Clinical course in patients with chronic carbon disulfide polyneuropathy

Clinical Neurology and Neurosurgery 104 (2002) 115– 120 www.elsevier.com/locate/clineuro

Clinical course in patients with chronic carbon disulfide polyneuropathy Chin-Chang Huang a,*, Chun-Che Chu a, Trong-Neng Wu b, Tung-Sheng Shih c, Nai-Shin Chu a b

a Department of Neurology, Chang Gung Memorial Hospital and Uni6ersity, 199, Tung Hwa North Road, Taipei, Taiwan, ROC Department of Health, Diseases Sur6eillance and Quarantine Ser6ice, Executi6e Yuan, and Kaohsiung Municipal Hsiaokang Hospital, Taipei, Taiwan, ROC c Institute of Occupational Safety and Health, Council of Labor Affairs, Executi6e Yuan, Taipei, Taiwan, ROC

Received 29 May 2001; received in revised form 31 October 2001; accepted 23 November 2001

Abstract The natural course of clinical manifestations and electrophysiological changes were studied in six patients with carbon disulfide (CS2) induced polyneuropathy. All of the six patients worked in the cutting-machine department of a viscose rayon plant. The environmental monitoring was also conducted in the initial stage and followed up 3 years later. In the 3-year follow-up period, the neurological symptoms and signs persisted. The highest concentration of CS2 in the cutting machine where these patients worked was about 100–200 ppm. Three years later, the highest concentration was decreased to between 10 and 20 ppm in the cutting machine of the new production line after the engineering control had been improved. Nerve conduction velocity (NCV) studies revealed persistent abnormality in motor and sensory NCVs. Although, a tendency to improvement was noted, it did not reach a statistical significance except for conduction velocity of sural nerve in sensory NCV. Sural nerve biopsy from one patient, 2 years after diagnosis showed degeneration of both axon and myelin and a predominant loss of large myelinated fibers. A remyelination process was also noted. We concluded that CS2 intoxication may induce a persistent damage to the peripheral nerves even after CS2 exposure had ceased for 3 years. © 2002 Elsevier Science B.V. All rights reserved. Keywords: CS2; Peripheral neuropathy; NCV; Sural nerve biopsy; Neurotoxic disease

1. Introduction Peripheral neuropathy has been documented in patients after long-term exposure to carbon disulfide (CS2) in viscose rayon workers [1 – 4]. The severity of peripheral nerve involvement may be correlated with the level of CS2 exposure [5– 10]. However, only a few electrophysiological studies have demonstrated the long-term effect of peripheral nerve damage after cessation of exposure [5,8,11]. Either good improvement, partial improvement or permanent changes have been reported [5,8,11]. In addition, most of these reports did not

* Corresponding author. Tel.: + 886-3-3281200x8418/82853; fax: 886-3-3287226. E-mail address: [email protected] (C.-C. Huang).

describe the clinical features. Therefore, the natural course and prognosis of the patients with CS2 polyneuropathy have remained obscure. The pathological findings of CS2 polyneuropathy showed multifocal axonal swelling due to accumulation of 10 nm neurofilaments and secondary demyelination only in experimental rats [12 –15]. The pathogenesis of axonal swelling is very similar to that of n-hexane intoxication [16 –18]. However, a comparison of the long-term prognosis of these two types of intoxications has not been discussed. In a recent outbreak of CS2 intoxication in 1992, we found a group of patients who developed axonal polyneuropathy [19,20]. The aims of the study were to follow-up the clinical manifestations, electrophysiological studies, and the natural course in these patients after cessation of CS2 exposure.

