The pathogenetic relationship of welding-related Parkinsonism (WP) and idiopathic Parkinson's disease (PD) is a matter of debate. Whereas some reports have proposed that welding is a risk factor for idiopathic PD,1, 2 others found that the incidence of PD, defined by the British Brain bank criteria,3 is not increased in welders.4–6 WP has been suggested to be caused by neurotoxicity of aerosolized manganese, an elementary metal that is released in welding plumes.1, 7 Parkinsonism can occur after chronic exposure to high levels of manganese, usually above the permissible exposure limit ceiling at 5 mg/m3 total dust set by the Occupational Safety and Health Administration.8, 9 Manganese-induced Parkinsonism has been reported in Chilean miners,10 but also in welders where they have worked in confined, unventilated spaces.7, 11
Transcranial brain sonography (TCS) is a neuroimaging method displaying tissue echogenicity of the brain parenchyma through the intact skull. In more than 90% of PD patients, TCS reveals characteristic SN hyperechogenicity, which is stable during the course of the disease and is thought to reflect increased amounts of iron, bound to proteins other than ferritin, but not the progressive neurodegeneration in the SN.12–14 Marked SN hyperechogenicity is also found in 10% of healthy adults aged between 20 and 80, and has been related to a functional impairment of the nigrostriatal dopaminergic system.15, 16 Therefore, SN hyperechogenicity has been proposed to indicate an increased risk for developing PD.12, 15, 16 On the other hand, in atypical parkinsonian syndromes typically a normal SN echogenicity is found, discriminating them highly specifically from idiopathic PD.14 Here, our aim was to find out whether TCS findings differ in WP and PD.
SUBJECTS AND METHODS
We studied 2 WP patients in whom history and clinical findings were suggestive of manganese-induced Parkinsonism (Table 1).
Table 1. Demographic and clinical data of the WP patients
WP Patient 1 was a 38-year-old man, who had worked for 12 years as a professional welder in a Chilean mine and in smelting plants for ∼12 hours daily. He reported that he worked in confined spaces with inadequate ventilation and without control of toxic fumes in the environment. At the age of 32, after having worked for 5 years as a welder, an akinetic-rigid syndrome developed as the first parkinsonian sign. After several months of disease progression his condition remained stable. There was no response to levodopa (L-dopa) at daily dosages of up to 1.5 g. His medical history was entirely unremarkable. In particular, there was no exposure to neuroleptics and no family history of neurological disorders. At the time of this study, neurological examination showed a severe symmetric akinetic-rigid syndrome. He had no rest tremor. There was no evidence of dystonia, pyramidal tract abnormality, or cerebellar dysfunction. Smell function as assessed with the sniffin stick test was normal. He had no signs of dysautonomy. Neuropsychological evaluation revealed impaired free recall in tests of episodic memory and a reduced reverse digit span (Table 1). Routine laboratory tests including liver function tests were normal and Wilson's disease could be ruled out. Serum manganese levels and 1.5-T MRI head scan (Magnetom; Siemens, Erlangen, Germany) performed 1 month after terminating welding were normal. In particular, no pallidal hyperintensities were detected on T1-weighted MR images.
WP Patient 2 was a 59-year-old man, who had a 20-year exposure to welding fumes in different mines in Chile. He reported that he has worked in places partly with inadequate ventilation and partly with seemingly adequate ventilation but with only irregular controls of toxic fumes load in the working environment. At age 49, he developed an akinetic-rigid syndrome somewhat more prominent on the right side. After a steady increase over 4 years the course of his illness was stable. He did not respond to L-dopa at daily dosages of up to 2 g. His medical history was entirely unremarkable. There was no exposure to neuroleptics and no family history of neurological disorders. At the time of this study, he presented with asymmetric bradykinesia and rigidity, pronounced on the right side. Armswing was reduced on the right side. He walked with short stride whereas starting of walk and turning were normal. There was moderate nuchal rigidity and a slightly prone posturing. Postural reflexes were normal. He showed a mild irregular myoclonic postural tremor but no rest tremor. Smell function was normal. Eye movements and cranial nerve function were regular. There was no evidence of dystonia, pyramidal tract abnormalities, dysautonomy, or cerebellar dysfunction. Neuropsychological evaluation revealed impaired free recall in tests of episodic memory and a reduced verbal fluency (Table 1). Blood and urine tests were normal and Wilson's disease was ruled out. Serum manganese levels and 1.5-T MRI head scan were normal 3 months after terminating welding.
