Alain Demers

Rate and Predictors of Early / Missed Colorectal Cancers After Colonoscopy in Manitoba: A Population-Based Study

OBJECTIVES:Many of the colorectal cancers (CRCs) diagnosed within 3 years after a colonoscopy are likely because of lesions missed on the initial colonoscopy. In this population-based study, we investigated the rate and predictors of CRCs diagnosed within 3 years of a colonoscopy.METHODS:We identified individuals 50–80 years of age diagnosed with CRC between 1992 and 2008 from the provincewide Manitoba Cancer Registry. Performance of colonoscopy and history of co-morbidities was determined by linkage to the provincial universal health care insurance providers physician billing claims and hospital discharges databases. CRCs diagnosed within 6 months of a colonoscopy were categorized as detected CRCs and those 6–36 months after a colonoscopy as early/missed CRCs. Logistic regression analysis was performed to identify the patient, endoscopist, colonoscopy, and CRC factors associated with early/missed CRCs.RESULTS:Of the 4,883 CRCs included in the study, 388 (7.9%) were early/missed CRCs, with a range of 4.5% of rectum/rectosigmoid cancers in men to 14.4% of transverse colon/splenic flexure cancers in women. Independent risk factors associated with early/missed CRCs included prior colonoscopy, performance of index colonoscopy by family physicians, recent year of CRC diagnosis, and proximal site of CRC.CONCLUSIONS:This study suggests that approximately 1 in 13 CRCs may be an early/missed CRC, diagnosed after an index colonoscopy in usual clinical practice. Women are more likely to have early/missed CRC. It is unclear if this relates to differences in procedure difficulty, bowel preparation issues, or tumor biology between men and women.

Predictors of colorectal cancer after negative colonoscopy: a population-based study.

OBJECTIVES:A higher proportion of colorectal neoplasia among women occurs in the proximal colon, which might be more frequently missed by colonoscopy. There are no data on predictors of developing colorectal cancer (CRC) after a negative colonoscopy in usual clinical practice. We evaluated gender differences and predictors of CRC occurring after a negative colonoscopy.METHODS:All individuals 40 years or older with negative colonoscopy were identified from Manitobas provincial physicians’ billing claims database. Individuals with less than 5 years of coverage by the provincial health plan, earlier CRC, inflammatory bowel disease, resective colorectal surgery, or lower gastrointestinal endoscopy were excluded. CRC risk after negative colonoscopy was compared to that in the general population by standardized incidence ratios. Cox regression analysis was performed to determine the independent predictors of CRC occurring after negative colonoscopy.RESULTS:A total of 45,985 individuals (18,606 men; 27,379 women) were followed up for 229,090 person-years. After a negative colonoscopy, men had a 40–50% lower risk of CRC diagnosis through most of the follow-up time. Risk among women was similar to that of women in the general population in the first 3 years and then was 40–50% lower. Older subject age and performance of index colonoscopy by non-gastroenterologists were independent predictors for early/missed CRC (cancers occurring within 3 years of negative colonoscopy).CONCLUSIONS:Women may have a higher rate of missed/early CRCs after negative colonoscopy. Predictors of missed/early CRCs after negative colonoscopy include older age and performance of index colonoscopy by a non-gastroenterologist.

Increased Risk of Nonmelanoma Skin Cancers Among Individuals With Inflammatory Bowel Disease

BACKGROUND & AIMS There are limited data on the risk of nonmelanoma skin cancer (NMSC) among individuals with inflammatory bowel disease (IBD), including those with or without exposure to immunosuppressant medications. METHODS Individuals with IBD (n = 9618) were identified from the University of Manitoba IBD Epidemiology Database and matched with randomly selected controls (n = 91,378) based on age, sex, and postal area of residence on the date of IBD diagnosis (index date). Groups were followed up from the index date until a diagnosis of any invasive cancer (including NMSC), death, migration from the province, or the end of the study (December 31, 2009), whichever came first. Cox regression analysis was performed to calculate the relative risk of NMSC among the individuals with IBD, adjusting for frequency of ambulatory care visits and socioeconomic status. RESULTS Of the individuals followed, 1696 were diagnosed with basal cell skin cancer (BCC) and 341 were diagnosed with squamous cell skin cancer (SCC). Individuals with IBD had an increased risk for BCC, compared with controls (hazard ratio, 1.20; 95% confidence interval [CI], 1.03-1.40). Among patients with IBD, use of thiopurines increased the risk of SCC (hazard ratio, 5.40; 95% CI, 2.00-14.56), compared with controls. Use of thiopurines also was associated with SCC in a case-control, nested analysis of individuals with IBD (odds ratio, 20.52; 95% CI, 2.42-173.81). CONCLUSIONS The risk of BCC could be increased among individuals with IBD. Use of thiopurines increases the risk of SCC among individuals with IBD.

