Associations Of Multiple Urinary Metal Concentrations With Arterial Stiffness In Adults

Background:Arteriosclerosis is a disease that thickens arterial vessel wall until it loses elasticity and narrows.Recently,studies have shown arterial stiffness may be linked to exposure to the environmental pollutants such as particulate matters,gaseous pollutants or heavy metals,in addition to the traditional factors such as lack of exercise,smoking,unhealthy eating habits,diabetes,hypertension or hyperlipidemia.Evidence from animal experiments suggest that exposure to Ti O₂,arsenic,and iron result in atherosclerosis in mice,and a few epidemiological studies have investigated the associations of long-term exposure to single metal with arteriosclerosis.Although a growing body of studies have shown that the associations of some metals with cardiovascular disease,but conclusions are still controversial and most studies only use single measurements of metals in spot urine samples as biomarkers.Objective:The associations of multiple metal concentrations with arterial stiffness are analyzed by cross-sectional in summer（n=127）,autumn（n=114）,winter（n=104）and panel methods in three seasons（n=345）,and we estimate the possible dose-response relationships using generalized addition model.Methods:We recruited 143 adults in the Baofeng community of Qiaokou district and Dijiao community of Jiang’an district in Wuhan.We invited them to complete the questionnaires and peform physical examinations.In addition,morning urine samples were conducted once a day for 4 consecutive days of each season.Urinary concentrations of 23 metals were detected by using inductively coupled plasma-mass spectrometry（ICP-MS）,and ankle brachial Index（ABI）and ankle brachial pulse wave velocity（ba PWV）were measured by automatic arteriosclerosis detector.A total of 127 adults free of serious health conditions such as coronary heart disease,stroke,cancer,and missing values of arteriosclerosis were included for statistical analysis.We used mixed linear models to analyze repeated measurements of urinary metal concentrations and arteriosclerosis,then use false discovery rate（FDR）multiple correction to select the metal whose P value<0.05.Finally,we used the generalized addition model to explore the dose-response relationships between the metals and ABI or ba PWV.Results:Cross-sectional study:In the single metal models,in summer:the levels of arsenic（As）at lag 1 day was positively associated with ABI,the levels of vanadium（V）,copper（Cu）,barium（Ba）,lead（Pb）at lag 3 day and 4-day average aluminum（Al）,manganese（Mn）was nagatively associted with ABI;In autumn:the levels of V at lag 0day was positively associated with ABI,the levels of molybdenum（Mo）at lag 0 day,the levels of selenium（Se）,Mo,Tin（Sn）at lag 1 day,the levels of strontium（Sr）at lag 3day,4-day average cobalt（Co）,Se was nagatively associted with ABI;In winter:the levels of manganese（Mg）,iron（Fe）,Ba,uranium（U）at lag 2 day,U at lag 3 day,4-day average Al was positively associated with ABI,the levels of titanium（Ti）and V at lag 0day,the levels of Ti,Co and Sn at lag 1 day,the levels of Ti,Se and U at lag 3 day was nagatively associted with ABI.The levels of Ti at 4-day average in winter were significantly associated with ABI（FDR<0.05）after adjusting for arteriosclerotic risk factors.Each 1%increase in Ti was associated with 0.051%（95%CI:0.082%,0.020%）decreases in ABI,but we found the dose-response relationships between Ti and ABI（P=0.015）,but no metal was associated with ba PWV.Panel study:In the single metal models,the levels of Ti,Co,Se,Mo,Sn at lag 0 day,the levels of Ti,Co,Sn at lag 1 day,the lavels of Co at lag 2 day,the levels of Ti,V,Co,Cu,zinc（Zn）,As,Se,rubidium（Rb）,Sr,Mo,cadmium（Cd）and Sn was nagatively associted with ABI.The levels of Ti,cobalt（Co）,copper（Cu）,selenium（Se）,molybdemum（Mo）at lag 3 day and the levels of Ti,Co at 4-day average were significantly associated with ABI（FDR<0.05）after adjusting for arteriosclerotic risk factors.In the multiple metal model,an 1%increase in Ti was associated with 0.019%（95%CI:0.038%,0.001%）decrease in ABI,and an 1%increase in Co was associated with 0.012%（95%CI:0.024%,0.001%）decrease in ABI,and we found the dose-response relationships between Ti and ABI（P=0.016）,but no metal was associated with ba PWV.Conclusions:To the best of our knowledge,this is the first panel study to report the effects of multiple metals on arterial stiffness among a community-based population.Our repeated measure analyses showed that urinary Ti and Co were significantly related to decreased ABI in both single and multiple metal models,and cross-sectional analyses by seasons further revealed that urinary Ti was inversely associated with ABI in the winter.This study provided a scientific basis for the early prevention of arteriosclerosis and even cardiovascular disease,but additional large and multiple center cohorts were required to confirm these findings.