SAN FRANCISCO— Manganese exposure is a known risk factor for parkinsonism. Welding rods contain manganese, which is vaporized during the welding process. Do the nation’s half million welders have an elevated risk for Parkinson’s disease? Much rides on the answer, including occupational health regulations and the outcome of class action lawsuits. Unfortunately, as a study presented at the 56th Annual Meeting of the American Academy of Neurology showed, finding the answer is not easy.

The study attempted to test whether the age at onset of Parkinson’s disease was affected by occupation. It was conducted by Samuel Goldman, MD, and colleagues from the Parkinson’s Institute in Sunnyvale, California. Their investigation was part of a larger effort to understand environmental risk factors for Parkinson’s disease. It was also specifically designed to explore a finding from another recent study by Brad Racette, MD, and colleagues of the Washington University School of Medicine in St. Louis, who described the clinical characteristics of 15 professional welders.

Racette and colleagues showed that Parkinson’s disease in the welders was clinically indistinguishable from idiopathic Parkinson’s disease, except for a younger age of onset: The welders had a mean age at onset of 46 versus 63 for nonwelder controls. “We speculate that the younger age at onset may be attributable to the effects of an accelerating agent from welding in a potentially ‘at-risk’ patient who might otherwise develop Parkinson’s disease with a later age at onset,” they wrote. The study has been widely used to support litigation against welding equipment manufacturers, although coauthor Joel Perlmutter, MD, has stated it is too early to draw any certain causal link between welding and Parkinson’s disease.

FACT OR ARTIFACT?

It was in this context that Dr. Goldman and colleagues set out to determine whether the welding/Parkinson’s disease association was, in Dr. Goldman’s words, “fact or artifact.” The Racette study did not indicate occupations of the controls, Dr. Goldman noted, which may have introduced some bias in interpretation of results.

“Our objective was to replicate the design of the Racette study in a population with different geographic and demographic characteristics,” he said. “If welding was truly associated with young onset parkinsonism in our population, it would support the association found by Racette and colleagues.”

Dr. Goldman and colleagues examined the medical records of 885 consecutive patients with Parkinson’s disease for whom occupation and age at diagnosis was recorded. Reflecting the largely white-collar population of the Silicon Valley, 12% of the patients were teachers, 10% were administrative workers, and 10% were engineers. Only two welders were found. Mean age at onset for the entire group was 64.3, and 14% had onset at 50 or younger. Age at onset was significantly higher for homemakers and military personnel, with mean onset ages of 68 and 72, respectively.

Age at onset was lower than the mean for “technical workers,” a group that included software developers, programmers, bioscientists, earth scientists, and electronics workers. Their mean age at onset was 58.6. This group did not include assembly workers or electricians, who might be more likely to have exposures to similar toxicants as welders. Age at onset for the two welders was near the mean for the entire cohort, Dr. Goldman noted.

“Caution must be exercised when interpreting cross-sectional data derived from a specialty clinic population,” said Dr. Goldman, including those from the current study. The problem with this type of study, he elaborated, is bias—a systematic error that results in an incorrect estimate of association between exposure and risk. He noted that the younger onset for technical workers could be explained not because there is any increased risk for these professions but because, as a whole, the technical workers of the Silicon Valley tend to be a young group. Furthermore, those who work with their hands for a living, whether at a keyboard or a machinist’s lathe, might be expected to notice symptoms that affect job performance earlier and to seek care, he said.

“Specialty clinic cross-sectional surveys may be useful for identifying putative risk factors,” Dr. Goldman remarked, “but they are not useful for determining causality. Carefully designed analytical studies are needed to test hypotheses regarding occupational factors and parkinsonism risk.”

NO CONCLUSIVE ANSWERS

Where do these results leave the question of welding and parkinsonism? According to Caroline Tanner, MD, PhD, Director of Clinical Research at the Parkinson’s Institute and senior coauthor of the Goldman study, neither the Racette study nor her own can provide any conclusive information about the risks of welding or any other occupation. Ascertainment bias is a potential confounder in both—in the Racette study because they chose to look specifically at welding and in the Goldman study because the Silicon Valley’s technical workers happen to be young. Her group is currently completing a study of occupational history in twins discordant for Parkinson’s disease, and so far no association with welding has arisen. Claims from personal injury lawsuits notwithstanding, she says, “Welding certainly isn’t a common cause of Parkinson’s disease.”

NR

—Richard Robinson

Suggested Reading
Racette BA, McGee-Minnich L, Moerlein SM, et al. Welding-related parkinsonism: clinical features, treatment, and pathophysiology. Neurology. 2001;56:8-13.

Return to table of contents