The association between renal dysfunction and osteopenia in environmental cadmium-exposed subjects

ENVIRONMENTALRESEARCH51, 71--82 (1990)

The Association between Renal Dysfunction and Osteopenia in Environmental Cadmium-Exposed Subjects TERUHIKO K I D O , KOJI NOGAWA, RYUMON H O N D A , IKIKO TSURITANI, MASAO ISHIZAKI, YUICHI YAMADA, AND HIDEAKI NAKAGAWA*

Departments of Hygiene and *Public Health, Kanazawa Medical University, Uchinada, Ishikawa 920-02, Japan Received April 8, 1989 Two hundred and three cadmium (Cd)-exposed subjects with renal dysfunction and 80 non-exposed subjects were examined to reveal the relationship between Cd-induced renal dysfunction and osteopenia. As biological indicators of renal function, urinary f32-microglobulin (132-mg), and serum creatinine, calcium, and phosphorus were selected. Cd in the urine and blood was also measured. The results indicated that significant differences exist between both sexes in Cd-exposed as well as nonexposed subjects. To evaluate the degree of osteopenia, a microdensitometrical (MD) method was used. The relationships between biological parameters and each index of the MD method were analyzed using multivariate analysis. Age, urinary 132-mg, and serum creatinine were significantly associated with indices of osteopenia in Cd-exposed men. In contrast, age showed the most significant association with MD indices in women of both groups. However, urinary ~2-mg was significantly associated with MD indices only in Cd-exposed women. In Cd-exposed subjects, after the number of predictor variables was increased, urinary 132-mg was also strongly associated with osteopenia. These results suggest a causal relationship between renal dysfunction and osteopenia in Cd-exposed subjects. © 1990AcademicPress, Inc.

INTRODUCTION

Cadmium (Cd) damages various human organs such as kidneys, lungs, and bones (Friberg et al., 1986; Foulkes, 1986). In chronic Cd intoxication, lowmolecular-weight proteins in the urine such as urinary [32-microglobulin, urinary oh-microglobulin, and retinol binding protein are useful indicators of early renal tubular dysfunction (Bernard et al., 1982; Nogawa et al., 1984; Kido et al., 1985). Itai-itai disease is considered to represent the most severe stage. The pathological bone findings of this disease are a combination of osteomalacia and osteoporosis (Kajikawa, 1978; Nogawa, 1981). Looser's zone is often found on X-ray photographs of bone as a characteristic feature of osteomalacia. Even if this typical finding cannot be seen, autopsy or bone biopsy is performed to diagnose osteomalacia since osteomalacia findings are indispensable for the diagnosis of Itai-itai disease (Mukawa et al., 1980; Kitagawa et al., 1983). On the other hand, osteoporosis is characterized by a reduction in bone mass exceeding the natural reduction occurring with age alone (Grech et al., 1985). Animal experiments have often shown that Cd induces osteoporotic changes (Ishizaki et al., 1966; Matsuda, 1967; Yoshiki et al., 1975). However, less research on osteoporosis in Cd-exposed humans has been carried out due to the lack of suitable equipment to measure bone density quantitatively. Recently a new method, called the microdensitometrical (MD) method, has 71 0013-9351/90 $3.00 Copyright© 1990by AcademicPress, Inc. All fightsof reproductionin any formreserved.

72

K I D O ET AL.

