http://std.sagepub.

com/
International Journal of STD & AIDS
http://std.sagepub.com/content/23/12/876
The online version of this article can be found at:

DOI: 10.1258/ijsa.2012.011464
2012 23: 876 Int J STD AIDS
K Sunchatawirul, K Tantiwongse, P Chathaisong, S Thongyen, N Chumpathat and W Manosuthi
Hypogonadism among HIV-infected men in Thailand

Published by:
http://www.sagepublications.com
On behalf of:

BASSH
IUSTI
can be found at: International Journal of STD & AIDS Additional services and information for

http://std.sagepub.com/cgi/alerts Email Alerts:

http://std.sagepub.com/subscriptions Subscriptions:
http://www.sagepub.com/journalsReprints.nav Reprints:

http://www.sagepub.com/journalsPermissions.nav Permissions:

What is This?

- Dec 1, 2012 Version of Record >>

, P Chathaisong MSc

, S Thongyen MPH

,
N Chumpathat MEd** and W Manosuthi MD

*Division of Urology, Department of Disease Control, Bamrasnaradura Infectious Diseases Institute, Ministry of Public Health, 126 Tiwanont Road,
Muang Nonthaburi;

Division of Urology, Department of Surgery, Faculty of Medicine, Chulalongkorn University;

Division of Nursing;

Research
Co-ordination Section, Department of Disease Control, Bamrasnaradura Infectious Diseases Institute, Ministry of Public Health; **Faculty of
Nursing, Huachiew Chalermprakiet University;

