- Email: [email protected]
Comparison of methods to enumerate white blood cells in semen
Vol. 60, No.2, August 1993
FERTILITY AND STERILITY
Printed on acid-free paper in U. S. A.
Copyright" 1993 The American Fertility Society
Comparison of methods to enumerate white blood cells in semen
Joseph A. Politch, Ph.D. Hans Wolff, M.D. Joseph A. Hill, M.D. Deborah J. Anderson, Ph.D.* Fearing Research Laboratory, Department of Obstetrics, Gynecology, and Reproductive Biology, Brigham and Women's Hospital, Harvard Medical Schaol, Boston, Massachusetts
High white blood cell (WBC) concentrations in semen, referred to as leukocytospermia (1) (also known as leukospermia, pyospermia, or pyosemia), have been associated with genital tract infection (2) and poor semen quality (3). Elucidating the interaction between leukocytospermia and genital tract infection and its potential role in infertility depends on accurate assessment of WBC in semen. Numerous techniques are used to detect seminal WBC. Direct counting of round cells in wet mounts is widely practiced but is highly inaccurate because WBC are not distinguished from immature germ cells, which are also commonly found in semen (4). The Papanicolaou and modified Bryan-Leishman staining techniques to detect WBC in semen are time-consuming, require an expert reader, and are imprecise because some forms of immature germ cells resemble WBC even after staining (1). In 1974, Endtz (5) described a simple peroxidase test for enumerating WBC in semen. This method is now widely used, and a similar technique has been recommended by the World Health Organization (WHO) (1). However, the peroxidase assay principally detects granulocytes, which comprise only a subpopulation of the WBC found in semen (6). More recently, immunohistologic techniques employing monoclonal anti-WBC antibodies have been used by several laboratories to enumerate total WBC and specific WBC subpopulations in semen (6). This approach is now considered by many
to be the gold standard of semen WBC tests. However, preparation of samples, immunohistologic staining, and evaluation of stained smears is timeconsuming and expensive. The purpose of this study was to compare results obtained from the peroxidase test with those obtained with the highly specific immunohistologic technique to determine whether the peroxidase test is a reliable screening method for WBC in semen. In addition, the association of total round cells with WBC in semen was determined to assess the accuracy of using total round cell counts either as a substitute for WBC counts or as an initial screen to select samples for subsequent WBC enumeration.
MATERIALS AND METHODS Semen
Semen samples were obtained from 87 men undergoing semen analysis for fertility evaluation by the Fertility and Endocrinology Unit at Brigham and Women's Hospital, Boston, Massachusetts, and from 25 volunteers of whom 16 were of proven fertility and 9 were of unknown fertility status. Samples were collected by masturbation into sterile urinalysis containers after a minimum abstinence period of 48 hours. Samples were processed within 4 hours of collection. Peroxidase Test
Received February 5,1993; revised and accepted April 7, 1993. * Reprint requests: Deborah J. Anderson, Ph.D., Fearing Research Laboratory, 250 Longwood Avenue, Boston, Massachusetts 02115.
372
Politch et aI.
