Immunoglobulin in Seminal Fluid of Fertile, Infertile, Vasectomy and Vasectomy Reversal Patients

0022-5347/83/1294-0869$02.00/0 Vol. 129, April

THE JOURNAL OF UROLOGY

Printed in U.S.A.

Copyright © 1983 by The Williams & Wilkins Co.

IMMUNOGLOBULIN IN SEMINAL FLUID OF FERTILE, INFERTILE, VASECTOMY AND VASECTOMY REVERSAL PATIENTS JACKSON E. FOWLER, JR.*

AND

MICHELE MARIANO

ABSTRACT

We measured the concentrations of IgG, IgA, and IgM, in the seminal fluid of 16 fertile men, 77 men of infertile marriages, 21 men who had undergone vasectomy reversal and 5 men who had undergone vasectomy only. The lower limits of sensitivity of the assay was 0.04 mg./dl. IgG (mean concentration 3.29 mg./dl., range 0.48 to 15.41 mg./dl.) and IgA (mean concentration 1.11 mg./dl., range 0.05 to 19.11 mg./dl.) were measurable in all specimens, but IgM (range 0.04 to 0.76 mg./dl.) was measurable in only 20 per cent. Intrasubject variability of IgG and IgA concentrations expressed as the coefficients of variation of serial determinations ranged from 18 to 40 per cent and 29 to 52 per cent, respectively. Discrepancies between the presence or absence of measurable IgM in serial determinations were unusual. The mean concentrations of seminal fluid IgG and IgA in the fertile group were not significantly different from the other patient groups. However, IgM was measurable in only 13 per cent of specimens from the fertile patients but in 62 per cent of specimens from the vasectomy reversal patients (p = 0.003). This suggests disruption of the blood-genital tract barrier following vasectomy and continuing after vasectomy reversal. Antibody directed against spermatozoa has been detected in the seminal fluid of infertile men and may be a cause of infertility. 1• 2 Since this phenomenon might be associated with increased seminal fluid immunoglobulin, several groups have compared class specific immunoglobulin concentrations in the seminal fluid of infertile and fertile men. Sullivan and Quinlivan3 reported that the concentrations of seminal fluid lgG and lgA, and the frequency of detection of secretory-lgA (S-IgA), were greater among infertile patients with azoospermia, oligospermia, and self-agglutinating sperm than among fertile controls. Kovary4 found that seminal fluid lgA concentrations were greater for patients with spontaneous sperm agglutination than for patients without such agglutination. Azim5 reported that the mean concentrations of seminal fluid lgG and lgA were greater among men with self-agglutinating sperm than among fertile controls with no sperm agglutination. Further, lgM was usually detectable in the seminal fluid of the former patients but usually undetectable in that of the latter. On the other hand, Uehling6 reported that seminal fluid lgG, lgA, lgM and S-IgA concentrations were similar for fertile and infertile men, and Soliman7 found no differences in S-lgA concentrations in the seminal fluid of azoospermic, subfertile or fertile men. Friberg8 was unable to demonstrate differences between the concentrations of seminal fluid lgG, lgA, and lgE in fertile men and in men with self-agglutinating seminal fluid. In addition to inconsistencies concerning the relationships between fertility status or semen analysis and seminal fluid immunoglobulin concentrations, the reported frequencies with which lgA and lgM are measurable in the seminal fluid of fertile and infertile men vary markedly. Uehling6 and Friberg8 quantitated lgA in almost all specimens assayed, whereas Azim 5 and Sullivan3 found measurable lgA in only 54 and 59 per cent of specimens, respectively. U ehling6 uniformily quantitated lgM in seminal fluid but Friberg8 was unable to detect lgM in any specimen. These conflicting findings are likely due to differences in the sensititivities of the immunoglobulin assays. We used an indirect solid phase radioimmunoassay9 (SPRIA) to quantitate lgG, lgA and lgM in the seminal fluid of fertile men, men of infertile marriages, and previously fertile men who had undergone vasectomy or vasectomy and subsequent vasectomy reversal. The lower limits of sensitivity of the assay for Accepted for publication-February 3, 1983. * Requests for reprints: Department of Urology, Box 422, University of Virigina School of Medicine, Charlottesville, Virginia 22908. 869

each immunoglobulin class was 0.04 mg./dl. lgG and lgA were uniformily measurable, but the mean concentrations of these immunoglobulins in each patient group were not significantly different. lgM was measurable in 29 per cent of the samples assayed and the frequency of measurable lgM was significantly greater among specimens from patients who had undergone vasectomy reversal than among specimens from the fertile patients. MATERIALS AND METHODS

