Bruce R. Gilbert, M.D., Ph.D.
George W. Cooper, Ph.D.
Marc Goldstein, M.D.
Department of Urology - The New York Hospital-Cornell Medical Center
and the Department of OB-GYN and Urology - North Shore University Hospital
INTRODUCTION
15% of couples have difficulty conceiving, and in 50-70% of these there is either
a primary male factor or an impaired semen quality sufficient to reduce the
probability of pregnancy in the subfertile female (1, 2). Accordingly, the first
and the easiest test in the evaluation of the subfertile couple should be the
semen analysis.
Although "seminal antimolecules" or spermatozoa were first described
in the human semen by Van Leeuwenhoek in 1677 (3), their morphologic evaluation
did not begin in earnest until the 1900's when normal and 'pathologic' forms
were found to co-exist in semen (4). To this day there is still controversy
as to what constitutes the "normal" spermatozoa in semen and the genetic
constitution of the "abnormal" forms. Three recent guidelines for
evaluation of human semen are described in the World Health Organization laboratory
manual for the examination of semen and semen-cervical mucus interaction (5),
in Keel (6), and in the atlas of Menkveld et al. (7).
Unlike routine serum biochemical tests, information obtained from the macroscopic
and microscopic examination of semen can be completely altered by events preceding
these examinations. Therefore, we will detail the collection protocols prior
to discussing the parameters for macroscopic examination, as well as microscopic
and biochemical tests. Particular emphasis will be given to sperm morphology,
since recent results of IVF have provided insight as to which morphologic characteristics
correlate with fertilization in vitro (8, 9). We will also discuss when and
how to test for antisperm antibodies, as well as the relative advantages, and
disadvantages, of computer assisted semen analysis. For the purposes of uniformity,
the WHO values (5) will be used as the normal semen values (Table 1),
unless otherwise stated. The nomenclature used for some semen variables is listed
in Table 2.
COLLECTION PARAMETERS
It is important to provide patients with clearly written instructions
on how to produce a specimen and as to the type of container to use. Routinely
we ask for 48-72 hours of abstinence. Longer periods usually produce a larger
volume, but often poorer motility quality. Occasionally when the percent motility
is reduced to 20% or less, we ask patients to produce a second sample within
1-2 hours after the first collection to rule out male tract storage problems.
This often results in a specimen with much improved motility, albeit with a
decreased total number of sperm though this can be within the normal range.
We also ask patients to evacuate their bladder prior to ejaculation and routinely
ask for a post-ejaculate urine to rule out retrograde ejaculation and/or possible
retention of semen in the urethra secondary to a noncompliant urethra.
The type of container used has always been of some concern in the past. Glass
containers were used for their supposed inertness, but frequent breakage has
led most semen analysis labs to use wide mouth polypropylene containers with
a screw top cap. Some plastics, such as polystyrene, have been found to cause
an increase in semen viscosity, and in fact, may be toxic to sperm. We ask for
the specimen to be delivered to the lab within an hour to 1 1/2 hours and to
be kept as close as possible to body temperature (preferably in a jacket pocket).
When received in the laboratory, we note the time of collection, the time that
we evaluate it and whether it is a complete sample or whether some loss occurred
during collection. We also ask for a post-ejaculate urine to be given at this
time. A sample of our work sheet is shown in Figure 1.
MACROSCOPIC EXAMINATION
Liquefaction usually occurs within 10-20 minutes of collection. On a scale
of 0-4 (4 being the normal value for a well liquefied sample), the failure to
liquefy is usually a sign that there is inadequate secretion by the prostate
of the proteolytic enzymes fibrinolysin, fibrinoginase and aminopeptidase (10).
