Correspondence
Comparison of Colonoscopy and Double-Contrast Barium Enema
N Engl J Med 2000; 343:1728-1730December 7, 2000
- Article
To the Editor:
The report by Winawer et al. (June 15 issue)1 on the sensitivity of double-contrast barium enema for detecting colonic adenomas requires comment. The overall sensitivity of double-contrast barium enema was 39 percent, with a sensitivity of 48 percent for detecting adenomas larger than 1 cm. Only 23 adenomas larger than 1 cm were detected among the seven participating institutions, and most were between 11 and 15 mm (Stewart E: personal communication). The measurement method used in the study has been shown to overestimate significantly the true size. In one review, half the lesions endoscopically classified as more than 1 cm were proved to be smaller.2 In the study by Winawer et al., the fact that the sensitivity for adenomas that were 6 to 10 mm was similar to the sensitivity for those classified as larger supports the belief that most of the “larger polyps” belonged in the group of smaller polyps. A table listing the size, histologic features, and location of adenomas larger than 1 cm and also noting those detected by double-contrast barium enema was not provided. With the small sample and likelihood of measurement error, conclusions about the sensitivity of double-contrast barium enema for detecting polyps larger than 1 cm are premature.
A more important issue is the necessity of detecting all polyps. Although the incidence of carcinoma was reduced in the study by Winawer et al., almost all the benefit was derived from the initial polypectomy.3 Almost 40 percent of the adenomas detected initially were larger than 1 cm. During surveillance, only 23 of 791 (2.9 percent) of the polyps removed exceeded 1 cm. Before one advocates routine and frequent colonoscopy, the morbidity and expense associated with the procedure must be weighed against the prevalence of clinically significant lesions. Waye, one of the coauthors of the present study, and colleagues reported that 2 percent of patients who underwent polypectomy required hospitalization for complications.4 In a review of nine post-polypectomy surveillance studies, the number of deaths from colonoscopic perforation exceeded the number of deaths from colon cancer.5 The prevalence of small polyps is high, approaching 50 to 70 percent by the age of 70 years, and almost half the lesions are not even adenomas. The efficiency and safety of an aggressive strategy to promote universal eradication of polyps in any group should be a major concern.
The choice of barium enema or colonoscopy in an individual patient should take into account the prevalence and clinical significance of different types of polyps, the patient's absolute risk of colon cancer, and the costs and complication rates associated with each approach. The ultimate goal is an acceptable balance between the reduction of disease-specific mortality, the efficient allocation of limited resources, and the potential dangers to the overwhelming majority of patients in whom the disease will never develop.
5 ReferencesSeth N. Glick, M.D.
MCP Hahnemann University, Philadelphia, PA 191021
Winawer SJ ,Stewart ET ,Zauber AG , et al. A comparison of colonoscopy and double-contrast barium enema for surveillance after polypectomy. N Engl J Med 2000;342:1766-1772
Full Text | Web of Science | Medline2
Schoen RE ,Gerber LD ,Margulies C . The pathologic measurement of polyp size is preferable to the endoscopic estimate. Gastrointest Endosc 1997;46:492-496
CrossRef | Web of Science | Medline3
Zauber AG ,Winawer SJ ,Loeve F ,Boer R ,Habbema D . Effect of initial polypectomy versus surveillance polypectomy on colorectal cancer incidence reduction: microsimulation modeling of National Polyp Study data. Gastroenterology 2000;118:Suppl 2:A187-A187 abstract.
CrossRef | Web of Science4
Waye JD ,Lewis BS ,Yessayan S . Colonoscopy: a prospective report of complications. J Clin Gastroenterol 1992;15:347-351
CrossRef | Web of Science | Medline5
Rex DK . Endoscopists, polyp size, and post-polypectomy surveillance: making a mountain out of a molehill? Gastrointest Endosc 1997;46:571-574
Web of Science | Medline
To the Editor:
There is no mention of the performance of barium enema with respect to the quality of the colonic cleansing achieved and the radiologist's level of confidence. If the results of barium enema were poor despite excellent preparation of the patients and a high level of confidence on the part of the radiologists, they would have considerably more clinical significance than the results of all the examinations lumped together regardless of the level of confidence.
