A NEW CHAMBER FOR RAPID SPERM COUNT AND MOTILITY ESTIMATION

FERTILITY AND STERILITY Copyright ' 1978 The American Fertility Society Vol. 30, No.3, September 1978 Printed in U.S A. A NEW CHAMBER FOR RAPID SPERM COUNT AND MOTILITY ESTIMATION AMNON MAKLER, M.D. Fertility Clinic, Departments of Obstetrics and Gynecology "A" and "B," Rambam Medical Center, Technion, Faculty of Medicine, Haifa, Israel A new chamber for sperm count and motility estimation is described. This chamber, which is only 10 pm deep, enables free horizontal movement of spermatozoa in one focal plane and provides conditions for the examination of undiluted samples. Therefore, with the aid of this instrument it is possible to compare sperm motility in various samples from the same person or in different samples at different times. This can be done either by simple estimation or with any other method of motility evaluation chosen by the examiner. The sperm count can be made rapidly and directly from an undiluted, preheated sample by counting spermatozoa in the area of a grid located within the eyepiece; the count is expressed in millions per milliliter. Thirty-seven specimens were analyzed with this chamber. Statistical evaluation of the results revealed high precision, accuracy, and reliability of sperm counts when compared with the hemocytometric method. Better results were obtained when motility estimation was compared with the ordinary slide technique. Easy performance, rapid sperm counts, and improvement of motility estimation make this chamber a useful tool where sperm analysis is carried out. Fertil Steril 30:313, 1978 The most popular technique for sperm count involves the use of a hemocytometer. 1-5 The method consists of several steps: preparing 1:20 or 1:100 dilutions of the original sample, thoroughly mixing the diluent, inserting the diluted sample in a special way into the chamber, and allowing it to rest for about 20 minutes to permit the precipitation of spermatozoa to the bottom of the chamber before the counting itself is performed. These steps are necessary because the chamber, which is 100 JLm deep, does not permit undiluted semen to be examined. When such a sample is inserted into the chamber and viewed through the microscope, the spermatozoa are found to be too crowded, multilayered, cannot be seen in one focal plane, and cannot be counted. These several steps also increase inaccuracy, 6 are time-consuming, and discourage the untrained technician from using this method. The most popular method of spermatozoal motility evaluation is by microscopic examination Received February 14, 1978; revised March 23, 1978; accepted April 21, 1978. of a drop of fresh semen pressed between a slide and a cover slip. 1-3 The motility rate is estimated in one of several ways as chosen by the examiner: by estimating the percentage of motile spermatozoa, grading motility from + 1 to +4, or by more sophisticated methods of speed determination with a stopwatch 7 or with still cameras, 8 9 movie cameras, 10 or a closed-circuit TV system.11 There is one common point which devalues the results of all of these systems: since the cover slip is pressed with an uncontrolled pressure by the examiner, the thickness of the examined drop is not a standard one, since it can range from a few micrometers to several tenths of them. 12 As a result, the more the spermatozoa are pressed, the less freely they move, and vice versa. Moreover, if the drop of semen is too thick some spermatozoa will be out of focus and remain unnoticed, or may gain nonhorizontal movement. In a previous preliminary study 13 we were able to demonstrate how changes in the thickness of the examined drop can affect extensively the motility and the subsequent results. Therefore, comparing motil- 313

314 MAKLER ity rates of different samples when using the simple slide technique has limited value. We report here the invention of a new counting chamber which solves the above-mentioned problems of sperm count and motility evaluation. This chamber is only 10 p.m deep; this depth was chosen after preliminary experiments had shown that it fulfilled the following requirements: (1) All of the spermatozoa are located in one focal plane and all are seen, thus none is omitted by the observer. (2) The chamber provides for the free movement of the spermatozoa with minimal friction between surfaces of the chamber, so that spermatozoal speed is pradically unaffected. (3) The spermatozoa move horizontally with no vertical component that might affect the observer's interpretation when watching them from above. As a result, the semen may be analyzed without dilution, in one step and quickly. Since motility is established from a sample with standard depth, comparing various samples is possible with the highest accuracy. MATERIALS AND METHODS Basically, the new chamber* is constructed from two discs of high quality, transparent, optical flat glass; each disc is about 30 mm in diameter. The lower disc, which is 2 mm thick serves as a slide; the upper disc, which is 1 mm thick, is the cover glass. The tips of three pins located equidistant on the periphery of the lower glass are raised exactly 10 p.m above the plane, and when the cover glass is placed on those tips, a chamber of that thickness is thus obtained (Fig. 1). The accuracy of this space was checked interferometrically with a monochromatic laser beam and a distortion of less than 0.3 p.m was found. Technique of