Secondary Logo

Journal Logo

Original Articles: Gastroenterology

Long-term cumulative incidence of metachronous advanced colorectal neoplasia after colonoscopy and a novel risk factor: a cohort study

Omata, Fumioa; Deshpande, Gautam A.b; Suzuki, Hidekazuc; Hayashi, Kuniyoshid; Ishii, Naokie; Matoba, Koheia; Ohmuro, Akemia; Rai, Fumiea; Takashima, Misakoa; Fukuda, Katsuyukia; Masuda, Katsunoria; Kumakura, Yasuhisaa

Author Information
European Journal of Gastroenterology & Hepatology: November 2021 - Volume 33 - Issue 11 - p 1341-1347
doi: 10.1097/MEG.0000000000002259
  • Free



Screening colonoscopy with polypectomy of colorectal adenoma (CRA) has been reported to reduce colorectal cancer (CRC) mortality [1,2]. The effectiveness of surveillance colonoscopy for improving CRC mortality, however, remains controversial [3]; guidelines regarding surveillance intervals for patients having undergone CRA removal [4–8] were updated after analysis of the data from the United States National Polyp Study [9].

Previous systematic reviews and meta-analyses reported a significant increase of advanced adenoma [10] and marginally higher risk of CRC, but no significant risk of CRC mortality [11] in people with low-risk adenoma (LRA), defined as 1–2 small (<10 mm) tubular adenomas. However, in the meta-analysis of advanced adenoma [10], pooled estimates were calculated using studies with relatively short follow-up periods.

Reliable data on the long-term cumulative incidence of metachronous advanced adenomas (≧10 mm, or with villous or high-grade dysplasia) is critical for optimizing surveillance colonoscopy strategies. Long-term follow-up studies focusing on metachronous advanced adenoma, especially in patients with LRA, are needed.

Metachronous advanced colorectal neoplasia (ACRN), including both advanced adenoma and CRC, is associated with CRC risk [12,13] and can be used as a surrogate marker for metachronous CRC. The cumulative incidence of ACRN can be useful information for refining surveillance colonoscopy intervals in order to prevent CRC mortality at the individual level.

In order to optimize surveillance strategies, we should also consider relevant risk factors for metachronous ACRN, which likely raise absolute risk for poor outcomes. Risk factors including age, obesity [14], smoking [15] and colorectal neoplasia (CRN) (both CRA and CRC) at index colonoscopy [16–18] have been previously reported. Recently, fecal immunochemical test (FIT) has been used to screen for CRC [19], though many patients with positive screening FIT are subsequently found to have no CRN on diagnostic colonoscopy. Nonetheless, we hypothesize that positive FIT with consecutive negative colonoscopy may be a risk factor for metachronous CRN.

The aim of our study was to determine cumulative incidence of metachronous ACRN over a 10-year period and evaluate risk factors, including FIT positivity, in a large screening population.


Study design

We conducted a cohort study using data from a preventive medicine facility affiliated with a tertiary referral medical center in Tokyo, Japan. The study population consisted of 6720 patients who had undergone an opportunistic general health check-up, including FIT, with an initial screening or diagnostic colonoscopy within the subsequent 180 days after general health check-up. All patients underwent colonoscopy at least twice. A total of 6720 patients received 21 275 colonoscopies between January 2004 and January 2019.

Patients diagnosed with CRC at the initial colonoscopy and those with a history of colectomy, incomplete colonoscopy and Crohn’s disease, or ulcerative colitis were excluded as were patients lacking information about the smoking status or alcohol consumption or FIT (Fig. 1). The number of patients meeting the above inclusion criteria during the study period determined the sample size.

Fig. 1.
Fig. 1.:
Study flow diagram. A total of 6720 patients underwent colonoscopy at least twice between January 2004 and January 2019. Patients with only one colonoscopy; missing information of smoking history or alcohol drinking or fecal immunochemical test data; incomplete colonoscopy; and history of inflammatory bowel disease, or partial colectomy, or colorectal cancer were excluded from the analysis.

Colonoscopy findings within 360 days after the diagnosis of CRN were disregarded, as those colonoscopies were assumed to have been therapeutic procedures. Additionally, two cases of postcolonoscopy CRC which were diagnosed with CRC within 360 days after index colonoscopy were excluded from the study population, as these were considered missed CRC at index colonoscopy.

