Impact of a CT colonography screening program on endoscopic colonoscopy in clinical practice

American Journal of Gastroenterology  C 2008 by Am. Coll. of Gastroenterology Published by Blackwell Publishing

ISSN 0002-9270 doi: 10.1111/j.1572-0241.2007.01586.x

Impact of a CT Colonography Screening Program on Endoscopic Colonoscopy in Clinical Practice Darren C. Schwartz, M.D.,1 Kevin J. Dasher, M.D.,1 Adnan Said, M.D.,1 Deepak V. Gopal, M.D.,1 Mark Reichelderfer, M.D.,1 David H. Kim, M.D.,2 Perry J. Pickhardt, M.D.,2 Andrew J. Taylor, M.D.,2 and Patrick R. Pfau, M.D.1 1 Department of Medicine, Section of Gastroenterology & Hepatology, and 2 Department of Radiology, University of Wisconsin Medical School-Madison, Madison, Wisconsin

OBJECTIVE:

The potential effect of CT colonography (CTC) on endoscopic colonoscopy (EC) has been the topic of much speculation. The aim of this study was to evaluate the impact of a CTC screening program on colonoscopy in clinical practice.

METHODS:

At our institution a third-party reimbursed CTC colorectal cancer (CRC) screening program was established in 2004. The number of CTC monthly exams performed, monthly EC total and screening exams performed, EC with polypectomy performed, and the number of referrals for EC screening exams requested were prospectively examined in the first 33 months after introduction of a CTC CRC screening program.

RESULTS:

The mean number of overall (378.5 vs 413.1) and screening (150.7 vs 162.9) colonoscopy exams performed per month did not change significantly after screening CTC was introduced. The mean number of monthly CTC exams performed rose significantly throughout the first year of the study from 39 initially to a peak of 147.6 cases per month but decreased slightly to 114.3 monthly exams at the end of 2006. A mean 10.0 patients per month were sent for EC after a positive CTC exam. The mean number of monthly colonoscopies with polypectomy remained constant after the introduction of CTC (197.0 vs 180.2). Monthly referrals for screening EC exams initially decreased but were unchanged 3 yr after institution of a CTC screening program (255.0 vs 253.5).

CONCLUSIONS: (a) In our tertiary care center the initiation of a screening CTC program did not result in a decrease in the number of total colonoscopy exams, screening colonoscopy exams performed, nor requests for screening colonoscopy. (b) Only a small number of CTC exams were referred for EC with polypectomy, therefore a CTC screening program may not increase the overall number of therapeutic colonoscopies performed. (Am J Gastroenterol 2008;103:346–351)

INTRODUCTION

EC (5, 6). Increased interest in CTC screening is reflected in a large increase in the number of studies examining the role of CTC, including the results of a large, prospective, multicentered study in which CTC detected polyps 6 mm or greater with a sensitivity and specificity comparable to EC (7) and two other recent large multi-institutional trials that yielded less favorable results (8, 9). The potential impact of CTC on EC has been the topic of much speculation. Quantitative mathematical modeling has been employed to predict changes in EC demand if colorectal cancer basis screening with CTC were implemented on a wide scale. These studies have suggested a potential decrease of over 20% in EC demand (10, 11). While a multidisciplinary task force concluded that CTC would likely have a “significant” impact on the practice of gastroenterology in America, its effect on EC utilization could not yet be ascertained due to insufficient data regarding CTC in clinical practice (12).

Endoscopic colonoscopy (EC) exam with polypectomy detects colorectal polyps in a highly sensitive and specific manner and appears to reduce CRC incidence (1–3). Colonoscopy has become the favored means of CRC screening because of its ability to both detect and remove colorectal polyps, thus interrupting the adenoma to carcinoma sequence (4). Based on a preponderance of data and demonstrated cost-effectiveness, Medicare and most third-party payers began providing coverage for screening with EC for individuals at average risk for CRC since 2001. The first reports of polyp detection with virtual colonoscopy (VC), also known as computed tomography colonography (CTC), date back over a decade, but it was not until recently that the possibility of wide-scale CRC screening using CTC began gaining interest as a viable alternative to 346

Impact of a CT Colonography Screening Program on EC

In April 2004, third-party reimbursed average-risk CRC screening with CTC was instituted at the University of Wisconsin Hospital and Clinics. Our center is the first, and only, to our knowledge, in the United States to offer primary CTC screening covered by third-party payers (13). In the first year alone, over 1,100 average risk screening CTC were performed, with over 99% covered under managed care agreements and continued increasing numbers of CTC exams being performed (14). Herein, we report the first data on the impact of CTC screening on EC in clinical practice.

