 |
|
Colon Polyp Recurrence in a Managed Care Population
Marianne Ulcickas Yood, DSc, MPH;
Susan Oliveria, ScD, MPH;
J. Gregory Boyer, PhD;
Karen Wells, BS;
Paul Stang, PhD;
Christine Cole Johnson, PhD, MPH
Arch Intern Med. 2003;163:422-426.
ABSTRACT
| |
Background Detection, treatment, and subsequent screening of patients with colorectal polyps have considerable implications for public health. Understanding the natural history of colorectal polyps from biologic and health care services standpoints is critical in determining optimal treatment.
Methods Using computerized data from a large Midwestern health maintenance organization, we identified patients 50 years or older who underwent a polypectomy (index polypectomy) between January 1, 1989, and December 31, 1999. We followed up patients to identify subsequent polypectomies through September 1, 2001.
Results Our final study sample consisted of 8865 individuals with an index polypectomy. Overall, 2704 patients (30.5%) were diagnosed as having recurrent polyps. Kaplan-Meier projections estimate that 50% of patients will have a recurrence within 7.6 years (95% confidence interval [CI], 7.2-8.1 years). Among patients who underwent colon screening at least 9 months after the index polypectomy (52% of original cohort), Kaplan-Meier projections estimate that 50% will have a recurrent polyp within 3.9 years (95% CI, 3.8-4.1 years).
Conclusions The results demonstrate that, even when screening and treatment are received by those who need it, the risk of colon polyp recurrence is high and many patients do not undergo additional screening. Efforts to increase and monitor ongoing screening of postpolypectomy patients are warranted. In addition, further research is needed to better identify patients at risk for recurrence so that continued screening and other interventions can be targeted at these groups.
INTRODUCTION
COLORECTAL CANCERS are the third most common cancers in men and women, with an estimated 135 400 new cases and 56 700 deaths in 2001.1 Adenomatous polyps of the colon or rectum have been identified as precursors to colorectal cancer2-6 and as such are of critical importance in the early identification and potential prevention of colorectal cancer. The progression of adenomatous polyps to adenocarcinoma, which has been termed the adenoma-adenocarcinoma sequence, is the basis of current clinical practice and is supported by autopsy, clinical, epidemiologic, and molecular genetic studies.4, 7-10
The prevalence of adenomas is approximately 30% to 40% in people 60 years and older. However, the factors that predict who will develop cancer or who is likely to have recurrent polyps are not yet clearly determined.11 Histologic change and morphologic factors have been identified as important characteristics of adenomas that are predictive of progression to cancer, but there are likely to be other factors that influence risk for the adenoma-adenocarcinoma sequence.1, 3, 10
The initial treatment of patients with polyps includes removal of all polyps visualized during colonoscopic screening to eliminate potential neoplastic lesions and reduce the likelihood of subsequent colorectal cancer morbidity and mortality.10 Results of observational studies provide consistent evidence for these recommendations.1, 3, 12 Given that it is routine clinical practice to remove all polyps, the rate of progression to cancer in the absence of polypectomy cannot be determined. Observational data collected before the routine use of sigmoidoscopy, colonoscopy, and polypectomy suggest that the cumulative risk of progression of polyps to cancer is 4% to 35% throughout 5 to 20 years.13 A 2- to 4-fold increased risk of colorectal cancer for patients diagnosed as having an adenomatous polyp (compared with the general population) has also been reported.14-15 A persistent risk of colorectal cancer exists even with polypectomy, because patients who undergo a colonoscopic polypectomy are likely to have additional polyps detected during follow-up screening; the percentage of adenomas detected after polypectomy can range from 37% to 60%.16-19
The detection, treatment, and subsequent screening of patients with colorectal polyps have considerable implications for public health. A comprehensive understanding of the natural history of colorectal polyps from biologic and health care services standpoints is critical in determining optimal treatment strategies. Current clinical management of polyps consists of removal of all polyps together with a plan for follow-up screening. However, considerable debate exists about appropriate follow-up intervals. Two randomized trials, the National Polyp Study (NPS)4 in the United States and the Funen study20 in Denmark, have investigated different screening intervals. In each trial, patients with adenoma and without a prior history of polyps or colorectal cancer were randomized to receive follow-up on 2 different postpolypectomy schedules: at 1 and 3 years or at 3 years only in the NPS and at 2 or 4 years in the Funen study. In both studies, a significant number of patients produced additional polyps. In the NPS, 27% of the patients experienced a polyp recurrence after 1 year and 32% after 3 years. The Funen study found that polyps recurred in 20% of patients after 2 years, 35% after 4 years, and 50% after 8 years.