Treatment of CRA and follow-up colonoscopy was performed per usual care protocols at the discretion of the provider, especially prior to 2015, when surveillance colonoscopy guidelines [6] were first published in Japan, as these guidelines did not suggest a risk stratification strategy based on CRA characteristics at initial colonoscopy. Any CRA should have been removed during follow-up period by biopsy, standard polypectomy, endoscopic mucosal resection or submucosal dissection.

The main outcome was metachronous ACRN, including both advanced adenomas and CRC. The diagnosis of advanced adenoma and CRC was confirmed by pathological examination of biopsy specimens obtained during colonoscopy.

Candidate risk factors included age, sex, BMI, CRC in a first-degree relative (FDR), more than two FDRs with CRC, at least one FDR with CRC under 50 years old, current smoking, alcohol drinking, FIT positivity and CRA with characteristics of either LRA or high-risk adenoma (HRA) at index colonoscopy. Following definitions from US guidelines [5], we defined LRA as 1–2 tubular adenomas <10 mm in size, and HRA as ≥3 tubular adenomas of any size or at least one advanced adenoma. Smoking history (never smoker, ex-smoker and current smoker) and alcohol drinking history (never, social and regular drinker) were obtained via dedicated nurse interview during the general health check-up. In statistical analysis, age and BMI were treated as continuous variables; smoking and alcohol drinking status were dichotomized as current smoking or not, and never drinker or not, respectively; FIT ≥150 ng/mL was defined as positive following commonly used cut-off value in screening institutions in Japan.

Patients were prepared for colonoscopy using 1.8 L of 3.7% magnesium citrate or 2 L polyethylene glycol solution on the morning of the procedure. Colonoscopies were performed in the afternoon using Olympus CF-Q, CF-H, PCF or PCF-H series instruments. All colonoscopies were performed by gastroenterologists certified by The Japanese Gastroenterological Endoscopy Society (JGES) or fellows supervised by a Society board-certified endoscopist. FIT was conducted with OC Sensor (Eiken Chemical Company, Tokyo, Japan).

This study was approved by the internal review board at St. Luke’s International University (study identification number 17-R091). Individual informed consent was waived via opt-out method per announcements of the study on our institutional web page.

Statistical analysis

Differences between continuous variables in three groups were tested by one-way analysis of variance or Wilcoxon rank sum test; the differences of proportion were tested with chi-square test.

Time-to-event analyses were conducted by Kaplan–Meier estimates with log-rank test and univariate and multivariate Cox-proportional hazard regression with time zero for the index colonoscopy. Nelson–Aalen estimates were used to confirm constant hazard over time, and Cox-proportional hazard assumption was checked by Shoenfeld residuals. Multicollinearity among independent variables was ruled out by variable inflation factor. In the cohort without CRA at initial colonoscopy, FIT positivity was evaluated as a predictor of metachronous ACRN.

Statistical inferences were expressed with two-sided 95% confidence intervals (CI). Two-sided P values less than 0.05 were considered to be statistically significant. All analyses were conducted using JMP 14 software (SAS Institute Inc, Cary, North Carolina, USA), Stata 16.1 software (Stata Corp, College Station, Texas, USA) or R (Windows version 3.4.1, R Foundation for Statistical Computing, Vienna, Austria).


The mean age (SD) of the 6720 patients was 54 (11) years; 4443 (66%) were men.

Table 1 presents the patients’ characteristics by CRA type. Analysis of one-way variance demonstrated that the means of age and BMI differed by CRA type. By chi-square test, the proportions of men, current smokers and alcohol drinkers were associated with CRA type, though CRC in FDR, more than two FDRs with CRC, or at least one FDR with CRC under 50 years old was not. By Wilcoxon rank sum test, median of follow-up days differed by CRA type. The prevalence of CRA was 43 and 62.5% in those with negative and positive FIT, respectively. Cecal intubation rate was 99%.

