| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Review |
From the Animal Reproduction and Biotechnology Laboratory, Colorado State University, Fort Collins, Colorado.
| Correspondence to: Dr Rupert P. Amann, 909 Centre Ave, #123, Fort Collins, CO 80526-2091 (e-mail: rpalra62{at}comcast.net). |
| Received for publication September 10, 2008; accepted for publication April 23, 2009. |
Total number of sperm per ejaculate (TSperm) is an important measure for
clinicians to provide advice to patient couples. However, TSperm per hour of
abstinence (TSperm/h) is a better measure for epidemiologist-andrologist teams
or clinicians to evaluate spermatogenesis because it is a rate function. This
review looks at the interplay and impacts of rate of sperm accumulation in the
excurrent duct system, abstinence interval, sexual arousal, and masturbation
vs intercourse on observed TSperm. It also examines why and when TSperm/h
might provide a meaningful quantitative evaluation of spermatogenesis (ie,
rate of sperm production). There is no doubt that TSperm increases with longer
abstinence, and in different men plateaus after 2–9 days. Clinicians
wishing to maximize number of fully functional sperm available during
intercourse, or for artificial insemination, might wish to recommend 6–7
days of abstinence. Diagnostically, the important feature is TSperm/h. After
abstinence interval exceeds 64–72 hours, TSperm/h has started to decline
in most nonoligozoospermic men as rate of sperm accumulation in the excurrent
ducts approaches zero; apparently increasingly more sperm are voided in urine.
Clinicians or epidemiologist-andrologist teams wishing to have optimal
distinction among individuals with high, typical, or low sperm production (ie,
normal or abnormal spermatogenesis) should accurately measure TSperm/h for
samples provided after 42–54 hours' abstinence (never
36 or >64
hours). Longer abstinence intervals reward men with poor sperm production,
because sperm accumulate in the excurrent ducts for 7 days or more of
abstinence, and penalize men with good sperm production, because after 3 days
or less of abstinence their excurrent ducts probably are full.
Key words: Evaluating spermatogenesis, total sperm per hour of abstinence, abstinence interval, excurrent ducts, sexual arousal
The introduction to most epidemiologic publications states or implies that the authors sought to determine whether "agent X" was associated with abnormal semen characteristics. This might seem logical because andrologists and clinicians assume that measured characteristics of ejaculated semen reflect illness or nonillness of the testes, epididymides, and accessory sex glands. However, methods for obtaining the semen sample(s) and decisions when selecting which seminal attributes to report usually preclude a meaningful evaluation of spermatogenesis. Reasons for this statement are evident throughout this review.
Consideration of evaluating spermatogenesis on the basis of ejaculated semen is timely because there is increasing emphasis on the fetal basis for adult disease, including "testicular dysgenesis syndrome" (Skakkebæk et al, 1998; Sharpe, 2006), and impact of environment or life style on testis function in adults. Agents studied for possible impact on testes function in adults include maternal consumption of beef or caffeine during pregnancy (Swan et al, 2007; Ramlau-Hansen et al, 2008); maternal obesity during pregnancy (Ramlau-Hansen et al, 2007); exposure of young adults to DDT (Aneck-Hahn et al, 2007); presence of metabolic syndrome in subjects (Kort et al, 2006; Aggerholm et al, 2008); or region within Europe, the United States, or China where subjects lived (Jørgensen et al, 2001; Swan et al, 2003; Gao et al, 2007). As this quest to understand reproductive problems afflicting many men continues, it is important to look at spermatogenesis (and other features of testis function) rather than look at semen.
Noninvasive evaluation of spermatogenesis is best accomplished via data on TSperm per hour of abstinence (TSperm/h; 106 sperm/h; calculated from TSperm and abstinence interval), for reasons summarized herein. For biological reasons, information on number of sperm per milliliter of semen (ie, sperm concentration) is not useful for evaluation of spermatogenesis (Amann, 2009). This review summarizes literature (including accompanying Amann and Chapman, 2009) important to an epidemiologist-andrologist team, clinician, or andrology laboratory director in planning a logical approach to request semen samples and explaining such requests to potential study participants or patients.
The review is organized as sections, from spermatogenesis to estimating sperm production, and culminating in the final section: the take-home message. To achieve the contradictory goals of readability/utility and thoughtful review of pertinent literature, for most sections only selected papers are summarized herein and details, critique, or contradictory evidence are presented in similarly organized Supplemental Material (available online at www.andrologyjournal.org).
Spermatogenesis and Semen
Spermatogenesis in humans is complex and our knowledge has limitations
(comprehensive review in Amann,
2008). The process requires 
74 days, regardless of number of
sperm produced. Many potential sperm never are produced because of cell death,
especially during the preleptotene/leptotene portion of meiosis-1 and during
meiosis-2. Many sperm released from the seminiferous epithelium are abnormal.
Hence, spermatogenesis can be deficient in quantity of sperm produced, quality
of sperm produced, or both.
The quantitative aspect of spermatogenesis is a rate (ie, number of sperm, or less mature cell type, produced per unit time). It is best expressed as daily sperm production, or daily sperm production per gram testis parenchyma, measured by morphometric analysis of fixed tissue or enumeration of germ cells in homogenates of fixed tissue (Amann, 2008). Both methods require access to testicular tissue (eg, via a biopsy) and can give similar results; the first method is very labor intensive. Because of differences among men in innate testis size, daily sperm production per gram of testis is a better measure of normalcy of a man's testes than daily sperm production.
Daily sperm production can be estimated noninvasively as TSperm/h since the
previous emission/ejaculation or as daily sperm output, but not as TSperm. The
difference between the 2 estimates for daily sperm production is that daily
sperm output is more precise because of better premeasurement stabilization of
epididymal sperm reserves and number of samples used to calculate mean TSperm
per hour or per day (1–3 vs 
6 samples). Definitions are in the
"Definitions" section. Provided a suitable abstinence interval is
used, either approach for estimating daily sperm production is valid because
essentially all sperm entering the efferent ducts leave during emission, and
can be recovered in ejaculated semen or urine (see "Fate of Epididymal
Sperm").
Considerations in obtaining a meaningful value for TSperm or TSperm/h are throughout this review and are brought together in "A Meaningful Value for TSperm or TSperm/h." A succinct recommendation is in "Take-Home Message." A protocol to measure daily sperm output is in Section 7 of the Supplemental Material. It is far less demanding than those suggested earlier (Freund, 1962, 1963; Amann, 1970; Johnson, 1982). One might be tempted to adjust TSperm/h for estimated testis size to give a noninvasive estimate of daily sperm production per gram of testis. This probably might be unwise, because it is difficult to accurately measure parenchyma volume in situ.
Quantifying daily sperm production as TSperm/h or daily sperm output is
important because only using these measures (assuming testes are not biopsied)
can one learn whether "agent X" affected number of sperm currently
being produced in a subset of individuals within a population. TSperm/h better
reflects testes function than TSperm, because on average TSperm changes by
48% if abstinence interval is 64 rather than 40 hours (2% per hour;
Amann and Chapman, 2009).
Logically, precision of measurement of TSperm/h for an individual should be
sufficient to:
Obviously, an imprecise value for TSperm/h can lead to questionable conclusions. The number of individuals needed to detect a difference of X% in TSperm/h because of action of a past or future "agent" on a population of men should be considered.
An answer to the qualitative aspect of "did agent X affect spermatogenesis?" requires an atypical mindset, restricting measurements of sperm quality to attributes that:
Because data to make back calculations are not available, noninvasive measurement of the quality of spermatogenesis must rely on attributes in the first category. The qualitative aspect of spermatogenesis is not considered herein.