Importantly, liquefaction should be differentiated from viscosity, as abnormalities
in viscosity can be the result of abnormal prostate function and/or the use
of an unsuitable type of plastic container. High Viscosity can usually
be reduced by repeated aspiration through an 18 or 19 gauge needle before or
after diluting the semen with a suitable buffer or culture medium to reduce
the amount of shear during physical disruption of the seminal gel. Volume
is measured with 2 to 5mL serologic pipettes; normal volume being between 2-6
mL. Color is normally opaque or opalescent with translucent semen usually
being oligospermic. The presence of a yellow hue or a very milky turbidity is
associated with pyospermia while a rust colored or reddish semen is indicative
of hematospermia. pH (normal 7.2 - 7.8) is also measured with colorpHast
pH 6.5-10.0 indicator strips or a standard pH meter. In acute infection (prostate,
seminal vesicles or epididymis) the pH of semen will be greater than 8.0 when
measured soon after liquefaction. In cases of obstruction of the ejaculatory
ducts or when only prostatic fluids are secreted the pH is usually less than
7.0.
MICROSCOPIC EXAMINATION
Sperm concentration (normal > 20 million/mL) is usually measured
either with a Makler chamber (Sefi Medical Industries, Haifa, Israel)(11, 12),
a Neubauer hemocytometer (American Optical Company, Buffalo, N.,Y.) or a disposable
device (Micro-Cell; Fertility Technologies, Inc., 215 Oak Street, Natick, MA.).
Various investigators have their preferences and certainly for normospermic
samples, each of these devices yields consistent results. However, controversy
exists as to which is the most accurate measurement device (13). The advantages
of the Makler chamber and Micro-Cell are that sperm counts can be done directly
on an undiluted semen after immobilization of the spermatozoa and evaluation
of the percent of motile sperm can be directly quantified. Use of the hemocytometer
requires dilution of the semen sample, due to chamber depth which can be done
with accuracy using volumetric pipettes with the diluent containing a fixative.
No matter which method is used, several measurements on a well mixed specimen
should be done and the mean of the results recorded. Adequate sampling is particularly
important in severely oligozoospermic semen. When fewer than 10 million sperm/mL
are present, we concentrate the specimen and re-suspend in a constant volume
to increase the sensitivity of our measurements. Total sperm count (normal
> 50 million) is then calculated by multiplying concentration times volume.
Motility (normal > 50%) is expressed as the percent of spermatozoa
that have motion and the forward progression (normal > 3) is noted.
A forward progression of 4 denotes spermatozoa rapidly moving in a straight
line with no yaw or lateral movement; unfortunately, the rapidity (or
linear velocity) of forward progression has not yet been standardized. A forward
progression of 3 denotes spermatozoa similarly moving linearly but at a slower
velocity. Sperm movement with a forward progression of 2 often exhibits angular
displacement or yaw to varying degrees while a progression of 1 denotes only
tail motion without progression. Zero progression being no movement at all.
The measurement of the duration of motility in seminal fluid is controversial.
MacLeod (14) has pointed out that the seminal plasma is only a temporary transport
medium and that evaluation of the duration of sperm motility in vitro
in semen, is probably not physiologically significant as sperm rapidly lose
their motility in the specimen container but remain motile for days in the mid
cycle cervical mucus.
If motility is less than 50% a viability stain is done using Eosin Y
with Nigrosin as a counterstain. Greater then 50% of the sperm should be viable
(i.e., non-stained). This is seen by bright field microscopy as red dye (Eosin
Y) taken up by the head of most non-motile sperm. Agglutination is also
noted if present (Figure 2). Though clumping of spermatozoa can mean
sperm antibodies are present, clumping usually only occurs around debris in
the sample. However, if motile dimers, of either head to head or tail to tail
associations are seen (Figure 2B), this is diagnostic for the presence
of antisperm antibodies in semen bound to the sperm.
Other than the spermatozoa, the presence of round cells should be noted
and when a count is made of spermatozoa, quantification of round cells by cell
type should also be made. Round cells should be differentiated between immature
germinal cells (Figure 3A), which usually have a single or double
round nucleus, are variable in size with a decreased nuclear to cytoplasmic
ratio as compared to polymorphonuclear leukocytes (PMN; should be 1 x
106 /mL) (Figure 3B), which are more constant in size, smaller than the
germinal cells, have interconnected lobes to their nucleus and have an increased
nuclear to cytoplasmic ratio. The classic method of distinguishing PMN's from
immature germinal forms is the peroxidase stain (15). PMN's are peroxidase-positive
while degranulated PMN's, lymphocytes and 'immature' germ cells are peroxidase
negative. The quantification of red blood cells, if present, should
also be made. In addition, bacterial contamination and presence or absence
of epithelial cells should be noted. If no sperm are seen, and/or the volume
is less than 1.5 cc, a fructose test should be done to confirm the presence
of seminal vesicle fluid, as well as a post ejaculate urine, to rule out retrograde
ejaculation.
