Abstract
Background
Inflammatory bowel disease [IBD] has been linked to an increased risk of colorectal neoplasia. However, the types and risks of specific polyp types in IBD are less clear.
Methods
We identified 41 880 individuals with IBD (Crohn’s disease [CD: n = 12 850]; ulcerative colitis [UC]: n = 29 030]) from Sweden matched with 41 880 reference individuals. Using Cox regression, we calculated adjusted hazard ratios [aHRs] for neoplastic colorectal polyps [tubular, serrated/sessile, advanced and villous] defined by histopathology codes.
Results
During follow-up, 1648 [3.9%] IBD patients and 1143 [2.7%] reference individuals had an incident neoplastic colorectal polyp, corresponding to an incidence rate of 46.1 and 34.2 per 10 000 person-years, respectively. This correlated to an aHR of 1.23 (95% confidence interval [CI] 1.12–1.35) with the highest HRs seen for sessile serrated polyps [8.50, 95% CI 1.10–65.90] and traditional serrated adenomas [1.72, 95% CI 1.02–2.91]. aHRs for colorectal polyps were particularly elevated in those diagnosed with IBD at a young age and at 10 years after diagnosis. Both absolute and relative risks of colorectal polyps were higher in UC than in CD [aHRs 1.31 vs 1.06, respectively], with a 20-year cumulative risk difference of 4.4% in UC and 1.5% in CD, corresponding to one extra polyp in 23 patients with UC and one in 67 CD patients during the first 20 years after IBD diagnosis.
Conclusions
In this nationwide population-based study, there was an increased risk of neoplastic colorectal polyps in IBD patients. Colonoscopic surveillance in IBD appears important, especially in UC and after 10 years of disease.
Keywords: Crohn’s disease, ulcerative colitis, polyp, dysplasia, cancer
1. Introduction
Inflammatory bowel diseases [IBD], comprising Crohn’s disease [CD] and ulcerative colitis [UC], are chronic inflammatory conditions of the gastrointestinal tract characterized by aberrant immune responses to an altered gut microbiome in genetically susceptible individuals. Individuals with IBD are at an increased risk of developing colorectal neoplasia (colorectal dysplasia and colorectal cancer [CRC]) due to long-standing chronic colonic inflammation.1 The incidence of CRC in IBD is lower today than in previous decades, probably due to more effective and safer therapies to control inflammation, and improved colonoscopic technologies and techniques.2,3 However, colorectal neoplasia remains a significant source of IBD-related morbidity and mortality.4,5
Many disease- and patient-specific factors are closely linked to colorectal neoplasia risk, including disease duration, extent, degree of inflammation, concomitant primary sclerosing cholangitis [PSC] and prior history of colorectal neoplasia.6 Colorectal neoplasia in patients with IBD was previously often thought to be invisible and non-adenomatous. Recent data, however, have confirmed that the vast majority of neoplasia, especially in the era of high-definition white light endoscopy, is polypoid and visible.7 There are few data characterizing the incidence and risk factors for specific types of colon polyp histological architecture in IBD. In a retrospective analysis of a histopathology database, 302 polyps were detected in 131 patients with IBD undergoing 178 colonoscopies. The histology was tubular adenoma in 76 [25%], tubulovillous adenoma in 14 [5%], hyperplastic in 112 [37%], post-inflammatory in 32 [11%], sessile serrated polyp in 31 [10%], traditional serrated adenoma in two [0.7%], flat high-grade dysplasia or cancer in two [0.7%] and other in 33 [11%].8 Previous studies were limited in their sample size, years of follow-up and assessment of histological subtypes, and lacked adjustment for critical IBD covariates.8–10
According to SCENIC consensus,11 the distinction between IBD-associated polypoid or visible dysplasia and sporadic adenomas is not important as endoscopic resection is typically sufficient to manage visible dysplasia. Furthermore, while current guidelines favour terminology based on endoscopic appearance and evidence of dysplasia, little is known regarding the risk of specific histological colon polyp types in IBD—both in absolute terms and compared to the general population. The primary objective of this study was to characterize and determine types and rates of neoplastic colorectal polyps in patients with IBD, and secondarily to compare these findings to the general non-IBD population who also underwent colonoscopy.
2. Methods
2.1. Setting and data sources
In Sweden, all contacts with the healthcare system are recorded in nationwide patient registries, with data in these registries linked to administrative registries by means of unique personal identity numbers with virtually no loss to follow-up.12
2.2. Study population
We identified patients in Sweden with a diagnosis of IBD using International Classification of Disease [ICD] codes in the Swedish Patient Register, which contains records of inpatient care since 1964 and non-primary outpatient care since 2001, and SNOMED codes in the Epidemiology Strengthened by istopathology Reports in Sweden [ESPRESSO] cohort,13,14 which contains histopathology since 1965 [Supplementary Table S1]. We used validated registry definitions to determine incident cases of IBD.15 We identified patients diagnosed between 1969 and 2016. We excluded several years after the start of the patient registers in 1964 [Patient Register] and 1965 [ESPRESSO] since IBD diagnosed in these years may not represent incident IBD, but rather patients with IBD who had been diagnosed prior but never as inpatients. We matched each CD and UC patient by sex, age, birth year and place of residence to one general population comparator who also had undergone a colonoscopic evaluation within 5 years before to 3 months after the matching date for IBD diagnosis. Reference individuals were required to have a record of colonoscopy to minimize the influence of surveillance in IBD. Matched reference individuals had to be alive and free of IBD at the start of follow-up of the index patients with IBD. They stopped contributing reference person-time if they were diagnosed with IBD [i.e. first ever National Patient Register diagnostic listing/SnoMed code consistent with IBD].16
Ascertainment of IBD subtype [UC, CD] was based on ICD codes and pathology reports at start of follow-up [Supplementary Table S1]. Disease phenotypes [maximum UC extent and CD localization and behaviour] was defined using the Montreal classification, available from 1997 [Supplementary Table S2].17 Ascertainment of histological IBD disease activity was based on our validated model using a dichotomized [high/low] SNOMED-based inflammation histology score.14 We also evaluated the presence of PSC and other extraintestinal manifestations during follow-up, available from 1987 [Supplementary Table S3]. We captured information on colon surgery through validated surgery codes [Supplementary Table S4] and extracted information on CRC history from validated cancer codes [Supplementary Table S5].18 Colonoscopy frequency was defined by procedure codes in the National Patient Register or topography codes from ESPRESSO in both IBD [UC and CD separately] and matched reference individuals [Supplementary Table S6].3
2.3. Colorectal polyps
Our primary outcome was the detection of any neoplastic polyp using histopathology codes [Supplementary Table S7]. Secondary outcomes included adenoma types (tubular, tubulovillous, villous, serrated/sessile serrated, and the presence of high-grade dysplasia [HGD]). Location of polyp [proximal colon, distal colon, rectum] was reported when available using topography codes.
Patients with a history of CRC, neoplastic polyp or colectomy prior to the start of follow-up were excluded. Polyps consisting of submucosal lesions were not included in the current analysis and patients were excluded if there was a history of non-neoplastic polyps.
2.4. Statistical analysis
We calculated the occurrence of the first neoplastic polyp during follow-up, stratified by IBD status and specific polyp type. The incidence of polyps was defined as the number of events per 1000 person-years, and the cumulative incidence was estimated at 10 and 20 years of follow-up. Follow-up started on the date of the second diagnostic record of an ICD code for IBD or histopathology report suggesting IBD to avoid immortal time bias and on the corresponding date for individually matched comparators. Stratification by age and year of IBD onset in the descriptive tables started on the date of the first diagnostic record of IBD or histopathology consistent with IBD. Follow-up ended at first polyp, emigration, death, colectomy, CRC or end of the follow-up period, whichever came first.
We stratified analyses by sex, age of IBD diagnosis [childhood-onset: <18 years, young adulthood-onset: 18 to <40 years, middle aged-onset: 40 to <60 years, and elderly-onset: ≥60 years], year of IBD onset, extent/location of disease, disease behaviour, presence/absence of colon surgery, drug exposures and years of follow-up. For calculations of stratum-specific hazard ratios [HRs] of time-dependent exposures such as PSC and other extraintestinal manifestations, follow-up started on the date of the first register record of PSC or extraintestinal manifestations for IBD patients and on the same date for matched reference individuals.
We used Cox regression, adjusted for age at IBD diagnosis and sex, to calculate HRs with 95% confidence intervals [CIs] as a measure of the relative rates for polyp type compared to a general population comparison cohort.
Sensitivity analyses were performed including polyps occurring on the same date as the start of follow-up and with adjustment for number of prior colonoscopic examinations at baseline. Further analyses were performed removing age as a matching condition, adjusting for family history of CRC in a first-degree relative identified through the Total Population Register, and censoring patients with genetic risk factors for CRC.
SAS software, version 9.4 [SAS Institute], was used in our analyses. We computed CIs by inverting likelihood ratio test statistics and considered a 95% CI that did not include 1 or a p-value <0.05 to be statistically significant.
3. Results
3.1. Study cohorts and IBD characteristics
We identified 69 241 individuals with IBD during the study period with 41 880 [12 850 CD; 28 030 UC] matched 1:1 to reference individuals in the general population by sex, age, birth year and place of residence, who underwent a colonoscopic evaluation but without IBD [Supplementary Figure S1; Table 1]. The male-to-female ratio was ~0.84, and 1.2% of IBD patients [n = 506] were diagnosed before 18 years of age. Median age at first IBD diagnosis was 50.4 years (interquartile range [IQR] 36.4–63.9) for those with a matched comparator and 28.2 years [IQR 18.6-35.8] for those without a matched comparator. Ileal or ileocolonic involvement was seen in 61.9% of all CD patients [n = 7 950] and of patients with UC, 51.9% had extensive colitis or unknown distribution [n = 15 064]. The majority of IBD patients had histological evidence of inflammation with a high SNOMED-based histology score [87.6%, n = 36 683]. PSC was observed in 1.0% of all IBD patients [n = 358]. Over a median follow-up of 7.3 years [IQR 3.4–12.5], patients with IBD provided 357 659 person-years of follow-up and reference individuals provided 333 936 person-years of follow-up. Of patients with IBD, 35.6% [n = 14 952] had >10 years of follow-up and the majority required a subsequent colonoscopic examination [Table 2]. During follow-up, 40.2% [n = 16 840] of patients with IBD had three or more colonoscopies whereas only 4.3% [n = 1810] of reference individuals had three or more colonoscopies.
Table 1.