been devised to assess the degree of bone density (Inoue et al., 1983). It uses an aluminum step-wedge, a densitometer, and a computer. This method measures the cortical width and bone mineral content, which are good indicators of the degree of osteopenia. At present, this method is widely utilized in Japan since it has relatively good reproducibility, is easy to use, and is not expensive (Yoshikawa et al., 1983; Miyamoto et al., 1984; Nakamura and Yoshikawa, 1984; Okano et al., 1985; Hayashi et al., 1986). Using the MD method, we have already shown that marked osteopenia is found in Itai-itai disease patients and Cd-exposed subjects compared to nonexposed subjects (Kido et al., 1989). The aim of the present study was to demonstrate the association between Cd-induced renal dysfunction and osteopenia using multivariate analysis. This study consists of two parts. First, Cd-exposed and nonexposed subjects are selected as a target population. The relationships between age, biological indicators such as urinary [32-microglobulin and serum creatinine, and osteopenia are investigated. Second, with the addition of renal function tests, the relationship between these indicators and osteopenia in Cd-exposed subjects is examined. MATERIALS AND METHODS Target Population Epidemiological studies for evaluating chronic Cd intoxication were conducted by Ishikawa Prefectural Health Authorities in the Cd-polluted Kakehashi River Basin in Ishikawa Prefecture in Japan twice in 1974, 1975, and in 1981, 1982. All inhabitants over 50 years of age in this polluted area were examined each time. Details about these two epidemiological studies have been described previously (Nogawa et al., 1978; Ishikawa Prefecture, 1984). Of the inhabitants, 207 were found to have Cd-induced renal tubular dysfunction and they were officially recognized as "subjects requiring observation" by the Research Committee organized by the Prefectural Health Authorities. In the present study 203 inhabitants (91 men and 112 women) from these 207 inhabitants were selected as Cd-exposed subjects. No patients were receiving renal dialysis or estrogen therapy at the time of this examination in 1983. The nonexposed subjects consisted of 25 men and 55 women living in an urban Cdnonpolluted area located about 30 km northeast of the Kakehashi River Basin. All of them were over 60 years and visited our hospital for a health check-up conducted annually by the Department of Public Health in our university. Means and standard deviations of ages in both groups of subjects are shown in Table 1. The M D M e t h o d To assess the degree of bone density, an X-ray of the hands alone with an aluminum step-wedge was taken. The bone density was measured at the middle site of the metacarpal bone II with a microdensitometer. Details regarding use of the microdensitometer have been described previously (Inoue et al., 1983). From the densitometric pattern shown in Fig. 1, the bone width D, marrow

Cd-INDUCED RENAL DYSFUNCTION AND OSTEOPENIA

Optical densily Step

73

No.

: radial side I_

~

ulnar side O

r

I

~'~d"l~///d'/-~/d~'f.4 U t

- 1.5

--~'.~---2

{---{ ................... I

~

~ / ~ , ~G//S~//r~//~G$1minI

".ce .....

(2): d (3): GSmin

--~-~---3

(4): (.3$rfla,(=GS, .I.('~$2

W/-_=__:=:=:t____4________ __ __:::__=::::_____=_____y___~_ - - - 6

P-o.s

--L'V-'

FiG. 1. Densitometer chart.

width d, and cortical width d 1 (cortical width on the radial side) and d 2 (cortical width on the ulnar side), and peak GS 1 (peak of the cortex on the radial side), GS2 (peak of the cortex on the ulnar side), and GSmin (peak of the middle positions of the bone marrow) were measured and converted into the number of steps on the aluminum wedge and the following five indices were studied: (1) M C I

-

dl + d E D

(2) d

(3) GSmax -

GS1 + GS2

(4) GSmin (5) "ZGS/D ZGS is the value obtained by the computer integrating the pattern area which is obtained optico-densitometrically and converted into the number of steps on the aluminum step-wedge. The value obtained by dividing EGS by bone width D corresponds to the bone mineral content determined by photon absorptiometry. For the five obtained indices, the degree of osteopenia was scored as point 0 for normal, point 1 for degree I, point 2 for degree II, and point 3 for degree III. On the basis of the sum of the given scores, a synthetic assessment was established.