MORPHOLOGY (Figure 4)
Sperm have a head, a mid-piece and a tail, each component of which has particular
morphologic characteristics. Figure 5 shows a comparison between the
strict morphologic criteria of Kruger, et al (9) and that of the WHO (5). The
head of the normal human spermatozoa is ovoid in the frontal view and pyriform
in the lateral view. Fixed and stained (Papanicolaou) sperm heads measure approximately
3-5 microns in length and 2-3 microns in width. The acrosome should make up
somewhere between 40-70% of the normal sperm head. The mid-piece has a mitochondrial
sheath and often excess cytoplasmic material from the developing spermatid.
The tail principal piece is approximately 50-55 micra in length and varies in
thickness from about 1 micron near the base to 0.1 microns at the tip of the
end-piece as shown by electron microscopy. The tail is composed of an axial
core consisting of two central singlet microtubules surrounded by nine pairs
of doublet microtubules , an outer ring of nine dense fibers surrounded by the
fibrous sheath which define the principal piece of the tail (Figure 6a,b).
The various structural alterations that can occur in each region of the spermatozoa
are shown in Figure 7. Menkveld et al. (7) have described an association
between the "strict" morphologic evaluation of spermatozoa and the
results of in vitro fertilization. Of particular note is the direct relationship
between acrosome size and the frequency of both pregnancy and fertilization
(Figure 8).
WHO TO TEST FOR ANTI-SPERM ANTIBODIES
In recent years, immunologic causes of fertility have been more frequently
recognized. As mentioned above, the presence of agglutination and in
particular motile dimers, should lead to investigations for anti-sperm
antibodies, as well as the standard indications of idiopathic infertility
or abnormal post coital tests. Multiple risk factors have been implicated
with increased risk of anti-sperm antibodies (Table 3). There are several
ways to test for antisperm antibodies, (Table 4). Good tests detect and
define binding sites to sperm surface antigens on motile, viable spermatozoa.
Non-motile and/or non-viable sperm are not normally involved in sperm/egg interactions
and it is the internal and subsurface antigens of these cells which have too
often been studied and which are not of clinical relevance. Thus, tests that
rely on sperm whose subsurface antigens are reveled as the consequence of chemical
fixation or freezing or air-drying all yield both false positive and false negative
results (16). In addition, assays that only evaluate the antisperm antibodies
in seminal fluid residual after removal of sperm or in serum by passive transfer
to donor third party sperm are of questionable clinical significance.
Two current methods of detecting antibodies bound to the surface of motile sperm
are the mixed agglutination reaction assay (MAR test; only for IgG's) and the
immunobead binding assay (for IgA, IgG and IgM's). Both tests require motile
sperm, the MAR test (SpermMar: Fertility Technologies, Inc., 215 Oak Street,
Natick, MA.;(17)) is a simple test that can be done with a light microscope
on unprocessed semen. It is highly sensitive and IgG specific and is performed
by mixing fresh, untreated semen with latex particles coated with human IgG.
To this mixture a monospecific antihuman IgG antiserum is added. The formation
of mixed agglutinates between latex particles and motile spermatozoa indicates
the presence of IgG antibodies on the spermatozoa. The binding to motile sperm
and can be easily seen by light microscopy and localized as binding to different
regions of the head, mid-piece or tail. The % of motile sperm, which are head
, mid-piece or tail bound or not bound are recorded. Greater than 40% binding
is considered to be significant. In the indirect SpermMar test, washed motile
donor spermatozoa are incubated with diluted decomplemented patient serum of
male or female origin. If the serum contains antisperm antibodies, these will
bind to the donor spermatozoa which will react with latex particles in the SpermMar
test. To-date the kit is available only for IgG antibodies. A similar test for
IgA, probably a clinically more important antibody, is presently under production.