Background on the Excurrent Ducts
General—
The human epididymal duct is 4.5–6.0 m long and continues as the vas
deferens with ampulla. The vas deferens and ampulla are 30–45 and
3–7 cm long, respectively. These structures might contain some sperm
during sexual abstinence and many more as precopulatory sexual arousal moves
sperm distally (detail in Supplemental Material, Section 2.a).
There are no data for concentrations of sperm in rete testis fluid entering the human efferent ducts or within the lumen of the human epididymis. However, examinations of histological sections reveal that the concentration of sperm within the epididymal duct increases substantially between the proximal caput and cauda (Turner, 2008). Sperm probably are transported through the caput and corpus epididymidis by frequent pendular contractions of smooth muscle fibers in the duct wall, and in the cauda (and vas deferens) by low-frequency segmental contractions (Amann et al, 1993; Robaire et al, 2006). To-and-fro movement is greater than net movement.
X-ray cinematography (Mitsuya et al, 1960) showed that during sexual excitement, the ampullae were drawn towards the pubis and the vasa were straightened. Early in emission, sperm apparently were moved from the caudae epididymides into the ampullae. Emission continued with discharge of prostatic fluid into the prostatic urethra, followed by sperm and contrast medium from the ampullae; presumably all were mixed. Ejaculation discharged this mixture from the urethra plus additional boluses of sperm from the ampullae concurrent with fluid from 10 or fewer contractions of the vesicular glands, followed by a little prostatic fluid. It is uncertain whether additional boluses of sperm are moved from the caudae epididymides during emission/ejaculation, or whether residual sperm are moved retrogradely from the ampullae after ejaculation. In any case, the ejaculate is best described by TSperm, quantifying the total number of sperm from the excurrent ducts, and seminal volume, summing the fluid accompanying sperm or contributed by accessory sex glands (Amann, 2009).
Number of Sperm in the Excurrent Ducts— Data for human epididymal sperm reserves are for tissues taken at autopsy shortly after a myocardial infarction, gunshot wound, or automobile accident. Based on available data (Table) it was estimated that for a hypothetical healthy man, 25–45 years old, with a daily sperm production of possibly 130–270 x 106 sperm, the 95% confidence limits (CLs) for number of sperm in paired capita, corpora, or caudae epididymides would be near 65–115, 65–116, and 165–275 x 106. These values are rough approximations, and do not exactly reflect the number of individuals or variance in any of the 3 groups of men used in the calculations. Calculating from the lower and upper ends of these ranges, for a typical man his 2 epididymides might contain between 296 and 506 x 106 sperm.
|
Number of sperm in paired human vasa deferentia, including the ampullae,
after 2–6 days of abstinence is unknown. However, it was estimated
(Supplemental Material, Section 2.b) that a pair of vasa deferentia and
ampullae might contain 50–100 x 106 sperm, before
sexual arousal, in a majority of nonoligozoospermic men. Hence, 70% of
the sperm in a typical ejaculate might have been in the caudae epididymides
prior to sexual arousal. After sexual arousal but before emission, the paired
ampullae might contain up to 380 x 106 sperm.
Fate of Epididymal Sperm— The fate of sperm within the epididymis has been a topic of research and discussion for >70 years. The topic remains controversial (eg, Cooper et al, 2002; Sutovsky et al, 2002) because authors are not explicit in distinguishing 1) fate(s) of >90% of sperm entering the epididymal duct in normal adult men or animals based on quantitative measures from 2) apparent fate of a few sperm in normal males (possibly more in neopubertal and abnormal males or aged males) based on nonquantitative visualization in tissue sections or luminal fluid expressed from the duct.
Current knowledge (Supplemental Material, Section 2.c) leads to 3
conclusions, applicable to men and other common mammals with patent excurrent
ducts. In the future, new robust, quantitative data might shift the magnitude
of sperm removal attributed to each of the following paths. 1) A few sperm
(<1%–5%??) might be removed from the excurrent ducts consequent to
apoptosis, dissolution, absorption, or phagocytosis. 2) Virtually all sperm
(likely 90%) leaving the testes pass through the excurrent ducts into the
pelvic urethra. 3) Essentially all sperm entering the epididymides accumulate
until 1 of 2 events occurs: i. emission and ejaculation, to reduce size of the
sperm population by removal of sperm from the distal ducts; or ii. the sperm
population reaches the limit of each duct's distensibility and restricting
musculature, a matter of less than 4 days for sperm leaving the testis during
a given hour in most nonoligozoospermic men having no emission, after which
some sperm from 1 or both distal ducts leave during unrecognized
"spilling out" into the pelvic urethra to be washed out in
urine.
These conclusions are consistent with what happens after vasectomy. When egress from the epididymal duct is surgically blocked, sperm accumulate until the epididymal duct ruptures and sperm are extravasated. The interval until duct rupture depends on the species, and ranges from days to months (Bedford, 1976; Supplemental Material, Section 2.c). For vasectomized men, duct rupture might occur every 2–3 months (Amann and Howards, 1980).
Some sperm apparently are tagged for apoptosis while they are spermatids,
but they seem to pass through the epididymis
(Huszar et al, 1998;
Cayli et al, 2004). Sperm are
ubiquitinated within the human efferent ducts and/or epididymal ducts, but
micrographs and statements that ubiquitinated sperm might be removed by
apoptosis, dissolution, or absorption (Sutovsky et al,
2001,
2004;
Baska et al, 2008) are not
supported by quantitative data. Indeed, ubiquitinated sperm are abundant in
ejaculated semen. Section 2.c in the Supplemental Material considers
apoptosis, dissolution, or absorption of sperm; documents that during
prolonged abstinence sperm apparently spill out into urine; and suggests why
the contrary report by Barratt and Cook
(1988) might be in error.
Certain published research reports stated that 50% of daily sperm
production was unobtainable in ejaculated semen, but Amann
(2008) and Section 2.c in the
Supplemental Material teach how procedural errors had led to erroneous
conclusions.
Transit Time of Sperm Through the Epididymis— The best estimates of time for transit of sperm through the human epididymis are based on enumerations of sperm within the excurrent ducts and measurement of daily sperm production of the attached testis. To calculate transit time, number of sperm found in a portion of the epididymis, or the entire organ, is divided by daily sperm production (Orgebin-Crist, 1962; Amann et al, 1974; Amann, 2008). This gives a typical or average value, but cannot provide minimal transit time. Also, there is some mixing of older and younger sperm. Estimates based on labeling spermatids with a marker (eg, Rowley et al, 1970; Misell et al, 2006) cannot provide accurate estimates (Supplemental Material, Section 2.d).
Meaningful values for transit time are summarized in the Table, with the top 3 rows showing group means of 6.0, 3.2, or 1.9 days. At least in part, transit time of sperm through a human epididymis is a function of daily sperm production (Figure 1). Arguably, men with a daily sperm production per testis of 60–150 x 106, and epididymal transit times less than 5 days, might best represent "normal" men in certain epidemiologic studies. The longest mean epididymal transit times were 6–14 days, all for individuals producing less than 60 x 106 sperm/testis. For 73% of men in Figure 1, duration of epididymal transit was estimated at less than or equal to 5 days. In addition, a few sperm remain in the vas deferens and ampulla for an unknown interval.
|
It seems unlikely that daily emission hastens transit of sperm through the
caput epididymidis, although there are no data for humans. However, it is
presumed that sperm transit distally into or through the human cauda
epididymidis, and probably into the vas deferens, is more rapid for 0–18
hours after ejaculation than subsequently during sexual abstinence without
emission (Amann and Chapman,
2009). This apparently positions up to 100 x
106 sperm to augment those otherwise available. Rapid
postejaculatory repositioning of sperm is evidenced by: 1) sperm number in the
first ejaculate after vasectomy (Supplemental Material, Section 2.b); and 2)
"extra sperm" in samples provided after 18–36 hours of
abstinence compared with what would be expected based on TSperm after
48–84 hours' abstinence (Amann and
Chapman, 2009). Impact of this rapid transit on TSperm and
TSperm/h is discussed in "Select a Restrictive Abstinence
Interval."