Baseline characteristics of patients with IBD and reference individuals from the general population matched 1:1 to each patient by sex, age, birth year and place of residence.
| IBD | UC | CD | Reference [IBD overall] | IBD [no matched comparator] | |
|---|---|---|---|---|---|
| Total | 41 880 | 29 030 | 12 850 | 41 880 | 27 361 |
| Sex | |||||
| Female | 22 802 [54.4%] | 15 279 [52.6%] | 7523 [58.5%] | 22 802 [54.4%] | 10 777 [39.4%] |
| Male | 19 078 [45.6%] | 13 751 [47.4%] | 5327 [41.5%] | 19 078 [45.6%] | 16 584 [60.6%] |
| Age at first diagnosis [years] | |||||
| Mean [SD] | 50.4 [17.5] | 51.1 [17.4] | 49.0 [17.6] | 50.5 [17.4] | 28.2 [12.9] |
| Median [IQR] | 50.5 [36.4–63.9] | 51.1 [37.1–64.5] | 49.2 [34.4–62.6] | 50.6 [36.5–64.0] | 25.6 [18.6–35.8] |
| Range, min–max | 1.1–98.2 | 1.4–95.9 | 1.1–98.2 | 1.3–97.0 | 0.4–96.1 |
| <18 | 506 [1.2%] | 295 [1.0%] | 211 [1.6%] | 460 [1.1%] | 6263 [22.9%] |
| 18 to <40 | 12 545 [30.0%] | 8403 [28.9%] | 4142 [32.2%] | 12 508 [29.9%] | 16 147 [59.0%] |
| 40 to <60 | 15 474 [36.9%] | 10 779 [37.1%] | 4695 [36.5%] | 15 481 [37.0%] | 4347 [15.9%] |
| ≥60 | 13 355 [31.9%] | 9553 [32.9%] | 3802 [29.6%] | 13 431 [32.1%] | 604 [2.2%] |
| Year of first IBD diagnosis | |||||
| 1969–1976 | 251 [0.6%] | 170 [0.6%] | 81 [0.6%] | 317 [1.2%] | |
| 1977–1990 | 2784 [6.6%] | 1837 [6.3%] | 947 [7.4%] | 4245 [15.5%] | |
| 1991–2002 | 11 118 [26.5%] | 7909 [27.2%] | 3209 [25.0%] | 11 331 [41.4%] | |
| 2003–2016 | 27 727 [66.2%] | 19 114 [65.8%] | 8613 [67.0%] | 11 468 [41.9%] | |
| Year of IBD index date | |||||
| 1969–1976 | 3 [0.0%] | 2 [0.0%] | 1 [0.0%] | 3 [0.0%] | 114 [0.4%] |
| 1977–1990 | 1677 [4.0%] | 1065 [3.7%] | 612 [4.8%] | 1677 [4.0%] | 3448 [12.6%] |
| 1991–2002 | 11 017 [26.3%] | 7805 [26.9%] | 3212 [25.0%] | 11 017 [26.3%] | 11 662 [42.6%] |
| 2003–2016 | 29 183 [69.7%] | 20 158 [69.4%] | 9 025 [70.2%] | 29 183 [69.7%] | 12 137 [44.4%] |
| Country of birth | |||||
| Nordic | 38 822 [92.7%] | 27 053 [93.2%] | 11 769 [91.6%] | 38 820 [92.7%] | 25 908 [94.7%] |
| Non-Nordic | 3058 [7.3%] | 1977 [6.8%] | 1081 [8.4%] | 3060 [7.3%] | 1452 [5.3%] |
| Missing | 0 | 0 | 0 | 0 | 1 [0.0%] |
| Montreal Classification CD a | |||||
| Number classified since 1998 | 10 521 [81.9%] | ||||
| L1 [terminal ileum]/L3/LX [ileocolonic, not defined] | 7953 [61.9%] | ||||
| L2 [colon] | 2568 [24.4%] | ||||
| B1 [non-stricturing, non-penetrating] | 9410 [89.4%] | ||||
| B2 [stricturing] | 1111 [10.6%] | ||||
| B3 [penetrating] | 63 [0.6%] | ||||
| Perianal | 72 [0.7%] | ||||
| Montreal Classification UC | |||||
| Number classified since 1998 | 24 425 [84.1%] | ||||
| E1 [ulcerative proctitis]/E2 [left sided UC] | 11 546 [39.8%] | ||||
| E3 [extensive UC]/EX [extent not defined] | 15 064 [51.9%] | ||||
| Extraintestinal manifestations * | |||||
| Number classified since 1998 | 34 946 [83.4%] | 24 425 [84.1%] | 10 521 [81.9%] | 18 378 [67.2%] | |
| Primary sclerosing cholangitis [PSC] | 358 [1.0%] | 298 [1.2%] | 60 [0.6%] | 303 [1.6%] | |
| Other extraintestinal manifestations | 2678 [7.7%] | 1640 [6.7%] | 1038 [9.9%] | 996 [5.4%] | |
| History of colorectal cancer in first-degree relative | |||||
| Yes | 290 [0.7%] | 224 [0.8%] | 66 [0.5%] | 328 [0.8%] | 85 [0.3%] |
| Number of previous endoscopies at baseline | |||||
| 0 | 0 | 0 | 0 | 0 | 0 |
| 1 | 8762 [20.9%] | 5313 [18.3%] | 3449 [26.8%] | 22 673 [54.1%] | 7804 [28.5%] |
| 2 | 13 783 [32.9%] | 9384 [32.3%] | 4399 [34.2%] | 13 089 [31.3%] | 9903 [36.2%] |
| ≥3 | 19 335 [46.2%] | 14 333 [49.4%] | 5002 [38.9%] | 6118 [14.6%] | 9654 [35.3%] |
| Inflammatory status at index date | |||||
| Low histology score | 5197 [12.4%] | 3227 [11.1%] | 1970 [15.3%] | 2726 [10.0%] | |
| High histology score | 36 683 [87.6%] | 25 803 [88.9%] | 10 880 [84.7%] | 24 635 [90.0%] | |
| SnoMed normal mucosa codea | 10 077 [24.1%] | 6 401 [22.0%] | 3 676 [28.6%] | 22 088 [52.7%] | 5738 [21.0%] |
| SnoMed inflammation codea | 16 925 [40.4%] | 12 226 [42.1%] | 4699 [36.6%] | 4330 [10.3%] | 10 935 [40.0%] |
| Colon surgery at baseline | |||||
| Resection of the colon | 2483 [5.9%] | 453 [1.6%] | 2030 [15.8%] | 2541 [6.1%] | 1868 [6.8%] |
| Other bowel surgery | 1014 [2.4%] | 464 [1.6%] | 550 [4.3%] | 1634 [3.9%] | 642 [2.3%] |
| IBD therapy exposure at baseline | |||||
| Corticosteroids | 13 251 [31.6%] | 8856 [30.5%] | 4395 [34.2%] | 5620 [13.4%] | 5320 [19.4%] |
| 5-ASA | 16 032 [38.3%] | 13 226 [45.6%] | 2806 [21.8%] | 1054 [2.5%] | 6243 [22.8%] |
| Immunomodulators | 3336 [8.0%] | 1317 [4.5%] | 2019 [15.7%] | 570 [1.4%] | 1960 [7.2%] |
| Anti-TNF | 726 [1.7%] | 344 [1.2%] | 382 [3.0%] | 108 [0.3%] | 433 [1.6%] |
| Vedolizumab | 5 [0.0%] | 5 [0.0%] | 0 | 0 | 3 [0.0%] |
| Ustekinumab | 2 [0.0%] | 1 [0.0%] | 1 [0.0%] | 2 [0.0%] | 0 |
| Tofacitinib | 5 [0.0%] | 5 [0.0%] | 0 | 0 | 3 [0.0%] |
aSnoMed normal mucosa codes: M00100 or M00110; SnoMed inflammation codes: M41, M42, M43; M44, M45, M463 or M47.
Totals may not equal 100% based on missing or unknown data.
Abbreviations: IBD = inflammatory bowel disease, CD = Crohn’s disease, UC = ulcerative colitis, IQR = interquartile range, NA = not applicable.
Table 2.
First incident polyp and follow-up characteristics in patients with IBD and in reference individuals from the general population matched to each patient by age, sex and calendar year.
