Human & Experimental Toxicology http://het.sagepub.com/
Chronic Occupational Lead Exposure and Testicular Endocrine Function A.J. McGregor and H.J. Mason Hum Exp Toxicol 1990 9: 371 DOI: 10.1177/096032719000900602 The online version of this article can be found at: http://het.sagepub.com/content/9/6/371
Published by: http://www.sagepublications.com
Additional services and information for Human & Experimental Toxicology can be found at: Email Alerts: http://het.sagepub.com/cgi/alerts Subscriptions: http://het.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://het.sagepub.com/content/9/6/371.refs.html
>> Version of Record - Jan 1, 1990 What is This?
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
Chronic Occupational Lead Function A. J.
McGregor
&
Exposure and Testicular Endocrine
H. J. Mason
Occupational Medicine and Hygiene Laboratories, Health and Safety Executive, 403 Edgware Road, London NW2
6LN, UK
The effects of moderate lead exposure on testicular endocrine function were evaluated in a group of 90 males who were occupationally exposed to inorganic lead. Lead concentrations in blood and bone were measured as indices of short-term and long-term, cumulative exposure, respectively. The results of this study show that the lead exposure levels encountered in the UK at present may result in a subclinical increase in follicle stimulating hormone (FSH), which is related to blood lead levels. This suggests that lead may be causing some subclinical primary damage to the seminiferous tubules in the testes. However, at blood lead levels of less than -1 this effect on serum FSH is not apparent. There was no significant effect on 47 μg dl serum testosterone concentrations or the free testosterone index. The mean luteinizing hormone (LH) level in the exposed group was found to be lower than in the control population. However, there appeared a confoundingly significant positive correlation between serum luteinizing hormone levels and the length of occupational lead exposure within the exposed group.
Introduction Several authors have reported that occupational exposure to lead has resulted in high frequencies of abnormal sperm production and effects on the hypothalamic-pituitary-testicular axis in men with long-term or high level lead exposure.l°z Rodamilans et a1.3 recently evaluated testicular endocrine function in several workers who work in the Spanish lead smelting industry. This study suggested that the first stage of exposure to lead involves testicular toxicity which affects synthesis of testosterone, and also that longer term exposure to lead may produce toxic effects at the hypothalamic or pituitary level. A recent revieW4 on the reproductive effects of occupational exposure to lead in males suggested that there was some association between blood lead in the range 40-70 vg dl-1 and reproductive effects on sperm counts and hormonal parameters. However the dose-effect relationships of these effects were still unclear. This paper reports a study of endocrine testicular function in 90 men at present day occupational exposure levels in the UK. Biochemical variables measured included testosterone, follicle stimulating hormone (FSH), luteinizing hormone (LH)
and
hormone binding globulin (SHBG). indices measured in the lead workers were blood lead, tibia lead and the length of occupational exposure. sex
Exposure
Population
and methods
Ninety men, who were all current lead workers aged between 16 and 60 years, with blood lead values between 17 and 77 u.g dl-1 and bone lead concentrations ranging up to 93 ~g g wet bone, formed the study group. None of the workers had clinical symptoms of lead poisoning and they were all being routinely monitored by blood lead measurements under the Approved Code of Practice for the Control of Lead at Work Regulations 1980 (revised 1985). A detailed medical and social questionnaire, including information on alcohol consumption (measured in units per week) was completed by each worker. Blood samples were taken between approximately 10 a.m. and 5 p.m. from resting subjects by the factory doctor or Employment Medical Adviser. The blood samples for hormone
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
372
analysis were centrifuged within 30 min of collection and sera stored at -20°C until transportation to the laboratory, where they were stored at -70°C until analysis. The referent subjects were male, manual workers of similar socio-economic
subjects had no known exposure to lead and were aged between 19 and 60 years. Blood lead measurements were performed on 58 of 86 (67%) of this control population; all except two had blood leads less than 10 [tg dl-1, the other two values were 12 and 14 ~g dl-1. These blood lead values are consistent with published data on subjects with environmental lead exposure. in the UK.S5 The in-vivo measurement of tibia lead was carried out using X-ray fluorescence (IVXRF). There was a measurement sensitivity in these studies of 10 wg g wet bone.6 Blood lead was measured on an EDTAanticoagulated blood sample using Electrothermal Atomization Atomic Absorption Spectrometry. The precision and accuracy of the blood lead measurements are monitored by internal quality control schemes and participation in the UKEQAS external quality assurance programme. Thus the effects of lead on testicular endocrine function can be related to blood lead which reflects short-term exposure, and to bone lead, which reflects long-term chronic cumulative status. These control
occupational
exposure.~7
(DPC Ltd., Wallingford, UK). The free testindex (FTI%) was calculated from the
osterone
and SHBG levels. The for SHBG, LH, FSH the of assays imprecisions and testosterone at normal and abnormally high values for each analyte were: 3.1%, 5.2%; 13.2%, 7.1%; 6.2%, 5.8% and 6.3%, 6.8%, serum
testosterone
respectively. analyses were carried out using V3.0 (Mercia Software, Birmingham, Statgraphics UK). Those variables showing significant nonnormal distributions were log-transformed before Statistical
