153
archives of andrology 47:153–156 (2001)
copyright ã 2001 taylor & francis
0148-5016 /01 $12.00 + .00

validation of a new disposable
counting chamber
k. coetzee
r. menkveld
reproductive biology unit, department of obstetrics
and gynecology, university of slenbosch and
tygerberg hospital, tygerberg, south africa
in a routine prospective study, 3 different counting chambers were compared for manual and computerassisted
evaluations. swim-up samples were used so as to remove all debris and other cells that may
contaminate the evaluation process when using a semen analyzer. the makler concentration determinations
(n = 20) were, on average, approximay 20 × 106 cells/ml higher compared to the corresponding
20 hemocytometer counts. the mean differences between the leja chambers and the hemocytomete r
counts (n = 320) were only around 1 × 106 cells/ml, with coefficients of variation around 20%. the
leja chambers for both manual and the computer-assisted sperm concentration determinations provided
consistent and accurate data on sperm concentration.
keywords computer assisted, concentration, human sperm, manual
sperm concentration is one of the major semen parameters that is evaluated as part of the
standard semen analysis. the clinical diagnosis of the male partner’s fertility primarily relies
on the outcome of the standard semen analysis. it is therefore imperative that the concentration
be determined accuray and reliably. although the initial sperm concentration of the sample
has been shown to have relatively low value [1, 4], postpreparation concentrations are routinely
used for the accurate insemination of oocytes in vitro. the parameter can be measured manually
or by a computer-assisted semen analyzer system (casa). a number of chambers, reusable
and disposable, are available to be used for both the manual and computerized evaluation.
before any chamber can be used for the routine evaluation of sperm concentration, its accuracy
and precision has to be evaluated.
a prospective study was therefore performed to compare manual and computer-assisted
sperm concentration measures using the hemocytometer (neubauer, west germany), makler
chamber (sefi medical instruments, israel), and the new disposable leja chamber (leja, nieuw-
vennep, the netherlands).
we thank leja products bv for their advice and for the materials used in conducting the experiment .
address correspondence to dr. r. menkveld, phd, andrology laboratory e3, tygerberg hospital, tygerberg,
7505, south africa. : rme@gerga.sun.ac.za
154 k. coetzee and r. menkveld
materials and methods
semen samples used for the study were obtained from patients presenting at our clinic for a
routine semen analysis. after liquefaction an aliquot of the semen sample was processed
according to our routine wash-and-swim-up procedure. the aspirates obtained after 1 h of
incubation were used for sperm concentration determinations. these swim-up samples were
used so as to remove all debris and other cells that may contaminate the evaluation process
when using a casa system, because without specific staining these debris may often be
recognized as sperm by the casa system. in theory, we would therefore mimic the evaluation
of latex beads. the resultant aspirates were diluted 1:10 with sterile water and a specific
aliquot was loaded for each evaluation, manual (leja, hemocytometer) or computer-assisted
(leja, makler). for the computer-assisted evaluations, at least 10 fields were evaluated. all
sperm concentration determinations were performed blind to the other outcomes.
the casa system used was the hamilton thorne research semen analyzer (ivos, version
10.8s, hamilton thorne research, beverly, ma). the standard parameter settings used were as
follows: frame acquired, 30; frame rate, 60 hz; minimum contrast, 85; minimum cell size, 2;
static size limits, 0.53–3.50; static intensity limits, 0.52–0.98; static elongation limits, 14–98
and temperature, 25ºc.
the sperm concentration outcomes produced by manual evaluation using the hemocytometer
were regarded as the standards (control). this method is used for the routine evaluation of
sperm concentration in our laboratory and is the method prescribed by the world health organisation
[5]. the sperm concentration outcomes of the other 3 methods were compared to the hemocytometer
results using the following statistical measures; coefficients of variation and bland and
altman plots.
results
the mean concentration (n = 20, 42.6 ± 37.4) obtained using the makler counting chamber
was significantly (p = .0078) higher than that obtained using the hemocytometer (n = 20,
22.7 ± 22.0). the individual concentration determinations were, on average, approximay
20 × 106 cells/ml (table 1) higher than the hemocytometer determinations. the high coefficient
of variation (» 50%) obtained for the makler outcomes may be due to the background
table 1. desc-riptive and comparative results of the sperm concentration outcomes measured.
