1 Divisions of Reproductive Medicine and Andrology, Department of Obstetrics and Gynaecology, Hacettepe University, Medical Faculty, Ankara, Turkey, 2 Reproductive Biology Unit, Department of Obstetrics and Gynaecology, University of Stellenbosch and Tygerberg Hospital and 3 Biostatistics Unit, National Programme for Health Systems and Policy, Medical Research Council, Tygerberg, South Africa
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: in-vivo fertilization/semen parameters/sperm morphology
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Although the 5% threshold (strict criteria) (Lindheim et al., 1996; Ombelet et al., 1997a
; Coetzee et al., 1998
; Montanaro-Gauci et al., 1999) has been shown to be of clinical significance in the situations mentioned above, it is still an unanswered question when considering the general population since thresholds are not well established for this group. In a recent publication, Ombelet et al. (1997a) pointed out that using the receiver operating characteristics (ROC) curve, thresholds have been established for a group of patients in Belgium. Regarding sperm morphology, a 10% threshold was used as an indicator of subfertility (when below this threshold) or fertile population when above this threshold.
The question arises of whether the threshold value might vary in populations in different parts of the world. Therefore, an attempt to study the threshold values in a subfertile population in Turkey was undertaken as a complementary study of the Ombelet study done in Belgium (Ombelet et al., 1997a). The differentiating aspect of the study is that the morphological evaluation of the spermatozoa was done by one of the original authors of strict criteria (T.K.). The aim of the study was to determine the semen parameters of a proven fertile population and to compare these parameters with that of a subfertile group in the same region.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
During an 8 month period (from September 1996 to May 1997) and at a gestational age of 823 weeks, 69 pregnant women who attended the antenatal clinic at a university hospital were asked to participate in a study where the husbands' semen was needed to define the fertile population in Turkey. The couples provided written consent after being given full background information. Inclusion criteria were as follows: pregnancy had to be achieved within 1 year of marriage and no habitual abortions (>3) were allowed. On the other hand, 93 consecutive infertile men were recruited into the study as they were consulted at the Andrology Division of the IVF Unit at Hacettepe University, Ankara, Turkey. There were no infertility factors in the female partners of the couples with no ovulatory dysfunction, with tubal patency in hysterosalpingography and normal laparoscopic findings.
This study took place at the following two centres: (i) Andrology Division and Antenatal Polyclinic, Department of Obstetrics and Gynaecology, Hacettepe University, Medical Faculty, Ankara, Turkey; (ii) Reproductive Biology Unit, Department of Obstetrics and Gynaecology, Tygerberg Hospital, Tygerberg, Republic of South Africa. Sperm parameters (except morphology) were collected and evaluated at the first centre. A single investigator (T.K.) at the second centre evaluated sperm morphology.
In short, the methods used for determination of the three variables in this study were as follows: the semen samples were obtained after 34 days of abstinence by masturbation at the laboratory. Immediately after liquefaction, a drop of the well-mixed specimen was placed on a clean and pre-warmed glass slide at 37°C, covered with a coverslip, and left for a few minutes. The microscope was equipped with a hot stage to keep the slides at 37°C. The preparation was examined under a magnification of both x10 and x40 objectives. The motility assessment was done according to World Health Organization guidelines (WHO, 1992) in at least 10 separate randomly selected high-power fields, as described by MacLeod (1965). The motility of each spermatozoon was graded a, b, c, or d according to whether it showed (a) rapid progressive motility, (b) slow or sluggish progressive motility, (c) non-progressive motility or (d) immotility. The percentage of the progressive motility was done according to the formula a + b/a + b + c + d We did not use a computer assisted sperm analysis (CASA) system in our laboratory. At the same time presence of agglutination was observed and an estimation of the sperm concentration made. The volume of the semen samples was determined using a sterile disposable 5 ml pipette.