0303-8467/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 3 - 8 4 6 7 ( 0 1 ) 0 0 1 8 7 - 1

116

C.-C. Huang et al. / Clinical Neurology and Neurosurgery 104 (2002) 115–120

2. Materials and methods

3. Sural nerve biopsy study

In a viscose rayon plant, 10 out of 163 workers developed a mild to moderate degree of polyneuropathy in 1992 [19]. All of them were chosen for the follow-up study from 1995 to 1996. The diagnosis of polyneuropathy was based on distal sensory impairment, muscle weakness, decreased or absent tendon reflexes, and abnormal nerve conduction velocity (NCVs) [21]. In addition, other causes such as diabetes, porphyria, Guillain– Barre syndrome, nutritional deficiency, and other toxic diseases could be excluded. After the diagnosis of CS2 polyneuropathy, eight patients had subsequently left their jobs. The remaining two patients were still working the same jobs but the environments had been improved. Most of the patients were followed-up and charted monthly. In 1995, we conducted a questionnaire on these workers concerning job history, age of onset, duration of exposure, and histories of hypertension, diabetes, smoking and alcohol consumption. Detailed changes in neurological symptoms including muscle pain, back pain, muscle cramps, fatigue, distal muscle weakness, distal numbness, gait disturbance, and difficulty walking or running were emphasized and compared with the previous symptoms. In the 10 patients previously studied, four patients who had developed stroke episodes were excluded from the follow-up study. There were neither frank psychosis, aphasia, nor personality disorders. Neurological examinations were performed by the same two neurologists (C.C. Chu and C.C. Huang) independently including muscle strength, deep tendon reflexes and sensory functions. The muscle strengths from biceps, flexor carpi ulnaris, abductor pollicis bre6is, hamstrings, tibialis anterior and extensor hallucis longus muscles were recorded according to the Medical Research Council of the Great Britain. MRC 4/5 was considered as mild weakness, 3/5 as moderate, and 0– 3/5 as marked weakness. Deep tendon reflexes were classified as generalized areflexia, hyporeflexia, absent ankle jerk and hyperreflexia. Impairments of sensations including pin pricks, temperature, touch, vibration and position were also classified as mild, moderate and marked. Cranial nerve abnormalities and coordination was also evaluated carefully. All patients underwent nerve conduction studies including motor nerve conduction study in bilateral median, ulnar, tibial and peroneal nerves and sensory nerve conduction study in bilateral median, ulnar and sural nerves. NCV was measured and carried out at a room temperature of 22– 24 °C by the standard method as previously described [22]. Control NCV data were obtained from 20 age- and sex-matched healthy adults. Student’s t-test was used for statistical analysis as compared between the initial test and the follow-up test and each test with the normal controls.

Sural nerve biopsy was obtained from patient 1. Approximately 2 cm of the sural nerve was examined with light and electron microscopy. The fiber sizes of each myelinated nerve fiber was measured in a morphometric analyzer (LEICA Q500MC image processing and analysis system). In addition about 100 nerve fibers were teased under a dissecting microscope and observed in a light microscope. Paranodal swelling, demyelination and internodal distance were specially observed.

4. Environmental and biological monitoring Fixed-point and personal air samples collected at different worksites were obtained and studied in the viscose rayon factory according to the methods previously published [23]. A biological marker of 2-thiothiazolidine-4-carboxylic acid (TTCA)— a metabolite of CS2 was obtained from urine samples of the workers before and after the shift and studied by gas chromatography (GC)/mass spectrometry (MS) [24]. The environmental monitoring studies were conducted in 1992 and 1995, respectively, before and after the engineering control was improved.

5. Results Table 1 summarized the clinical manifestations of six patients with CS2 intoxication. All patients were males aged between 43 and 50 years. The age of onset of neurological symptoms was from 31 to 46 years. The duration of exposure ranged from 4 to 23 years. Diabetes and alcoholism were not found, but hypertension was found in two patients. Three patients were smokers. Table 1 shows the changes of neurological symptoms and signs. Most of the neurological symptoms were improved except for patient 1. The neurological sign of muscle strength revealed symmetrical distal weakness in four limbs and was improved in three, stationary in two, and deteriorated in one. Tendon reflexes were changed from hyporeflexia to absent ankle jerk in one, and remained stationary in the remaining five. The sensory impairments demonstrated a glove and stocking-like pattern and were improved in three, stationary in two and deteriorated in one. The cranial abnormality of neurosensory type hearing impairment was only found in patient 1. The coordination was normal in all six patients.

