A comparative study of zinc, copper, cadmium, and lead levels in fertile and infertile men

Vol. 40, No.5, November 1983 Printed in U.sA.

FERTILITY AND 8TEmLITY Copyright c 1983 The American Fertility Society

A comparative study of zinc, copper, cadmium, and lead levels in fertile and infertile men

Rosalind Stanwell-Smith, M.R.C.O.G.* Simon G. Thompson, M.A., Dip.Stat.* Andrew P. Haines,·M,R.C.P., M.R.C.G.P.*t Roberta J. Ward:j: Geoffrey Cashmore:j: Jitka Stedronska§ William F. Hendry, F.R.C.S.§ Northwick Park Hospital, Harrow, Middlesex, and Chelsea Hospital for Women, London, England

Eighty infertile men and 38 men of known fertility were studied for investigation of both the importance of zinc, copper, cadmium, and lead to fertility and the possible interrelationships between these trace elements. The infertile men had higher mean concentrations of plasma copper than those of proven fertility. The difference was statistically significant (P < 0.0l) but was of small magnitude (- 1.5 fLmol mean difference). The concentrations of plasma zinc, erythrocyte zinc, whole blood lead and cadmium, and seminal plasma zinc and copper did not differ significantly between infertile and fertile men. There was a significant positive relationship between sperm density and seminal plasma zinc concentration in the fertile, but not in the infertile, men. The infertile men with antisperm antibodies or counts > 20 millionlml had significantly higher mean levels of seminal plasma zinc than infertile men with oligospermia. The higher semen zinc in these two groups may reflect an abnormal fragility of the spermatozoa, resulting in the release of zinc, but the absence of significant overall differences between fertile and infertile men suggests that measurement of the concentration of zinc in plasma or zinc and copper in seminal plasma has little value in the routine investigation of infertility. Fertil Steril 40:670, 1983

Received December 13, 1982; revised and accepted June 29, 1983. *Medical Research Council Epidemiology and Medical Care Unit, Northwick Park Hospital. tReprint requests: Dr. Andrew P. Haines, Medical Research Council Epidemiology and Medical Care Unit, Northwick Park Hospital, Harrow, Middlesex HAl 3UJ, England. :j:Division of Clinical Chemistry, Clinical Research Centre, Northwick Park Hospital. §Chelsea Hospital for Women.

670

Stanwell-Smith et aI. Trace elements and infertility

The human prostate gland contains a higher concentration of zinc than any other organ in the body, which suggests that this element has an important role in male reproduction. There is evidence that zinc is required for normal testicular development, 1 spermatogenesis, 2 and sperm motility,3,4 although the precise mechanisms for these actions have not yet been defined. Competitive interaction of zinc with copper5 and of both zinc and copper with the nonessential Fertility and Sterility

elements cadmium and lead has been described. 6 Both cadmium and lead have been implicated as particularly harmful to fertility.7 In previous studies on infertile men, the possible interrelationships between zinc, copper, cadmium, and lead have not been investigated. In much of the previous work on zinc and other trace metals in infertile populations, differences in sperm count (or sperm density in millions per milliliter) have been used to separate and compare the groups of men studied. Such comparisons are based on the assumption that semen quality is a reliable guide to potential fertility. However, it has been observed that seminal analysis can seldom be relied upon to indicate definite infertility in the male, and never to indicate definite fertility.s Thus, comparing groups defined by semen quality alone is of limited value. Although rapid progress is being made in the recognition of disorders of testicular function, the only sure measure of fertility remains the achievement of pregnancy. In the present s.tudy, the differences in concentrations of zinc, copper, cadmium, and lead were investigated in groups with observed differences in fertility, irrespective of apparent semen quality, in addition to analyses based on sperm density and clinical differences between the infertile men. Hematologic indices and the concentration of substances known to affect trace element levels, such as serum proteins, were -also measured. MATERIALS AND METHODS

