Kim Ekblom

High iron levels can increase the formation of noxious oxygen radicals, which are thought to promote carcinogenesis. The aim of this prospective study was to determine whether iron biomarkers and HFE genotypes, which influence iron regulation, constitute risk factors for colorectal cancer. This is a prospective nested case-referent study. The study was performed within the population-based Northern Sweden Health and Disease Study. The study included 226 cases of colorectal cancer and 437 matched referents. Conditional regression analysis was performed. Adjustments for smoking, smoking and BMI, and HFE C282Y and H63D were performed. The highest quintile of total iron-binding capacity showed significantly higher risk for colorectal cancer, unadjusted OR 2.35 (95% CI 1.38-4.02). When stratified by sex, the findings were only present in women (OR 3.34 (95% CI 1.59-7.02)). Ferritin was associated with reduced risk throughout quintiles 2 to 5 both in univariate and multivariate models. Colorectal cancer may influence iron markers because of occult bleeding. Homozygotes for HFE C282Y were too few to make conclusions for this group. The relatively short follow-up time might be insufficient to detect risk of iron biomarkers. High iron levels do not increase the risk of colorectal cancer. HFE genotypes influencing iron uptake had no effect on colorectal cancer risk.

Background: Bilirubin, an antioxidant, has been associated with reduced cardiovascular disease risk. A major cause of elevated plasma bilirubin is the common UGT1A1*28 promoter polymorphism in the gene of the bilirubin-conjugating enzyme UDP-glucuronosyltransferase 1A1, which reduces transcription by 70%. Earlier studies reporting a protective effect of bilirubin on stroke have not included analysis of UGT1A1*28. The purpose of this study is to investigate if bilirubin and UGT1A1*28 are protective against ischemic stroke in a prospective case-referent setting. Methods: Cases with first-ever ischemic stroke (n = 231; median lag time 4.9 years) and 462 matched referents from the Northern Sweden Health and Disease Study Cohort were included. Plasma bilirubin was measured and UGT1A1*28 was analyzed by fragment analysis. Results: Plasma bilirubin was lower in cases than in referents, but the difference reached significance only for women. The UGT1A1*28 polymorphism (allele frequency 30%) showed a strong gene-dose relationship with bilirubin levels both among cases and referents, but was not associated with risk for stroke. Among multiple other variables analyzed, the strongest correlation with bilirubin was found for plasma iron. Conclusions: There was no evidence for a protective effect of the UGT1A1*28 polymorphism against stroke and consequently neither for bilirubin. The findings suggest that other factors influencing the risk for stroke might also affect bilirubin levels.

Background— Bilirubin, an effective antioxidant, shows a large variation in levels between individuals and has been positively associated with reduced cardiovascular disease risk. A major reason for the variability is a common promoter polymorphism, UGT1A1*28, which reduces the transcription of the enzyme that conjugates bilirubin, UDP-glucuronosyltransferase 1A1. The aim of the study was to evaluate a possible protective effect of plasma bilirubin and the UGT1A1*28 polymorphism against myocardial infarction in a prospective case-referent setting. Methods and Results— Subjects (n=618) with a first-ever myocardial infarction (median event age, 60.5 years; median lag time, 3.5 years) and 1184 matched referents were studied. Plasma bilirubin was lower in cases versus referents. Despite a strong gene-dosage effect on bilirubin levels in both cases and referents, the UGT1A1*28 polymorphism did not influence the risk of myocardial infarction. Among multiple other variables, serum iron showed one of the strongest associations with bilirubin levels. Conclusions— We found no evidence for a protective effect of the UGT1A1*28 polymorphism against myocardial infarction and consequently neither for bilirubin. The lower bilirubin levels in cases might be caused by decreased production, increased degradation, or increased elimination.

Background: High iron levels can increase the formation of noxious oxygen radicals, which are thought to contribute to cerebrovascular disease. The aim of this prospective study was to determine if iron status and HFE genotypes constitute risk factors for stroke. Methods: First-ever stroke cases (231 ischemic and 42 hemorrhagic) and matched double referents from the population-based Northern Sweden cohorts were studied in a nested case-referent setting. Results: For total iron binding capacity, an increased risk of ischemic stroke was seen in the highest quartile (OR 1.80; 95% CI 1.14–2.83; p for trend 0.012). The highest quartile of transferrin iron saturation showed a decreased risk of ischemic stroke in men (OR 0.44; 95% CI 0.22–0.87; p for trend 0.028), but not in women. There was an increased risk of hemorrhagic stroke in the second (OR 4.07; 95% CI 1.09–15.20) and third quartile (OR 4.22; 95% CI 1.08–16.42) of ferritin. Neither quartiles of plasma iron concentrations nor the HFE C282Y and H63D genotypes were associated with ischemic or hemorrhagic stroke. Conclusions: Iron stores were not positively related to increased risk of ischemic stroke. Furthermore, HFE genotypes did not influence the risk of ischemic or hemorrhagic stroke.

The indications for continuous oral anticoagulant treatment, the target interval and the procedures for withdrawing treatment have changed in the last 10 years. Patients on continuous oral anticoagulant treatment at the Outpatient Anticoagulant Clinic at Umeå University Hospital in 2002 were included in a descriptive study (n = 900). 263 of those had a mechanical heart valve prosthesis. Only patient records for patients with other indications than mechanical heart valve prosthesis were examined. 582 of those records were found. In the 55 remaining patients some clinical information was retrieved from the computerised warfarin dosage database. These latter, more unsure clinical data, are presented separately. Anticoagulant treatment was discontinued if lack of proper indication or presence of too high risk for hemorrhagic complications were found. The prevalence of continuous oral anticoagulant treatment in the uptake area was 0.65%. The most common target interval was INR 2.1–3.0, but patients with a mechanical heart valve prosthesis were often treated more aggressively, i.e. with a higher INR target interval. Of the patients on continuous treatment, 26.6% of the INR values were outside 2.0–3.0. The most common reasons for oral anticoagulant treatment were atrial fibrillation or mechanical heart valve prosthesis, in contrast to earlier findings in studies of our population in 1987 and 1990. We found 90 patients (10.0%) without proper indication for oral anticoagulant treatment or too high risk, and their treatment was discontinued. In patients on oral anticoagulant therapy, re-evaluation of indications and risks resulted in a substantial number of treatment withdrawals. There have been major changes in treatment indications during the last decade, possibly due to rapid development of knowledge in the field of thrombosis risk factors. Treatment should be re-considered once a year.