C.-C. Huang et al. / Clinical Neurology and Neurosurgery 104 (2002) 115–120

117

6. Nerve conduction studies

7. Sural nerve biopsy

Serial motor and sensory nerve conduction studies were carried out in six patients. At the initial study of motor nerve conduction, distal latency (DL) of compound muscle action potentials (CMAP) were prolonged in all motor nerves tested except for tibial nerve. The amplitudes of CMAP were reduced except for ulnar and tibial nerves. The motor NCV were slowed in all nerves tested as compared with the normal controls. There was generally a tendency for improvement in the follow-up study including DL, amplitudes of CMAP and MNCV in all nerves tested. However, none had a statistically significant improvement in the follow-up study. Most of the findings were still abnormal in the follow-up study except for the amplitudes of CMAP in the ulnar and tibial nerves and the DL of the tibial nerve as compared with the normal controls (Table 2). The serial sensory nerve conduction study was shown in Table 3. In the initial study, the DL, amplitudes, and NCV of the SNCV were significantly abnormal except for the DL of the sural nerve as compared with the normal controls. In the follow-up study, a tendency of improvement was noted in most items which was significant for the NCV of the sural nerve only. In addition most of the data still had a statistically significant abnormality as compared with the normal controls.

Light microscopic examination revealed axonal degeneration, and thin myelin in some fibers. A reduction of fiber density in large myelinated fibers was noted. Morphometric analysis also confirmed a decrease of myelinated fiber density (4580–5600/mm2, reference ranges 6000–10,000/mm2). The teased-fiber study showed shortening of internodal length in some axons. There was no obvious paranodal swelling. At ultrastructural examination, neurofilaments and microtubules were disorganized and abnormal mitochrondria, glycogen granules and segregation of axoplasmic organelles were observed. In addition, reduction and splitting of myelin lamellae and intramyelinic vacuoles were also noted. In histogram of myelinated fiber diameter, small myelinated fibers were relatively preserved but large myelinated fibers were decreased.

8. Air and urinary analysis The concentrations of CS2 in fixed-point air samples in the cutting machine of the old production line where these patients worked were about 100–200 ppm in 1992. The geometric mean of personal samples in the

Table 1 Clinical follow-up of six patients with CS2 intoxication Patient

1

2

3

4

5

6

Current age (yr) Age of onset (yr) Exposure duration (yr) Hypertension Alcohol Smoking Diabetes Symptoms Muscle pain Back pain Muscle cramp Fatigue Distal muscle weakness Distal numbness Gait disturbance Difficulty walking or running Signs Muscle strength Tendon reflexes Sensation Cranial nerve abnormality Coordination

50 46 23 − − − −

46 44 21 + − − −

46 31 14 + − + −

46 44 4 − − + −

44 33 17 − − + −

43 41 17 − − − −

+ + + + + + + +

( ) ( ) ( ) ( ) (¡) (¡) (¡) (¡)

Mi“ Mo 1“1 Mi“Mo Hearing impairment N

+ + + + + + + +

(S) (S) ( ) ( ) ( ) ( ) ( ) ( )

Mi“Mi 2 “2 Mi“N N N

+ + + + + + + +

( ) (S) ( ) ( ) (¡) (S) (S) (S)

Mi“Mi 1“1 Mi“Mi N N

+ + + + + + + +

( ) (S) ( ) ( ) ( ) ( ) ( ) ( )

Mi“ N 3“3 Mi“Mi N N

+ + + + + + + +

( ) (S) (S) ( ) (S) (S) (S) (S)

Mi“N 1“1 Mi“N N N

+ + + + + + + +

( ) (S) ( ) ( ) ( ) ( ) ( ) ( )

Mi“ N 2“ 3 Mi“ N N N

yr, years; +, presence; −, absence;  , improved; ¡, deteriorated; S, stationary; 1, diffuse areflexia; 2, hyporeflexia; 3, absent ankle jerks; Mi, mild; Mo, moderate; N, normal; L, left; R, right.

