brain MRI in manganese exposed workers and HEPATOPATIC patients

Roberto Lucchini *, Elisa Albini, Laura Benedetti, Lorenzo Alessio (Institute of Occupational Health, University of Brescia, Italy)

Roberto Gasparotti (Department of Neuroradiology, University of Brescia, Italy)

ABSTRACT

Brain magnetic resonance image (MRI) can visualize an elective accumulation of manganese (Mn) in the globus pallidus among intoxicated workers following high exposure to Mn and in hepatopatic patients suffering from hepatic encephalopathy, due to a lack of the biliaric excretion of Mn.

To assess the relationship between the level of Mn exposure and the degree of MRI hyperintensity, a study was conducted on 7 male workers exposed to Mn, 5 hepatopatic patients and a group of 7 age matched controls. The subjects were examined with brain MRI, blood Mn (MnB) and serum prolactin (PRL) dosage. The MRI hyperintensities were assessed with the “pallidal index”, which is considered as a “semi-quantitative” estimate of the amount of Mn accumulated in the globus pallidus.

MRI hyperintensity was evident among Mn workers and hepatopatic patients, with identical patterns of bilateral and symmetric images, in the globus pallidus and the pituitary gland. A significant association was observed towards the increase of the PI and the PRL as a function of MnB levels and also between the PI and the PRL levels.

INTRODUCTION

Mn is an essential nutrient and is a natural component of most food. The highest concentration of manganese has generally been detected in cereals, grain products, nuts, unpolished rice and tea. As an essential element, manganese is normally present in human and animal fluids and tissues. The highest tissue concentrations are in the liver, pancreas and kidney, and lower levels in bone and fat. The brain has a small amount of manganese, but retention time is long, with the highest concentrations in the basal ganglia (nucleus caudatus, globus pallidus and putamen). Manganese is almost totally excreted via the biliary tract. The biological half life is, for the most part, extremely high regarding the elimination from the brain, and lower from the liver.

Soft tissues such as the central nervous system (CNS) can be effectively visualized by using proton nuclear MRI technique. In overload conditions, Mn selectively accumulates in the basal ganglia and can produce toxic effects in this region. MRI has provided evidence of Mn deposition in the brain of Mn intoxicated workers, who had been exposed to high concentrations of Mn in the air, generally above 1 mg/m³ (Alessio and Lucchini, 1996). These workers developed a clinical intoxication called "manganism" which is characterized by extrapyramidal dysfunction and neuropsychiatric features. An identical clinical picture has been described in patients receiving Mn as part of a long-term total parenteral nutrition, with similar MRI findings (Mirowitz et al., 1991). Another group of studies on patients suffering from end-stage liver disease has found the clinical features of “hepatic encephalopathy”, which resembles manganism and shows increased MnB levels due to a lack of biliaric excretion of Mn. Additionally, it was also observed in these patients, increased MRI signals in the globus pallidus (Krieger et al., 1995).

More recently, field studies have been conducted on asymptomatic Mn exposed workers, confirming the presence of MRI hyperintensity without any clinical manifestation of manganism (Kim et al., 1999; Dietz et al., 1999). The aim of our research was to compare the results of brain MRI and the MnB levels of Mn exposed asymptomatic workers and with the resultsof hepatic patients using the same methodology.

METHODS

The target population was composed of 3 groups of subjects: Mn exposed workers, hepatopatic patients and control subjects.

1. Seven workers were selected from a group of 10 male subjects employed at a ferroalloy plant, who were exposed to airborne Mn concentrations below the TLV-TWA of 200 mg/m³. They were not affected by any neurological or hepatological sickness.

2. Five hepatopatic patients were selected from inpatients at our hospital and were affected by liver cirrosis classified as B and C according to the Child-Pugh classification.

3. Seven hospital workers not exposed to any neurotoxic agent nor affected by neurological or hepatological sickness volunteered to participate in the study.

All the subjects were examined with brain MRI and the dosage of MnB and serum PRL. MRI scan was performed with the Siemens Magnetom apparatus with a field strength of 1,5 Tesla. T1-weighted trasversal scans had a slice thickness of 3 mm and 0.5 mm of interslice gap. with a repetition time (TR) of 580 msec and an echo time (TE) of 15 msec. Since the region of interest (ROI) for MRI was the globus pallidus, the signal intensities were evaluated in this area using the semi-quantitative method named pallidal index (PI), according to Krieger et al. (1995), where the PI is defined as the ratio of the signal intensity of the globus pallidus (ROI1) to subcortical white-matter (ROI2 and ROI3) multiplied by 100.

The measures of PI, MnB and PRL in the three groups were analysed with ANOVA test and linear regression between MnB levels vs. PI and PRL values was assessed using the Pearson's coefficient.

RESULTS

The results of the measures in the three groups of subjects are summarized in table 1. MnB was significantly higher in the workers and the patients compared to the control subjects. Overall, MnB was higher in the Mn exposed workers than in the liver patients, but it must be specified that liver patients with Child-C classification showed higher MnB levels compared to the patients with Child-B classification, and the Mn workers. The PI and the PRL levels, showed a significant difference among the three groups, with higher PI and PRL in the liver patients.