| IBD | UC | CD | Reference [IBD overall] | IBD [no matched comparator] | |
|---|---|---|---|---|---|
| Number of events | 1648 [3.9%] | 1241 [4.3] | 407 [3.2%] | 1143 [2.7%] | 833 [3.0%] |
| Person-years of follow-up | 357 659 | 243 752 | 113 907 | 333 936 | |
| Polyps | |||||
| Tubular adenoma | 1 177 [71.4%] | 877 [70.7%] | 300 [73.7%] | 839 [73.4%] | 600 [72.0%] |
| Sessile serrated polyps | 13 [0.8%] | 9 [0.7%] | 4 [1.0%] | 1 [0.1%] | 8 [1.0%] |
| Polyp unspecified with LGD | 32 [1.9%] | 27 [2.2%] | 5 [1.2%] | 11 [1.0%] | 24 [2.9%] |
| Any advanced polyp | 487 [29.6%] | 379 [30.5%] | 108 [26.5%] | 340 [29.7%] | 229 [27.5%] |
| Sessile serrated polyps with LGD | 7 [0.4%] | 6 [0.5%] | 1 [0.2%] | 1 [0.1%] | 2 [0.2%] |
| Traditional serrated adenomas | 58 [3.5%] | 42 [3.4%] | 16 [3.9%] | 27 [2.4%] | 29 [3.5%] |
| Traditional serrated adenomas with LGD | 19 [1.2%] | 13 [1.0%] | 6 [1.5%] | 9 [0.8%] | 11 [1.3%] |
| Tubulovillous | 381 [23.1%] | 296 [23.9%] | 85 [20.9%] | 278 [24.3%] | 181 [21.7%] |
| Tubulovillous with HGD | 16 [1.0%] | 14 [1.1%] | 2 [0.5%] | 14 [1.2%] | 7 [0.8%] |
| Villous | 29 [1.8%] | 25 [2.0%] | 4 [1.0%] | 21 [1.8%] | 11 [1.3%] |
| Villous with HGD | 2 [0.1%] | 0 | 2 [0.5%] | 1 [0.1%] | 1 [0.1%] |
| Tubulous with HGD | 16 [1.0%] | 14 [1.1%] | 2 [0.5%] | 17 [1.5%] | 6 [0.7%] |
| Polyp unspecified with HGD | 2 [0.1%] | 2 [0.2%] | 0 | 0 | 0 |
| Age at polyp diagnosis [years] | |||||
| Mean [SD] | 63.5 [12.0] | 63.9 [11.6] | 62.5 [12.9] | 64.0 [12.0] | 50.3 [14.1] |
| Median [IQR] | 64.9 [56.3–71.9] | 65.2 [57.3–72.0] | 63.9 [54.3–71.3] | 64.9 [57.4–72.3] | 51.2 [40.2–60.9] |
| Range, min–max | 20.1–93.2 | 20.1–93.2 | 30.2–92.1 | 19.4–96.4 | 14.5–83.3 |
| <18 | 0 | 0 | 0 | 0 | 7 [0.8%] |
| 18 to <40 | 70 [4.2%] | 46 [3.7%] | 24 [5.9%] | 36 [3.1%] | 197 [23.6%] |
| 40 to <60 | 504 [30.6%] | 369 [29.7%] | 135 [33.2%] | 341 [29.8%] | 403 [48.4%] |
| ≥60 | 1 074 [65.2%] | 826 [66.6%] | 248 [60.9%] | 766 [67.0%] | 226 [27.1%] |
| Year of polyp diagnosis | |||||
| 1969–1976 | 0 | 0 | 0 | 0 | 0 |
| 1977–1990 | 14 [0.8%] | 11 [0.9%] | 3 [0.7%] | 19 [1.7%] | 21 [2.5%] |
| 1991–2002 | 189 [11.5%] | 145 [11.7%] | 44 [10.8%] | 234 [20.5%] | 171 [20.5%] |
| 2003–2017 | 1445 [87.7%] | 1085 [87.4%] | 360 [88.5%] | 890 [77.9%] | 641 [77.0%] |
| Inflammatory status at polyp diagnosis | |||||
| Low histology score | 208 [12.6%] | 155 [12.5%] | 53 [13.0%] | — | 89 [10.7%] |
| High histology score | 1440 [87.4%] | 1086 [87.5%] | 354 [87.0%] | — | 744 [89.3%] |
| Polyp location | |||||
| Colon | 1525 [92.5%] | 1156 [93.2%] | 369 [90.7%] | 940 [82.2%] | 780 [93.6%] |
| Proximal colon | 165 [10.0%] | 131 [10.6%] | 34 [8.4%] | 71 [6.2%] | 73 [8.8%] |
| Distal colon | 161 [9.8%] | 122 [9.8%] | 39 [9.6%] | 104 [9.1%] | 73 [8.8%] |
| Rectum | 647 [39.3%] | 495 [39.9%] | 152 [37.3%] | 339 [29.7%] | 302 [36.3%] |
| Length of follow-up [years] | |||||
| Mean [SD] | 8.5 [6.4] | 8.4 [6.2] | 8.9 [6.8] | 8.0 [6.4] | 13.3 [8.1] |
| Median [IQR] | 7.3 [3.4–12.5] | 7.2 [3.4–12.3] | 7.3 [3.4–12.8] | 6.6 [2.8–11.9] | 13.0 [7.0–17.9] |
| Range, min–max | 0.0–41.5 | 0.0–41.5 | 0.0–37.7 | 0.0–39.5 | 0.0–47.0 |
| 0 to <1 | 2 913 [7.0%] | 2 092 [7.2%] | 821 [6.4%] | 4 739 [11.3%] | 871 [3.2%] |
| 1 to <5 | 12 143 [29.0%] | 8415 [29.0%] | 3728 [29.0%] | 12 098 [28.9%] | 3918 [14.3%] |
| 5 to <10 | 11 872 [28.3%] | 8273 [28.5%] | 3599 [28.0%] | 11 364 [27.1%] | 5391 [19.7%] |
| 10 to <20 | 12 714 [30.4%] | 8972 [30.9%] | 3742 [29.1%] | 11 607 [27.7%] | 12 156 [44.4%] |
| ≥20 | 2238 [5.3%] | 1278 [4.4%] | 960 [7.5%] | 2072 [4.9%] | 5025 [18.4%] |
| No. of endoscopies during follow-up, n [%] | |||||
| 0 | 13 787 [32.9%] | 9214 [31.7%] | 4573 [35.6%] | 31 767 [75.9%] | 5397 [19.7%] |
| 1 | 5688 [13.6%] | 3722 [12.8%] | 1966 [15.3%] | 5577 [13.3%] | 3028 [11.1%] |
| 2 | 5565 [13.3%] | 3755 [12.9%] | 1810 [14.1%] | 2726 [6.5%] | 3182 [11.6%] |
| ≥3 | 16 840 [40.2%] | 12 339 [42.5%] | 4501 [35.0%] | 1810 [4.3%] | 15 754 [57.6%] |
| Colon surgery during follow-up | |||||
| Colectomy or segmental resection | 3618 [8.6%] | 2038 [7.0%] | 1580 [12.3%] | 576 [1.4%] | 4338 [15.9%] |
| Other bowel surgery | 712 [1.7%] | 472 [1.6%] | 240 [1.9%] | 507 [1.2%] | 937 [3.4%] |
| IBD therapy exposure during follow-up | |||||
| Corticosteroids | 20 200 [48.2%] | 14 011 [48.3%] | 6189 [48.2%] | 10 619 [25.4%] | 14 575 [53.3%] |
| 5-ASA | 22 167 [52.9%] | 17 907 [61.7%] | 4260 [33.2%] | 1148 [2.7%] | 14 900 [54.5%] |
| Immunomodulators | 10 757 [25.7%] | 6034 [20.8%] | 4723 [36.8%] | 1144 [2.7%] | 10 548 [38.6%] |
| Anti-TNF | 3776 [9.0%] | 1939 [6.7%] | 1837 [14.3%] | 237 [0.6%] | 4299 [15.7%] |
| Vedolizumab | 366 [0.9%] | 230 [0.8%] | 136 [1.1%] | 0 | 402 [1.5%] |
| Ustekinumab | 43 [0.1%] | 14 [0.0%] | 29 [0.2%] | 10 [0.0%] | 67 [0.2%] |
| Tofacitinib | 365 [0.9%] | 229 [0.8%] | 136 [1.1%] | 0 | 399 [1.5%] |
| Reasons for end of follow-up | |||||
| Event [any polyp] | 1648 [3.9%] | 1241 [4.3%] | 407 [3.2%] | 1143 [2.7%] | 833 [3.0%] |
| Death | 6296 [15.0%] | 4351 [15.0%] | 1945 [15.1%] | 6768 [16.2%] | 1434 [5.2%] |
| Emigration | 387 [0.9%] | 275 [0.9%] | 112 [0.9%] | 276 [0.7%] | 708 [2.6%] |
| Colorectal cancer | 393 [0.9%] | 275 [0.9%] | 118 [0.9%] | 459 [1.1%] | 243 [0.9%] |
| Colectomy | 1841 [4.4%] | 1584 [5.5%] | 257 [2.0%] | 69 [0.2%] | 2344 [8.6%] |
| IBD in control | 0 | 0 | 0 | 3125 [7.5%] | 0 |
| End of follow-up | 31 315 [74.8%] | 21 304 [73.4%] | 10 011 [77.9%] | 30 040 [71.7%] | 21 799 [79.7%] |
Abbreviations: IBD = inflammatory bowel disease, CD = Crohn’s disease, UC = ulcerative colitis, IQR = interquartile range, NA = not applicable.
3.2. Incident colorectal polyp lesions by histology
During follow-up, 1648 [3.9%] IBD patients (n = 407 [3.2%] CD patients and n = 1241 [4.3%] UC patients) and 1143 [2.7%] reference individuals had an incident neoplastic colorectal polyp [Table 2]. The median age at the incident polyp in patients with IBD was 64.9 years [63.9 years in CD, 65.2 years in UC] and 64.9 years in reference individuals. The majority were tubular adenomas (n = 1177 [71.4%] in IBD and n = 839 [73.4%] in reference individuals). Advanced neoplastic lesions, including sessile serrated, traditional serrated, tubulovillous, villous and tubulous with HGD, comprised 29.6% [n = 487] of polyps in patients with IBD compared to 29.7% [n = 340] in the general population comparison cohort. Patients with IBD had a higher proportion of sessile serrated polyps and traditional serrated adenomas [n = 84, 5.1%] compared to reference individuals [n = 37, 3.2%; p = 0.035]. Compared to the reference group for available data, there was a higher proportion of polyps in the proximal colon in patients with IBD [n = 165, 10%, vs n = 71, 6.2%].
These data yielded an incidence rate of neoplastic colorectal polyps of 46.1 per 10 000 person-years in IBD patients and 34.2 per 10 000-person-years in reference individuals [Table 3]. Focusing on advanced polyps, this yielded an incidence rate of 14.9 per 10 000 person-years in IBD patients and 11.2 per 10 000 person-years in reference individuals. In patients with IBD, the adjusted HR [aHR] for any polyp and advanced polyps, adjusted for number of colonoscopic assessments at baseline, was 1.23 [95% CI 1.12–1.35] and 1.18 [95% CI 1.00–1.39], respectively. The highest aHRs were for sessile serrated polyps [8.50, 95% CI 1.10–65.90], traditional serrated adenomas [1.72, 95% CI 1.02–2.91] and tubular adenomas [1.19, 1.06–1.33]. aHRs for polyps were particularly elevated after 20 years of follow-up [2.60, 95% CI 1.24–5.46], in males [1.26, 95% CI 1.10–1.45], and in those diagnosed with IBD age 18–40 years [1.40, 95% CI 1.02–1.92] and in recent years [2003–2016, 1.39, 95% CI 1.22–1.58; Table 4]. We found a 1.3-fold increased risk of polyps in patients with both low [aHR 1.25, 95% CI 0.94–1.66] and high inflammation histology scores [aHR 1.23, 95% CI 1.11–1.36].
Table 3.