Analysis of Blood and Urine oq-mg and 132-mgin urine and serum(U-Ctl-mg, S-Ctl-mg, U-132-mg, and S-132-mg) were determined by enzyme immunoassay using kits ( a r m EIA kit and ~32-mEIA kit, FujiRebio Co., Japan). Creatinine in urine and serum was analyzed by Jaffe's method (Bonsnes and Taussky, 1945). Calcium in urine and serum (U-Ca and S-Ca) was measured by atomic absorption spectrophotometry (Nogawa et al., 1979, 1987). Inorganic phosphorus in urine was determined by Allen's method (Allen, 1940) and that in serum (S-P) was analyzed by Taussky's method using a kit (P-Test Wako, Wako, Inc., Japan). Cd in urine and whole blood (U-Cd and

74

KIDO ET AL.

B-Cd) was directly measured with a Zeeman effect flameless atomic absorption spectrometer (Hitachi Co., Japan) without ashing or extraction (Kido et al., 1984). Measurement of creatinine clearance (Ccr), percentage tubular reabsorption of phosphate (%TRP), and base excess (BE) were performed as previously described (Nogawa et al., 1981). Using the same samples as for Ccr, clearances of calcium, phosphorus, oq-mg, and [32-mg (Cca, Cp, Cal-mg, C[3z-mg) were measured. Statistical Analysis

Differences between two means were tested with Student's t test. Stepwise backward regression analysis was performed to discover the predictor variables relating to each index of the MD method (Okuno et al., 1976). The boundary values of Fin = Fout = 2 were set to find a good subset. Among the finally selected predictor variables, those which showed significant standard partial regression coefficients were considered to have associations with dependent variables. Other predictor variables seemed to be less related to dependent variables.

RESULTS Mean age and biological parameters in both sexes of Cd-exposed and nonexposed subjects are seen in Table 1. Mean U-Cd and B-Cd concentrations in both sexes of Cd-exposed subjects were significantly higher than those in nonexposed subjects. U-[32-mg and S-Cr in Cd-exposed subjects were also significantly higher than those in nonexposed subjects. S-Ca and S-P in Cd-exposed subjects were significantly lower than those in nonexposed subjects. TABLE 1 BIOLOGICAL PARAMETERSIN Cd-ExPOSED AND NONEXPOSED SUBJECTS Cd-exposed subjects N Men Age (year) a U-Cd (la,g/gcr)b U-132-mg (~g/gcr)b S-Cr (rag/100 ml) b S-Ca (mg/100 ml)a S-P (mg/100 ml) a B-Cd (p~g/liter)b Women Age (year) a U-Cd (txg/gcr)b U-132-mg (ixg/gcr)b S-Cr (mg/100 ml) b S-Ca (mg/100 ml) a S-P (mg/t00 ml)a B-Cd (p,g/liter) b

Nonexposed subjects

Mean

SD

N

Mean

SD

91 91 91 91 91 91 91

70.8 7.82 2138.8 1.16 9.13 2.53 7.92

9.7 1.96" 8.0* 1.26" 0.60* 0.32* 1.87"

25 25 25 25 25 25 25

72.4 2.53 139.2 1.00 9.53 3.30 3.10

6.6 1.59 4.0 1.14 0.16 0.11 1.72

112 112 112 112 112 112 112

70.7 11.02 3831.5 0.96 9.05 2.75 8.74

7.9 2.09* 8.2" 1.37" 0.65* 0.40* 1.91 *

55 55 55 55 55 55 55

69.0 4.03 174.3 0.87 9.60 3.41 3.22

5.8 1.45 3.5 1.21 0.18 0.11 1.51

a Arithmetric mean and arithmetric standard deviation are calculated. b Geometric mean and geometric standard deviation are calculated. * Significant difference (P < 0.05) compared with nonexposed subjects.