One of the workers repeated the MRI at three different times: immediately before retirement and 2 and 12 months after. The high signals and the MnB concentration decreased gradually: the PI from 113.4, to 107.5 and 104.8 respectively and the MnB concentration from 15 µg/l to 13 µg/l and 8 µg/l respectively.

Table 1: results of MnB, PI and PRL in the three groups, with the significance levels at the ANOVA test.

	Liver patients	Mn workers	Controls	p
MnB (µg/l) Mean (SD)	9.88 (5.44)	12.26 (3.18)	5.02 (1.18)	0.007
Median	8.10	13.00	5.25
PI
Mean (SD)	118.78 (6.61)	113.64 (5.80)	97.62 (3.38)	<0.0001
Median	119.2	113.4	97.7
PRL (ng/ml)
Mean (SD)	19.50 (0.71)	12.65 (4.55)	6.51 (0.98)	0.0009
Median	19.50	12.58	6.54

Considering all the subjects in the three groups, the simple regression analysis evidenced a positive association between MnB and both PI and PRL, and also a strong relationship between PI and PRL: r=0.71, r²=0.51, p=0.0006, PRL=-27,888 + 0,363 x PI (figure 1).

Figure 1: Relationship between the PI and the PRL (in ng/ml) levels in the three groups

DISCUSSION

Mn can be effectively visualized by the use of MRI. This is an important feature because it allows a better definition of kinetics and neurological effects of this essential element. Mn overload can be determined by an excessive absorption of a high external dose, as in occupational exposure, or by a lack of biliaric excretion, as in the hepatic failure conditions. No matter what the basic abnormal condition is, the result is commonly represented by an accumulation of Mn in the brain area where it shows an unusual affinity to the globus pallidus. In this area peculiar hyperintense MRI signals can be visualized.

Mn overload in globus pallidus can determine neurotoxic effects which are likely to be dose-dependent and based on a continuum from early changes in neurobehavioral functions including motor, memory and mood impairments, to clinical signs of overt extrapyramidal and psychiatric manifestations. This continuum has been fairly well documented by the literature of occupational (Lucchini et al., 1999) and environmental studies and now can be hypothesized within the framework of the literature on hepatic encephalopathy. This pathogenesis can be soundly recognized as a decisive role played by Mn.

The mechanisms of action involved in Mn neurotoxicity that may primarily involve the dopaminergic system. In fact, in autopsied brain tissues from cirrhotic patients, a selective loss of postsynaptic dopamine D2 binding sites was observed in pallidum, with a concomitant two- to seven-fold increase of Mn concentrations in this brain region (Butterworth et al., 1995). The interference of the dopaminergic system, which is a tonic modulator of prolactin secretion, could explain the significantly increased prolactin secretion as a function of MnB concentrations both in Mn exposed workers (Smargiassi and Mutti, 1999), than in hepatopatic patients, as appears from the results of our study.

The results of this study confirm the presence of a dose-dependent deposition of Mn in the globus pallidus visualized by MRI. Previous observations were obtained by separate examination of occupationally exposed subjects and liver patients respectively, whereas in this research these different type of subjects were examined at the same time with the same methodology. Moreover, the results have evidenced an increase of PRL secretion which is dose-dependent as a function of MnB, and is also associated with the intensity of MRI signals in the globus pallidus. This finding appears to be of relevant importance in view of the explanation of a possible mechianism of action of Mn that involve the dopaminergic system and deserves further research.

REFERENCES

Alessio L and Lucchini R (1996) In: Data profiles for selected chemical and pharmaceuticals: 2,4-dinitrophenol, manganese and manganese compounds, tricloroethylene. (F Argentesi, R Roi, M Sevilla), Ispra, Italy, Joint Research Center, European Commission, pp. 75-162.

Butterworth RF, Spahr L, Fontaine S, Layrargues GP (1995) Metabolic Brain Disease. Dec; 10 (4): 259-67.

Dietz MC, Ihrig A, Bader M, Wradzillo W, Triebig G (1999) . In: Proc. The 7th International Symposium on Neurobehavioral Methods and Effects in Occupational and Environmental Health; June 20-23 1999, Stockholm.

Kim Y, Kim KS, Yang JS, Shin YC, Park IJ, Kim E, Jin Y, Kwon KR, Chang KH, Kim JW, Park SH, Lim HS, Cheong HK, Park J, Moon Y (1999), Neurotoxicology, 20(6): 901-908.

Krieger D, Krieger S, Jansen O, Gass P, Theilmann L, Lichtnecker H (1995), Lancet Jul 29; 346(8970): 270-4.

Lucchini R, Apostoli P, Perrone C, Placidi D, Albini E, Migliorati P, Mergler D, Sassine MP, Palmi S, Alessio L (1999) Neurotoxicology,; 20 (2-3): 287-298

Mirowitz SA, Westrich TJ, Hirsch JD (1991) Radiology, 181: 117-120.

Smargiassi A, Mutti A (1999) Neurotoxicology, 20(2-3):401-406.