Communications-in-brief
The peroxidase method used was adapted from Endtz (5). A stock solution was prepared by mixing 50 mL distilled water, 50 mL 96% ethanol, and 125 Fertility and Sterility
mg benzidine (Sigma Chemical Co., St. Louis, MO). The working solution was obtained by adding 5 JLL 30% H 2 0 2 to 4 mL of stock solution. Twenty microliters of working solution were mixed with 20 JLL of liquefied ejaculate and incubated for 5 minutes at room temperature; 20 JLL of this solution were diluted into 20 JLL of phosphate-buffered saline immediately before counting. Then, 10 JLL were placed in a hemocytometer, and peroxidase-positive (i.e., intensely brown-stained round cells the size of granulocytes) as well as peroxidase-negative (unstained) round cells were counted at 400X magnification on a phase-contrast microscope. Immunohistologic Test
The immunohistochemistry technique employed was a modification of a method we have previously described (6). White blood cells were identified by a monoclonal antibody (mAb) , HLe-1, that reacts with the leukocyte common antigen CD45 (Becton Dickinson, Mountain View, CA). Antibody binding was visualized with a commercial alkaline phosphatase antialkaline phosphatase test kit (Dako Corp., Carpinteria, CA). Using this method, WBC were stained deep red, whereas immature germ cells and epithelial cells revealed only a light blue hematoxylin counterstain. The number of immature germ cells was approximated by subtracting the number of HLe-1-positive (WBC) from the total round cell count. Data Analysis
For data analysis, Friedman two-way analysis of variance (AN OVA) by ranks, Wilcoxon's signed rank test, and Spearman rank correlation coefficient were used where applicable. Nonparametric statistics were used because all semen cells examined exhibited unimodal, leptokurtic, and positively skewed (non-normal) distributions. Statistical analysis was performed on a microcomputer (Macintosh IIci; Apple Computer, Inc., Cupertino, CA) with the StatView II statistical program (Abacus Concepts, Inc., Berkeley, CA).
RESULTS Subjects were combined for statistical analysis (n 112) because it was not the purpose ofthis paper to compare infertility patients with laboratory donors. The combined group is representative ofheterogeneous populations encountered in most infertility laboratories. =
Vol. 60, No.2, August 1993
Among the semen samples from this group, the concentration of total round cells (median, 1.18 X 106/mL) was the highest, followed by immature germ cells (median, 7.73 X 105 /mL), HIe-I-positive cells (median, 1.11 X 105 /mL), andperoxidase-positive cells (median, 2.40 X 104 /mL). These concentrations were significantly different from one another, as determined by Friedman ANOV A by ranks (P < 0.0001) and post hoc Wilcoxon signed rank tests with the Bonferroni correction (all Ps < 0.0001). White blood cell concentrations obtained by the peroxidase and immunohistology methods were highly associated (p = +0.70, P < 0.0001). However, as indicated above, WBC concentrations obtained with the immunohistology technique were significantly higher than those obtained with the peroxidase method. Thus, when applying the threshold for leukocytospermia recommended by the WHO (l06 WBC/mL) to the two tests, a number of samples were discordant. By the peroxidase test, 10 of 112 semen samples (8.9%) contained >106 WBC/ mL, whereas by the immunohistology test, 17 of 112 (15.2%) had >106 WBC/mL. When compared with the immunohistology test and using 106 WBC/mL as the threshold, the specificity of the peroxidase method to detect leukocytospermia was 100% (10/ 10), but the sensitivity was only 58.8% (10/17). Lowering the threshold for leukocytospermia to 5 X 105 /mL for the peroxidase assay, had no effect on the specificity, which remained 100% (16/16); however, the sensitivity was increased to 94.1 % (16/17). Raising the threshold for leukocytospermia by the immunohistology technique to 2 X 106/mL and using a threshold of 1 X 106/mL for the peroxidase assay gave both a sensitivity and specificity of 90% (9/10). The raw data for these two assays are graphically depicted in Figure 1. For most of the samples, the relationship between peroxidase-positive cells and HLe-1-positive cells was linear. In the low range «106 WBC/mL), there was disagreement between the two techniques in some cases, probably because of counting inaccuracy when few WBC were present. The correlations between total round cell and HLe-1-positive cell concentrations (p = +0.58, P < 0.0001) and between total round cell and peroxidase-positive cell concentrations (p = +0.42, P < 0.0001) were statistically significant. However, these relationships were much weaker than that between the two WBC methods (p = +0.70, P < 0.0001). The correlation coefficients (p) between immature germ cell concentrations and total Politch et a1.