Semen Analysis. With the aid of a glass rod or a wooden stick, a drop of a well-mixed fresh seminal specimen is put.in the center ofthe lower glass and immediately covered. For observation, a microscope with a x20 objective and x 10 eyepiece is used. A grid calculated to cover an area of 0.033 sq mm is inserted into the eyepiece to aid counting and motility estimation. Motility is estimated immediately after the drop is placed in the chamber. Although one can use any method for estimating motility subjectively or obj.ectively, the grid can be used for this purpose in the following way: First the nonmotile spermatozoa within the area of the grid are counted, then the motile spermatozoa are *Pending patent no. 52276, manufactured by EL-OP, Rehovoth, Israel. September 1978 FIG. 1. Schematic side view of the main part of the instrume?t. An undilute~ drop of specimen is placed between two optical flat glass shdes separated exactly 10 /Lm by the tips of three pins (one pin is omitted from the figure). counted, and their quality of movement is determined either by simple grading or by measuring the time elapsed for each spermatozoon to cross the area of the grid (about 150 x 200 p.m). After repeating this process in several fields of the examined drop to make it sufficient for statistical purposes, the percentage ofmotile spermatozoa is calculated and the quality of motility is also determined. These estimations can be repeated in several drops of the same specimen. Sperm Count. When the seminal fluid is not too concentrated and the spermatozoa are not too active, the count of live spermatozoa can be easily done directly in a fresh specimen. All spermatozoa whose heads are included within the borders of the grid (Fig. 2) are counted in the same way as blood cells are counted in a hemocytometer. This procedure is repeated in two more areas of.the drop by shifting the instrument on the stage of the microscope. The sum in the areas. counted represents the number of spermatozoa in onemillionth milliliter according to the following calculations: 0.01 mm x 0.033 sq mm x 3 = 0.001 cu mm = O.OOOOOlml. Therefore, adding six zeros to the total number in three different grid areas (e.g., 22 + 26 + 24 =.72) gives their concentration (72,000,000) in 1 ml. When the spermatozoa are very active and concentrated, counting of live specimens is more difficult. In this case immobilization of the spermatozoa can be achieved by transferring part of the specimen into another test tube and inserting it into 50 to 60 C heated water for about 5 minutes (a cup containing a mixture of two parts boiling water and one part tap water is sufficient for this purpose). It is unwise to add spermicidal material such as formaldehyde or any detergent since it induces clumping, and errors result. Increasing the accuracy of the count can be achieved by counting more grid areas in the same drop or in several drops from the same specimen. In this

Vol. 30, No.3 A NEW CHAMBER FOR RAPID SPERM COUNT AND MOTILITY ESTIMATION 315 FIG. 2. Photomicrograph of an undiluted semen specimen located in the chamber, while a grid is placed within the eyepiece. The area covered by the grid is exactly 0.033 sq mm when a x20 objective and a x10 eyepiece are used. By multiplying by 3 the-number of spermatozoa in the grid, or by counting them in three different grid areas, the concentration.in millions per milliliter is obtained. All spermatozoa are in one focal plane; none is blurred or omitted by the observer even in this undiluted specimen. case, calculation of the concentration is done according to the formula C = [(T x 3)/G] x 106, where T is the total number of spermatozoa within all grid a:reas and G is the number of grid areas explored in one or all drops examined. For example: 275 spermatozoa are counted in four grid areas in each of five drops from the same specimen. Then C = [(275 x 3)/20] x 10 6 = 41,250,000/ml. Difficulties. Real difficulties are extremely rare. Since samples need not be aspirated or inserted into capillary_ spaces, drops from very viscid samples can also be applied on the lower part of the chamber, being pressed uniformly into the 10~/Lm space by the cover glass. Specimens that remain persistently unliquefied induce problems which are common to any technique that requires sampling, such as the ordinary hemocytometric method. In such cases liquefying agents should be added to the specimen before counting is done. Difficulties in mixing spermatozoa homogenously in a viscid sample may lead to an uneven distribution of spermatozoa within the chamber. In this case more grid areas from the same drop or from several drops should be explored and more spermatozoa should be counted. When spermagglutination occurs, counting is done in areas devoid of agglutinated spermatozoa, with a special note about the. existence of this phenomenon and its estimated intensity. Very oligospermic or hyperspermic specimens do not present any difficulties. In the first case, one may find many "empty" grid areas among others containing one or two spermatozoa per grid. If 20 to 30 grid areas are selected randomly for exploration and those "empty" grids are intable 1. Frequency of Distribution in Fifteen Counts in Five Drops of a Normospermic Specimen Drop no. 2 3 4 5 millions spermatozoa/ml Mean 107 119 126 124 126 118 138 144 136 131 133 129 127 127 134 117.3 122.6 139.3 131.0 129.3 Mean sperm count = 127 million; SD = 8.7; CV = 6.69.