The immunobead assay (18) has the advantage of being extremely sensitive and
specific and can evaluate binding in three immunoglobulin classes ( IgG, IgA
and IgM). The immunobead assay also can be performed as a direct test on sperm
isolated from semen as well as an indirect test of immunoglobulin binding following
incubation in seminal fluid or semen. The procedure requires one additional
step, washing sperm prior to incubation with the polyacrylamide microspheres
which also have been bound with anti-human IgG, IgA, or IgM antiserum. The immunobead
study requires a phase contrast microscope for evaluation.
TESTS OF SPERM FUNCTION
No review of semen analysis can be complete without mentioning the tests of
sperm function. Although spermatozoa may look normal, move with good forward
progression and lack surface antibodies, this is no assurance that they perform
properly. Of the many tests of sperm function available (Table 5), none
unequivably identify functional spermatozoa. (1) The spermatozoa penetration
assay (SPA) utilizes the golden hamster egg which is unusual in that removal
of it's zona pellucida results in loss of all species specificity to egg penetration.
The SPA can therefore be used to yield information regarding the fertilizing
capacity of human spermatozoa. The concept of the sperm penetration assay was
introduced by Yanagamachi in 1972 and modified for use in humans in 1976 (19,
20). The spermatozoan must undergo several changes prior to penetrating a zona
pellucida-free hamster egg. These are capacitation and the acrosome reaction
which require still poorly understood biophysical and biochemical changes induced
both in vitro and in vivo in the presence of mature oocytes. Thus
the zona-less egg penetration assay measures sperm capacitation, the acrosome
reaction, sperm/oolemma fusion, sperm incorporation into the ooplasm and the
decondensation of the sperm chromatin. However, penetration of the zona pellucida
and normal embryonic development are not tested. Thus, a positive SPA does not
guarantee fertilization of intact human eggs nor their embryonic development
while conversely, a negative SPA has not been found to correlate with poor fertilization
in human IVF. (2) The acrosin assay measures acrosin, a trypsin
like serine proteinase found in the acrosome which may be responsible for penetration
of the zona pellucida and also triggering the acrosome reaction (21). Measurement
of acrosin is thought to be correlated with physiologic facilitation of the
acrosome reaction and sperm binding to and penetration of the zona pellucida.
(3) The hemizona assay (22) measures the binding of sperm to the
zona pellucida of human eggs. Human spermatozoa will bind firmly to only the
human zona pellucida and that of closely related higher primates (23). For ethical
reasons one does not perform diagnostic "functional" assessments of
binding of human spermatozoa to intact viable human oocytes. In the human hemizona
assay, unfertilized oocytes obtained through donation are bisected and the number
of sperm tightly bound to the outer surface counted. The major advantages of
the hemizona assay is that: 1) the two halves of the hemizona are functionally
(qualitatively), equivalent surfaces allowing controlled comparison of binding,
and 2) the limited number of available human oocytes is amplified. The results
of this test show a good correlation with "strict" morphology (9)
and IVF fertilization rate (24). (4) The hypoosmotic swelling test
(HOS) (25) measures the membrane integrity of sperm, important not only
for metabolism, but successful union of male and female gametes. A critical
property of all cell membranes is their ability to selectively transport molecules.
Therefore, when viable spermatozoa are exposed to hypoosmotic conditions, water
enters the sperm resulting in swelling. This places the tail fibers under tension
causing curling of the tails, which can be readily visualized by phase contrast
microscopy. The HOS test correlates highly (r=0.9), with the outcome of zona
free hamster penetration assay when normal ejaculates are used (25). A distinct
advantage of this test is that it is simple, cheap and readily applicable clinically.
(5) The cervical mucus penetration assay quantitates the sperm-mucus
interaction and yields information regarding the motility of the spermatozoa.
Although millions of sperm are deposited in the vagina, only hundreds ever reach
the ampulla of the fallopian tubes. A major factor in this dramatic reduction
is the cervical mucus. After the estrogen surge, the cervical mucus is about
98% water, compared to its composition at times under control of progesterones
when the water content falls to about 90%. At midcycle, during the more aqueous
phase, cervical mucus is less of a barrier to sperm and it is at this point
that the post coital test (PCT) is performed. Alliquots of cervical mucus
are collected within 12 hours after coitus at midcycle. Good PCT being defined
as greater than 10 motile sperm/high powered field. Fewer sperm may indicate
collection of non midcycle mucus or poor mucus sometimes referred to as hostile
cervical factors. The presence of antibodies, oligozoospermia or sperm with
poor motility also yield poor test results . Female factors that contribute
to a poor PCT include inadequate estrogen priming, deficient endocervical tissue
or previous surgery to the cervix. There are also other in vitro tests,
using either human or bovine cervical mucus, again to test the penetration by
motile sperm. These are subject to variable mucus quality and are difficult
to standardize and interpret.
