unique symptom pro
fi
les associated with endogenously low
serum TT and free testosterone levels.
Patients and Methods
After approval by the Institutional Review Board at Baylor
College of Medicine (Houston, TX, USA), we evaluated 352
men (aged
<
40 years) seen consecutively between May 2013
and March 2014. These men presented with a chief
complaint of
‘
low testosterone
’
. Men using testosterone or
other androgenic anabolic steroids (AAS) either at, or
6 months prior to, the time of the survey were excluded.
We also excluded men who had presented with a primary
diagnosis of infertility, Kleinfelter
’
s syndrome, or secondary
hypogonadism after appropriate assessment of history and
endocrine evaluation with gonadotrophins. All men
answered the Androgen De
fi
ciency in the Aging Male
(ADAM) questionnaire [9,10] and on the same day their
testosterone levels were measured. The ADAM
questionnaire consists of 10
‘
Yes/No
’
questions about
symptoms, although not all of them necessarily indicate
treatable symptoms of testosterone de
fi
ciency. All venous
blood samples were obtained under standardised conditions
before 10:00 h from fasting patients. Serum or plasma were
separated at 800
g
. Serum TT and sex hormone binding
globulin measurements were done using the
radioimmunoassay Beckman Access II platform (Beckman
Coulter, Fullerton, CA, USA). TT levels were
fi
tted on a
scatter plot to determine in
fl
ection points and threshold
levels. Data was analysed using Microsoft Excel (Microsoft,
Redmond, WA, USA) and
SPSS
(SPSS Inc., Chicago, IL,
USA). Chi-squared tests were used to compare percentages
and the Student
’
s
t
-test was used to compare means. The
Q
–
Q test was used to verify that the variables were
continuously distributed. Univariate and multivariable
analysis was performed for age and the 10 symptoms
identi
fi
ed on the ADAM questionnaire. Variables that were
statistically signi
fi
cant on the univariate analysis were
included in the multivariable analysis. All values are
reported as the mean (
SD
) and
t
-tests were used to evaluate
differences in means between groups. A
P
≤
0.05 was
considered to indicate statistical signi
fi
cance.
Results
Of the 352 men aged
<
40 years, 210 men had a TT level of
<
300 ng/dL; 67 men had levels of 300
–
400 ng/dL; and 75
men had levels of
>
400 ng/dL. The mean (
SD
) age of the 352
men was 33.2 (4.2) years, and the mean (
SD
; range) TT level
was 308 (170; 0.86
–
1537) ng/dL. Of the 10 hypogonadal
symptoms, the probability of having
fi
ve symptoms decreased
(
P
<
0.05) in men with testosterone levels of
>
400 ng/dL
(Table 1 and Fig. 1). These
fi
ve symptoms included two
psychological (
‘
decreased energy
’
,
‘
feeling sad
’
), and three
physical (
‘
decreased strength and endurance
’
,
‘
decreased
ability to play sports
’
, and
‘
deterioration in work
performance
’
). The probabilities of having the 10
hypogonadal symptoms that are part of the ADAM
questionnaire, were similar in men with TT levels of
<
300 ng/dL and men whose levels were between 300 and
400 ng/dL (Fig. 1). On a univariate analysis, the presence of
the same
fi
ve symptoms predicted a testosterone level of
<
400 ng/dL. On multivariable analysis, only
‘
lack of energy
’
predicted a testosterone level of
<
400 ng/dL. Of note, sexual
symptoms (libido and erectile function) commonly thought to
be associated with low testosterone did not identify men with
testosterone levels of
<
400 ng/dL. In addition, none of the 10
symptoms evaluated predicted a testosterone level of
<
300 ng/dL.
Discussion
Hypogonadism is caused by insuf
fi
cient concentrations of
testosterone in the blood, resulting in symptoms of androgen
de
fi
ciency. The reference range for most assays of TT is 300
–
800 ng/dL [11,12], meaning that only 2.5% of healthy men
have concentrations of
<
300 ng/dL. A clinical threshold of
300 ng/dL is often cited in the literature as the biochemical
de
fi
nition of hypogonadism [2]. One argument for using
300 ng/dL as the threshold for diagnosing male
hypogonadism is that there is a functional correlation with
erectile dysfunction (ED). This relationship was determined
in a study of 162 elderly (mean age 64.1 years) men with ED
(mean duration 45.6 months), where a Korean group
reported that hypogonadism (serum TT level of
<
300 ng/dL)
was among the strongest independent predictors of a poor
response to sildena
fi
l25
–
100 mg for 8 weeks [13]. The
threshold TT level below which signs and symptoms of
androgen de
fi
ciency occur and testosterone replacement is
bene
fi
cial is not known and varies among individuals
depending on age and comorbid conditions, and among
affected target organs. Therefore, there is no absolute value of
the TT level below which clinical androgen de
fi
ciency or
hypogonadism can be con
fi
rmed in all patients, especially in
young men. Consequently, we evaluated the association
between hypogonadal symptoms and serum TT levels in men
who came to our outpatient men
’
s health clinic with a chief
complaint of
‘
low testosterone
’
.
We identi
fi
ed a serum TT level threshold of 400 ng/dL by
evaluating probabilities of the different hypogonadal
symptoms across various TT levels. Most previous studies
have identi
fi
ed an association of low testosterone with sexual
symptoms, such as poor erectile function [7,14,15] and low
libido [16]. However, in our present study we found that
physical and psychological symptoms were most often closely
associated with testosterone levels of
<
400 ng/dL.
Interestingly, only
‘
lack of energy
’
remained statistically
signi
fi
cant on a multivariable analysis. None of the sexual
ª
2014 The Authors
BJU International
ª
2014 BJU International
143
Hypogonadal symptoms in young men with TT
<
400 ng/dL