Seminal fluid antisperm antibodies measured by direct flow cytometry do not correlate with those measured by indirect flow cytometry, the indirect immunobead test, and the indirect mixed antiglobulin reaction

p

FERTILITY AND STERILITY Copyright

~

1996 American Society for Reproductive Medicine

Vol. 65, No.1, January 1996

Printed on acid·free paper in U. S. A

Seminal fluid antisperm antibodies measured by direct flow cytometry do not correlate with those measured by indirect flow cytometry, the indirect immunobead test, and the indirect mixed antiglobulin reaction

Marita Rasanen, M.D.*t Yash P. Agrawal, M.D.:j:§ Seppo Saarikoski, M.D. * Kuopio University Hospital, Kuopio, Finland

Objective: To compare the direct flow cytometric (FCM) measurement of sperm-bound antibodies with the indirect FCM measurement of unbound antisperm antibodies in seminal plasma of the same ejaculates. To compare the FCM measurements with the indirect mixed antiglobulin reaction (MAR) and the indirect immunobead test (IBT) performed on the same seminal plasmas. Setting: University hospital-based infertility service. Patients: Eleven infertile men with a positive immunoglobulin (Ig) G direct MAR test and antibody positivity on the direct FCM measurements. Four control patients with a negative IgG-direct MAR test. Main Outcome Measures: The proportion of spermatozoa positive for IgG and IgA antibodies as measured by various tests. Results: The direct and indirect FCM measurements for sperm-bound antibodies from seminal plasma did not correlate with each other. In general, IgG antibodies were not detectable by indirect FCM whereas IgA were detected in lower proportions than by direct FCM measurements. Weak correlation was observed between the indirect FCM, indirect MAR, and indirect IBT measurements. Conclusion: Nearly all ofthe IgG and some of the IgA antisperm antibodies in seminal fluid bind to spermatozoa. Thus, indirect tests to measure antisperm antibodies from seminal plasma are likely to miss the presence of antis perm IgG antibodies while detecting some cases of IgA antibodies. Fertil Steril1996;65:170-5 Key Words: Flow cytometry, mixed antiglobulin reaction, immunobead test, antisperm antibodies, seminal plasma

Antisperm antibodies on the sperm surface and in the genital secretions are implicated in the pathogenesis of infertility, whereas the clinical significance of serum antis perm antibodies is controversial (1, 2). The evaluation of the role of antisperm antiReceived January 5, 1995; revised and accepted August 17, 1995. * Department of Obstetrics and Gynecology. t Reprint requests and present address: Marita Riisiinen, M.D., Department of Dermatology, Bluemle Life Science Building, Thomas Jefferson University, 233 South 10th Street, Philadelphia, Pennsylvania 19107 (FAX: 215-923-9354). Department of Clinical Chemistry. § Present address: Stem Cell Laboratory, University of California San Diego Cancer Center, La Jolla, California.

bodies in infertility is complicated by the lack of standardization of antibody assays and the concern regarding the specificity, sensitivity, and objectivity of many methods (1, 3, 4). The commonly used methods for detection of antisperm antibodies are the mixed antiglobulin reaction (MAR) (5, 6) and the immunobead binding test (IBT) (7). The direct MAR and the direct IBT measure antibodies that already are bound to the patient's spermatozoa. A limitation of the direct tests is the requirement of fresh semen samples having good sperm motility. This requirement may not be met always in pathological samples or when samples are transported to the laboratory over long distances. To overcome these limitations, indirect ver-