had no previous history of infertility. Eleven men were on drug treatment for infertility (mesterolone, tamoxifen, or clomiphene) at the time of sampling. The results were analyzed both with and without the inclusion of this group. Classification of sperm density for the infertile men was based on two or three semen specimens, including the sample analyzed for copper and zinc. Oligospermia was defined as a sperm density of> 0 but not> 20 million/ml on all samples. The fertile control subjects were recruited from the partners of a random sample of women attending Queen Charlotte's Maternity Hospital for antenatal care. Selection was restricted to men who had no history of previous infertility in themselves or in their partners. One hundred fertile men were selected, but only 40 of these men agreed to take part in the study. Two men had to be excluded from the analysis: one was found to have undergone investigation for infertility in the past, and the other attended on the wrong day. The results for the remaining 38 are considered in this report. They had achieved their pregnancies within an average of 8 months. The low response rate in control subjects (40%) was partly due to a high rate of refusal by men who lived some distance from the hospital and those with unskilled occupations. Response also appeared to be affected by parity (less refusals from couples during their first pregnancy [54%] than those with two or more children [76%]) and by religious denomination (high refusal range [93%] among Roman Catholics).

SELECTION CRITERIA FOR SUBJECTS AND CONTROLS

PROCEDURES

Caucasian men in the age range of 20 to 45, with no apparent chronic or acute disease, were selected for the study. The 80 infertile subjects were recruited consecutively from those referred to the Male Fertility Clinic at the Chelsea Hospital for Women (CHW). The duration of infertility, defined as failure to achieve pregnancy for at least 1 year, ranged from 1 to 14 years (mean, 3.7). All the female partners underwent standard investigations, including hormonal evaluation and laparoscopy, and men were included in the study only if the female partner had no diagnosed cause of infertility or a minor disorder of ovulation. Men whose wives conceived within 8 months after study entry were excluded because the aim of the study was to compare men who had been infertile for at least 1 year with fertile men who

One sample of venous blood was obtained from each of the 80 infertile and 38 fertile men over a 14-month period (May 1980 to June 1981). All the blood samples were taken between 8:30 A.M. and 1:30 P.M. The exact time of taking the blood sample was recorded. Each subject and control was asked to bring a semen specimen to the clinic, collected by masturbation not more than 4 hours before attending the clinic and after at least 48 hours' abstinence. Each man was interviewed in a standard manner at the research clinic. The following information was collected during the interview: occupation, from which social class was determined (based on the six categories of the Registrar General's classification9 ); exposure at work to heated metals, high temperatures, fumes, or dust; height, weight, smoking and alcohol con-

Vol. 40, No.5, November 1983

Stanwell-Smith et aI. Trace elements and infertility

671

Table 1. Group Classification of Men Included in the Study Description of group

Group

Fertile control Fertile men (wives booked for antenatal care; no history of subjects infertility in either partner) Infertile men "Oligospermic" (men with sperm A densities of up to 20 million/ ml, excluding men with azoospermia, antisperm antibodies, or a varicocele) Men with antisperm antibodies B (titer> 1:32) (± other factors such as azoospermia) Men whose sperm densities exC ceeded 20 million/ml but who did not have antisperm antibodies or a varicocele Men with azoospermia (but no D antisperm antibodies) (n = 4), and men later treated for a varicocele (n = 9)

No. of men

38

TRACE METAL DETERMINATIONS

29

14 24

13

80

Total infertile men Total fertile and infertile men

to spermatozoa using three methods: gelatin agglutination,l1 tray agglutination,12 and sperm immobilization. 13

118

All glassware used in the collection and analysis of trace metal concentrations had been previously processed using standard washing and heating techniques to eliminate any contamination. All reagents were of Aristar grade. Blood or semen samples with an insufficient volume for analysis or plasma samples that had stood for more than 4 hours at room temperature were di~­ carded. The concentrations of zinc and copper in heparinized blood and seminal plasma and of cadmium and lead in whole blood were measured by atomic absorption spectroscopy. The methods for zinc and copper were based on those described by Smith et al. 14 and Ward et al. 15 The concentrations of zinc in the erythrocyte was determined by means of the formula (PCV, packed cell volume): Whole blood zinc

sumption; whether or not they were vegetarian; whether or not they had a current or recent illness; duration of infertility (or time taken to conceive in the control cases); drug therapy; and previous pregnancies in their partners. SEMINAL ANALYSIS AND ANTISPERM ANTIBODY DETERMINATIONS