Ejaculate collection and analysis. The ejaculate was obtained by masturbation into clean dry glass containers. Patients were asked to abstain from ejaculation for 2 days before specimen collection. Following liquefaction the volume of ejaculate was measured in a graduated cylinder and sperm concentration was determined in a Neubauer counting chamber. Motility was graded on a scale of O to 4 by examination of sperm on a standard microscope slide. For the purposes of this analysis, however, only the percentages of motile sperm were considered. Normal seminal parameters were considered as follows: sperm concentration :'>20 X 106 /ml., total sperm count :'>60 X 106 and per cent motility :'>50 per cent. Oligospermia was defined as <20 X 106 sperm/ml. or <60 X 106 total sperm, and azoospermia as the total absence of identifiable sperm. Decreased motility was defined as <50 per cent motile sperm. Patients. We analyzed the seminal fluid from 16 fertile patients whose wives had conceived within 2 years prior to study, from 77 patients of infertile marriages whose wives had been unable to conceive during 1 to 9 years (mean 3.8 years) prior to study, from 5 patients who had undergone vasectomy 1 to 8 months (mean 3.0 months) prior to study, and from 21 patients who had undergone vasectomy 2 to 15 years (mean 7.2 years) and subsequent vasectomy reversal 2 to 5 months (mean 4.0 months) prior to study. All vasectomy and vasectomy reversal patients had previously fathered at least 1 child but none of the vasectomy reversal patients had proven restoration of fertility. No patients had histories of genitourinary infections. The mean age and mean seminal parameters for the patient groups are shown in table 1. Each of the fertile patients had normal seminal parameters. The infertile patients were subgrouped according to the presence or absence of abnormal seminal parameters. No infertile patient with normal seminal parameters was married to a woman with a documented cause for infertility, and no infertile patient with azoospermia had evidence of ductal obstruction.

870

FOWLER AND MARIANO TABLE

1. Patient characteristics and results of semen analyses* Semen Analyses

Patient Group

No.

Fertile Infertile Normal parameters Oligospermic Decreased motility Oligospermic and decreased motility Azoospermic Vasectomy reversal Vasectomy

Mean Age

16

77 33 20 7 14 3

21 5

Volume (ml.)

Sperm Concentration (X10 6 /ml.)

Total Sperm Count (XlO'')

Motility (%)

29.4 30.9 29.7 30.6 34.1 32.7

:3.19 ± 1.48

70 ± 44

202 ± 105

60 ± 13

1.26 1.29 0.92 1.54

56 ± 27 12 ± 5 57 ± 25 8±7

162 30 127 22

15 12

19

71 67 34 23

31.0 34.5 30.8

2.23 ± 2.00 2.75 ± 1.75 3.08 ± 1.51

0±0 17 ± 13 0±0

0±0 36 ± 20 0±0

0 20 0

0 21 0

2.82 2.90 2.56 3.03

± ± ± ±

± ± ± ±

96 12 26

11

17

-, Values are expressed as the mean ± 1 standard deviation.

TABLE

Patient

No. Determinations

It 2t 3

4 5 6

10

5 5 5 4 4

2. Variability of seminal fluid immunoglobulin in 2 fertile and 4 infertile patients*

Mean lgG (mg./dl.)

CV(%)

Mean Total lgG (mg.)

CV(%)

Mean lgA (mg./dl.)

CV(%)

Mean Total lgA (mg.)

CV(%)

1.59 3.06 1.59 2.21 0.98 2.97

32 27 40 33 32 18

6.01 4.68 6.35 6.02 5.13 15.79

38 43 51 39 22 23

0.24 0.83 0.44 0.42 0.46 4.77

33 38 41 29 43 52

0.89 1.13 1.60 1.15 2.51 24.94

33 54 23 23 45 50

* CV = Coefficient of variation. 1" Fertile patient.