Communications-in-brief
373
10 8
o Spearman Rank Order Correlation Coefficient (n=112) Rho=+O.70 (p
10 7
10 6
~
6
o 8 10 5
~
+
'"QJ'" I
~
e
10 4
o
::c
-
bO 10 3
0
10 2
N=3
/ log Peroxidase+/mL Figure 1 Correlation of log concentrations of HLe-l-positive cells and peroxidase-positive cells in semen of 112 men. HLe-l-positive cells were determined by immunohistology and peroxidase-positive cells by a modification of the method of Endtz (5).
round cell, HLe-1-positive cell, and peroxidase-positive cell concentrations were +0.82 (P < 0.0001), +0.23 (P < 0.05), and +0.07 (nonsignificant), respectively. Using various threshold values (106 /mL, 5 X 106 /mL, and 10 X 106 /mL) for total round cell concentrations and the 106 /mL cutoff for HLe-1-positive cell concentrations, there was very poor concordance between the two counting methods. When a threshold of 106 /mL round cells was used, sensitivity was 100% (17/17), but specificity was only 26.6% (17/64). Using 5 X 106 /mL round cells, sensitivity was 58.82% (10/17), whereas specificity was 76.92% (10/13). At 10 X 106 /mL round cells, sensitivity was 11.76% (2/17) and specificity was 50% (2/4). DISCUSSION
Simple and reliable methods for the detection of WBC in semen are needed to better understand the prevalence and significance of male genital tract infiammation. The WHO laboratory manual for semen analysis defines leukocytospermia as seminal WBC counts> 106 /mL but recommends the use of either peroxidase or immunohistology assays for 374
Politch et al.
Communications-in-brief
enumerating WBC in semen (1). The results ofthis study indicate that WBC counts obtained by these two methods are highly correlated but numerically different. The two methods give comparable results in screening for leukocytospermia only if different threshold concentrations are used. Threshold values of 1 X 106 /mL for the immunohistology assay and 5 X 105 /mL for the peroxidase assay or 2 X 106 /mL for immunohistology and 1 X 106 /mL for peroxidase resulted in good interassay sensitivity and specificity. More research is needed to establish which WBC types and concentrations are associated with genital tract infection and/or reduced fertility in male infertility patients. Seminal round cells are comprised almost exclusively of WBC and immature germ cells (1). Thus, immature germ cell concentration was calculated by subtracting WBC from total round cell concentration. Because the conventional histochemistry staining techniques for differentiating immature germ cells from WBC are time-consuming and problematic, the current method may prove valuable in future research. Because immature germ cells usually comprise a significantly larger proportion of round cells in semen than do WBC (median valFertility and Sterility
ues in this study, 7.73 X 105 /mL versus 1.11 X 105 / mL) and do not correlate well with WBC counts (p = +0.23), it is not surprising that total round cell concentrations correlated well with immature germ cell concentrations (p = +0.82) and weakly with WBC concentrations (p = +0.58). The poor sensitivity and specificity of total round cell counts in predicting WBC counts by either the peroxidase or immunohistology methods indicate that total round cell counts are of no diagnostic value for enumerating WBC in semen.
WBC/mL for the peroxidase assay and 2 X 106 WBC/mL for the immunohistology assay. We conclude that either peroxidase or immunohistology assays can be used to screen for leukocytospermia, but that more research is needed to establish thresh01ds for pathological levels of WBC in semen using these two approaches. Total round cell counts are of no value for enumerating WBC in semen.
Key Words: Semen, white blood cells, leukocytospermia, leukocytes, male genital tract inflammation, peroxidase, monoclonal antibodies, immunohistochemistry, immature germ cells, infertility.
