F'ERTIIJTY AND STERILITY Copyright c 1982 The American Fertility Society Vol. 38, No.2, August 1982 Printed in U.SA. Scoring sperm morphology using the scanning electron microscope John Liakatas, M.D.* Anthony E. Williams, Ph.D.t Timothy B. Hargreave, F.R.C.S.:J: University of Edinburgh, Western General Hospital, Edinburgh, Scotland The scanning electron microscope (SEM) was used in an attempt to accumtely score morphologic differences in sperm, but early attempts were confounded by tangled prepamtions. In order to obtain an evenly spread sample, a new technique was evolved where a suspension of sperm cells was filtered and the whole filter prepared for the microscope. Sperm samples were then examined from men with and without obvious clinical and Doppler evidence of a varicocele. All samples were coded and scored; statistically significant differences in sperm morphology between the two groups of men were found. Our conclusions are that the SEM is worthy of further evaluation as a tool in the accumte scoring of sperm morphology, and the presence of a varicocele does appear to have an effect on sperm structure. Fertil Steril 38:227, 1982 Morphologic abnormalities of sperm have been reported in men with infertility secondary to varicocele and in normal men after viral illness and other insults. A predominance of tapering spermatozoal forms are said to indicate testicular stress following any of these insults. 1 The resolving power of the light microscope is such that even skilled observers may disagree about morphologic patterns. The purpose of this study has been to evaluate the use of the scanning electron microscope (SEM) in scoring sperm morphology. A previous study2 using SEM has described the detailed morphology of normal spermatozoa in man, and changes associated with disease were Received December 29, 1981; revised and accepted April 16; 1982. *The Infertility Clinic, University of Edinburgh, Department of SurgerylUrology, Western General Hospital, Edinburgh, Scotland. tteaching and Research Centre;' University of Edinburgh, Western General Hospital, Edinburgh, Scotland. treprint requests: Mr. T. B. Hargreave, The Infertility Clinic, University of Edinburgh, Department of Surgery! Urology, Western General Hospital, Edinburgh, EH42XU, Scotland. also reported. However, we know of no previous study that has attempted to score these changes as a measure of the disease state. MATERIALS AND MEmODS Specimens from 20 men attending a general infertility clinic were examined. These samples were chosen from men with and without evidence of varicocele as judged by clinical examination and Doppler analysis. This selection was made in order to obtain samples that were likely to have widely differing morphologic characteristics. The samples were coded, and at the time of scoring the observer did not know the origin of individual samples. Clinical examination was carried out independently by two doctors, and a diagnosis was subsequently agreed upon. Doppler analysis was carried out with the patient in the erect position, and blood flow was graded in three categories: grade 1, Val salva-induced Doppler activity only; grade 2, intermittent activity; and grade 3, continuous activity. The sperm density was measured with the use of a Neubauer Chamber (Hawksley, London, En- Vol. 38, No.2, August 1982 Liakatas et ai. Scoring morphology using scanning microscope 227
ti QI) Table 1. Frequency of Occurrence of Forms of the Head, Midpiece, and Tail of Human Spermatozoa in Relation to the Presence or Absence of Clinically Detectable Varicoceles t:: = I'l" Sperm morphologic types observed in 100 sperm per patient Head Midpiece Tail Total til Patient Sperm count abnormality (II... no. x lo&/ml Abnormalities Normal Abnormalities Normal Abnormalities score Normal oval Tapering Amorphous Other Abnormal Blebs Short Coiled Bent No varicocele ;:I. 1 3.5" 65 13 0 22 73 15 12 90 5 4 1 72 2 5.5" 56 17 1 26 80 6 14 85 0 11 4 79 3 32 58 20 4 18 70 13 17 82 0 14 4 90 c 4 52 81 7 2 10 70 12 18 89 0 3 8 60 'ti 5 64 69 9 15 7 82 13 5 93 1 2 4 56 [ 6 69 59 13 11 17 74 16 10 95 1 1 3 72 '<!': Mean 37.7 ± 28.9. '" Total 388 79 33 100 449 75 76 534 7 35 24 429 '" (') Mean (%) 64.7 13.2 5.5 16.7 74.8 12.5 12.7 89 1.2 5.8 4 71.5 ;:! ;:!. Small varicocele 7 4 45 15 11 29 73 16 11 85 4 0 11 97 8 7 48 27 11 14 68 20 12 81 1 0 18 103 1';' 9 60 40 23 0 37 74 10 16 88 3 0 9 98 g.g Total 133 65 22 80 215 46 39 254 8 0 38 298 '" Mean (%) 44.3 21.7 7.3 26.7 71.7 15.3 13 84.7 2.7 0 12.7 99.3 (') Mean 23.7 ± 31.50 Large varicocele 10 1.5 35 57 5 3 62 10 28 84 5 4 7 119 11 10" 42 24 23 11 76 16 8 91 1 2 6 91 12 10" 41 36 15 8 82 9 9 85 4 4 7 92 13 14 43 32 3 22 66 16 18 90 1 1 8 101 14 16 34 46 12 8 61 20 19 86 0 0 14 119 15 26 56 17 6 21 76 11 13 91 0 0 9 77 16 32 24 58 14 4 68 9 23 80 7 6 7 128 17 35 19 68 7 6 77 8 15 75 4 13 8 129 18 55 35 27 10 28 62 16 22 88 0 3 9 115 19 56 41 22 14 23 60 19 21 90 2 0 8 109 20 65 37 29 4 30 66 15 19 91 0 2.7 106 Mean 29.1 ± 21.51 Total 407 416 113 164 756 149 195 951 24 35 90 1186 Q" ;:! R. t'jj.., :::.:. Mean (%) 37.0 37.8 10.3 14.9 69.6 13.6 17.7 86.5 2.2 3.2 8.2 107.8 "Patient's wife became pregnant.