To quantify the risk of and time to recurrent polyps in real-world clinical practice, we conducted a population-based retrospective cohort study of 8865 individuals who were a part of a large health maintenance organization (HMO) and for whom colon cancer screening was a covered benefit. This setting provided a unique opportunity to meaningfully assess real-world clinical practice because of the comprehensive sets of patient information maintained electronically and because of the racial diversity of its patient population.
METHODS
DATA SOURCES AND STUDY POPULATION
Patients for this study were identified from a large, vertically integrated health system serving the primary and specialty health care needs of Midwestern residents. This system is affiliated with a multispecialty, salaried physician group that provides most of the care for health system patients. The health system also owns a large, nonprofit, mixed-model HMO. To optimize the computerized data available for this study, the population was limited to HMO members assigned to the medical group physicians.
COMPUTERIZED DATA SOURCES
The health system maintains an extensive, centralized system of computerized databases. Data for this study primarily came from 2 computerized sources: encounter and claims records and the HMO membership file. The encounter and claims databases include comprehensive patient data for outpatient, emergency department, and inpatient care delivered by the medical group. For each outpatient encounter, information on date of visit, diagnoses, physician delivering care, procedures delivered, and clinic in which the care was delivered is compiled. Likewise, for each inpatient stay, information on admission and discharge date, diagnoses, and procedures is compiled. Procedure data, which include polypectomies, are recorded using Current Procedural Terminology codes and codes specifically created for internal system purposes. In addition to these encounter-specific fields, data on the patient's age, sex, and race are available. The HMO also maintains detailed contractual information for each of its enrollees. These data are available via the HMO membership database and include historical and current coverage dates, covered benefits, and the amount of copayments for which the enrollee is responsible.
STUDY POPULATION
Using the computerized data described herein, we identified all patients 50 years or older who were members of the HMO, received care from the medical group physicians, and underwent a polypectomy between January 1, 1989, and December 31, 1999. This first polypectomy within the study period was termed the index polypectomy. From this group, we excluded (1) patients diagnosed as having Crohn disease or ulcerative colitis anytime before or within 3 months after the index polypectomy, (2) patients diagnosed as having colorectal cancer anytime before or within 3 months after the index polypectomy, and (3) patients enrolled in the HMO for less than 1 year after the time of the index polypectomy.
ASCERTAINMENT OF FOLLOW-UP DATA
We followed up all patients who met the eligibility criteria to determine whether a subsequent polypectomy occurred through September 1, 2001. Because we were interested in identifying new polyps (ie, not additional procedures related to the index polypectomy), we included as follow-up events only those polypectomies that occurred at least 9 months after the index date. In addition to analyzing the entire study population, we also conducted an analysis among the subset of patients who had subsequent colon-related screening during the follow-up period, because those undergoing screening had the opportunity to have a recurrence detected.
To evaluate the impact of health care–related resource use among patients who underwent polypectomies, we identified a comparison group of patients, sampled from the same base population as the polypectomy patients, matched on sex, race, and age (±5 years). The date of index polypectomy served as the index date from which resource use was measured in the matched comparison group. Resource use, including all outpatient visits and inpatient stays, was captured from the index date until the end of HMO enrollment or the end of the study period, whichever came first.