Table 1. - Patients’ characteristics (n = 6720)
No CRA (n = 3423) LRA (n = 2067) HRA (n = 1230) P value
Age, mean (SD) 52 (11) 56 (11) 58 (11) <0.001a
Men, n (%) 2172 (63) 1419 (69) 852 (69) <0.001b
BMI, mean (SD) 23 (3.2) 23 (3.3) 23 (3.5) <0.001c
CRC in FDRs, n (%) 385 (11) 221 (11) 140 (11) 0.77b
More than two FDRs with CRC, n (%) 25 (0.73) 14 (0.68) 7 (0.57) 0.84b
At least one FDR with CRC under 50 years old, n (%) 30 (0.88) 19 (0.92) 9 (0.73) 0.85b
Current smoker, n (%) 428 (13) 344 (17) 282 (23) <0.001b
Alcohol Drinker, n (%) 2210 (65) 1427 (69) 858 (70) 0.001b
FIT≧150 ng/mL, n (%) 739 (22) 631 (31) 601 (49) <0.001b
Follow-up days, median (range) 1779 (365–3650) 1555 (365–3650) 1570 (365–3650) <0.001d
CRA, colorectal adenoma; CRC, colorectal cancer; FDRs, first-degree relatives; FIT, fecal immunochemical test; HRA, high-risk adenoma; LRA, low-risk adenoma.
aAnalysis of one-way variance (ANOVA) with equal variance;
bChi-square test;
cANOVA with unequal variance (Welch’s F test);
dWilcoxon rank sum test.

The Kaplan–Meier estimates of metachronous ACRN are illustrated in Fig. 2. For patients who had no initial CRA, the cumulative incidence [95% CI] of ACRN at 3, 5, 7 and 10 years, respectively, was 1.8% [1.4–2.3], 3.4% [2.7–4.2], 4.8% [3.9–5.8] and 7% [5.7–8.6]. In patients with LRA, the corresponding values were 3.2% [2.4–4.1], 5.7% [4.6–7.1], 8% [6.5–9.7] and 11% [8.9–14]. In patients with HRA, the values were 6.4% [5.1–8.1], 10% [8.6–13], 13% [11–16] and 17% [14–21]. Cumulative incidence of metachronous ACRN differed significantly by CRA type (P < 0.001, log-rank test). Nelson–Aalen estimates showed constant hazard over time.

Fig. 2.
Fig. 2.:
Kaplan–Meier estimates of metachronous advanced colorectal neoplasia. Kaplan–Meier estimates of metachronous advanced colorectal neoplasia were depicted over a 10-year horizon with number at risk. The cohort was divided into three groups by colorectal adenoma (CRA) related findings at initial colonoscopy: no CRA; low-risk adenoma (LRA) defined as 1–2 tubular adenoma ≦10 mm; high-risk adenoma (HRA) defined as ≧3 tubular adenomas of any size or at least one advanced adenoma (tubular adenoma ≧10 mm, high-grade dysplasia or villous adenoma). Cumulative incidence differed between groups (P < 0.001, log-rank test).

Crude and adjusted hazard ratios (HR) [95% CI] of age, men, BMI, CRC in an FDR, more than two FDRs with CRC, at least one FDR with CRC under 50 years old, current smoking, alcohol consumption, LRA (vs. no CRA) and HRA (vs. no CRA) and positive FIT are presented in Table 2. For LRA, sensitivity analysis showed an HR [95% CI] of 1.37 [0.93–2.02] (P = 0.11) at 3 years in contrast to 1.34 [1.04–1.74] (P = 0.027) at 10 years (Table 3). Variance inflation factor of all candidate risk factors was less than 1.5.

Table 2. - Cox-proportional hazard regression for metachronous advanced colorectal neoplasia after 6720 index colonoscopies (10 years’ follow-up)
Crude HR 95% CI P value Adjusted HR 95% CI P value
Age 1.04 1.03–1.05 <0.001 1.04 1.03–1.05 <0.001
Men 1.28 1.02–1.61 0.034 1.06 0.81–1.39 0.68
BMI 1.03 1.00–1.06 0.049 1.02 0.98–1.05 0.29
CRC in FDRs 1.05 0.76–1.45 0.78
More than two FDRs with CRC 1.81 0.68–4.86 0.28
At least one FDR with CRC under 50 years-old 0.69 0.17–2.77 0.58
Current smoking 1.59 1.24–2.04 <0.001 1.55 1.2–2.02 0.001
Drinking alcohol 1.21 0.96–1.52 0.1 1.2 0.93–1.54 0.16
Initial CRA
 LRA (vs. no CRA) 1.68 1.3–2.17 <0.001 1.34 1.04–1.74 0.027
 HRA (vs. no CRA) 3 2.33–3.87 <0.001 1.94 1.48–2.55 <0.001
FIT≧150 ng/mL 1.89 1.53–2.33 <0.001 1.69 1.35–2.1 <0.001
P value was calculated by likelihood ratio test.
CI, confidence interval; CRA, colorectal adenoma; CRC, colorectal cancer; FDRs, first degree relatives; FIT, fecal immunochemical test; HR, hazard ratio; HRA, high-risk adenoma; LRA, low-risk adenoma.