Rasmen et al. Seminal plasma antisperm antibodies

Fertility and Sterility

*

170

sions of the MAR and the IBT have been used for the measurement of unbound antisperm antibody from the patient's serum or seminal plasma using spermatozoa from normal donors (8-10). Using indirect IBT measurements on seminal plasma, it has been reported that antis perm antibodies are more prevalent in men with low sperm motility (9) and the presence of immunoglobulin (lg) G or both IgG and IgA antibodies is associated with a poor post coital test and a reduced fertilization rate of human oocytes (10) Although seminal plasma commonly is used to diagnose autoimmunity to sperm, seminal plasma antisperm antibodies may not be reflective of the Ig bound to the sperm surface and may not be related to the patient's fertility (11). The correlation between sperm-bound and unbound antibodies has given conflicting results and also a significant disagreement between the indirect MAR and the indirect IBT measurements from seminal plasma has been reported (12). The tray agglutination test (TAT) (13) commonly is used for the detection of unbound antibodies. The TAT, however, does not detect antibody class, and agglutination may be nonspecific, as in some cases seminal plasma TAT does not correlate with fertility (14). Flow cytometry (FCM) is an objective and reliable method for the measurement of sperm-bound antibodies (15-17) for diagnostic and therapeutic applications. The direct FCM method correlates well with the direct MAR test (18) on fresh semen samples and also can be used on semen samples with poor motility. Flow cytometry also has been applied for the indirect measurements of serum antisperm antibodies (16). In this study, we wanted to test whether direct FCM would correlate with the indirect FCM measurement of seminal plasma antisperm antibodies and with the commonly used indirect IBT and the indirect MAR tests done from the same sample. The use of indirect FCM would obviate the need of same day flow cytometric analysis. It also would allow the seminal plasma samples to be collected and mailed from smaller hospitals to a specialized laboratory, where they could be frozen and objectively analyzed when convenient. In our study, the direct FCM measurements from 11 strongly antibody-positive semen samples correlated only weakly with the corresponding indirect FCM, indirect IBT, and indirect MAR measurements from seminal plasma.

MAR test (SpermMar test; Ortho Diagnostic Systems, Beerse, Belgium) and in whom the direct FCM analysis of spermatozoa revealed strong antibody positivity for both IgG and IgA. Four patients with a negative screening IgG-direct MAR test served as controls. The seminal plasmas from these ejaculates were frozen and stored at -20°C. The unbound antisperm antibodies in seminal plasma were determined using three indirect tests: indirect MAR, indirect IBT, and indirect FCM. Fresh donor semen for indirect tests was obtained from patients who attended our semen laboratory for infertility investigations. All donor semen samples had good motility, forward progression> 50%, and no sperm-bound antibodies as measured by the direct MAR test for IgG and IgA. Direct Flow Cytometry

Semen samples were obtained from 11 infertile men, with a strongly positive screening IgG-direct

The washing of spermatozoa, the antibody labeling procedures, the calibration of the flow cytometer, and the estimation of the proportion of spermatozoa positive for antibodies have been described in detail previously (13). Briefly, a fresh semen sample was washed on Ham's F-10 medium (GIBCO, Paisley, United Kingdom) and the sperm pellet was resuspended in 1 mL of Ham's F-10. Fluorescein isothiocyanate (FITC)-conjugated F(ab')2 fragments of polyclonal rabbit anti-human IgG and IgA antibodies (Dakopatts AlS, Glostrup, Denmark) were added and FITC-conjugated F(ab'h rabbit Ig served as the negative control. Dead spermatozoa were excluded from analysis by propidium iodide staining. Spermatozoa were analyzed on F ACS-Star flow cytometry (Becton-Dickinson, Rochelle Park, NJ) by using a 488-nm line from argon ion laser at 150 m W. Data from at least 10,000 cells were collected using forward scatter as the trigger signal and a threshold of 48 of 1,024 channels. Using the combination of forward and side scatter, the sperm population was gated, while excluding debris and other contaminating cells, and its fluorescence was determined. The FITC signal was detected using standard (DF 530/ 30) filters (Becton-Dickinson) and the propidium iodide fluorescence was detected using standard phycoerytherin filters (DF 584/42). Color compensation was done by labeling sperm cells separately and together with propidium iodide- and FITC-conjugated antibodies. The fluorescence data were collected using the logarithmic amplifier, where fluorescence one height (F11-H) represents the FITC fluorescence measured by detector one and, similarly, F12-H represents the propidium iodide fluorescence measured by the second detector. The control quadrant was marked by using the negative control to include < 1% of cells as positive in other quadrants except lower

Vol. 65, No.1, January 1996

Riisiinen et al. Seminal plasma antisperm antibodies

MATERIALS AND METHODS Patients

171

left quadrant (i.e., viable antibody-negative cell). The control marking was set with all the cells displayed on the screen. Ofthe four quadrants, the antibody-negative dead or membrane-disrupted sperm appear in the upper left quadrant and the antibodynegative viable sperm appear in the lower left quadrant. Sperm positive for both fluorochromes (i.e., dead with nonspecific antibody binding) are displayed in the upper right quadrant and viable antibody-positive sperm are displayed in the lower right quadrant. Only the viable sperm fraction was used for antisperm antibody detection and all data reported are from antibody-positive live spermatozoa. Immunoglobulin G or IgA were considered to be present on spermatozoa if ~ 10% oflive spermatozoa stained positively. Indirect MAR and Indirect IBT