DATA ANALYSIS
Among the cohort of patients who underwent a polypectomy, we calculated the proportion of patients who had a recurrent polyp and the relative risk (RR) of recurrence, with corresponding 95% confidence intervals (CIs). We fit Kaplan-Meier survival curves to estimate the proportion of patients who were free of recurrence throughout the study period, overall and limited to patients who had a second colon-related procedure (colonoscopy, sigmoidoscopy, or barium enema). To evaluate resource use, we tabulated primary care visits, specialty visits, and inpatient stays for the polypectomy and comparison groups.
RESULTS
POLYP RECURRENCE
We identified 12 351 patients 50 years or older who underwent a polypectomy between January 1, 1989, and December 31, 1999. From this group of 12 351 index polypectomies, we excluded 2242 patients with a diagnosis of Crohn disease or ulcerative colitis and 524 patients with a diagnosis of colorectal cancer anytime before or within 3 months after the index polypectomy. To ensure that we obtained complete information on study patients, we further limited the study population to individuals who had HMO insurance for at least 1 year, for a final study sample of 8865 individuals.
The characteristics of the study population are given in Table 1. Almost 60% of the study population was male, nearly 24% of the population was 70 years or older, and approximately 25% of the population was African American. Almost 60% of the individuals were enrolled in the health plan for 3 years or more after the index polypectomy.
|
|
|
|
Table 1. Age and Sex Distribution of Final Study Population
|
|
|
Table 2 presents the number of patients who had a polyp recurrence at least 9 months after the index polypectomy through September 2001. The mean follow-up time for the study cohort was 4.6 years (median, 4.1 years), with a minimum follow-up time of 1 year after the index polypectomy and a maximum of 11.8 years. Overall, 2704 patients (30.5%) were diagnosed as having recurrent polyps. Men were slightly more likely to be diagnosed as having a recurrent polyp (RR, 1.2; 95% CI, 1.1-1.3), and African American patients were less likely to have a recurrence identified (RR, 0.75; 95% CI, 0.70-0.80).
|
|
|
|
Table 2. Polyp Recurrence by Demographic Characteristics for Entire Study Population
|
|
|
Figure 1 shows the Kaplan-Meier projections for a recurrent polyp diagnosis over time. These figures indicate that 50% of patients who had an initial polyp will have a recurrence within 7.6 years (95% CI, 7.2-8.1 years). These projections further estimate that 25% of patients will be diagnosed as having recurrence within 3.3 years (95% CI, 3.2-3.4 years).
|
|
|
|
Figure 1. Kaplan-Meier estimates of the proportion of patients without recurrence over time.
|
|
|
The estimate of incidence of recurrence depends not only on the natural history of the disease but also on the proportion of the population that receives additional screening and therefore has the opportunity to have an asymptomatic recurrence detected. To examine this issue, we also calculated the proportion of patients with recurrence, limiting the patient population to those individuals who had a second colon-related procedure during the study period. We found that 4628 patients (52% of the original cohort) had colon screening at least 9 months after the index date. Among this group of patients, Kaplan-Meier projections estimate that half (50%) will have a recurrent polyp detected within 3.9 years (95% CI, 3.8-4.1 years; Figure 2).
|
|
|
|
Figure 2. Kaplan-Meier estimates of the proportion of patients without recurrence over time among patients who had a second colon-related procedure.
|
|
|
RESOURCE USE
The proportions of patients in the polyp and comparison groups who had primary care visits, specialty visits, and hospital stays are presented in Table 3. The polyp cohort had more health care use than the comparison cohort. Patients who underwent polypectomy were 30% more likely to have a specialty visit, 20% more likely to have a primary care visit, and almost 40% more likely to have a hospital stay. During the study period, the polyp cohort had an average of 12.5 primary care visits and 15.6 specialty visits compared with 8.6 primary care and 13.8 specialty visits in the comparison group.