Relationships Among Daily Sperm Production, TSperm, TSperm/Hour, and Daily Sperm Output
Definitions—
Daily sperm production is the number of sperm produced per day by
a testis or 2 testes of an individual (Amann,
1970,
2008); that is, a rate
describing spermatogenesis. It is a quantitative endpoint for success in
spermatogenesis: success in maintenance of the population of progenitor
Apale-spermatogonia, production of committed
Apale-spermatogonia, and proliferation of progeny from each
committed Apale-spermatogonium to spermatozoa released from the
seminiferous epithelium after completing spermiogenesis. Prolonged or severe
reduction, or failure, in production of committed
Apale-spermatogonia is one cause of reduced daily sperm production.
The other is an unusually high rate of apoptosis or death of differentiating
germ cells. Either or both problems would reduce the number of sperm produced
per day per gram of testicular parenchyma, termed efficiency of sperm
production, and overall daily sperm production. Testis size (or weight) also
might diminish. Hence, daily sperm production per gram of testis parenchyma is
the best measure of success in spermatogenesis.
Daily sperm production can be measured directly via morphometric analysis of fixed testis tissue, enumeration of elongated spermatids in homogenates of testis parenchyma, or quantitation of sperm passing through an indwelling catheter placed in the rete testis (Amann, 1970, 2008; Amann et al, 1974). Daily sperm production is not altered by sexual arousal or emission/ejaculation frequency (Amann, 1970, 1981). Measurements of daily sperm production have high precision, and CVs for within-individual measurements usually are less than 15% (Amann, 2008:482). This is far less than among-individual variation in daily sperm production per gram of testis (30%–60%; Berndtson, 2008).
TSperm is a measured attribute of ejaculated semen (Amann, 2009). It is not a rate. Accurate measurement of TSperm requires gravimetric determination of seminal volume before semen is removed from the container receiving a masturbation sample or recovery and enumeration of sperm remaining in a silicone collection device, worn during intercourse, after most of the semen is aspirated. To be meaningful, the abstinence interval associated with a semen sample should be known (recorded in hours). TSperm per se is important to a clinician, provided it is averaged over 3 or more samples.
Measurements of TSperm in repeated samples from an individual usually give a standard deviation approximately proportional to its mean and a high CV around the mean (Schwartz et al, 1979; Baker et al, 1981; Amann and Chapman, 2009). Depending on the report, within-individual CVs average between 35% and 94%. Authors usually recognized that these values were misleading because data had a right skew, raw data should be transformed, and CVs for transformed data would be smaller than reported values. For example, for 48 donors (18–20 samples per donor) in Amann and Chapman (2009), the within-individual CV for log10-transformed values for TSperm averaged 7% and among donor CV was 6%, vs 35% and 36% for raw data.
TSperm/h of abstinence is a rate attribute reflecting
quantitative success of spermatogenesis. Although some might term it a derived
attribute of an ejaculate, TSperm/h should receive greater attention by
clinicians and should be adopted by epidemiologist-andrologist teams as the
measure most descriptive of the quantitative aspect of spermatogenesis.
Correctness of the value is dependent on accurate measurement of TSperm and
truthful recording of abstinence interval. TSperm/h should be calculated for
individual samples obtained after 42–54 hours of abstinence (never
36 or >64 hours) and then used in statistical analyses or averaged
across 2–3 samples. Log10 transformation of data for TSperm/h
somewhat reduces the within- and among-individual variation. CVs for
transformed data were 24% and 26%, respectively, vs 33% and 39% for
nontransformed data (Amann and Chapman,
2009).
Daily sperm output is a traditional term (Amann, 1970) and is simply mean TSperm/h for 6 or more successive samples, produced at a uniform interval (eg, 48 hours) without intervening ejaculations, expressed on a per-day or per–24 hours basis. Daily sperm output is measured after stabilization of the number of sperm in the excurrent duct system (ie, extragonadal sperm reserves) in a special research setting. A modern protocol is in the Supplemental Material, Section 7.
Figure 2 (left) illustrates
relationships among these measures. Daily sperm production represents the gold
standard for the other measures, but the true value for any measure based on
TSperm must be lower because 10% of sperm moved from the excurrent ducts
remain in the pelvic urethra to be washed out in the first postejaculatory
urination (Sigman et al, 2009; near 8%, 23%, or 38% of all sperm for 54%, 25%,
and 13% of fertile men). Assumptions in respect to TSperm include no
elimination of sperm within the excurrent duct system, ejaculations at 48-hour
intervals, and (for simplicity) that each mean for TSperm/24 hours was
identical with the true value. Although the observed mean was depicted equal
to the true value, it might have fallen anywhere within the box or even higher
or lower. The right panel in Figure
2 is a density plot for the right-skewed theoretical distributions
of the box for n = 1 in the left panel. This illustrates that box plots of
skewed data might be deceptive, because individual values for TSperm are more
likely to be near the true value than near the limits of the distribution
depicted by the box (when mean TSperm is based on 6 or more samples, the
theoretical distribution becomes bell-shaped; see Supplemental Figure 1 in
Amann and Chapman, 2009,
available online at
www.andrologyjournal.org).
|
From Figure 2 it is evident that: 1) daily sperm production usually is measured with high precision; 2) an observed value for TSperm/24 hours is more likely to be above the true value than below; 3) observed TSperm/24 hours is twice as likely to be near the true value when the mean is based on 3 samples rather than 1 sample; and 4) daily sperm output based on 6 or 10 samples reduces imprecision, but even then daily sperm output provides an imprecise estimate of daily sperm production (ignoring the depicted 10% offset for sperm moved into the pelvic urethra but not ejaculated). Daily sperm output provides the best noninvasive estimate of daily sperm production for an individual, but in most studies with humans the improved precision of the estimate will not be worth the effort. With animals, however, measurement of daily sperm output is practical and common, and might be calculated from 10 or more samples.
TSperm and Abstinence Interval—
It is illogical to consider TSperm while ignoring abstinence interval,
because for 70 years it has been known that abstinence interval
influences quantitative characteristics of semen
(Hotchkiss, 1941;
MacLeod and Gold, 1956).
Perhaps underappreciated is the early conclusion that sperm accumulate in a
linear manner for only 2–3 days of abstinence
(MacLeod and Gold, 1952b).
Also, the differing effects of 36 hours or less vs 72 hours or more of
abstinence on TSperm are not generally recognized. Reasons for these divergent
effects are included in this section and highlighted in "Select a
Restrictive Abstinence Interval."
Population-based studies of TSperm as a function of abstinence interval are reviewed in the Supplemental Material, Section 3.b. The first meaningful planned study of within-subject effects of abstinence interval on TSperm involved 18 medical students and 29 patients (most oligozoospermic). Each provided samples after 3, 6, and 10 days abstinence (MacLeod and Gold, 1952a,b). For the respective abstinence intervals, TSperm averaged 393, 614, and 834 x 106 for students and 185, 293, and 318 x 106 for patients. Differences between days 3 and 6 or 6 and 10 averaged 221 and 220 x 106 sperm for students and 108 and 25 x 106 sperm for patients; all differences except the last were statistically significant. At least in this study, it seems that TSperm might increase for 7 days or more of abstinence.
The important question, however, is, what is the rate of increase in
available sperm over time after the last emission/ejaculation? Based on the
above TSperm values, during the first 3 days of abstinence, accumulation rate
might have averaged 5.5 x 106 sperm/h (ie, 393 x
106/72) vs 3.1 and 2.3 x 106 sperm/h during days
4–6 and 7–10 [ie, (614 – 393)/72 and (834 – 614)/96].
