Frank Schoonjans

MedCalc: a new computer program for medical statistics

In recent years, the use and abuse of statistics in the medical literature has extensively been reviewed. Amongst others, the importance of the P-value has been challenged and the use of misleading graphics, including 3-dimensional displays, has been criticized. The ease of access to more complex statistical procedures, since the introduction of several statistical software packages for personal computers, has been identified as one of the factors involved in the misuse of statistics. Therefore, we have developed a new computer program that includes those statistical procedures commonly encountered in the medical literature and in statistical textbooks for medical researchers. More complex statistical analyses are not implemented in the software. If researchers with limited statistical training require more sophisticated statistical analyses, they should refer to a statistician, not to a more complete statistical software package.

Deterioration of sperm quality in young healthy Belgian men

We have retrospectively analysed the sperm characteristics of 416 consecutive healthy young men who presented themselves in the past 19 years as candidate sperm donors. Ejaculate volume increased slightly (P = 0.067), and average sperm concentration decreased (P = 0.035) by 12.4 x 10(6)/ml over the observation period, so that sperm count per ejaculate remained unchanged (P = 0.91). In contrast, sperm morphology (r = - 0.23, P < 0.0001), rapid progressive motility (r = - 0.42, P < 0.0001) and total motility (r = - 0.33, P < 0.0001) presented an important and time-related decrease. When a quadratic model was used rather than a linear one to analyse the data on rapid progressive motility, there appeared to have been no further decline since 1990. The average proportion of spermatozoa with normal morphology decreased from 39.2% in the period 1977-1980 to 26.6% in 1990-1995 (P < 0.0001), and the mean percentage of spermatozoa with rapid progressive motility decreased from 52.7 to 31.7% (P < 0.0001). The percentage of candidate donors with sperm characteristics below the 5th percentile cut-off value of a normal fertile population increased from 13 to 54% during the observation period (P < 0.0001). Since the technique of semen analysis has remained essentially unchanged in-so-far as has been practically possible, as has the method of recruitment of candidate sperm donors, the observed deterioration of sperm characteristics is considered to reflect degeneration of sperm production among men aged between 20 and 40 years.

Cytokines in semen of normal men and of patients with andrological diseases

PROBLEM: The potential value of assessment of cytokine concentrations for the diagnosis of certain pathological conditions of male reproduction has not been fully evaluated.

The effective cumulative pregnancy rate of different modes of treatment of male infertility

Summary The clinical efficacy of conventional and advanced methods of treatment was assessed in 814 couples with infertility due to a male factor. The monthly and effective cumulative rate of ongoing or term pregnancies was calculated during 4712 couple‐months. Treatment of varicocele by transcatheter embolization, resulting in 3.9% pregnancies per cycle and an effective cumulative pregnancy rate of 41% after 1 year, is more effective than counselling and timed intercourse (9% pregnancies after 12 months). Intrauterine insemination (IUI) of washed spermatozoa produced 17% pregnancies in the initial 4 months, but the success rate of the subsequent cycles (1.7% per cycle) was not different from that of the controls.

Placebo-controlled trial of high-dose Mesterolone treatment of idiopathic male infertility

The possible effect of Mesterolone (Schering N.V., Brussels, Belgium) (l α -methyl-5- α -androstane-17 β -ol-3-one) on semen quality and fertility of men with idiopathic oligoasthenospermia and/or teratozoospermia has been evaluated in a double-blind trial. The study included 52 patients who were treated during 12months with either 150mg/d of Mesterolone or placebo. The overall pregnancy rate was similar in the Mesterolone-treated cases (26%) and in the placebo control cases (48%), although a significant increase in motility and in the proportion of spermatozoa with normal morphology was recorded in the Mesterolone-treated cases. Because similar semen improvement also occurred in the placebo controls, our findings cast doubt on the possible usefulness of high-dose Mesterolone treatment of idiopathic male infertility.

Methodological aspects of sperm morphology evaluation: comparison between strict and liberal criteria*

OBJECTIVE To compare the methodological accuracy of different sperm morphology criteria. SETTING A multicenter study including 10 laboratories with high expertise in semen analysis. PATIENTS Semen preparations of subfertile men with a variable degree of teratozoospermia and of fertile semen donors. INTERVENTIONS Detailed assessment of sperm morphology on 10 air-dried semen smears, of which 3 originated from the same ejaculate. RESULTS The average coefficient of variation calculated on the three smears of the same ejaculate was higher when strict criteria for morphological normality were used and borderline cells were classified as abnormal than when criteria of abnormality were applied and borderline cells were considered normal. The correspondence between individual centers mutually and of each center with the average result of all centers was better when the latter approach was taken. The performance of two computer-assisted systems was intermediate between that of the two approaches, whereas one system gave unreliable results. CONCLUSIONS Strict criteria for normality of sperm morphology, with borderline cells considered abnormal, gives results that are less reproducible and less accurate than the approach that classifies sperm as abnormal, with borderline cells considered normal.