analysis. Results mean blood lead value in the group 45.9 wg g (range 17-77 vg dl- ), the mean tibia lead value was 17.5 tLg g wet bone (range from below the detection limit to 93 vg g wet bone), and for length of occupational exposure to lead the mean was 11.5 years (range 1-45 years). The differences in age and biochemical variables between the exposed and control groups are summarized in Table 1. The mean serum LH value is significantly lower in the occupationally exposed subjects than in the referent group (P = 0.014). Age was also significantly different between the two groups (P < 0.001). No other variable was significantly different on a group mean basis. Age-related biochemical variables were identified in the control group by linear regression analyses between these measurements and age. Those variables found to be significantly age related were testosterone (r = -0.532, P = < 0.001), the free testosterone index
The
exposed
was
=
Samples from the exposed and referent groups were randomized before analysis for reproductive hormones. Plasma testosterone was measured by radio-immunoassay, FSH and LH were measured by immuno-radiometric assays (IDS Ltd., Washington, UK). Sex hormone binding globulin was measured by immuno-radiometric assay
Age and biochemical variables: a summary of the means, standard deviations (s.d.) and levels of statistical significance between exposed and control groups (P) are shown. Variables marked were log-transformed before analysis, geometric means and s.d.’s shown. Table 1
*
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
373
(FTI%) (r -0.462, P < 0.001) and follicle stimulating hormone (r 0.314, P 0.006). LH and SHBG did not show a significant age 0.9348 and r 0.009, P relationship (r =
=
=
=
=
0.2193,
P
=
=
0.0327, respectively)
=
in the control
group. As mean age was significantly different between the two groups further analyses relating lead exposure to FSH, testosterone and the FTI were carried out using the observed measurement for each exposed worker minus the expected value for his age, calculated from the age regression analysis within the control population. Table 2 shows the means and their 95% confidence intervals for these observed minus expected differences and their correlation with possible explanatory variables, namely blood lead, tibia lead and alcohol consumption. For these variables to have had no significant effect the observed minus expected value should have a mean close to zero and 95% confidence intervals should encompass zero. However for variables requiring log-transformation before analysis, the ratio of observed to expected value was used, hence for no significant effect the mean should lie close to one and the 95% confidence intervals encompass one. FSH showed a significant increase in the exposed worker compared to the expected value for his age. No significant differences were found for testosterone or the FTI (P > 0.05), although both variables showed a decreased mean difference between the observed value in the exposed subject and his age-expected value. This increase in FSH was significantly related to blood lead levels. However further analysis showed that this relationship was not significant (P > 0.05) after removal of all subjects with blood lead concentrations greater than 47 vg dl-1. Fifty-one subjects had blood lead levels of less than 47 ag dl7l. The geometric means (and standard deviations) for FSH in those subjects with either blood lead levels of less than 47 vg dl-1 or greater than 47 vg dl-1 were 3.6(1.52) and
Table 3 shows the correlations within the for LH and SHBG which were not shown to be significantly related to age in the control population. LH shows a more positive correlation with length of occupational lead exposure than with tibia lead or age, although these correlation coefficients are not significantly different from each other and it had previously been shown (Table 1) that the mean LH was significantly reduced in the exposed group compared with the controls. SHBG showed a significant correlation with age within the exposed group although this correlation was not apparent within the referent population. Multiple regression analyses using age, blood and tibia lead, years exposure and present alcohol intake as explanatory variables were also performed within the exposed subgroup. The following equations were obtained:
exposed population
0.59 + 0.0134 (age) + 0.0076 (blood 1.615 + 0.0190 (years exposure) Testosterone 30.76 - 0.2034 (age) SHBG = 21.11 + 0.446 (age) log (FTI%) 0.322 - 0.0204 (age)
log (FSH) log (LH)
=
Pb)
=
=
=
These
multiple regression
models
suggested
that, of the testicular endocrine measurements, Table 3 Correlation analyses within the exposed population for those variables not age related in the referent population.
4.5(1.63), respectively. Table 2 Observed minus expected differences of the age-related biochemical variables in the lead exposed group: Means and 95% confidence intervals (C.I.). Variables marked * were log-transformed before analysis. Data are shown antilogged i.e. as geometric mean ratios and their confidence intervals.
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
374
influenced by a lead dose indicator, The model for LH suggested that the number of years exposure was an influencing factor rather than the co-linearly related variables tibia lead or age. The proportion of the exposed population which fell outside the 95% boundaries calculated from the referent population were calculated for testosterone, FSH and LH. The x-square test with Yates correction was applied to this proportion of subjects lying outside the 95% boundaries. There was no significant difference (P > 0.05) between the numbers of exposed and referent subjects lying outside these boundaries for any of these three biochemical measureFSH
was
namely blood lead.