Table 3. - Cox-proportional hazard regression for metachronous advanced colorectal neoplasia after 6720 index colonoscopies (3 years’ follow-up)
Crude HR 95% CI P value Adjusted HR 95% CI P value
Age 1.05 1.03–1.06 <0.001 1.04 1.03–1.06 <0.001
Men 1.38 0.98–1.92 0.056 1.12 0.7–1.65 0.58
BMI 1.02 0.98–1.07 0.37 1 0.95–1.05 0.94
CRC in FDRs 1.34 0.88–2.05 0.19 1.4 0.91–2.14 0.14
Current smoking 1.84 1.31–2.59 <0.001 1.78 1.24–2.55 0.003
Drinking alcohol 1.46 1.04–2.06 0.025 1.44 0.99–2.08 0.052
Initial CRA
 LRA (vs. no CRA) 1.8 1.23–2.63 0.003 1.37 0.93–2.02 0.11
 HRA (vs. no CRA) 3.79 2.64–5.43 <0.001 2.22 1.51–3.27 <0.001
FIT≧150 ng/mL 2.2 1.63–2.96 <0.001 1.91 1.4–2.6 <0.001
P value was calculated by likelihood ratio test.
CI, confidence interval; CRA, colorectal adenoma; CRC, colorectal cancer; FDRs, first degree relatives; FIT, fecal immunochemical test; HR, hazard ratio; HRA, high-risk adenoma; LRA, low-risk adenoma.

In a subgroup analyses of the patients with normal initial colonoscopy, a positive FIT was a significant predictor of metachronous ACRN (P = 0.003, log rank test) (Fig. 3). The adjusted HR [95% CI] of positive FIT was 1.88 [1.29–2.74] (Table 4).

Table 4. - Cox-proportional hazard regression for metachronous advanced colorectal neoplasia (10 years’ follow-up) Subgroup analysis of the patients with positive fecal immunochemical test in whom colorectal adenoma was ruled out by colonoscopy (n = 3423)
Crude HR 95% CI P value Adjusted HR 95% CI P value
Age 1.05 1.03–1.07 <0.001 1.05 1.03–1.07 <0.001
Men 1.18 0.81–1.72 0.39
BMI 1.02 0.96–1.08 0.54
CRC in FDRs 1.29 0.77–2.15 0.35
Current smoking 1.07 0.63–1.8 0.81
Drinking alcohol 0.96 0.66–1.38 0.81
FIT≧150 ng/mL 1.76 1.21–2.57 0.005 1.88 1.29–2.74 <0.001
P value was calculated by likelihood ratio test.
CI, confidence interval; CRC, colorectal cancer; FDRs, first degree relatives; FIT, fecal immunochemical test; HR, hazard ratio.

Fig. 3.
Fig. 3.:
Kaplan–Meier estimates of metachronous advanced colorectal neoplasia in subgroup of the patients with positive fecal immunochemical test in whom colorectal adenoma was ruled out by colonoscopy. Kaplan–Meier estimates of metachronous advanced colorectal neoplasia (ACRN) are shown by fecal immunochemical test (FIT) status. FIT was performed as part of general health check-up before colonoscopy. The interval between FIT and initial colonoscopy was less than 180 days. Patients with positive FIT were more likely to develop metachronous ACRN (P = 0.003, log rank test).


Our study confirmed that the CRA at index colonoscopy is a significant predictor for metachronous ACRN over long-term follow-up. In contrast to a previous study [20] in a Western population that identified only HRA as a risk factor, LRA was also significantly associated with metachronous ACRN at 10-year follow-up in our population. Our study also showed that current smoking and positive FIT were significant predictors for metachronous ACRN. In addition, subgroup analysis showed that positive FIT was a significant predictor of metachronous ACRN, despite a normal index colonoscopy.

We chose ACRN as the primary outcome in our study given that cumulative incidence of metachronous ACRN is an important factor to consider when deciding upon a surveillance interval at the individual level. While CRC incidence is also undoubtedly an important outcome, our study population was not large enough to estimate this statistical inference with precision.