Semen screening was carried out as soon as possible after the specimen arrived at the CHW seminology laboratory and after liquefaction had occurred. The age of the semen sample on arrival at the laboratory and at the time of centrifugation was recorded. The semen screening included estimation of semen volume, total sperm count, sperm density, percent motility (on arrival and 5 to 10 hours later), pH, clumping, the mixed antiglobulin reaction test for antibody screening, 10 and culture to exclude infection. The semen specimens were then centrifuged at 2000 rpm for 10 minutes, and the seminal plasma was removed. If the mixed antiglobulin reaction test was positive, the antibody titer was measured in the seminal plasma. The remaining seminal plasma was transferred to containers for transport to Northwick Park Hospital and refrigerated at 4° C, zinc and copper analyses were carred out within 1 week. Serum specimens from all subjects and controls were screened for the presence of antibodies 672

Stanwell-Smith et aI. Trace elements and infertility

X

100 - ([100 - PCV]

X

plasma zinc)

PCV

The concentrations of lead and cadmium were determined in whole blood by the method described by Stoeppler and Brandt. 16 The samples were analyzed in batches throughout the study, and samples were assayed from infertile and fertile cases together so that the influence of interassay variation could be reduced. The atomic absorption signals were compared with six freshly prepared matrix-matched standards for the whole blood analyses and with aqueous standards for the plasma and seminal plasma estimations. OTHER INVESTIGATIONS

PCV, mean corpuscular volume, alkaline phosphatase, aspartate transaminase, serum albumin, and globulin were determined by standard laboratory techniques as part of the routine investigations at the fertility clinic. STATISTICAL METHODS

The results of the 80 infertile men were analyzed first as one group (including men with azoospermia or a varicocele [group D]) and second as three separate groups: A, oligospermia, defined as sperm density up to 20 millionlml; B, antisperm antibodies; and C, sperm density> 20 millionlml Fertility and Sterility

n y r y y :

c)

re ees y, rsbly ole for

osbuard in-

anazoOd as ed as erm n/ml

erility

Table 2. Age, Sperm Density, Motility, Semen Volume Infertile men

Fertile control subjects

All cases (A, B, C, D)

Group A

Group B

Group C

32.8 4.5 38

32.5 5.3 80

32.0 5.3 29

32.7 5.6 14

32.7 5.9 24

78.3 b 54.7 38

30.5 b 48.9 72

4.2 5.3 25

27.8 31.7 14

70.9 67.8 22

48 b 12 38

35 b 19 64

27 18 24

30 16 11

43 17 22

3.5 2.0 38

3.2 1.9 72

3.0 1.8 25

3.0 2.2 14

3.1 1.7 22

Age (yr) Mean SDa n Sperm density (million! m!) Mean SD n Sperm motility (%) Mean SD n Volume of semen (m!) Mean SD n

aStandard deviation. bp < 0.001 for differences between fertile and all infertile men.

(Table 1). The results were analyzed both with and without the group of 11 infertile men on drug therapy to investigate whether their inclusion affected the overall results. The results for the men in group D were not analyzed separately, because this was a heterogeneous group with small numbers. The results for fertile control subjects were analyzed as one group in all analyses; we compared them first with the results of all 80 infertile men, then with the results of the 29 oligospermic men in group A, because the men in this group were thought to be most likely to show trace element differences. These comparisons were made with the unpaired t-test (or Mann-Whitney U test for variables, such as sperm count and alcohol consumption, which were cleariy not normally distributed), or the chi-square test for categorical variables (such as smoking habit). Alcohol consumption was analyzed as drinks per week, where drinks were calculated as approximate equivalents of pints of beer, glasses of wine, or measures of spirits. The logarithms of certain variables (alkaline phosphatase, aspartate transaminase, whole blood cadmium, seminal plasma zinc, and seminal plasma copper) were used in the statistical analysis to improve the normality of their distributions. The means presented in the tables of results are from unlogged data. An additional comparison was made between the infertile groups A, B, and C (Table 1) using one-way analysis of variance, or chi-square analysis for categorical variables. Vol. 40, No.5, November 1983

