Downloaded from www.ajronline.org by 37.44.207.185 on 01/22/17 from IP address 37.44.207.185. Copyright ARRS. For personal use only; all rights reserved

Transrectal Sonography of Anal Sphincter Infiltration in Lower Rectal Carcinoma Andrea G. Maier 1 Soeren H. Kreuzer 1 Friedrich Herbst 2 Fritz Wrba 3 Wolfgang Schima 1 Martin A. Funovics 1 Bela K. Teleky 2 Gerhard L. Lechner 1

OBJECTIVE. The purpose of the study was to assess the accuracy of transrectal sonography in determining invasion by lower rectal cancer into the anal canal. SUBJECTS AND METHODS. Thirty-eight patients (14 women, 24 men; mean age, 65 years) with lower rectal cancer underwent transrectal sonography before surgery. Both depth of infiltration and tumor invasion into the anal canal were assessed, and results were compared with histopathology of the resected specimens. RESULTS. Infiltration into the anal canal was found histopathologically in 12 (32%) of 38 patients. Transrectal sonography revealed a true-positive diagnosis in 11 of these 12 patients. A falsepositive diagnosis of anal canal infiltration was made in two patients for a sensitivity of 91%, a specificity of 85%, and an accuracy of 92%. In the 11 patients diagnosed correctly on transrectal sonography, the depth of tumor infiltration into the anal canal corresponded with histopathology. CONCLUSION. This study shows that transrectal sonography is an accurate method for assessment of anal canal infiltration in lower rectal cancer.

I

Received December 17, 1999; accepted after revision February 16, 2000. 1 Department of Radiology, University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria. Address correspondence to A. G. Maier. 2

Department of Surgery, University of Vienna, A-1090 Vienna, Austria.

3

Department of Clinical Pathology, University of Vienna, A-1090 Vienna, Austria.

AJR 2000;175:735–739 0361–803X/00/1753–735 © American Roentgen Ray Society

AJR:175, September 2000

n the treatment of lower rectal cancer, maintenance of transanal defecation and continence is a major concern for both patients and surgeons. Total mesorectal excision has become widely accepted as a standard procedure for mid and lower rectal tumors because it results in low rates of regional tumor recurrence [1, 2]. In recent years, better knowledge of longitudinal intramural spread beyond the distal palpable tumor border has led to narrow longitudinal resection margins. About 90% of lower rectal cancers show no distal spread or limited distal spread of less than 1 cm, allowing a distal resection margin of 2.0–2.5 cm [3, 4]. Using mechanical stapling devices and surgical techniques such as intersphincteric resection with coloanal anastomosis [5], the anal sphincters may be preserved in most cancers of the middle or even the lower third of the rectum. However, for surgery of tumors with a distal edge below 8 cm from the anal verge (i.e., the distal rectal third), a special problem arises. The mesorectum that surrounds the upper and middle rectum tapers out distally, leaving the last few centimeters of rectum above the anorectal junction covered only by connective tissue. This anatomical situation provides a narrow margin of safety circumferentially. Accordingly, it is of the utmost importance to

know before the operation the depth of tumor invasion into the bowel wall and beyond, especially close to the anorectal junction. Digital examination and functional proctoscopy give only indirect evidence regarding infiltration of the levators or the anal canal [6, 7]. Tumor infiltration into the anal sphincters precludes a resection that preserves the anal sphincters, making an abdominoperineal resection with permanent colostomy necessary. Transrectal sonography is now regarded as useful for determination of the depth of transversal invasion of rectal cancer with an accuracy of 60–94% [8– 11]. Only one recent study has shown that, in addition to circumferential invasion, distal submucosal spread of rectal cancer may be determined on transrectal sonography [12]. It was the aim of this study to assess the accuracy of transrectal sonography in determining invasion of distal rectal cancer into the anal sphincters and surrounding organs by comparing results of these techniques with the histopathology of resected specimens. Subjects and Methods Thirty-eight consecutive patients (14 women, 24 men; age range, 27–90 years; mean age, 65 years) with adenocarcinoma of the mid and lower rectum (within 12 cm from the anal verge) were prospectively examined between January 1995 and October 1997 before