Several observational studies, systematic reviews and meta-analyses [10,21] have suggested an association between baseline colonoscopy findings and metachronous ACRN. Heisser et al. [22] reported a 10-year pooled probability [95% CI] of advanced adenoma of 7% [5.3–8.7] with little heterogeneity in patients without initial CRN. That result aligns closely with the 6.9% [5.6–8.4] that we identified, despite differences in the outcomes used. Based on these results, we believe that a 10-year interval after normal colonoscopy is appropriate for this cohort.

Dube et al. [10] reported a pooled 5-year cumulative incidence [95% CI] of advanced adenoma of 4.9% [3.18–6.97], a 5-year risk ratio [95% CI] of 1.64 [1.24–2.17], and a risk ratio of 1.41 [0.97–2.05] after 5 years in patients with LRA; cumulative incidences of advanced adenoma after 5 years were not reported. Lieberman et al. [20], using 10-year follow-up surveillance colonoscopy data from 1950 Veterans Affairs patients, confirmed predictive validity of baseline screening colonoscopy, finding that HRA was a significant risk factor for metachronous ACRN, while LRA was not. However, as most of the population in their study was male, applicability to the general population is limited.

Regarding CRC incidence in patients with LRA history, two studies [23,24] reported no risk of initial LRA removal. However, a significant risk (odds ratio [OR], 1.26 [95% CI, 1.06–1.51]) was reported in a recent systematic review and meta-analysis by Duvvuri et al. [11]. Our study showed that the cumulative incidence [95% CI] of metachronous ACRN was 5.7% [4.6–7.1] at 5 years and 11% [8.9–14] at 10 years in those patients with initial LRA, with an HR [95% CI] of 1.35 [1.04–1.76] over 10 years. The HR of LRA for metachronous ACRN was statistically significant, compatible with Duvvuri’s findings [11].

European guidelines [8] suggest treating patients with 1–4 tubular adenomas <10 mm as non-high-risk, recommending no colonoscopic surveillance and only invited bowel screening. In contrast, US guidelines [7] consider 1–2 tubular adenomas <10 mm as LRA and recommend surveillance colonoscopy in 5–7 years. Based on our results, we consider US guidelines more appropriate than European guidelines, given a relatively substantial absolute risk [95% CI] of 11% [8.9–14] at 10 years in patients with LRA.

Jacobs et al. [25] suggested a significant association between obesity and metachronous adenoma, but nonsignificant association between obesity and advanced adenoma. Our study did not suggest any statistically significant association between BMI increase and metachronous ACRN, either. Regarding the risk for metachronous ACRN associated with CRC in an FDR and its risk for metachronous ACRN, Jacobs et al. [26] reported an OR [95% CI] of 1.15 [0.96–1.37], while Kim et al. [27] reported an HR [95% CI] of 0.75 [0.39–1.47] for metachronous ACRN. Our results did not demonstrate these associations despite a rigorous analysis of both multiple FDRs with CRC and occurrence in an FDR under 50 years old. The lack of association suggests that the information regarding CRC in FDRs may not be necessary for risk stratification in metachronous ACRN surveillance.

Though several observational studies [28,29] and meta-analyses [30,31] have suggested that both smoking and alcohol consumption are risk factors for CRN, few studies have investigated an association between these habits and metachronous CRN. Jacobson et al. [15] conducted a case-control study, reporting an OR for metachronous or recurrent CRA [95% CI] of 1.8 [1–3.4] in men and 3.6 [1.7–7.6] in women who smoke. We could not find any previous studies using time-to-event analysis. Our study showed that the HR [95% CI] of current smoking and alcohol drinking was 1.55 [1.5–2.02] and 1.2 [0.93–1.54], respectively. Although our study did not consider changes in smoking or drinking status during the follow-up period, it is reasonable to consider these as ongoing candidate risk factors for metachronous ACRN on the assumption that changes in smoking and drinking status are rare and difficult to achieve even with aggressive intervention.

Few studies have taken into account FIT status. Our study showed that positive FIT was an independent risk factor for metachronous ACRN. The concern for possible collinearity between positive FIT and initial CRA was excluded by calculating variance inflation factor. In addition, positive FIT was a risk factor for metachronous ACRN even in those in whom CRA was ruled out by colonoscopy (Table 3). In real practice, clinicians often encounter patients with positive FIT in whom subsequent diagnostic colonoscopy reveals no CRN. The absolute risk of metachronous ACRN in these patients is shown in Fig. 3. This information has important clinical implications, especially as FIT becomes an increasingly popular CRC screening tool for CRC worldwide.