Todd Fibus, M.D.
Emory University, Atlanta, GA 30322To the Editor:
Most physicians would agree that in expert hands with optimal effort, colonoscopy is better than double-contrast barium enema for detecting insignificant polyps. The real questions are whether colonoscopy is significantly better for detecting lesions that would be incurable by polypectomy at the time of subsequent examination, and whether its use could still be justified after a full consideration of the substantial differences in safety and cost between the two approaches.
There is also the problem of generalizing from the results of optimally performed colonoscopies in a research setting to those performed in general practice — a pivotal issue with respect to any policy on the use of an examination that is so dependent on time and effort. Colonoscopies performed without time constraints by motivated experts in prospective studies involving a carefully selected population of patients1 can be a far cry from the typical situation. The results of these studies, which usually fail to report the duration of the examination, can be quite misleading. In a prospective study, Hixson et al. reported that one of two examiners spent an average of 51 minutes performing initial colonoscopy2 — three times as long as the typical duration of colonoscopy (15 to 20 minutes) noted in a review by Winawer et al.3 The other examiner, who spent an average of 26 minutes performing the evaluation, missed nearly twice as many lesions.2 These findings led Hixson et al. to conclude that spending more time on the examination would be a practical way of improving the detection of polyps.
The endoscope is a wonderful instrument, and unusually vigilant endoscopists can detect nearly 100 percent of polyps. The larger question is whether endoscopists are willing to spend the time to do so. This question has already been answered by Rex et al. (who cited rates of missed polyps from 17 to 48 percent, depending on the examiner),4 Winawer et al.,1,3 Hixson et al.,2 and various retrospective studies. The limitation of colonoscopy is not the technique, but the examiner.
4 ReferencesMark R. Fister, M.D.
Dennis M. Balfe, M.D.
Mallinckrodt Institute of Radiology, St. Louis, MO 631101
Winawer SJ ,Stewart ET ,Zauber AG , et al. A comparison of colonoscopy and double-contrast barium enema for surveillance after polypectomy. N Engl J Med 2000;342:1766-1772
Full Text | Web of Science | Medline2
Hixson LJ ,Fennerty MB ,Sampliner RE ,McGee D ,Garewal H . Prospective study of the frequency and size distribution of polyps missed by colonoscopy. J Natl Cancer Inst 1990;82:1769-1772
CrossRef | Web of Science | Medline3
Winawer SJ ,Fletcher RH ,Miller L , et al. Colorectal cancer screening: clinical guidelines and rationale. Gastroenterology 1997;112:594-642[Erratum, Gastroenterology 1997;112:1060, 1998;114:625.]
CrossRef | Web of Science | Medline4
Rex DK ,Cutler CS ,Lemmel GT , et al. Colonoscopic miss rates of adenomas determined by back-to-back colonoscopies. Gastroenterology 1997;112:24-28
CrossRef | Web of Science | Medline
To the Editor:
Currently, most gastroenterologists use barium enema, a test previously thought to have at least adequate sensitivity (70 to 90 percent) for detecting clinically significant polyps,1,2 to complete a colonic evaluation in a patient with an incomplete examination. Evaluating patients with incomplete examinations was not an important issue in the study by Winawer et al. because of the high overall completion rate (99 percent). As the authors indicated, this rate, which is higher than those cited in the literature, may be a result of the inclusion criterion of a previous completed colonoscopic examination. However, this high rate may be due to a sexual imbalance in the trial, because the majority of the patients (74 percent) in the paired-examination group were men. We recently reviewed the records of 2000 colonoscopic procedures performed in a variety of practice settings and found that women had a significantly lower rate of complete examination than men (94.8 percent vs. 98.2 percent, P<0.01).3
Given the poor sensitivity of barium enema for the detection of clinically significant polyps in this study, physicians may now be reluctant to rely on barium enema to complete the colonic evaluation. Perhaps virtual colonoscopy, with an apparently higher sensitivity for detecting clinically significant polyps (91 percent),4 will be considered the best method to complete the evaluation if conventional colonoscopy has failed.
