MH, 36. Schaffzin DM, Wong WD. Surgeon-performed ultrasound: endorectal Fordtran JS, eds. Gastrointestinal Disease: Pathophysiology, Diagnosis, ultrasound. Surg Clin North Am. 2004;84:1127Y1149. Management. 5th ed. Philadelphia, PA: Saunders; 1993:1276. 37. Oberwalder M, Thaler K, Baig MK, et al. Anal ultrasound and 63. Gordon PH. Anorectal abscesses and fistula-in-ano. In: Gordan PH, endosonographic measurement of perineal body thickness. Surg Nivatvongs S, eds. Principles and Practice of Surgery for the Colon, Endosc. 2004;18:650Y654. Rectum and Anus. St Louis, MO: Quality Medical; 1992:222. 38. Titi MA, Jenkins JT, Urie A, et al. Perineum compression during EAUS 64. Parks AG, Gordon PH, Hardcastle JD. A classification of fistula-in-ano. enhances visualization of anterior anal sphincter defects. Colorectal Br J Surg. 1976;63:1Y12. Dis. 2009;11:625Y630. 65. Deen KI, Williams JG, Hutchinson R, et al. Fistulas in ano: endoanal 39. Karlbom U, Lundin E, Graf W, et al. Anorectal physiology in relation ultrasonographic assessment assists decision making for surgery. to clinical subgroups of patients with severe constipation. Colorectal Dis. Gut. 1994;35:391Y394. 2004;6:343Y349. 66. Hamzaoglu I, Hodin RA. Perianal problems in patients with ulcerative 40. Nielsen MB, Rasmussen OO, Pedersen JF, et al. Anal endosonographic colitis. Inflamm Bowel Dis. 2005;11:856Y859. findings in patients with obstructed defecation. Acta Radiol. 67. Hughes LE. Surgical pathology and management of anorectal Crohn’s 1993;34:35Y38. disease. J R Soc Med. 1978;71:644Y651. 41. Marshall M, Halligan S, Fotheringham T, et al. Predictive value of 68. Scott HJ, Northover JM. Evaluation of surgery for perianal Crohn’s internal anal sphincter thickness for diagnosis of rectal intussusception fistulas. Dis Colon Rectum. 1996;39:1039Y1043. in patients with solitary rectal ulcer syndrome. Br J Surg. 2002;89: 69. Williamson PR, Hellinger MD, Larach SW, et al. Twenty-year review 1281Y1285. of the surgical management of perianal Crohn’s disease. Dis Colon 42. Dvorkin LS, Chan CL, Knowles CH, et al. Anal sphincter morphology Rectum. 1995;38:389Y392. in patients with full thickness rectal prolapse. Dis Colon Rectum. 70. Singh B, McC Mortensen NJ, Jewell DP, et al. Perianal Crohn’s disease. 2004;47:198Y203. Br J Surg. 2004;91:801Y814. 43. Stoker J, Bartram CI, Halligan S. Imaging of the posterior pelvic floor. 71. Radcliffe AG, Ritchie JK, Hawley PR, et al. Anovaginal and rectovaginal Eur Radiol. 2002;12:779Y788. fistulas in Crohn’s disease. Dis Colon Rectum. 1988;31:94Y99. 44. Kruskal JB, Kane RA, Sentovich SM, et al. Pitfalls and sources of error 72. Malouf AJ, Williams AB, Halligan S, et al. Prospective assessment in staging rectal cancer with endorectal US. Radiographics. of accuracy of endoanal MR imaging and endosonography in patients 1997;17:609Y626. with fecal incontinence. AJR Am J Roentgenol. 2000;175:741Y745. 45. Edge SB, Byrd DR, Compton CC, et al. eds. AJCC Cancer Staging 73. Dobben AC, Felt-Bersma RJ, ten Kate FJ, et al. Cross-sectional imaging Manual. 7th ed. New York, NY: Springer; 2010. of the anal sphincter in fecal incontinence. AJR Am J Roentgenol. 46. Beynon J, Foy DMA, Roe AM, et al. Endoluminal ultrasound in the 2008;190:671Y682. assessment of local invasion of rectal cancer. Br J Surg. 1986;73: 74. Cazemier M, Terra MP, Stoker J, et al. Atrophy and defects detection 474Y477. of the external anal sphincter: comparison between three-dimensional 47. Hager T, Gall FP, Hermanek P. Local excision of cancer of the rectum. anal endosonography and endoanal magnetic resonance imaging. Dis Colon Rectum. 1983;26:149Y151. Dis Colon Rectum. 2006;49:20Y27. 48. Hu KS, Harrison LB. Adjuvant therapy for resectable rectal 75. Piloni V, Spazzafumo L. Evacuation sonography. Tech Coloproctol. adenocarcinoma. Semin Surg Oncol. 2000;19:336Y349. 2005;9:119Y125. 49. Kim HJ, Wong WD. Role of endorectal ultrasound in the conservative 76. Schwandner O, Poschenrieder F, Gehl HB, et al. Differential diagnosis management of rectal cancers. Semin Surg Oncol. 2000;19:358Y366. in descending perineum syndrome. Chirurg. 2004;75:850Y860. 50. Billingham RP. Conservative treatment of rectal cancer: extending the 77. Brown G, Davies S, Williams GT, et al. Effectiveness of preoperative indication. Cancer. 1992;70:1355Y1363. staging in rectal cancer: digital rectal examination, endoluminal 51. Kudo S. Endoscopic mucosal resection of flat and depressed types of ultrasound or magnetic resonance imaging? Br J Cancer. 2004;91: early colorectal cancer. Endoscopy. 1993;25:455Y461. 23Y29.