MATERIALS AND METHODS Beginning in April 2004, wide-scale CRC screening with CTC was implemented at the University of Wisconsin Hospital and Clinics in Madison, Wisconsin. The three major third-party payers in the area agreed to coverage for screening CTC in patients at average risk for CRC. Primary care providers belonging to one of the three insurance carriers in the area were given the opportunity to refer patients for CRC screening with either EC or CTC. No other insurers in the area or state covered CTC for CRC screening, thus the impact of CTC only affected our institution. Patients were referred to either CTC or EC primarily based upon the preference of the primary care physician and/or the preference of the patient. Any physician in the medical foundation is able to schedule a colonoscopy through an open access endoscopy system without need for the patient to be seen by a specialist physician prior to their screening exam. Almost all screening exams are referred from the primary care physicians in the medical foundation with referrals being made simply by contacting either the gastroenterology or radiology sections. The CTC technique and interpretation have been previously described in a detailed manner (14). CTC interpretation is employed using the software system V3D Colon (Viatronix, Stony Brook, NY). A biphasic interpretative approach is used in which a three-dimensional “fly-through” of the colon is used for primary detection of polyps and twodimensional imaging is used to confirm findings. CTC exams are generally performed in the morning with the option of same day EC and polypectomy if recommended. CTC exams were read by one of three radiologists all with at least 3 yr experience in interpreting CTC. The Institutional Review Board approved CTC screening protocol at our institution is primarily determined by the size of the polyps visualized on CTC exam. Patients with polyps of size 10 mm or greater or patients with 3 or more polyps ≥6 mm are recommended to undergo same day colonoscopy and polypectomy. Patients with polyps in the intermediate range of 6–9 mm are given the option of same day polypectomy or entering ongoing CTC surveillance, with polyps measuring 6–7 mm having a repeat CTC every 2 yr and polyps measuring 8–9 mm having a repeat CTC every year. Polyps measuring 5 mm or less are not reported per CTC protocol. The patients choosing polyp surveillance are part of a

347

research protocol and must sign a written consent form to do so. Data on EC was tracked on all exams performed for screening, diagnostic, and therapeutic (polypectomy) indications at the hospital-owned endoscopy suite, which includes 7 endoscopy rooms. Colonoscopies were performed by a total of 10 gastroenterology faculty members. The number of CTC and EC tests performed and referrals for screening colonoscopy (number of tests requested by primary care physicians) were collected prospectively. We reviewed and compared data from the 3–6 month period before the introduction of screening CTC to our institution to the time period 33 months after the CTC program began. Data analyzed were monthly referrals/requests for screening EC, number of monthly screening EC performed, total EC (screening, diagnostic, and therapeutic) performed, and EC with polypectomy (therapeutic) exams during the above selected time periods. We also analyzed monthly screening CTC performed, and number of CTC referred to EC for polypectomy during the same time period. Comparisons were made between the time period 3–6 months before the introduction of the CTC screening program and the 33 months after the initiation of the CTC screening program through the end of 2006. The study was approved by the Institutional Review Board (IRB).

RESULTS The mean number of total colonoscopy exams performed in the 6 months prior to the CTC program was 378.5 exams per month. Total monthly colonoscopy exams performed did not change significantly after the introduction of screening CTC, with a mean 413.1 total colonoscopy procedures performed per month after the CTC program began (378.5 vs 413.1) (Fig. 1). Likewise, there was no difference in the mean number of monthly screening EC exams performed during the period before (150.7 monthly exams performed) CTC compared to the productivity following introduction of screening CTC, with a mean 162.9 monthly screening colonoscopy exams performed after introduction of the CTC program (150.7 vs 162.9) (Fig. 2). The introduction of CTC had no effect on the monthly total number of EC with polypectomy performed at our institution. The mean number of monthly EC with polypectomy did not vary across study periods. Before the introduction of CTC screening 197.0 monthly exams with polypectomy were performed as compared to 180.2 monthly exams with polypectomy after the CTC program began (197. 0 vs 180.2) (Fig. 3). The CTC program as stated was initiated in April of 2004. The mean number of screening CTC exams performed in the first three months of the program was 39 cases per month. Monthly CTC cases performed rose to a peak of 147.6 cases per month in mid-2005 but has decreased slightly to a level of

348

Schwartz et al.

Figure 1. Total monthly endoscopic colonoscopies performed before and after screening CTC program began. CTC program began in April 2004 (indicated by arrow).