735

Downloaded from www.ajronline.org by 37.44.207.185 on 01/22/17 from IP address 37.44.207.185. Copyright ARRS. For personal use only; all rights reserved

Maier et al. surgery. Digital palpation, rigid rectosigmoidoscopy, and biopsy were performed before surgery to establish the diagnosis and the site of the tumor. All transrectal sonography was performed routinely with commercially available sonographic scanners (Diagnostic Ultrasound Systems 1846 and 3535; B&K Medical, Naerum, Denmark). Both scanners used a mechanical rotating endoprobe (type 1850, 5–7 MHz) with a focal zone (penetration depth) of 2–4 cm, an axial resolution of 0.4 mm, and a lateral resolution of 1.0 mm. A 360° display of the rectal wall and surrounding tissue was provided. A latex balloon was attached to the end of the probe over the transducer. After a cleansing enema, the patients were placed in the lithotomy position. The endoprobe was inserted blindly, the balloon was then filled with degassed water, and the probe was slowly withdrawn while the rectal wall was being scanned. For scanning of the anal canal, the balloon was replaced with a small hard plastic cap. The upper end of the anal canal was defined as being at the level of the puborectalis muscle. The normal sonographic appearance of the mucosa and the sphincter apparatus is shown in Figure 1. The transrectal sonographic classification of tumor invasion into bowel wall (uT) corresponds to the pathologic classification (pT) of the TNM system, in which uT1 represents mucosal or submucosal disease, uT2 is disease involving the hypoechoic muscularis propria, uT3 is disease extension into the hyperechoic perirectal fat, and uT4 is disease into adjacent organs or sidewall structures. Real-time transrectal sonograms were interpreted by one observer. Hard-copy images of the investigations were also reviewed by a second radiologist. Discrepancies were resolved by consensus. Histopathologic examinations were assessed by a single observer who was unaware of the transrectal sonographic findings.

Preoperative transrectal sonographic results were compared with the histopathologic results of the resected specimens from all patients. For histopathology, all surgical specimens were dissected as outlined by Hruban et al. [13]. After overnight fixation in neutral buffered formalin (7.5%), the entire tumor was transversely dissected into 4-mm slices. All tumor slices were blocked sequentially, dehydrated in a series of graded alcohol, and embedded in paraffin wax. Two 4-µm sections of each paraffin block were stained with H and E. The depth of local tumor spread was classified histopathologically according to the TNM classification [14] as follows: pT1 is tumor confined to the submucosa, pT2 is tumor confined to the muscularis propria, pT3 is tumor that penetrates through the rectal wall into the perirectal fat or the subserosa, and pT4 is tumor that invades adjacent organs or structures or penetrates the serosa. Infiltration of the mucosa beyond the anorectal junction and penetration into sphincter muscles were noted separately. The distal resection margin was measured on the formalin-fixed specimens. In cases of abdominoperineal excision, the distance of tumor from the perianal skin incision was recorded. Of 38 patients, 17 underwent preoperative hyperfractionated radiotherapy with 25 Gy over 5 days [15]. Surgical procedures included 19 abdominoperineal excisions, 15 low-anterior resections, three transanal disk excisions, and one intersphincteric resection. Differences between groups were analyzed using the two-tailed t test with Yates correction and the chi-square test as appropriate. A p value of less than 0.05 was considered significant.