Donor spermatozoa were washed with Ham's F10 medium (37°C) and centrifuged at 300 X g for 10 minutes. The supernatant was aspirated and the sperm pellet was overlaid with 1 mL Ham's F-10 medium. The sperm were incubated at 37°C for 60 minutes. The highly motile sperm from the top layer were used for analysis. The mean concentration of donor sperm after incubation was 13.5 X 10 6/mL. Seminal plasmas were incubated at 56°C for 30 minutes and were diluted to 1:4 in Ham's F-10 medium. A 100-J..tL aliquot of motile donor sperm was added to 100 J..tL of diluted seminal plasma and incubated at 37°C for 60 minutes. The sperm suspension was washed with 2 mL Ham's F-10, and the pellet was resuspended into 200 to 300 J..tL of the medium. Parallel samples were used for indirect IBT, indirect MAR, and indirect FCM measurements. SpermMar-IgG and -IgA were used as by the direct tests according to the manufacture's instructions. Briefly, 10 J..tL of latex particles coated with human IgG or IgA and 10 J..tL of SpermMar IgG or IgA antiserum was mixed with 10 J..tL of donor sperm suspension incubated in patient's seminal plasma. The reaction was examined by light microscopy using 400x magnification. Two hundred spermatozoa were counted and the percentage of motile spermatozoa carrying one or more particles were determined. The MAR reaction > 10% was considered positive. For indirect IBT testing, anti-IgG and anti-IgA beads were obtained from Bio-Rad Laboratories (Irvine Scientific, Santa Ana, CA). The immunobead reagents were reconstituted with deionized water and the working solution of 10 mg/mL was stored at 4°C. When required, an aliquot was washed and resuspended in Earle's medium containing 1% bovine serum albumin at the same concentration. For analysis, 10 J..tL of beads and 10 J..tL of sperm suspen172

Rasiinen et a1. Seminal plasma antisperm antibodies

sion were mixed and the positivity of the sample was determined as with the MAR test. The IBT reaction > 20% was considered positive. Indirect Flow Cytometry

The FITC-conjugated F(ab')z fragments of rabbit anti-human IgG or IgA antibodies were used for indirect FCM. Ten microliters of 200 J..tg/mL IgG and 100 J..tg/mL IgA were added to 1 X 106 spermatozoa, which incubated with 1:4 diluted seminal plasma. An isotype control antibody was used with every sample. Mter 30 minutes incubation in the dark, at room temperature, the samples were washed with 3 mL of phosphate-buffered saline and centrifuged at 300 X g for 5 minutes. The pellet was resuspended in 0.5 mL of phosphate-buffered saline and 5 J..tL of 2 mg/mL propidium iodide was added. The flow cytometry was done as described before (15). Immunoglobulin G or IgA were considered to be present on spermatozoa if > 10% of live spermatozoa stained positively. Also the sperm incubation with undiluted seminal plasma was tested; because the results were similar to 1:4 dilution, the latter was used as with other tests. Statistical Analysis

Statistical analysis was done using statistical software, StatXact-Turbo 1992 (Cytel Software Corporation, Cambridge, MA). Linear by linear association test, exact method was used to correlate the methods (see Table 2), and the significance between the antibody positivity measured by direct FCM as compared with that measured by indirect FCM, indirect MAR, and indirect IBT was tested pairwise with Wilcoxon signed rank test, computed by an exact method. RESULTS Control Samples

Four control samples were negative by IgGSpermMar test. They also were, as expected, negative by direct FCM for IgG and IgA. On indirect FCM, some control samples unexpectedly had up to a 20% positivity for IgA, but all were negative for IgG. The control patients were negative for IgG and IgA binding as measured by indirect MAR and indirect IBT tests. Direct Antibody Measurements

The IgG positivity by direct MAR was studied in 11 semen samples and was 86% ± 25% (mean ± SD, range 19% to 100%). Although the proportion ofspermatozoa positive for IgG by direct FCM was lower, Fertility and Sterility