Ours is the first study to show that positive FIT is a significant predictor of metachronous ACRN despite a normal colonoscopy. As 17 % of colon polyps are reported to be missed per one colonoscopy [32], positive FIT may be a useful surrogate marker for occult CRN which could be missed by colonoscopy.

We can not disregard quality metrics of colonoscopy [33]. We assume that the quality of colonoscopy was adequate in our study, given a cecal intubation rate of 99% and a CRA prevalence (adenoma detection rate) of 43% among FIT negative participants.

In a cost-effective analysis using Markov models, Sekiguchi et al. [34] showed that risk-stratified colonoscopy surveillance with shorter intervals was more cost-effective than surveillance without risk stratification. Specifically, they reported that 3-years surveillance intervals after LRA polypectomy were more cost-effective than 10-year surveillance intervals. Our results suggesting that LRA was a significant risk for metachronous ACRN corroborate these previous findings.

Our study has some limitations. First, we disregarded serrated lesions as a candidate risk factor, as the definition was introduced by the WHO only in 2010 [35]. Second, we did not consider longitudinal status of CRC in an FDR, BMI, smoking or drinking in the follow-up period, potentially biasing the demonstrated HR for these factors. Third, patients may have had interventions during follow-up colonoscopies such as biopsy, polypectomy, endoscopic mucosal resection, or endoscopic submucosal dissection, which we were unable to adjust for in our study. As such, the cumulative incidence of metachronous ACRN might be underestimated. Fourth, lead-time bias is inevitable in our study as those patients with a history of CRA tend to have surveillance colonoscopy more frequently, leading to an overestimation of metachronous ACRN in CRA group.

Nevertheless, our study has several strengths. First, given the large study population and sufficient number of patients remaining at risk at 10 years after initial colonoscopy, we could make precise statistical inferences regarding long-term absolute risk particularly of ACRN; in patients with LRA at initial colonoscopy, these metrics have not been previously reported in a general population. Second, we were able to account for not only baseline colonoscopy findings, but also other salient risk factors including family CRC history, BMI, smoking, alcohol drinking history and FIT status.

In conclusion, our study showed that colorectal adenoma at the baseline colonoscopy is a valid predictor for metachronous advanced colorectal neoplasia over a 10-year period. Colorectal adenomas at initial colonoscopy (both LRA and HRA), current smoking and positive FIT were independent risk factors for metachronous advanced colorectal neoplasia. In patients with either LRA or HRA at index colonoscopy, the absolute risk of metachronous advanced colorectal neoplasia at 10 years remained substantially high. Interestingly, positive FIT was a significant risk factor for metachronous advanced colorectal neoplasia despite normal baseline colonoscopy, suggesting that patients with these findings warrant surveillance earlier than 10 years, similar to those with initial LRA. Future surveillance guideline should take into account of not only baseline colonoscopy findings but also smoking status and FIT results when available.


We thank Ms. Eri Yoshino for her excellent work of obtaining critical data from endoscopy and pathological reports, Ms. Naomi Masui for dedicated data preparation, and Ms. Reiko Nakayama for secretarial work. We thank all endoscopy nurses and technicians for their support of endoscopic examination and Dr. William R. Brown for professional edits and continuing mentorship.

This study was supported by JSPS KAKENHI Grant Number JP17K09094.

O.F. conceived and designed the study, performed data analysis and drafted the manuscript. D.G.A. and S.H. critically revised the manuscript for important intellectual content. H.K. conducted a part of statistical analysis. I.N., M.K., O.A., R.F., T.M., F.K., M.K. and K.Y. interpreted the data in results and approved the manuscript.

Conflicts of interest

There are no conflicts of interest.