RESULTS Figure 1 Representative view to show quality of preparation. The spennatozoa are evenly spread onto the filter, thus allowing scoring to be done. (Magnification x 2100.) gland), and in the light of this figure an aliquot containing 5 x 106 spermatozoa was prepared. This aliquot was diluted with 0.5 ml phosphatebuffered saline (ph 7.2). Ten milliliters of 2.5% glutaraldehyde in sucrose-cacodylate buffer was added slowly with continuous gentle agitation, and the suspension in fixative was left at 40 C overnight. The spermatozoa were gently resuspended, and the suspension was filtered through a Nucleopore membrane filter (0.45-j.Lm pore size; 35-mm effective filtering diameter; Sterilin Limited, Teddington, Middlesex, England). This produced an even distribution of spermatozoa at a density of approximately 1 sperm/100 j.lm. 2 The fixative was removed by suction through this filter, and the collected sperm were washed with cacodylate-buffered sucrose, care being taken to ensure that the filter did not dry out. The filter was removed from the holder and processed for SEM by dehydration through graded ethanols, critical-point-dried from carbon dioxide, and sputter-coated with gold (- 20 j.lm). (Critical-point dryer and sputter coater from Polaron Equipment Ltd., Watford, Hertfordshire, England.) All spec imens were examined in ISI-60 SEM (International Scientific Instruments Incorporated, Newmarket, Suffolk, England). One hundred sperm were scored from each specimen according to morpnologic features of the head, midpiece, and tail (Table 1). Initially, scoring was done from electronmicrographs, but with experience, time could be saved by scoring directly from the SEM display screen. Vol. 38, No.2, August 1982 The quality of specimens resulting from the membrane filter technique allowed accurate repeated morphologic assessment. In order not to prejudge the observations, the term "normal" was not used during assessment for any particular form. The most frequent combination of sperm features in patients with or without varicocele was a smooth, oval head with a well-developed acrosome tapering into a well-defined short, smooth, cylindrical midpiece, followed by a long (- 50 j.lm), smooth, fairly straight tail of even thickness (Fig. 1). For convenience of description in this paper, these forms of head, midpiece, and tail have been termed "normal," and any variations from these are referred to as "abnormal." Variations were encountered in the head, midpiece, and/or tail, and Figures 2 to 6 show some of the forms: tapering and amorphous heads, head bent at an angle to midpiece, narrowing and blebbing of the midpiece (cytoplasmic droplets), and twin tails. The results (Table 1) show the frequency of the various forms of the sperm of patients with varicoceles, compared with those without. In this small sample, there was no relationship between the sperm density and frequency of morphologic abnormality. There was, however, a significant increase (by X2 test) in abnormal morphologic forms in those patients with a varicocele. This was particularly apparent in relation to changes in head shape, notably an increase in tapering form, but there was an overall increase in other abnormalities of the head, midpiece, and tail in Figure 2 Spenn with a nonnal head and midpiece. The smooth bend in the tail is almost certainly artifactual. (Magnification x 13,000.) Liakatas et al. Scoring morphology using scanning microscope 229
Figure 3 Tapering form. (Magnification x 4000.) Figure 5 Cytoplasmic droplet. (Magnification x 14,000.) the patients with varicoceles. There were significantly more tapered heads in patients with large, compared with small, varicoceles. Only one sperm with a double tail was noted in all the preparations. The overall ((abnormality" of a sample was assessed as the total number of abnormalities (head, midpiece, and tail) seen in 100 spermatozoa. Thus a sperm with a tapering head, cytoplasmic droplets, and a coiled tail had an abnormality score of 3, whereas a sperm normal except for a coiled tail scored only 1. In Table 1 the abnormality score is related to the presence or absence of clinically detectable varicocele, and in Table 2, to Doppler findings. In both cases there was a significant increase in the abnormality score associated with the presence of a varicocele or the presence of Doppler activity. Figure 4 Amorphous head. (Magnification x 13,600.) Figure 6 Double tail. (Magnification x 2750.) 230 Liakatas et al. DISCUSSION In spite of modern staining techniques and improved optics, the resolution of the best light microscope is such that it is difficult for one to examine abnormalities in detail, and sperm samples are often graded according to the predominant abnormality. Even skilled observers may disagree about morphologic features seen with the light microscope, and thus one observer must be used to make scientific comparisons: The method of preparation of samples described above produced samples that could be scored directly from Scoring morphology using scanning microscope Fertility and Sterility