4 ReferencesJoseph C. Anderson, M.D.
John W. Birk, M.D.
Robert D. Shaw, M.D.
University at Stony Brook, Stony Brook, NY 117941
Winawer SJ ,Fletcher RH ,Miller L , et al. Colorectal cancer screening: clinical guidelines and rationale. Gastroenterology 1997;112:594-642[Erratum, Gastroenterology 1997;112:1060, 1998;114:625.]
CrossRef | Web of Science | Medline2
Steine S ,Stordahl A ,Lunde OC ,Loken K ,Laerum E . Double-contrast barium enema versus colonoscopy in the diagnosis of neoplastic disorders: aspects of decision-making in general practice. Fam Pract 1993;10:288-291
CrossRef | Web of Science | Medline3
Anderson JC ,Gonzalez JD ,Messina CR ,Pollack BJ . Factors that predict incomplete colonoscopy: thinner is not always better. Am J Gastroenterol 2000;95:2784-2787
CrossRef | Web of Science | Medline4
Fenlon HM ,Nunes DP ,Schroy PC III ,Barish MA ,Clarke PD ,Ferrucci JT . A comparison of virtual and conventional colonoscopy for the detection of colorectal polyps. N Engl J Med 1999;341:1496-1503[Erratum, N Engl J Med 2000;342:524.]
Full Text | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: We agree with Dr. Glick that total eradication of all adenomas is not necessary, desirable, or possible. Dr. Glick does not dispute the desirability of identifying and removing adenomas larger than 1 cm but states that the 24 large adenomas (>1 cm) among the 791 polyps detected by colonoscopy may actually have been smaller, and therefore the accuracy of double-contrast barium enema could not have been evaluated. The adenomas larger than 1 cm had a mean size of 1.7 cm, and 21 of the 24 were 1.5 to 3.0 cm; 14 (58 percent) had villous features (Table 1Table 1
Characteristics of 24 Adenomas Larger Than 1.0 cm That Were Detected by 23 of 862 Surveillance Colonoscopies.). In addition, 25 of the 81 adenomas that ranged in size from 0.6 to 1.0 cm were 1.0 cm; 10 of these 25 adenomas (40 percent) were detected by double-contrast barium enema. Consequently, 21 of 49 adenomas (43 percent) that were greater than or equal to 1.0 cm were detected by double-contrast barium enema. In two studies reviewed by Rex,1 endoscopists visually classified polyps correctly as less than or greater than 1 cm in 91 percent and 95 percent of cases. Even with the observed variance of 3 mm among endoscopists, 21 of the 24 adenomas in the National Polyp Study that were larger than 1 cm would still be classified as large. Double-contrast barium enema detected 10 (48 percent) of these 21 adenomas. The detection rate for adenomas larger than 1 cm was similar whether the size was determined by endoscopic examination (46 percent) or pathological measurements (53 percent).There were 1.8 major complications per 1000 examinations in our cohort.2 Rex3 reviewed nine studies of surveillance colonoscopy, but deaths were cited for only one study.4 In that study, there were two colonoscopy-related deaths among approximately 4000 examinations, and there was one death from colon cancer. The expected number of deaths from colorectal cancer in the absence of polypectomy was 7.58.
In response to Drs. Fister and Balfe: the median duration of colonoscopy was 20 minutes. In spite of the known rate of missed polyps, primarily small adenomas, there was a 76 to 90 percent reduction in the incidence of colorectal cancer in our cohort.5 We have shown that colonoscopy is more effective than double-contrast barium enema in detecting the lesions (i.e., adenomas larger than 1 cm) that are most likely to result in cancer. We have an opportunity to change the focus of colorectal-cancer screening from the detection of early-stage cancers to the removal of clinically significant premalignant adenomas. With regard to extrapolation, trials are designed to demonstrate the best outcome possible. We should strive for this in practice.