For oligozoospermic patients the respective values were 2.6, 1.5, and 0.3
x 106 sperm/h. Clearly, for both students and patients, after
3 days the rate at which additional sperm accumulated in the excurrent
duct system, as reflected in TSperm/h, declined. This study was confirmed by
others (Supplemental Material, Sections 3.b and 3.d).
Apparently there is no report wherein individuals provided replicate
samples for each of several preplanned abstinence intervals, with each
abstinence interval preceded by a single ejaculation 3–4 days earlier.
However, a hint of what might be found is provided by 2 studies reporting
multiple samples per subject, but not at stipulated abstinence intervals.
First, Schwartz et al (1979)
reported on 36 men who provided 5 or more ejaculates after abstinence
intervals of up to 6 days. Only 1 or 2 different abstinence intervals were
reported by 30% of subjects, and 61% of individuals provided only 5 or 6
samples. The slope for TSperm vs abstinence interval for each man was
calculated; no slope departed from linearity. They also reported mean TSperm
as a function of abstinence interval. Inspection of their
Figure 2c suggests that
accumulation rate was linear for 4 days (4.9 x 106
sperm/h), but was much slower after 5–6 days of abstinence. Because
abstinence interval accounted for only 29% of the variation in TSperm for a
given individual, Schwartz et al
(1979) concluded that most of
the variation in TSperm was "biological" in origin.
Second, a retrospective study of data for TSperm in 18–20 samples from each of 48 semen donors (Amann and Chapman, 2009) allowed meaningful analyses of intraindividual variation in TSperm. Individual values for TSperm vs abstinence interval for some semen donors in Figure 1 of Amann and Chapman (2009) show that TSperm often ranged widely after a given abstinence interval. The best perspective of the change in TSperm as a function of abstinence interval was provided by pooling across donors (using a mixed-model analysis). It was found that mean log10-TSperm at 48 hours of abstinence was not significantly different from means for 36 or 60 hours' abstinence (Figure 2 in Amann and Chapman, 2009). Also, log10-TSperm did not differ significantly after 72, 84, 96, or 120 hours of abstinence. From this study and Schwartz et al (1979), it is evident that after several days of abstinence, sperm representing at least 50% of the accumulation rate over the first 3 days of abstinence are "missing" in samples obtained after 4–6 days of abstinence, assuming a constant supply into the epididymis (ie, daily sperm production). Presumably the missing sperm are lost in urine (Supplemental Material, Section 2.c).
Integrating Information on TSperm and Epididymal Sperm Reserves— How do data on TSperm reconcile with number of sperm in the paired caudae epididymides and vasa deferentia, the immediate source for sperm in an ejaculate? The conclusion that on average TSperm increases in a linear manner through 2.5–3.5 days of abstinence (Amann and Chapman, 2009) seemingly is consistent with data on number of sperm in an epididymis in the Table. Information on transit time of sperm through the epididymis (Figure 1) suggests that a half-empty epididymis should refill in less than 3 days for virtually all men if the attached testis is producing more than 60 x 106 sperm/d. This calculation ignores sperm in the vasa deferentia and ampullae, possibly relatively few during sexual abstinence. Further, if TSperm reflects some proportion of the number of sperm available in the excurrent ducts, then after 3–4 days of abstinence sperm must be moved out of the excurrent ducts and voided with urine in increasing numbers. This has not been convincingly demonstrated for humans, but has been for several other species (Supplemental Material, Section 2.c).
TSperm/H More Informative Than TSperm to Look at Spermatogenesis— For meaningful comparisons of spermatogenesis among individuals, based on semen, a rate measure is required. Calculation of TSperm/h provides necessary rate information, even though it does not correct for most of the differences in TSperm for an individual after different abstinence intervals. Indeed, abstinence interval accounted for only 21% or 29% of the variation in TSperm within a given individual (Schwartz et al, 1979; Amann and Chapman, 2009). Sources of unaccounted-for variation might include fallacious reporting of abstinence interval, excessively long abstinence interval relative to sperm production, or other factors discussed below.
What does the literature teach about TSperm/h or per day? For 48 donors studied by Amann and Chapman (2009), the estimate was 5.4 x 106 sperm/h. Other values (calculated from published values for TSperm in population studies) for abstinence intervals of 1–3 days include 5.5, 5.5, and 4.9 x 106 sperm/h (Handelsman et al, 1984; MacLeod and Gold, 1952b; Schwartz et al, 1979) and possibly values of 4.4, 3.9, or 3.6 x 106 sperm/h (Freund, 1962; Tyler et al, 1985; Eliasson, 2003). However, these values are higher than the 1.8 x 106 sperm/h after 2–3 days' abstinence reported by Elzanaty et al (2005). Other values for TSperm/h are low (see Supplemental Material, Section 3.d), but some were based on a longer abstinence interval. Loss of sperm in the collection cup, or during transfer to a measuring device, was not accounted for in these studies. With a nonstandard procedure allowing enumeration of all sperm within a collection cup, Johnson (1982) reported a value of 6.9 x 106 sperm/h for samples collected at 24-hour intervals.
Considerations in Use of TSperm/H to Evaluate Spermatogenesis
Stabilize Number of Sperm in the Excurrent Ducts—
In early efforts to estimate sperm production in animals on the basis of
ejaculated semen, the need to account for changes in the extragonadal sperm
reserves between the start and end of a series of experimental ejaculations
was recognized (Ortavant,
1958). This led to the concept that number of sperm in the
excurrent ducts should be stabilized by appropriately spaced ejaculations
before collection of samples to be used to evaluate spermatogenesis as daily
sperm output (Amann, 1970).
Although the goal of providing a uniform starting point is unattainable, an
attempt to stabilize number of sperm in the excurrent ducts via preliminary
ejaculates was included in classic studies of humans
(Freund, 1963; Supplemental
Material, Section 4.a) and still merits consideration.
In the 1980s, direct measurement of numbers of sperm in the human epididymis revealed that relatively few sperm are present in the excurrent duct system (Table; "Number of Sperm in the Excurrent Ducts") and that the ratio between daily sperm production and extragonadal sperm reserves is far narrower in humans than in common mammals (evident in Table). An epididymis in a man producing more than 60 x 106 sperm/d rarely contains more than 5 days' production of sperm (Figure 1).
Johnson (1982) suggested
that 1 masturbation sample on each of 2 successive days was sufficient to
stabilize extragonadal sperm reserves, based on data for 3 individuals. Tyler
et al (1985) studied 14 men
(Supplemental Figure) and reported the same conclusion. However, inspection of
their tabulated data suggested that for 5 of the 14 subjects, 3 rather than 2
daily ejaculates were required to stabilize extragonadal sperm reserves. On
the other hand, Freund (1962)
concluded that 1 ejaculation every other day removed all accumulating sperm.
Stabilization of extragonadal sperm reserves by provision of 1 sample every
48 hours would be less demanding than 1 sample each day, and, for reasons
given in "Select a Restrictive Abstinence Interval," should give
comparable data for most men.
|
|
48-hour abstinence interval
between prestudy samples is recommended. Select a Restrictive Abstinence Interval— When evaluating quantitative success of spermatogenesis, the stipulated abstinence interval should be shorter than the time required for replenishing sperm reserves in the epididymides (Figures 1 and 3, Table), vasa deferentia, and ampullae in the majority of subjects. This interval depends on: 1) their normal storage capacity; 2) number of sperm removed in the most recent emission(s); and 3) daily sperm production of the attached testes. Unfortunately, all 3 variables range substantially among individuals and there is no accurate method to directly measure 1) or 3) in a living man.