Risk of Cervical Abnormalities in Women With Inflammatory Bowel Disease: A Population-Based Nested Case-Control Study

BACKGROUND & AIMS We evaluated the risk of cervical abnormalities in women with inflammatory bowel disease (IBD) in a population-based, nested, case-control study. METHODS Cases with abnormal Papanicolaou (Pap) smears or cervical biopsies were matched with up to 3 controls (normal Pap smears) by year of birth, year of first health care coverage, and number of Pap smears in the preceding 5 years. A diagnosis of IBD before the index date was identified from the University of Manitoba IBD Epidemiology Database. Exposures to immunosuppressant drugs and corticosteroids were determined from the provincial drug prescription database. Analyses were adjusted for socioeconomic status and exposure to oral contraceptives and nonsteroidal anti-inflammatory drugs. RESULTS 19,692 women with cervical cytologic and/or histologic abnormalities were matched with 57,898 controls with normal Pap smears. There was no association between cervical abnormalities and ulcerative colitis (odds ratio [OR], 1.03; 95% confidence interval [CI], 0.77-1.38). The increase in risk in women with Crohns disease was limited to those exposed to 10 or more prescriptions of oral contraceptives (OR, 1.66; 95% CI, 1.08-2.54). The combined exposure to corticosteroids and immunosuppressants was associated with increased risk of cervical abnormalities (OR, 1.41; 95% CI, 1.09-1.81). There was no interaction between the effect of IBD and corticosteroids and/or immunosuppressants. CONCLUSIONS These findings do not support an association between IBD itself and the risk of developing cervical abnormalities. An increased risk in patients given a combination of corticosteroids and immunosuppressants should be considered in managing women with IBD.

Risk of Second Primary Cancer and Death Following a Diagnosis of Nonmelanoma Skin Cancer

Cancer-free patients diagnosed with a first primary nonmelanoma skin cancer (NMSC) offer an opportunity for studying the risk of a second primary cancer without the confounding effect of systemic treatment. The objective of the study was to estimate the risk of second primary cancer in people with a history of basal cell carcinoma (BCC) or squamous cell carcinoma (SCC) and the risk of dying in cancer patients with a NMSC history. BCC and SCC cases diagnosed between 1956 and 2000 in Manitoba, Canada were followed-up for second primaries (other than NMSC). Standardized incidence and mortality ratios (SIR and SMR) were calculated. Men [SIR, 1.06; 95% confidence interval (95% CI), 1.02-1.10] and women (SIR, 1.07; 95% CI, 1.02-1.12) with a BCC history as well as men (SIR, 1.15; 95% CI, 1.08-1.22) with a SCC history were at greater risk of a second primary cancer. Overall, the increased risk was observed only in the first 4 years following a NMSC, although it remained increased for specific cancer sites. The risk remained higher in all age groups up to 75 years of age. People with a history of BCC (males: SMR, 1.09; 95% CI, 1.04-1.14; females: SMR, 1.24; 95% CI, 1.16-1.32) or SCC (males: SMR, 1.18; 95% CI, 1.09-1.27; females: SMR, 1.55; 95% CI, 1.35-1.79) had a greater risk of death following their second primaries. Even if NMSC patients are at greater risk of a second cancer, it is not recommended to follow them up beyond the generally accepted periodic examination of the skin.

Long-term use of statins and risk of colorectal cancer: a population-based study.

