C:\Users\UNIVERSA MEDICINA\Docu 63 ABSTRACT UNIVERSA MEDICINA Parkinson's disease and the welding job: are both related? Iqbal Mochtar* and Muchtaruddin Mansyur** Up to the present, there has been a controversy on the relationship between Parkinson’s disease and the welding job. The aim of this study was to obtain evidence-based information regarding the relationship between Parkinson’s disease and the welding job through an evidence-based case report derived from a literature review. The review was conducted through a method of search and selection of articles in the Pubmed, Cochrane Library and JSTOR databases aimed at answering the study question. The process of searching articles used the keywords “Welding” OR “Welder” AND “Parkinson”. Article selection was performed using the defined inclusion and exclusion criteria. At the initial search, 117 articles were retrieved from the three databases. Following the selection process, three articles remained, which consisted of one systematic review and two observational studies. Comparing the selected articles, the systematic review of Mortimer et al. is more relevant and appropriate for answering the clinical question. Mortimer et al. used a meta-analytical method, applied strict inclusion and exclusion criteria and excluded studies that potentially led to bias effects, lack of validity or inadequate statistical methods. Based on the selected evidence-based resources, Parkinson’s disease is not related to the welding job. The welding job, however, may produce clinical symptoms and signs resembling Parkinson’s disease, known as manganism. Keywords: Welding; Parkinson’s disease; evidence-based; case report CASE REPORT pISSN: 1907-3062 / eISSN: 2407-2230 DOI: http://dx.doi.org/10.18051/UnivMed.2020.v39.63-70 January-April, 2020 Vol.39- No.1 Cite this article as: Mochtar I, Mansyur M. Parkinson's disease and the welding job: are both related?. Univ Med 2020; 39:6 3 -7 0 . doi: 10 .18 051/ Uni vMed. 2020.v39.63-70 *Division of Occupational Medicine, Department of Community Medicine, Universitas Indonesia, Indonesia **Department of Community Medicine, Universitas Indonesia, Indonesia Correspondence: Iqbal Mochtar Division of Occupational Medicine, Department of Community Medicine, Universitas Indonesia, Indonesia Phone: 0974-66554913 Email: muhammad.mochtar03@alumni. imperial.ac.uk ORCID ID: 0000-0002-1057-3916 Date of first submission, June 7, 2019 Date of final revised submission, March 8, 2020 Date of acceptance, March 16, 2020 This open access article is distributed under a Creative Commons Attribu- tion-Non Commercial-Share Alike 4.0 International License 64 Mochtar, Mansyur Parkinson's disease and the welding job INTRODUCTION Parkinson’s disease (PD) is the second most common neurodegene rative disease affecting 6.1 million people worldwide in 2016.(1) The pathological basis of the disease is related to the aggregation of the alpha-synuclein protein of Lewy bodies and the loss of pigmented mel a nin in t he mi d br a in, le a d ing t o th e degeneration of dopaminergic neurons in the substantia nigra and dopamine insufficiency in the striatum.(2) It is the most common cause of parkinsonism, a syndrome characterized by postural instability, tremor, bradykinesia and rigidity.(3) Pe s t i c i de e xp os u r e , da ir y p r o du c t consumption, and brain trauma have been associated with an increased risk of PD, while smoking, caffeine intake and exercise have been linked with reduced risk of the disease. (4,5) Some studies linked PD with the welding job, due to manganese exposure in the welding works. Those studies identified neuropsychomotor symptoms and manganese deposition on MRI in manganese-exposed workers. Other studies refuted the finding and argued that PD was not related to welding works. It still remains c o nt r ove rs i a l wh e t h e r we l d er s h a ve a n i nc r e a s e d r i sk of PD a nd whe t he r t h e neuropsychomotor symptoms observed in manganese-exposed workers are related to PD.