C d - I N D U C E D RENAL D Y S F U N C T I O N A N D O S T E O P E N I A

75

Simple correlation coefficients between age, biological parameters, and MD indices are shown in Table 2. In Cd-exposed subjects, age showed significant correlations with each MD index. U-[32-mg was also significantly correlated with each MD index. In nonexposed women, age and B-Cd concentration were significantly correlated with most MD indices. Table 3 shows predictor variables selected from age and biological parameters using stepwise backward regression and the significance of their standard partial regression coefficients. In Cd-exposed men, the multiple correlation coefficients between selected biological parameters and MD indices were significant. Age, U-[32-mg, and S-Cr showed significant standard partial regression coefficients. In women, significant multiple correlation coefficients between selected biological parameters and MD indices were found in both Cd-exposed and nonexposed subjects. Age showed significant standard partial regression coefficients in both groups of subjects. However, U-132-mg showed a significant standard partial regression coefficient only in Cd-exposed women. In addition to the biological parameters in Table 1, other biological parameters and renal function tests were examined in Cd-exposed subjects, the results of which are seen in Table 4. Significant sex difference was found only in BE. Simple correlation coefficients between biological parameters, renal function tests in Table 4, and each MD index are found in Table 5. U-oq-mg and C[32-mg showed significant correlations with most MD indices in both sexes. S-[32-mg, Ccr, Ca~-mg, and BE showed significant correlations with each MD index only in Cd-exposed women. Items selected from age, biological parameters, and renal function tests in Tables 1 and 4 using stepwise backward regression analysis are shown in Table 6. Significant multiple correlation coefficients were found in both sexes. In men, U-132-mg, S-Cr, S-[32-mg, and C[32-mg showed significant standard partial regression coefficients. In women, age, U-132-mg, S-Cr, S-[32-mg, Cp, and %TRP showed significant standard partial regression coefficients. Compared with the values of adjusted R 2 corresponding to each MD index in Tables 3 and 6, the values in Table 6 were always higher than those in Table 3. DISCUSSION There are many factors which may lead to a reduction of bone content, including such common factors as malnutrition, prolonged periods of immobilization due to accident or illness, steroid therapy, and estrogen deficiency (Grech et al., 1985). Decrease in bone mass is also common in chronic renal failure. Bone resorption slightly exceeds formation after the fourth decade. There is no doubt that postmenopausal women lose bone at a great rate than do men. Therefore, age and sex are important factors to consider when studying osteoporosis. In the present study, subjects were divided by sex. Multiple regression analysis including age was used to predict variables to reveal the influence of age. The stepwise procedure which we used is a popular tool for selecting a candidate regression model. It is not, however, without problems and Wetherill has stated that standard stepwise strategies, with F levels used, are open to very serious criticisms if any meaning is given to significance (Wetherill et al., 1986).

76

KIDO ET AL.

Z <

II

I I I

I I I

I

II

I I I I

I I I

I

II

II

II

I

t~

z

z < i-

I I I

II

0

O

Z <

II

I

II

I I I I

I

oq [-

M < I I I

I l l

I l l l l

I

I I I

I11

I I I I I

I

I I I

I I I

I I I l l

I

0

e

< Z 0 < Z

r,1 0 r..)

I

I

2; < M

vv I I I

I

eo~,

d~

0

I

I I I I I

I

55

Cd-INDUCED

RENAL

DYSFUNCTION

AND

77

OSTEOPENIA

I

II

I Z

H

u~

."2.

=

I

]

I

I

0 m

z

Z <

I

I

I

I

I

I

m.m.

I Z

/

I

Z

i

I

I

I

I

Z

I

I

I

e~ e~ N N M N r~

] "0 t~tt~ mm

,,~--

~

~ o

I

M ~ O

0 M

i

~ ' ,"~

i

,

t J ~

i

= ,-~

o-o - N ~

Z ~

N

78

K I D O ET AL. TABLE 4 BIOLOGICAL PARAMETERS IN Cd-ExPOSED MEN AND WOMEN Men

S-13z-mg (~g/liter) ~ S - a l - m g (~g/liter) b U - a l - m g (ixg/gcr) b Ccr (ml/min) a Cca (ml/min) b Cp (ml/min) b % T R P (%)a C[32-mg (ml/min) b C a l - m g (ml/min) b BE d