hemacytometer leja leja makler
manual manual computer computer
number 35 35 35 20a
mean (sd) 21.2 (20.2) 20.3 (19.5) 19.9 (16.5) 42.6 (37.4)
range 1.1–88.2 1.0–81.0 1.5–72.9 7.3–152.7
coefficient of varianceb 19.2% 19.1% 47.9%
bland and altmanc 1.0 (–10.1–12.0) 1.3 (–12.2–9.6) –19.9 (-53.0–13.2)
note. values, if not indicated are, ×106cells/ml.
asee results for details.
bvalues as compared to the outcomes of the measured manually with the hemocytometer.
cbland and altman: mean difference (95% confidence interval).
disposable counting chamber 155
clutter (although this was limited by adjusting the size and elongation limits) and sample
calculation (dilution and chamber fill). for this reason only 20 samples were used for comparison.
the leja counting chambers outcomes, for manual and computer-assisted evaluations,
produced concentration outcomes extremely similar to the outcomes produced by the hemocytometer
(n = 35). the mean differences between the leja chambers, for manual and casa
determinations, and the hemocytometer were only around 1 × 106 cells/ml, with coefficients
of variation around 20% (table 1).
discussion
any method that is used to count the number of sperm using an aliquot from the original
sample is only an estimation of the true sperm concentration. variability is therefore inevitable
in the determination of sperm concentration. the origins of the errors have been well documented
(i.e., dilution, chamber, semen properties, human). it is therefore imperative to choose
a practical and reliable method that will limit the level of variability.
one of the obvious advantages of disposable chambers over reusable chambers is that there
is no damage to contend with. the damage that may occur to a hemocytometer or makler with
continued use can be largely overcome with manual evaluation, but this compensation is not
possible with computer-assisted evaluations. if a system like the indent system (hamilton
thorne research) is not being used, damage to the surface of the counting chambers may
generate a large amount of background clutter, which may negatively influence the outcome
being measured.
the significant differences we found using the makler corroborate a number of previously
conducted studies [2, 3]. the leja chambers showed a relatively low variation when compared
to the hemocytometer. the mean differences for both the manual and computer evaluations
using the leja chambers as compared to the hemocytometer were only approximay 1 × 106
sperm/ml. we feel confident that a 20% coefficient of variation in the context of sperm concentration
determinations is acceptable. no difficulties were encountered in the filling of both
types of leja chambers (manual and computer-assisted). the filled chambers, after a settling
period, exhibited good sperm distribution, evident from the excellent correlation with the hemocytometer
sperm concentration outcomes.
measurements performed by the manufacturer (leja) and an independent manufacturer of
precision assisted reproduction technology have determined the chamber height to vary between
18 and 20 μm. this small variance in chamber height may result in a small underestimation
of concentration.
the leja chambers for both manual and computer-assisted sperm concentration determinations
provided consistent and accurate data on sperm concentration with the use of swim-up
sperm samples.
references
1. badenoch df, evans sj, mccloskey dj (1989): sperm density measurement: should this be abandoned.
br j urol 64:521–523.
2. ginsburg ka, armant dr (1990): the influence of chamber characteristics on the reliability of sperm
concentration and movement measurements obtained by manual and videomicrographic analysis. fertil
steril 53:882–887.
156 k. coetzee and r. menkveld
3. sukcharoen n, ngeamjirawat j, chanprasit y, aribarg a (1994): a comparison of makler counting
chamber and improved neubauer hemocytometer in sperm concentration measurement. j med assoc
thai 77:471–476.
4. tomlinson mj, kessopoulou e, barrat clr (1999): the diagnostic and prognostic value of traditional
semen parameters. j androl 20:588–593.
5. world health organisation (1992): who laboratory manual for the examination of human semen
and sperm-cervical mucus interaction, ed 3. cambridge, uk: cambridge university press.