(6) The aim of this evidence-based case report was to obtain evidence-based information regarding the relationship between Parkinson’s disease and the welding job through an evidence- based case report derived from a literature review. Case Illustration A 55-year-old male Qatari patient came with resting tremor in both hands, r igid movement, slow and unstable gait since two yea rs ago. Due to the symptoms, he had difficulty in doing work and daily activities. The initial symptom was only tremor in the left hand, which was mild in quality and on-off. Later, the tremor occurred in both hands and was more severe and frequent. His general movement was bradykinesia without slurred speech. He was a non-smoker and had controlled diabetes and hypertension. He had been working as a welding inspector in an oil and gas company for the last 12 years. Previously, he worked as a welding technician for 8 years. His latest work was to supervise welding activities, to test welding materials, to inspect and control welding equipment and consumables such as filler wire, fluxes and electrodes. He had under gone treatment in some hospitals, including one in Germany. The final diagnosis was idiopathic PD and he had been given some medications, including co-beneldopa and co-careldopa. Following se ver al mont hs of tre atment , the worker observed some improvement in his symptoms. The worker believed that his condition was due to his job and therefore he planned to claim financial compensation. The clinical question for this case was: in the workers, is the welding job, compared to the non-welding jobs, associated with PD? Evidence Following the search of articles using the keywords (Table 1), the selection was performed using the inclusion and exclusion criteria (Figure 1). The inclusion criteria included human studies, publications within the last 10 years, in the English language, using adult subjects, and relevant to the clinical question. Exclusion criteria included pathophysiological, phar macological and radiological studies as well as opinion papers. Applying the inclusion and exclusion criteria, the article list was shortened, and eventually only three articles remained for appraisal, consisting of one systematic review (Mortimer et al.(9)) and two observational studies (Racette et al.(10) and Racette et al.(11) The selecte d ar ticl es were critic ally appraised using the appraisal tools f or a systematic review (7) and for cross-sectional and 65 prospective cohort studies.(8) The results of the appraisal are presented in Tables 2, 3 and 4. Mortimer et al.(9) performed a meta-analysis a n d s c r u t ini ze d 10 5 s t ud i e s to se e k t he association of the welding job and manganese exposure with PD.(9) The authors analyzed two main questions, namely the association between the welding job and PD and the association between manganese exposure and PD. The authors demonstrated that the pooled relative risks of the welding job to PD disease was 0.86 (95% CI: 0.80-0.92) and that of manganese exposure to PD was 0.76 (95% CI: 0.41-1.42). The authors concluded that both the welding job and manganese exposure were not related to an increased risk of PD. Racette et al.(10) compared the prevalence of Parkinsonism in welding-exposed workers to the reference group. The study utilized clinical assessment including the use of the Unified Parkinson’s Disease Rating Motor Scale 3 (UPDRS3) to categorize the subjects into Parkinsonism (score >15), intermediate (score 6- 14) and normal (score <6). The exposure assessment was performed using a questionnaire. The prevalence of Parkinsonism in the welding Figure 1. The process of article selection Univ Med Vol. 38 No.1 66 Mochtar, Mansyur Parkinson's disease and the welding job job was 15.6% and in the non-welding jobs 0%. The study concluded that welding-exposed workers had a high prevalence of Parkinsonism compared to the non-exposed workers. In anothe r ar ticle , Racette et al. (1 1) assessed the progression of Parkinsonism in the manganese-exposed welders in three worksites. They measured the clinical assessment of 886 workers using UPDRS3 questionnaire and clinical evaluation while the exposure assessment was measured by questionnaire. The authors followed- up the patients, with a mean follow up of 4.2 years and revealed that in all examined workers, there was a change of 0.24 (0.10-0.38) in UPDRS3 score. The study concluded that the progression of Parkinsonism increased with increased cumulative manganese exposure. Up to now, there has been a controversy on whether welding-exposed workers are associated with PD. Some studies supported and other studies refuted the association. Park pointed out that one of the factors that induced these conflicting findings were the pitfalls in various methodological issues in the studies, including a healthy worker effect bias, statistical modeling and the use of litigation studies. In addition, a number of studies also did not carefully consider confounder factors in manganese work such as the exposure to carbon-monoxide, aluminum and heat stress.