Ejaculation-induced transit of sperm results in an atypically high value
for TSperm in a sample provided 24 hours after a previous ejaculation. As
shown in Figure 2 of Amann and Chapman
(2009), geometric mean TSperm
after the first 24 hours of abstinence was higher than expected on the basis
of values after 48, 60, or 72 hours' abstinence. Apparently, 80 x
106 or even 100 x 106 "extra sperm"
move distally in the excurrent ducts during the first 0–18 hours after
an ejaculation (Supplemental Material, Section 3.b). Consequently, mean
TSperm/h after 24 hours of abstinence was 6.9 x 106, but was
6.0, 4.8, 4.6, and 4.6 x 106 sperm/h after 36, 48, 60, or 72
hours' abstinence (data from Amann and
Chapman, 2009). Hence, during successive 12-hour intervals the
impact of rapid sperm transit was attenuated and the distortion from these
extra sperm can be overcome by calculating TSperm/h for samples obtained after
42–64 hours of abstinence.
The possibility of fallacious conclusions when abstinence interval is
outside the recommended 42–54 hours, or acceptable 42–64 hours, is
illustrated in Figure 4. Most
nonoligozoospermic men would fall somewhere between plots A and C, as
evidenced by studies summarized in "TSperm and Abstinence
Interval" and immediately below. Also, the range in accumulation rates
for the 48 individuals summarized in the Donors plot covers the area between
plot A and below plot C. With abstinence near 80 hours, 50% of individuals in
such nonoligozoospermic populations could not be distinguished one from
another because TSperm would have maximized 0–40 hours earlier. Hence,
TSperm/h would be incorrectly low (as in row 1 in
Figure 3). Restriction of
abstinence interval to 42–54 hours (hatched area in
Figure 4) would allow correct
measurement of TSperm and TSperm/h for most men. However, 10%–15% of men
might have a more rapid sperm accumulation rate than represented by plot A
(based on Johnson and Varner,
1988). Shortening abstinence interval below 40 hours to avoid the
plateau in total sperm for a few individuals is not recommended because of the
transitory impact of postejaculatory sperm relocation. After 80 hours,
for possibly 50% of nonoligozoospermic individuals, any incremental increase
in number of sperm ejaculated is progressively smaller and approaches
zero.
In epidemiologic reports, it was recognized that the increase in TSperm slowed after several days of abstinence. Commonly a series of linear-spline functions was included in a multivariate analysis to standardize values for TSperm to 96 hours (detail in Section 3.d, Supplemental Material). Use of several linear splines probably correctly modeled TSperm data from such a study, but by including long abstinence intervals seemingly ignored the underlying biology ("Background on the Excurrent Ducts" herein and Supplemental Material, Sections 2–4).
Nonexclusion of samples obtained after longer than 64 hours of
abstinence could result in underestimation of differences in spermatogenesis
between 2 or more populations. Iwamoto et al
(2006) reported that Japanese
men had lower TSperm than men in 4 European cities (P < .02);
differences were 17%, 18%, 30%, and 45%. However, differences in sperm
production might be greater than suggested by the values for TSperm. Median
abstinence interval was 64–96 hours for men in the 4 European cities,
but 134 hours for Japanese men (detail in Supplemental Material, Section 3.d).
Abstinence interval was less than 64 hours for only 5% of Japanese men. For
European men TSperm did not increase after 4 days abstinence, whereas for
Japanese men TSperm increased through 9 days of abstinence. It appears that
more European men might have lost sperm via urine sooner after a previous
nonstudy ejaculation than Japanese men. If abstinence interval had been
similar at 42–60 hours for all populations, differences in TSperm might
have been far greater than reported. This general problem is common.
Some might opine that censoring samples provided after 64 hours of
abstinence is too extreme, because when TSperm/h is estimated from information
in epidemiologic reports some population values are less than or equal to 2.5
x 106 sperm/h (Section 3.d, Supplemental Material). For such
men, sperm might accumulate for 140 hours before spilling out (plot C in
Figure 4 represents 2.9 x
106 sperm/h). However, for other publications values are between
3.6 and 5.5 x 106 sperm/h ("TSperm/h More Informative
Than TSperm to Look at Spermatogenesis"). Further, in any population
TSperm/h for 50% of individuals is above the population estimate. For
example, during the winter in populations representing 4 European cities
(Jørgensen et al,
2001), the upper 95% CLs range up to 5.8 x 106
sperm/h (similar to plot for Donors in
Figure 4). Hence, to insure a
meaningful TSperm/h for most individuals and leeway because the ratio of
available sperm to daily sperm production might be narrower than depicted in
Figure 4, a cutoff at 64 hours
of abstinence is appropriate.
For an initial clinical evaluation or an epidemiologic study, one would rather have an abstinence interval that allows for incomplete refilling of the excurrent ducts at a rate reflecting daily sperm production, than spillage of sperm out in urine from many men. The exception to this recommendation is when a clinician has detected oligozoospermia and then wishes to evaluate number of sperm in ejaculates after 6–7 days abstinence.
Effects of Sexual Arousal and Method for Sample Collection on TSperm
Emission and ejaculation are reflex actions mediated in the spinal cord,
although emission also involves cerebral inputs
(Newman et al, 1982). Given
neuronal inputs involved in emission, it is plausible that prolonged foreplay
would increase TSperm in a man's ejaculate. However, this is hard to prove,
especially when only 2 samples per individual are compared, because if TSperm
in a first sample is relatively high or low there is a tendency for TSperm in
the second sample to be closer to the true value for that individual
(Baker and Kovacs, 1985).
Sexual Arousal— Over 50 years ago it was demonstrated that sexual arousal can substantially increase the number of sperm in a given ejaculate collected by artificial vagina from bulls (Collins et al, 1951; Hale and Almquist, 1960). Thereafter, increasing and/or prolonging sexual arousal (termed sexual preparation) became a standard procedure when collecting semen from boars, bulls, or rabbits.
Speculation that a man's sexual arousal is greater with intercourse than during masturbation is common. However, arousal is hard to quantify and might not be measured by self-perceived sexual satisfaction scores recorded on a 4- or 10-point scale, and scores might be influenced by instructions given to subjects. In 2 studies (Zavos and Goodpasture, 1989; Sofikitis and Miyagawa, 1993), sexual satisfaction scores were higher after providing samples by intercourse, wearing a silicone collection device, than after masturbation. Details are in the Supplemental Material, Section 5.a.
The practical question, however, is whether use of special erotic materials consistently increases TSperm in masturbation samples. As detailed in the Supplemental Material, van Roijen et al (1996) directly examined effect of sexual arousal on quantitative attributes of semen obtained by masturbation. Based on 3 studies using different paradigms, they concluded that TSperm was independent of sexual arousal during masturbation, although provision of an erotic video increased sexual arousal and ease with which a sample was produced.
Pound et al (2002) studied within-individual benefit of sexual arousal on TSperm by having semen donors watch a sexually explicit video before/while obtaining a sample after 3 days of abstinence. Time spent in the private room was recorded and TSperm was measured. The study involved 292 samples from 25 donors. On a within-donor basis, TSperm was not affected (P = .07) by time spent in the room for samples produced in 30 minutes or less. Nevertheless, it might be appropriate to consider time spent producing the sample as a covariate in population studies.
Factors that might contribute to psychological stress should be minimized, and this might be facilitated by having men provide masturbation samples at home rather than in a clinic. Elzanaty and Malm (2008) reported on consecutive patients undergoing fertility assessment; 106 men provided a home sample and 273 a clinic sample (only 1 sample per male). TSperm was greater for samples provided at home (270 vs 175 x 106 sperm; P < .05). It is not clear whether this was because of reduced stress, more effective foreplay, or other factors.