Women

N

Mean

SD

N

Mean

SD

91 91 91 91 91 91 91 9t 9t 91

2540.2 21525.6 18269.7 58.4 0.64 13.21 76.2 0.52 0.53 -0.46

1.4 1.3 2.4 22.9 2.18 5.05 7.4 6.82 2.18 1.66

112 112 112 112 112 112 112 112 112 112

2455.3 19870.7 20660.9 60.3 0.76 12.41 77.5 0.83 0.55 - 1.08

1.5 1.3 3.2 22.4 2.18 4.68 10.0 6.22 2.67 2.59*

a Arithmetric m e a n and arithmetric standard deviation are calculated. b Geometric m e a n and geometric standard deviation are calculated. * Significant difference (P < 0.05) c o m p a r e d with Cd-exposed men.

One possible solution to this dilemma is to use an "all possible regressions" approach. However, it is more problematic for the models in Table 6 where the number of variables has been increased. Stepwise procedures were designed to solve this problem. We think that our procedure setting the boundary values as Fin = Fout = 2 is appropriate since selected variables using this procedure were closely consistent with those selected by other procedures such as calculating prediction sum of squares or Akaike's information criterion (Okuno et al., 1976). As results, it was found that U-[32-mg and S-Cr showed significant associations with each MD index in Cd-exposed men and that on the contrary, age was significantly associated with each MD index in both Cd-exposed and nonexposed women. However, U-[32-mg showed significant associations with each MD index only in Cd-exposed women. Further study increasing the number of predictor variables revealed significant associations among S-[32-mg, Ccr, CI32-mg, and some MD indices in Cd-exposed men and also among S-132-mg, Cp, %TRP, and some MD indices in Cd-exposed women, in addition to the variables selected in Table 3. The values of adjusted R 2 in Table 6 are higher than those in Table 3. Therefore, a relationship between renal dysfunction and osteopenia in Cdexposed subjects was demonstrated in our study. Regarding this relationship, another researcher demonstrated significant associations in Cd-exposed subjects in the Jinzu River Basin where Itai-itai disease patients are found (Aoshima et al., 1988). She divided the subjects into five groups on the basis of the level of fractional excretion of [32-mg. MD indices demonstrated deterioration of osteopenia in proportion to the severity of renal dysfunction. With the exception of one group in men, there were no significant differences in the mean age in any group. However, the relationship between osteopenia and age was not calculated directly in her study. Regarding this point, our present study for the first time reveals a relationship between osteopenia and Cd-induced renal dysfunction independent of age.

Cd-INDUCED

RENAL

DYSFUNCTION

AND

79

OSTEOPENIA

TABLE 5 SIMPLE CORRELATION COEFFICIENTS AMONG BIOLOGICAL PARAMETERS, RENAL FUNCTION TESTS AND MICRODENSITOMETRICAL INDICES IN C d - E x P O S E D SUBJECTS MCI

d

GSmax

GSmin

EGS/D

Men S-132-mg

- 0.080

0.046

- 0.176

- 0.175

- 0.215"

S-cq-mg

- 0.010

0.016

0.038

- 0.018

- 0.014

U-al-mg/Cr

- 0.257*

0.237*

- 0.121

- 0.174

- 0.161

Ccr Cca Cp

0.017 -0.178

-0.039 0.157

0.012 -0.139

-0.001 -0.175

0.014 -0.149

0.090

- 0.104

- 0.010

0.019

%TRP

- 0.101

0.095

- 0.016

- 0.014

- 0.014 0.038

CI3z-mg

- 0.260*

0.231"

- 0.200

- 0.246*

- 0.237*

Ccq-mg

- 0.224*

0.199

- 0.126

- 0.160

- 0.152

BE

-0.144

0.184

-0.092

-0.151

-0.065

S-132-mg

- 0.319"*

0.244**

- 0.257**

- 0.328**

- 0.326**

S-eq-mg

-0.195"