SUMMARY
Seminal WBC counts obtained by an mAb-based immunohistologic method correlated well with seminal granulocyte counts obtained with a simple peroxidase method (p = +0.70; P < 0.0001). However, total WBC counts were significantly higher than granulocyte counts for most samples. With the immunohistologic method, 17 of 112 samples (15.2%) contained> 106 WBC/mL semen, whereas the peroxidase method resulted in only 10 samples (8.9%) with >10 6 WBC/mL. When the threshold defining leukocytospermia was set at 1 X 106 positive cells/ mL for both methods, the specificity of the peroxidase test compared with the immunohistology technique was 100% (10/10), but the sensitivity was only 58.8% (10/17). When the threshold for leukocytospermia in the peroxidase test was lowered to 5 X 105 positive cells/mL semen, the sensitivity relative to the immunohistology technique increased to 94.1 % (16/17), and specificity remained 100% (16/ 16). Likewise, good interassay sensitivity and specificity values were obtained with thresholds of 106
Vol. 60, No.2, August 1993
Acknowledgments. The authors thank Amy Abbott, B.S., Adriana Martinez, B.S., and Ai-Wei Wang, M.D., for technical assistance.
REFERENCES 1. World Health Organization. WHO Laboratory manual for the examination of human semen and sperm-cervical mucus interaction. 3rd ed. Cambridge: The Press Syndicate of the University of Cambridge, 1992:8-50. 2. Comhaire F, Verschraegen G, Vermeulen L. Diagnosis of accessory gland infection and its possible role in male infertility. Int J Androl 1980;3:32-45. 3. Wolff H, Politch JA, Martinez A, Haimovici F, Hill JA, Anderson DJ. Leukocytospermia is associated with poor semen quality. Fertil Steril 1991;53:528-36. 4. Comhaire F. Diagnosis and treatment of male adnexitis in relation to infertility. In: Negro-Vilar A, editor. Male reproduction and fertility. New York: Raven Press, 1983:265-72. 5. Endtz AW. A rapid staining method for differentiating granulocytes from "germinal cells" in Papanicolaou-stained semen. Acta CytoI1974;18:2-7. 6. Wolff H, Anderson DJ. Immunohistological characterization and quantitation of leukocyte subpopulations in human semen. Fertil Steril 1988;49:497-504.
Politch et al.
Communications-in-brief
375
FERTILITY AND STERILITY
Printed on acid-free paper in U. S. A.
Copyright" 1993 The American Fertility Society
Comparison of methods to enumerate white blood cells in semen
Joseph A. Politch, Ph.D. Hans Wolff, M.D. Joseph A. Hill, M.D. Deborah J. Anderson, Ph.D.* Fearing Research Laboratory, Department of Obstetrics, Gynecology, and Reproductive Biology, Brigham and Women's Hospital, Harvard Medical Schaol, Boston, Massachusetts
High white blood cell (WBC) concentrations in semen, referred to as leukocytospermia (1) (also known as leukospermia, pyospermia, or pyosemia), have been associated with genital tract infection (2) and poor semen quality (3). Elucidating the interaction between leukocytospermia and genital tract infection and its potential role in infertility depends on accurate assessment of WBC in semen. Numerous techniques are used to detect seminal WBC. Direct counting of round cells in wet mounts is widely practiced but is highly inaccurate because WBC are not distinguished from immature germ cells, which are also commonly found in semen (4). The Papanicolaou and modified Bryan-Leishman staining techniques to detect WBC in semen are time-consuming, require an expert reader, and are imprecise because some forms of immature germ cells resemble WBC even after staining (1). In 1974, Endtz (5) described a simple peroxidase test for enumerating WBC in semen. This method is now widely used, and a similar technique has been recommended by the World Health Organization (WHO) (1). However, the peroxidase assay principally detects granulocytes, which comprise only a subpopulation of the WBC found in semen (6). More recently, immunohistologic techniques employing monoclonal anti-WBC antibodies have been used by several laboratories to enumerate total WBC and specific WBC subpopulations in semen (6). This approach is now considered by many
to be the gold standard of semen WBC tests. However, preparation of samples, immunohistologic staining, and evaluation of stained smears is timeconsuming and expensive. The purpose of this study was to compare results obtained from the peroxidase test with those obtained with the highly specific immunohistologic technique to determine whether the peroxidase test is a reliable screening method for WBC in semen. In addition, the association of total round cells with WBC in semen was determined to assess the accuracy of using total round cell counts either as a substitute for WBC counts or as an initial screen to select samples for subsequent WBC enumeration.