8.8*,a 2.6

4.4**,a 0.7

3.0 0.4

(II) Mean SD (n=12)

(III) Mean SD (n= 40)

60.5 4.0

50.2**,a 4.9 2.4 0.4

3.0**,a 0.5

3.1** 0.5

9.1 1.8

9.0a 1.6

8.6 2.2

Amp (mV)

60.5 4.2

52.3**,a 6.5

48.9** 7.3

NCV (m/s)

4.2 0.5

5.0*,a 1.7

5.7** 1.2

DL (ms)

5.9 2.5

3.4**,a 1.8

3.0** 1.6

Amp (mV)

Peroneal nerve

DL, distal latency; Amp, amplitude; NCV, nerve conduction velocity; (I) initial study; (II) follow-up study; (III) normal controls. a P\0.05 in I vs. II. * PB0.05, ** PB0.01 in I vs. III or II vs. III.

10.9 2.4

8.9* 3.6

4.5** 0.9

(I) Mean SD (n=12)

49.0** 6.3

DL (ms)

NCV (m/s)

DL (ms)

Amp (mV)

Ulnar nerve

Median nerve

Table 2 Serial motor nerve conduction studies in six patients with CS2 intoxication

51.1 3.2

41.7**,a 6.0

39.3** 9.6

NCV (m/s)

5.3 1.0

5.0a 1.5

5.6 1.5

DL (ms)

Tibial nerve

9.8 2.9

9.7a 3.9

7.5 5.7

Amp (mV)

49.1 3.4

42.7**,a 8.4

40.5** 8.0

NCV (m/s)

118 C.-C. Huang et al. / Clinical Neurology and Neurosurgery 104 (2002) 115–120

C.-C. Huang et al. / Clinical Neurology and Neurosurgery 104 (2002) 115–120

119

Table 3 Serial sensory nerve conduction studies in six patients with CS2 intoxication Group

Median nerve

Ulnar nerve

Sural nerve

DL (ms)

Amp (mV)

NCV (m/s)

DL (ms)

Amp (mV)

NCV (m/s)

DL (ms)

Amp (mV)

NCV (m/s)

(I) Mean SD (n= 12)

3.2** 0.8

23.9** 14.0

55.8** 7.0

2.9** 0.7

17.7** 9.5

57.8** 7.0

2.9 0.5

12.7** 10.2

27.0** 18.5

(II) Mean SD (n=12)

3.3**,a 0.5

19.5**,a 10.5

57.1**,a 5.5

2.8*,a 0.4

15.3**,a 6.0

57.8**,a 7.5

2.7*,a 0.6

12.0**,a 5.0

42.7**,b 11.0

(III) Mean SD (n=40)

2.5 0.4

42.0 18.0

65.9 3.6

2.1 0.3

41.0 14.4

66.0 5.1

3.0 0.3

29.0 15.1

49.4 4.2

DL, distal latency; Amp, amplitude; NCV, nerve conduction velocity; (I) initial study; (II) follow-up study; (III) normal controls. a P\0.05, b PB0.05 in I vs. II. * PB0.05, ** PB0.01 in I vs. III or II vs. III.

cutting machine was 30.349 2.46 ppm with a range of 7.5 –251 ppm (n=15). Since the engineering control was improved, the fixed-point concentrations of CS2 in the cutting machine of the new production line were as low as 10–20 ppm in 1995. The personal air samples in the cutting machine were between 20 and 40 ppm. The urinary concentration of 2-thiothiazolidine-4-carboxylic acid (TTCA) was measured in 1992. The geometric mean was 0.59 3.6 mg/g creatinine (range: 0.01–6.2 mg/g creatinine, n= 60) at the beginning of the shift. At the end of the shift, the geometric mean was 1.4894.74 mg/g creatinine (range: 0.01– 18.2 mg/g creatinine, n= 57). About 14 urinary samples (26.3%) of the workers had a level above the biological exposure index 5 mg/g creatinine, although most of the directly exposed workers wore respirators, long rubber apron, gloves, and boots during our sampling.

