Original Article
Diagnosis and Staging of Colonic Tumors by Conventional Abdominal Sonography as Compared with Hydrocolonic Sonography
N Engl J Med 1992; 327:65-69July 9, 1992
- Abstract
- Abstract
Background.
Ultrasonic examination has become increasingly important in the diagnosis of diseases of the gastrointestinal tract. In neoplastic diseases of the large intestine, however, the diagnostic value of conventional transabdominal sonography is limited. The purpose of this investigation was to determine whether the evaluation and staging of colonic tumors would be improved by the retrograde instillation of water into the colon in a procedure called hydrocolonic sonography.
Methods.
Three hundred patients were examined in a prospective study in which both conventional transabdominal sonography and transabdominal hydrocolonic sonography were performed before the diagnosis was verified by colonoscopy. The indication for the examinations was abdominal pain, diarrhea, weight loss, or the presence of occult blood in a stool specimen.
Results.
With the instillation of water into the colon, it was possible to display the colon sonographically from the rectosigmoid transition to the cecum in 97 percent of the patients examined. In addition to permitting the evaluation of the colonic lumen, the procedure allowed the five layers of the colonic wall and the connective tissue surrounding the colon to be examined in detail. Only 9 of 29 carcinomas were diagnosed by conventional abdominal sonography (31 percent), whereas hydrocolonic sonography permitted the diagnosis of 28 (97 percent). No polyps could be detected by conventional abdominal sonography, whereas hydrocolonic sonography permitted the diagnosis of 38 of 42 polyps ≥7 mm (91 percent) and 3 of 12 polyps <7 mm (25 percent). In addition, detailed evaluation of the structure of the bowel wall with hydrocolonic sonography permitted a more precise staging of colonic tumors.
Conclusions.
Hydrocolonic sonography, a new diagnostic procedure, can facilitate the diagnosis and staging of colonic tumors. (N Engl J Med 1992;327:65–9.)
Media in This Article
Figure 1
Normal Colon as Displayed by Hydrocolonic Sonography.Figure 2
Colonic Carcinoma as Displayed by Hydrocolonic Sonography.Article Activity
- Article
ULTRASONIC examination has become a well-established diagnostic tool in many abdominal diseases. In recent years, the degree of resolution attainable with ultrasonic equipment has been considerably improved, making more accurate ultrasonic diagnosis possible. Thus, conventional transabdominal sonography has become increasingly important in the diagnosis of diseases of the gastrointestinal tract. Today it is even possible to diagnose acute appendicitis by ultrasonic examination.1 , 2 Nonetheless, conventional abdominal sonography does have limitations. With the use of sonography alone, it is impossible to detect diseases of the colon reliably, since the large bowel cannot be visualized in its entirety and detailed evaluation of wall structures and intraluminal lesions is difficult. Transabdominal sonography after the retrograde instillation of water into the colon, a method known as hydrocolonic sonography, can considerably improve conditions for ultrasonic imaging and evaluation of the colon. Studies have shown that this method permits not only the diagnosis of colonic tumors3 but also that of inflammatory colonic diseases, such as Crohn's disease and ulcerative colitis.4 , 5
The present study compared the value of hydrocolonic sonography with that of conventional transabdominal sonography in the diagnosis and staging of colonic tumors.
Methods
Three hundred consecutive patients (mean [±SD] age, 56±12 years) referred for colonoscopy to evaluate abdominal pain, diarrhea, weight loss, or a positive test for fecal occult blood were the subjects of the study. They were examined both by conventional abdominal sonography and by hydrocolonic sonography with retrograde instillation of water into the colon before verification of the diagnosis by colonoscopy.