|
|
|
|
Table 3. Health Care Use Among Polyp and Comparison Cohorts
|
|
|
COMMENT
In this study, conducted using current clinical practice–based data in a racially diverse managed care environment, 30.5% of patients with polyps had subsequent polyps. We found no large differences in recurrence by sex, race, or age. Kaplan-Meier projections showed that, in the overall study population (including both patients who returned for screening and those who did not), 50% of patients will have a recurrence within 7.6 years. These results are similar to those obtained in important studies in this clinical area, including the 2 randomized trials: the NPS and the Funen study; in the NPS, 27% of patients experienced a polyp recurrence after 1 year and 32% after 3 years; in the Funen study, polyps recurred in 20% of patients after 2 years, 35% after 4 years, and 50% after 8 years.3, 20
Several other observational studies have examined polyp recurrence rates. However, patient accrual for the most recently published studies occurred between 1975 and 1990,21-25 which was before or during the initial use of endoscopic removal of polyps in clinical practice. Nonetheless, 3 of these studies21-23 also had recurrence rates similar to those found in both the NPS and the present study (28%-37% between 1.4 and 5 mean years of follow-up). One study25 found a much lower recurrence rate (6% of patients with single adenomas after 4 years; 5.7% of patients with multiple adenomas after 2 years), but the definition of recurrence used was much more restrictive. Another study, with a small sample size (n = 44), found a high recurrence rate (59%) after a median follow-up time of 12 months.
The results presented in this article demonstrate that even when screening and treatment are received by those who need it, the public health burden of subsequent polyps is large, not only because of the biologically inherent high risk of patients with an initial polyp but also because many patients then require surveillance. In this study, only 52% of patients had subsequent colon screening. However, the proportion of patients who received screening in this population is actually higher than the reported 19% of patients 50 years and older in the National Health Interview survey who, in 1998, reported use of screening endoscopy within the previous 3 years.26 As expected, in the present study, health care–related resource use among patients with an index polyp was greater than an age- and sex-matched comparison group. Subsequent work may focus on determining the extent to which the difference is attributable to their polyps or to comorbidities.
This study is limited by the fact that data were not available to classify polyps by histologic findings. Therefore, the patient population consists of a heterogeneous sample with differing risks for subsequent colorectal cancer. Although there is some evidence suggesting that a proportion of hyperplastic polyps may be a premalignant condition,27-28 sharing risk factors similar to adenomas,29 current clinical consensus is that hyperplastic polyps are benign lesions.30 Nonetheless, this study was designed to estimate the time to polyp recurrence and measured the frequency of additional polypectomy. Therefore, regardless of histologic findings, the end points in this analysis included patients who did in fact have an additional polyp removed. Furthermore, the institutional guidelines for patients with hyperplastic polyps recommend additional colonoscopy in 1 year, reverting to colonoscopy or barium enema and sigmoidoscopy every 3 to 5 years if the colon is cleared.
In this study, among the group of patients who underwent additional screening procedures, Kaplan-Meier projections estimated that 50% of patients would have recurrent polyps detected within 3.9 years. These results, in a real-world clinical practice in which insurance-related barriers to screening are not an issue, indicate that patients with polyps are at high risk for polyp recurrence. The results of this study also demonstrate that half of patients received additional colon screening, despite the fact that surveillance of this population is generally recommended. Impediments to colon screening in this population are not definitely known. However, factors such as suboptimal patient knowledge and awareness, assumption that colorectal cancer risk is symptom dependent, competing health risks, and perceptions that sigmoidoscopy is dangerous and undignified have been cited as barriers to screening; among patients who have been screened, barriers include perception that a previous negative screening result or a screening in which polyps were removed makes subsequent screening unnecessary.31 Efforts to increase and monitor ongoing screening of postpolypectomy patients and efforts to improve appropriate colorectal cancer screening activities are warranted. In addition, further research is needed to better identify patients at risk for recurrence so that continued screening and other interventions can be targeted at these groups.
AUTHOR INFORMATION
Corresponding author and reprints: Marianne Ulcickas Yood, DSc, MPH, Galt Associates, 46040 Center Oak Plaza, Suite 100, Sterling, VA 20166 (e-mail: marianneu{at}galt-assoc.com).