Results

A total of 38 patients with adenocarcinoma of the lower rectum were examined with transrectal

sonography. One patient had a pT1 tumor, 13 patients were staged as pT2, 21 patients as pT3, and three patients as pT4. In 34 patients (89%), the endosonographic diagnosis corresponded to histopathology. In the four cases of discrepancy between transrectal sonography and histopathology, two were overstaged on transrectal sonography (pT1–uT2, pT2–uT3) and two patients were understaged (pT3–uT2, pT4–uT3). In 12 patients (32%), infiltration into the anal canal was histopathologically verified (Table 1). Eleven patients (92%) were correctly diagnosed on transrectal sonography. In two patients, a false-positive diagnosis of tumor infiltration of the anal sphincter was made on transrectal sonography for a sensitivity of 91%, a specificity of 85%, and an accuracy of 92%. In the 11 patients diagnosed correctly on transrectal sonography, the depth of tumor infiltration into the anal canal corresponded to histopathology. One case of tumor infiltration of the anal canal mucosa was seen (Fig. 2). Seven cases of mucosa and internal sphincter muscle infiltration were found (Fig. 3), and in three cases the mucosa and the internal and external sphincter muscles were infiltrated (Fig. 4). No isolated submucosal tumor spread was found in any of these patients. Discrepancies in interpretation between the two observers occurred in five patients. These discrepancies were always resolved in favor of the real-time observer, resulting in two true-positive and three true-negative diagnoses. In all 17 patients who received preoperative radiotherapy of 25 Gy, the staging and

A

B

Fig. 1.—Normal appearance of anal canal. A, Coronal diagram of anal canal and sphincter apparatus. Longitudinal line (1) shows level of transrectal sonogram. B, Transrectal sonogram depicts mucosa (white arrow ), internal sphincter (black arrow ), and sling of puborectalis and deep external sphincter (arrowheads).

736

AJR:175, September 2000

Sonography of Lower Rectal Carcinoma

TABLE 1

Assessment of Sphincter Infiltration on Transrectal Sonography: Correlation with pT Stage Present

Absent

Downloaded from www.ajronline.org by 37.44.207.185 on 01/22/17 from IP address 37.44.207.185. Copyright ARRS. For personal use only; all rights reserved

Cancer Stage Sonography

Histopathology

Sonography

Histopathology

0 3 8a 2

0 3 7 2

1 10 13b 1

1 10 14 1

pT1 (n = 1) pT2 (n = 13) pT3 (n = 21) pT4 (n = 3)

Note.—Cancer stage is based on pathohistologic staging of the resected specimens. a

Includes two false-positive cases.

b

Includes one false-negative case.

assessment of anal canal infiltration made on transrectal sonography correlated with the histopathologic results. All patients with infiltration and seven additional patients without infiltration into the anal canal underwent an abdominoperineal excision. The parameters of tumor grading and distal resection margin were investigated separately. The distance between the lower margin of the tumor and the anal verge was 2.4 cm in tumors that invaded the anal canal and 4.14 cm in tumors without infiltration into the anal canal ( p = 0.0007). No statisti-

A

B

Fig. 2.—40-year-old man with lower rectal cancer (pT3, N1, G3) and superficial spreading into mucosa of anal canal. A, Transrectal sonogram shows hypoechoic lesion from 3-o’clock to 6-o’clock positions (arrowheads ) limited to mucosa, representing tumor extension. Internal sphincter (black arrow ) and external sphincter (white arrow ) are not affected. B, Corresponding photomicrograph of histopathologic slice shows tumor to be restricted to mucosa of anal canal (arrowhead ). Internal sphincter (short arrow ) and external sphincter (long arrow ) are intact. Specimen was opened anteriorly by surgeon. (H and E, ×1.5)

Fig. 3.—74-year-old man with lower rectal cancer (pT3, N0, G2) with extension into anal canal. A, Transrectal sonogram shows infiltration of mucosa and internal sphincter (arrowhead ) from 11o’clock to 2-o’clock positions. (black arrows ). External sphincter (white arrow ) is intact. B, Photomicrograph of histopathologic specimen shows tubular tumor structures infiltrating smooth muscle structure (arrows ) of internal sphincter muscle. (H and E, ×200)

A AJR:175, September 2000

B 737

Maier et al.