Risk of incident polyp by type in patients with IBD [n = 41 880] and matched general population comparators [n = 41 880], UC [n = 29 030] and matched general population comparators [n = 29 030], and CD [n = 12 850] and matched general population comparators [n = 12 850].
| Group | IBD | UC | CD | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N events [%] | Incidence rate [95% CI] per 10 000 person-years | aHRa [95% CI] | No. of events [%] | Incidence rate [95% CI] per 10 000 person-years | aHRa [95%CI] | No. of events [%] | Incidence rate [95% CI] per 10 000 person years | aHRa [95% CI] | |||||||
| IBD | Ref | IBD | Ref | UC | Ref | UC | Ref | CD | Ref | CD | Ref | ||||
| Any polyp | 1648 [3.9%] | 1143 [2.7%] | 46.1 [43.9–48.3] | 34.2 [32.2–36.2] | 1.23 [1.12–1.35] | 1241 [4.3%] | 820 [2.8%] | 50.9 [48.1–53.7] | 35.6 [33.1–38.0] | 1.31 [1.17–1.46] | 407 [3.2%] | 323 [2.5%] | 35.7 [32.3–39.2] | 31.2 [27.8–34.6] | 1.06 [0.89–1.27] |
| Tubular adenoma | 1241 [3.0%] | 885 [2.1%] | 34.5 [32.6–36.4] | 26.4 [24.6–28.1] | 1.19 [1.06–1.33] | 926 [3.2%] | 633 [2.2%] | 37.7 [35.3–40.2] | 27.3 [25.2–29.5] | 1.25 [1.09–1.42] | 315 [2.5%] | 252 [2.0%] | 27.5 [24.5–30.6] | 24.2 [21.3–27.2] | 1.06 [0.86–1.30] |
| Sessile serrated polyps | 16 [0.0%] | 4 [0.0%] | 0.4 [0.2–0.6] | 0.1 [0.0–0.2] | 8.50 [1.10–65.90] | 12 [0.0%] | 3 [0.0%] | 0.5 [0.2–0.7] | 0.1 [–0.0–0.3] | 6.13 [0.63–59.66] | 4 [0.0%] | 1 [0.0%] | 0.3 [0.0–0.7] | 0.1 [−0.1 to 0.3] | — |
| Polyp unspecified with LGD | 40 [0.1%] | 12 [0.0%] | 1.1 [0.8–1.4] | 0.4 [0.2–0.6] | 2.09 [0.97–4.49] | 35 [0.1%] | 12 [0.0%] | 1.4 [0.9–1.9] | 0.5 [0.2–0.8] | 1.67 [0.75–3.71] | 5 [0.0%] | 0 | 0.4 [0.1–0.8] | 0.0 [0.0–0.0] | — |
| Any advanced polyp | 544 [1.3%] | 381 [0.9%] | 14.9 [13.7–16.2] | 11.2 [10.1–12.4] | 1.18 [1.00–1.39] | 428 [1.5%] | 269 [0.9%] | 17.2 [15.6–18.9] | 11.5 [10.1–12.9] | 1.38 [1.13–1.68] | 116 [0.9%] | 112 [0.9%] | 10.0 [8.2–11.9] | 10.7 [8.7–12.7] | 0.79 [0.57–1.09] |
| Sessile serrated poyps with LGD | 8 [0.0%] | 3 [0.0%] | 0.2 [0.1–0.4] | 0.1 [–0.0–0.2] | 8.27 [0.59–116.09] | 7 [0.0%] | 2 [0.0%] | 0.3 [0.1–0.5] | 0.1 [–0.0–0.2] | 8.73 [0.46–166.97] | 1 [0.0%] | 1 [0.0%] | 0.1 [−0.1 to 0.3] | 0.1 [−0.1 to 0.3] | — |
| Sessile serrated polyps with HGD | 0 | 0 | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — | 0 | 0 | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | – | 0 | 0 | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — |
| Traditional serrated adenomas | 74 [0.2%] | 37 [0.1%] | 2.0 [1.6–2.5] | 1.1 [0.7–1.4] | 1.72 [1.02–2.91] | 58 [0.2%] | 23 [0.1%] | 2.3 [1.7–2.9] | 1.0 [0.6–1.4] | 2.64 [1.34–5.19] | 16 [0.1%] | 14 [0.1%] | 1.4 [0.7–2.1] | 1.3 [0.6–2.0] | 0.76 [0.31–1.86] |
| Traditional serrated adenomas with LGD | 23 [0.1%] | 12 [0.0%] | 0.6 [0.4–0.9] | 0.4 [0.2–0.6] | 1.76 [0.69–4.50] | 17 [0.1%] | 8 [0.0%] | 0.7 [0.4–1.0] | 0.3 [0.1–0.6] | 2.35 [0.74–7.52] | 6 [0.0%] | 4 [0.0%] | 0.5 [0.1–0.9] | 0.4 [0.0–0.8] | 0.84 [0.13–5.60] |
| Traditional serrated adenomas with HGD | 0 | 0 | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — | 0 | 0 | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — | 0 | 0 | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — |
| Tubulovillous | 437 [1.0%] | 315 [0.8%] | 12.0 [10.9–13.1] | 9.3 [8.3–10.3] | 1.13 [0.94–1.36] | 341 [1.2%] | 226 [0.8%] | 13.7 [12.2–15.2] | 9.6 [8.4–10.9] | 1.28 [1.03–1.58] | 96 [0.7%] | 89 [0.7%] | 8.3 [6.6–10.0] | 8.5 [6.7–10.3] | 0.82 [0.58–1.17] |
| Tubulovillous with HGD | 24 [0.1%] | 20 [0.0%] | 0.7 [0.4–0.9] | 0.6 [0.3–0.8] | 1.66 [0.75–3.70] | 19 [0.1%] | 16 [0.1%] | 0.8 [0.4–1.1] | 0.7 [0.3–1.0] | 1.93 [0.75–4.95] | 5 [0.0%] | 4 [0.0%] | 0.4 [0.1–0.8] | 0.4 [0.0–0.8] | 1.05 [0.21–5.28] |
| Villous | 35 [0.1%] | 28 [0.1%] | 1.0 [0.6–1.3] | 0.8 [0.5–1.1] | 1.41 [0.71–2.81] | 30 [0.1%] | 17 [0.1%] | 1.2 [0.8–1.6] | 0.7 [0.4–1.1] | 2.15 [0.89–5.16] | 5 [0.0%] | 11 [0.1%] | 0.4 [0.1–0.8] | 1.0 [0.4–1.7] | 0.67 [0.19–2.36] |
| Villous with HGD | 3 [0.0%] | 1 [0.0%] | 0.1 [–0.0–0.2] | 0.0 [−0.0 to 0.1] | — | 0 | 0 | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — | 3 [0.0%] | 1 [0.0%] | 0.3 [−0.0 to 0.6] | 0.1 [−0.1 to 0.3] | — |
| Tubulous with HGD | 21 [0.1%] | 25 [0.1%] | 0.6 [0.3–0.8] | 0.7 [0.4–1.0] | 0.54 [0.22–1.33] | 19 [0.1%] | 21 [0.1%] | 0.8 [0.4–1.1] | 0.9 [0.5–1.3] | 0.68 [0.26–1.80] | 2 [0.0%] | 4 [0.0%] | 0.2 [−0.1 to 0.4] | 0.4 [0.0–0.8] | 0.00 [0.00–.] |
| Polyp unspecified with HGD | 3 [0.0%] | 0 | 0.1 [–0.0–0.2] | 0.0 [0.0–0.0] | — | 3 [0.0%] | 0 | 0.1 [−0.0 to 0.3] | 0.0 [0.0–0.0] | – | 0 | 0 | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — |
aAdjusted for age, sex, calendar period, and number of endoscopies at baseline.
Table 4.
Risk of any polyp overall and by subgroups in patients with IBD, UC and CD, and matched general population comparators.
| IBD | UC | CD | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Group | No. of events [%] | Incidence rate [95% CI] per 10 000 person-years | aHRa [95% CI] | No. of events [%] | Incidence rate [95% CI] per 10 000 person-years | aHRa [95% CI] | No. of events [%] | Incidence rate [95% CI] per 10 000 person-years | aHRa [95% CI] | ||||||
| IBD | Ref | IBD | Ref | UC | Ref | UC | Ref | CD | Ref | CD | Ref | ||||
| Overall | 1648 [3.9%] | 1143 [2.7%] | 46.1 [43.9–48.3] | 34.2 [32.2–36.2] | 1.23 [1.12–1.35] | 1241 [4.3%] | 820 [2.8%] | 50.9 [48.1–53.7] | 35.6 [33.1–38.0] | 1.31 [1.17–1.46] | 407 [3.2%] | 323 [2.5%] | 35.7 [32.3–39.2] | 31.2 [27.8–34.6] | 1.06 [0.89–1.27] |
| Follow-up | |||||||||||||||
| 0 to <1 year | 202 [0.5%] | 267 [0.6%] | 49.9 [43.0–56.8] | 68.4 [60.2–76.6] | 0.72 [0.58–0.89] | 151 [0.5%] | 196 [0.7%] | 53.9 [45.3–62.5] | 72.5 [62.3–82.6] | 0.73 [0.57–0.93] | 51 [0.4%] | 71 [0.6%] | 40.9 [29.7–52.2] | 59.2 [45.5–73.0] | 0.70 [0.46–1.04] |
| 1 to <5 years | 506 [1.3%] | 371 [1.0%] | 38.6 [35.3–42.0] | 30.1 [27.0–33.1] | 1.13 [0.95–1.34] | 382 [1.4%] | 261 [1.0%] | 42.2 [38.0–46.4] | 30.4 [26.7–34.1] | 1.23 [1.01–1.50] | 124 [1.0%] | 110 [1.0%] | 30.7 [25.3–36.0] | 29.3 [23.8–34.7] | 0.91 [0.66–1.25] |
| 5 to <10 years | 482 [1.8%] | 280 [1.1%] | 46.9 [42.7–51.1] | 29.4 [25.9–32.8] | 1.54 [1.27–1.87] | 362 [2.0%] | 205 [1.2%] | 51.1 [45.8–56.3] | 30.8 [26.6–35.1] | 1.64 [1.31–2.06] | 120 [1.4%] | 75 [1.0%] | 37.7 [30.9–44.4] | 26.0 [20.2–31.9] | 1.30 [0.91–1.87] |
| 10 to <20 years | 407 [2.7%] | 203 [1.5%] | 55.8 [50.3–61.2] | 30.6 [26.4–34.8] | 1.82 [1.44–2.30] | 307 [3.0%] | 148 [1.6%] | 63.2 [56.1–70.3] | 32.7 [27.4–38.0] | 1.92 [1.44–2.56] | 100 [2.1%] | 55 [1.3%] | 40.9 [32.9–48.9] | 25.9 [19.1–32.8] | 1.66 [1.11–2.48] |
| ≥20 years | 51 [2.3%] | 22 [1.1%] | 48.7 [35.3–62.1] | 22.5 [13.1–31.9] | 2.60 [1.24–5.46] | 39 [3.1%] | 10 [0.8%] | 67.6 [46.4–88.9] | 17.0 [6.5–27.6] | 5.77 [1.74–19.10] | 12 [1.3%] | 12 [1.5%] | 25.5 [11.1–40.0] | 30.7 [13.3–48.1] | 1.03 [0.35–3.03] |
