Longitudinal study examining the neurotoxicity of occupational exposure to aluminium-containing welding fumes

Objectives. The neurotoxicity of occupational exposure to aluminium (Al)-containing welding fumes has been discussed with controversial results. The aim of the longitudinal study was to examine a group of Al welders for significant central nervous changes in comparison with a non-exposed cohort. Methods. A group of 98 Al welders (mean age 37 years) in the car-body construction industry, with a median of 6 years of occupational exposure to Al welding fumes, and an education-matched, gender-matched, age-matched control group of 50 car-production workers (mean age 36 years) at the same plant, were included in this longitudinal study. Two cross-sectional studies were done in 1999 and 2001. In the second cross-sectional study 97 welders and 50 controls could be examined. The examination programme consisted, for example, of a standardised anamnesis, focussing on occupational history, education, illnesses, medication, accidents and current alcohol consumption, a physical examination that included neurological status, and the assessment of Al concentration in plasma and urine. The neurobehavioral methods included a symptom questionnaire, modified Q16, and computerised and non-computerised tests: psychomotor performance (steadiness, line tracing, aiming, tapping), verbal intelligence (WST), simple reaction time, digit span, block design (HAWIE), symbol-digit substitution, digit span, switching attention (European neurobehavioral evaluation system, EURO-NES), and standard progressive matrices. The data were analysed by multivariate analysis of covariance (MANCOVA) for repeated measurements with covariates age, education, and carbohydrate-deficient transferrin in plasma (CDT). Results. The median Al urine concentration (mean preshift/postshift) was 52.4 mug/g creatinine (2001) and 57.6 mug/g creatinine (1999). Median respirable air dust was 0.67 mg/m(3) (2001) and 0.47 mg/m(3) (1999). Welders and controls did not report significantly more symptoms in the modified Q16. Furthermore, no significant differences in psychomotor performance and other neurobehavioral tasks, except for reaction time, were seen between welders and non-welders. Regression analyses reveal a significant relationship between reaction time and Al excretion in urine that was confounded by other factors. Conclusions. At present the outcome for reaction time has to be interpreted as a single result. However, as the modified Q16 questionnaire and the rest of the psychomotor performance showed no significant changes, the next cross-sectional study, in 2003, will provide further information on which a final conclusion can be based.