To ensure blinding of the sonographers, the WP patients were randomly mixed with three age-matched male patients with clinically definite idiopathic akinetic-rigid PD.3 The PD patients had a mean age of 55.3 ± 4.2 years, a disease duration of 7.0 ± 1.0 years, and a disease severity of 20.7 ± 3.1 on the motor part of the Unified PD Rating Scale.
Transcranial Brain Sonography
TCS was performed through the preauricular acoustic bone windows using a phased-array ultrasound system with a 2.5-MHz transducer (Acuson Antares; Siemens, Erlangen, Germany). The ultrasound parameters chosen were penetration depth 16 cm, dynamic range 50 dB, postprocessing preset G. SN echogenic size measurements were performed on axial TCS scans automatically after manually encircling the outer circumference of SN′s echogenic area. SN echogenic sizes of less than 0.2 cm2 are classified as normal, sizes of 0.25 cm2 and above as markedly hyperechogenic, and sizes in-between as moderately hyperechogenic.14 Additionally, echogenicity of lenticular and caudate nuclei was investigated and classified as hyperechogenic when it was more intense than the surrounding white matter.14, 17 Widths of third ventricle and of frontal horns of lateral ventricles were measured on a standardized diencephalic axial scanning plane.14 TCS examinations were performed in the Clinica Las Condes, Santiago, Chile, independently by two sonographers (1: U.W.; 2: C.L.) who were blind to the clinical diagnosis of the patients.
For group comparison of SN echogenic sizes the Mann-Whitney U test was used. Categorial data were analyzed by Fisher exact test.
All patients studied were adequately assessable with TCS. Typical TCS images are shown in the Figure 1. Measurements of both sonographers resulted in classifying SN echogenicity as normal in both WP patients (individual larger SN echogenic size, measured by sonographer 1, ranging from 0.09 to 0.13 cm2) and as markedly hyperechogenic in all 3 PD patients (0.32–0.34 cm2). Bilaterally measured SN echogenic sizes were larger in PD patients than in WP patients (Mann-Whitney U test, nWP = 4, nPD = 6, P = 0.010). The finding of at least unilateral SN hyperechogenicity discriminated patients with PD from those with WP in this study; yet, this difference was statistically not significant due to low patient number (Fisher exact test, P = 0.10). Lenticular nucleus TCS revealed unilateral hyperechogenicity in WP Patient 1, bilateral hyperechogenicity in WP Patient 2, unilateral hyperechogenicity in one of the PD patients, and normal echogenicity in the remaining 2 PD patients (P = 0.30). Caudate nuclei were normal in both WP patients and 1 PD patient but hyperechogenic in the remaining 2 PD patients (P = 0.30). Widths of third ventricle and of frontal horns were normal in all patients and did not differ between groups WP and PD.