114.3 monthly screening CTC cases being performed at the end of 2006 (Fig. 4). A mean 10.0 patients were sent because of polyps seen on CTC exams to gastroenterology per month for EC and polypectomy during the course of the study. Referrals or requests for screening colonoscopy dropped slightly during the first fourteen months after the introduction of screening CTC (255.0 mean monthly requests for screening colonoscopy prior to CTC versus 217.5 mean monthly requests for screening colonoscopy in first fourteen months). However, examining all 33 months after the introduction of a screening CTC program, there has been no change in the mean monthly requests for screening colonoscopy (255.0 preCTC vs 253.5 post-CTC) (Fig. 5).

DISCUSSION The impact of wide-spread screening CTC on EC demand has been a matter of conjecture. Studies using mathematical modeling have suggested that CTC may affect the current practice of CRC screening with endoscopy, with a predicted

decrease in EC exams ranging from 9 to 22% (10). However, such studies are riddled with the inherent inaccuracies plaguing all hypothetical models. Moreover, speculative studies such as these cannot quantify the effects patient and referring provider preferences have on utilization of a new modality such as CTC. It is therefore crucial to acquire data from clinical practice before assessing the impact a novel technology like CTC may have on a gold standard, in this case EC. Despite the introduction of screening CTC to our institution, the monthly numbers of EC performed did not suffer, both in terms of total overall diagnostic and therapeutic (polypectomy) EC or screening EC performed. The possible reasons why CTC did not have an immediate affect on colonoscopy numbers include the fact that when CTC was introduced there was a surplus of patients waiting for colon cancer screening with a waiting list for screening colonoscopy greater than 12 months, which has been essentially eliminated after the initiation of the CTC program. Further, the addition of CTC may have resulted in patients being screened by CTC

Figure 2. Total monthly screening endoscopic colonoscopies performed before and after screening CTC program began. CTC program began in April 2004 (indicated by arrow).

Impact of a CT Colonography Screening Program on EC

349

Figure 3. Total monthly endoscopic colonoscopies with polypectomy before and after screening CTC program began. CTC program began in April 2004 (indicated by arrow).

that would not have been screened by EC and thus did not affect the number of EC exams performed. Thus, we initially believed referral numbers or requests for screening colonoscopy may represent a more accurate assessment of the early impact of CTC on real-time screening EC demand. Despite stable EC numbers being performed with our endoscopy unit working at full capacity, we did experience a decrease in referrals for screening EC during the first year after the screening CTC program began. This occurred once screening CTC productivity hit a “steady state” of approximately 100–150 cases per month. The absolute reduction in screening EC requests in the first year after the introduction of a CTC program was approximately 51 less requests a month, or a 20% drop in requests for screening EC compared to the period immediately preceding CTC. The reasons for this early decrease in requests for screening EC was likely multifactorial. With any new technology, there is initial curiosity and intrigue, which may affect early referral patterns. Moreover, a recent survey study of our primary care providers suggests that some medical providers

Figure 4. Mean monthly CTC exams performed from April 2004 through December 2006. Data are presented in three-month quarters.

perceived CTC as a cheaper, more tolerable, safe, and accessible alternative to EC (15). The primary reason, however, that requests for EC screening decreased initially likely was that CTC provided another screening option to primary care providers that was readily and quickly available. However, the initial decrease in requests for screening colonoscopy once the CTC program was instituted has not lasted for the 3 yr that the CTC program has been in place. Requests for screening colonoscopy now at a time 3 yr past the beginning of the CTC screening program show no significant change or decrease and in fact have slightly increased as compared to requests for screening colonoscopies prior to CTC. The number of CTC screening exams performed per month peaked at year one after the start of the program, with approximately 150 CTC exams being done per month, and has dropped off slightly in the third year of the program to about 115 CTC cases per month. The end result is that the CTC screening program has had no impact at all on the number of total colonoscopies and screening colonoscopies requested for or performed at our institution even after almost 3 yr. The fact that the CTC program has not affected EC at our institution after 3 yr is contrary to our original assumption and hypothesis that CTC would result in a decrease in requests for screening colonoscopy and eventually result in a decrease in colonoscopy exams performed. The reasons why our screening colonoscopy numbers and requests likely have not suffered is screening CTC is a different test from screening colonoscopy and thus has not replaced screening colonoscopy but merely added an additional screening option with its own advantages and disadvantages. While CTC offers the option of a screening test without sedation and without the very small but still present risk of the complications of perforation and bleeding associated with screening colonoscopy and polypectomy, it also has obvious disadvantages as compared to colonoscopy. The most notable is that CTC is not a therapeutic test and cannot offer the same test therapy and polypectomy as a colonoscopy can. Radiation with CTC has been a concern, with an estimation

350

Schwartz et al.