In reply to Dr. Fibus: we indicated the radiologists' level of confidence in Table 2 of our article. The radiologists had a high level of confidence in the findings in all anatomical segments for only 20 percent of the double-contrast enemas. Even these had a poor sensitivity (33 percent). The low level of confidence and poor sensitivity were the best achieved by a superb group of radiologists with excellent preparation, technique, and interpretation.
In response to Anderson et al.: the rate of completed colonoscopy was 98.6 percent for men and 98.2 percent for women. Although less accurate, double-contrast barium enema is the recommended method to complement incomplete colonoscopy in the follow-up surveillance of patients after polypectomy.
Our study was sponsored by the American Gastroenterological Association, the American Society for Gastrointestinal Endoscopy, and the American College of Gastroenterology.
5 ReferencesAnn G. Zauber, Ph.D.
Sidney J. Winawer, M.D.
Memorial Sloan-Kettering Cancer Center, New York, NY 10021Edward T. Stewart, M.D.
Froedtert Memorial Lutheran Hospital, Milwaukee, WI 53226-35961
Rex DK . Endoscopists, polyp size, and post-polypectomy surveillance: making a mountain out of a molehill? Gastrointest Endosc 1997;46:571-574
Web of Science | Medline2
Schnoll-Sussman F ,Carlson MDA ,Zauber AG ,Winawer SJ . Benefits and harms of colonoscopy surveillance following colonoscopic polypectomy: report from the National Polyp Study. Am J Gastroenterol 2000;95:2549-2550
CrossRef3
Rex DK . Colonoscopy: a review of its yield for cancers and adenomas by indication. Am J Gastroenterol 1995;90:353-365
Web of Science | Medline4
Jorgensen OD ,Kronborg O ,Fenger C . The Funen Adenoma Follow-up Study: incidence and death from colorectal carcinoma in an adenoma surveillance program. Scand J Gastroenterol 1993;28:869-874
CrossRef | Web of Science | Medline5
Winawer SJ ,Zauber AG ,Ho MN , et al. Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med 1993;329:1977-1981
Full Text | Web of Science | Medline
- Citing Articles (5)
Citing Articles
1
Mohammadreza Khanmohammadi, Amir Bagheri Garmarudi, Keyvan Ghasemi, Hadigheh Kazemi Jaliseh, Ahmad Kaviani. (2009) Diagnosis of colon cancer by attenuated total reflectance-fourier transform infrared microspectroscopy and soft independent modeling of class analogy. Medical Oncology 26:3, 292-297
CrossRef2
Bernard Levin, David A. Lieberman, Beth McFarland, Kimberly S. Andrews, Durado Brooks, John Bond, Chiranjeev Dash, Francis M. Giardiello, Seth Glick, David Johnson, C. Daniel Johnson, Theodore R. Levin, Perry J. Pickhardt, Douglas K. Rex, Robert A. Smith, Alan Thorson, Sidney J. Winawer. (2008) Screening and Surveillance for the Early Detection of Colorectal Cancer and Adenomatous Polyps, 2008: A Joint Guideline From the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. Gastroenterology 134:5, 1570-1595
CrossRef3
Marc J. Gollub, Lawrence H. Schwartz, Tim Akhurst. (2007) Update on Colorectal Cancer Imaging. Radiologic Clinics of North America 45:1, 85-118
CrossRef4
Christoph Wald, Christopher D. Scheirey, Tai M. Tran, Nazli Erbay. (2006) An Update on Imaging of Colorectal Cancer. Surgical Clinics of North America 86:4, 819-847
CrossRef5
Shmuel Argov, Jagannathan Ramesh, Ahmad Salman, Igor Sinelnikov, Jed Goldstein, Hugo Guterman, Shaul Mordechai. (2002) Diagnostic potential of Fourier-transform infrared microspectroscopy and advanced computational methods in colon cancer patients. Journal of Biomedical Optics 7:2, 248
CrossRef