RESULTS The distribution of values was inspected for each variable: there were no greatly outlying values in the distribution of trace element results. Although the full range of tests described was planned for each subject and control, complete results were not obtained in all cases. The number of men for each test is shown in the results (Tables 2 to 4). Sperm density and motility were significantly higher in the fertile than in the infertile men (both P < 0.001) (Table 2). The very low mean sperm density of the infertile men in group A is explained by the inclusion of men with severe oligospermia (sperm density < 1 million/mI). There were no convincing differences in erythrocyte zinc between any of the groups (Table 3). Plasma zinc levels from samples taken later in the morning were lower than those taken earlier, and this trend (average 4 fLmolll over 5 hours) was significant (P < 0.001) (Fig. 1). The fertile men tended to be seen earlier than the infertile patients; and, when adjustment was made for time of sampling, there were no significant differences in plasma zinc between the groups. Plasma zinc was positively related to albumin levels (r = 0.28, P = 0.005). There was no consistent relationship between the concentrations of plasma and seminal plasma zinc. Seminal plasma zinc did not differ convincingly between fertile and infertile men. Among the fertile men, those with Stan well-Smith et aI. Trace elements and infertility

673

Table 3. Blood Levels of Zinc, Copper, and Lead; Semen Levels of Zinc and Copper; Serum Aspartate Transaminase Infertile men

Erythrocyte zinc (mmol/l) Mean SD n Plasma zinc (ILmoll!)a Mean SD n Seminal plasma zinc (mmoll!) Mean SD n Plasma copper (ILmol/l) Mean SD n Seminal plasma copper (ILmol/l) Mean SD n Whole blood lead (lLg/100 m!) Mean SD n Aspartate transaminase

Fertile control subjects

All cases (A. B. C. D)

Group A

Group B

Group C

1.17 0.20 35

1.14 0.19 61

1.15 0.24 20

1.06 0.16 12

1.16 0.18 17

13.5 2.1 38

12.9 1.6 68

13.0 1.6 27

12.9 2.0 14

13.2 1.6 20

1.85 1.25 37

2.42 1.70 49

1.50 b 0.83 18

14.5C 2.9 38

16.1c 2.8 78

15.6 2.8 29

1.65 1.21 33

1.64 1.21 46

1.89 1.89 17

1.50 0.58 8

1.56 0.83 16

0.80 0.24 24

0.79 0.24 52

0.79 0.22 20

0.76 0.16 11

0.78 0.24 13

24.7 d 8.4 26

19.9d 4.7 45

20.8 5.3 13

22.7 4.1 11

16.8 2.9 12

2.90 b 2.07 9

17.4 2.8 14

2.92b 2.05 16

16.3 3.0 22

(IUIl)

Mean SD n

aValues adjusted to 10:30 A.M. to allow for the trend with sampling time. bp = 0.01 for differences between the three groups A, B, and C. cp = 0.007 for differences between fertile and all infertile men. dp = 0.01 for differences between fertile and all infertile men.

antisperm antibodies, or with sperm density> 20 million/ml, had significantly higher seminal plasma zinc than those with oligospermia (P = 0.01). There was a clear positive relationship between sperm density and seminal plasma zinc within the fertile control subjects (r = 0.49, P = 0.002), but this relationship was not so consistent within the infertile men (r = 0.10). Plasma copper was significantly higher in the infertile men, compared with the control subjects (P = 0.007) (Table 3) and concentrations did not vary with sampling time. There were no consistent differences in the concentrations of seminal plasma copper between groups and no significant relationship between the levels of copper in seminal plasma and plasma. There were no significant differences in the concentrations of whole blood lead or cadmium between the groups (Table 3). Smokers had high674

Stan well-Smith et aI.

Trace elements and infertility

er mean levels of blood cadmium than nonsmokers (0.22 nmolll as against 0.08 nmol/I) (Table 4), and there was a significant positive relationship between blood cadmium concentration and the reported number of cigarettes smoked per day (P = 0.005) (Fig. 2). No significant associations were found between the concentrations of zinc, copper, cadmium, and lead in blood or between zinc and copper in seminal plasma. Nearly 50% of the men were in classes I and II of the Registrar General's classification. The social class distribution of the fertile control subjects was significantly different from that of the infertile subjects (P = 0.006), with a higher proportion of the fertile men in classes I and II; but there was no relationship between social class and trace elements. Reported alcohol consumption was somewhat lower in the infertile men than in the fertile conFertility and Sterility