| ≥1 year | 1 446 [3.7%] | 876 [2.4%] | 45.6 [43.2–47.9] | 29.7 [27.7–31.7] | 1.42 [1.27–1.59] | 1 090 [4.0%] | 624 [2.4%] | 50.5 [47.5–53.5] | 30.7 [28.3–33.1] | 1.53 [1.35–1.75] | 356 [3.0%] | 252 [2.2%] | 35.1 [31.4–38.7] | 27.5 [24.1–30.9] | 1.19 [0.97–1.45] |
| Sex | |||||||||||||||
| Females | 742 [3.3%] | 552 [2.4%] | 36.9 [34.2–39.5] | 29.0 [26.6–31.4] | 1.19 [1.04–1.37] | 525 [3.4%] | 373 [2.4%] | 39.6 [36.2–43.0] | 29.6 [26.6–32.6] | 1.20 [1.01–1.41] | 217 [2.9%] | 179 [2.4%] | 31.5 [27.3–35.7] | 27.8 [23.8–31.9] | 1.20 [0.95–1.51] |
| Males | 906 [4.7%] | 591 [3.1%] | 58.0 [54.2–61.7] | 41.2 [37.9–44.5] | 1.26 [1.10–1.45] | 716 [5.2%] | 447 [3.3%] | 64.4 [59.6–69.1] | 42.9 [38.9–46.8] | 1.41 [1.20–1.64] | 190 [3.6%] | 144 [2.7%] | 42.2 [36.2–48.1] | 36.8 [30.8–42.8] | 0.89 [0.67–1.19] |
| Age | |||||||||||||||
| <18 years | 2 [0.4%] | 1 [0.2%] | 4.5 [0.0–10.7] | 2.7 [0.0–8.1] | — | 2 [0.7%] | 1 [0.4%] | 7.7 [0.0–18.3] | 4.5 [0.0–13.2] | 6212.37 [0.00] | [0.0%] | [0.0%] | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — |
| 18 to <40 years | 185 [1.5%] | 91 [0.7%] | 17.2 [14.7–19.7] | 9.3 [7.4–11.2] | 1.40 [1.02–1.92] | 131 [1.6%] | 62 [0.7%] | 18.8 [15.5–22.0] | 9.6 [7.2–12.0] | 1.37 [0.93–2.04] | 54 [1.3%] | 29 [0.7%] | 14.3 [10.5–18.1] | 8.8 [5.6–12.0] | 1.44 [0.84–2.47] |
| 40 to <60 years | 776 [5.0%] | 504 [3.3%] | 51.4 [47.8–55.0] | 35.7 [32.6–38.9] | 1.32 [1.14–1.52] | 581 [5.4%] | 351 [3.3%] | 56.1 [51.5–60.6] | 35.9 [32.1–39.6] | 1.40 [1.18–1.67] | 195 [4.2%] | 153 [3.3%] | 41.2 [35.4–47.0] | 35.4 [29.8–41.0] | 1.15 [0.89–1.48] |
| ≥60 years | 685 [5.1%] | 547 [4.1%] | 72.4 [67.0–77.8] | 59.7 [54.7–64.6] | 1.10 [0.95–1.28] | 527 [5.5%] | 406 [4.2%] | 77.8 [71.2–84.5] | 61.6 [55.6–67.6] | 1.17 [0.99–1.39] | 158 [4.2%] | 141 [3.7%] | 58.7 [49.5–67.8] | 54.6 [45.6–63.6] | 0.89 [0.65–1.20] |
| Year of IBD onset/index year | |||||||||||||||
| 1969–1976 | 0 | 0 | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — | [0.0%] | [0.0%] | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — | [0.0%] | [0.0%] | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — |
| 1977–1990 | 98 [5.8%] | 77 [4.6%] | 33.4 [26.8–40.0] | 28.2 [21.9–34.5] | 1.31 [0.88–1.95] | 74 [6.9%] | 59 [5.5%] | 43.3 [33.4–53.2] | 35.5 [26.4–44.5] | 1.38 [0.86–2.20] | 24 [3.9%] | 18 [2.9%] | 19.6 [11.8–27.5] | 16.9 [9.1–24.7] | 1.14 [0.52–2.50] |
| 1991–2002 | 723 [6.6%] | 518 [4.7%] | 48.9 [45.4–52.5] | 38.9 [35.5–42.2] | 1.04 [0.89–1.21] | 568 [7.3%] | 370 [4.7%] | 56.1 [51.5–60.8] | 39.8 [35.7–43.9] | 1.19 [0.99–1.43] | 155 [4.8%] | 148 [4.6%] | 33.3 [28.0–38.5] | 36.8 [30.9–42.7] | 0.74 [0.56–0.99] |
| 2003–2016 | 827 [2.8%] | 548 [1.9%] | 45.8 [42.7–48.9] | 31.6 [29.0–34.2] | 1.39 [1.22–1.58] | 599 [3.0%] | 391 [1.9%] | 47.8 [43.9–51.6] | 32.4 [29.2–35.6] | 1.40 [1.20–1.63] | 228 [2.5%] | 157 [1.7%] | 41.4 [36.0–46.8] | 29.9 [25.2–34.5] | 1.37 [1.07–1.74] |
| Year—First 5 years of follow-up | |||||||||||||||
| 1969–1976 | 0 | 0 | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — | [0.0%] | [0.0%] | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — | [0.0%] | [0.0%] | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — |
| 1977–1990 | 18 [1.1%] | 28 [1.7%] | 24.5 [13.2–35.9] | 41.6 [26.2–57.0] | 0.56 [0.27–1.15] | 14 [1.3%] | 25 [2.3%] | 30.6 [14.6–46.7] | 59.5 [36.2–82.8] | 0.48 [0.21–1.09] | 4 [0.7%] | 3 [0.5%] | 14.4 [0.3–28.6] | 11.9 [0.0–25.3] | 0.79 [0.13–4.84] |
| 1991–2002 | 206 [1.9%] | 251 [2.3%] | 41.4 [35.7–47.0] | 55.2 [48.4–62.0] | 0.56 [0.44–0.72] | 168 [2.2%] | 174 [2.2%] | 48.1 [40.9–55.4] | 54.1 [46.0–62.1] | 0.67 [0.50–0.90] | 38 [1.2%] | 77 [2.4%] | 25.5 [17.4–33.6] | 57.8 [44.9–70.7] | 0.35 [0.20–0.59] |
| 2003–2012 | 421 [2.0%] | 359 [1.2%] | 43.9 [39.7–48.1] | 32.6 [29.2–35.9] | 1.20 [1.01–1.43] | 310 [2.1%] | 258 [1.3%] | 46.5 [41.3–51.7] | 33.7 [29.6–37.9] | 1.25 [1.02–1.53] | 111 [1.8%] | 101 [1.1%] | 38.1 [31.0–45.1] | 29.9 [24.1–35.8] | 1.08 [0.78–1.50] |
| Country of birth | |||||||||||||||
| Nordic | 1 537 [4.0%] | 1 041 [2.7%] | 45.7 [43.4–48.0] | 33.3 [31.3–35.3] | 1.22 [1.10–1.35] | 1 166 [4.3%] | 743 [2.8%] | 50.7 [47.8–53.6] | 34.5 [32.0–36.9] | 1.32 [1.16–1.49] | 371 [3.2%] | 298 [2.5%] | 34.9 [31.4–38.5] | 30.7 [27.3–34.2] | 1.01 [0.83–1.22] |
| Other | 111 [3.6%] | 102 [3.3%] | 51.9 [42.3–61.6] | 47.8 [38.5–57.0] | 1.70 [0.59–4.92] | 75 [3.8%] | 77 [3.6%] | 54.5 [42.2–66.9] | 52.1 [40.5–63.8] | 2.38 [0.70–8.13] | 36 [3.3%] | 25 [2.7%] | 47.3 [31.8–62.7] | 38.0 [23.1–52.9] | 0.00 [0.00–.] |
| Montreal Classification CD | |||||||||||||||
| L1/L3/LX | — | — | — | — | — | — | — | — | — | — | 214 [2.7%] | 163 [2.1%] | 38.3 [33.2–43.4] | 31.5 [26.6–36.3] | 1.13 [0.88–1.45] |
| L2 | — | — | — | — | — | — | — | — | — | — | 88 [3.4%] | 63 [2.5%] | 43.9 [34.7–53.1] | 34.0 [25.6–42.4] | 1.27 [0.86–1.87] |
| Montreal Classification UC | |||||||||||||||
| E1/E2 | — | — | — | — | — | 338 [2.9%] | 244 [2.1%] | 41.7 [37.2–46.1] | 32.5 [28.4–36.5] | 1.24 [1.01–1.53] | — | — | — | — | — |
| E3/EX | — | — | — | — | — | 616 [4.1%] | 386 [2.6%] | 53.8 [49.5–58.0] | 35.4 [31.8–38.9] | 1.39 [1.18–1.65] | — | — | — | — | — |
| Extraintestinal manifestations | |||||||||||||||
| PSC | 23 [6.4%] | 10 [2.8%] | 102.1 [60.4–143.8] | 41.0 [15.6–66.4] | 1.90 [0.71–5.05] | 23 [7.7%] | 10 [3.4%] | 120.1 [71.0–169.2] | 49.5 [18.8–80.3] | 1.90 [0.71–5.05] | [0.0%] | [0.0%] | 0.0 [0.0–0.0] | 0.0 [0.0–0.0] | — |
| Other EMF | 89 [3.3%] | 53 [2.0%] | 55.3 [43.8–66.8] | 33.5 [24.5–42.5] | 1.29 [0.85–1.97] | 70 [4.3%] | 37 [2.3%] | 70.3 [53.8–86.7] | 37.3 [25.3–49.3] | 1.41 [0.84–2.37] | 19 [1.8%] | 16 [1.5%] | 31.1 [17.1–45.0] | 27.2 [13.9–40.5] | 1.09 [0.53–2.24] |
| Inflammatory status at index | |||||||||||||||
| Low histology | 208 [4.0%] | 123 [2.4%] | 47.5 [41.0–53.9] | 30.5 [25.1–35.9] | 1.25 [0.94–1.66] | 155 [4.8%] | 92 [2.9%] | 54.7 [46.1–63.3] | 34.8 [27.7–42.0] | 1.32 [0.95–1.83] | 53 [2.7%] | 31 [1.6%] | 34.2 [25.0–43.4] | 22.3 [14.4–30.1] | 0.98 [0.53–1.80] |
| High histology | 1 440 [3.9%] | 1 020 [2.8%] | 45.9 [43.5–48.3] | 34.7 [32.6–36.9] | 1.23 [1.11–1.36] | 1 086 [4.2%] | 728 [2.8%] | 50.4 [47.4–53.4] | 35.7 [33.1–38.3] | 1.31 [1.16–1.48] | 354 [3.3%] | 292 [2.7%] | 36.0 [32.2–39.7] | 32.6 [28.9–36.3] | 1.07 [0.88–1.29] |
| Colon surgery at baseline | |||||||||||||||
| Resection of the colon | 72 [2.9%] | 96 [3.9%] | 24.3 [18.7–29.9] | 37.6 [30.1–45.2] | 0.52 [0.35–0.77] | 9 [2.0%] | 20 [4.4%] | 25.2 [8.8–41.7] | 57.8 [32.5–83.1] | 0.54 [0.20–1.51] | 63 [3.1%] | 76 [3.7%] | 24.2 [18.2–30.2] | 34.5 [26.7–42.2] | 0.51 [0.33–0.78] |
| Other bowel surgery | 25 [2.5%] | 29 [2.9%] | 29.8 [18.1–41.5] | 36.8 [23.4–50.2] | 1.17 [0.52–2.63] | 16 [3.4%] | 16 [3.4%] | 46.2 [23.6–68.9] | 47.0 [24.0–70.1] | 9.85 [1.51–64.23] | 9 [1.6%] | 13 [2.4%] | 18.3 [6.3–30.2] | 29.0 [13.3–44.8] | 0.25 [0.05–1.22] |
aAdjusted for age, sex, calendar period and number of endoscopies at baseline.