Division of Medicine, Department of Disease Control, Bamrasnaradura Infectious Diseases
Institute, Ministry of Public Health, Nonthaburi, Thailand
Summary: This study assessed prevalence and associated factors of hypogonadism among 491 HIV-infected Thai men who
visited the HIV outpatient clinic. All participants were interviewed and data were collected from medical records, including
demographic and HIV-related illness characteristics. They also completed questionnaires relevant to hypogonadal symptoms,
sexual function and depression. All participants blood samples were obtained to check for total testosterone, sex hormone-binding
globulin (SHBG) and albumin levels, and free testosterone (cFT) was calculated. Hypogonadism was diagnosed if a cFT level of
,0.225 nmol/L was detected. The median age of the participants was 37 years old (ranging from 34 to 44 years old). HIV infection
was diagnosed for a median of 77 (4799) months. Eight of 491 participants (2%) had hypertension and 1% had diabetes mellitus
(DM). Fourteen (3%) used methadone and 23% had SHBG level over 70 nmol/L. Of the 491 participants, 123 (25%) men were
diagnosed with hypogonadism. The univariate analyses indicated that DM, hypertension, methadone use, SHBG level .70 nmol/L
group and lack of antiretroviral therapy were associated with hypogonadism. In multivariate analysis, a SHBG level .70 nmol/L
was the only factor that was signicantly associated with hypogonadism (odds ratio [OR] 1.922, P 0.007).
Keywords: HIV, AIDS, men, hypogonadism, prevalence, free testosterone, SHBG, Thailand
INTRODUCTION
Hypogonadism has previously been studied among
HIV-infected men, as well as other endocrine disorders includ-
ing hypothyroidism, adrenal insufciency, hypercortisolism,
diabetes mellitus (DM) or dyslipidaemia. Hypogonadism can
cause bothersome clinical manifestations for HIV-infected
men such as sexual dysfunction, fatigue, depressive mood,
sleep problems, decrease in muscular strength and lean body
mass, diminished bone mineral density, dyslipidaemia and
anaemia.
1
Some studies reported that hypogonadism was the
most frequent endocrine disorder in men with AIDS.
2,3
Hypogonadism was found in up to 50% of AIDS patients in
the pre-highly active antiretroviral therapy ( pre-HAART) era
and was related to HIV-associated wasting and advanced
stage of AIDS.
2,4
During the era after HAART, the prevalence
of hypogonadism decreased to 1925% but it still remained
higher than the previous reports in American men of the
same age.
3,5,6
The predictors of hypogonadism in
HIV-infected men are not well dened and may be HIV infec-
tion itself, concurrent opportunistic infections, malnutrition or
medications such as methadone, morphine sulphate, ketocona-
zole, megestrol acetate, glucocorticoids and anabolic steroid.
3,7,8
Determining the prevalence and associated factors for
hypogonadism among HIV-infected men is important,
because it results in many clinical sequelae and deterioration
in the quality of life.
1
The aim of the present study was to deter-
mine the prevalence and associated factors of hypogonadism
among HIV-infected Thai men.
MATERIALS AND METHODS
The design of the present study was an observational
descriptive cross-sectional study. The sample size for preva-
lence estimation was calculated by the following formula:
n
(Z
a=2
)
2
pq
d
2
where n is the minimum sample size required; P the estimated
prevalence of hypogonadism 0.25; q 1P 0.75; d the
acceptable error 0.04; and a is the probability of type I
error 0.05.
Z
a/2
Z
0.025
1.96
(1:96
2
0:25 0:75
(0:04)
2
n 450 cases.
Correspondence to: K Sunchatawirul
Email: [email protected]
International Journal of STD & AIDS 2012; 23: 876881. DOI: 10.1258/ijsa.2012.011464
by guest on May 15, 2013 std.sagepub.com Downloaded from
After obtaining approval from the Ethics Committee for
Research in Human Subjects, Department of Disease Control,
the present study was carried out between August 2007 and
May 2008 at Bamrasnaradura Infectious Diseases Institute.
The study sample included 491 HIV-infected Thai men (age
18 years) who visited the HIV outpatient clinic with evidence
of HIV infection (conrmed positive ELISA and documented
history of measureable HIV-RNA) during the period of this
study. Patients were excluded if they had a learning disability
(including dyslexia), acute severe illness, or androgen replace-
ment therapy prior to or during the study period. After obtain-
ing informed consent, each participant was interviewed and
the information was recorded in the designed record
forms, including age, current tobacco use, current alcohol use,
physical exercise, current illicit drug use and current psychiatric
illness.