, p e

P

e r, d

I oe out s

at n-

ty

Table 4. Whole Blood Cadmium Concentrations in Smokers and Nonsmokers (in Both Fertile and Infertile Men) Infertile men Fertile control

subjects

Whole blood cadmium (nmolll)a Smokers 0.20 Mean SD 0.12 n 6 Nonsmokers 0.07 Mean SD 0.05 n 14

All cases (A,B,C,D)

Group A

GroupB

Group C

Fertile and infertile men

0.22 0.11 23

0.26 0.07 9

0.20 0.15 6

0.22 0.11 7

0.11 29

O.OS 0.09 29

O.OS 0.09 9

0.16 0.17 5

0.06 0.04 S

O.OS 43

0.22b

O.OSb

aValues in nanomoles per liter. bp < 0.001 for differences between smokers and nonsmokers.

trol subjects (median intakes 7 and 11 drinks respectively, P = 0.03). A greater proportion of the infertile men smoked (45% as against 32%), and the infertile men tended to smoke more cigarettes,although neither of these differences was statistically significant. Aspartate transaminase levels were significantly lower in the infertile men than in the control subjects (P = 0.01) and were particularly low in group C (those with sperm densities> 20 million/mI). The differences in semen zinc and plasma copper between the groups were not related :to differences in reported alcohol consumption or aspartate transaminase. Eleven of the infertile men were taking drug therapy for infertility at the time of sampling (mostly mesterolone or tamoxifen), and they were found to have a significantly lower mean erythrocyte zinc concentration (mean, 1.00 mmolll) than that of other infertile men (mean, 1.17 mmolll). There were no other effects of hormone therapy on trace metal levels, and reanalysis of the results after exclusion of the small number of men on drug therapy did not materially affect the differences between the mean concentrations of plasma copper or seminal plasma zinc. No significant differences were observed between fertile and infertile men with respect to age, age of wife, height, weight, type of diet (only one man was a vegetarian), semen volume, age of s~men sample at analysis, serum proteins, alkalme phosphatase, hemoglobin, mean corpuscular volume, or PCV. Analysis of work exposure to metals, fumes, and dust from information given on the questionnaire neither revealed convincing differences between the fertile and infertile groups nor showed any effect of reported exposure OIl metal leyels. Vol. 40, No.5, November 19S3

DISCUSSION

This study showed no evidence of significantly lower plasma zinc levels in infertile men compared with those of proven fertility, in cont~ast to previous reportsP' 18 Although the study was not designed to investigate diurnal variation of trace element concentrations, particular care was taken to exclude temporal effects as a confounding variable. When sampling time was taken into account, apparent differences in plasma zinc levels between fertile and infertile men disappeared. None of the men had clinical evidence of zinc deficiency or plasma zinc concentrations below 20

o

18

o

fitted regression line: y • 21.3 - O.77x ; p< 0.001

•

o '0

16

E

Eu

c 'N

14

'"

E

'" 12

'" 0:::

10 8

0

• • • • ~

I

I

I

I

8:00

9:00

10:00

11:00

.

I 12:00

I 13:00

I 14:00

Time of sampling Figure 1 Var~ation o~plas~a f~rble and ~nfert~le

zinc concentration with sampling time in men. Fertile men are indicated by open clrcies and mfertile men by closed circles.

Stanwell-Smith et a1. Trace elements and infertility

675

0.50

r = 0.50 P = 0.005

•

0.40 :::: o E c:: E

::::I

0.30

•

E ~ u

g :c

•

0.20

I

I•

• •

I

•

•

• • •

Q)

'0

•

..c:::

:s:

I

0.10

• • • •

o~----~------.-----,------,

o

10

20

30

40

Reported number of cigarettes per day Figure 2 The relationship between whole blood cadmium concentration and the reported number of cigarettes smoked per day in male fertile and infertile smokers.