316 MAKLER September 1978 TABLE 2. Means± Standard Deviation often to Fifteen Counts in Each of Thirty-Seven Specimens Divided into Five Subgroups According to Concentration at Bottom Range of counts ~10 1~20 3.0± 2.2 13.6± 3.5 5.4± 2.1 15.8± 2.5 5.7± 2.5 16.1 ± 4.1 8.3± 2.5 19.5± 3.6 Counted grids 15 15 Mean± SD 6.3± 2.-3 16.3± 3.4 cv 46.3 21.6 a CV = (SD X 100)/x. eluded in the above formula, accurate results are obtained even in cases where concentration is below 100,000 spermatozoa/mi. It never occurred to us that any hyperspermic specimen could not be counted in its own undiluted medium. There is enough space for all sperm heads to be seen clearly within the area of the grid, with no overlapping heads. Statistical Evaluation. For this purpose, 37 specimens from normal, oligospermic, and asthenospermic males were examined. Sperm count and motility were evaluated two or three times in five drops of every specimen as described above. The results were statistically analyzed for accuracy and precision. In 5 of 37 cases, the same specimens were examined by the same technician in three different instruments of 10 p,m, and the results of the semen analyses were compared for reproducibility. In 18 cases the same specimens were also examined by a qualified technician using the ordinary method of semen analysis: in every sample the percentage of motility was estimated from 10 view fields of a regular slide according to the technique described by Eliason. 12 A count was done in four different diluted drops (1:20 or 1:100), using the Neubauer hemocytometer where spermatozoa were counted in 25 squares of0.04 sq mm, which is roughly equivalent to the area covered in our counts. All results were compared statistically. RESULTS Table 1 describes the frequency of distribution of 15 counts performed in five different drops in 2~50 5~100 10~200 millionslml 22.0± 3.5 50.6± 7.4 105.6± 7.7 22.0± 3.6 50.6± 3.1 127.0± 8.6 24.1 ± 5.6 53.0± 6.2 170.8± 3.9 25.2± 3.5 53.1± 2.2 183.5± 2.6 25.3± 4.9 67.1 ± 10.3 186.7 ± 10.6 33.6± 3.2 73.3± 5.0 189.0± 22.4 34.0± 5.8 77.7± 5.2 190.2± 14.5 34.4± 7.0 194.3± 12.0 34.8± 3.4 195.0± 10.3 39.8± 4.7 39.9 ± 7.1 42.0± 5.1 47.6± 7.8 10 10 10 32.6± 5.0 60.8± 5.7 171.2 ± 10.3 15.4 9.3 6.1 one of the normospermic specimens, the average of the counts of each drop, and the average of all 15 counts as well as the standard deviation and coefficient of variance. From Table 1 it can be seen that the distribution of the counts in this concentration is very small. Table 2 shows the average and standard deviation of 10 to 15 counts carried out for each of the 37 specimens divided into subgroups according to their concentration. For each subgroup, the average, standard deviation (SD), and coefficient of variance (CV) are given. It can be seen that, whereas the values of SD are increasing with the concentration, the values of CV are decreasing. This means that, practically, fewer counts are necessary in samples of 50 x 10 6 or more. Oligospermic samples need more counts to increase accuracy, although results such as 5 ± 3 million/ ml are clinically adequate. Table 3 compares results of 15 counts from 5 various specimens carried out in three different 10-p,m instruments. The difference between instruments is very small and reliability is fairly high. Figure 3 shows the correlation of 18 sperm counts each performed both with the 10-p,m TABLE 3. Results of Fifteen Counts of Five Specimens Carried Out in Three Different Chamber Instruments Chamber Mean (X) Mean Mean II III r,;- x) r,; ->Mx% 25 24 22 24 1 4.1 34 34 42 36.6 3.5 9.6 53 47 49 49.6 2.2 4.4 184 195 190 189.6 3.8 2.0 195 186 194 191.6 3.8 2.0

Vol. 30, No. 3 A NEW CHAMBER FOR RAPID SPERM COUNT AND MOTILITY ESTIMATION 317 10 6 1601r-------------------, 140 120 100 N=18 Y=1.4 0.92x Fra. 3. Results of18 sperm counts obtained with the aid of a 10-1-'m chamber plotted against results of counts performed with an ordinary hemocytometer. Each point indicates the results of two counts for each specimen as described in the text. chamber and with an ordinary 100-11-m hemocytometer. The average of differences between the instruments did not exceed 5.4% and was almost independent of sperm concentration except in very low concentrations. The results are highly correlated, and the reliability of the 10-11-m chamber is evident. Table 4 reveals results of motility estimation in 50 view fields of 3 different specimens each examined by the ordinary slide method and also with the 10-J.Lm chamber. The average motility and SD of each specimen using both