Preparation of the bowel consisted of a laxative orthograde intestinal lavage on the morning of the examination. A total of up to 1500 ml of water was instilled into the colon after an intravenous injection of 20 mg of scopolamine-N-butyl bromide (Buscopan; Boehringer Ingelheim) for relaxation of the bowel. The relaxant was not only necessary to achieve optimal distention of the bowel during the instillation of water, but was also useful in suppressing the sense of urgency for elimination. For the instillation of water, a commercially available system usually employed for barium enemas was used (Pneumocolon; Barnes—Hind Barium Products). Continuous transabdominal sonographic examination of the large intestine was carried out beginning at the time of water instillation, with use of a real-time scanning device with 3.5-, 5.0-, and 7.5-MHz transducers (Picker CS 9500). The examination was started with the 3.5-MHz transducer; for the detailed evaluation of the structure of the colon wall, the 5.0- and 7.5-MHz transducers were used. A tilting table operated manually allowed the patient's position to be adjusted. To achieve optimal filling and distention of the sigmoid colon, the examination was initially performed with the patient in an upright position; 300 to 500 ml of water was instilled. After the examination of the sigmoid, the patient was placed in the supine position for the examination of the remainder of the colon, during which a further 1000 ml of water was instilled. For the examination of the colonic flexures, a lateral intercostal route was chosen. Deep inhalation moves the colonic flexures down and facilitates their sonographic visualization. The diagnosis of colonic tumors was based on sonographic evidence of intraluminal masses fixed to the wall, the appearance of lesions within the intestinal-wall structure and the surrounding connective tissue, or both. The depth of tumor infiltration into the colon wall was measured, and the tumor stage was determined according to the system of classification devised by the International Union against Cancer.6 A tumor was classified as T1 when infiltration was confined to layers 1 and 2 of the colon, as T2 when the tumor had infiltrated into layers 1 to 3, as T3 when it had infiltrated into layers 1 to 5 and the surrounding connective tissue, and as T4 when the tumor had spread to all layers and infiltrated other organs.
Conventional transabdominal sonography was performed with 3.5- and 5.0-MHz transducers. A colonic tumor was considered to be present when there was sonographic evidence of a pathologic bowel loop, which could be identified by the presence of a segment of thick, echo-poor colon wall wider than 4 mm, the shape of which remained constant even when the abdomen was compressed by the transducer.
In all cases colonoscopy was performed after hydrocolonic sonography had been completed and after the water had been expelled. Neither the ultrasound operator nor the colonoscopist knew the suspected diagnosis beforehand, and each operator was unaware of the examination performed by the other. Whenever a polyp was diagnosed, a polypectomy and a histologic examination were performed. Whenever a colonic carcinoma was diagnosed, surgical resection and histologic examination of the resected tissue were carried out. The histologically confirmed tumor stage was then compared with the sonographically assessed stage. The statistical analysis was performed with Fisher's exact test. A two-tailed P value of less than 0.05 was considered to indicate statistical significance. Informed consent was obtained from all patients, and the study was approved by the local institutional ethics board.
Results
In 97 percent of the patients examined, the entire length of the colon, starting at the rectosigmoid border and ending at the cecum, could be visualized with hydrocolonic sonography. The rectum could not be examined in detail because of the overlying accumulation of air and the limited penetration of the transducers. The total time required for the examination was approximately 15 minutes. The examination was well tolerated by the patients, there were no side effects, and even the older patients were able to retain the water that was instilled. The images of the colon displayed by hydrocolonic sonography showed an echo-free intestinal lumen 4 to 5 cm in width, surrounded by ribbon-shaped structures — the haustra — that projected as echogenic lamellae into the lumen. With a high-frequency transducer it was possible not only to evaluate the haustra and the intestinal lumen, but also to view in detail the structure of the intestinal wall, within which five layers differing in echogenicity were distinguishable (Fig. 1Figure 1