Table 1 Antisperm IgG and IgA Measured by Direct Flow Cytometry* Patient

IgG

IgA

Control

IgG

IgA

1 2 3 4 5 6 7 8 9 10 11

31 4 91 21 71 90 22 59 86 62 65

57 90 69 73 53 86 43 96 63 18 60

1 2 3 4

4 2 1 1

7 9 5 3

with lower IgG binding (43% : :': : 31%), whereas indirect FCM was positive only in 5 of 11 (45%) samples with a very low proportion ofIgG-positive spermatozoa (13% : :': : 12%), which was a statistically significant finding (P = 0.001). For detecting IgA on seminal plasma, both indirect MAR and indirect IBT were positive in 10 of 11 (91%), with mean antibody binding of 54% : :': : 27% and 58% : :': : 24%, respectively. All samples were positive for IgA by indirect FCM but with lower mean binding (40% : :': : 19%). U sing different methods of measurement, the 11 individual samples also were compared with each other by linear by linear association test, exact method (Table 2). No correlation was found with the direct FCM and indirect FCM antis perm antibody measurements of the samples. However, there was a significant correlation between the IgG-direct FCM and the IgG-indirect MAR (P < 0.05). The result of the IgG-indirect FCM correlated significantly with that of the IgG-indirect IBT (P < 0.01) but not with the results of the IgG-indirect MAR. However, the results of the IgG-indirect IBT and the IgG-indirect MAR showed a good correlation (P < 0.05). The results ofIgA measurements by indirect FCM, indirect IBT, and indirect MAR did not correlate significantly.

* Results are expressed as the percentage of spermatozoa positive for antibodies.

55 : :': : 25% (range 4% to 91%), there was a good correlation between the direct MAR and the direct FCM measurements (r = 0.77, P < 0.01). One of the samples was IgG negative by direct FCM, correspondingly a low positivity also was observed by the IgGdirect MAR (19%). However, the same sample was strongly IgA positive by direct FCM (90%). The proportion of spermatozoa positive for IgA by direct FCM was 64% : :': : 22% (range 18% to 96%). The percentage of spermatozoa positive for IgG and IgA antibodies in the 11 antisperm antibody-positive semen samples and four control semen samples as measured by direct FCM is shown (Table 1).

DISCUSSION

Despite the limited number of patients studied in this work, our results clearly suggest that direct FCM does not correlate with indirect FCM for the measurement of IgG antisperm antibodies and correlates only weakly for the measurement ofIgA antisperm antibodies (Table 2). Also the correlation between the indirect IBT and indirect MAR with direct FCM with respect to the clinically relevant IgA antibodies was weak. A possible reason for the discrepancy between direct FCM and indirect FCM is that tail tip-directed antibody binding is detected less accurately than head-directed antibody binding in FCM. This expla-

Indirect Antibody Measurements of Seminal Plasma and the Comparison of Tests

The corresponding measurements from seminal plasma of 11 antibody-positive semen samples by indirect MAR, indirect IBT, and indirect FCM for IgG and IgA are shown in Figure 1. For detecting IgG in seminal plasma, the indirect MAR test was most sensitive, showing positivity in 9 of 11 (82%) of the samples and the highest proportion of spermatozoa with IgG (71% : :': : 34%). Correspondingly, indirect IBT was positive in 8 of 11 (73%) samples but

100

Figure 1 The comparison of the percentage of spermatozoa positive for IgG (A) (., MAR IgG; Ll, IBT IgG; !iJJ!, FCM IgG) and IgA (B) (., MAR IgA; Ll, IBT IgA; !iJJ!, FCM IgA) antibodies measured by indirect MAR, indirect IBT, and indirect FCM from seminal plasma of 11 antibody-positive patients. Vol. 65, No.1, January 1996

A

100

80

80

60

60

40

40

20

20

B

Q)

Cl

'"Q) C 0

li;

c..

0 2

3

4

5

6

7

8

9

1011

Riisiinen et a}.

o 234567891011

Seminal plasma antisperm antibodies

173

Table 2 The Correlation Between the Different Methods of Measuring IgG and IgA Antisperm Antibodies* IgG

IgA

Compared methods

r

P

r

P

Direct FCM and indirect FCM Direct FCM and indirect MAR Direct FCM and indirect IBT Indirect FCM and indirect MAR Indirect FCM and indirect IBT Indirect MAR and indirect IBT

0.3 0.7 0.4 0.5 0.8 0.7

0.39 0.01 0.28 0.06 0.001 0.01

0.6 0.1 0.1 0.5 0.4 0.6

0.05 0.77 0.77 0.14 0.22 0.07

* The methods were compared using linear by linear association test, exact method, where r = Pearson correlation coefficient.