Seminal fluid immunoglobulin determinations. After microscopic semen analysis, the ejaculate was diluted 1:3 with phosphate buffered saline (PBS buffered to 7.2) and centrifuged at 750 X g for 10 minutes to separate spermatozoa and debris from the seminal fluid. The latter was then frozen in multiple aliquots and stored at -20C until assayed. Quantitation of IgG, IgA and IgM was performed with an indirect SPRIA using previously described methods. 9 Purified heavy chain specific goat anti-human lgG, IgA or IgM (GAHlg) (Tago Inc., Burlingame, California) was used as the antigen, seminal fluid in serial 2-fold dilutions was used as the primary antibody, and 125I-labelled GAH lg of the corresponding class was used as the secondary antibody. We performed all assays in duplicate and assessed non-specific binding of immunoglobulin for each sample using bovine serum albumin (BSA) as the antigen. The concentration of class specific immunoglobulin in each sample was determined by reference to the linear portion of a standard SPRIA curve generated on each assay day with immunoglobulin standards of known concentration (Calbiochem-Behring Corp., La Jolla, California). We have previously demonstrated that this assay is specific for each immunoglobulin class and that the GAH-IgA reacts with both lgA and SIgA. The coefficients of variation range from 15.5 to 17.7 per cent for the 3 immunoglobulin assays and the lower limits of sensitivity are 0.01 mg./ dl. However, because the seminal fluid was diluted 1:3 before centrifugation, the lower limits of sensitivity in this study are 0.04 mg./dl. Statistical analyses. Differences between the concentrations of seminal fluid lgG and IgA and between total seminal fluid lgG and IgA for various patient groups or subgroups were identified by calculation of the T-statistic from the means and the standard deviations. The Welch approximation for the Tstatistic and the appropriate degrees of freedom was used when groups had unequal variance at the 0.05 level or better. The Fisher's exact test was used to determine differences in the frequency of detection of seminal fluid IgM in various patient groups or subgroups. P values <0.05 were considered significant. RESULTS

Longitudinal studies of seminal fluid immunoglobulin. Serial ejaculates were obtained over a 4-month period from 6 patients (2 fertile and 4 infertile) to assess the variability of

seminal fluid immunoglobulin for an individual subject. Ten specimens were obtained from 1 patient, 5 specimens were obtained from 3 patients, and 4 specimens were obtained from 2 patients. None of the infertile patients received treatment during the period of study. The coefficients of variation for lgG and IgA concentrations, and total IgG and IgA (lg concentration x seminal fluid volume) are shown in table 2. The coefficients of variation for IgG and lgA concentrations were similar to the coefficients of variation for total IgG and IgA, respectively. Only 1 of these patients (no. 6) had measurable seminal fluid lgM (mean concentration 0.33 ± 0.31 mg./dl.). This was present in each of the 4 specimens assayed. To assess better the intrasubject variability in detectable seminal fluid IgM, 2 specimens obtained at 1- to 12-month intervals from 23 different subjects were assayed. In 14 cases there was no measurable IgM in either specimen, in 7 cases there was measurable IgM in both specimens, and in 2 cases

10

~

c,,

o (15.4)

l

• (12.3)

8

E l!)

Cl

0

6

•

0

Fertile

Norm

4

2

Oligo

DM

Oligo+DM Azo Vas Rev

Vos

f-----lnfertile-----; Patient Groups

FIG. 1. Seminal fluid IgG concentrations in each group or subgroup. Open circles indicate seminal fluid specimens with detectable IgM. Open squares with bars indicate means ± standard deviation.

871

IMMUNOGLOBULIN IN SEMINAL FLUID

there was measurable IgM (0.11 and 0.06 mg./dl.) in 1 specimen and no measurable IgM in the other. Seminal fluid IgG and IgA. The concentrations of seminal fluid IgG and IgA for patients in each group or subgroup are shown in figures 1 and 2, and the mean IgG and IgA concentrations and mean total IgG and IgA levels are shown in table 3. There were in general no significant differences (p >0.05) between these parameters in the fertile group compared to the other groups or subgroups. The mean concentrations of IgG and IgA, and the mean total IgA for the azoospermic infertile patients, however, were significantly greater than those of the fertile patients, p <0.01, p <0.01 and p <0.02, respectively. Seminal fluid IgM. The frequencies with which seminal fluid IgM was detectable in each group or subgroup, and the mean concentrations of detectable IgM ± 1 standard deviation are shown in table 4. The proportions of patients with measurable IgM in the azoospermic infertile subgroup and the vasectomy reversal group were significantly greater than in the fertile group, p = 0.01 and 0.003, respectively. As demonstrated in figures 1 and 2, detection of seminal fluid IgM was often associated with elevated levels of IgG or IgA in the same specimen. In table 5 is shown the mean IgG and IgA concentrations for seminal fluid specimens that were or were not associated with measurable IgM for each group or subgroup. When stratified in this manner there were no significant differences between the mean concentrations of IgG or IgA in the

1

o(5.7)

o(7.4)

0(4.6) (4.3)

3.o'-

0(19.1) • r6.0) (5.4)

0

. . .. . . . .. - • Ht.i •I ;j .•. 0

2.5 ~

~

c,,

.§

2.0 -

1.5 ,-

1.0 ,-

0.5 ,-

)

ii

I

Fertile

I)

..