Figure 5. Total monthly requests or referrals for screening colonoscopy before and after initiation of the CTC screening program. CTC screening program began in April 2004 (indicated by arrow).

of a 0.14% risk of cancer with CTC screening (16). This estimation, however, does not take into consideration repeated CTC imaging if polyp surveillance is employed as it is at our institution, which may make the lifetime risk of radiation clinically significant. Finally, extracolonic findings on CTC has generated concern from primary care providers in our institution because of generation of patient anxiety and generation of increased workload for the primary-care physicians attempting to follow-up on the extracolonic findings. Thus, while a number of primary-care practitioners and patients may request and be interested in screening with CTC, a still very large number of patients and primary physicians will still choose colonoscopy as the screening option. Some have speculated that any potential decrease in average risk screening EC productivity caused by CTC would be offset by an increase in EC with polypectomy resulting from CTC polyp detection (11, 12). However, this supposition is based upon a practice of universal polypectomy (i.e., removal of all colorectal polyps found at CTC, regardless of size). Our data demonstrate that if universal polypectomy is abandoned in favor of “selective polypectomy” (i.e., CTC surveillance for polyps 6–9 mm in size, and polyps 5 mm or smaller are not reported), then only a very small percentage of screening CTC cases will be referred for subsequent EC and thus not increase the number of “therapeutic” colonoscopies. For example, with CTC performing almost 120 screening exams per month in the later months of the study, only 10 patients per month were sent for colonoscopic polypectomy after CTC exam. The low number of patients sent to EC is reflective of the fact that the majority of patients with medium size polyps (6–9 mm) on CTC chose CTC surveillance instead of polypectomy and diminutive polyps 5 mm or less are not reported by CTC at our institution’s CTC screening program. Evidence is presently unclear and mixed about the percentage of subcentimeter polyps that may harbor advanced histology (17–19). At our institution an IRB protocol is in place, wherein patients with polyps of 6–9 mm are closely followed with yearly or every other year repeat CTC exams

with close tracking of all these patients by the involved radiologists. Until the natural history of subcentimeter colorectal polyps is better understood, it remains undecided whether a strategy of “selective polypectomy” will be widely endorsed outside of a study protocol, particularly by practicing gastroenterologists.

CONCLUSIONS The initiation of a screening CTC program at our institution did not result in a decrease in the numbers of endoscopic colonoscopies performed in the first 33 months since its inception. CTC thus has not replaced colonoscopy or even impacted colonoscopy as a screening tool but appears to have provided an additional screening option. Only a small percentage of CTC exams were referred for EC with polypectomy, therefore a CTC screening program employing a strategy of “selective polypectomy” of subcentimeter polyps does not appear to significantly increase the overall number of therapeutic colonoscopies performed. Further studies will be needed to determine how CTC may affect other academic screening programs as well as a community-based colon cancer screening program and how a CTC program that has a different policy on CTC polyp referral to EC may affect colonoscopy with polypectomy numbers.

STUDY HIGHLIGHTS What Is Current Knowledge

r r

Computed tomography (CT) colonography is an emerging total colon-screening tool that has the potential to detect polyps at similar accuracies and sensitivities to endoscopic colonoscopy. Various mathematical modeling studies have assessed the impact of CT colonography on screening for colorectal cancer with endoscopic colonoscopy.

Impact of a CT Colonography Screening Program on EC

What Is New Here

r

r

This study presents the first clinical data on the impact of CT colonography on endoscopic colonoscopy and suggests that the introduction of a CT colonography program does not appear to affect the number of colonoscopies performed nor demand for screening colonoscopy. If a CT colonography program employs only selective polyp referral to endoscopic colonoscopy for polypectomy based upon size, CT colonogarphy does not result in an increase in therapeutic colonoscopy or the number of polypectomies.

13. 14.

15.

16. 17.

Reprint requests and correspondence: Patrick R. Pfau, M.D., Assistant Professor of Medicine, Director of Gastrointestinal Endoscopy, Section of Gastroenterology and Hepatology, University of Wisconsin Medical School, H6/516 Clinical Science Center, 600 Highland Avenue, Madison, WI 53792-5124. Received January 7, 2007; accepted August 27, 2007.