(6) Couper fir st reported the association b e t we e n ma n ga nes e e x po s ur e a n d neurobehavioral symptoms in 1873.(12) The report was the first to mention the toxicity of manganese in humans, which is now known as manganism. Manganism is a neurobehavioral syndrome due to an excessive accumulation of manga nese in t he brain fr om manganese exposure. Manganism has clinical features resembling PD, and therefore many physicians mi sd i a gn ose ma n ga n i s m a s PD. In f a c t , manganism and PD are two different entities wi t h d i f f er e n t c l i n i c a l a nd pat h o lo gi c a l ma ni f e s ta ti o ns . ( 1 3 ) T he di s t i ngui s h in g characteristics of manganism are the absence of resting tremor, the appearance of gait disorder as an initial symptom and the lack of response to L-dopa. Also, in contrast to PD, neuroimaging in patients with manganism does not reveal any degeneration of dopaminergic neurons in the substantia nigra and dopamine insufficiency in the striatum.(14) Comparing the three selected articles, the study of Mortimer et al.(9) is more relevant and appropriate for answering the clinical question. First, Mortimer used a meta-analytical method in scrutinizing relevant studies. The authors applied strict inclusion and exclusion criteria and excluded studies that potentially led to bias effects, lacked validity, and inadequate statistical method. The authors also calculated the heterogeneity of study effects to argue that the existing variation in the studies was due to chance. Unlike Mortimer, Racette et al.(10) supported the relationship between welding or manganese exposure and Parkinsonism. In their article, the authors showed an increased prevalence of Parkinsonism in welding-exposed workers compared to the reference population (15% vs. 0%). However, Racette’s study has some potential bias, mainly due to the use of the questionnaire- based metric for exposure calculation. Recall bias might occur because the subjects might provide inadequate information when estimating the type, quantity and the duration of exposure. A selection bias might also occur because the authors utilized the reference group sample provided by the trade union without clear subject criteria. In another article Racette et al.(11) followed up the manganese-exposed workers to observe the progression of Parkinsonism. During a mean follow-up period of 4.2 years, the authors found that in the manganese-exposed workers there A B C Database Keyword Hits Filter PubMed “Welding” OR “welder” AND “Parkinson” 26 Title/abstract Cochrane Library “Welding” OR “welder” AND “Parkinson” 88 Title/abstract JSTOR “Welding” OR “welder” AND “Parkinson” 3 Title/abstract Table 1. Search strategy using keywords 67 Article Mortimer JA, et al. Association of welding and manganese exposure with Parkinson’s disease – review and meta-analysis. (9) Level 1 (Systematic review of clinical trials, cohort, case-control studies) (7) Questions Does the systematic review address a focused question (PICO) and use it to direct the search and select articles for inclusion? Yes The authors analyzed the association between welding and PD and the association between manganese exposure and PD. The authors applied the focused questions to select the most relevant articles. Find Did the research find all the relevant evidence? Yes The authors searched PubMed and Cochrane databases, published reviews and abstracts of scientific meetings. Appraise Have the studies been critically appraised and was the overall quality adequate? Yes The authors indicated some appraisal criteria for the focused questions, such as types of exposure, outcomes, and number of person-years followed up. Synthesize Have the results been synthesized with appropriate summary tables and plot? Were the results similar between studies? Yes Yes The authors listed the details of each study in the appropriate tables and plot/pooled relative risk. The results of the studies were similar, in that the relative risk (RR) for both welding and manganese exposure to PD were less than 1. Result What measure was used, how large was the effect (could it have been due to chance?) Uncertain The RR between welding and PD was 0.86 (0.80-0.92) and between manganese exposure and PD 0.76 (0.41-1.42). The association between welding and PD lacked heterogeneity, indicating that the different results