Downloaded from www.ajronline.org by 37.44.207.185 on 01/22/17 from IP address 37.44.207.185. Copyright ARRS. For personal use only; all rights reserved

Fig. 4.—73-year-old woman with lower rectal cancer (pT3, N1, G3). Transrectal sonogram shows massive infiltration of internal sphincter (arrowhead), external sphincter (arrow), and all layers of anal canal between 11-o’clock and 2-o’clock positions (white arrows).

cally significant differences occurred between the grading of the tumors with or without anal canal infiltration (p = 0.78). Discussion

The maintenance of intestinal continuity is a major goal in rectal cancer surgery to avoid permanent colostomy. Tumor invasion into the anal sphincter muscles or pelvic floor muscles is a contraindication for sphincter preservation. In recent years, it has been shown that by reducing distal longitudinal safety margins and with the introduction of the total mesorectal excision concept, more sphincter-saving procedures may be accomplished without compromising oncologic results [2, 4]. The choice of the optimal therapeutic approach is based on several factors including tumor location, depth of penetration, lymph node status, and histopathologic grading [1]. A number of studies have established the validity of transrectal sonography as a means of tumor staging; and for the assessment of invasion depth, an accuracy of 60–94% has been reported [8–11]. In addition, longitudinal tumor spread is of importance for tumors located in the very distal rectum. Although longitudinal spread has been reported in about 26% of cases on the basis of histopathologic examination of surgical specimens [3, 16–18], data concerning the preoperative assessment of longitudinal tumor spread on transrectal sonography are scarce. In a recent study addressing this particular issue, it was shown that distal longitudinal tumor spread to a maximum of 2 cm was detectable

738

with an accuracy of 86% compared with histopathology [12]. However, infiltration into the anal sphincters was not mentioned. Our study shows that transrectal sonography can be regarded as an appropriate method to determine both longitudinal tumor spread and infiltration of the anal sphincters with a sensitivity, specificity, and accuracy of 91%, 85%, and 92%, respectively. Of the 38 patients investigated, the results of transrectal sonography were interpreted incorrectly in only three patients (8%). Disease of two of these patients was overstaged and of one patient was understaged. Accordingly, inadequate treatment might have resulted for only one patient (3%). The reason for the two cases of overstaging was a massive peritumoral inflammatory reaction that was confirmed on histopathology. Although it has been shown that a peritumoral inflammatory reaction may be differentiated from the tumor by a characteristic endosonographic appearance in most cases [19], differentiation was not possible in those two cases. Both were pT3 lesions located close to the anorectal junction, where precise focusing of sonography is difficult. In addition, because of angulation between the rectum and the anal canal, other factors, such as reverberation and refraction, may have been responsible for these misinterpretations [20]. Infiltration of the sphincter muscles in rectal cancer is not classified in the current TNM system. Discussion in the TNM committee over this point is underway (Sobin LH, personal communication). Some pathologists favor staging of anal canal invasion as pT3 because involvement

of the external sphincter by lower rectal carcinoma does not influence the prognosis, similar to pT3 tumors if an adequate abdominoperineal resection is done (Wittekind C, Sobin LH, personal communication). On the other hand, pathologists have reported that infiltration of the anal canal should be considered stage pT4 because the sphincters should be regarded as another anatomic structure (Wittekind C, Sobin LH, personal communication). In recent publications, MR imaging with the help of endoanal coils has been shown to be effective in the visualization of the structures of the anal canal and its diseases [21]. Endoanal coils might also be helpful in detecting invasion of the sphincter muscles in lower rectal cancer. However, visualization of the extent of a rectal tumor may be impossible because of the limited field of view of an endoanal coil. In conclusion, this study shows that both local tumor staging and anal sphincter infiltration can be assessed precisely on transrectal sonography, which also helps to determine the appropriate surgical strategy in lower rectal cancer.