MATERIALS AND METHODS Semen
Semen samples were obtained from 87 men undergoing semen analysis for fertility evaluation by the Fertility and Endocrinology Unit at Brigham and Women's Hospital, Boston, Massachusetts, and from 25 volunteers of whom 16 were of proven fertility and 9 were of unknown fertility status. Samples were collected by masturbation into sterile urinalysis containers after a minimum abstinence period of 48 hours. Samples were processed within 4 hours of collection. Peroxidase Test
Received February 5,1993; revised and accepted April 7, 1993. * Reprint requests: Deborah J. Anderson, Ph.D., Fearing Research Laboratory, 250 Longwood Avenue, Boston, Massachusetts 02115.
372
Politch et aI.
Communications-in-brief
The peroxidase method used was adapted from Endtz (5). A stock solution was prepared by mixing 50 mL distilled water, 50 mL 96% ethanol, and 125 Fertility and Sterility
mg benzidine (Sigma Chemical Co., St. Louis, MO). The working solution was obtained by adding 5 JLL 30% H 2 0 2 to 4 mL of stock solution. Twenty microliters of working solution were mixed with 20 JLL of liquefied ejaculate and incubated for 5 minutes at room temperature; 20 JLL of this solution were diluted into 20 JLL of phosphate-buffered saline immediately before counting. Then, 10 JLL were placed in a hemocytometer, and peroxidase-positive (i.e., intensely brown-stained round cells the size of granulocytes) as well as peroxidase-negative (unstained) round cells were counted at 400X magnification on a phase-contrast microscope. Immunohistologic Test
The immunohistochemistry technique employed was a modification of a method we have previously described (6). White blood cells were identified by a monoclonal antibody (mAb) , HLe-1, that reacts with the leukocyte common antigen CD45 (Becton Dickinson, Mountain View, CA). Antibody binding was visualized with a commercial alkaline phosphatase antialkaline phosphatase test kit (Dako Corp., Carpinteria, CA). Using this method, WBC were stained deep red, whereas immature germ cells and epithelial cells revealed only a light blue hematoxylin counterstain. The number of immature germ cells was approximated by subtracting the number of HLe-1-positive (WBC) from the total round cell count. Data Analysis
For data analysis, Friedman two-way analysis of variance (AN OVA) by ranks, Wilcoxon's signed rank test, and Spearman rank correlation coefficient were used where applicable. Nonparametric statistics were used because all semen cells examined exhibited unimodal, leptokurtic, and positively skewed (non-normal) distributions. Statistical analysis was performed on a microcomputer (Macintosh IIci; Apple Computer, Inc., Cupertino, CA) with the StatView II statistical program (Abacus Concepts, Inc., Berkeley, CA).