of the reviewed studies were due to chance. Table 2. Critical appraisal checklist for systematic review was a progression of Parkinsonism evidenced by a 0.32 (SD 2.1) change in UPDRS3 score. However, this article also has potential bias, such as information, recall and selection bias, mainly due to the use of questionnaire findings that were converted into weighted exposure years. In relation to the patient concerned in this case illustration, there are some reasons to argue that his PD is not related to his job. First, aging – not manganese exposure - is the largest contributing factor for PD; PD affects 1% of the population above 60 years and 5% of the population over 85 years. Around 5% of PD, however, occurs at the age below 60 years and this is associated with mutations in a number of genes such as PINK-1, Parkin, DJ-1 and alpha- nuclein.(15) Second, PD is a multifactorial disease, where various risk factors may play a role in the occurrence of the disease. The patient has a number of risk factors, such as diabetes and hypertension, which may play a role in the onset of his PD. Third, based on current and relevant evidence-based medicine publications available, including meta-analytical studies, no increased PD risk is observed in welders or welding-exposed workers.(9) Univ Med Vol. 38 No.1 68 Mochtar, Mansyur Parkinson's disease and the welding job Article Racette BA, et all. Increased risk of Parkinsonism with welding exposure. (10) Level Level 2 c (Case-control study) (8) Did the study address a clearly focused question? Yes To determine the prevalence and phenotype of Parkinsonism in shipyard and fabrication welders and compare these with non-exposure subjects comprising PD patients from Movement Disorders Center, Washington. Is the research method (study design) appropriate for answering clinical questions? Yes The study type (cross-sectional and nested case- control) is appropriate for the prevalence estimation. Is the method of selection of the subjects (employees, team, division, organization) clearly described? Uncertain The authors did not mention the reason for the recruitment from the places. Could the way the sample was obtained introduce (selection) bias? Yes In the study subjects, the authors took the list of welders and contacted them for participation. In the reference group 1, the subjects were recruited by the local trade union and the authors were not given the list of the workers. In the reference group 2, the subjects were taken from the clinical database of newly diagnosed untreated PD at Movement Disorders Center, without any subject selection criteria. Was the sample of subjects representative with regard to the population to which the findings will be referred to? Yes for the study group but not for the reference groups In the study group, the subjects represented the population to which the finding referred. In the reference groups 1 and 2, the recruited samples did not represent the reference population. Was the sample size based on pre- study considerations of statistical power? No No predetermined sample size. Was a satisfactory response rate achieved? Yes 811 workers out of 1,612 welding-exposure workers (50.3%) agreed to participate. Were the measurements (questionnaires) likely to be valid and reliable? For the clinical diagnosis they were valid and reliable but for the exposure assessment, they lacked validity and reliability. Clinical diagnosis of Parkinsonism was based on a neurological examination that included UPDRS3 questionnairre. The clinical assessment is considered valid and reliable. Exposure assessment utilized welding exposure questionnaire. Based on the questionnaire, weighted exposure-years was calculated for each participant. Table 3. Critical appraisal checklist for a cross-sectional study 69 Table 4. Critical appraisal checklist for a prospective cohort study Article Racette BA, et all. Dose-dependent progression of Parkinsonism in manganese-exposed welders. (11) Level Prospective cohort (2b) (8) Are the results of the study valid? Was the defined representative samples of patients assembled at a common (usually early) point in the course of their diseases? Yes Baseline examination on the study subjects was conducted from January 2006 to September 2013. Was patient follow up sufficiently long and complete? Yes Each subject had a different period or number of follow up. 61% had 1 follow-up, 26% had 2 follow-ups, 12% had 3 follow-ups and 1% had 4-5 follow-ups. Mean time