Determination of reference intervals of oxidative stress biomarkers in healthy Italian population

Background. Recent interest to investigation of oxidative stress (OS) linked disorders is due to its high involvement in many chronic and acute pathologies: cardiovascular, lung, hematological, diabetes, cancer and many others. Oxidative stress is an imbalance between production of reactive oxygen species (ROS) and antioxidant defensive capacity of organism. Normally ROS are generated in human body and are considered to be natural byproducts of metabolism of oxygen. But in the framework of oxidative stress, ROS suppress antioxidant capacity in vivo, damaging DNA, lipids, and proteins. ROS are highly reactive molecules with a short half-life and therefore it is very difficult to detect them in human biological fluids. Instead, the products of their oxidizing reactions, denoted as biomarkers of oxidative stress, are used widely to detect the level of OS in a human body. There are statistical techniques which allow estimating non-adjusted for determinants or adjusted for determinants reference intervals of biomarkers of oxidative stress. Aim. In this work we aimed at identifying main demographic and laboratory determinants of urinary biomarkers of oxidative stress: DNA-derived 8-oxodG and lipid membrane-derived 8-isoprostane, and at defining their reference intervals, adjusted for main determinants, in a sample of healthy people from the general Italian population. Methods. In current study, the data on 281 subjects from general Italian population, gathered during the Gene Environment Interactions in Respiratory Diseases (GEIRD) project who were never-and ex- (not smoking over the last year) smokers and who did not report respiratory symptoms at the screening questionnaire and at the clinical survey, either other comorbidities (heart disease, ictus, high blood pressure, diabetes and cancer) at the clinical interview and with normal lung function and allergologic test have been used. Results. Non adjusted reference intervals of 8-oxodG and of 8-isoprostane were evaluated first in this work. The main determinants of both biomarkers of oxidative stress were 'distance from collection' (DFC, the period from the moment of urine collection and its laboratory processing), and season, the period of year when urine was collected and split into warm (April - September) and cold (October - March). The reference intervals of 8-oxodG and of 8-isoprostane stratified by season and adjusted for DFC were predicted using GAMLSS (generalized additive models for location, scale and shape) regression analysis and showed slight but statistically significant degradation of both biomarkers with increase of DFC in both seasons, except 8-oxodG biomarker during the warm season, which provided unchanged values with increased DFC. Conclusion. To our knowledge it is for the first time when it was shown that both OS biomarkers 8-oxodG and 8-isoprostane should be evaluated in association with DFC and season, when urine has been collected. It is especially important in large epidemiological studies when long-term conservation of urine is stipulated. (Semi)parametric GAMLSS regression analysis is a new useful technique that can be used for estimating reference intervals of urinary biomarkers (8-oxodG and 8-isoprostane) from general adult population and adjusted for appropriate determinants.