Search

Abstract Purpose Recent studies have suggested that higher postmenopausal follicle stimulating hormone (FSH) may be associated with lower risk of diabetes. However, relatively little is known about postmenopausal FSH levels, including the level of variation between women and whether reproductive factors are associated with this variation. Methods We assessed the relationship of multiple reproductive factors with FSH levels among 588 postmenopausal women in the Kuopio Ischaemic Heart Disease Risk Factor Study. Participants were aged 53 to 73 years and not using hormone therapy at study enrollment (1998–2001) when reproductive factors were assessed and FSH was measured. Results After adjustment for age, menopause timing, sex steroid levels, adiposity and behavioral factors, we observed numbers of pregnancies and age at first birth were each inversely associated with FSH levels. For example, women with ≥ 3 births and an age at first birth ≥ 25 years had mean FSH levels that were 7.8 IU/L lower than those of women with 1–2 births and an age at first birth ≤ 24 years (P = 0.003). Number of miscarriages was inversely associated with FSH levels (-2.7 IU/L per miscarriage; P = 0.02). Women reporting 4 or more years of past hormone therapy use had significantly higher mean FSH levels than women who had never used hormone therapy (P for trend = 0.006). Conclusion Multiple reproductive factors were associated with postmenopausal FSH, independent of estradiol, adiposity and other confounders. These findings warrant replication and further exploration of potential underlying mechanism.

Abstract Background Lower urinary sodium concentrations (UNa) may be a biomarker for poor prognosis in chronic heart failure (HF). However, no data exist to determine its prognostic association over the long-term. We investigated whether UNa predicted major adverse coronary events (MACE) and all-cause mortality over 28–33 years. Methods One hundred and eighty men with chronic HF from the Kuopio Ischaemic Heart Disease Risk Factor Study (KIHD) were included. Baseline data was collected between 1984 and 1989. MACE and all-cause outcomes were obtained using hospital linkage data (1984–2017) with a follow-up of 28–33 years. Cox proportional hazards models were generated using 24-h UNa tertiles at baseline (1 ≤ 173 mmol/day; 2 = 173-229 mmol/day; 3 = 230-491 mmol/day) as a predictor of time-to-MACE outcomes, adjusted for relevant covariates. Results Overall, 63% and 83% of participants (n = 114 and n = 150) had a MACE event (median 10 years) and all-cause mortality event (median 19 years), respectively. On multivariable Cox Model, relative to the lowest UNa tertile, no significant difference was noted in MACE outcome for individuals in tertiles 2 and 3 with events rates of 28% (HR:0.72; 95% CI: 0.46–1.12) and 21% (HR 0.79; 95% CI: 0.5–1.25) respectively.. Relative to the lowest UNa tertile, those in tertile 2 and 3 were 39% (HR: 0.61; 95% CIs: 0.41, 0.91) and 10% (HR: 0.90; 95% CIs: 0.62, 1.33) less likely to experience to experience all-cause mortality. The multivariable Cox model had acceptable prediction precision (Harrell's C concordance measure 0.72). Conclusion UNa was a significant predictor of all-cause mortality but not MACE outcomes over 28–33 years with 173–229 mmol/day appearing to be the optimal level. UNa may represent an emerging long-term prognostic biomarker that warrants further investigation.

Abstract Background Iron stores, estimated as ferritin levels, and type 2 diabetes (T2D) have been associated previously, while findings regarding coronary heart disease (CHD) and cerebrovascular disease (CEVD) are still inconclusive. No study has focused on simultaneous evaluation of associations between iron stores and the above cardiometabolic diseases (CMD) in the same population. We aim to evaluate the association between serum ferritin and risk of T2D, CHD and CEVD in Scottish population over a wide range of ferritin levels. Methods Longitudinal study in 6,497 participants of the 1995 and 1998 Scottish health surveys, who were followed-up until 2011. Cox regression models were conducted adjusting for age, sex/menopausal status, fibrinogen, GGT levels, smoking, alcohol consumption, total cholesterol, HDL-cholesterol, blood pressure, and BMI. Ferritin was used as continuous (sex/menopausal status-specific Z score) and categorical variable (sex/menopausal status-specific quartiles, quintiles and sextiles). Results During follow-up, 4.9% of the participants developed T2D, 5.3% CHD, and 2.3% CEVD. By using ferritin quartiles, serum ferritin was positively associated with T2D, CHD and CEVD but only the association with T2D remained after adjustment for covariates [Quartile 4 v. 1: adjusted HR 95% CI 1.59 (1.10–2.34); P = 0.006]. When ferritin sextiles were used (6 v. 1), the ferritin-CEVD association became slightly stronger and significant [adjusted HR 95% CI 2.08 (1.09–3.94); P = 0.024]. Conclusions Iron stores relate differently to each CMD. Serum ferritin levels were positively and independently associated with incident T2D, and with incident CEVD if higher cut-off points for high ferritin levels were considered.

Abstract Background A Finnish joint research effort Kuopio Birth Cohort (KuBiCo) seeks to evaluate the effects of genetics, epigenetics and different risk factors (medication, nutrition, lifestyle factors and environmental aspects) during pregnancy on the somatic and psychological health status of the mother and the child. Methods KuBiCo will ultimately include information on 10,000 mother-child pairs who have given their informed consent to participate in this cohort. Identification of foetal health risk factors that can potentially later manifest as disease requires a repository of relevant biological samples and a flexible open up-to-date data handling system to register, store and analyse biological, clinical and questionnaire-based data. KuBiCo includes coded questionnaire-based maternal background data gathered before, during and after the pregnancy and bio-banking of maternal and foetal samples that will be stored in deep freezers. Data from the questionnaires and biological samples will be collected into one electronic database. KuBiCo consists of several work packages which are complementary to each other: Maternal, foetal and placental metabolism and omics; Paediatrics; Mental wellbeing; Prenatal period and delivery; Analgesics and anaesthetics during peripartum period; Environmental effects; Nutrition; and Research ethics. Discussion This report describes the set-up of the KuBiCo and descriptive analysis from 3532 parturients on response frequencies and feedback to KuBiCo questionnaires gathered from June 2012 to April 2016. Additionally, we describe basic demographic data of the participants (n = 1172). Based on the comparison of demographic data between official national statistics and our descriptive analysis, KuBiCo represents a cross-section of Finnish pregnant women.

Abstract Background The redundancy of information is becoming a critical issue for epidemiologists. High-dimensional datasets require new effective variable selection methods to be developed. This study implements an advanced evolutionary variable selection method which is applied for cardiovascular predictive modeling. The epidemiological follow-up study KIHD (Kuopio Ischemic Heart Disease Risk Factor Study) was used to compare the designed variable selection method based on an evolutionary search with conventional stepwise selection. The sample contains in total 433 predictor variables and a response variable indicating incidents of cardiovascular diseases for 1465 study subjects. Results The effectiveness of variable selection methods was investigated in combination with two models: Generalized Linear Logistic Regression and Support Vector Machine. We managed to decrease the number of variables from 433 to 38 and save the predictive ability of the models used. Their performance was evaluated with an F-score metric. At most, we gained 65.6% and 67.4% of the F-score before and after variable selection respectively. All the results were averaged over 5-folds of a cross-validation procedure. Conclusions The presented evolutionary variable selection method allows a reduced set of variables to be chosen which are relevant to predicting cardiovascular diseases. A reference list of the most meaningful variables is introduced to be used as a basis for new epidemiological studies. In general, the multicollinearity of variables enables different combinations of predictors to be used and the same performance of models to be attained.