1. Zauber AG, Winawer SJ, O’Brien MJ, Lansdorp-Vogelaar I, van Ballegooijen M, Hankey BF, et al. Colonoscopic polypectomy and long-term prevention of colorectal-cancer deaths. N Engl J Med 2012; 366:687–696.
2. Nishihara R, Wu K, Lochhead P, Morikawa T, Liao X, Qian ZR, et al. Long-term colorectal-cancer incidence and mortality after lower endoscopy. N Engl J Med 2013; 369:1095–1105.
3. Doubeni CA, Fedewa SA, Levin TR, Jensen CD, Saia C, Zebrowski AM, et al. Modifiable failures in the colorectal cancer screening process and their association with risk of death. Gastroenterology 2019; 156:63–74.e6.
4. Winawer SJ, Zauber AG, Fletcher RH, Stillman JS, O’Brien MJ, Levin B, et al.; US Multi-Society Task Force on Colorectal Cancer; American Cancer Society. Guidelines for colonoscopy surveillance after polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer and the American Cancer Society. Gastroenterology 2006; 130:1872–1885.
5. Lieberman DA, Rex DK, Winawer SJ, Giardiello FM, Johnson DA, Levin TR. Guidelines for colonoscopy surveillance after screening and polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer. Gastroenterology 2012; 143:844–857.
6. Tanaka S, Saitoh Y, Matsuda T, Igarashi M, Matsumoto T, Iwao Y, et al.; Japanese Society of Gastroenterology. Evidence-based clinical practice guidelines for management of colorectal polyps. J Gastroenterol 2015; 50:252–260.
7. Gupta S, Lieberman D, Anderson JC, Burke CA, Dominitz JA, Kaltenbach T, et al. Recommendations for follow-up after colonoscopy and polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer. Gastrointest Endosc 2020; 91:463–485.e5.
8. Rutter MD, East J, Rees CJ, Cripps N, Docherty J, Dolwani S, et al. British Society of Gastroenterology/Association of Coloproctology of Great Britain and Ireland/Public Health England post-polypectomy and post-colorectal cancer resection surveillance guidelines. Gut 2020; 69:201–223.
9. Winawer SJ, Zauber AG, O’Brien MJ, Ho MN, Gottlieb L, Sternberg SS, et al. Randomized comparison of surveillance intervals after colonoscopic removal of newly diagnosed adenomatous polyps. The National Polyp Study Workgroup. N Engl J Med 1993; 328:901–906.
10. Dubé C, Yakubu M, McCurdy BR, Lischka A, Koné A, Walker MJ, et al. Risk of advanced adenoma, colorectal cancer, and colorectal cancer mortality in people with low-risk adenomas at baseline colonoscopy: a systematic review and meta-analysis. Am J Gastroenterol 2017; 112:1790–1801.
11. Duvvuri A, Chandrasekar VT, Srinivasan S, Narimiti A, Dasari C, Nutalapati V, et al. Risk of colorectal cancer and cancer related mortality after detection of low-risk or high-risk adenomas, compared with no adenoma, at index colonoscopy: a systematic review and meta-analysis. Gastroenterology 2021; 160:1986–1996.e3.
12. 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.
13. Otchy DP, Ransohoff DF, Wolff BG, Weaver A, Ilstrup D, Carlson H, Rademacher D. Metachronous colon cancer in persons who have had a large adenomatous polyp. Am J Gastroenterol 1996; 91:448–454.
14. Kim NH, Jung YS, Park JH, Park DI, Sohn CI. Impact of obesity and metabolic abnormalities on the risk of metachronous colorectal neoplasia after polypectomy in men. J Gastroenterol Hepatol 2019; 34:1504–1510.
15. Jacobson JS, Neugut AI, Murray T, Garbowski GC, Forde KA, Treat MR, et al. Cigarette smoking and other behavioral risk factors for recurrence of colorectal adenomatous polyps (New York City, NY, USA). Cancer Causes Control 1994; 5:215–220.
16. Yamaji Y, Mitsushima T, Ikuma H, Watabe H, Okamoto M, Kawabe T, et al. Incidence and recurrence rates of colorectal adenomas estimated by annually repeated colonoscopies on asymptomatic Japanese. Gut 2004; 53:568–572.
17. van Heijningen EM, Lansdorp-Vogelaar I, Kuipers EJ, Dekker E, Lesterhuis W, Ter Borg F, et al. Features of adenoma and colonoscopy associated with recurrent colorectal neoplasia based on a large community-based study. Gastroenterology 2013; 144:1410–1418.
