ANDROLOGIA zyxwvutsrqpon 30 (Supp. l), zyxwvut 49-53 (1998) Sperm morphology and male urogenital infections R. Menkveld and T. F. Kruger Reproductive Biology Unit, Department of Obstetrics and Gynaecology, Tygerberg Hospital and University of Stellenbosch, Tygerberg, South Africa Key words. Urogenital infections-leukocytospermia-strict sperm morphology-human zyxw Summary. The aim of this study was to investi- gate the influence of urogenital infections as indi- cated by leukocytospermia on human sperm morphology, diagnosed cytologically and by means of a leukocyte peroxidase test. A basic semen analysis, including a leukocyte peroxidase test, was prospectively performed on 150 consecutive semen samples. Cytology smears were microscopically investigated for the presence of WBC and the results expressed on a 4 point scale as zyxwvu f to zyxwvu + + + WBCs/high power field (HPF). ROC curve analy- sis indicated that for cases with more than WBC/HPF the peroxidase determined WBC count cut-off value was >0.25 x lo6 WBC ml-' with a sensitivity of 75% and specificity of 90%. The presence of more than +WBC/HPF was negatively correlated with sperm morphology characteristics studied. The mean ( + SD) percent- age of morphological normal spermatozoa was 7.0% (SD 4.4) in the WBC negative group zyxwvut (n= 134) compared to 4.3% (SD 3.5) in the WBC positive (n = 16) group (P< 0.000 1). There was also an associated increase, 15.3% (SD 13.3) to 23.6% (SD 13.8),in the percentage of spermatozoa with elongated head forms in the WBC positive group (P=0.0218). No other effect on sperm and acrosome morphology could be found. With the peroxidase determinations there was also a tend- ency in the WBC positive group (n = 10) towards poorer sperm morphology characteristics, but these changes were not statistically significant. The presence of urogenital infections as diagnosed cytologically was associated with statistically poorer sperm morphology characteristics. This statistical relationship was not found in the peroxi- dase diagnosed leukocytospermia positive groups. Correspondence: Dr Roelof Menkveld PhD, Andrology Laboratory E3, Tygerberg Hospital, 7505 Tygerberg, South Africa. E-mail: Rme@maties.sun.ac.za. Introduction According to the literature, the clinical diagnosis of urogenital infections appears to be a complex process (Weidner et al., 1991). Clinical diagnoses are mostly made in combination with laboratory tests (Rowe et al., 1993)or in many cases, diagnoses can be based on many different laboratory tests alone (Wolff, 1995).Although controversial (Aitken & Baker, 1995), the presence of leukocytospermia in semen samples is also regarded as indicators or mediators of genital tract infections (Comhaire, 1980; WHO, 1992; Wolff et al., 1990). One of the main reasons for this controversy is that no corre- lation exists between outcome of semen cultures for microorganisms and leukocytospermia (El-Demiry et al., 1986). To add to the complexity, many different methods also exist for the detection of leukocytospermia as well as for the identification of different leukocyte subsets, such as immunocyto- chemical (Tomlinson et al., 1992), the use of monoclonal antibodies (Eggert-Kruse zyx et al., 1992), different leukocyte peroxidase tests (Ludwig & Fick, 1990; WHO, 1992; Politch et al., 1993), and cytological (Riedel, 1980; Riedel & Semm, 1980; Leib et al., 1994). Furthermore, bad correlations exist between the different methods to determine the presence of WBC (Politch et al., 1993). Cytological identification of white blood cells has always been regarded as an insufficient method in much of the literature. The argument is the inac- curate diagnosis of various leukocyte subpopula- tions or even the ability to distinguish between different WBC forms and immature germinal epi- thelium cells (Wolff, 1995). However, positive identification of both groups is possible with a good staining method like the Papanicolaou stain (Riedel & Semm, 1980). To add further to the confusion, the WHO (1992) cut-off value of zy (2 1,0 x lo6 WBC ml-') for a positive diagnosis of leukocytospermia is an empirical one (Aitken & Baker, 1995).