REFERENCES 1. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med 1993;329:1977–81. 2. Lieberman DA, Weiss DG, Bond JH, et al. Use of colonoscopy to screen asymptomatic adults for colorectal cancer. Veterans Affairs Cooperative Study Group 380. N Engl J Med 2000;343:162–8. 3. Walsh JM, Terdiman JP. Colorectal cancer screening: Scientific review. JAMA 2003;289:1288–96. 4. Bond JH. Clinical evidence for the adenoma-carcinoma sequence, and the management of patients with colorectal adenomas. Semin Gastrointest Dis 2000;11:176–84. 5. Vining DJ, Gelfand DW, Bechtold RE, et al. Technical feasibility of colon imaging with helical CT and virtual reality. AJR Am J Roentgenol 1994;162(Suppl):104. 6. Hara AK, Johnson CD, Reed JE, et al. Detection of colorectal polyps with CT colonography: Initial assessment of sensitivity and specificity. Radiology 1997;205:59–65. 7. Pickhardt PJ, Choi JR, Hwang I, et al. Computed tomographic virtual colonoscopy to screen for colorectal neoplasia in asymptomatic adults. N Engl J Med 2003;349:2191– 200. 8. Cotton PB, Durkalski VL, Pineau BC, et al. Computed tomographic colonography (virtual colonoscopy): A multicenter comparison with standard colonoscopy for detection of colorectal neoplasia. JAMA 2004;291:1713–9. 9. Rockey DC, Paulson E, Niedzwiecki D, et al. Analysis of air contrast barium enema, computed tomographic colonography, and colonoscopy: Prospective comparison. Lancet 2005;365:305–11. 10. Hur C, Gazelle S, Zalis ME, et al. An analysis of the potential impact of computed tomographic (virtual colonoscopy) on colonoscopy demand. Gastroenterology 2004;127:1312–21. 11. Ladabaum U, Song K. Projected national impact of colorectal cancer screening on clinical and economic outcomes and health services demand. Gastroenterology 2005;129:1151– 62. 12. Van Dam J, Cotton P, Johnson CD, et al. AGA future trends

18. 19.

351

report: CT colonography. Gastroenterology 2004;127;970– 84. Barnes E. HMO pays for screening virtual colonoscopy. AuntMinnie.com, June 4, 2004. Available at: http://www. auntminnie.com. Accessed June 4, 2004. Pickhardt PJ, Taylor AJ, Reichelderfer M, et al. Primary colorectal screening with virtual colonoscopy: Program results from the first year of coverage by third-party payers. Radiology 2006;241:417–25. Schwartz DC, Said A, Gopal DV, et al. Perceptions and referral practices of primary care providers given a choice including both screening virtual colonoscopy and optical colonoscopy. Gastrointest Endosc 2006;63:AB204. Brenner DJ, Georgsson MA. Mass screening with CT colonography: Should the radiation exposure be of concern? Gastroenterology 2005;129:328–37. Butterly LF, Chase MP, Pohl H, et al. Prevalence of clinically important histology in small adenomas. Clin Gastroenterol Hepatol 2006;4:343–8. Chen SC, Mouchli A, Chadalawada V, et al. Histopathology of small polyps removed in the videoendoscopic era. Gastrointest Endosc 2006;63:AB199. Odom SR, Duffy SD, Barone JE, et al. The rate of adenocarcinoma in endoscopically removed colorectal polyps. Ann Surg 2005;71:1024–6.

CONFLICT OF INTEREST Guarantor of the article: Patrick Pfau, M.D. Specific author contributions: Darren C. Schwartz: conception and design, acquisition of data, analysis and interpretation of the data, drafting of the manuscript, critical revision of the manuscript. Kevin J. Dasher: conception and design, acquisition of data, analysis and interpretation of the data, drafting of the manuscript, critical revision of the manuscript. Adnan Said: conception and design, analysis and interpretation of the data, drafting of the manuscript, critical revision of the manuscript, statistical analysis. Deepak V. Gopal: conception and design, acquisition of data, analysis and interpretation of the data, drafting of the manuscript, critical revision of the manuscript. Mark Reichelderfer: conception and design, acquisition of data, analysis and interpretation of the data, drafting of the manuscript, critical revision of the manuscript. David H. Kim: conception and design, acquisition of data, analysis and interpretation of the data, drafting of the manuscript, critical revision of the manuscript. Perry J. Pickhardt: conception and design, acquisition of data, analysis and interpretation of the data, drafting of the manuscript, critical revision of the manuscript. Andrew J. Taylor: conception and design, acquisition of data, analysis and interpretation of the data, drafting of the manuscript, critical revision of the manuscript. Patrick R. Pfau: conception and design, acquisition of data, analysis and interpretation of the data, drafting of the manuscript, critical revision of the manuscript, supervision. Financial support: None. Potential competing interests: Dr. Pickhardt is a consultant for Viatronix. No other potential competing interests exist for the authors.