the normal laboratory range, and the results suggest that measurement of plasma zinc would not be a useful addition to standard infertility investigations. Previous reports of differences in semen zinc levels between groups of infertile men are inconsistent, and in some cases it has not been clearly stated whether seminal plasma or whole semen was used. A positive relationship between sperm count and seminal plasma zinc concentration has been reported,4, 19,20 although in contrast to .. these previous reports, the relationship observed in the present study was only significant in the group of fertile control subjects. Other surveys of seminal plasma or whole semen zinc in infertile men have failed to find any relationship between zinc levels and sperm density or motility.2I, 22 The higher mean concentration of seminal plasma zinc in men with antisperm antibodies or sperm counts > 20 million/ml could be related to a higher proportion of abnormally fragile spermatozoa in these men. Sperm have a high zinc concentration,3 and leakage of zinc into seminal plasma may have occurred either spontaneously or during specimen processing. An association has been described previously between oligospermia, 676

Stanwell·Smith et aI. Trace elements and infertility

particularly in association with prostatitis and low concentrations of seminal plasma zinc: both sperm count and motility increased when oligospermic men with relatively low seminal zinc concentrations were given zinc supplementation. 19, 20 However, although the men with oligospermia (group A) in the present study had significantly lower concentrations of seminal plasma zinc than other infertile men, none showed clinicalor bacterial evidence of prostatitis. Previously described relationships between semen zinc concentration and sperm motility3, 4 have been based on comparisons between semen samples with differing sperm density. There was no evidence in this study of a relationship between zinc concentration in plasma or seminal plasma and sperm motility . The difference in plasma copper between fertile and infertile groups is statistically significant but of small magnitude (- 1.5-,...,mol mean difference). An inverse relationship between copper and zinc has previously been ,suggested, 5 but in this study there was no evidence of such a relationship and no difference in the concentrations of seminal plasma copper between the groups. Copper has been identified as one of the elements most toxic to sperm,23 and the previously undescribed difference in plasma copper reported in the present study may be of pathogenic significance. Although its relationship to infertility is unclear, because there were no differences in seminal plasma copper concentrations between fertile and infertile men. Inclusion of 11 men on mesterolone, tamoxifen, or clomiphene did not affect the overall significant differences, and the observation of a lower erythrocyte zinc concentration in this group may reflect an effect of these drugs on the bone marrow, although the therapy was mixed and the small size of the group reduces any conclusions which can be drawn. Higher levels of cadmium in the blood of smokers have been previously reported. 24 The present study confirmed the association between blood cadmium and smoking and also demonstrated a relationship between cadmium levels and amount smoked, although there were no significant differences in cadmium concentrations between fertile and infertile men. A possible association between smoking and infertility has been described. 25 There was, however, no evidence to suggest that the higher concentration of cadmium in the blood of smokers affects their fertility. Fertility and Sterility

Infertile men had a significantly lower mean concentration of aspartate transaminase than that of fertile men, but this is almost certainly related to the lower reported alcohol consumption in the infertile group rather than to any association with infertility. Although the original control group was randomly selected, the men who agreed to participate were, on average, of higher social class than the men who refused. However, there was no evidence that social class was related to differences in plasma copper and semen zinc. The inconvenience involved in taking part in the study was the commonest stated reason for refusal: the men recruited as control subjects had to take time off work and travel some distance to the hospital where the study took place. A study of this size comparing 38 fertile control subjects with 80 infertile men has a 90% chance of detecting, as significant at. the 5% level,. a true difference between the two groups of - 0.65 of a standard deviation (i.e., - 1.2 t-tmol/lfor plasma zinc and - 0.82 t-tmo1l1 for seminal plasma copper). This excludes the· existence of true differences larger than this with reasonable certainty. Because of the difficulties in the interpretation of differences in semen quality and testicular function, future investigations of the relationship of zinc and copper to male infertility should use control subjects of known fertility. This may allow associations that are independent of fertility but dependent on sperm density to be distinguished from associations that may be helpful in the investigation and treatment of infertile men. Acknowledgments. We gratefully acknowledge the help of Dr. Tom Meade in the design of this study and preparation of the report. We also wish to thank the laboratory and clerical staff of Chelsea Hospital for Women, Northwick Park Hospital, and the MRC Epidemiology and Medical Care Unit for their assistance, Dr. Mary Campbell-Brown for her help in the Fertility Clinic, and Professor Geoffrey Chamberlain for his advice and encouragement throughout the study.