The enroled patients medical records were retrieved and
reviewed for data including duration of HIV infection, current
fasting cholesterol and triglyceride levels, current CD4 cell
count, current plasma HIV-1 RNA, peak HIV-1 RNA, Center
for Disease Control (CDC) clinical category, current medical
history (DM, hypertension, stroke, seizure, myocardial infarc-
tion), any previous HIV-related illness, current opportunistic
infection-related drug use, current receipt of antiretroviral
therapy (ART), current body mass index (BMI) and the diagno-
sis of lipodystrophy (when the participants and treating phys-
ician agreed that peripheral lipoatrophy and central
lipohypertrophy had occurred). Participants who had history
of abnormal symptoms or signs of the testes were examined
for further investigation if indicated by the physicians or urolo-
gists and the report returned to the authors.
Participants were asked to complete questionnaires that
consisted of three parts. The rst part was an aging males
symptom (AMS) questionnaire, which was developed to
assess severity of symptoms of hypogonadism or partial andro-
gen deciency in ageing men (PADAM). The total score over 26
indicated positive hypogonadism symptoms.
9
The AMS ques-
tionnaire was culturally and linguistically validated from
English into Thai with good correlation.
10
The second part
was a bridged ve-item version of the International Index of
Erectile Function (IIEF), also known as the Sexual Health
Inventory for Men (SHIM). A classication-tree analysis sug-
gested an optimal cut-off score of 21 or less for diagnosis of
erectile dysfunction (ED) (sensitivity 0.98, specicity 0.88)
for the clinical trial patients from the USA and UK.
11
The par-
ticipants were classied as having ED if they had an SHIM
total score less than 22.
11
The Thai version of IIEF was used
similar to the study of prevalence of ED in Thailand by Thai
ED epidemiological study group.
12
The third part was a
Thai Hospital Anxiety and Depression Scale (Thai HADS) for
evaluation of depression. The participants who had a total
score higher than 10 calculated from all the even numbered
questions of the 14 measurement items were considered to be
depressed.
13
Blood samples were obtained between 8:00 and 11:00 hours
to measure total testosterone, sex hormone-binding globulin
(SHBG) and albumin levels. The assays were performed all in
the laboratory of the King Chulalongkorn memorial hospital.
The electrochemiluminescence immunoassay (ECLIA) was
intended for use on Elecsys and Cobas Immunoassay Analyzers
(Roche Diagnostic Systems, Indianapolis, IN, USA) to determine
serum total testosterone (normal range 9.628.8 nmol/L) and
SHBG levels (normal range 14.548.4 nmol/L). Calculated
value of free testosterone (cFT) was derived by putting the
data of serum total testosterone, SHBG and albumin levels
into the calculator in www.issam.ch/freetesto.htm.
Hypogonadism was diagnosed as a calculated free testosterone
level of less than 0.225 nmol/L according to the standards,
guidelines and recommendations of The International Society
for the Study of the Aging Men (ISSAM).
14,15
According to
ISSAM, a deciency in serum testosterone levels is dened by
the level below the young healthy adult male reference range.
This is the same denition for diagnosis of hypogonadism
according to the Endocrine Society clinical practice guideline
by Bhasin et al.
16
The prevalence of hypogonadism was reported. Descriptive
statistics were presented in contingency tables for categorical
variables and continuous variables. For non-normally distribu-
ted data of continuous variables, median and range were
reported. Univariate analysis to compare variables between
the patients with and without hypogonadism was performed
using Chi-squared or Fishers exact test as appropriate. For con-
tinuous variables, t-tests or non-parametric counterparts were
used depending upon observed distributions. Multivariable
logistic regression analysis was used to assess the strength of
association of potential associated factors for hypogonadism.
The variables in the multivariable logistic regression
analysis comprised the variables with a P value 0.05 from
the univariate analysis. The P value ,0.05 indicates statistical
signicance.
RESULTS
A total of 491 HIV-infected Thai men were enroled. The
descriptive characteristics of the participants are shown in
Table 1. The median age of the participants was 37 (range
3444) years of age. Participants had known of their seroposi-
tivity for a median of 77 (range 4799) months. Of 491 partici-
pants, 0.8% had DM, 1.6% had hypertension and 25.5% had
diagnosed lipodystrophy. Fourteen of 491 participants (2.9%)
used methadone and 2.6% used antidepressant drugs. Of 491