Accuracy of sperm characteristics in predicting the in vitro fertilizing capacity of semen

Based on the results of in vitro fertilization (IVF) in 56 couples, the power was assessed of traditional sperm characteristics of native semen to discriminate between in vitro fertile and in vitro infertile semen. The number per ejaculate of spermatozoa with regular oval heads was the best discriminant, followed by the concentration of progressively motile spermatozoa. This contrasts with the in vivo fertilizing capacity, which depends mostly on the proportion and concentration of spermatozoa with rapid linear progression. The lower limit of sperm characteristics was assessed as the fifth percentile of in vitro fertile semen and was compared with the lower limit of semen of fertile men and of subfertile men who achieved spontaneous or treatment-related conception in vivo. It appeared that the semen quality needed for in vitro fertilization is inferior to that of fertile men but not remarkably different from that of subfertile men who achieved spontaneous conception during 1-year follow-up after consultation. If conventional methods for semen preparation are used, there seems to be no major advantage in favor of IVF for the treatment of male infertility due to sperm deficiency. An increased success rate may, however, be attained, thanks to improved techniques of semen collection, semen preparation, and oocyte insemination.

Evaluation of the hypo‐osmotic swelling test in relation with advanced methods of semen analysis

Summary. The hypo‐osmotic swelling test was claimed to assess an independent functional characteristic of human spermatozoa bearing relevance to their fertilizing capacity. To test this claim, we have studied the relationship between the result of the hypo‐osmotic swelling test with that of conventional semen analysis and sperm motility patterns, the semen content of adenosine triphosphate, the staining pattern to acidified aniline blue, and the zona‐free hamster oocyte test. The result of the HOS test is significantly correlated with all sperm characteristics except for the aniline blue stainability and the hamster oocyte test. The capacity of spermatozoa to react in a hypo‐osmotic environment expresses the same functional information as the viability test using eosine staining.

Declining sperm quality in European men

In the early 1980s, we tried to determine the limits of ‘normal semen quality’. In order to do so, we recruited a number of men who had impregnated their partner within the first 12 months of ‘exposure to the risk of pregnancy’ (Rowe et al., 1993). The fifth percentile of the sperm characteristics of these men was accepted as the lower limit of normality (Fig. 1). The limits suggested by WHO (World Health Organization, 1987), and below

Estimation of population percentiles.

Percentiles play an important part in descriptive statistics of continuous data, and their use is recommended for reference interval estimation.1 We have selected various methods for the calculation of percentiles based on recommendations in the literature or use in popular software, and evaluated the accuracy of the percentile calculated in the sample as an estimate of the true population percentile2 using Monte-Carlo techniques. All selected methods calculate a rank or an index that points to a number in the sorted array of sample data, and linear interpolation is applied when the index does not correspond to an integer value. One method (method A1,3) calculates a rank or an index p(n+1) with p representing the centile (which is the percentile divided by 100) and n the sample size. Method B4 calculates an index 0.5+pn. Method C5 (commonly used in spreadsheets) uses p(n-1)+1 and method D6 uses p(n+1/3)+1/3. Details of the use of these 4 methods are given in the eAppendix (http://links.lww.com/EDE/A488). Experimental population data were obtained using a normal distribution pseudo-random number generator, programmed to generate a data set of 106 numbers with mean 0 and standard deviation 1. From our population data, 6 sets of 100 000 random samples each were drawn using a pseudo-random number generator with uniform distribution. Each of these sets consisted of 100 000 random samples with sample size 20, 120, 500 and 1000. The average of the 5th and 95th percentiles obtained with the 4 methods in these sample sets were calculated and compared with the population values. For each sample, the relative difference with the population values was expressed as a percentage, and the mean and standard deviation of these percentages were calculated. Next, the population data were transformed exponentially (base 10) to obtain a log-normal distribution and the experiments as described above were repeated. The results for 95th percentile in the normally distributed data are represented in table 1 (more comprehensive tables with figures are available in the eAppendix, http://links.lww.com/EDE/A488). The results for the 5th percentile were symmetrical to the results for the 95th percentile and are not shown. Method B presents the highest accuracy, followed by method D, A and C. Table 1 Accuracy of percentiles calculated in samples with various sizes from normally and log-normally distributed population data. Results are presented as average percentile, the average of the relative differences (%) and their standard deviation (SD). The results for 5th and 95th percentile in the log-normally distributed data are represented in the Table. For the 5th percentile, method A has a higher accuracy than the methods D, B and C, especially in small sample sizes, whereas for the 95th percentile method C presents the highest accuracy, followed by method B, D and A. We find that, for the calculation of percentiles, it may still be advantageous to transform log-normally distributed data. For example, the 95th percentile in the log-normal data should be about 44.16 (= 101.65). With method B and n = 20 we find an average of 114.69. But if we first log-transform the data we find, on average, 1.64, which back-transforms to 101.64 or 43.65, which is much closer to the true population value of 44.16. The effect of the log-transformation may be explained by the fact that linear interpolation is applied in the calculations of percentiles, and the transformation changes the distribution model within the interpolated interval. We conclude that method B is the preferred method in general for continuous data, taking into account the recommendation to transform the data to a normal distribution if necessary. Finally the large standard deviations of the observed differences illustrate the large statistical uncertainty associated with the estimated percentiles in small sample sizes. Therefore we stress the importance of reporting percentiles with their 95% confidence interval.