Normal Colon as Displayed by Hydrocolonic Sonography.). The total width of the intestinal wall was 3 to 4 mm.A colonic carcinoma was diagnosed by colonoscopy in 29 of the patients examined. Twelve polyps <7 mm in diameter and 42 polyps ≥7 mm were found (of the 42, 8 were 7 to 10 mm in diameter, 13 were 11 to 15 mm, 9 were 16 to 20 mm, 8 were 21 to 25 mm, and 4 were 26 to 30 mm) (Tables 1Table 1
Diagnosis of Colon Cancer by Colonoscopy, Hydrocolonic Sonography, and Conventional Abdominal Sonography.* and 2Table 2
Diagnosis of Colonic Polyps by Colonoscopy and Hydrocolonic Sonography.*). The carcinomas ranged in size from 20 to 69 mm (4 were 20 to 29 mm in diameter, 12 were 30 to 39 mm, 6 were 40 to 49 mm, 3 were 50 to 59 mm, and 4 were 60 to 69 mm).Sonographically, colonic polyps and carcinomas appeared as echogenic structures fixed to the intestinal wall and projecting into the intestinal lumen. With hydrocolonic sonography, it was possible to diagnose and localize 28 of the 29 carcinomas correctly (sensitivity, 97 percent; specificity, 100 percent). The one carcinoma not detected by this method was located in the transverse colon (Table 1). During hydrocolonic sonography, tumor staging was performed on the basis of the sonographically assessed depth of tumor infiltration into the bowel wall and the surrounding connective tissue (Fig. 2Figure 2
Colonic Carcinoma as Displayed by Hydrocolonic Sonography.). The carcinomas were then classified as T1, T2, T3, or T4 according to their sonographic appearance. No T1 carcinomas were found in the patients examined. Seventy-five percent of the T2 tumors, 85 percent of the T3 tumors, and 75 percent of the T4 tumors were correctly classified by hydrocolonic sonography (Table 3Table 3
Staging of Colonic Tumors on the Basis of Hydrocolonic Sonography.). One of the T2 tumors was incorrectly judged to be T3, three of the T3 tumors were misclassified as T2, and one of the T4 tumors was misclassified as T3.Sonographically, polyps appeared as echogenic structures fixed to the wall. In the case of pedunculated polyps, the peduncle could be identified at the base of the polyp. During external compression or increased instillation of fluid into the colon, these polyps floated in the intestinal lumen. They were classified as benign if the structure of the wall, in particular the submucosa and the muscularis propria near the base of the tumor, appeared intact (Fig. 3Figure 3
Small Colonic Polyp (P) as Displayed by Hydrocolonic Sonography.). All 38 polyps ≥7 mm and the 3 polyps <7 mm that were diagnosed by hydrocolonic sonography were correctly classified with these criteria (Table 2). As compared with colonoscopy, hydrocolonic sonography thus had a sensitivity of 91 percent and a specificity of 100 percent for the diagnosis of colonic polyps ≥7 mm, but the sensitivity of this technique for the detection of polyps <7 mm was only 25 percent (Table 2). The polyps that were not demonstrated sonographically ranged in size from 3 to 10 mm. All the polyps <7 mm were sessile; two of the undetected polyps ≥7 mm were pedunculated, and two were sessile.With conventional abdominal sonography, pathologically thickened bowel loops were correctly identified in nine cases as indications of the presence of a tumor. Four of these nine were T4 carcinomas, and five were classified as being in stage T3. In three cases, false positive results were obtained with conventional abdominal sonography (sensitivity, 31 percent; specificity, 99 percent) (Table 1). None of the polyps were correctly identified by conventional abdominal sonography.
Discussion
Carcinoma of the colon is the second most common cancer in the Western world. In recent years, numerous investigations have attempted to find a method of diagnosing carcinomas in an early stage.7 8 9 10 11 Unfortunately, the traditional screening examinations (e.g., stool examination for occult blood and sigmoidoscopy)12 , 13 are imperfect. Experience has shown that up to 40 percent of carcinomas and up to 72 percent of polyps cannot be detected by examining the stool for occult blood9 and that up to 40 percent of colon carcinomas are overlooked when sigmoidoscopy alone is used.14 Furthermore, a negative test by these methods may give a false sense of security. Because of the increased frequency of tumors located proximal to the sigmoid colon,15 , 16 a method of examination that covers the proximal as well as the distal colon is clearly needed. Undoubtedly, colonoscopy is the most sensitive method for the accurate diagnosis of colonic tumors, but it is unsuitable for widespread use.17 Hydrocolonic sonography, on the other hand, is performed easily, is better tolerated by patients, and may eventually prove to be more practical than colonoscopy in screening for colonic tumors.