0.151

-0.223*

-0.133

-0.191"

U-cq-mg/Cr

- 0.343**

0.299**

- 0.335**

- 0.387**

- 0.377**

0.263**

0.257**

- 0.012

0.015

0.017

Women

Ccr Cca

0.341"*

-0.271'*

0.082

- 0.061

Cp

0.199"

- 0.188"

0.078

0.096

0.090

%TRP

0.167

- 0.071

0.221"

0.202*

0.244**

0.206*

CI3z-mg

- 0.293**

0.275**

- 0.303**

- 0.357**

- 0.338**

Ccq-mg

- 0.298**

0.256**

- 0.287**

- 0.356**

- 0.329**

0.281"*

0.314"*

0.331"*

BE

0.297**

-0.211"

* S i g n i f i c a n t ( P < 0.05). ** S i g n i f i c a n t ( P < 0.01).

In addition, some Japanese researchers, who dispute the effects of Cd on Itaiitai disease, have stressed that this disease appears only in endemic districts (Takeuchi, 1978; Kajikawa, 1978). From this view, it is very important that this study clarify the relationship between Cd-induced renal dysfunction and bone effects in another Cd-polluted area in Japan. We have already described that there may be three main mechanisms for the development of Cd-induced bone damage. The first mechanism is that Cd causes renal damage since vitamin D metabolism occurs in the kidneys. The kidney damage can lead in turn to an internal vitamin D deficiency which, then, induces a reduction of calcium in the bone. The second mechanism is that Cd decreases gastrointestinal calcium absorption which, in turn, leads to reduced availability of calcium in the body and may induce the bone decalcification found is osteoporosis. The third mechanism is that Cd affects bone collagen metabolism directly which is shown by a reduction in lysyl-oxidase actiTity. In the present study, we could find no significant associations between osteopenia and U-Cd, B-Cd, or S-Ca. Only the relationship between renal dysfunction and osteopenia was significant. Besides, our former study reported that loL,25(OH)zD levels are lower in Itai-itai disease patients and Cd-exposed subjects

80

KIDO E T A L .

t~ to

¢5¢5 Z

2~

Z

,d

I

I

l

I

I

I

I

I

I

I

o

~o oo

Z

I

I

I

I

I

I

i

It

I

I

I

II

z~ r~

U~

,.J

¢)

@ ,d

¢.

~_~_~

~ooo

__=

~

Cd-INDUCED RENAL DYSFUNCTION AND OSTEOPENIA

81

with renal damage than in nonexposed subjects. These facts suggest a close relationship among renal dysfunction, vitamin D metabolism, and osteopenia, and thus support the first mechanism. In some animal experiments, it was demonstrated that Cd-induced skeletal changes occur independent of Cd-induced renal dysfunction (Yoshiki et al., 1975; Bhattacharyya et al., 1988). However, the Cd exposure levels were much higher than those reported for the diets of inhabitants exposed to Cd. Therefore, in humans exposed to environmental Cd, the mechanism of Cd-induced bone effects may not be necessarily consistent with the mechanism studied by high-Cd-dose experiments. Other factors, such as low intake of calcium, vitamin D, and protein, pregnancy, and menopause, may also affect the development of bone disease. ACKNOWLEDGMENTS The authors are grateful to Fujisawa Pharmaceutical Co, Ltd. and Teijin Bio-Laboratories Inc. for their assistance in microdensitometry.