.
ments.
Discussion
’
Previous investigations into the effects of lead on testicular endocrine function have tended to look at men who show evidence of depressed endocrine function2 and depressed spermatogenesis.88 However it has been suggested that lead can have adverse effects on reproductive functions at levels of exposure which are within legal recommendations.44 This study attempted to reveal any possible effects of present UK occupational exposure levels on testicular endocrine function. Ninety occupationally exposed males were studied who were all being routinely monitored by blood lead measurements as described previously. Tibia lead was also measured on these men as it has been shown that this reflects long-term exposure.~ The measurements carried out to detecr testicular endocrine dysfunction were serum testosterone, sex hormone binding globulin, the derived free testosterone index, FSH and LH. It is well known that testicular function declines with age.9,1O Data showing age-related changes in total plasma testosterone, free testosterone, SHBG, LH and FSH have been reviewed The data from our non-occupationally lead-exposed referents showed a significant decrease in testosterone and the FTI with a significant increase in FSH with age. LH and SHBG were not significantly age related in the referent population over the span of ages encompassed. In view of this we have been careful in this study either to remove the influence of age by comparing the measured value in an exposed subject with the expected value obtained from age regression in the referents, or to use age as a confounding factor in multiple regression analyses in the exposed subjects. Our results indicate that moderate occupational
lead exposure results in a subclinical increase in FSH which is related to blood lead levels. This relationship between FSH and blood lead is not apparent in those subjects with a blood lead of less than 47 lAg dl-I. This effect of blood lead on FSH was confirmed by multiple regression analysis within the exposed subjects. Cullen et al. (1983)8 reported two moderately chronically lead patients (blood leads were 40 vg dl and 66 vg dl-1, exposed for 7 and 11 years, respectively) who had elevated FSH levels, the second of these two also had elevated LH levels. The elevation of basal FSH may be caused by a deficiency of the secretion by the Sertoli cells in the seminiferous tubules of the protein ’inhibin’~ ~~~~ which is believed to be involved in the negative feedback control at the pituitary of FSH secretion. The subclinical increases in FSH levels found for lead-exposed men in this study may support the idea of possible lead-induced toxic damage to the seminiferous tubules. Serum testosterone and the FTI were reduced, but not significantly in the exposed group compared to the age expected values. We also found that LH, which was shown to be unrelated to age in the control population, was significantly decreased in the occupationally exposed subjects compared to the referent group. However, regression analysis within the exposed group suggested that length of occupational exposure was related to increased LH levels. It has been reported elsewhere 10 that LH is positively age related thus the lower mean LH value which we found in the younger, exposed group (mean age 31.5 years) compared to the referents (mean age 40.6 years) may be due to the significant age difference between the two groups. Braunstein et al.2 reported that both lead poisoned and lead exposed subjects had reduced basal serum testosterone levels and normal SHBG, LH and FSH levels. This suggests a defect in the regulation of LH secretion as this normally rises as testosterone levels fall. However this was a small study, only comprising ten men, and may not be representative of a larger population. Also, all of the lead exposed and lead poisoned subjects had been treated with calcium EDTA in order to reduce their total body lead burden, therefore it is possible that some of the hormonal abnormalities noted in these people could have resulted from this
exposed
therapy. The results of our study showed that there was significant effect on testosterone or the FTI at the lead exposure levels encountered. In contrast, a study of 23 workers from a Spanish lead smelter reported by Rodamilans et a1.33 showed a decrease in testosterone levels in ten men described as having severe lead poisoning no
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
375
lead values 76 ± 11 wg dl-1, > 5 years exposure. He also suggested that the FTI was the best marker of testicular endocrine dysfunction in lead-exposed subjects. It was found that any alteration of this parameter correlated with the duration of exposure. They also found that the exposed subjects had an increase in LH levels, although there was no difference in FSH between exposed and control groups. In contrast we have found no significant effect on testosterone production but an effect on FSH which is related to blood lead. Lancranjan et a/.