RESULTS Subjects were combined for statistical analysis (n 112) because it was not the purpose ofthis paper to compare infertility patients with laboratory donors. The combined group is representative ofheterogeneous populations encountered in most infertility laboratories. =
Vol. 60, No.2, August 1993
Among the semen samples from this group, the concentration of total round cells (median, 1.18 X 106/mL) was the highest, followed by immature germ cells (median, 7.73 X 105 /mL), HIe-I-positive cells (median, 1.11 X 105 /mL), andperoxidase-positive cells (median, 2.40 X 104 /mL). These concentrations were significantly different from one another, as determined by Friedman ANOV A by ranks (P < 0.0001) and post hoc Wilcoxon signed rank tests with the Bonferroni correction (all Ps < 0.0001). White blood cell concentrations obtained by the peroxidase and immunohistology methods were highly associated (p = +0.70, P < 0.0001). However, as indicated above, WBC concentrations obtained with the immunohistology technique were significantly higher than those obtained with the peroxidase method. Thus, when applying the threshold for leukocytospermia recommended by the WHO (l06 WBC/mL) to the two tests, a number of samples were discordant. By the peroxidase test, 10 of 112 semen samples (8.9%) contained >106 WBC/ mL, whereas by the immunohistology test, 17 of 112 (15.2%) had >106 WBC/mL. When compared with the immunohistology test and using 106 WBC/mL as the threshold, the specificity of the peroxidase method to detect leukocytospermia was 100% (10/ 10), but the sensitivity was only 58.8% (10/17). Lowering the threshold for leukocytospermia to 5 X 105 /mL for the peroxidase assay, had no effect on the specificity, which remained 100% (16/16); however, the sensitivity was increased to 94.1 % (16/17). Raising the threshold for leukocytospermia by the immunohistology technique to 2 X 106/mL and using a threshold of 1 X 106/mL for the peroxidase assay gave both a sensitivity and specificity of 90% (9/10). The raw data for these two assays are graphically depicted in Figure 1. For most of the samples, the relationship between peroxidase-positive cells and HLe-1-positive cells was linear. In the low range «106 WBC/mL), there was disagreement between the two techniques in some cases, probably because of counting inaccuracy when few WBC were present. The correlations between total round cell and HLe-1-positive cell concentrations (p = +0.58, P < 0.0001) and between total round cell and peroxidase-positive cell concentrations (p = +0.42, P < 0.0001) were statistically significant. However, these relationships were much weaker than that between the two WBC methods (p = +0.70, P < 0.0001). The correlation coefficients (p) between immature germ cell concentrations and total Politch et a1.
Communications-in-brief
373
10 8
o Spearman Rank Order Correlation Coefficient (n=112) Rho=+O.70 (p
10 7
10 6
~
6
o 8 10 5
~
+
'"QJ'" I
~
e
10 4
o
::c
-
bO 10 3
0
10 2
N=3
/ log Peroxidase+/mL Figure 1 Correlation of log concentrations of HLe-l-positive cells and peroxidase-positive cells in semen of 112 men. HLe-l-positive cells were determined by immunohistology and peroxidase-positive cells by a modification of the method of Endtz (5).
round cell, HLe-1-positive cell, and peroxidase-positive cell concentrations were +0.82 (P < 0.0001), +0.23 (P < 0.05), and +0.07 (nonsignificant), respectively. Using various threshold values (106 /mL, 5 X 106 /mL, and 10 X 106 /mL) for total round cell concentrations and the 106 /mL cutoff for HLe-1-positive cell concentrations, there was very poor concordance between the two counting methods. When a threshold of 106 /mL round cells was used, sensitivity was 100% (17/17), but specificity was only 26.6% (17/64). Using 5 X 106 /mL round cells, sensitivity was 58.82% (10/17), whereas specificity was 76.92% (10/13). At 10 X 106 /mL round cells, sensitivity was 11.76% (2/17) and specificity was 50% (2/4). DISCUSSION
Simple and reliable methods for the detection of WBC in semen are needed to better understand the prevalence and significance of male genital tract infiammation. The WHO laboratory manual for semen analysis defines leukocytospermia as seminal WBC counts> 106 /mL but recommends the use of either peroxidase or immunohistology assays for 374
Politch et al.