REFERENCES Allen, R. J. L. (1940). The estimation of phosphorus. Biochem. J. 34, 858--865. Aoshima, K., Iwata, K., and Kasuya, M. (1988). Environmentalexposure to cadmium and effects on human health. 2. Bone and mineral metabolism in inhabitants of the cadmium-polluted Jinzu River basin in Toyama Prefecture. Japan. J. Hyg. 43, 864-871. (in Japanese) Bhattacharyya, M. H., Whelton, B. D., Peterson, D. P., Carnes, B. A., Morett, E. S., Toomey, J. M., and Williams, L. L. (1988). Skeletal changes in multiparous mice fed a nutrient-sufficient diet containing cadmium. Toxicology 50, 193-204. Bernard, A. M., Moreau, D., and Lauwarys, R. (1982). Comparison of retinol-binding protein and 132-microglobulindetermination in urine for the early detection of tubular proteinuria. Clin. Chim. Acta 126, 1-7. Bonsnes, R. W., and Taussky, H. H. (1945). On the colorimetric determination of creatinine by the Jaffe reaction. J. Biol. Chem. 158, 581-591. Foulkes, E. C. (1986). "Cadmium," pp. 135-178. Springer-Verlag, Berlin. Friberg, L., Elinder, C. G., Kjellstr/~m, T., and Nordberg, G. F. (1986). "Cadmium and Health: A Toxicological and Epidemiological Appraisal," Vol. II, "Effects and Response," pp. 21-204. CRC Press, Boca Raton, FL. Grech, P., Martin, T. J., Barrington, N. A., and Ell, P. J. (1985). "Diagnosis of Metabolic Bone Disease," pp. 54-62. Chapman & Hall Medical, London. Hayashi, Y., Kamidate, T., Miyata, K., Furukawa, T., and Shoji, T. (1986) Analytical reproducibility of microdensitometry-online-computer system for quantitative analysis of atrophy in metacarpal bone. J. Bone Miner. Metab. 4, 31-39. lnoue, T., Kusida, K., Miyamoto, S., and Sumi, Y. (1983). Quantitative assessment of bone density on X-ray picture. J. Japan. Orthop. Assoc. 57, 1923-1936. lshikawa Prefecture. (1984). "Result of the Epidemiological Study on the Health Effects of Cadmium on the Population in The Kakehashi River Basin, pp. 1-56. Ishikawa Prefecture, Kanazawa. (in Japanese) Ishizaki, A., Tanabe, S., Matsuda, S., and Sakamoto, M. (1966). Experimental study on the chronic cadmium poisoning in relation to calcium deficiency. Japan. J. Hyg. 20, 398-404. (in Japanese) Kajikawa, K. (1978). Pathogenesis of Itai-itai disease based on post-mortem studies. In "Cadmium Studies in Japan: A Review" (K. Tsuchiya, Ed.), pp. 286-295. Kodansha LTD and Elsevier, Tokyo and Amsterdam. Kido, T., Honda, R., Yamada, Y., Tsuritani, I., Ishizaki, M., and Nogawa, K. (1985). ~rMicroglobalin determination in urine for the early detection of renal tubular dysfunctions caused by exposure to cadmium. Toxicol. Lett. 24, 195-201. Kido, T., Nogawa, K., Yamada, Y., Honda, R., Tsuritani, I., Ishizaki, M., and Yamaya, H. (1989).

82

K I D O ET A L .

Osteopenia in inhabitants with renal dysfunction induced by exposure to environmentalcadmium. Int. Arch. Occup. Environ. Health 61, 271-276.