~ reported on the deleterious effects of lead on human reproduction at high and subtoxic (moderate) levels of exposure. The fertility of lead poisoned (mean 74 [tg lead dl-1 blood) and moderate absorption (mean 53 [tg lead dl-1 blood) groups was decreased. We have not addressed the question of fertility in this study; however the detection of an effect on FSH levels suggests that further study of the spermatogenesis process at current exposure levels would be valid. The reduction in mean LH levels noted in our exposed group is difficult to explain unless, as suggested previously,2 it is caused by an alteration in the central regulation of LH in the hypothalamic-pituitary axis. However, without stimulation tests of the hypothalamus and pituitary such an interpretation is speculative, especially given the positive relationship between LH and years exposure within the exposed subgroup. Blood lead levels may not always be a true indicator of susceptibility to lead toxicity. Raghaven et al. 14 identified a low molecular ratio lead-binding protein in the erythrocytes of industrially exposed workers which appeared to
(blood
non-toxic compartmentalization of lead. It has been suggested that workers with a diminished capacity for synthesizing this leadbinding protein may be at increased risk of developing lead toxicity at relatively low blood lead levels. 14 Inducible lead-binding proteins have also been found in rat kidney and brain,16 both of which are target organs for lead toxicity. Thus the relationship between indices of exposure and toxic effect may depend on an individual’s adaptive response to the lead dose. It is of interest that no testicular effects appeared to be related to tibia lead which has been shown to be a good indicator of chronic, cumulative lead exposure,~ the only testicular effect appears to be related to blood lead which is the current biological monitor of occupational lead exposure. The diverse published findings on male reproductive and endocrine function caused by lead may be explained both by the differing levels and duration of the exposure and by the differing responses of subjects to the toxin, such as the ability to induce intracellular binding proteins. None of the subjects we have studied show any clinical symptoms of lead toxicity, and their exposure histories represent exposure levels found within the UK at present. act as a
CytoSOI15
Acknowledgements We would like to thank the Toxic Metals section of the Occupational Medicine and Hygiene Laboratories, Health and Safety Executive, London for the blood lead analyses. This work arises from a collaborative programme with the Medical Physics Group, School of Physics and Space Research, University of Birmingham, whose contributions are gratefully acknowledged.
References 1
2
Lancranjan I, Popescu HI, Gavanescu O, Klepsch I & Serbanescu M. Reproductive ability of workmen occupationally exposed to lead. Archives of environmental health 1975; 30: 396-401. Braunstein GD, Dahlgren J &
chronically lead-poisoned
Loriaux DL.
inRodamilans M, Osaba MJ Mtz & 3
men.
Hypogonadism Infertility 1978; 1: 33-51. To-Figuaras J et al. The
toxicity on endocrine testicular function in an occupationally exposed population. Human Toxicology effects of lead
4
5
1988; 7: 125-8. Winder C. Reproductive and chromosomal effects of occupational exposure to lead in the male. Reproductive Toxicology 1989; 3: 221-33. Quinn MJ, Delves HT & Davies DHA. UK blood lead monitoring programme-results for 1987. In: Heavy Metals in the Environment, Volume 2, Geneva, International ConferEdinburgh CEP. Consultants, 1989. Somervaille LJ, Chettle DR & Scott MC. In-vivo measurement of lead in bone using X-ray fluorescence. Physics in medicine and biology 1985; 30: 929-43.
ence. 6
7
8
9 10
11
12
Somervaille LJ, Chettle DR & Scott MC et al. In vivo tibia lead measurements as an index of cumulative exposure in occupationally exposed subjects. British Journal of Industrial Medicine 1988; 45: 174-81. Cullen MR, Robins JM & Eskenazi B. Adult inorganic lead intoxication: Presentation of 31 new cases and a review of recent advances in the literature. Medicine 1983; 62: 221-47. Wilson JD & Foster DW (eds) Textbook of Endocrinology pp. 259-311. Philadelphia: W.B. Saunders, 1983. Harmon SM. Clinical aspects of the male reproductive system. In: The Aging Reproductive System, ed. Schneider EL. New York: Raven Press, 1978. Lipshultz LI, Greenberg SH, Caminos-Torres R & Snyder PS. Supranormal FSH response to Gonadotrophinreleasing hormone in oligospermic men with a normal basal serum FSH concentration. Clinical Endocrinology 1977; 7: 103-9. Steinberger A & Steinberger E. Secretion of an FSHinhibiting factor by cultured Sertoli cells. Endocrinology 1976; 99: 918.
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
376
13 Alvares AP. Lead and polychlorinated biphenyls: effects on heme and drug metabolism. Drug Metabolism Reviews 1979; 10: 91-106. 14 Raghaven SRV, Culver BD & Gonick HO. Erythrocyte
binding protein after occupational exposure. I. Relationship to lead toxicity. Environmental Research 1980; lead
22: 264-70. Lucier GW & Fowler BA. High affinity lead binding proteins in rat kidney cytosol mediate cell-free nuclear translocation of lead. Journal of Pharmacology and
16
Experimental Therapeutics 1985; 232: 462-9. Goering PL, Mistry P & Fowler BA. A low molecular weight lead-binding protein in brain attenuates lead inhibition of gamma-aminolevulanic acid dehydratase: comparison with a renal lead binding protein. Journal of Pharmacology and Experimental Therapeutics 1986; 237: 220-5.