Accepted for publication June 14, 2002.
This study was funded by a grant from Pharmacia Corporation, Peapack, NJ.
This study was presented at the American Society for Preventive Oncology meeting, Bethesda, Md, March 11, 2002.
We thank Sharon Hensley-Alford, MPH, and Lynn H. Hoffman, MA, MPH, for their assistance with literature review and manuscript preparation.
From Galt Associates, Sterling, Va (Drs Yood, Oliveria, and Stang); Department of Epidemiology and Public Health, Yale University School of Medicine, New Haven, Conn (Dr Yood); Josephine Ford Cancer Center, Henry Ford Health System, Detroit, Mich (Drs Yood and Johnson and Ms Wells); and Pharmacia Corporation, Peapack, NJ (Dr Boyer).
REFERENCES
| |
1. American Cancer Society. Cancer Prevention and Early Detection: Facts and Figures 2001. Atlanta, Ga: American Cancer Society; 2001.
2. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med. 1993;329:1977-1981.
FREE FULL TEXT
3. Winawer SJ, O'Brien MJ, Waye JD, et al. Risk and surveillance of individuals with colorectal polyps. Bull World Health Organ. 1990;68:789-795.
WEB OF SCIENCE
| PUBMED
4. Winawer SJ, Zauber AG, O'Brien MJ, et al. Randomized comparison of surveillance intervals after colonoscopic removal of newly diagnosed adenomatous polyps. N Engl J Med. 1993;328:901-906.
FREE FULL TEXT
5. Muto T, Bussey HJR, Morson BC. The evolution of cancer of the colon and rectum. Cancer. 1975;36:2251-2257.
WEB OF SCIENCE
| PUBMED
6. Eide TJ. Risk of colorectal cancer in adenoma-bearing individuals within a defined population. Int J Cancer. 1986;38:173-176.
WEB OF SCIENCE
| PUBMED
7. Correa P, Strong JP, Reif A, et al. The epidemiology of colorectal polyps: prevalence in New Orleans and international comparisons. Cancer. 1977;39:2258-2264.
PUBMED
8. Morson BC. The polyp-cancer sequence in the large bowel. Proc R Soc Med. 1974;67:451-457.
WEB OF SCIENCE
| PUBMED
9. Jackman RJ, Mayo CW. The adenoma-carcinoma sequence in cancer of the colon. Surg Gynecol Obstet. 1951;93:327-330.
10. Vogelstein B, Fearon ER, Hamilton SR, et al. Genetic alterations during colorectal-tumor development. N Engl J Med. 1988;319:525-532.
ABSTRACT
11. Zauber AG, Winawer SJ. Initial management and follow-up surveillance of patients with colorectal adenomas. Gastroenterol Clin North Am. 1997;26:85-101.
FULL TEXT
| PUBMED
12. Jorgensen OD, Kronborg O, Fenger C. The Funen adenoma follow-up study: incidence and death from colorectal carcinoma. Scand J Gastroenterol. 1993;28:869-874.
PUBMED
13. Stryker SJ, Wolff BG, Culp CE, et al. Natural history of untreated colonic polyps. Gastroenterology. 1987;93:1009-1013.
WEB OF SCIENCE
| PUBMED
14. Otchy DP, Ransohoff DF, Wolff BG, et al. Metachronous colon cancer in persons who have had a large adenomatous polyp. Am J Gastroenterol. 1996;91:448-454.
WEB OF SCIENCE
| PUBMED
15. Atkin WS, Morson BC, Cuzick J. Long-term risk of colorectal cancer after excision of rectosigmoid adenomas. N Engl J Med. 1992;326:658-662.
ABSTRACT
16. Macrae FA, Williams CB. A prospective colonoscopic follow-up study of 500 adenoma patients with multivariate analysis to predict risk of subsequent colorectal tumors [abstract]. Gastroenterology. 1982;82:1122.