Table 2. Correlation of Sperm Morphologic Abnormality Score with Doppler Grade No varicocele Patients with varicocele Doppler grade 1 2 3 Abnormality score 72 97 91 92 79 103 77 101 90 98 128 119 60 119 129 115 56 106 109 72 Mean abnor- 71.5 104.3 106.2 107.2 mality score the screen or photographed, allowing two or more observers to confirm observations. In cases of doubt the SEM can be zoomed to magnifications of between 10 and 50 times those obtained with the best light microscope. With practice, each sample could be scored in 60 to 90 minutes. For the future, this method lends itself to computerized image processing, which could greatly increase the speed and objectivity of this procedure. This model of patients with and without varicocele was used for an evaluation of the preparation method. It was notable that an increasing number of abnormal sperm were seen in patients with varicocele, compared with those without, and that this increased abnormality particularly reflected an increase in the number of tapered forms. There was a trend suggesting that patients with a large varicocele on clinical examination had a greater abnormality score than patients with small varicocele, but Doppler analysis of the severity of reflux showed no obvious correlation, because patients with minimal reflux had the same abnormality score as patients with continuous reflux. Varicocele ligation has been reported to improve semen quality in 81% of patients and to result in a pregnancy rate of 48%,3 but recently this accepted view of the efficacy of varicocele ligation has been challenged in a controlled trial where no benefit was seen in those patients who received operation, compared with those who did not 4 ; and it has also been suggested that efforts should be concentrated on treating the female partner, since varicocele ligation has little effect. 5 It is pertinent to note the case described in 1952 by Mr. Tulloch from this department,6 where spermatogenesis and fertility were restored in a man with azoospermia and bilateral varicoceles following bilateral varicocele ligation. In our present study, the increase in abnormal morphologic forms supports previous work l showing that unilateral varicocele has an effect on sperm structure, and this work thus supports the conventional view that varicoceles are likely to interfere with fertility. It was not, however, our intention to resolve the current dispute over the relevance of varicocele to fertility, but rather, the model was chosen to evaluate this new method of sperm scoring. In routine infertility practice SEM morphologic assessment of sperm is unlikely to add greatly to diagnosis or choice of therapy, and it is not suggested that this accurate but time-consuming technique should be used routinely. The potential may lie in the assessment of sperm samples from workers in hazardous environments where the early signs of hazard may be an increase in abnormal morphologic forms and where objective detection of abnormalities could be of importance. There is increasing realization that certain occupations are associated with damaged spermatogenesis, e.g., lead workers,7 pesticide workers,s and those handling cadmium. 9 Further work is necessary to evaluate this technique including the accurate assessment of a control group. Recent advances in back-scattered electron imaging in the SEM lo allow visualization of the internal contents of spermatozoa as well as the surface forms, and it may be possible to adapt this scoring to internal abnormalities. Acknowledgments. We wish to thank Miss Janis L. Tocher, Mrs. Irene Bell, and Mrs. Mary Torrance for excellent technical assistance. REFERENCES 1. Macleod J: Seminal cytology in the presence of varicocele. Fertil Steril 16:735, 1965 2. Martin DE, Gould KG: Normal and abnormal hominoid spermatozoa. J Reprod Med 14:204, 1975 3. Dublin L, Amelar RD: Varicocele size and results ofvaricocelectomy in selected subfertile men with varicocele. Fertil Steril 21:606, 1970 4. Nilsson S, Edvinsson A, Nilsson B: Improvement of semen and pregnancy rate after ligation and division of the internal spermatic vein: fact or fiction? Br J Urol 51:591, 1979 5. Rodriguez-Rigau LJ, Smith KD, Steinberger E: Relationship of varicocele to sperm output and fertility of male partners in infertile couples. J Urol 120:691, 1978 6. Tulloch WS: A consideration of sterility factors in the light of subsequent pregnancies. II. Subfertility in the male. Trans Edinburgh Obstet Soc 104:29, 1952 7. Lancranjan I, Popescu HI, Gavenescu 0, Klepsch I, Serbanescu M: Reproductive ability of workmen occupationally exposed to lead. Arch Environ Health 30:396, 1975 Vol. 38, No.2, August 1982 Liakatas et al. Scoring morphology using scanning microscope 231
8. Whorton D, Krauss RM, Marshall S, Milby TH: Infertility in male pesticide workers. Lancet 2:1259,1977 9. Gunn SA,.Gould TC, Anderson WA: Mechanism of zinc cysteine and selenium protection against cadmium induced vascular injury to mouse testis. J Reprod Fertil 15:65,1968 10. Becker RP, Sogard M: Visualization of subsurface structures in cells and tissues by back-scattered electron imaging. Scan Electron Microsc 2:835, 1979 LiakBtaset al. Scoring morphology using scanning microscope Fertility and Sterility