15 Mistry P,
(Received
14 June
1990; accepted 15 June 1990)
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
Chronic Occupational Lead Exposure and Testicular Endocrine Function A.J. McGregor and H.J. Mason Hum Exp Toxicol 1990 9: 371 DOI: 10.1177/096032719000900602 The online version of this article can be found at: http://het.sagepub.com/content/9/6/371
Published by: http://www.sagepublications.com
Additional services and information for Human & Experimental Toxicology can be found at: Email Alerts: http://het.sagepub.com/cgi/alerts Subscriptions: http://het.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://het.sagepub.com/content/9/6/371.refs.html
>> Version of Record - Jan 1, 1990 What is This?
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
Chronic Occupational Lead Function A. J.
McGregor
&
Exposure and Testicular Endocrine
H. J. Mason
Occupational Medicine and Hygiene Laboratories, Health and Safety Executive, 403 Edgware Road, London NW2
6LN, UK
The effects of moderate lead exposure on testicular endocrine function were evaluated in a group of 90 males who were occupationally exposed to inorganic lead. Lead concentrations in blood and bone were measured as indices of short-term and long-term, cumulative exposure, respectively. The results of this study show that the lead exposure levels encountered in the UK at present may result in a subclinical increase in follicle stimulating hormone (FSH), which is related to blood lead levels. This suggests that lead may be causing some subclinical primary damage to the seminiferous tubules in the testes. However, at blood lead levels of less than -1 this effect on serum FSH is not apparent. There was no significant effect on 47 μg dl serum testosterone concentrations or the free testosterone index. The mean luteinizing hormone (LH) level in the exposed group was found to be lower than in the control population. However, there appeared a confoundingly significant positive correlation between serum luteinizing hormone levels and the length of occupational lead exposure within the exposed group.
Introduction Several authors have reported that occupational exposure to lead has resulted in high frequencies of abnormal sperm production and effects on the hypothalamic-pituitary-testicular axis in men with long-term or high level lead exposure.l°z Rodamilans et a1.3 recently evaluated testicular endocrine function in several workers who work in the Spanish lead smelting industry. This study suggested that the first stage of exposure to lead involves testicular toxicity which affects synthesis of testosterone, and also that longer term exposure to lead may produce toxic effects at the hypothalamic or pituitary level. A recent revieW4 on the reproductive effects of occupational exposure to lead in males suggested that there was some association between blood lead in the range 40-70 vg dl-1 and reproductive effects on sperm counts and hormonal parameters. However the dose-effect relationships of these effects were still unclear. This paper reports a study of endocrine testicular function in 90 men at present day occupational exposure levels in the UK. Biochemical variables measured included testosterone, follicle stimulating hormone (FSH), luteinizing hormone (LH)
and
hormone binding globulin (SHBG). indices measured in the lead workers were blood lead, tibia lead and the length of occupational exposure. sex
Exposure
Population
and methods
Ninety men, who were all current lead workers aged between 16 and 60 years, with blood lead values between 17 and 77 u.g dl-1 and bone lead concentrations ranging up to 93 ~g g wet bone, formed the study group. None of the workers had clinical symptoms of lead poisoning and they were all being routinely monitored by blood lead measurements under the Approved Code of Practice for the Control of Lead at Work Regulations 1980 (revised 1985). A detailed medical and social questionnaire, including information on alcohol consumption (measured in units per week) was completed by each worker. Blood samples were taken between approximately 10 a.m. and 5 p.m. from resting subjects by the factory doctor or Employment Medical Adviser. The blood samples for hormone
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
372
analysis were centrifuged within 30 min of collection and sera stored at -20°C until transportation to the laboratory, where they were stored at -70°C until analysis. The referent subjects were male, manual workers of similar socio-economic
subjects had no known exposure to lead and were aged between 19 and 60 years. Blood lead measurements were performed on 58 of 86 (67%) of this control population; all except two had blood leads less than 10 [tg dl-1, the other two values were 12 and 14 ~g dl-1. These blood lead values are consistent with published data on subjects with environmental lead exposure. in the UK.S5 The in-vivo measurement of tibia lead was carried out using X-ray fluorescence (IVXRF). There was a measurement sensitivity in these studies of 10 wg g wet bone.6 Blood lead was measured on an EDTAanticoagulated blood sample using Electrothermal Atomization Atomic Absorption Spectrometry. The precision and accuracy of the blood lead measurements are monitored by internal quality control schemes and participation in the UKEQAS external quality assurance programme. Thus the effects of lead on testicular endocrine function can be related to blood lead which reflects short-term exposure, and to bone lead, which reflects long-term chronic cumulative status. These control
occupational
exposure.~7
(DPC Ltd., Wallingford, UK). The free testindex (FTI%) was calculated from the
osterone
and SHBG levels. The for SHBG, LH, FSH the of assays imprecisions and testosterone at normal and abnormally high values for each analyte were: 3.1%, 5.2%; 13.2%, 7.1%; 6.2%, 5.8% and 6.3%, 6.8%, serum
testosterone