In this study, findings on two Chilean welders who developed a L-dopa-resistant akinetic-rigid parkinsonian syndrome at age <50 years are reported. The diagnosis of an atypical parkinsonian disorder such as multiple-system atrophy is unlikely since they had, after several months of initial disease progression, a stable course for many years and did not show any signs of cerebellar or autonomic dysfunction.18 However, the pattern of motor and neuropsychological symptoms, the course of the disease over time, and lack of L-dopa responsivity are in line with previous reports on manganese-induced Parkinsonism in welders.7, 9, 11, 19 Both WP patients were career welders with reasonable cause for manganese toxicity. They performed metal inert gas welding in different mines in confined spaces with insufficient ventilation, without using personal respirators, and were exposed to welding fumes over many years. MRI head scan and laboratory workup were performed several weeks after terminating welding and, possibly, several years after the major exposure to the toxic fumes as control of working conditions in Chilean plants has improved in recent years. This may explain why MRI and laboratory findings were normal in the WP patients studied. It has been reported earlier that, in neurologically affected welders with chronic exposure to toxic fumes, MRI globus pallidus T1 hyperintensity and serum manganese levels may normalize already within few weeks after terminating welding.7 However, although there are strong indications for chronic manganese intoxication in the WP patients studied here, it should be stressed that there is no definite proof of this presumption.
Data obtained in this study show that SN hyperechogenicity, characteristic for idiopathic PD, is absent in WP. The WP patients, however, consistently exhibited hyperechogenicity of the lenticular nucleus in the region of the globus pallidus internus, unlike the PD patients who rarely show lenticular nucleus hyperechogenicity.14, 20, 21 A number of studies have demonstrated that accumulation of trace metals in the SN and the basal ganglia can be detected by TCS more sensitively than with routine MRI.12, 17, 22, 23 In patients with idiopathic dystonia and with Wilson's disease, TCS detected accumulation of manganese and copper in the lenticular nucleus, even when routine MRI was normal.17, 22, 23 On the other hand, LN hyperechogenicity in the WP patients may have caused not only by trace metal accumulation but also by minor calcification or gliosis.
The TCS findings in the WP patients resemble findings in patients with multiple-system atrophy and with Wilson's disease who also typically show normal SN echogenicity and lenticular nucleus hyperechogenicity.14, 20, 23 These diagnoses, however, were excluded. Normal SN echogenicity was previously found also in a small percentage of idiopathic PD patients, however, normal SN echogenicity was related to disease onset at ages >50 years in European subjects.21, 24, 25 In these studies, early onset of PD was associated with pronounced SN hyperechogenicity, irrespective of disease duration and severity,21, 24 indicating a characteristic alteration of the nigrostriatal system. SN hyperechogenicity is also typically found in patients with monogenetic forms of hereditary Parkinsonism but can be less pronounced than in idiopathic PD.16, 26 Interestingly, in a recent study of Taiwanese PD patients larger SN echogenic sizes were found to be associated with late rather than early disease onset which might be due to different environmental or genetic factors in Asian compared to European PD patients,27 e.g., a larger frequency of parkin mutations in the European early-onset PD patients studied with TCS. Parkin mutations were found to be related to SN hyperechogenicity but their prevalence had not been estimated in the European TCS studies on sporadic early-onset PD patients.16, 21, 24, 27 Both WP patients studied here had onset of their motor symptoms at ages <50 years but exhibited normal SN echogenicity, in contrast to Chilean and European idiopathic PD patients studied so far.
Assuming manganese intoxication as a relevant cause of Parkinsonism in our WP patients, TCS findings are in line with previously reported findings of positron emission tomography studies in manganese-induced Parkinsonism. These studies suggested that motor symptoms in manganese-induced Parkinsonism are caused by an alteration of the postsynaptic striatal dopaminergic projections rather than affection of the nigrostriatal system, unlike PD where the nigrostriatal system is primarily affected.28 Also, postmortem neuropathologic examinations of patients with documented manganese-induced Parkinsonism showed predominant involvement of the basal ganglia with neuronal degeneration and gliosis, mainly in the globus pallidus, with sparing of the nigrostriatal system and other regions typically affected in PD.29
Finally, it remains debatable whether the Parkinsonism not only in our WP patients, but also in previously reported WP patients, was caused by manganese neurotoxicity, or by welding at all.9 Nevertheless, TCS findings of the present study further support the idea of a different pathophysiology of WP and idiopathic PD. Further studies are warranted to search for a genetically determined predisposition for developing Parkinsonism in welders.