Carcinoma in a bowel segment is demonstrated sonographically by a concentric or eccentric thickening of the bowel wall that produces a target-like appearance on transverse sections and a kidney-shaped appearance on longitudinal sections.18 , 19 This wall infiltration can be visualized by conventional abdominal sonography only when the involved segments exceed a certain size. That is why polyps that have not infiltrated the colonic wall cannot be identified by conventional abdominal sonography and why so few colonic carcinomas are correctly diagnosed with this procedure. Indeed, most of the carcinomas diagnosed by conventional abdominal sonography are in advanced stages.20 Even though the resolution of images obtainable with sonographic equipment has been greatly improved, the conventional method continues to play a subordinate part in the diagnosis of colonic tumors. Although patients should always be checked routinely for pathologic bowel loops during abdominal sonography, one should remember that the sensitivity of this method for the diagnosis of colonic tumors is poor and that conclusive diagnosis is impossible.
With the retrograde instillation of water into the colon, sonographic imaging of the colon is considerably improved, and the accuracy of diagnoses can be greatly enhanced without the need for more invasive diagnostic techniques. Because of the notable differences in impedance between the fluid-filled lumen of the large bowel and the colonic wall, colonic tumors and lesions in the wall are easy to recognize during hydrocolonic sonography when high-resolution transducers are used. This technique allows the sonographic differentiation of five layers within the colon wall. The images of wall architecture displayed during hydrocolonic sonography correspond to those obtained by endorectal sonography and endosonography of the upper gastrointestinal tract.21 , 22 The wall layers display different patterns of echogenicity that correspond to the known anatomical layers.23 The first two layers adjacent to the lumen correspond to the mucosa; the third, hyperechoic layer to the submucosa; the fourth, hypoechoic layer to the muscularis propria; and the fifth, echogenic layer to the serosa and the subserous fatty tissue.
The identification of these layers during hydrocolonic sonography makes possible not only the diagnosis but also the staging of colonic tumors. Carcinomas and polyps appear on sonograms as hyperechoic intraluminal structures fixed to the colon wall. Polyps with a small peduncle float in the intestinal lumen during external compression with the transducer or during the retrograde instillation of water. At the base of the polyps, the structure of the wall layers remains intact. In the case of carcinomas, however, destruction of the wall structure can be clearly seen, and the depth of infiltration into the surrounding connective tissue can also be demonstrated by hydrocolonic sonography. The system used by the International Union against Cancer to classify tumors is based on the assessment of the depth of tumor infiltration.6 In the present study there were no T1 carcinomas. It is therefore impossible to conclude whether or not hydrocolonic sonography is capable of differentiating between a benign polyp and a carcinoma in this stage, in which infiltration is limited to the submucosa. Nor can polyps smaller than 7 mm always be visualized by hydrocolonic sonography. Small colonic polyps are difficult to demonstrate, because differentiating them definitively from the nearby haustra is not always possible. The missing of small colonic polyps does not appear to be of much clinical importance, since several studies suggest that patients with such polyps have no increased risk of colon cancer.24 25 26
Colonic tumors can be confused with stool particles, which appear during hydrocolonic sonography as echogenic structures within the bowel lumen. With the use of real-time equipment, it is possible to observe the movement of particulate material or gas bubbles within the lumen. Compression of the abdomen with the transducer may cause stool particles to move, thereby allowing rapid differentiation between artifacts and tumors of the large bowel. Although examination of the transverse colon and detailed evaluation of the structure of the bowel wall are especially difficult in obese patients because of the penetration limits of the higher-frequency transducer, a 3.5-MHz transducer allows clear visualization of colonic tumors in such patients.