nation appears to be unlikely, because we have shown previously that sperm tails can be detected easily by FCM (15). However, the most frequent type of binding on the indirect MAR and the indirect IBT (data not shown) was tail tip binding, which is in agreement with the study of Bronson et al. (11). At variance to these findings, Sinton et al. (19) have reported that seminal plasma antisperm antibodies bind predominantly to sperm head. In agreement with our study, Bronson et al. (11) showed the discrepancy between sperm-bound and unbound antibodies. Comparing the direct IBT to the indirect IBT from the same ejaculates they found that the unbound antibodies remaining in the seminal fluid did not reflect the Ig bound on the sperm surface, because head-directed antibodies were detected on spermatozoa but not from seminal fluid in 18 of 29 cases. The reasons for this discrepancy are not understood completely, but antigen densities on the sperm head might be higher as compared with the tail and as a result more Ig bind to spermatozoa relative to those remaining free in the seminal fluid (11). This finding also was confirmed by Menkveld et al. (20) who have shown that the site ofimmunobead binding, whether to sperm head or tail, does not always agree with the results observed on direct IBT on spermatozoa and indirect IBT of seminal plasma. These discrepancies are significant, because IgA binding to tails and/or heads affects the penetration of spermatozoa in endocervical mucus. Taken together, these results suggest that the antisperm antibodies measured by the indirect methods differ from those measured by the direct ones. It is possible, however, that the location of antibody binding dictates the sensitivity of the detection methods used and the antibody detection by solidphase assays may contribute to greater sensitivity. Because we did not localize the site of antisperm antibody binding in this study, it is difficult to make any definite conclusions. Haas et al. (21) have reported that the indirect IBT is more sensitive than 174

Rasiinen et al. Seminal plasma antisperm antibodies

FCM for detecting low titers of serum antisperm antibodies. Further studies are needed to confirm this finding with respect to seminal fluid antibodies. However, we found a good correlation between indirect FCM and indirect IBT for IgG measurements in seminal plasma. Our finding that indirect IBT and indirect MAR correlated for IgG but not for IgA antibodies is in agreement with Andreou et al. (22), who reported that MAR and IBT showed a good agreement with regard to the detection ofIgG antibodies on spermatozoa and in seminal plasma. However, the IgGMAR was found to be more sensitive than the IBT, which was also the case in our study. When evaluating data such as in Table 2, it is useful to appreciate that antisperm antibodies in a patient's seminal plasma bind to his own spermatozoa and that only those unbound antibodies that are unable to bind to a patient's own spermatozoa (although being capable of binding to donor spermatozoa) are detected by the indirect tests. Although the probable reason for the presence of unbound antibodies in seminal plasma traditionally is believed to be due to the excess of antibodies (11) relative to the patient's sperm specific antigens, the presence of unbound antibodies due to a reactivity different than that found on the patient's spermatozoa cannot be excluded automatically. Interestingly, although direct FCM does not correlate with either indirect FCM or indirect IBT, the indirect IBT correlates moderately both with indirect FCM and with indirect MAR. We believe that the anomalous correlations observed in Table 2 are due to the differences in reactivities as well as in the amounts ofIgG and IgA antibodies that are coupled by the manufacturer to the immunobeads or latex beads in the indirect tests. If this hypothesis is correct, then uniformity between the indirect tests might be obtained by using the same antibodies in defined amounts to coat the beads in the indirect MAR and indirect IBT tests. In conclusion, using an objective FCM method with high sensitivity and specificity (15, 16), this study shows that the indirect measurement of sperm antibodies may be grossly unreliable as compared with the results obtained by direct FCM. This finding provides an objective basis for a broader discussion of the role of indirect antisperm antibody testing. At present, indirect tests for seminal plasma antisperm antibodies should be interpreted with caution because of the disagreement between different methods of measuring sperm antibodies as compared with direct FCM.

Acknowledgments. We thank Outi Hovatta, M.D., from the Fertility and Sterility

Finnish Population and Family Welfare Foundation, Helsinki, Finland, for scientific advice. Mr. Simo Pohjavaara is thanked for skillful assistance in performing the antibody tests and Ilkka Penttila, M.D., from the Department of Clinical Chemistry, Kuopio University Hospital, Kuopio, Finland, for allowing use of department facilities. The skillful statistical assistance of Mikko HippelaYnen, M.D., from the Department of Surgery, Kuopio University Hospital, Kuopio, Finland, also is gratefully acknowledged.

12.

13.

14.