9. Discussion The present study showed that long-term improvement of the clinical manifestations in CS2 induced polyneuropathy was not so prominent. Most of the neurological symptoms and signs persisted, although some symptoms might improve. However, the serial nerve conduction studies also confirmed a persistent abnormality. A tendency to improvement was noted in both motor and sensory NCV, but it did not reach a statistical significance in most parameters. Review of the literature also reveals that some patients did not improve in the course of years. Vigliani [25] had reported that one patient recovered, four improved, five unchanged and 10 deteriorated suggesting a variable prognosis. However, a detailed clinical and electrophysiological study on long-term prognosis has not yet been published. In the patients’ group we

studied, variable but persistent symptoms and signs were observed. Therefore, our data show that the neuropathy may persist for a long period of time. The electrophysiological studies have been extensively studied in a large group of occupationally exposed workers [5–9]. In patients with high level of CS2 exposure, there were slowing of motor and sensory NCV, prolongation of DL and reduction of amplitudes in both CMAPs and SNAPs. Furthermore, slowing of NCV was correlated statistically in a dose-related way to an average exposure level of 2.0 ppm [10]. In the electrophysiological studies, an improvement or incomplete recovery had been reported in subjects who had ceased exposure to CS2 for a long period [5,8]. Corsi et al. [11] had also reported that some patients had a persisted NCV abnormality 10 years later and concluded that the damage of CS2 was permanent or only partially recovered. However, the clinical manifestations were not clear in their studies. The electrophysiological study in our patients suggested an axonal and demyelinating polyneuropathy. In the follow-up studies, the clinical improvement was slow and incomplete. Serial MNCV and SNCV also suggested a potential reversible change in the nerve conduction. However, most patients did not return to normal values even when they have already stopped exposure to CS2 3 years later. The neuropathological changes in animal studies after CS2 exposure have indicated a central-peripheral distal axonopathy [12–15]. Light microscopic examination revealed multifocal axonal swelling particularly in the paranodal region and secondary demyelination. In electron microscopic examination, an accumulation of 10 nm neurofilaments was found very similar to those with n-hexane intoxication [16–18]. In our study, sural nerve biopsy also showed an axonal degeneration, a decrease of fiber density and relative loss of large myelinated fibers. In addition, some large axons con-

120

C.-C. Huang et al. / Clinical Neurology and Neurosurgery 104 (2002) 115–120

tained thin myelin indicating a remyelinating process [20]. The mechanisms of the toxicity of CS2 remained obscure. However, it is thought that CS2 may react with amines to form dithiocarbamate and result in enzyme inhibition similar to 2,5-hexanedione to induce neuropathy. In addition, the ability of cross-link proteins covalently is required for the production of neurofilament accumulation by g-diketones [17,26]. Further investigation of the mechanisms of CS2 toxicity may elucidate the problems.

[11]

[12]

[13]

[14]

Acknowledgements This study was supported in part by a grant from the Department of Health, Executive Yuan (DOH 84-TD098). The authors are grateful to Y.-C. Hsieh for typing the manuscript.

[15]

[16]

[17]

References [18] [1] Peters HA, Levine RL, Matthews CG, et al. Carbon disulfide-induced neuropsychiatric changes in grain storage workers. Am J Ind Med 1982;3:373 –91. [2] Peters HA, Levine RL, Matthews CG, et al. Synergistic neurotoxicity of carbon tetrachloride/carbon disulfide (80/20 fumigants) and other pesticides in grain storage workers, Acta Pharmacol Toxicol 1986;59(7):535 – 46. [3] Lewey FM. Neurological and biochemical signs and symptoms indicating chronic industrial carbon disulfide absorption. Ann Intern Med 1941;15:869 –83. [4] Aaserud O, Hommeren OJ, Tvedt B, et al. Carbon disulfide exposure and neurotoxic sequelae among viscose rayon workers. Am J Ind Med 1990;18:25 –37. [5] Manu P, Lilis R, Lancranjan J, Jonescu S, Vasilescu CJ. The value of electromyographic changes in the early diagnosis of carbon disulphide peripheral neuropathy. Med Lav 1970;61:102 – 8. [6] Seppalainen AM, Tolonen M. Neurotoxicity of long-term exposure to carbon disulfide in the viscose rayon industry. A neurophysiological study. Work Environ Health 1974;11:145 – 53. [7] Vasilescu C. Sensory and motor conduction in chronic CS2 poisoning. Eur Neurol 1976;14:447 –57. [8] Giglioli R, Bulgheroni C, Bertazzi PA, et al. Study of neurological and neurophysiological impairment in carbon disulphide workers. Med Lav 1978;69:130 –43. [9] Vasilescu C, Florescu A. Clinical and electrophysiological studies of carbon disulfide polyneuropathy. J Neurol 1980;224:59 – 70. [10] Hirata M, Sugimoto K, Misumi J, et al. A neurophysiological