COMPUTER AUTOMATED ANALYSIS OF SEMEN
Automated semen analysis is increasingly being used for the routine evaluation
of semen quality, as well as for assessing the role motility plays in fertilizing
potential in vivo and in vitro. To meet these goals an automated
analyzer must accurately count sperm, and percent motility, evaluate motion
characteristics based on their ability to quantitate both the angular and linear
motion (26, 27) of sperm populations and analyze sperm morphology. However,
the present 'gold standard' remains manual evaluation performed by trained technicians.
Automation of this seeks to increase not only the rapidity of measurement but
also the accuracy allowing standardization of semen parameters between laboratories
by eliminating inter-technician biases.
Automated semen analyzers, although doing a superb job with motion evaluation,
suffer from many technical problems (28) including their inability to adequately
differentiate small particles ('debris') from both sperm or round cells (29-32).
In addition, the evaluation of sperm morphology, is still not well developed
for automated analyzers (33). Although, the hope was to standardize the measurement
of semen parameters by automated devices, this has not yet been obtained. Jasko
et al. (34) found when comparing two automated semen analyzers that the results
were often dissimilar making direct comparisons of results difficult. Lastly,
even functional assessment is questioned since the need for 'human' intervention
in the analysis (31) has made inter and intra laboratory comparisons of motion
related phenomena difficult.
Therefore, although the potential for automated semen analysis is great. It
has not yet reached its goals of providing a highly reliable method to evaluate
a semen specimen. Until the technique becomes further developed it remains primarily
a research tool.
REFERENCES:
1. Simmons F. Human infertility. N Engl J Med 1956;255:1140.
2. MacLeod J. Human male infertility. Obstet Gynecol Surv 1971;26:325.
3. Mayer AW. The discovery and earliest representations of spermatozoa. Bull
Inst Hist Med 1938;6:89.
4. Hotchkiss R. Fertility in Men.London: 1945(Heinemann W, ed.
5. WHO. WHO Laboratory manual for the Examination of Human Semen and Semen-Cervical
Mucous Interaction.Cambridge: University Press, 1987
6. Keel BA. The semen analysis. In: Keel B, Webster B, ed. CRC Handbook of the
Laboratory Diagnosis and Treatment of Infertility. Boca Raton: CRC Press, 1990:
27-69.
7. Menkveld R, Oettlee, TF K, RJ S, Acosta A, Oehninger S. Atlas of Human Sperm
Morphology.Baltimore: Williams and Wilkins, 1991
8. Hershlag A, Cooper G, Rosenfeld D. Optimizing fertilization of sperm populations
containing acrosome defective sperm by increasing the number of normal spermatozoa
per oocyte. 7th World Congress on IVF and Assisted Reproduction. Paris, France:
, 1991: 388.
9. Kruger T, Acosta A, Simmons K, Swanson R, Matta J, Oehninger S. Predictive
value of abnormal sperm morphology in in vitro fertilization. Fertil Steril
1988;49:112.
10. Amelar R. Coagulation, Liquefaction and Viscosity of Human Semen. J. Urol.
1962;87:187.
11. Makler A. The improved 10-micrometer chamber for rapid sperm count and motility
determination. Fertil. Steril. 1980;33:337.
12. Makler A. A new chamber for rapid sperm count and motility determination.
fertil. Steril. 1978;30:313.
13. Ginsburg K, Armant R. The influence on chamber characteristics on the reliability
of sperm concentration and movemnet measurements obtained by manual and videomicrographic
analysis. Fertil.Steril 1990;53:882.
14. MacLeod J. The semen analysis. Clin.Obstet.Gynecology 1965;8:115.
15. Nahoum C, Cardozo D. Staining for volumetric count of leukocytes in semen
and prostate-vesicular fluid. Fertil Steril 1980;34:68.