Focusing on UC, the aHRs for any neoplastic polyp [1.31, 95% CI 1.17–1.46], tubular adenoma [1.25, 95% CI 1.09–1.42], any advanced polyp [1.38, 95% CI 1.13–1.68], traditional serrated adenomas [2.64, 95% CI 1.34–5.19], tubulovillus polyps [1.28, 95% CI 1.03–1.58] and sessile serrated polyps [6.13, 95% CI 0.63–59.66] were all elevated whereas no difference in risk was noted between patients with CD and reference individuals [Table 3]. In UC, aHRs were particularly elevated in those with more than 20 years of follow-up [5.77, 95% CI 1.74–19.10] and extensive colitis [1.39, 95% CI 1.18–1.65].
The 20-year cumulative incidence of any neoplastic polyp was 9.4% [8.8–10.1%] in IBD, 10.7% [9.9–11.6%] in UC and 6.9% [6.1–7.8%] in CD, compared with 6.1% [5.6–6.6%], 6.4% [5.8–7.0%] and 5.4% [4.7–6.3%] in the respective general population comparison cohorts undergoing colonoscopy, with 20-year cumulative risk differences of 3.3% in IBD, 4.4% in UC and 1.5% in CD [Table 5; Supplementary Tables S8–S10]. This corresponded to one extra case of a polyp in 23 patients with UC and in 67 patients with CD, followed up for 20 years after an IBD diagnosis. Over the entire follow-up period, patients with IBD remained more likely to have a colorectal polyp, particularly patients with UC, compared to reference individuals [Figure 1].
Table 5.
Cumulative incidence of any polyp at 10 and 20 years of follow-up in patients with IBD, UC and CD.
| Group | IBD | UC | CD | ||||||
|---|---|---|---|---|---|---|---|---|---|
| IBD | Comparators | Risk difference | UC | Comparators | Risk difference | CD | Comparators | Risk difference | |
| 10-year cumulative incidence [95% CI] | |||||||||
| Any polyp | 4.3% [4.1–4.6] | 3.3% [3.1–3.6] | 1.0% [0.9; 1.1] | 4.7% [4.4–5.0] | 3.4% [3.2–3.7] | 1.3% [1.1; 1.4] | 3.4% [3.1–3.9] | 3.0% [2.7–3.4] | 0.4% [0.2; 0.6] |
| Tubular adenoma | 3.2% [3.0–3.5] | 2.6% [2.4–2.8] | 0.6% [0.5; 0.8] | 3.5% [3.2–3.7] | 2.7% [2.5–2.9] | 0.8% [0.6; 0.9] | 2.7% [2.4–3.1] | 2.4% [2.1–2.8] | 0.3% [0.1; 0.5] |
| Sessile serrated polyps | 0.0% [0.0–0.1] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.1] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.1] | 0.0% [0.0–0.1] | 0.0% [0.0; 0.1] |
| Polyp unspecified with LGD | 0.1% [0.1–0.2] | 0.0% [0.0–0.1] | 0.1% [0.0; 0.1] | 0.1% [0.1–0.2] | 0.1% [0.0–0.1] | 0.1% [0.0; 0.1] | 0.1% [0.0–0.1] | 0.0% [0.0–0.0] | 0.1% [0.0; 0.1] |
| Any advanced polyp | 1.4% [1.3–1.6] | 1.1% [0.9–1.2] | 0.3% [0.3; 0.4] | 1.6% [1.4–1.8] | 1.1% [0.9–1.2] | 0.5% [0.4; 0.6] | 0.9% [0.8–1.2] | 1.0% [0.8–1.3] | –0.1% [–0.2; 0.0] |
| Sessile serrated polyps with LGD | 0.0% [0.0–0.1] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.1] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.1] | 0.0% [0.0–0.1] | –0.0% [–0.0; 0.0] |
| Sessile serrated polyps with HGD | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] |
| Traditional serrated adenomas | 0.1% [0.1–0.2] | 0.1% [0.1–0.1] | 0.1% [0.0; 0.1] | 0.2% [0.1–0.2] | 0.1% [0.0–0.1] | 0.1% [0.1; 0.1] | 0.1% [0.1–0.2] | 0.1% [0.1–0.3] | –0.0% [–0.1; 0.0] |
| Traditional serrated adenomas with LGD | 0.0% [0.0–0.1] | 0.0% [0.0–0.1] | 0.0% [–0.0; 0.0] | 0.1% [0.0–0.1] | 0.0% [0.0–0.1] | 0.0% [–0.0; 0.0] | 0.0% [0.0–0.1] | 0.0% [0.0–0.1] | –0.0% [–0.0; 0.0] |
| Traditional serrated adenomas with HGD | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] |
| Tubulovillous | 1.1% [1.0–1.3] | 0.9% [0.8–1.0] | 0.3% [0.2; 0.3] | 1.3% [1.1–1.5] | 0.9% [0.8–1.1] | 0.4% [0.3; 0.5] | 0.8% [0.6–1.0] | 0.8% [0.7–1.1] | –0.0% [–0.1; 0.1] |
| Tubulovillous with HGD | 0.1% [0.0–0.1] | 0.0% [0.0–0.1] | 0.0% [–0.0; 0.0] | 0.1% [0.0–0.1] | 0.1% [0.0–0.1] | 0.0% [–0.0; 0.0] | 0.1% [0.0–0.1] | 0.0% [0.0–0.1] | 0.0% [0.0; 0.0] |
| Villous | 0.1% [0.1–0.1] | 0.1% [0.0–0.1] | 0.0% [0.0; 0.1] | 0.1% [0.1–0.2] | 0.1% [0.0–0.1] | 0.1% [0.0; 0.1] | 0.0% [0.0–0.1] | 0.1% [0.0–0.1] | –0.0% [–0.1; –0.0] |
| Villous with HGD | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | –0.0% [–0.0; 0.0] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.1] | 0.0% [0.0–0.1] | –0.0% [–0.0; 0.0] |
| Tubulous with HGD | 0.0% [0.0–0.1] | 0.1% [0.0–0.1] | –0.0% [–0.0; 0.0] | 0.1% [0.0–0.1] | 0.1% [0.0–0.1] | –0.0% [–0.0; –0.0] | 0.0% [0.0–0.1] | 0.0% [0.0–0.1] | 0.0% [–0.0; 0.0] |
| Polyp unspecified with HGD |
0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.1] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] |
| 20-year cumulative incidence [95% CI] | |||||||||
| Any polyp | 9.4% [8.8–10.1] | 6.1% [5.6–6.6] | 3.3% [3.2; 3.5] | 10.7% [9.9–11.6] | 6.4% [5.8–7.0] | 4.4% [4.1; 4.6] | 6.9% [6.1–7.8] | 5.4% [4.7–6.3] | 1.5% [1.2; 1.8] |
| Tubular adenoma | 7.4% [6.9–8.0] | 4.8% [4.4–5.3] | 2.6% [2.4; 2.8] | 8.3% [7.6–9.1] | 5.0% [4.5–5.6] | 3.3% [3.1; 3.5] | 5.6% [4.9–6.5] | 4.4% [3.7–5.1] | 1.3% [1.0; 1.5] |
| Sessile serrated polyps | 0.1% [0.1–0.2] | 0.0% [0.0–0.1] | 0.1% [0.1; 0.1] | 0.2% [0.1–0.4] | 0.0% [0.0–0.1] | 0.1% [0.1; 0.2] | 0.0% [0.0–0.1] | 0.0% [0.0–0.1] | 0.0% [0.0; 0.1] |
| Polyp unspecified with LGD | 0.3% [0.2–0.4] | 0.1% [0.0–0.1] | 0.2% [0.2; 0.2] | 0.4% [0.2–0.6] | 0.1% [0.1–0.2] | 0.3% [0.2; 0.3] | 0.1% [0.0–0.1] | 0.0% [0.0–0.0] | 0.1% [0.0; 0.1] |
| Any advanced polyp | 3.2% [2.9–3.6] | 2.1% [1.9–2.4] | 1.1% [1.0; 1.2] | 3.9% [3.4–4.4] | 2.2% [1.9–2.6] | 1.7% [1.5; 1.8] | 1.9% [1.5–2.4] | 1.9% [1.5–2.5] | 0.0% [–0.2; 0.2] |
| Sessile serrated polyps with LGD | 0.1% [0.0–0.2] | 0.0% [0.0–0.1] | 0.1% [0.0; 0.1] | 0.1% [0.0–0.3] | 0.0% [0.0–0.1] | 0.1% [0.1; 0.1] | 0.0% [0.0–0.1] | 0.0% [0.0–0.1] | –0.0% [–0.0; 0.0] |
| Sessile serrated polyps with HGD | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] |
| Traditional serrated adenomas | 0.6% [0.4–0.8] | 0.3% [0.2–0.4] | 0.3% [0.3; 0.4] | 0.8% [0.5–1.1] | 0.3% [0.2–0.5] | 0.5% [0.4; 0.5] | 0.3% [0.2–0.6] | 0.2% [0.1–0.4] | 0.1% [0.1; 0.2] |
| Traditional serrated adenomas with LGD | 0.2% [0.1–0.3] | 0.1% [0.0–0.1] | 0.1% [0.1; 0.2] | 0.2% [0.1–0.4] | 0.1% [0.0–0.1] | 0.1% [0.1; 0.2] | 0.1% [0.1–0.4] | 0.0% [0.0–0.1] | 0.1% [0.1; 0.1] |
| Traditional serrated adenomas with HGD | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] |
| Tubulovillous | 2.4% [2.1–2.8] | 1.7% [1.5–2.0] | 0.7% [0.6; 0.8] | 2.9% [2.5–3.3] | 1.8% [1.5–2.1] | 1.1% [1.0; 1.3] | 1.6% [1.2–2.0] | 1.5% [1.1–2.1] | 0.0% [–0.1; 0.2] |
| Tubulovillous with HGD | 0.1% [0.1–0.2] | 0.2% [0.1–0.3] | –0.0% [–0.1; –0.0] | 0.2% [0.1–0.3] | 0.2% [0.1–0.4] | –0.1% [–0.1; –0.0] | 0.1% [0.0–0.2] | 0.1% [0.0–0.2] | 0.0% [–0.0; 0.1] |
| Villous | 0.2% [0.1–0.2] | 0.2% [0.1–0.4] | –0.1% [–0.1; –0.0] | 0.2% [0.1–0.3] | 0.1% [0.1–0.3] | 0.0% [0.0; 0.1] | 0.1% [0.0–0.2] | 0.3% [0.1–0.7] | –0.3% [–0.3; –0.2] |
| Villous with HGD | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [–0.0; 0.0] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.1] | 0.0% [0.0–0.1] | 0.0% [–0.0; 0.0] |
| Tubulous with HGD | 0.2% [0.1–0.3] | 0.2% [0.1–0.3] | –0.0% [–0.0; 0.0] | 0.2% [0.1–0.4] | 0.2% [0.1–0.3] | 0.0% [–0.0; 0.1] | 0.0% [0.0–0.1] | 0.1% [0.0–0.3] | –0.1% [–0.1; –0.0] |
| Polyp unspecified with HGD | 0.0% [0.0–0.1] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] | 0.0% [0.0–0.1] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.1] | 0.0% [0.0–0.0] | 0.0% [0.0–0.0] | 0.0% [0.0; 0.0] |
Figure 1.