0

~

<[

~

0

0 0

Norm

0

•

Oligo

DM Oligo+DM Azo

These data indicate that IgG and IgA are uniformly present in human seminal fluid, occurring in concentrations of approximately 3.0 and 1.0 mg./dl., respectively. IgM, if present, is usually in concentrations of <0.04 mg./dl. The validity of these findings are supported by several considerations. Approximately 15 to 30 per cent of the total ejaculate volume appears to be derived from prostatic secretions. 10• 11 In addition, since the concentrations of IgG and IgA are greater in the initial than in the terminal portions of the ejaculate, it is thought that seminal fluid immunoglobulin is derived largely from prostate secretions. 12' 13 We recently measured the concentrations of immunoglobulin in the prostatic fluid of 16 adult men (mean age 38.8 years) who had no evidence of prostatic disorders or clinically detectable prostatic hyperplasia (Fowler, J.E., Jr. and Mariano, M., unpublished data). The mean concentrations± 1 standard deviation of IgG and IgA were 9.2 ± 8.5, and 3.9 ± 4.2 mg./dl., respectively. There was sufficient fluid to measure IgM in 10 cases. In 2 the concentration was <0.08 mg./dl., and the mean concentration ± 1 standard deviation in the remaining 8 was 0.28 ± 0.32 mg./ dl. Wishnow and associates, 14 using similar methodology, found the mean concentrations of prostatic fluid IgG and IgA in a comparable population to be 15.0 ± 4.0 and 5.3 ± 3.4 mg./dl., respectively. The seminal fluid IgG and IgA concentrations reported herein, therefore, are consistent with the anticipated relationships between prostatic fluid and semTABLE

0

4. Frequency of measurable IgM* in seminal fluid

Patient Group

i

)

0 0

§

.

0 0

0

'

Vos

•

Vos Rev

Infertile Patient Groups FIG. 2. Seminal fluid IgA concentrations in each group or subgroup. Open circles indicate seminal fluid specimens with detectable IgM. Open squares with bars indicate means ± standard deviation. TABLE

No.

IgG (mg./dl.)

Fertile Infertile Normal parameters Oligospermic Decreased motility Oligospermic and decreased motility Azoospermic Vasectomy reversal Vasectomy

16 77 33 20 7 14

2.74 3.72 4.01 3.20 2:94 3.59

3 21 5

6.14 2.23 2.94

± ± ± ± ± ±

No. With Measurable IgM

Fertile Infertile Normal parameters Oligospermic Decreased motility Oligospermic and decreased motility Azoospermic Vasectomy reversal Vasectomy

2/16 16/77 7/33 4/20 0/7 2/14

(13%) (21%) (21%) (20%) (0%) (14%)

3/3 (100%):j: 13/21 (62%)§ (60%) 3/5

IgM (mg./dl.)t 0.09 ± 0.01 0.16 ± 0.10 0.29 ± 0.33 0.18 ± 0.08 0.11 ± 0.05 0.12 ± 0.06 0.06 ± 0.02

* Measurable IgM > 0.04 mg./dl. t Values are expressed as the mean of detectable levels ± 1 standard deviation. :j: Significantly greater than fertile group, p = 0.01. § Significantly greater than fertile group, p = 0.003.

3. Seminal fluid IgG and IgA *

Patient Group

* Values are expressed as the mean ± 1 standard deviation. t Significantly greater than fertile group, p <0.01. :j: Significantly greater than fertile group, p <0.01. § Significantly greater than fertile group, p <0.02.

DISCUSSION

0

0 0

0 0

fertile group compared to the other groups or subgroups. In general, the mean concentrations of IgG and IgA associated with measurable IgM within each group or subgroup were greater than those not associated with measurable IgM. These differences were statistically significant for the infertile group, where the mean concentrations of IgG and IgA associated with measurable IgM were significantly greater than those not associated with measurable IgM, p <0.001 and <0.01, respectively, and for the oligospermic and decreased motility infertile subgroup, where the mean concentration oflgA associated with measurable IgM was significantly greater than that not associated with measurable IgM, p <0.01.

Total IgG (mg.)

IgA (mg./dl.)

Total IgA (mg.)