period between baseline and any follow-up was 3.9 (1.8) years. Were outcomes criteria either objective or applied blindly? Yes The primary outcomes were the rates of changes of UPDRS3 score. The secondary outcomes were the annual rates of changes of UPDRS3 sub-score. The exposure assessment was blinded to the neurologists who examined the subjects. If subgroups with different prognosis are identified, did an adjustment for important prognostic factors take place? Yes The authors adjusted the prognostic factors; e.g. duration of exposure to baseline examination was divided into <5 years and ± 5 years. What are the results? How likely are the outcomes over time? Yes The authors provided the estimation of expected additional increase of UPDRS3 score over a 10-year time period How precise are the prognostic estimates? Uncertain The precision of prognostic value can be based on the confidence interval. The confidence intervals of findings were mixed, some of them were narrow and the rest were wide. Can I apply this valid, important evidence about prognosis to my patient? No The study subjects were from the shipyard and fabrication unit while our patient was from a gas and petroleum company. The race also matters, since the study subjects were non-Hispanic Caucasian and our patient was Arabic. Our patient had consumed PD medication and therefore the estimation of PD deterioration using a time-based estimation cannot be applied. LEARNING POINTS Based on the selected evidence-based resources, it is concluded that: (i) Parkinson’ disease is not related to the welding job; (ii) The welding job, however, may produce clinical symptoms and signs resembling PD, known as manganism. Univ Med Vol. 38 No.1 70 Mochtar, Mansyur Parkinson's disease and the welding job CONFLICTS OF INTEREST None. ACKNOWLEDGEMENTS We would like to thank the staff of the Department of Community Medicine, Universitas Indonesia for the support for this publication. CONTRIBUTORS IM contributed to conceptualization and methodology. IM and MM contributed to preparing and writing the original draft. MM contributed to supervision. IM contributed to review and editing. All authors have read and approved the final manuscript. REFERENCES 1. Rocca WA. The burden of Parkinson’s disease: a worldwide perspective. Lancet Neurol 2018;17: 928-9. doi: 10.1016/S1474-4422(18)30355-7. 2. Oertel WH. Recent advances in treating Parkinson’s disease. F1000Res 2017;6:260. doi: 10.12688/f1000research.10100.1. 3. DeMaagd G, Philip A. Parkionson’s disease and its management. PT 2015;40:504-32. 4. Ascherio A, Schwarzschild MA. T he epidemiology of Parkinson’s disease: risk factors and prevention. Lancet Neurol 2016;15:1257-72. DOI: https://doi.org/10.1016/S1474-4422(16) 30230-7. 5. Xu Y, Yang J, Shang H. Meta-analysis of risk factors for Parkinson’s disease dementia. Transl Neurodegener 2016;5:1-8. 6. Park RM. Neurobehavioral deficits and Parkinsonism in occupations with manganese exposure: a review of methodological issues in the epidemiological literature. Saf Health Work 2013;4:123-35. 7. Glasziou PP, Mar CD, Salisbury J. Evidence-Based Practice Workbook. 2nd ed. London: Blackwell Publishing - BMJ Books; 2007. 8. Oxford Centre for Evidence Based Medicine (CEBM). Critical Appraisal Tools. Oxford;2019. 9. Mortimer JA, Borenstein AR, Nelson LM. Associations of welding and manganese exposure with Parkinson’s disease: review and meta-analysis. Neurology 2012;79:1174-80. 10. Racette BA, Criswell SR, Lundin JI, et al. Increased risk of Parkinsonism with welding exposure. Neurotoxicology 2012;33:1356-61. 11. Racette BA, Nielsen SS, Criswell SR, et al. Dose- dependent progression of Parkinsonism in manganese-exposed welders. Neurology 2017;88: 344-51. 12. Couper J. On the effects of black oxide of manganese when inhaled in the lungs. Br Ann Med Pharmacol 1873;1:41–2. Cited by: Blanc PD. T he early history of manganese and the recognition of its neurotoxicity, 1837-1936. Neurotoxicology 2018;64:5-11. doi: 10.1016/ j.neuro.2017.04.006. 13. Kwakye GF, Paoliello MMB, Mukhopadhyay S, et al. Manganese-induced Parkinsonism and Parkinson’s disease: shared and distinguishable features. Int J Environ Res Public Health 2015;12: 7519-40. 14. Guilarte TR. Manganese and Parkinson’s disease: a critical review and new findings. Environ Health Perspect 2010;118:1071-81. 15. Reeve A, Simcox E, Turnbull D. Ageing and Parkinson’s disease: why is advancing age the biggest risk factor? Ageing Res Rev 2014;14:19- 30.