Kaplan–Meier failure curves of time to diagnosis of adenomatous polyp in patients with IBD [A], CD [B] and UC [C] with matched comparators [reference group].
3.3. Sensitivity analyses
To account for the exclusion of many young IBD patients without a general population matched comparator who also underwent colonoscopy, we adjusted for age as a continuous variable in the matched cohort, producing similar risk estimates for any incident polyp [HR 1.33, 95% CI 1.22–1.45]. Removing age as a matching variable also yielded similar risk estimates for any incident polyp [HR 1.20, 95% CI 1.09–1.33]. Further adjusting for CRC in a first-degree relative [HR 1.32, 95% CI 1.22–1.44] or by removing individuals with genetic risk factors for CRC [HR 1.27, 95% CI 1.15–1.34] again demonstrated consistent findings for any incident polyp.
4. Discussion
In this population-based study of patients with IBD from Sweden, there was an increased risk for neoplastic colorectal polyps compared to the general population, particularly in those with UC. The incidence rate of neoplastic colorectal polyps and advanced polyps was 46.1 and 14.9 per 10 000 person-years, respectively, in IBD patients. In UC, we found a 1.3-fold increased risk of any neoplastic polyp, 1.4-fold increased risk of advanced polyps and 2.6-fold increased risk of traditional serrated adenomas, with the highest risks occurring after 20 years of follow-up, and in those with extensive colitis. While reinforcing previous literature on the risk of colorectal neoplasia in IBD, this is one of the first nationwide studies to report the rate, incidence and adjusted HRs for various neoplastic polyps in IBD.
Although SCENIC and AGA guidelines11,19 have simplified the nomenclature of colorectal neoplasia in IBD, there are few data characterizing the specific histological neoplastic architecture of colorectal lesions in IBD, in comparison to the general population, or with various important covariates, such as histological inflammation.11,19 In one of a few studies, in a cross-sectional, case-control study among 130 204 patients undergoing colonoscopy for the work-up of diarrhoea, in 4 435 patients with IBD, polyp histology was tubular adenoma in 51%, hyperplastic in 41% and serrated adenoma in 8%.9 The prevalence of hyperplastic polyps, serrated adenomas and tubular adenomas were all reduced.9 The authors posited that colorectal inflammation may suppress polyp formation as most neoplasia in IBD does not arise from typical adenomas. These cross-sectional data, however, were probably heavily biased with respect to inadequate selection of appropriate controls. In a recent study of 317 dysplastic lesions from 168 IBD patients in a histopathology database, over 75% had adenomatous features (tubular adenoma-like [n = 183] and tubulovillous/villous adenoma-like [n = 56]) with fewer [n = 78] having non-conventional dysplasia, including polyps with increased Paneth cell differentiation, crypt cell dysplasia, goblet cell deficient, hypermucinous, sessile serrated lesion [SSL]-like and traditional serrated adenoma [TSA]-like.10
While the distribution of neoplastic polyp types was similar between IBD patients and reference individuals, IBD patients more often had polyps in the proximal colon, which may be driven by the increased risk of traditional serrated/sessile serrated polyps due to inflammation.20 However, we cannot exclude the influence of repeated endoscopic evaluations, particularly favouring colonoscopy over sigmoidoscopy in patients with IBD. In addition, previous data have demonstrated that a histological diagnosis of a polyp with adenomatous features in IBD confers a higher risk of subsequent advanced neoplasia compared to an adenomatous polyp in a patient without IBD, or to a patient with IBD but without adenomas.21
The cumulative risk differences between IBD patients and matched reference individuals with colonoscopy were small after 10 years of follow-up, but substantially greater at 20 years of follow-up, especially in UC. The long follow-up and the large number of patients provide substantial power and allowed us to carry out important subanalyses. While we had limited power to examine polyp development in childhood-onset IBD [seven individuals with childhood IBD had a polyp], we found a clear age pattern in adulthood-onset IBD with the highest HRs in patients aged 18–40 years at IBD diagnosis. This age gradient is consistent with our earlier findings on CRC in IBD.2,3 Our findings of a continuously increased HR of colorectal lesions among patients diagnosed with IBD in the last calendar period [2003–2016], despite the introduction of biological therapies, is notable, and may be due to improved endoscopic technology. However, the highest HR was seen in UC, which is consistent with our earlier CRC data.3 Although we did not find an association with PSC, this was probably due to our small sample of patients with PSC.
We suggest several explanations for the increased risk of colorectal neoplastic polyps in IBD. First, although we adjusted for number of colonoscopies at baseline, individuals with IBD underwent significantly more endoscopic evaluations than matched reference individuals of our study, increasing the discovery of colorectal lesions. Second, patients with IBD are likely to have more inflammation than others undergoing a colonoscopy. This was reflected by our findings that 40% of IBD patients had a histopathological diagnosis of mucosal inflammation, compared to only 10% of reference individuals. Furthermore, IBD patients had a higher inflammation histology score than reference individuals14 and colonic inflammation is an established risk factor for colorectal neoplasia.6,22 Several studies have shown that histological healing is a negative predictor of disease relapse and of colorectal neoplasia risk.23–28
Our study has several strengths. We used a validated definition of IBD with a positive predictive value [PPV] of 93–95%.15,16 To ensure that we only examined incident neoplastic polyps, we excluded anyone with CRC, colectomy or a prior record of polyp before diagnosis, and censored follow-up at colectomy. Since IBD patients often undergo surveillance while non-IBD individuals may not, we restricted our comparison cohort to individuals with a record of colonoscopy.3 Also, as screening colonoscopy is not universally performed in Sweden, reference individuals who underwent colonoscopy may have been done for diagnostic purposes and, as such, follow-up colonoscopies were not required, and our analyses may underestimate the risk of neoplasia in IBD compared to the general population. Of note, reference individuals were censored in case they themselves were diagnosed with IBD. Through linkage to the Total Population Register with data on death and emigration, we had a virtually complete follow-up of both IBD patients and reference individuals. We used histopathology data from Sweden’s all 28 pathology departments to ascertain our outcome13 and have previously found a high PPV for serrated/sessile serrated polyps in this cohort.29
Our study also has limitations. These data are not randomized, with some estimates containing wide CIs, and we cannot completely rule out residual confounding. Since polyps are generally asymptomatic, the ideal study to compare rates and types of polyps in IBD and the general population would require equally frequent colonoscopies over long follow-up for unbiased estimates, although we limited our comparison cohort to non-IBD patients who had undergone colonoscopy. We did not have data on colonoscopy quality metrics, visible lesion descriptors, dietary factors, smoking, the microbiome, laboratory measures or medications. At baseline, 31.6% of IBD patients vs 13.4% of reference individuals had a record of steroid medication, and during follow-up the corresponding percentages were 48.4% and 25.4% respectively, but it was beyond the scope of this study to examine the influence of medications on polyp development. Nearly 93% of both IBD patients and reference individuals were born in a Nordic country, and while this decreases the risk of ethnic confounding, it limits generalizability. Due to matching, many IBD patients, particularly younger patients, were excluded. Our data on disease extent and location were based on ICD code data. Moreover, an indication for colonoscopy was not recorded. As biopsies were done at the discretion of providers and performed non-randomly, we cannot accurately link biopsy location and degree of inflammation. Also, a predicted Mayo endoscopic subscore was only available in a subset of IBD patients and our inflammation status data in referred individuals were limited to SnoMed codes, although confirming less inflammation, probably due to diverticular or non-immune-mediated inflammatory diseases, and more often normal mucosa compared to IBD patients.
In summary, in this population-based study of patients with IBD from Sweden, there was an increased risk for incident colorectal neoplastic polyps, particularly those of adenomatous, traditional serrated and tubulovillous architecture in patients with UC, compared to matched reference individuals. Future studies should evaluate the influence of colorectal neoplasia and associated specific lesional architecture on the risks of subsequent advanced colorectal neoplasia and related outcomes.
Supplementary Material
Contributor Information
Jordan E Axelrad, Inflammatory Bowel Disease Center at NYU Langone Health, Division of Gastroenterology, Department of Medicine, NYU Grossman School of Medicine, New York, NY>, USA.
Ola Olén, Clinical Epidemiology Division, Department of Medicine Solna, Karolinska Institutet, Stockholm, Sweden; Sachs’ Children and Youth Hospital, Stockholm South General Hospital, Stockholm, Sweden; Department of Clinical Science and Education Södersjukhuset, Karolinska Institutet, Stockholm, Sweden.
Jonas Söderling, Clinical Epidemiology Division, Department of Medicine Solna, Karolinska Institutet, Stockholm, Sweden.
Bjorn Roelstraete, Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.
Hamed Khalili, Division of Gastroenterology, Department of Medicine, Massachusetts General Hospital, Boston, MA, USA.
Mingyang Song, Inflammatory Bowel Disease Center at NYU Langone Health, Division of Gastroenterology, Department of Medicine, NYU Grossman School of Medicine, New York, NY>, USA.
Adam Faye, Inflammatory Bowel Disease Center at NYU Langone Health, Division of Gastroenterology, Department of Medicine, NYU Grossman School of Medicine, New York, NY>, USA.
Michael Eberhardson, Department of Gastroenterology and Hepatology, Linköping University Hospital, Linköping University and Karolinska Institutet, Linköping, Sweden.
Jonas Halfvarson, Department of Gastroenterology, Faculty of Medicine and Health, Örebro University, Örebro, Sweden.
Jonas F Ludvigsson, Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden; Department of Pediatrics, Orebro University Hospital, Orebro, Sweden; Department of Medicine, Columbia University College of Physicians and Surgeons, New York, NY, USA.