17. Waye JD, Braundfeld SF. Surveillance intervals after colonoscopic polypectomy. Endoscopy. 1982;14:79-81.
PUBMED
18. Fowler DL, Hedberg SE. Follow-up colonoscopy after polypectomy [abstract]. Gastrointest Endosc. 1980;26:67.
19. Neugut AI, Jacobsen JS, Ahsan H, et al. Incidence and recurrence rates of colorectal adenomas: a prospective study. Gastroenterology. 1995;108:402-408.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
20. Jorgensen OD, Kronborg O, Fenger C. A randomized surveillance study of patients with pedunculated and small sessile tubular and tubulovillous adenomas. Scand J Gastroenterol. 1995;30:686-692.
WEB OF SCIENCE
| PUBMED
21. Hintze RE, Adler A, Veltzke W. Endoscopic resection of large colorectal adenomas: a combination of snare and laser ablation. Endoscopy. 1995;27:665-670.
PUBMED
22. Eckardt VF, Fuchs M, Kanzler G, et al. Follow-up of patients with colonic polyps containing severe atypia and invasive carcinoma. Cancer. 1988;61:2552-2557.
FULL TEXT
| PUBMED
23. Olsen HW, Lawrence WA, Snook CW, Mutch WM. Review of recurrent polyps and cancer in 500 patients with initial colonoscopy for polyps. Dis Colon Rectum. 1988;31:222-227.
WEB OF SCIENCE
| PUBMED
24. Nava H, Carlsson G, Petrelli NJ, et al. Follow-up colonoscopy in patients with colorectal adenomatous polyps. Dis Colon Rectum. 1987;30:465-468.
PUBMED
25. Matek W, Guggenmoos-Holzmann I, Demling L. Follow-up of patients with colorectal adenomas. Endoscopy. 1985;17:175-181.
PUBMED
26. Breen N, Wagener DK, Brown ML, Davis WW, Ballard-Barbash R. Progress in cancer screening over a decade: results of cancer screening from the 1987, 1992 and 1998 National Health Interview Surveys. J Natl Cancer Inst. 2001;93:1704-1713.
FREE FULL TEXT
27. Otori K, Sugiyama K, Hasebe T. High frequency of K-ras mutations in human colorectal hyperplastic polyps. Gut. 1997;40:691-692.
FREE FULL TEXT
28. Fraser GM, Niv Y. Hyperplastic polyp and colonic neoplasia: is there an association? J Clin Gastroenterol. 1993;16:278-280.
PUBMED
29. Martinez ME, McPherson RS, Levin B, Glober GA. A case-control study of dietary intake and other lifestyle risk factors for hyperplastic polyps. Gastroenterology. 1997;113:423-429.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
30. Bensen SP, Cole BF, Mott LA, et al. Colorectal hyperplastic polyps and risk of recurrence of adenomas and hyperplastic polyps. Lancet. 1999;354:1873-1874.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
31. Weitzman ER, Zapka J, Estabrook B, Valentine Goins K. Risk and reluctance: understanding impediments to colorectal cancer screening. Prev Med. 2001;32:502-513.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati  Twitter
What's this?
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
Colorectal cancer (CRC) screening using sigmoidoscopy followed by colonoscopy: a feasibility and efficacy study on a cancer institute based population
Bleiberg et al.
Ann Oncol 2006;17:1328-1332.
ABSTRACT
| FULL TEXT
Project PREVENT: A Randomized Trial to Reduce Multiple Behavioral Risk Factors for Colon Cancer
Emmons et al.
Cancer Epidemiol. Biomarkers Prev. 2005;14:1453-1459.
ABSTRACT
| FULL TEXT
Surveillance Patterns and Polyp Recurrence following Diagnosis and Excision of Colorectal Polyps in a Medicare Population
Amonkar et al.
Cancer Epidemiol. Biomarkers Prev. 2005;14:417-421.
ABSTRACT
| FULL TEXT
|