References 1. Williams NS. Surgical treatment in rectal cancer. In: Keighley MRB, Williams NS, eds. Surgery of the anus, rectum, and colon. London: Saunders, 1993:939–1053 2. MacFarlane JK, Ryall RDH, Heald RJ. Mesorectal excision for rectal cancer. Lancet 1993;341: 457–460 3. Cross SS, Bull AD, Smith JHF. Is there any justification for routine examination of bowel resection margins in colorectal adenocarcinomas? J Clin Pathol 1989;42:1040–1042 4. Karanjia ND, Schache DJ, North WRS, Heald RJ. Close shave in anterior resection. Br J Surg 1990;77:510–512 5. Schiessel R, Karner-Hanusch J, Herbst F, Teleky B, Wunderlich M. Intersphincteric resection for low rectal tumours. Br J Surg 1994;81:1376–1378 6. Williams NS, Durdey P, Quirke P, et al. Pre-operative staging of rectal neoplasm and its impact on clinical management. Br J Surg 1985;72:868–874 7. Nicholls RJ, Maxon AY, Morson BC, Dixon AK, Fry IK. The clinical staging of rectal cancer. Br J Surg 1982;69:404–409 8. Beynon J, Mortensen NJ, Rigby HS. Rectal endosonography, a new technique for the preoperative staging of rectal carcinoma. Eur J Surg Oncol 1988;14:297–309 9. Feifel G, Hildebrandt U. New diagnostic imaging in rectal cancer: endosonography and immunoscintigraphy. World J Surg 1992;16:841–847 10. Konishi F, Ugajin H, Ito K, Kanazawa K. Endorectal ultrasonography with a 7.5 MHz linear array scanner for the assessment of invasion of rectal carcinoma. Int J Colorectal Dis 1990;5:15–20

AJR:175, September 2000

Downloaded from www.ajronline.org by 37.44.207.185 on 01/22/17 from IP address 37.44.207.185. Copyright ARRS. For personal use only; all rights reserved

Sonography of Lower Rectal Carcinoma 11. Rifkin MD, Ehrlich SM, Marks G. Staging of rectal carcinoma: prospective comparison of endorectal US and CT. Radiology 1989;120:319–320 12. Yanagi H, Kusunoki M, Shoji Y, Yamamura T, Utsunomiya J. Preoperative detection of distal intramural spread of lower rectal carcinoma using transrectal ultrasonography. Dis Colon Rectum 1996;39:1210–1214 13. Hruban RH, Westra WH, Phelps TH, Isacson C. Surgical pathology dissection: an illustrated guide. New York: Springer-Verlag, 1996:9 14. Sobin LH, Wittekind CH, eds. TNM classification of malignant tumors, 5th ed. New York: WileyLiss, 1997:66–73

AJR:175, September 2000

15. Pahlman L, Dahlberg M, Glimelius B. Clinicopathological staging and management of colorectal cancer: perioperative radiation therapy. World J Surg 1997;21:733–740 16. Williams NS, Dixon MF, Johnston D. Reappraisal of the 5 centimeter rule of distal excision for carcinoma of the rectum: a study of distal intramural spread and of patients’ survival. Br J Surg 1993;70: 150–154 17. Lazorthes F, Voigt JJ, Roques J, Chiotasso P, Chevreau P. Distal intramural spread of carcinoma of the rectum correlated with lymph nodal involvement. Surg Gynecol Obstet 1990;170:45–48 18. Grinnel RS. Distal intramural spread of carci-

noma of the rectum and rectosigmoid. Surg Gynecol Obstet 1954;99:421–430 19. Maier AG, Barton PP, Neuhold NR, Herbst F, Teleky BK, Lechner GL. Peritumoral tissue reaction at transrectal US as a possible cause of overstaging: histopathologic correlation. Radiology 1997;203:785–789 20. Kruskal JB, Kane RA, Sentovich SM, Longmaid HE. Pitfalls and sources of error in staging rectal cancer with endorectal US. RadioGraphics 1997; 17:609–626 21. Hussain SM, Stoker J, Lameris JS. Anal sphincter complex: endoanal MR imaging of normal anatomy. Radiology 1995;197:671–677

739