In summary, there is no persuasive evidence that sexual arousal increases TSperm, even though it might reduce interval to ejaculation via masturbation. Nevertheless, a facility should provide appropriate instructions, and request use of visual and auditory stimulation to maximize sexual arousal before and during masturbation.
Masturbation vs Intercourse— Freund (1962) sought an answer to the question of whether a masturbation sample is representative of a sample produced at intercourse. It is evident that the answer is not straight forward and probably depends on details of the study protocol. It is hard to conduct a study where sexual arousal is not a potential confounding factor. As detailed in Section 5.b of the Supplemental Material, there has been negligible effort to control this variable or to recover residual sperm from the collection device or jar and include these residual sperm in reported TSperm.
Freund (1962) and Purvis et al (1986) both found that average values for TSperm were not different for samples obtained by masturbation or intercourse. Together, their studies involved 19 medical students and 7 patients. An opposite conclusion was reported by Zavos and Goodpasture (1989), based on single samples by each method from 35 nonoligozoospermic men. Intercourse using a silicone collection device after prolonged foreplay on average provided a 39% greater TSperm than masturbation at home. None of these reports mentions special effort to provide visual or auditory stimulation during masturbation.
The only study including within-subject/method replication involved 38 infertility patients (Sofikitis and Miyagawa, 1993). Each patient provide 6 masturbation samples after 48–60 hours' abstinence and then 6 samples collected in a silicone collection device during coitus at similar intervals. Variation in TSperm within method/individual was not presented. Averaged across subjects, TSperm was greater for samples collected during intercourse than by masturbation (99 vs 44 x 106 sperm; P < .01). They concluded that samples collected by masturbation might lead to diagnostic mistakes, because samples were not representative of what might be provided to a female partner during intercourse.
Results in some of the above studies might be confounded by variable sexual arousal and sexual stimulation before and during emission. The studies also suffer from failure to recover and enumerate all sperm deposited in the jar or condom. For condom samples, sperm adhering to the device (ie, residual sperm) usually are not included in the total, but they might represent 6%–10% of the number aspirated (Zavos and Goodpasture, 1989). Also, occasionally some sperm might have missed the container.
It seems unlikely that masturbation samples typically will provide a TSperm exceeding that for coitus samples. Although the seemingly greater TSperm in intercourse samples might be consequent to prolonged foreplay, as discussed in the previous section, sexual arousal seems to have no consistent effect on TSperm in masturbation samples. Perhaps clinicians or epidemiologist-andrologist teams should consider the use of modern silicone collection devices to obtain samples for evaluation of TSperm and sperm quality, as suggested by Sofikitis and Miyagawa (1993).
A Meaningful Value for TSperm or TSperm/H
Obtaining Meaningful Individual Values—
Both biologically and analytically correct individual values for TSperm and
TSperm/h are crucial for validity of clinical or epidemiological data. To
correctly portray the quantitative aspect of spermatogenesis on the basis of
TSperm/h (Figure 4), one should
request an abstinence interval of 42–54 hours (ie, 6 hours around 48
hours) with a range of 42–60 hours to provide more leeway
(Amann and Chapman, 2009). This
abstinence interval enhances detection of men producing a reduced number of
sperm, because in many nonoligozoospermic men the number of sperm ejaculated
per hour of abstinence might be artifactually low if abstinence interval is
more than 3 days. Sperm can accumulate longer in individuals producing few
sperm than in individuals producing many sperm. Any sample provided with an
abstinence interval less than or equal to 36 hours or more than 64 hours
should be excluded from a study.
Although 42–60 hours is more restrictive than often suggested, this
abstinence interval enhances the likelihood of correctly describing an
individual as producing a high, moderate, or low number of sperm daily. Also,
42–60 hours' abstinence would not disrupt the 3x weekly coital
frequency of many married couples (Sherins
et al, 1977). Note that a few men might accumulate sperm, as
occasionally happens in stallions (Pickett
and Voss, 1975), so that the epididymides and ampullae contain an
unexpected number. For example, Johnson
(1982) reported single
instances in which a man ejaculated 4.0 x 109 sperm after 13
days of abstinence or in which 300 x 106 sperm were
voided in urine in less than 2 days.
Carryover effects can alter observed TSperm, as discussed in conjunction with Figure 3. A clinician or epidemiologist-andrologist team should recognize that stipulating an abstinence interval for the sample(s) evaluated (eg, 42–54 hours) is insufficient. At least 2, and in some cases 3, preliminary ejaculations at the stipulated abstinence interval should proceed the first sample to be evaluated. Requesting the stipulated abstinence interval for at least 2 preliminary ejaculations is especially important when a patient or some study subjects might have a relatively low daily sperm production (eg, bottom plot in Figure 3).
The preliminary ejaculations could be at home and not evaluated, followed without interruption by sample(s) to be evaluated. Request that intervening ejaculations be avoided. Accurate information on abstinence interval (in hours) is important and an error greater than or equal to 5 hours, longer or shorter, will affect calculated TSperm/h by more than 10%. An epidemiologist-andrologist team might automatically reject the concept of 2 preliminary samples (not evaluated) at the stipulated abstinence interval, but if so they should reconsider the importance of having meaningful data as the basis for their conclusions.
Number of Samples per Individual—
Because of within-individual variation in TSperm, Amann and Chapman
(2009) concluded that a single
sample could not provide a meaningful evaluation of an individual's TSperm or
TSperm/h. In 25% of cases, the observed value would be more than 16%
below the individual's true value, and in 25% of cases the observed value
would be more than 30% above the true value. For most uses, they recommended
calculation of an individual's mean TSperm/h based on 3 samples. The reason
for this recommendation should be obvious from
Figure 2; the associated text
notes that the 90% CLs are twice as wide for 1 sample as for 3 samples per
mean. Only for special research, such as to determine if a drug has a moderate
but real effect on sperm production, might one consider evaluation of 6 or
more samples for each study block and calculation of daily sperm output. A
review of pertinent literature and an appropriate protocol to obtain daily
sperm output data are in the Supplemental Material, Sections 6.d and 7.
Detection of Differences Among Individuals— Two premises underlying acquisition of data on TSperm or TSperm/h from patients or a series of subjects in a research study are that: 1) there will be differences among men; and 2) detection of biologically meaningful differences is possible. How small a difference can be detected? In part this is determined by the precision of the value or mean value for each individual and also the range in mean TSperm or TSperm/h among individuals. This important and neglected topic was addressed in Supplemental Table 3 of Amann and Chapman (2009). Interest to look there might be stimulated by a quotation from the associated text: "When only 1 sample per individual is available, an individual with TSperm of 100 x 106 would not be clearly separated from an individual with TSperm as high as 250 x 106 based on their CLs, but would be separated from one with 300 x 106." With 3 samples per individual, 2 individuals with TSperm values of 50 and 100, 100 and 200, or 200 and 300 x 106 would be separated.
Obviously, a research study will use many individuals. In contrast to some diagnostic measures of body function, the CV for TSperm or TSperm/h for typical individuals is not substantially less than the among-individual CV except when the study population includes more than a few oligozoospermic men. For the data set of Amann and Chapman (2009), mean TSperm/h (based on 18–20 samples) for the 48 seminal donors ranged from 2.4 to 10.0 x 106 sperm/h, which is equivalent to 115–480 x 106 sperm per ejaculate when abstinence interval is 48 hours. Within- and among-donor CVs based on log10-transformed data were 24% and 26%.
Take-Home Message
To evaluate spermatogenesis on the basis of TSperm/h, the approach outlined
below is recommended. It is based on information in this review and the
associated Supplemental Material. A clinician can see how her/his standard
procedure measures up to what is suggested. An epidemiologist-andrologist team
planning a large study can see what should be done (or rejected with adequate
justification included in the study plan). The outline is aimed at providing
meaningful data on TSperm/h with minimal demand on patients or subjects.