OBJECTIVES:We conducted a population-based cohort study to determine the effect of long-term regular use of statins on the risk of colorectal cancer (CRC).METHODS:Individuals who were dispensed statins regularly were identified from Manitobas population-based prescription drug database and followed up until diagnosis of CRC, migration out of province, death, or December 2005. The incidence of CRC in this group was compared with that among individuals who were never dispensed statins. Stratified analysis was performed to determine the risk after 5 years of regular statin use. Multivariate Poisson regression models were used to adjust for potential confounding by age, sex, and history of diabetes, inflammatory bowel disease, coronary heart disease, lower gastrointestinal endoscopy, resective colorectal surgery, use of nonsteroidal anti-inflammatory drugs, hormone replacement therapy (among women), and median household income. The dose effect was evaluated in defined daily dose units.RESULTS:In total, 35,739 individuals were dispensed statins regularly. In all, 10,287 (49% males; 51% females) long-term (≥5 years) regular statin users were followed up for up to 5 additional years. In multivariate analysis, the incidence rate ratio (IRR) of CRC among those dispensed statins regularly compared with those who were never dispensed statins (n=377,532) was 1.13 (95% confidence interval (CI): 1.02–1.25). The CRC risk among the long-term regular statin users was similar to that for individuals never dispensed statins (IRR, 0.89; 95% CI: 0.70–1.13). A statistically nonsignificant risk reduction was observed among high-dose long-term regular statin users.CONCLUSIONS:These findings suggest that long-term regular use of statins for the current clinical indications does not protect against CRC. The benefit of high-dose long-term statin use needs further evaluation.

Twenty-year trends in the incidence and prevalence of diagnosed anogenital warts in Canada.

Background: A vaccine has recently been licensed in many countries that protects against the human papillomavirus types 6, 11, 16, and 18. Types 6 and 11 account for approximately 90% of anogenital warts (AGWs). We describe the 20-year trends in the incidence and prevalence of AGWs in Manitoba, Canada. Methods: We used linked population-based hospital and physician databases for Manitoba for 1984 to 2004. Cases were identified using tariff (billing) and ICD codes. A case was considered to be incident if it was preceded by a 12-month interval free period of AGWs care. Otherwise, it was deemed to be prevalent. An episode was considered over once a 12-month interval had elapsed without an AGW claim. Results: Approximately 25,000 Manitobans were diagnosed with AGWs between 1985 and 2004. The annual age-standardized incidence rates peaked in 1992 (men, 149.9/100,000; women 170.8/100,000). In recent years, the rates have been increasing again, particularly for men. The male:female incidence rate ratio increased from 0.76 in 1985 to 1.25 in 2004. The highest incidence rate tended to be in those aged 20 to 24 years. Trends in prevalence were similar. Prevalence in 2004 was 165.2/100,000 for men and 128.4/100,000 for women. Conclusions: These population-based findings suggest that AGWs are a substantial burden to Manitobans and that their pattern has changed over time, with incidence and prevalence becoming higher in men than women. Monitoring the future trends in AGWs will provide an early marker of the effectiveness and duration of protection of human papillomavirus vaccination at a population level.

Time trends in colon cancer incidence and distribution and lower gastrointestinal endoscopy utilization in Manitoba.

OBJECTIVES:There are limited data on recent trends in subsite-specific colon cancer incidence and utilization of lower gastrointestinal endoscopy from Canada. The aim of our study was to determine the concomitant trends in right-sided colon cancer incidence and utilization of colonoscopy and flexible sigmoidoscopy (FS) in Manitoba.METHODS:Cases of colon cancer diagnosed from 1964 to 2004 were identified from the Manitoba Cancer Registry. Lower gastrointestinal endoscopies performed between 1984 and 2003 were identified from Manitoba Healths Physician Claims database. Trends of age-standardized incidence rates were determined using Joinpoint analyses.RESULTS:Rates of right-sided colon cancer showed a monotonic increase in both sexes (annual percent change [APC] in both sexes 1.04%, P < 0.001). The most rapid increase (200%) occurred in individuals of 70 yr of age and older. While rates of colonoscopies without polypectomies quadrupled between 1985 (257 per 100,000) and 2003 (1,083 per 100,000, APC 8.89%, P < 0.001), rates of colonoscopies with polypectomies quadrupled from 35 per 100,000 in 1985 to 140 per 100,000 in 2000, and then increased more rapidly in the subsequent 4 yr (233 per 100,000 in 2003, APC 20%, P < 0.001). Rates of FS declined between 1999 (342 per 100,000) and 2003 (257 per 100,000, APC −6.68%, P = 0.01).CONCLUSIONS:The rates of right-sided colon cancer are continuing to increase in Manitoba, with the most rapid increase occurring in older individuals. Reasons for the increasing incidence of right-sided colon cancer despite increasing rates of colonoscopy need to be further explored, and may reflect increased detection of asymptomatic cancers or a real increase in right-sided colon cancer incidence.