Depending on the estimated sperm concentration, a 1/10, 1/20, or 1/100 dilution of the semen sample was made with the use of a glass tuberculin syringe, instead of a white blood cell pipette (Menkveld et al., 1984), using HAM's F10 + 10% BSA medium. An improved, double-ruled Neubauer haemocytometer (FRG) was used for counting the spermatozoa. Two dilutions were made for every sample. The difference between the two dilutions for each sample was not >10% for low concentrations and not >20% for concentrations of >60x106/ml (Eliasson, 1971
).
The following procedures were used for the assessment of the morphological characteristics of the spermatozoa. The slides were thoroughly cleaned, washed in alcohol and dried before use. For a reliable and reproducible assessment, a thin and well-spread smear was made so that each spermatozoon could be clearly and individually visualized. The smears were air-dried and the following day fixed and stained according to the Papanicolaou method (Papanicolaou, 1942). The morphological classification used at the Tygerberg Hospital unit was based on a modification of the methods described by Eliasson (1971) and MacLeod (1962). This system takes the whole spermatozoon as well as the presence of germinal epithelial cells into consideration.
In this laboratory, a spermatozoon is considered normal when the head has a smooth, oval configuration with a well-defined acrosome comprising about 4070% of the sperm head. Also there must be no neck, mid-piece, or tail defect and no cytoplasmic droplets of more than one-half the size of the sperm head. In contrast with other authors (Papanicolaou, 1942; MacLeod, 1962
, 1965
; Eliasson, 1971
; WHO, 1980), borderline forms were considered abnormal (Kruger et al., 1986
; Menkveld et al., 1990
). At least 100, but preferably 200, spermatozoa with tails were classified into one of seven groups: normal (head and tail normal), normal head but with another abnormality present, large heads, small heads, tapering heads, duplicated heads or amorphous heads all with or without tail, neck or mid-piece defects. Tail, neck, and mid-piece defects, loose head, immature germinal cells and unknown cells were recorded separately and reported per 100 spermatozoa. The size of the spermatozoa was evaluated in five different areas on the slide to ensure a more randomized evaluation (WHO, 1999).
Statistical analysis
Basic descriptive statistics such as the mean, standard deviation and range were calculated for the two groups and compared using Fisher's t-test. Linear regression analysis was used to investigate the effect of age on the mean difference between the semen parameters of the two groups. This analysis led to the restriction of the diagnostic analysis to subgroups with adequate overlap in the age distribution of the male and female partners in the study. The diagnostic ability of the individual semen parameters to differentiate between the fertile or infertile status of the male was analysed using ROC curve analysis (Hanley et al., 1983). The sensitivity, specificity, positive and negative predictive value, positive and negative likelihood ratio and area under the curve of each semen parameter was estimated assuming a 15% prevalence of infertility (Ombelet et al., 1997a). Pairwise comparison of the area under the curve was done for the semen parameters of volume, concentration, motility, progressive motility and morphology.
Definitions used in this study
Sensitivity of a test is the percentage of individuals with infertility (disease) who are classified as having infertility. Specificity of a test is the percentage of individuals without the disease who are classified as not having the disease (infertility in this study). Positive predictive value is where the test is positive and the subject has the problem (subfertility). A negative predictive value is where the test is negative and the subject does not have the disease (subfertility), thus the subject is fertile. The last two definitions can also be applied to screen the general population with a given incidence of infertility (e.g., 15%) (Fisher et al., 1993).
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The statistics for the two subsets are given in Table I. The fertile group consisted of 61 subjects and the infertile group of 62 subjects. After remodelling, the mean age for infertile males was 31.3 years, for fertile males 29.9 (not significant) and for females 27.3 versus 26.3 (not significant).
|
The ROC analysis was done on the subset (Figure 1). The best performing semen parameters were progressive motility and morphology which had nearly identical predictive power as indicated by areas under the ROC curve (AUC), 0.707 and 0.697 respectively, and there was no significant difference between them, P = 0.972 (Table II
). With respect to the AUC, progressive motility differed significantly from motility (P = 0.002) and concentration (P = 0.011) as did morphology for the latter (P = 0.014).
|
|
Assuming a prevalence of 15% infertility in the population the positive and negative predictive values for the semen parameters are reported in Table III. The corresponding sensitivity and specificity at the reported cut-off point are also given to describe the performance of the parameters in the study population. The positive predictive value of the semen parameters is low overall, using the mean values of the infertile group (Table I
) as threshold values, i.e., progressive motility at 35 with a 33.6% chance of subfertility, morphology at 10% normal forms with a 29% chance and motility at 30 with a 28% chance. At 5% normal forms sperm morphology has a positive predictive value of 64.7% and this is similar to the 57.4% achieved by progressive motility at 14%.