[19]

[20]

[21]

[22]

[23]

[24]

[25] [26]

study among Chinese CS2-exposed workers. G Ital Med Lav 1984;6:107 – 11. Corsi G, Maestrelli P, Picotti G, Manzoni S, Negrin P. Chronic peripheral neuropathy in workers with previous exposure to carbon disulphide. Br J Ind Med 1983;40:209 – 11. Juntunen J, Linnoila I, Haltia M. Histochemical and electron microscopic observation on the myoneural junctions of rats with carbon disulfide induced polyneuropathy. Scand J Work Environ Health 1977;3:36 – 42. Colombi A, Maroni M, Picchi O, Rota E, Castano P, Foa V. Carbon disulfide neuropathy in rats. A morphological and ultrastructural study of degeneration and regeneration. Clin Toxicol 1981;18:1463 – 74. Jirmanova I, Lukas E. Ultrastructure of carbon disulfide neuropathy. Acta Neuropathol (Berl) 1984;63:255 – 63. Gottfried MR, Graham DG, Morgan M, Casey HW, Bus JS. The morphology of carbon disulfide neurotoxicity. Neurotoxicology 1985;6:89 – 96. Spencer PS, Schaumburg HH. Ultrastructural studies of the dying-back process III. The evaluation of experimental peripheral giant axon degeneration. J Neuropathol Exp Neurol 1977;36:276 – 99. Graham DG, Amarnath V, Valentine WM, Pyle SJ, Anthony DC. Pathogenetic studies of hexane and carbon disulfide neurotoxicity. Crit Rev Toxicol 1995;25:91 – 112. Johnson DJ, Graham DG, Amarnath V, Amarnath K, Valentine WM. Release of carbon disulfide is a contributing mechanism in the axonopathy produced by N,N-diethyldithiocarbamate. Toxicol Appl Pharmacol 1998;148:288 – 96. Chu CC, Huang CC, Chen RS, Shih TS. Carbon disulfide-induced polyneuropathy in viscose rayon workers. Occup Environ Med 1995;52:404 – 7. Chu CC, Huang CC, Chu NS, Wu TN. Carbon disulfide induced polyneuropathy: sural nerve pathology, electrophysiology, and clinical correlation. Acta Neurol Scand 1996;94:258 – 63. Juntunen J, Haltia M. Polyneuropathies in occupational neurology: pathogenetic and clinical aspects. Acta Neurol Scand 1982;Suppl. 92:59 – 73. Huang CC, Chu NS, Cheng SY, Shih TS. Biphasic recovery in n-hexane polyneuropathy. A clinical and electrophysiological study. Acta Neurol Scand 1989;80:610 – 5. National Institute for Occupational Safety and Health (NIOSH). Carbon disulfide method 1600: 1 – 4. Manual of analytical methods, 3rd ed., vol. 2. Cincinnati, OH: HIOSH, 1984. pp. 84 –100. Ogata M, Taguchi T. Determination of urinary 2-thiothiazolidine-4-carboxylic acid by automated high performance liquid chromatography as an index of carbon disulfide exposure. Ind Health 1989;29:31 – 5. Vigliani EC. Carbon disulfide poisoning in viscose rayon factories. Br J Ind Med 1954;11:235 – 44. Genter St Clair MB, Amarnath V, Moody MA, et al. Pyrrole oxidation and protein cross-linking as necessary steps in the development of g-diketone neuropathy. Chem Res Toxicol 1988;1:179 – 85.