Incidence and anatomic presentation of cutaneous malignant melanoma in central Canada during a 50-year period: 1956 to 2005

BACKGROUND Incidence rates of cutaneous malignant melanoma (CMM) have increased worldwide. Long-term studies examining rates and anatomic site-specific incidence on a population-based level are infrequent. OBJECTIVE We sought to examine the historical changes in the incidence and anatomic site presentation of CMM during a 50-year period in Manitoba, Canada. METHODS Using population-based data, all first diagnoses of CMM reported between 1956 and 2005 were identified. Age-specific rates, age-standardized incidence rates, and anatomic sites were recorded. RESULTS Incidence rates of CMM slowed for each sex beginning in 1981 for female patients and 1992 for male patients. Annual percent change revealed decreasing rates among male patients younger than 40 years (1992-2005: -5.3% [P = .03]) and female patients younger than 40 years (1987-2005: -1.8% [P = .15]). Similarly, middle-aged individuals (age 40-59 years) also had diminished annual percent change (men 1992-2005: 0.6% [P = .65]; women 1983-2005: -0.3% [P = .68]). The annual percent change for older men and women (60-79 and > or =80 years) continued to increase. Anatomic site-specific analyses revealed that the trunk was the most frequent site of CMM for young male patients (<60 years) whereas the lower extremities were the most common among young female patients (<60 years). Incidence rates for each site, however, are slowing. Among those aged 60 years and older, the rates for each anatomic site increased. LIMITATIONS Determining changes in tumor thickness would have been useful in determining whether the nature of tumors have changed over time; however, this is not recorded in our registry. CONCLUSION The rates of CMM are slowing; however, this change is confined to younger individuals. Anatomic site-specific CMMs are changing; rates among older individuals continue to increase for both sexes.

Effectiveness of the Quadrivalent Human Papillomavirus Vaccine Against Cervical Dysplasia in Manitoba, Canada

PURPOSE Effectiveness of the quadrivalent human papillomavirus (QHPV) vaccine against cervical dysplasia has not been estimated using population-based individual level data. We assessed the vaccine effectiveness (VE) of the QHPV vaccine against cervical dysplasia using data collected routinely in Manitoba. METHODS Females ≥ 15 years old who received the QHPV vaccine in Manitoba between September 2006 and April 2010 privately (n = 3,541) were matched on age to up to three nonvaccinated females (n = 9,594). We used Cox regression models to estimate the hazard ratios for three outcomes: atypical squamous cells of undetermined significance (ASCUS), low-grade squamous intraepithelial lesions (LSILs), and high-grade SILs (HSILs). RESULTS Among the 15- to 17-year-olds, the adjusted VE estimates were 35% (95% CI, -19% to 65%), 21% (-10% to 43%), and -1% (-44% to 29%) against the detection of HSILs, LSILs, and ASCUS, respectively. The corresponding estimates were higher (46% [0% to 71%], 35% [10% to 54%], and 23% [-8% to 45%]) among those who had ≥ one Pap smear after enrollment. The QHPV vaccine was associated with 23% (-17% to 48%) reduction in HSIL risk among those ≥ 18 with no history of abnormal cytology, but there was no evidence of protection among those with such a history (-8% [-59% to 27%]). CONCLUSION A significant percentage of vaccinated women may not be protected against HSIL and lesser dysplasia especially if they were vaccinated at older age (≥ 18) or had abnormal cytology before vaccination. These findings affirm the importance of vaccination before any significant exposure to HPV occurs and underscore the need for screening programs that cover all sexually active women, even if they were vaccinated.