respectively. analyses were carried out using V3.0 (Mercia Software, Birmingham, Statgraphics UK). Those variables showing significant nonnormal distributions were log-transformed before Statistical
analysis. Results mean blood lead value in the group 45.9 wg g (range 17-77 vg dl- ), the mean tibia lead value was 17.5 tLg g wet bone (range from below the detection limit to 93 vg g wet bone), and for length of occupational exposure to lead the mean was 11.5 years (range 1-45 years). The differences in age and biochemical variables between the exposed and control groups are summarized in Table 1. The mean serum LH value is significantly lower in the occupationally exposed subjects than in the referent group (P = 0.014). Age was also significantly different between the two groups (P < 0.001). No other variable was significantly different on a group mean basis. Age-related biochemical variables were identified in the control group by linear regression analyses between these measurements and age. Those variables found to be significantly age related were testosterone (r = -0.532, P = < 0.001), the free testosterone index
The
exposed
was
=
Samples from the exposed and referent groups were randomized before analysis for reproductive hormones. Plasma testosterone was measured by radio-immunoassay, FSH and LH were measured by immuno-radiometric assays (IDS Ltd., Washington, UK). Sex hormone binding globulin was measured by immuno-radiometric assay
Age and biochemical variables: a summary of the means, standard deviations (s.d.) and levels of statistical significance between exposed and control groups (P) are shown. Variables marked were log-transformed before analysis, geometric means and s.d.’s shown. Table 1
*
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
373
(FTI%) (r -0.462, P < 0.001) and follicle stimulating hormone (r 0.314, P 0.006). LH and SHBG did not show a significant age 0.9348 and r 0.009, P relationship (r =
=
=
=
=
0.2193,
P
=
=
0.0327, respectively)
=
in the control
group. As mean age was significantly different between the two groups further analyses relating lead exposure to FSH, testosterone and the FTI were carried out using the observed measurement for each exposed worker minus the expected value for his age, calculated from the age regression analysis within the control population. Table 2 shows the means and their 95% confidence intervals for these observed minus expected differences and their correlation with possible explanatory variables, namely blood lead, tibia lead and alcohol consumption. For these variables to have had no significant effect the observed minus expected value should have a mean close to zero and 95% confidence intervals should encompass zero. However for variables requiring log-transformation before analysis, the ratio of observed to expected value was used, hence for no significant effect the mean should lie close to one and the 95% confidence intervals encompass one. FSH showed a significant increase in the exposed worker compared to the expected value for his age. No significant differences were found for testosterone or the FTI (P > 0.05), although both variables showed a decreased mean difference between the observed value in the exposed subject and his age-expected value. This increase in FSH was significantly related to blood lead levels. However further analysis showed that this relationship was not significant (P > 0.05) after removal of all subjects with blood lead concentrations greater than 47 vg dl-1. Fifty-one subjects had blood lead levels of less than 47 ag dl7l. The geometric means (and standard deviations) for FSH in those subjects with either blood lead levels of less than 47 vg dl-1 or greater than 47 vg dl-1 were 3.6(1.52) and
Table 3 shows the correlations within the for LH and SHBG which were not shown to be significantly related to age in the control population. LH shows a more positive correlation with length of occupational lead exposure than with tibia lead or age, although these correlation coefficients are not significantly different from each other and it had previously been shown (Table 1) that the mean LH was significantly reduced in the exposed group compared with the controls. SHBG showed a significant correlation with age within the exposed group although this correlation was not apparent within the referent population. Multiple regression analyses using age, blood and tibia lead, years exposure and present alcohol intake as explanatory variables were also performed within the exposed subgroup. The following equations were obtained:
exposed population
0.59 + 0.0134 (age) + 0.0076 (blood 1.615 + 0.0190 (years exposure) Testosterone 30.76 - 0.2034 (age) SHBG = 21.11 + 0.446 (age) log (FTI%) 0.322 - 0.0204 (age)
log (FSH) log (LH)
=
Pb)
=
=
=
These
multiple regression
models
suggested
that, of the testicular endocrine measurements, Table 3 Correlation analyses within the exposed population for those variables not age related in the referent population.
4.5(1.63), respectively. Table 2 Observed minus expected differences of the age-related biochemical variables in the lead exposed group: Means and 95% confidence intervals (C.I.). Variables marked * were log-transformed before analysis. Data are shown antilogged i.e. as geometric mean ratios and their confidence intervals.
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
374
influenced by a lead dose indicator, The model for LH suggested that the number of years exposure was an influencing factor rather than the co-linearly related variables tibia lead or age. The proportion of the exposed population which fell outside the 95% boundaries calculated from the referent population were calculated for testosterone, FSH and LH. The x-square test with Yates correction was applied to this proportion of subjects lying outside the 95% boundaries. There was no significant difference (P > 0.05) between the numbers of exposed and referent subjects lying outside these boundaries for any of these three biochemical measureFSH
was
namely blood lead.
