Longitudinal study on potential neurotoxic effects of aluminium: II. Assessment of exposure and neurobehavioral performance of Al welders in the automobile industry over 4 years

Objectives This is the second of two parallel longitudinal studies investigating Al exposure and neurobehavioral health of Al welders over 4 years. While the first published study in the trail and truck construction industry examined the neurobehavioral development of Al welders from age 41-45 in the group mean (Kiesswetter et al. in Int Arch Occup Environ Health 81: 41-67, 2007), the present study in the automobile industry followed the development from 35 to 39. Although no conspicuous neurobehavioral developments were detected in the first study, which furthermore exhibited the higher exposure, it cannot be excluded that exposure effects appear in earlier life and exposure stages. Methods The longitudinal study is based on a repeated measurement design comprising 4 years with three measurements in 2 years intervals. 92 male Al welders in the automobile industry were compared with 50 non-exposed construction workers of the same industry and of similar age. The repeated measurements included total dust in air, and Al pre- and post-shift plasma and urine samples. Neurobehavioral methods comprised symptoms, verbal intelligence, logic thinking, psychomotor behavior, memory, and attention. The computer aided tests came from the Motor Performance Series and the European Neurobehavioral Evaluation System. The courses of neurobehavioral changes were analyzed with multivariate covariance-analytical methods considering the covariates age, indicators of 'a priori' intelligence differences ( education or markers of 'premorbid' intelligence), and alcohol consumption (carbohydrate-defficient transferrin in plasma). Additionally, the interrelationship, reliability and validity of biomonitoring measures were examined. Results The mean environmental dust load during welding, 0.5-0.8 mg/m(3), and the mean internal load of the welders (pre-shift: 23-43 mu g Al/g creatinine in urine; 5-9 mu g Al/l plasma) were significantly lower than in the parallel study. Under low exposure, the stability of biomonitoring measures was reduced, but the Al load differed significantly between Al welders and referents. It could not be shown that the development of neurobehavioral performances over the 4-year period differed between both groups. Mainly, markers of 'premorbid' intelligence and age were related to neurobehavioral performance differences but not Al exposure. Conclusions The biomonitoring and neurobehavioral results are in line with the results of the first published study. The repeated measurement models of both studies showed no adverse neurobehavioral effects of Al welding. A modular lifetime-oriented research concept is outlined aiming at the investigation of sequential periods of exposure life with special focus on the biologically most sensitive phases like first exposure and old age.