The results of the present study demonstrate that marked improvement in the ultrasonic diagnosis of colonic tumors can be achieved with the use of hydrocolonic sonography, as compared with conventional abdominal sonography. Moreover, with hydrocolonic sonography it is possible to examine structures of the bowel wall with a clarity comparable to that achieved with endosonography and to determine the stage of tumors with a high degree of accuracy. This new sonographic method reveals alterations of bowel-wall layers not only in neoplastic diseases but also in other common diseases of the colon, such as Crohn's disease and ulcerative colitis.4 , 5 Hydrocolonic sonography is well tolerated by patients and is thus particularly well suited for the diagnosis of colonic tumors in outpatients. It may also find a place among the recommended strategies for the early diagnosis of these tumors.
Supported by a grant from Deutsche Krebshilfe.
I am indebted to Mrs. Helga Lange, Mrs. Ursula Osann, and Mrs. Ute Wilutzki for technical assistance.
Source Information
From the 2nd Department of Internal Medicine, Klinikum Darmstadt, Academic Hospital of the University of Frankfurt, Grafenstrasse 9, 6100 Darmstadt, Germany, where reprint requests should be addressed to P.D. Dr. Limberg.
- References
References
1
Puylaert
Full Text | Web of Science | Medline2
Schwerk
Web of Science | Medline3
Limberg
CrossRef | Web of Science | Medline4
Limberg
CrossRef | Web of Science | Medline5
Limberg
CrossRef | Web of Science | Medline6
Hermanek P, Sobin LH, eds. TNM classification of malignant tumours. 4th ed. Berlin, Germany: Springer-Verlag, 1987.
7
Ransohoff
Full Text | Web of Science | Medline8
Rex
Web of Science | Medline9
Simon
Web of Science | Medline10
Allison
Web of Science | Medline11
Eddy
Web of Science | Medline12
Working guidelines for early cancer detection: rationale and supporting evidence to decrease mortality. Bethesda, Md.: National Cancer Institute, 1987.
13
Guidelines for the cancer-related checkup: recommendations and rationale . CA Cancer J Clin 1980;30:194–240.
CrossRef | Web of Science14
Neugut
Web of Science | Medline15
Mamazza
CrossRef | Web of Science | Medline16
Rosato
CrossRef | Web of Science | Medline17
Neugut
Web of Science | Medline18
Lutz
CrossRef | Medline19
Bluth
Web of Science | Medline20
Sianesi
CrossRef | Web of Science | Medline21
Hildebrandt
CrossRef | Web of Science | Medline22
Tio
Web of Science | Medline23
Kimmey
Web of Science | Medline24
Spencer
Web of Science | Medline25
Eide
CrossRef | Web of Science | Medline26
Atkin
- Citing Articles (25)
Citing Articles
1
Keiichi Onji, Shigeto Yoshida, Shinji Tanaka, Yoshito Takemura, Shiro Oka, Masaharu Yoshihara, Hiroyasu Yamada, Masazumi Okajima, Kazuaki Chayama. (2011) Microvascular structure and perfusion imaging of colon cancer by means of contrast-enhanced ultrasonography. Abdominal Imaging
CrossRef2
H. Zhuang, Z-G. Yang, H.-J. Chen, Y.-L. Peng, L. Li. (2011) Time-intensity curve parameters in colorectal tumours measured using double contrast-enhanced ultrasound: correlations with tumour angiogenesis. Colorectal Diseaseno-no
CrossRef3
Satoru Seo, Yuhei Hamaguchi, Yukihiro Okuda, Yutaka Babazono, Yasuhiro Fujimoto, Akira Mitsuyoshi. (2011) Transabdominal Ultrasound Efficacy in Determining Colorectal Cancer Invasion Depth. The Japanese Journal of Gastroenterological Surgery 44:1, 100-107
CrossRef4
Rakesh Sinha. 2011. Appendix, colon and rectum. , 388-410.