participants, 114 (23.2%) had a SHBG level over 70 nmol/L.
Among the 491 HIV-infected participants, 459 (93.5%) men
were receiving ART.
Out of the 491 participants in this study, 123 (25.0%, 95%
condence interval (CI) 21.329.2) were diagnosed with hypo-
gonadism due to a cFT level 0.225 nmol/L. The median
(range) of serum total testosterone was 19.1 (13.425.5) nmol/
L. The univariate analyses of factors associated with hypogo-
nadism are also shown in Table 1. The factors, including DM,
hypertension, methadone use, SHBG level .70 nmol/L group
and not currently receiving ART were associated with hypogo-
nadism (P 0.05). The authors also examined currently receiv-
ing ART according to non-nucleoside reverse transcriptase
inhibitors-based (NNRTI-based), protease inhibitor-based
(PI-based) and individual ARV drugs such as ritonavir,
indinavir, lopinavir, efavirenz, nevirapine, didanosine, tenofo-
vir, zidovudine, lamivudine or stavudine but found no corre-
lation (P . 0.05). HIV-related illnesses were studied and
categorized into pulmonary tuberculosis, extra-pulmonary
tuberculosis, Pneumocystis jiroveci pneumonia, cytomegalovirus
disease or toxoplasmosis of the brain. However, the results
showed that such illnesses had no relation to hypogonadism.
Almost all of these illnesses were inactive conditions. Current
opportunistic infection-related drug use including uconazole,
................................................................................................................................................
Sunchatawirul et al. Hypogonadism in HIV-infected Thai men 877
by guest on May 15, 2013 std.sagepub.com Downloaded from
cotrimoxazole and antituberculosis drugs (isoniazid, rifampicin
pyrazinamide, ethambutol) were separately analysed and also
indicated no association with hypogonadism.
The multivariate analyses of the selected factors with their
odds ratio (OR), 95% CI and P values are shown in Table 2.
The SHBG level .70 nmol/L was the only factor that was
signicantly associated with hypogonadism (OR 1.922,
P 0.007).
The data collection by using the AMS questionnaire indicated
that 72.7% of all participants had hypogonadism symptoms,
with a sensitivity level of 78.0%, specicity of 29.1%, positive
predictive value of 26.9% and negative predictive value of
79.9%. There was no correlation between hypogonadism and
positive hypogonadism symptoms.
Out of 491 participants, 130 (26.5%) were diagnosed with ED
by total score of SHIM ,22. There was no correlation between
ED and hypogonadism. Of 491 participants, 22 (4.5%) had
depression indicated by positive Thai HADS but no correlation
was found between depression and hypogonadism.
DISCUSSION
The prevalence of hypogonadism among HIV-infected Thai
men in the present study is 25%. This rate of hypogonadism
in the HAART era is lower than the rates in the previous
studies conducted in the pre-HAART era by Dobs et al.
(50%),
4
Raf et al. (29%)
17
and Grinspoon et al. (49%).
18
The
rst two aforementioned studies diagnosed hypogonadism by
determining total testosterone levels which might even under-
estimate the prevalence of hypogonadism
3
but Grinspoon
et al.
18
used free testosterone levels which were recommended
according to increased SHBG levels in HIV-infected patients.
3
The reduction in the prevalence of hypogonadism in the
present study is comparable with the ndings of other studies
in the HAART era, including studies by Berger et al. (17%),
5
Rietschel et al. (19%),
3
Fisher et al. (20%),
6
Crum et al. (17%)
19
and Rochira et al. (16%).
20
The availability of HAART leading
to a lower prevalence of hypogonadism is expected to be
related to the reduction in the number of patients with
advanced HIV/AIDS. However, hypogonadism remains a sig-
nicant problem even among the patients in the early stages of
HIV infection, particularly when the prevalence is still higher
than the average rate for the general population.
1
Kaufman
and Vermeulen
21
found only one subnormal serum testosterone
in 105 non HIV-infected men aged 2040 years old. Araujo
et al.
22
reported 4.1% of the age-specic prevalence of hypo-
gonadism in general population aged 4049 years old. By com-
parison, the present study indicated that the prevalence of
hypogonadism in HIV-infected men was higher than that of
the general population at the same age. These ndings
propose that serum testosterone might begin to decrease at an
early age in HIV-infected men. This conclusion is similar to
the nding of Rochira et al.
20
In the present study, although the median SHBG level
(42 nmol/L) of the participants was within the normal range;
Table 1 Characteristics and univariate analyses of HIV-infected Thai men with or without hypogonadism