Epidemiologist-andrologist teams might conclude (based on thoughtful analysis rather than a preconceived decision) that the best approach was use of 1 rather than 2 prestudy samples and calculation of mean TSperm/h from data for 2 samples after 42–60 hours' abstinence because more potential study subjects would agree to that plan and comply with stipulations. That would be far better than the common practice of a single sample after 36–168 hours' abstinence, and standardization of TSperm to some abstinence interval (eg, 96 hours), which would result in loss of discrimination for reasons discussed under "Select a Restrictive Abstinence Interval."
References
Aggerholm AS, Thulstrup AM, Toft G, Ramlau-Hansen CH, Bonde JP. Is overweight a risk factor for reduced semen quality and altered serum sex hormone profile? Fertil Steril. 2008; 90: 619 –626.[CrossRef][Medline]
Amann RP. Sperm production rates. In: Johnson AD, Gomes WR, VanDemark NL, eds. The Testis. Vol 1 . New York, NY: Academic Press; 1970: 433 –482.
Amann RP. A critical review of methods for evaluation of spermatogenesis from seminal characteristics. J Androl. 1981;2: 37 –58.
Amann RP. Reproductive physiology and endocrinology of the dog. In: Morrow DA, ed. Current Therapy in Theriogenology. 2nd ed. Philadelphia, PA: WB Saunders Co; 1986: 532 –538.
Amann RP. The cycle of the seminiferous epithelium in humans: a
need to revisit? J Androl. 2008; 29: 469
–487.
Amann RP. Evaluating spermatogenesis using semen: the biology of
emission tells why reporting total sperm per sample is important, and why
reporting only number of sperm per milliliter is irrational. J
Androl. 2009;30: 623
–625.
Amann RP. Tests to measure quality of sperm at spermiation. Asian J Androl. In press.
Amann RP, Chapman PL. Total sperm per ejaculate of men: obtaining a
meaningful value or a mean value with appropriate precision. J
Androl. 2009;30: 642
–649.
Amann RP, Hammerstedt RH, Veeramachaneni DNR. The epididymis and sperm maturation: a perspective. Reprod Fertil Dev. 1993; 5: 361 –381.[CrossRef][Medline]
Amann RP, Howards SS. Daily spermatozoal production and epididymal spermatozoal reserves of the human male. J Urol. 1980; 124: 211 –215.[Medline]
Amann RP, Johnson L, Thompson DL Jr, Pickett BW. Daily spermatozoal production, epididymal spermatozoal reserves and transit time of spermatozoa through the epididymis of the rhesus monkey. Biol Reprod. 1976;15: 586 –591.[Abstract]
Amann RP, Kavanaugh JF, Griel LC Jr, Voglmayr JK. Sperm production
of Holstein bulls determined from testicular spermatid reserves, after
cannulation of the rete testis or vas deferens, and by daily ejaculation.
J Dairy Sci. 1974; 57: 93
–99.
Aneck-Hahn NH, Schulenburg GW, Bornman MS, Farias P, de Jager C.
Impaired semen quality associated with environmental DDT exposure in young men
living in a malaria area in the Limpopo Province, South Africa. J
Androl. 2007;28: 423
–434.
Baker HWG, Burger HG, de Kretser DM, Lording DW, McGowan P, Rennie GC. Factors affecting the variability of semen analysis results in infertile men. Int J Androl. 1981; 4: 609 –622.[Medline]
Baker HWG, Kovacs GT. Spontaneous improvement in semen quality: regression towards the mean. Int J Androl. 1985; 8: 421 –426.[Medline]
Barratt CL, Cooke ID. Sperm loss in the urine of sexually rested men. Int J Androl. 1988; 11: 201 –207.[CrossRef][Medline]
Baska KM, Manandhar G, Feng D, Agca Y, Tengowski MW, Sutovsky M, Yi Y-J, Sutovsky P. Mechanism of extracellular ubiquitination in the mammalian epididymis. J Cell Physiol. 2008; 215: 684 –696.[CrossRef][Medline]
Bedford JM. Adaptations of the male reproductive tract and the fate of spermatozoa following vasectomy in the rabbit, rhesus monkey, hamster and rat. Biol Reprod. 1976; 14: 118 –142.[Abstract]
Berndtson WE. Comparative reliability and sensitivity of different methods for assessing treatment effects on sperm production. Anim Reprod Sci. 2008;105: 5 –22.[Medline]
Cayli S, Sakkas D, Vigue L, Demir R, Huszar G. Cellular maturity
and apoptosis in human sperm: creatine kinase, caspase-3 and Bcl-XL
levels in mature and diminished maturity sperm. Mol Hum
Reprod. 2004;10: 365
–372.
Collins WJ, Bratton RW, Henderson CR. The relationship of semen
production to sexual excitement of dairy bulls. J Dairy
Sci. 1951;34: 224
–227.
Cooper TG, Yeung C-H, Jones R, Orgebin-Crist M-C, Robaire B.
Rebuttal of a role for the epididymis in sperm quality control by phagocytosis
of defective sperm. J Cell Sci. 2002; 115: 5
.
Eliasson R. Basic semen analysis. In: Matson P, ed. Current Topics in Andrology. Perth, Australia: Ladybrook Publishing; 2003: 35 –89.
Elzanaty S, Malm J. Comparison of semen parameters in samples collected by masturbation at a clinic and at home. Fertil Steril. 2008;89: 1718 –1722.[CrossRef][Medline]
Elzanaty S, Malm J, Giwercman A. Duration of sexual abstinence:
epididymidal and accessory sex gland secretions and their relationship to
sperm motility. Hum Reprod. 2005; 20: 221
–225.
Freund M. Interrelationships among the characteristics of human
semen and factors affecting semen-specimen quality. J Reprod
Fertil. 1962;4: 143
–159.
Freund M. Effect of frequency of emission on semen output and an
estimate of daily sperm production in man. J Reprod
Fertil. 1963;6: 269
–286.
Gao J, Gao ES, Yang Q, Walker M, Wu JQ, Zhou WJ, Wen SW. Semen
quality in a residential, geographic and age representative sample of healthy
Chinese men. Hum Reprod. 2007; 22: 477
–484.
Hale EB, Almquist JO. Relationship of sexual behavior to germ cell output in farm animals. J Dairy Sci. 1960; 43(suppl): 145 –169.
Handelsman DJ, Conway AJ, Boylan LM, Turtle JR. Testicular function in potential sperm donors: normal ranges and the effects of smoking and varicocele. Int J Androl. 1984; 7: 369 –382.[Medline]
Hotchkiss RS. Factors in stability and variability of semen specimens: observations on 640 successive samples from 23 men. J Urol. 1941;45: 875 –888.
Huszar G, Patrizio P, Vigue L, Willets M, Wilker C, Adhoot D,
Johnson L. Cytoplasmic extrusion and the switch from creatine kinase B to M
isoform are completed by the commencement of epididymal transport in human and
stallion spermatozoa. J Androl. 1998; 19: 11
–20.
Iwamoto T, Nozawa S, Yoshiike M, Hoshino T, Baba K, Matsushita T,
Tanaka SN, Naka M, Skakkebæk NE, Jørgensen N. Semen quality of
324 fertile Japanese men. Hum Reprod. 2006; 21: 760
–765.