18. Shono T, Oyama S, Oda Y, Yokomine K, Murakami Y, Miyamoto H, et al.; Kumamoto Colon Cancer Study Group. Risk stratification of advanced colorectal neoplasia after baseline colonoscopy: Cohort study of 17 Japanese community practices. Dig Endosc 2020; 32:106–113.
19. Rex DK, Boland CR, Dominitz JA, Giardiello FM, Johnson DA, Kaltenbach T, et al. Colorectal cancer screening: recommendations for physicians and patients from the U.S. Multi-Society Task Force on colorectal cancer. Gastroenterology 2017; 153:307–323.
20. Lieberman D, Sullivan BA, Hauser ER, Qin X, Musselwhite LW, O’Leary MC, et al. Baseline colonoscopy findings associated with 10-year outcomes in a screening cohort undergoing colonoscopy surveillance. Gastroenterology 2020; 158:862–874 e8.
21. Hassan C, Gimeno-García A, Kalager M, Spada C, Zullo A, Costamagna G, et al. Systematic review with meta-analysis: the incidence of advanced neoplasia after polypectomy in patients with and without low-risk adenomas. Aliment Pharmacol Ther 2014; 39:905–912.
22. Heisser T, Peng L, Weigl K, Hoffmeister M, Brenner H. Outcomes at follow-up of negative colonoscopy in average risk population: systematic review and meta-analysis. BMJ 2019; 367:l6109.
23. Click B, Pinsky PF, Hickey T, Doroudi M, Schoen RE. Association of colonoscopy adenoma findings with long-term colorectal cancer incidence. JAMA 2018; 319:2021–2031.
24. Lee JK, Jensen CD, Levin TR, Doubeni CA, Zauber AG, Chubak J, et al. Long-term risk of colorectal cancer and related death after adenoma removal in a large, community-based population. Gastroenterology 2020; 158:884–894.e5.
25. Jacobs ET, Ahnen DJ, Ashbeck EL, Baron JA, Greenberg ER, Lance P, et al. Association between body mass index and colorectal neoplasia at follow-up colonoscopy: a pooling study. Am J Epidemiol 2009; 169:657–666.
26. Jacobs ET, Gupta S, Baron JA, Cross AJ, Lieberman DA, Murphy G, Martínez ME. Family history of colorectal cancer in first-degree relatives and metachronous colorectal adenoma. Am J Gastroenterol 2018; 113:899–905.
27. Kim NH, Jung YS, Park JH, Park DI, Sohn CI. Association between family history of colorectal cancer and the risk of metachronous colorectal neoplasia following polypectomy in patients aged <50 years. J Gastroenterol Hepatol 2019; 34:383–389.
28. Lieberman DA, Prindiville S, Weiss DG, Willett W; VA Cooperative Study Group 380. Risk factors for advanced colonic neoplasia and hyperplastic polyps in asymptomatic individuals. JAMA 2003; 290:2959–2967.
29. Omata F, Brown WR, Tokuda Y, Takahashi O, Fukui T, Ueno F, Mine T. Modifiable risk factors for colorectal neoplasms and hyperplastic polyps. Intern Med 2009; 48:123–128.
30. Zhu JZ, Wang YM, Zhou QY, Zhu KF, Yu CH, Li YM. Systematic review with meta-analysis: alcohol consumption and the risk of colorectal adenoma. Aliment Pharmacol Ther 2014; 40:325–337.
31. McNabb S, Harrison TA, Albanes D, Berndt SI, Brenner H, Caan BJ, et al. Meta-analysis of 16 studies of the association of alcohol with colorectal cancer. Int J Cancer 2020; 146:861–873.
32. Ahn SB, Han DS, Bae JH, Byun TJ, Kim JP, Eun CS. The miss rate for colorectal adenoma determined by quality-adjusted, back-to-back colonoscopies. Gut Liver 2012; 6:64–70.
33. Rex DK, Schoenfeld PS, Cohen J, Pike IM, Adler DG, Fennerty MB, et al. Quality indicators for colonoscopy. Gastrointest Endosc 2015; 81:31–53.
34. Sekiguchi M, Igarashi A, Sakamoto T, Saito Y, Esaki M, Matsuda T. Cost-effectiveness analysis of postpolypectomy colonoscopy surveillance using Japanese data. Dig Endosc 2019; 31:40–50.
35. Snover D, Ahnen DJ, Burt RW, Odze RD. Serrated polyps of the colon and rectum and serrated polyposis. Bozman FT, Carneiro F, Hruban RH, et al., editors. In: WHO Classification of Tumours of Digestive System. 4 ed. Geneva, Switzerland: WHO PRESS; 2010.

alcohol; colonoscopy; colorectal neoplasia; colorectal adenoma; colorectal cancer; family history; fecal immunochemical test; metachronous; smoking; surveillance

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.