REFERENCES 1. Marshburn PB, Kutteh WH. The role of antisperm antibodies in infertility. Fertil Steril 1994;61:799-811. 2. Collins JA, Burrows EA, Yeo J, YoungLai EV. Frequency and predictive value of antisperm antibodies among infertile couples. Hum Reprod 1993;8:592-8. 3. Haas GG Jr. How should sperm antibody tests be used clinically? Am J Reprod Immunol MicrobioI1987;15:106-11. 4. Comhaire FH. Sperm antibodies-"gold standard" [letterl. Fertil Steril 1993;59:242. 5. Jager S, Kremer J, van Slochteren-Draaisma T. A simple method of screening for antisperm antibodies in human male: detection of spermatozoal surface IgG with the direct mixed antiglodulin reaction carried out on untreated fresh human semen. Int J Fertil 1978;23:12-21. 6. Hendry WF, Stedronska J, Lake RA. Mixed erythrocyte-spermatozoa antiglobulin reaction (MAR test) for IgA antisperm antibodies in subfertile males. Fertil Steril1982;37:108-12. 7. Bronson RA, Cooper GW, Rosenfeld DL. Correlation between regional specificity of antisperm antibodies to the spermatozoan surface and complement mediated sperm immobilization. Am J Reprod ImmunoI1982;2:222-4. 8. Shulman S, Pretorius E, Keane T. Antibodies to spermatozoa: XI. The use of immunobeads for the detection of sperm antibodies in serum. Am J Reprod Immunol Microbiol 1985;9: 62-6. 9. Barratt CLR, Havelock ML, Harrison PE, Cooke ID. Antisperm antibodies are more prevalent in men with low sperm motility. Int J Androl 1989; 12:110-6. 10. Matson PL, Junk SM, Spittle JW, Yovich JL. Effect of antispermatozoal antibodies in seminal plasma upon spermatozoal function. Int J Androl 1988; 11:101-6. 11. Bronson RA, Cooper GW, Rosenfeld DL. Seminal fluid anti-

Vol. 65, No.1, January 1996

15.

16.

17.

18.

19.

20.

21.

22.

sperm antibodies do not reflect those present on the sperm surface. Fertil Steril 1987;48:505-6. Hellstrom WJG, Samuels SJ, Waits AB, Overstreet JW. A comparison of the usefulness of SpermMar and Immunobead tests for the detection of antisperm antibodies. Fertil Steril 1989;52:1027-31. Friberg J. A simple and sensitive micromethod for demonstration of sperm agglutinating activity in serum from infertile men and women. Acta Obstet Gynecol Scand Suppl 1974;36:21-9. Meinertz H, Linnet L, Fogh-Andersen P, Hjort T. Antisperm antibodies and fertility after vasovasostomy: a follow-up study of 216 men. Fertil Steril 1990;54:315-21. Rasanen ML, Hovatta OL, Penttila 1M, Agrawal YA. Detection and quantitation of sperm-bound antibodies by flow cytometry of human semen. J Androl 1992; 13:55-64. Nikolaeva MA, Kulakov VI, Ter-Avanesov GV, Terekhina LN, Pshenichnikova TJ, Sukhikh GT. Detection of antisperm antibodies on the surface of living spermatozoa using flow cytometry: preliminary study. Fertil Steril 1993;59:639-44. Rasanen M, Saarikoski S, Penttila I, Agrawal YP. A flow cytometric study of low dose cyclic prednisolone treatment on sperm-bound antibody levels. Hum Reprod 1994;9: 889-93. Rasanen M, Lahteenmaki A, Saarikoski S, Agrawal YP. Comparison of flow cytometric measurement of seminal antisperm antibodies with the mixed antiglobulin reaction and the serum tray agglutination test. Fertil Steril 1994;61:143-50. Sinton EB, Riemann DC, Aston ME. Antisperm antibodies detection using concurrent cytofluorometry and indirect immunofluorescence microscopy. Am J Clin Pathol 1991;95: 242-6. Menkveld R, Kruger TF, Kotze TJvW, Windt M-L, Pretorius E. Detection of sperm antibodies on unwashed spermatozoa with the immunobead test: a comparison of results with the routine method and seminal plasma TAT titers and SCMC test. Am J Reprod Immunol 1991;25:88-91. Haas GG Jr, D'Cruz OJ, DeBault LE. Comparison ofthe indirect immunobead, radiolabeled, and immunofluorescence assays for immunoglobulin G serum antibodies to human sperm. Fertil Steril 1991;55:377-88. Andreou E, Mahmoud A, Vermeulen L, Schoonjans F, Comhaire F. Comparison of different methods for the investigation of antisperm antibodies on spermatozoa, in seminal plasma and in serum. Hum Reprod 1995;10:125-31.

Riisiinen et a1. Seminal plasma antisperm antibodies

175