Johnson L. A reevaluation of daily sperm output of men. Fertil Steril. 1982; 37: 811 –816.[Medline]
Johnson L, Varner DD. Effect of daily sperm production but not age on transit time of spermatozoa through the human epididymidis. Biol Reprod. 1988;39: 812 –817.[Abstract]
Jørgensen N, Andersen A-G, Eustache F, Irvine DS, Suominen
J, Petersen JH, Andersen AN, Auger J, Cawood EHH, Horet A, Jensen TK, Jouannet
P, Keiding N, Vierula M, Toppari J, Skakkebæk NE. Regional differences
in semen quality in Europe. Hum Reprod. 2001; 16: 1012
–1019.
Judd JE, Berndtson WE, Castro ACS. Extragonadal sperm reserves,
sperm-depletion rates, numbers of sperm per mating, and fertility with
successive matings by intact or unilateral vasectomized rats. J
Androl. 1997;18: 698
–707.
Kort HI, Massey JB, Elsner CW, Mitchell-Leef D, Shapiro DB, Witt
MA, Roudebush WE. Impact of body mass index values on sperm quantity and
quality. J Androl. 2006; 27: 450
–452.
MacLeod J, Gold RZ. The kinetics of human spermatogenesis as revealed by changes in the ejaculate. Ann N Y Acad Sci. 1952a;55: 707 –724.[CrossRef][Medline]
MacLeod J, Gold RZ. The male factor in fertility and infertility. V. Effect of continence on semen quality. Fertil Steril. 1952b;3: 297 –315.[Medline]
MacLeod J, Gold RZ. The male factor in fertility and infertility. VIII. A study of variation in semen quality. Fertil Steril. 1956;7: 387 –410.[Medline]
Misell LM, Holochwost D, Boban D, Santi N, Shefi S, Hellerstein MK, Turek PJ. A stable isotope-mass spectrometric method for measuring human spermatogenesis kinetics in vivo. J Urol. 2006; 175: 242 –246.[CrossRef][Medline]
Mitsuya H, Asai J, Suyama K, Ushida T, Hosoe K. Application of X-ray cinematography in urology: I. Mechanism of ejaculation. J Urol. 1960; 86 –92.
Newman HF, Reiss H, Northup JD. Physical basis of emission, ejaculation, and orgasm in the male. Urology. 1982; 19: 341 –350.[CrossRef][Medline]
Orgebin-Crist MC. Recherches expérimentales sur la durée de passage des spermatozoides dans l'épididyme du Taureau. Ann Biol Anim Bioch Biophys. 1962; 2: 51 –108.[CrossRef]
Ortavant R. Le Cycle Spermatogénétique chez le Bélier [DSc thesis]. Paris, France; University of Paris; 1958.
Pickett BW, Voss JL. Abnormalities of mating behavior in domestic stallions. J Reprod Fertil Suppl. 1975; 23: 129 –134.[Medline]
Pound N, Javed MH, Ruberto C, Shaihk MA, Del Valle AP. Duration of sexual arousal predicts semen parameters for masturbatory ejaculates. Physiol Behav. 2002; 76: 685 –689.[CrossRef][Medline]
Purvis K, Magnus Ø, Mørkås L, Åbyholm T, Rui H. Ejaculate composition after masturbation and coitus in the human male. Int J Androl. 1986; 9: 401 –406.[Medline]
Ramlau-Hansen CH, Nohr EA, Thulstrup AM, Bonde JP, Storgaard L,
Olsen J. Is maternal obesity related to semen quality in the male offspring? A
pilot study. Hum Reprod. 2007; 22: 2758
–2762.
Ramlau-Hansen CH, Thulstrup AM, Bonde JP, Olsen J, Bech BH. Semen
quality according to prenatal coffee and present caffeine exposure: two
decades of follow-up of a pregnancy cohort. Hum
Reprod. 2008;23: 2799
–2805.
Robaire B, Hinton BT, Orgebin-Crist MC. The epididymis. In: Neill JD, ed. Knobil and Neill's Physiology of Reproduction. 3rd ed. New York, NY: Elsevier; 2006: 1071 –1148.
Rowley MJ, Teshima F, Heller CG. Duration of transit of spermatozoa through the human male ductular system. Fertil Steril. 1970; 21: 390 –396.[Medline]
Schwartz D, Laplanche A, Jouannet P, David G. Within-subject
variability of human semen in regard to sperm count, volume, total number of
spermatozoa and length of abstinence. J Reprod Fertil. 1979; 57: 391
–395.
Sharpe RM. Pathways of endocrine disruption during male sexual differentiation and masculinization. Best Pract Res Clin Endocrinol Metab. 2006;20: 91 –110.[CrossRef][Medline]
Sherins RJ, Brightwell D, Sternthal PM. Longitudinal analysis of semen of fertile and infertile men. In: Troen P, Nankin HR, eds. The Testis in Normal and Infertile Men. New York, NY: Raven Press; 1977: 473 –488.
Sigman M, Boyle K, Jarow JP. Prevalence of sperm in the postejaculatory urine of fertile and subfertile men. Urology. 2008;71: 110 –112.[CrossRef][Medline]
Skakkebæk NE, Rajpert-De Meyts E, Jørgensen N, Carlsen E, Petersen P, Giwercman A, Andersen A-G, Jensen TK, Andersen A-M, Müller J. Germ cell cancer and disorders of spermatogenesis: an environmental connection? APMIS. 1998; 106: 3 –12.[Medline]
Sofikitis NV, Miyagawa I. Endocrinological, biophysical, and
biochemical parameters of semen collected via masturbation versus sexual
intercourse. J Androl. 1993; 14: 366
–373.
Sutovsky P, Hauser R, Sutovsky M. Increased levels of ubiquitin
correlate with semen quality in men from an andrology laboratory clinic
population. Hum Reprod. 2004; 19: 628
–638.
Sutovsky P, Moreno R, Ramalho-Santos J, Dominko T, Thompson WE, Schatten G. A putative, ubiquitin-dependent mechanism for the recognition and elimination of defective spermatozoa in the mammalian epididymis. J Cell Sci. 2001;114: 1665 –1675.[Abstract]
Sutovsky P, Moreno RD, Ramalho-Santos J, Dominko T, Thompson W. Reply. J Cell Sci. 2002; 115: 6 –7.
Swan SH, Brazil C, Drobnis EZ, Liu F, Kruse RL, Hatch M, Redmon JB, Wang C, Overstreet JW. Geographic differences in semen quality of fertile U.S. males. Environ Health Perspect. 2003; 111: 414 –420.[Medline]
Swan SH, Liu F, Overstreet JW, Brazil C, Skakkebæk NE. Semen
quality of fertile US males in relation to their mother's beef consumption
during pregnancy. Hum Reprod. 2007; 22: 1497
–1502.
Turner TT. De Graaf's thread: the human epididymis. J
Androl. 2008;29: 237
–250.
Tyler JPP, Matson PL, Crockett NG. The effects of frequent ejaculation on seminal spermatozoal number and calculation of daily spermatozoal output. Clin Reprod Fertil. 1985; 3: 145 –149.[Medline]
van Roijen JH, Slob AK, Gianotten WL, Dohle GR, van der Zon ATM,
Vreeburg JTM, Weber RFA. Sexual arousal and the quality of semen produced by
masturbation. Hum Reprod. 1996; 11: 147
–151.
Zavos PM, Goodpasture JC. Clinical improvements of specific seminal deficiencies via intercourse with a seminal collection device versus masturbation. Fertil Steril. 1989; 51: 190 –193.[Medline]
This article has been cited by other articles:
|
|
 |
|
  R.P. Amann Evaluating testis function non-invasively: how epidemiologist-andrologist teams might better approach this task Hum. Reprod., January 1, 2010; 25(1): 22 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. P. Amann Evaluating Spermatogenesis Using Semen: The Biology of Emission Tells Why Reporting Total Sperm per Sample Is Important, and Why Reporting Only Number of Sperm per Milliliter Is Irrational J Androl, November 1, 2009; 30(6): 623 - 625. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