CrossRef5
Franklin Greif, David Aranovich, Veacheslav Zilbermints, Nisim Hannanel, Alexander Belenky. (2010) Intraoperative hydrocolonic ultrasonography for localization of small colorectal tumors in laparoscopic surgery. Surgical Endoscopy 24:12, 3144-3148
CrossRef6
Christoph F. Dietrich, Barbara Braden. (2009) Sonographic assessments of gastrointestinal and biliary functions. Best Practice & Research Clinical Gastroenterology 23:3, 353-367
CrossRef7
Koichi Yabunaka, Shigeru Sanada, Hiroya Fukui, Shinji Tamate, Masayuki Fujioka. (2006) Transabdominal sonographic appearance of adult colonic polyps. Journal of Medical Ultrasonics 33:4, 231-237
CrossRef8
Marilee L. Freitas, Spiros G. Frangos, Heidi L. Frankel. (2006) The Status of Ultrasonography Training and Use in General Surgery Residency Programs. Journal of the American College of Surgeons 202:3, 453-458
CrossRef9
Jin-ichi Hida, Kiyotaka Okuno, Masayuki Yasutomi, Takehito Yoshifuji, Toshihiro Uchida, Tadao Tokoro, Hitoshi Shiozaki. (2005) Optimal Ligation Level of the Primary Feeding Artery and Bowel Resection Margin in Colon Cancer Surgery: The Influence of the Site of the Primary Feeding Artery. Diseases of the Colon & Rectum 48:12, 2232-2237
CrossRef10
Franklin Greif, Alexander Belenky, David Aranovich, Igal Yampolski, Nisim Hannanel. (2005) Intraoperative ultrasonography: a tool for localizing small colonic polyps. International Journal of Colorectal Disease 20:6, 502-506
CrossRef11
2005. Anticoagulants. .
CrossRef12
2005. Ultrasonics. .
CrossRef13
Nadia Pallotta, Flavia Baccini, Enrico Corazziari. (2001) Small intestine contrast ultrasonography (sicus) in the diagnosis of small intestine lesions. Ultrasound in Medicine & Biology 27:3, 335-341
CrossRef14
Nadia Pallotta, Flavia Baccini, Enrico Corazziari. (1999) Contrast ultrasonography of the normal small bowel. Ultrasound in Medicine & Biology 25:9, 1335-1340
CrossRef15
Christian Kuntz, Christian Herfarth. (1999) Imaging diagnosis for staging of gastric cancer. Seminars in Surgical Oncology 17:2, 96-102
CrossRef16
HENRIK J. DITZEL. (1999) Human monoclonal antibodies: A tool for cancer detection in vivo. APMIS 107:S94, 5-42
CrossRef17
Jin-ichi Hida, Masayuki Yasutomi, Tadao Tokoro, Ryuichi Kubo. (1999) Examination of nodal metastases by a clearing method supports pelvic plexus preservation in rectal cancer surgery. Diseases of the Colon & Rectum 42:4, 510-514
CrossRef18
Nadia Pallotta, Flavia Baccini, Enrico Corazziari. (1999) Ultrasonography of the small bowel after oral administration of anechoic contrast solution. The Lancet 353:9157, 985-986
CrossRef19
Francesc Tous, Marcos Busto. (1997) Assessment of abdominal sonography in the diagnosis of tumors of the gastroduodenal tract. Journal of Clinical Ultrasound 25:5, 243-247
CrossRef20
(1995) Hydrocolonic Ultrasonography. New England Journal of Medicine 332:23, 1581-1582
Full Text21
Toribara, Neil W., Sleisenger, Marvin H., . (1995) Screening for Colorectal Cancer. New England Journal of Medicine 332:13, 861-867
Full Text22
J. R. Siewert, A. Sendler, H. J. Dittler, U. Fink, H. H
fler. (1995) Staging gastrointestinal cancer as a precondition for multimodal treatment. World Journal of Surgery 19:2, 168-177
CrossRef23
Chui, David W.Gooding, GretchenMcQuaid, Kenneth R.Griswold, VirginiaGrendell, James H.. (1994) Hydrocolonic Ultrasonography in the Detection of Colonic Polyps and Tumors. New England Journal of Medicine 331:25, 1685-1688
Full Text24
B. David Collier. (1993) New radiographic techniques for colorectal cancer. Cancer 71:S12, 4214-4216
CrossRef25
(1992) Diagnosis of Colonic Tumors by Hydrocolonic Sonography. New England Journal of Medicine 327:20, 1459-1460
Full Text
- Letters