1.79 2.57 3.03 2.19 1.29 2.38

8.34 9.66 10.25 8.68 7.68 9.50

6.36 6.46 6.75 5.80 5.27 5.28

0.49 1.05 1.06 1.16 0.65 0.83

0.22 1.22 1.07 1.66 0.47 0.74

1.68 2.81 2.75 3.66 1.50 1.87

1.16 4.50 2.78 7.81 0.88 1.01

0.95t 1.88 2.65

14.97 6.95 6.60

14.17 4.77 4.75

2.24 1.84 0.86

2.08:j: 4.11 0.83

5.14 5.15 1.86

5.26§ 8.42 1.59

872

FOWLER AND MARIANO TABLE

5. Relationships between seminal fluid IgG and IgA and measurable IgM* Measurable IgM

No Measurable IgM

Patient Group No. Fertile Infertile Normal parameters Oligospermic Decreased motility Oligospermic and decreased motility Azoospermic Vasectomy reversal Vasectomy

IgG (mg./dl.)

IgA (mg./dl.)

No.

3.36 5.80 6.66 4.44

0.72 ± 0.28 2.42 ± 1.96:j: 1.97 ± 1.81 3.49 ±: 2.75

2

5.01 ± 2.55

2.10 ± 0.57§

14 61 26 16 7 12

3

6.14 ± 0.95 2.21 ± 2.85 4.17±1.14

2.24 ± 2.08 2.14 ± 5.11 1.24 ± 0.89

0 8 2

2

16 7 4 0

13 3

± ± ± ±

2.60 3.15t 4.26 2.41

IgG (mg./dl.) 2.64 ± 3.16 ± 3.30 ± 2.90 ± 2.94 ± 3.35 ±

1.76 2.10 2.22 2.09 1.29 2.38

2.26 ± 2.81 1.11 ± 0.80

IgA (mg./dl.) 0.56 0.70 0.82 0.58 0.65 0.61

± ± ± ± ± ±

0.20 0.54 0.62 0.44 0.47 0.52

1.35 ± 1.75 0.29 ± 0.34

* Values are expressed as the mean ± 1 standard deviation. t Significantly greater than corresponding mean IgA concentration in same group, p <0.01 :j: Significantly greater than corresponding mean IgA concentration in same group, p <0.001 § Significantly greater than corresponding mean IgA concentration in same subgroup, p < 0.01

inal fluid immunoglobulin concentrations. As would be expected, seminal fluid IgM was usually <0.04 mg./dl. However, these observations suggest that IgM is normally present in seminal fluid, albeit in concentrations unmeasurable with our assay. The intrasubject variability of seminal fluid lgG and IgA concentrations, while larger than the assay variability, seem acceptable for meaningful interpretations of the data. Discrepancies between the presence or absence of measurable IgM in serial specimens from the same individual were unusual. Uehling6 has reported minimal intrasubject variability in seminal fluid immunoglobulin concentrations, and the intrasubject variability of prostatic fluid immunoglobulin concentrations appear to be small. 9 ' 15 We were generally unable to demonstrate significant differences between the seminal fluid immunoglobulin of fertile men and men of infertile marriages or men who had undergone vasectomy or vasectomy reversal. Although the mean concentrations of lgG and IgA and the frequency of measurable lgM were significantly greater among the azoospermic infertile patients than among the fertile patients, only 3 of the former were studied. The frequency of measurable lgM was significantly greater among men who had undergone vasectomy reversal than among fertile men. This is of interest because 1) a relatively large number of vasectomy reversal patients were studied, 2) a similar frequency of detectable IgM was observed in a limited number of vasectomy patients, 3) measurable lgM in the vasectomy reversal group was not associated with significantly elevated mean concentrations of lgG and IgA, and 4) vasectomy often results in a specific antibody response to sperm antigen that may be detected in seminal fluid. 1 Disruption of the bloodgenital tract barrier following vasectomy and continuing after vasectomy reversal is a likely cause of the increased seminal fluid IgM. It is reasonable to believe that this change occurs in the epididymis and proximal vas deferens. However, fluid from the vas deferens constitutes only a fraction of the seminal fluid volume, and seminal fluid IgM is derived primarily from prostatic secretions. For these reasons, large increases in the concentration of IgM in fluid from the vas deferens would be necessary to alter the concentration of seminal fluid IgM. Our observations suggest that this may occur following vasectomy. Sperm agglutinating antibody has been detected in the serum of as many as 70 per cent of men who have undergone vasectomy.16' 17 These antibodies are primarily of the lgG and IgM class. 18 Sperm agglutinating antibodies have also been detected in the seminal fluid following vasectomy, and may be a cause of persistent infertility following technically successful vasectomy reversal. 19 Although the class or classes of agglutinating antibody in post-vasectomy seminal fluid is not well established, these antibodies appear to be transudates from serum. 1 We do not know as yet whether IgM in the seminal fluid of vasectomy or vasectomy reversal patients binds to spermatozoa. However, our observations raise the possibility that seminal fluid lgM

may contribute to antibody mediated infertility following vasectomy reversal. REFERENCES