Funding
JEA was supported by grants from the Crohn’s and Colitis Foundation, the Judith & Stewart Colton Center for Autoimmunity, and the NIH NIDDK Diseases K23DK124570. OO was supported by grants from the Swedish Research Council [Dnr 2020-02002], the Young Scholar Award from the Strategic Research Area Epidemiology Program at Karolinska Institutet, and the Regional Agreement on Medical Training and Clinical Research between Stockholm County Council [ALF] and the Karolinska Institutet [ALF]. JFL was supported by the FORTE Foundation and the Swedish Cancer Foundation. HK is supported by the American College of Gastroenterology Senior Research Award and the Beker Foundation. MS is supported by the American Cancer Society Mentored Research Scholar Grant and NIH grants [R00CA215314 and U01CA261961]. AF is supported by the Crohn’s and Colitis Foundation. None of the funding organizations had any role in the design and conduct of the study; in the collection, management and analysis of the data; or in the preparation, review and approval of the manuscript.
Conflict of Interest
JEA has received research grants from BioFire Diagnostics; consultancy, advisory board fees or honorarium from BioFire Diagnostics, Abbvie, Adiso, Pfizer, BMS, Fresnius and Janssen; and holds U.S. patent 2012/0052124A1. OO has been PI on projects at Karolinska Institutet financed by grants from Janssen, Takeda, AbbVie and Ferring. He also reports a grant from Pfizer in the context of a national safety monitoring programme. None of those studies have any relation to the present study. The Karolinska Institutet has received fees for OO’s lectures and participation on advisory boards from Janssen, Ferring, Takeda, Galapagos, Bristol Myers Squibb and Pfizer regarding topics not related to the present study. HK has received consulting fees from Abbvie and Takeda; HK has also received grant funding from Pfizer and Takeda. JH has served as speaker, consultant and/or advisory board member for AbbVie, Aqilion, Celgene, Celltrion, Ferring, Galapagos, Gilead, Hospira, Janssen, MEDA, Medivir, MSD, Novartis, Pfizer, Prometheus Laboratories Inc, Sandoz, Shire, Takeda, Thermo Fisher Scientific, Tillotts Pharma and Vifor Pharma, and has also received grant support from Janssen, MSD and Takeda. JFL has coordinated a study unrelated to the present study on behalf of the Swedish IBD Quality Register [SWIBREG]. That study received funding from Janssen. ME has received honoraria for lectures and consultancy from AbbVie, Merck [MSD], Takeda, Ferring, Orion Pharma, Otsuka, Tillotts, Novartis, Pfizer, Bristol Myers Squibb and Janssen, and received research funding from AbbVie and MSD. AF has received consulting fees for GLG, M3 and Janssen. Authors not named here have disclosed no conflicts of interest.
Author Contributions
Guarantor: Olén had full access to all the data in the study and takes responsibility for the integrity of the data. Söderling takes responsibility for the accuracy of the data analysis. Study concept and design: all co-authors. Acquisition of data: Olén, Ludvigsson. Analysis: Soderling. Funding: Olén, Ludvigsson and Axelrad. Writing first draft of the manuscript: Axelrad and Ludvigsson. Critical revision of the manuscript for important intellectual content and approval of final version: all co-authors.
Ethics Approval
Swedish data—2007/785-31/5, 2011/1509-32, 2014/1287-31/4, 2015/0004-31 and 2016/192-31/2.
Data Availability
According to Swedish regulations, data will not be made available to other researchers, but can be requested from the National Board of Health and Welfare, and Swedish pathology departments.
References
- 1. Beaugerie L, Itzkowitz SH.. Cancers complicating inflammatory bowel disease. N Engl J Med 2015;372:1441–52. [DOI] [PubMed] [Google Scholar]
- 2. Olén O, Erichsen R, Sachs MC, et al. Colorectal cancer in Crohn’s disease: a Scandinavian population-based cohort study. Lancet Gastroenterol. Hepatol 2020;5:475–84. [DOI] [PubMed] [Google Scholar]
- 3. Olén O, Erichsen R, Sachs MC, et al. Colorectal cancer in ulcerative colitis: a Scandinavian population-based cohort study. Lancet 2020;395:123–31. [DOI] [PubMed] [Google Scholar]
- 4. Olén O, Askling J, Sachs MC, et al. Mortality in adult-onset and elderly-onset IBD: a nationwide register-based cohort study 1964-2014. Gut 2019. [DOI] [PubMed] [Google Scholar]
- 5. Olén O, Askling J, Sachs MC, et al. Increased mortality of patients with childhood-onset inflammatory bowel diseases, compared with the general population. Gastroenterology 2019;156:614–22. [DOI] [PubMed] [Google Scholar]
- 6. Axelrad JE, Shah SC.. Diagnosis and management of inflammatory bowel disease-associated neoplasia: considerations in the modern era. Therap Adv Gastroenterol 2020;13:1756284820920779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. van den Broek FJC, Stokkers PCF, Reitsma JB, et al. Random biopsies taken during colonoscopic surveillance of patients with longstanding ulcerative colitis: low yield and absence of clinical consequences. Am J Gastroenterol 2014;109:715–22. [DOI] [PubMed] [Google Scholar]
- 8. Lahiff C, Mun Wang L, Travis SPL, East JE.. Diagnostic yield of dysplasia in polyp-adjacent biopsies for patients with inflammatory bowel disease: a cross-sectional study. J Crohns Colitis 2018;12:670–6. [DOI] [PubMed] [Google Scholar]
- 9. Laine L, Kaltenbach T, Barkun A, McQuaid KR, Subramanian V, Soetikno R; SCENIC Guideline Development Panel. SCENIC international consensus statement on surveillance and management of dysplasia in inflammatory bowel disease. Gastrointest Endosc 2015;81:489–501.e26. [DOI] [PubMed] [Google Scholar]
- 10. Ludvigsson JF, Otterblad-Olausson P, Pettersson BU, Ekbom A.. The Swedish personal identity number: possibilities and pitfalls in healthcare and medical research. Eur J Epidemiol 2009;24:659–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Ludvigsson JF, Lashkariani M.. Cohort profile: ESPRESSO (Epidemiology Strengthened by histoPathology Reports in Sweden). Clin Epidemiol 2019;11:101–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Axelrad JE, Sachs MC, Ludvigsson JF, Olén O; SWIBREG Study Group. A novel method for quantifying intestinal inflammatory burden in inflammatory bowel disease using register data. Clin Epidemiol 2020;12:1059–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Nguyen LH, Örtqvist AK, Cao Y, et al. Antibiotic use and the development of inflammatory bowel disease: a national case-control study in Sweden. Lancet Gastroenterol. Hepatol 2020;5:986–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Jakobsson GL, Sternegård E, Olén O, et al. Validating inflammatory bowel disease (IBD) in the Swedish National Patient Register and the Swedish Quality Register for IBD (SWIBREG). Scand J Gastroenterol 2017;52:216–21. [DOI] [PubMed] [Google Scholar]
- 15. Shrestha S, Olén O, Eriksson C, et al. The use of ICD codes to identify IBD subtypes and phenotypes of the Montreal classification in the Swedish National Patient Register. Scand J Gastroenterol 2020;1:6. [DOI] [PubMed] [Google Scholar]
- 16. Barlow L, Westergren K, Holmberg L, Talbäck M.. The completeness of the Swedish Cancer Register: a sample survey for year 1998. Acta Oncol 2009;48:27–33. [DOI] [PubMed] [Google Scholar]
- 17. Murthy SK, Feuerstein JD, Nguyen GC, Velayos FS.. AGA clinical practice update on endoscopic surveillance and management of colorectal dysplasia in inflammatory bowel diseases: expert review. Gastroenterology 2021;161:1043–51.e4. [DOI] [PubMed] [Google Scholar]
- 18. Sonnenberg A, Genta RM.. Low prevalence of colon polyps in chronic inflammatory conditions of the colon. Am J Gastroenterol 2015;110:1056–61. [DOI] [PubMed] [Google Scholar]
- 19. Lee H, Rabinovitch PS, Mattis AN, Lauwers GY, Choi W-T.. Non-conventional dysplasia in inflammatory bowel disease is more frequently associated with advanced neoplasia and aneuploidy than conventional dysplasia. Histopathology 2021;78:814–30. [DOI] [PubMed] [Google Scholar]
- 20. Wang L, Lo C-H, He X, et al. Risk factor profiles differ for cancers of different regions of the colorectum. Gastroenterology 2020;159:241–256.e13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. van Schaik FDM, Mooiweer E, van der Have M, et al. Adenomas in patients with inflammatory bowel disease are associated with an increased risk of advanced neoplasia. Inflamm Bowel Dis 2013;19:342–9. [DOI] [PubMed] [Google Scholar]
- 22. Rutter M, Saunders B, Wilkinson K, et al. Severity of inflammation is a risk factor for colorectal neoplasia in ulcerative colitis. Gastroenterology 2004;126:451–9. [DOI] [PubMed] [Google Scholar]
- 23. Zenlea T, Yee EU, Rosenberg L, et al. Histology grade is independently associated with relapse risk in patients with ulcerative colitis in clinical remission: A prospective study. Am J Gastroenterol 2016;111:685–90. [DOI] [PubMed] [Google Scholar]
- 24. Gupta RB, Harpaz N, Itzkowitz S, et al. Histologic inflammation is a risk factor for progression to colorectal neoplasia in ulcerative colitis: a cohort study. Gastroenterology 2007;133:1099–105; quiz 1340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Bryant RV, Burger DC, Delo J, et al. Beyond endoscopic mucosal healing in UC: histological remission better predicts corticosteroid use and hospitalisation over 6 years of follow-up. Gut 2016;65:408–14. [DOI] [PubMed] [Google Scholar]
- 26. Frøslie KF, Jahnsen J, Moum BA, Vatn MH; IBSEN Group. Mucosal healing in inflammatory bowel disease: results from a Norwegian population-based cohort. Gastroenterology 2007;133:412–22. [DOI] [PubMed] [Google Scholar]
- 27. Christensen B, Erlich J, Gibson PR, Turner JR, Hart J, Rubin DT.. Histologic healing is more strongly associated with clinical outcomes in ileal Crohn’s disease than endoscopic healing. Clin Gastroenterol Hepatol 2020;18:2518–2525.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Colman RJ, Rubin DT.. Histological inflammation increases the risk of colorectal neoplasia in ulcerative colitis: a systematic review. Intest Res 2016;14:202–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Bozorg SR, Song M, Emilsson L, Ludvigsson JF.. Validation of serrated polyps (SPs) in Swedish pathology registers. BMC Gastroenterol 2019;20:3. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
According to Swedish regulations, data will not be made available to other researchers, but can be requested from the National Board of Health and Welfare, and Swedish pathology departments.