The AUC of AGD for each urologic diagnosis is listed
(Table
2
).
As most men with semen data were evaluated for infertility,
a majority had abnormal semen parameters as assessed by low
sperm concentration, sperm motility, total sperm count, and
total motile sperm count. Compared to fathers, childless men
had a significant shorter AGD. Moreover, men with low
sperm concentration, total sperm count, and total motile sperm
count had a shorter AGD compared to men with normal
semen parameters (Table
3
).
In order to evaluate the discriminating ability of AGD,
ROC curves were created comparing AGD and total testis
volume (Fig.
1
). The area under the curve (AUC) was signif-
icantly larger for total testis volume compared to AGD when
evaluating fertility (0.71 vs 0.63,
p
=0.02). Similarly, there
was a trend towards a higher AUC for testis volume compared
to AGD for sperm concentration and total sperm count
(Table
4
).
Next, we stratified men based on genital measure-
ment to compare the implications of short/long AGD
and small/large testes. While small testes were associat-
ed with lower sperm concentration, men with short
AGD but large testes had relatively higher sperm
counts. Sperm motility and volume did not vary based
on genital measures. Fatherhood status was also best
correlated with testis volume compared to AGD but a
trend was detected so that two abnormal genital mea-
sures led to a higher rate of childlessness compared to
only one (Table
5
).
After adjusting for age, race, body mass index, and date of
measurement, multivariable logistic regression models dem-
onstrated that testis size and AGD were significantly associ-
ated with fertility and sperm concentration (
p
<0.01). When
both genital measures were incorporated into the model
to examine the odds of oligospermia (concentration<
15 M/mL), testis size remained significant (
p
<0.01)
but AGD did not (
p
=0.09).
Discussion
AGD is associated with infertility, low sperm concentrations,
and low total sperm counts. However, there was a trend
suggesting testis volume was more reliable for an individual
man. While AGD is associated with sperm production on a
population level, at the individual level the distinction based
AGD alone cannot accurately distinguish between fertile and
subfertile men enough for clinical use. Indeed, the AUC is
below the level of 0.75 which is generally considered a fair
discriminator of a given outcome.[
19
] Moreover, on
subanalyses, testis volume appeared to be a better discrimina-
tor of male fertility compared to AGD.
During sexual development the immature genital precur-
sors migrate ventrally via an androgen mediated pathway.[
20
]
The anogenital distance has been used to sex animals, since
males have longer lengths than females.[
1
–
3
] Moreover, hu-
man studies in infants have also established that boys have
longer perineal lengths than girls.[
4
–
7
] Investigators have
used the anogenital distance as a marker for normal genital
development. In humans, girls with CAH have been shown to
have longer perineal lengths than their normal
counterparts.[
21
] Normal androgen exposure during defined
gestational time periods, have been correlated with normal
genital lengths and reproductive function in rodents.[
22
,
23
]
Thus a more masculinized phenotype would display longer
genital lengths and higher sperm counts.
The goal of the study was to assess the relationship be-
tween AGD and fertility on an individual basis in order to
determine clinical utility. To accomplish this, fertility was
defined in several ways. Fatherhood status (which does have
limitations) as well as fertility cutoffs from the WHO 5th
edition were utilized. Using different measures of male fertil-
ity, AGD failed to perform adequately to routinely guide
clinical care.
Prior literature has demonstrated that AGD is associated
with sperm production. Mendiola et al. found a positive linear
relationship between sperm concentration and AGD in college
aged men.[
9
] A similar association among an older
cohort presenting to a urology clinic was also
demonstrated.[
8
] Examining infertile men, it was shown
that a longer AGD suggests improved varicocelectomy
outcomes and a lower likelihood of spermatogenic fail-
ure in azoospermic men.[
12
,
13
] However, all prior
studies examined large cohorts of men to determine
the association between testicular function and AGD.
In the current report, we examined the predictive value
of AGD for an individual.
Table 3
Distribution of AGD based on reproductive variables.
Subfertile semen parameters defined by WHO 5th edition of the manual
on semen analyses
Characteristic
n
(%)
AGD
(mean, SD)
p
Father
No
246 (52) 36.4 (12.9)
<0.01
Yes
227 (48) 41.9 (13.5)
Semen Volume
<1.5
29 (15)
35.8 (15.0)
0.99
≥
1.5
160 (85) 35.8 (11.6)
Sperm Concentration
<15 M/mL 127 (66) 34.3 (11.5)
<0.01
≥
15 M/mL 66 (34)
39.4 (12.9)
Sperm Motility
<40 %
121 (63) 35.3 (12.4)
0.38
≥
40 %
70 (37)
36.9 (11.8)
Total Sperm Count
<39 M
121 (63) 33.6 (12.0)
<0.01
≥
39 M
70 (37)
39.7 (11.4)
Total Motile Sperm Count <9 M
109 (58) 33.8 (12.4)
<0.01
≥
9 M
79 (42)
38.4 (11.2)
J Assist Reprod Genet (2015) 32:479
–
484
481