1. Rumke, P.: Autoantibodies against spermatozoa. In: Spermatozoa, Antibodies and Infertiity. Edited by J. Cohen and W. F. Hendry. Oxford: Blackwell Scientific Publications, p. 67, 1978. 2. Haas, G. G., Jr., Weiss-Wik, R. and Wolf, D. P.: Identification of antisperm antibodies on sperm of infertile men. Fertil. Steril., 38: 54, 1982. 3. Sullivan, H. and Quinlivan, W. L. G.: Immunoglobulins in the semen of men with azoospermia, oligospermia, or self-agglutination of spermatozoa. Fertil. Steril., 34: 465, 1980. 4. Kovar , P. M., DyKgers, A. and Niermann, H.: Seminal proteins in p.:,L~nts with agglomerations of spermatozoa. Int. J. Fertil., 22: 251, 1977. 5. Azim, A. A., Fayad, S., Fattah, A. A. and Habieb, M.: Immunologic studies of male infertility. Fertil. Steril., 30: 426, 1978. 6. Uehling, D.: Secretory IgA in seminal fluid. Fertil. Steril., 22: 769, 1971. 7. Soliman, H. A. and Olesen, H.: Concentration of the secretory IgA of seminal fluid in normal subjects in decreased fertility and in aspermia. Clin. Chim. Acta., 69: 543, 1976. 8. Friberg, J.: Immunoglobulin concentration in serum and seminal fluid from men with and without sperm-agglutinating antibodies. Am. J. Obstet. Gynecol., 136: 671, 1980. 9. Fowler, J. E., Jr., Kaiser, D. L. and Mariano, M.: Immunologic response of the prostate to bacteriuria and bacterial prostatitis. I. Immunoglobulin concentrations in prostatic fluid. J. Urol., 128: 158, 1982. 10. Lundquist, F.: Aspects of the biochemistry of human semen. Acta. Physiol. Scand., suppl., 19: 66, 1949. 11. Zaneveld, L. J. D.: The human ejaculate and its potential for fertility control. In: Control of Male Fertility. Edited by H. Fritz. New York; Harper & Row, 1975. 12. Rumke, P.: The origin of immunoglobulins in semen. Clin. Exp. Immunol., 17: 287, 1974. 13. Tauber, P. F., Zaneveld, L. J. D., Propping, D. and Schumacher, G. F. B.: Components of human split ejaculates. I. Spermatozoa, fructose, immunoglobulins, albumin, lactoferrin, transferrin, and other plasma proteins. J. Reprod. Fert., 43: 249, 1975. 14. Wishnow, K. I., Wehner, N. and Stamey, T. A.: The diagnostic value of the immunologic response in bacterial and non-bacterial prostatitis. J. Urol., 127: 689, 1982. 15. Shortliffe, L. M. D., Wehner, N. and Stamey, T. A.: Use of a solidphase radioimmunoassay and formalin-fixed whole bacterial antigen in the detection of antigen-specific immunoglobulin in prostatic fluid. J. Clin. Invest., 67: 790, 1981. 16. Hellema, H. W. J. and Rumke, P.: Sperm antibodies as consequence of vasectomy. I. Within 1 year post-operation. Clin. Exp. Immunol., 31: 18, 1978. 17. Ansbacher, R., Hodge, P., Williams, A. and Mumford, D. M.: Vas ligation: humoral sperm antibodies. Int. J. Fertil., 211: 258, 1976. 18. Rose, N. R., Hjort, T., Rumke, P., Harper, M. J. Kand Uyazor, 0.: Techniques for detection of iso- and auto-antibodies to human spermatozoa. Clin. Exp. Immunol., 23: 175, 1976. 19. Linnet, L., Hjort, T. and Fogh-Anderson, P.: Associaton between failure to impregnate after vasovasectomy and sperm agglutinins in semen. Lancet, 1: 117, 1981.