.
ments.
Discussion
’
Previous investigations into the effects of lead on testicular endocrine function have tended to look at men who show evidence of depressed endocrine function2 and depressed spermatogenesis.88 However it has been suggested that lead can have adverse effects on reproductive functions at levels of exposure which are within legal recommendations.44 This study attempted to reveal any possible effects of present UK occupational exposure levels on testicular endocrine function. Ninety occupationally exposed males were studied who were all being routinely monitored by blood lead measurements as described previously. Tibia lead was also measured on these men as it has been shown that this reflects long-term exposure.~ The measurements carried out to detecr testicular endocrine dysfunction were serum testosterone, sex hormone binding globulin, the derived free testosterone index, FSH and LH. It is well known that testicular function declines with age.9,1O Data showing age-related changes in total plasma testosterone, free testosterone, SHBG, LH and FSH have been reviewed The data from our non-occupationally lead-exposed referents showed a significant decrease in testosterone and the FTI with a significant increase in FSH with age. LH and SHBG were not significantly age related in the referent population over the span of ages encompassed. In view of this we have been careful in this study either to remove the influence of age by comparing the measured value in an exposed subject with the expected value obtained from age regression in the referents, or to use age as a confounding factor in multiple regression analyses in the exposed subjects. Our results indicate that moderate occupational
lead exposure results in a subclinical increase in FSH which is related to blood lead levels. This relationship between FSH and blood lead is not apparent in those subjects with a blood lead of less than 47 lAg dl-I. This effect of blood lead on FSH was confirmed by multiple regression analysis within the exposed subjects. Cullen et al. (1983)8 reported two moderately chronically lead patients (blood leads were 40 vg dl and 66 vg dl-1, exposed for 7 and 11 years, respectively) who had elevated FSH levels, the second of these two also had elevated LH levels. The elevation of basal FSH may be caused by a deficiency of the secretion by the Sertoli cells in the seminiferous tubules of the protein ’inhibin’~ ~~~~ which is believed to be involved in the negative feedback control at the pituitary of FSH secretion. The subclinical increases in FSH levels found for lead-exposed men in this study may support the idea of possible lead-induced toxic damage to the seminiferous tubules. Serum testosterone and the FTI were reduced, but not significantly in the exposed group compared to the age expected values. We also found that LH, which was shown to be unrelated to age in the control population, was significantly decreased in the occupationally exposed subjects compared to the referent group. However, regression analysis within the exposed group suggested that length of occupational exposure was related to increased LH levels. It has been reported elsewhere 10 that LH is positively age related thus the lower mean LH value which we found in the younger, exposed group (mean age 31.5 years) compared to the referents (mean age 40.6 years) may be due to the significant age difference between the two groups. Braunstein et al.2 reported that both lead poisoned and lead exposed subjects had reduced basal serum testosterone levels and normal SHBG, LH and FSH levels. This suggests a defect in the regulation of LH secretion as this normally rises as testosterone levels fall. However this was a small study, only comprising ten men, and may not be representative of a larger population. Also, all of the lead exposed and lead poisoned subjects had been treated with calcium EDTA in order to reduce their total body lead burden, therefore it is possible that some of the hormonal abnormalities noted in these people could have resulted from this
exposed
therapy. The results of our study showed that there was significant effect on testosterone or the FTI at the lead exposure levels encountered. In contrast, a study of 23 workers from a Spanish lead smelter reported by Rodamilans et a1.33 showed a decrease in testosterone levels in ten men described as having severe lead poisoning no
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
375
lead values 76 ± 11 wg dl-1, > 5 years exposure. He also suggested that the FTI was the best marker of testicular endocrine dysfunction in lead-exposed subjects. It was found that any alteration of this parameter correlated with the duration of exposure. They also found that the exposed subjects had an increase in LH levels, although there was no difference in FSH between exposed and control groups. In contrast we have found no significant effect on testosterone production but an effect on FSH which is related to blood lead. Lancranjan et a/.~ reported on the deleterious effects of lead on human reproduction at high and subtoxic (moderate) levels of exposure. The fertility of lead poisoned (mean 74 [tg lead dl-1 blood) and moderate absorption (mean 53 [tg lead dl-1 blood) groups was decreased. We have not addressed the question of fertility in this study; however the detection of an effect on FSH levels suggests that further study of the spermatogenesis process at current exposure levels would be valid. The reduction in mean LH levels noted in our exposed group is difficult to explain unless, as suggested previously,2 it is caused by an alteration in the central regulation of LH in the hypothalamic-pituitary axis. However, without stimulation tests of the hypothalamus and pituitary such an interpretation is speculative, especially given the positive relationship between LH and years exposure within the exposed subgroup. Blood lead levels may not always be a true indicator of susceptibility to lead toxicity. Raghaven et al. 14 identified a low molecular ratio lead-binding protein in the erythrocytes of industrially exposed workers which appeared to