Kido, T., Tsuritani, I., Honda, R., Ishizaki, M., Yamada, Y., and Nogawa, K. (1984). A direct determination of urinary cadmium by graphite furnace atomic absorption spectrometry using the zeeman effect. J. Kanazawa Med. Univ. 9, 70-75. (in Japanese) Kitagawa, M., Miwa, A., and Mural, Y. (1983). Pathological study on twenty seven autopsied cases with Itai-itai disease in Toyama Prefecture. Kankyo Hoken Report 49, 140-145. (in Japanese) Matsuda, S. (1967). Experimental study of chronic cadmium poisoning by cadmium compound insoluble in water. J. Juzen Med. Soc. 76, 239-253. (in Japanese) Miyamoto, S., Inoue, T., Kushida, K., Murata, H., Denda, M., Suzuki, H., and Yamashita, G. (1984). Osteoporosis in rheumatoid arthritis particularly with quantitative assessment on X-ray picture. J. Bone Miner. Metab. 2, 21-28. (in Japanese) Mukawa, A., Nogawa, K., and Hagino, N. (1980). Bone biopsy performed on women living in the cadmium-polluted Jinzu River basin. Japan. J. Hyg. 35, 761-773. (in Japanese) Nakamura, T., and Yoshikawa, S. (1984). Quantitative analysis on the changes of bone density in patients with renal osteodystrophy treated with 1 alpha-hydroxyvitamin D 3. J. Bone Miner. Metab. 2, 35-43. Nogawa, K. (1981). Itai-itai disease and follow-up studies. In "Cadmium in the Environment. Part II" (J. O. Nriagu, Ed.), pp. 1-37. Wiley, New York. Nogawa, K., Ishizaki, A., and Kawano, S. (1978). Statistical observations of the dose-response relationships of cadmium based on epidemiological studies in the Kakehashi River Basin. Environ. Res. 15, 185-198. Nogawa, K., Kido, T., Yamada, Y., Tsuritani, I., Honda, R., Ishizaki, M., and Terahata, K. (1984). oq-Microglobulin in urine as an indicator of renal tubular damage caused by environmental cadmium exposure. Toxicol. Lett. 22, 63-68. Nogawa, K., Kobayashi, E., Honda, R., and Ishizaki, A. (1979). Clinico-chemical studies on chronic cadmium poisoning. 1. Results of urinary examinations. Japan. J. Hyg. 34, 407-414. (in Japanese) Nogawa, K., Kobayashi, E., Honda, R., Ishizaki, A., Kawano, S., Ohmura, T., Nakagawa, H., Toga, H., and Matsuda, H. (1981). Clinico-chemical studies on chronic cadmium poisoning. 5. Renal functions. Japan. J. Hyg. 36, 512-517. (in Japanese) Nogawa, K., Tsuritani, I., Kido, T., Honda, R., Yamada, Y., and Ishizaki, M. (1987). Mechanism for bone disease found in inhabitants environmentally exposed to cadmium: Decreased lc~,25-dihydroxyvitamin D level. Int. Arch. Occup. Environ. Health 59, 21-30. Okano, K., Shimizu, M., Yamada, Y., Kou, S., Nawa, C., Ohira, K., Sasagawa, K., and Someya, K. (1985). Effect of 1,25-dihydroxyvitamin D3 on senile osteoporosis. J. Bone Miner. Metab. 3, 51-58. (in Japanese) Okuno, T., Haga, T., Yazima, K., Okuno, C., Hashimoto, S., and Furukawa, Y. (1976). "Multivariate Analysis. Part II," pp. 17-75. Nikkagiren, Tokyo. (in Japanese) Takeuchi, J. (1978). Cadmium vs. nutrition in Itai-itai disease. In "Cadmium Studies in Japan: A Review" (K. Tsuchiya, Ed.), pp. 283-286. Kodansha LTD and Elsevier, Tokyo and Amsterdam. Wetherill, G. B. (1986). "Monographs on Statistics and Applied Probability. Regression Analysis with Applications. Chapman & Hall, New York. Yoshikawa, S., Shiba, M., Hoshino, T., Igarashi, M., Orimo, H., Sakuma, A., and Tsuyama, N. (1983). Effect of eel calcitonin derivative (Elcatonin) in osteoporosis. J. Japan. Orthop. Assoc. 57, 1717-1728. (in Japanese) Yoshiki, S., Yanagisawa, T., Kimura, M., Otaki, N., Suzuki, M., and Suda, T. (1975). Bone and kidney lesions in experimental cadmium intoxication. Arch. Environ. Health 30, 559-562.