techniques are compared. Although the results were closely related, those of the slide method were more widely distributed, almost twice as much as those with the 10-J.Lm chamber technique. DISCUSSION The above results show that sperm counts using the new 10-J.Lm-chamber are accurate, precise, and reliable even though the technique is very TABLE 4. Percentage of Motility Estimated by Observing Fifty Fields in Each of Three Different Specimens in Ordinary Slide and 10-~ Chamber Estimated 10-JLm Chamber Estimated Ordinary slide motility SD cv motility SD cv 18% 4.8 27.6% 14% 5.7 46.4% 40% 5.1 13.0% 36% 8.18 24.4% 49% 5.4 11.0% 47% 10.0 26.1% simple and rapid. Moreover, motility estimation is more accurate than estimation done with the ordinary slide method. The constant thickness of the examined drop excludes external physical factors which may induce differences between samples of the same specimen or between various specimens that need to be compared, whatever the method of estimation chosen by the observer. The fact that the chamber is shallow enough to enable undiluted samples to be examined is very important for estimation of sperm motility in its own natural medium, unless dilution is purposely desired. Therefore, this instrument can be an important tool for every clinical or research study, when the effect of a certain factor on sperm motility needs an accurate evaluation or comparison of motility rates. When one compares the technique for use of this chamber with that for use of a hemocytometer the following advantages can be mentioned: 1. While both sperm count and motility estimation can be done in the same 10-11-m chamber, only the sperm count can be done with the hemocytometer. 2. As sperm dilution is not necessary, no accessories such as pipettes, solutions, and facilities for pipette flushing and drying are needed. Nor is special training for these procedures necessary. 3. As the chamber is very thin and all spermatozoa can be seen immediately and in one focal plane, there is no need to wait for sperm to be precipitated on the bottom of the chamber, a process which takes about 20 minutes when using the hemocytometer. 4. The grid located within the eyepiece allows one to count in the same drop as many grid areas as one wishes. This can be accomplished by shifting the instrument on the stage of the microscope from one area to another. Therefore, artefacts, bubbles, and pieces of dirt or clumped spermatozoa can be ignored and do not interfere with counts as might occur with the fixed scale in the ordinary hemocytometer. All of these advantages make this instrument a helpful tool in every institute, laboratory, or office where semen analysis is done. Acknowledgments. The author wishes to thank Professor Lipson from the Haifa Technion for his interferometric analysis of the instruments, Mr. Amnon Levy for his useful advice, Mr. N. Kipperman for his skillful assistance in constructing the model of the chamber, Mrs. Oma Heidenfeld for preparing the results, and my wife, Y osepha, for her statistical analysis of the results.

318 MAKLER REFERENCES 1. Freund M, Peterson RN: In Human Semen and Fertility Regulation in Men, Edited by ESE Hafez. St. Louis, CV Mosby Co, 1976, p 344 2. Eliasson R: Standards for investigation of human semen. Andrologia 3:49, 1971 3. Amelar RD: Infertility in Men. Philadelphia, FH Davis Co, 1966, p 31 4. Schirren C: Practical Andrology. Berlin, Verlag Bruder, 1972, p 21 5. Bauer JD, Ackerman PG, Toro G: Clinical Laboratory Methods. St. Louis, CV Mosby Co, 1974, p 73 6. Freund M, Carol B: Factors affecting hemocytometer counts of sperm concentration in human semen. J Reprod Fertil 8:149, 1964 7. Harvey C: The speed of human spermatozoa and the effect on it of various diluents. J Reprod Fertil1:84, 1 }60 September 1978 8. Rothschild L: The movement of spermatozoa. In Mammalian Germ Cells, Edited by GEW Wolstenholme. Boston, Little, Brown and Co., 1953, p 58 9. Janick J, MacLeod J: Measurement of human spermatozoa motility. Fertil Steril 21:140, 1970 10. Rikmenspoel R, Harren GV: A microscopic film of the movement of bull sperm cells in dark-field illumination. Experimenter Xll:160, 1956 11. Jecht EW, Russo JJ: A system for the quantitative analysis ofhuman sperm motility. Andrologia 5:215, 1973 12. Eliasson R: Analysis of semen. In Progress in Infertility, Edited by SJ Behrman, RW Kistner. Boston, Little, Brown and Co, 1975, p 696 13. Makler A: The thickness of the microscopically examined sample and its relationship to motility estimation. Int J Androl 1:213, 1978