(blood
non-toxic compartmentalization of lead. It has been suggested that workers with a diminished capacity for synthesizing this leadbinding protein may be at increased risk of developing lead toxicity at relatively low blood lead levels. 14 Inducible lead-binding proteins have also been found in rat kidney and brain,16 both of which are target organs for lead toxicity. Thus the relationship between indices of exposure and toxic effect may depend on an individual’s adaptive response to the lead dose. It is of interest that no testicular effects appeared to be related to tibia lead which has been shown to be a good indicator of chronic, cumulative lead exposure,~ the only testicular effect appears to be related to blood lead which is the current biological monitor of occupational lead exposure. The diverse published findings on male reproductive and endocrine function caused by lead may be explained both by the differing levels and duration of the exposure and by the differing responses of subjects to the toxin, such as the ability to induce intracellular binding proteins. None of the subjects we have studied show any clinical symptoms of lead toxicity, and their exposure histories represent exposure levels found within the UK at present. act as a
CytoSOI15
Acknowledgements We would like to thank the Toxic Metals section of the Occupational Medicine and Hygiene Laboratories, Health and Safety Executive, London for the blood lead analyses. This work arises from a collaborative programme with the Medical Physics Group, School of Physics and Space Research, University of Birmingham, whose contributions are gratefully acknowledged.
References 1
2
Lancranjan I, Popescu HI, Gavanescu O, Klepsch I & Serbanescu M. Reproductive ability of workmen occupationally exposed to lead. Archives of environmental health 1975; 30: 396-401. Braunstein GD, Dahlgren J &
chronically lead-poisoned
Loriaux DL.
inRodamilans M, Osaba MJ Mtz & 3
men.
Hypogonadism Infertility 1978; 1: 33-51. To-Figuaras J et al. The
toxicity on endocrine testicular function in an occupationally exposed population. Human Toxicology effects of lead
4
5
1988; 7: 125-8. Winder C. Reproductive and chromosomal effects of occupational exposure to lead in the male. Reproductive Toxicology 1989; 3: 221-33. Quinn MJ, Delves HT & Davies DHA. UK blood lead monitoring programme-results for 1987. In: Heavy Metals in the Environment, Volume 2, Geneva, International ConferEdinburgh CEP. Consultants, 1989. Somervaille LJ, Chettle DR & Scott MC. In-vivo measurement of lead in bone using X-ray fluorescence. Physics in medicine and biology 1985; 30: 929-43.
ence. 6
7
8
9 10
11
12
Somervaille LJ, Chettle DR & Scott MC et al. In vivo tibia lead measurements as an index of cumulative exposure in occupationally exposed subjects. British Journal of Industrial Medicine 1988; 45: 174-81. Cullen MR, Robins JM & Eskenazi B. Adult inorganic lead intoxication: Presentation of 31 new cases and a review of recent advances in the literature. Medicine 1983; 62: 221-47. Wilson JD & Foster DW (eds) Textbook of Endocrinology pp. 259-311. Philadelphia: W.B. Saunders, 1983. Harmon SM. Clinical aspects of the male reproductive system. In: The Aging Reproductive System, ed. Schneider EL. New York: Raven Press, 1978. Lipshultz LI, Greenberg SH, Caminos-Torres R & Snyder PS. Supranormal FSH response to Gonadotrophinreleasing hormone in oligospermic men with a normal basal serum FSH concentration. Clinical Endocrinology 1977; 7: 103-9. Steinberger A & Steinberger E. Secretion of an FSHinhibiting factor by cultured Sertoli cells. Endocrinology 1976; 99: 918.
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
376
13 Alvares AP. Lead and polychlorinated biphenyls: effects on heme and drug metabolism. Drug Metabolism Reviews 1979; 10: 91-106. 14 Raghaven SRV, Culver BD & Gonick HO. Erythrocyte
binding protein after occupational exposure. I. Relationship to lead toxicity. Environmental Research 1980; lead
22: 264-70. Lucier GW & Fowler BA. High affinity lead binding proteins in rat kidney cytosol mediate cell-free nuclear translocation of lead. Journal of Pharmacology and
16
Experimental Therapeutics 1985; 232: 462-9. Goering PL, Mistry P & Fowler BA. A low molecular weight lead-binding protein in brain attenuates lead inhibition of gamma-aminolevulanic acid dehydratase: comparison with a renal lead binding protein. Journal of Pharmacology and Experimental Therapeutics 1986; 237: 220-5.
15 Mistry P,
(Received
14 June
1990; accepted 15 June 1990)
Downloaded from het.sagepub.com at HUNTER COLLEGE LIB on May 25, 2014