* 2011 Lippincott Williams & Wilkins www.ultrasound-quarterly.com 103

78. Akasu T, Kondo H, Moriya Y, et al. Endorectal ultrasonography and 85. Otto SD, Lee L, Buhr HJ, et al. Staging anal cancer: prospective treatment of early stage rectal cancer. World J Surg. 2000;24:1061Y1068. comparison of transanal endoscopic ultrasound and magnetic resonance 79. Halefoglu AM, Yildirim S, Avlanmis O, et al. Endorectal imaging. Gastrointest Surg. 2009;13:1292Y1298. ultrasonography versus phased-array magnetic resonance imaging 86. Toyanaga T, Tanaka Y, Song JF, et al. Comparison of accuracy of for preoperative staging of rectal cancer. World J Gastroenterol. physical examination and endoanal ultrasonography for preoperative 2008;14:3504Y3510. assessment in patients with acute and chronic anal fistula. Tech 80. Kim H, Lim JS, Choi JY, et al. Rectal cancer: comparison of accuracy Coloproctol. 2008;12:217Y223. of local-regional staging with two- and three-dimensional preoperative 87. Tepes B, Cerni I. The use of different diagnostic modalities in diagnosing 3-T MR imaging. Radiology. 2010;254:485Y492. fistula-in-ano. Hepatogastroenterology. 2008;55:912Y915. 81. Mezzi G, Arcidiacono PG, Carrara S, et al. Endoscopic ultrasound 88. Gustafsson UM, Kahvecioglu B, Astro¨m G, et al. Endoanal ultrasound and magnetic resonance imaging for re-staging rectal cancer after or magnetic resonance imaging for preoperative assessment of anal radiotherapy. World J Gastroenterol. 2009;15:5563Y5567. fistula: a comparative study. Colorectal Dis. 2001;3:189Y197. 82. Heriot AG, Hicks RJ, Drummond EG, et al. Does positron emission 89. Ratto C, Grillo E, Parello A, et al. Endoanal ultrasound-guided surgery tomography change management in primary rectal cancer? A prospective for anal fistula. Endoscopy. 2005;37:722Y728. assessment. Dis Colon Rectum. 2004;47:451Y458. 90. SchratterYSehn AU, Lochs H, Vogelsang H, et al. Endosonography versus 83. Dobos N, Rubesin SE. Radiologic imaging modalities in the diagnosis computed tomography in the differential diagnosis of perianorectal and management of colorectal cancer. Hematol Oncol Clin North Am. complications in Crohn’s disease. Endoscopy. 1993;25:582Y586. 2002;16:875Y895. 91. Fishman EK, Wolf EJ, Jones B, et al. CT evaluation of Crohn’s disease: 84. Barbaro B, Vitale R, Leccisotti L, et al. Restaging locally advanced rectal effect on patient management. AJR Am J Roentgenol. 1987;148:537Y540. cancer with MR imaging after chemoradiation therapy. Radiographics. 92. YousemDM, Fishman EK, Jones B. Crohn disease: perirectal and perianal 2010;30:699Y716. findings at CT. Radiology. 1988;167:331Y334.

104 www.ultrasound-quarterly.com * 2011 Lippincott Williams & Wilkins