REFERENCES 1. Prasad AS: Trace Elements and Iron in Human Metabolism. New York, John Wiley & Sof!S, 1978, p 296 2. Abbasi AA, Prasad AS, Rabbair P, DuMouchelle G: Experimental zinc deficiency in man. J Lab Clin Med 96:544, 1980 3. Stankovic H, Mikac-Devic D: Zinc and copper in human semen. Clin Chim Acta 70:123, 1976

Vol. 40, No.5, November 1983

4. Skandhan KP, Skandhan S, Mehta YB: Semen electrolytes in normal and infertile subjects. II. Zinc. Experientia.34:1476, 1978 5. Bremner I: The toxicity of cadmium, zinc and molybdenum and their. effects on copper metabolism. Proc Nutr Soc 38:235, 1979 6. Petering HG: Some observations on the interaction of zinc, copper and iron metabolism in lead and cadmium toxicity. Environ Health Perspect 25:141, 1978 7. Henkin RI: Trace metals in endocrinology. Med Clin North Am 60:779, 1976 8. Pryor JP: Seminal analysis. Clin Obstet Gynecol 8:571, 1981 9. Office of Population Censuses and Surveys (OPCS): Classification of occupations. London, HMSO, 1970 10. Jager. S, Kremer J, van Slochteren-Draaisma T: A simple me~hod of screening for antisperm antibodies in the human body. Int J Fertil 23:12, 1978 11. Kibrick S, Belding DL, Merrill B: Methods for the detection of antibodies against mammalian spermatozoa. II. A gelatin agglutination test. Fertil Steril 3:430, 1952 12. Friberg J: A simple and sensitive micromethod for demonstration of sperm-agglutinating activity in serum from infertile men and women. Acta Obstet Gynecol Scand (SuppD 36:21,1974 13. Isojima S, Li TS, Ashitaka Y: Immunologic analysis of sperm-immobilizing factor found in sera of women with unexplained sterility. Am J Obstet Gynecol101:677, 1968 14. Smith JC, Butrimovitz GP, Purdy WC, Boeckx RL, Chu R, McIntosh ME, Lee K-D, Lynn JK, Dinovo EC, Prasad AS, Spencer H: Direct measurement of zinc in plasma by atomic absorption spectroscopy. Clin Chern 25:1487,1979 15. Ward RJ, Danpure CJ, Fyfe DA: Determination of gold in plasma and plasma fractions by atomic absorption spectroscopy and by neutron activation analysis. Clin Chim Acta 81:87, 1977 16. Stoeppler M, Brandt K: Contributions to automated trace analysis. V. Determinations of cadmium in whole blood and urine by electrothermal atomic absorption spectrophotometry. Fresenius Zeitschrift fur Analytische Chemie 300:372, 1980 17. Hartoma TR, Nahoul K, Netter A: Zinc, plasma androgens and male sterility. Lancet 2:1125, 1977 18. Hartoma TR: Serum testosterone compared with serum zinc in man. Acta Physiol Scand 101:336, 1977 19. Marmar JL, Katz S, Praiss DE, DeBenedictis TJ: Semen zinc levels in infertile and postvasectomy patients and patients with prostatitis. Fertil Steril 26:1057,1975 20. Caldamone AA, Freytag MK, Cockett ATK: Seminal zinc and male infertility. Urology 13:280, 1979 21. Lindholmer CH, Eliasson R: Zinc and magnesium in human spermatozoa. IntJ Fertil17:153, 1972 22. Wood BJ, Lawrence DM, McGarrigle HHG: Similar zinc levels in seminal fluid from normospermic, oligospermic and azoospermic men. Clin Chim Acta 123:329, 1982 23. Holland MK, White IG: Heavy metals and spermatozoa. I. Inhibition of the motility and metabolism of spermatozoa by metals related to copper. Fertil Steril 34:483, 1980 .24. Ward RJ, Fisher M, Tellez-Yudilevich M: Significance of blood cadmium concentrations in patients with renal disease or essential hypertension and the normal population. Ann Clin Biochem 15:197, 1978 25. Campbell JM, Harrison KL: Smoking and infertility. Med J Aust 1:342, 1979

Stanwell-Smith et al. Trace elements and infertility

677