16. Bronson R, Cooper G, Rosenfeld D, Witkin SS. Detection of spontaneously
occurring sperm-directed antibodies in infertile couples by immunobead binding
and enzyme-linked immunosorbent assay. Ann. NY Acad. Sci. 1984;438:504.
17. Comhaire F, Hinting A, Vermeulen L, Schoonjans F, Goethals I. Evaluation
of the direct and indirect mixed antiglobulin reaction with latex partivles
for the diagnosis of immunological infertility,. Int.J.Androl. 1987;11:37.
18. Bronson R, Cooper G, Rosenfeld D. Sperm antibodies: their role in infertility.
Fertil Steril 1984;(42):171-183.
19. Yanagimachi R. Penetration of guinea-pig spermatozoa into hamster eggs in
vitro. J. Reprod. Fertil. 1972;28:477.
20. Yanagimachi R, Yanagimachi H, Rogers B. The use of zona-free animal ova
as a test-system for the assessment of the fertilizing capacity of human spermatozoa.
Biol Reprod 1976;15:471.
21. Rogers B, Brentwood J. Capacitation, acrosome reaction and fertilization.
In: Zaneveld L, Chattterton T, ed. Biochemistry of Mammalian Reproduction. New
York: Wiley & Sons, 1982: 203.
22. Burkman L, Coddington C, Franken D, Kruger T, Rosenwaks Z, Hodgen G. The
hemizona assay (HZA): development of a diagnostic test for the binding of human
spermatozoa to human hemizona pellucida to predict fertilization potential.
Fertil Steril 1988;49:688.
23. Bedford J. Sperm/egg interactions: The specificity of human spermatozoa.
Anat Rec 1977;188:477.
24. Franken D, Burkman L, Coddington C, Oehninger S, Hodgen G. Human hemizona
attachment assay. In: Acosta A, Swanson R, Ackerman S, Kruger T, van Zyl J,
Menkveld R, ed. Human Spermatozoa in Assisted Reproduction. Baltimore: William
and Wilkins, 1990: 355.
25. Jeyendran R, Van der Ven H, Perez-Pelaez M, Crabo B, Zaneveld L. Development
of an assay to assess the functional integrity of the human sperm membrane and
its relationship to the other semen charateristics. J Reprod Fertil 1984;70:219.
26. Bongso TA, Ng SC, Mok H, et al. Effect of sperm motility on human in vitro
fertilization. Arch Androl 1989;22(3):185-90.
27. Fetterolf PM, Rogers BJ. Prediction of human sperm penetrating ability using
computerized motion parameters. Mol Reprod Dev 1990;27(4):326-31.
28. Mortimer D. Objective analysis of sperm motility and kinematics. In: Keel
B, Webster B, ed. CRC Handbook of the Laboratory Diagnosis and Treatment of
Infertility. Boca Raton: CRC Press, 1990: 97-133.
29. Chan SY, Wang C, Song BL, et al. Computer-assisted image analysis of sperm
concentration in human semen before and after swim-up separation: comparison
with assessment by haemocytometer. Int J Androl 1989;12(5):339-45.
30. Neuwinger J, Knuth UA, Nieschlag E. Evaluation of the Hamilton-Thorn 2030
motility analyser for routine semen analysis in an infertility clinic. Int J
Androl 1990;13(2):100-9.
31. Levine RJ, Mathew RM, Brown MH, et al. Computer-assisted semen analysis:
results vary across technicians who prepare videotapes. Fertil Steril 1989;52(4):673-7.
32. Ginsburg KA, Moghissi KS, Abel EL. Computer-assisted human semen analysis.
Sampling errors and reproducibility. J Androl 1988;9(2):82-90.
33. Wang C, Leung A, Tsoi WL, et al. Computer-assisted assessment of human sperm
morphology: comparison with visual assessment. Fertil Steril 1991;55(5):983-8.
34. Jasko DJ, Lein DH, Foote RH. A comparison of two computer-automated semen
analysis instruments for the evaluation of sperm motion characteristics in the
stallion. J Androl 1990;11(5):453-9.