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The threshold of sperm morphology assisting the clinician was based on research done on the IVF model (Coetzee et al., 1998). There were also studies dealing with strict criteria and IUI where it was shown that the threshold of 5% was of value as indicated by Lindheim (1996). Ombelet et al. (1997b) added a dimension to the findings of Lindheim by bringing into play the concentration retrieved. No pregnancy was reported in patients where less than 1x106 spermatozoa were retrieved after swim-up and with a morphology <5%. Toner et al. (1994) used a threshold of 14% with a significant lower pregnancy rate below that threshold. The 5% threshold was also shown to be of value in the IUI programme at Tygerberg Hospital in a recent publication (Montanaro-Gauci and Kruger, 2000
). In this study where all females were classified as normal, the pregnancy rate in the group 04% normal morphology was 2.63% per cycle compared to 24% in the group >14% normal morphology (P = 0.003).
It is clear that with the threshold indicated for strict criteria in IUI and IVF programmes, one of the problem areas is still the threshold for the in-vivo situation. The 14% threshold in the latest WHO manual was based on data from IVF studies (WHO, 1999). When using the principle of screening the population with the positive predictive value as indicator, it was clear from the data set that as far as sperm morphology is concerned the threshold is at the 5% level. With these data one can assume that 65% of patients below the threshold will be subfertile. On the other hand, the negative predictive value at this threshold was 89.4%, indicating that the probability of being fertile is above the 5% threshold. When using the same principle, the threshold for progressive motility was 14% or lower. For motility it was 30% or lower and for concentration 10x106/ml or lower (indicating the subfertile group below these thresholds) (Table III). The negative predictive values of the parameters are good and achieve 90% in most cases (Table III
). The sensitivity of the semen parameters at the reported thresholds is poor and indicates a large overlap in the distributions of these variables in the fertile and infertile groups.
Based on the current data obtained by ROC analysis, the best predictors of subfertility were the normal sperm morphology and progressive motility. By statistical comparison of these two parameters there was no significant difference between them (Table II). By using the ROC curve and the population under discussion in this article (fertile and subfertile), the best threshold for sperm morphology indicating a possible problem in this group was at the 12% threshold (Figure 1
) where the sensitivity was 69.1% and the specificity 66.7%. In the article by Ombelet et al. (1997a) they indicated a 10% threshold for sperm morphology using this method to evaluate their data. Barratt et al. (1995) also questioned the WHO (1992) 30% threshold and suggested a 10% threshold level for sperm morphology based on clinical considerations (Barratt et al., 1995
). The question is, which population do we want to identify? With respect to the general population's point of view (positive or negative predictive value), the 5% morphology threshold can be used with theoretically a 64.7% accuracy to identify a patient with subfertility (Table III
). Above this threshold the negative predictive value was 89.4%. Progressive motility at the 14% level is also of clinical help, but concentration and motility are not (Table III
).
Although there is scanty information concerning the geographical distribution of fertility and subfertility, it is interesting to note that with strict criteria applied in Turkey, Belgium and Singapore (Ombelet et al., 1997a; Chia et al., 1998
) the trend is very much the same in these studies. We also tried to establish an index to predict the fertilization potential of the semen sample with three basic sperm parameters, namely sperm concentration, motility and morphology, but we were not able to achieve this.