Communications-in-brief
enumerating WBC in semen (1). The results ofthis study indicate that WBC counts obtained by these two methods are highly correlated but numerically different. The two methods give comparable results in screening for leukocytospermia only if different threshold concentrations are used. Threshold values of 1 X 106 /mL for the immunohistology assay and 5 X 105 /mL for the peroxidase assay or 2 X 106 /mL for immunohistology and 1 X 106 /mL for peroxidase resulted in good interassay sensitivity and specificity. More research is needed to establish which WBC types and concentrations are associated with genital tract infection and/or reduced fertility in male infertility patients. Seminal round cells are comprised almost exclusively of WBC and immature germ cells (1). Thus, immature germ cell concentration was calculated by subtracting WBC from total round cell concentration. Because the conventional histochemistry staining techniques for differentiating immature germ cells from WBC are time-consuming and problematic, the current method may prove valuable in future research. Because immature germ cells usually comprise a significantly larger proportion of round cells in semen than do WBC (median valFertility and Sterility
ues in this study, 7.73 X 105 /mL versus 1.11 X 105 / mL) and do not correlate well with WBC counts (p = +0.23), it is not surprising that total round cell concentrations correlated well with immature germ cell concentrations (p = +0.82) and weakly with WBC concentrations (p = +0.58). The poor sensitivity and specificity of total round cell counts in predicting WBC counts by either the peroxidase or immunohistology methods indicate that total round cell counts are of no diagnostic value for enumerating WBC in semen.
WBC/mL for the peroxidase assay and 2 X 106 WBC/mL for the immunohistology assay. We conclude that either peroxidase or immunohistology assays can be used to screen for leukocytospermia, but that more research is needed to establish thresh01ds for pathological levels of WBC in semen using these two approaches. Total round cell counts are of no value for enumerating WBC in semen.
Key Words: Semen, white blood cells, leukocytospermia, leukocytes, male genital tract inflammation, peroxidase, monoclonal antibodies, immunohistochemistry, immature germ cells, infertility.
SUMMARY
Seminal WBC counts obtained by an mAb-based immunohistologic method correlated well with seminal granulocyte counts obtained with a simple peroxidase method (p = +0.70; P < 0.0001). However, total WBC counts were significantly higher than granulocyte counts for most samples. With the immunohistologic method, 17 of 112 samples (15.2%) contained> 106 WBC/mL semen, whereas the peroxidase method resulted in only 10 samples (8.9%) with >10 6 WBC/mL. When the threshold defining leukocytospermia was set at 1 X 106 positive cells/ mL for both methods, the specificity of the peroxidase test compared with the immunohistology technique was 100% (10/10), but the sensitivity was only 58.8% (10/17). When the threshold for leukocytospermia in the peroxidase test was lowered to 5 X 105 positive cells/mL semen, the sensitivity relative to the immunohistology technique increased to 94.1 % (16/17), and specificity remained 100% (16/ 16). Likewise, good interassay sensitivity and specificity values were obtained with thresholds of 106
Vol. 60, No.2, August 1993
Acknowledgments. The authors thank Amy Abbott, B.S., Adriana Martinez, B.S., and Ai-Wei Wang, M.D., for technical assistance.
REFERENCES 1. World Health Organization. WHO Laboratory manual for the examination of human semen and sperm-cervical mucus interaction. 3rd ed. Cambridge: The Press Syndicate of the University of Cambridge, 1992:8-50. 2. Comhaire F, Verschraegen G, Vermeulen L. Diagnosis of accessory gland infection and its possible role in male infertility. Int J Androl 1980;3:32-45. 3. Wolff H, Politch JA, Martinez A, Haimovici F, Hill JA, Anderson DJ. Leukocytospermia is associated with poor semen quality. Fertil Steril 1991;53:528-36. 4. Comhaire F. Diagnosis and treatment of male adnexitis in relation to infertility. In: Negro-Vilar A, editor. Male reproduction and fertility. New York: Raven Press, 1983:265-72. 5. Endtz AW. A rapid staining method for differentiating granulocytes from "germinal cells" in Papanicolaou-stained semen. Acta CytoI1974;18:2-7. 6. Wolff H, Anderson DJ. Immunohistological characterization and quantitation of leukocyte subpopulations in human semen. Fertil Steril 1988;49:497-504.
Politch et al.
Communications-in-brief
375