To conclude, the available data on the fertile and subfertile groups in the in-vivo situation is of value. As far as sperm morphology is concerned it correlates with the findings of Ombelet et al. (1997a) and one can use these thresholds at least to identify a `high risk' group with greater chance of having an infertility problem. One can also indirectly draw the line to IUI as well as IVF data, especially as far as sperm morphology is concerned, and use these thresholds at the 5% level practically to assist in clinical practice.
![]() |
Notes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Chia, S.-E., Tay, S.K. and Lim, S.T. (1998) What constitutes a normal seminal analysis? Semen parameters of 243 fertile men. Hum. Reprod., 13, 33943398.[Abstract]
Coetzee, K., Kruger, T.F. and Lombard, C.J. (1998) Predictive value of normal sperm morphology: a structured literature review. Hum. Reprod., 4, 7382.[Abstract]
Coetzee, K., De Villiers, A., Kruger, T.F. and Lombard, C.J. (1999) Clinical value of using an automated sperm morphology analyser (IVOS). Fertil. Steril., 71, 222225.[ISI][Medline]
Eliasson, R. (1971) Standards for the investigation of human semen. Andrologia, 3, 4964.
Fisher, L.D. and Van Belle, G. (1993) Biostatistic: Methodology for the Health Sciences. Wiley, New York, p. 207.
Hanley, J.A. and McNeil, B.J. (1983) A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology, 148, 839843.[Abstract]
Kruger, T.F., Menkveld, R., Stander, F.S.H. et al. (1986) Sperm morphologic features as a prognostic factor in in-vitro fertilization. Fertil. Steril., 46, 11181123.[ISI][Medline]
Lindheim, S., Barad, D., Zinger, M. et al. (1996) Abnormal sperm morphology is highly predictive of pregnancy outcome during controlled ovarian hyperstimulation and intrauterine insemination. J. Assist. Reprod. Genet., 13, 569572.[ISI][Medline]
MacLeod, J. (1962) A possible factor in the etiology of human male infertility: preliminary report. Fertil. Steril., 13, 2933.[ISI][Medline]
MacLeod, J. (1965) The semen examination. Clin. Obstet. Gynecol., 8, 115127.
Menkveld, R., Stander, F.S.H., Kotze, J.W. et al. (1990) The evaluation of morphological characteristics of human spermatozoa according to the stricter criteria. Hum. Reprod., 5, 586592.[Abstract]
Menkveld, R., van Zyl, J.A. and Kotze, T.J.v.W. (1984) A statistical comparison of three methods for counting human spematozoa. Andrologia, 16, 554.[ISI][Medline]
Montanaro-Gauci, M. and Kruger, T.F. (2000) Stepwise regression analysis on 495 cycles to study male and female factors impacting on pregnancy rates in an IUI programme. Andrologia, in press.
Ombelet, W., Bosmans, E., Janssen, M. et al. (1997a) Semen parameters in a fertile versus subfertile population: a need for change in the interpretation of semen testing. Hum. Reprod., 12, 987993.[ISI][Medline]
Ombelet, W., Vandeput, H., Van de Putte, G. et al. (1997b) Intrauterine insemination after ovarian stimulation with clomiphene citrate: predictive potential of inseminating motile count and sperm morphology. Hum. Reprod., 12, 14581463.[Abstract]
Papanicolaou, G.N. (1942) A new procedure for staining of vaginal smears. Science, 95, 438439.
Toner, J., Mossad, H., Grow, D. et al. (1994) Value of sperm morphology assessed by strict criteria for prediction of the outcome of artificial (intrauterine) insemination. Andrologia, 27, 143148.[ISI]
World Health Organization (1980) WHO Laboratory Manual for the Examination of Human Semen and Semencervical Mucus Interaction. Press Concern, Singapore, p. 16.
World Health Organization (1992) WHO Laboratory Manual for the Examination of Human Semen and Semencervical Mucus Interaction, 3rd edn. Cambridge University Press, Cambridge, p. 44.
World Health Organization (1999) WHO Laboratory Manual for the Examination of Human Semen and Semencervical Mucus Interaction, 4th edn. Cambridge University Press, Cambridge, p. 17.
Submitted on November 5, 1999; accepted on September 22, 2000.