Factor Total (n 5 491)
Hypogonadism
(n 5 123)
Non-hypogonadism
(n 5 368) P value
Age, years; median (range) 37 (3444) 39 (3543) 38 (3444) 0.516
Duration of known HIV infection, month; median (range) 77 (4799) 80 (47100) 77 (4599) 0.765
Body mass index, kg/m
2
; median (range) 21.9 (20.123.9) 22.1 (19.923.5) 21.8 (20.124) 0.645
Current tobacco use; n (%) 162 (32.9) 44 (35.8) 118 (32.1) 0.258
Current alcohol consumption; n (%) 151 (30.8) 31 (25.2) 120 (32.6) 0.075
Current physical exercise; n (%) 369 (75.2) 88 (71.5) 281 (76.4) 0.171
Illicit drug use; n (%) 30 (6.1) 6 (4.9) 24 (6.5) 0.339
Methadone use; n (%) 14 (2.9) 9 (7.3) 5 (1.4) 0.002
Antidepressant use; n (%) 13 (2.6) 3 (2.4) 10 (2.7) 0.583
DM; n (%) 4 (0.8) 3 (2.4) 1 (0.3) 0.050
Hypertension; n (%) 8 (1.6) 5 (4.1) 3 (0.8) 0.026
Psychiatric illness; n (%) 15 (3.1) 6 (4.9) 9 (2.4) 0.146
Lipodystrophy; n (%) 125 (25.5) 37 (30.1) 88 (23.9) 0.108
Cholesterol, mg/dL; median (range); n 13 193 (168223) 211 (175241) 221 (206295) 0.414
Triglycerides, mg/dL; median (range); n 13 168 (105266) 263 (232323) 313 (213452) 0.940
Current CD4 cell count; median (range); n 468 320 (200462) 321 (189456) 318 (207460) 0.811
Current CD4 cell count ,200 cells/mm
3
; n (%), n 468 113 (24.1) 29 (25.4) 84 (23.7) 0.399
Current plasma HIV-1 RNA ,50 copies/mL; n (%), n 417 364 (87.3) 79 (84.0) 285 (88.2) 0.183
Peak plasma HIV-1 RNA level; median (range), n 417 157 (50105,000) 54 (5075,700) 281 (50130,000) 0.296
SHBG level median (range) 42.0 (29.467.1) 46.0 (30.685.1) 41.8 (29.364.2) 0.059
SHBG level .70 nmol/L; n (%) 114 (23.2) 43 (35.0) 71 (19.3) ,0.001
CDC 0.095
0.005
Stage A; n (%) 184 (37.5) 56 (45.5) 128 (34.8)
Stage B; n (%) 99 (20.1) 20 (16.3) 79 (21.5)
Stage C; n (%) 208 (42.4) 47 (38.2) 161 (43.8)
Currently receiving ART; n (%) 459 (93.5) 108 (87.8) 351 (95.4)
HIV-related illness; n (%) 254 (51.7) 56 (45.5) 198 (53.8) 0.069
Current opportunistic infection-related drug use; n (%) 42 (8.5) 7 (5.7) 35 (9.5) 0.128
Hypogonadism by total score of AMS . 26; n (%) 357 (72.7) 96 (78.0) 261 (70.9) 0.076
ED by total score of SHIM ,22; n (%) 130 (26.5) 30 (24.4) 100 (27.2) 0.315
Depression by positive Thai HADS; n (%) 22 (4.5) 7 (5.7) 15 (4.1) 0.300
DM diabetes mellitus; SHBG sex hormone-binding globulin; CDC Centers for Disease Control and Prevention; ART antiretroviral therapy; AMS ageing mens
symptom; ED erectile dysfunction; SHIM Sexual Health Inventory for Men; HADS Hospital Anxiety and Depression Scale
................................................................................................................................................
878 International Journal of STD & AIDS Volume 23 December 2012
by guest on May 15, 2013 std.sagepub.com Downloaded from
however, 23.2% of participants were found to have the SHBG
level over 70 nmol/L, which was signicantly a predictive
factor for hypogonadism in HIV-infected Thai men. SHBG is
the major binding protein for testosterone, primarily syn-
thesized in the liver as well as in breast or prostatic tissue and
it renders testosterone unavailable to most tissues. In total,
6069% of total testosterone is tightly bound to SHBG,
3038% of total testosterone is loosely bound to albumin and
12% of testosterone is unbound (free testosterone). The last
two previously stated types of testosterone (albumin-bound
and free testosterone) are called bioavailable testosterone,
whichis metabolicallyactive.
23
Specically, SHBGlevels increase
with ageing, hepatic cirrhosis, hyperthyroidism, anticonvulsant
use and HIV infection, causing decline of circulating albumin-
bound and free testosterone, resulting in hypogonadism.
24
This result of increase in SHBG level in the present study is sup-
ported by Martin et al.
25
The participants with SHBG levels
over 70 nmol/L had 1.9 times the risk of being diagnosed
with hypogonadism than those participants without.
The study by Martin et al.
25
reported increased SHBG levels
among HIV-infected men. Based on the study, they advised
the measurement of free testosterone or bioavailable testoster-
one in these patients. The previous study of Rietschel et al.
3
reported that SHBG level was highly associated with total
testosterone level but not with free testosterone level and they
found no association between SHBG levels and free testosterone
levels. In contrast, Wasserman et al.
26
reported reduction in
testosterone and SHBG levels in three men with AIDS
wasting and hypogonadism after prolonged exposure to testos-
terone and oxandrolone. They concluded that rst pass metab-
olism of orally administered oxandrolone might decrease
hepatic synthesis of SHBG, allowing exogenously supplied
testosterone to be excreted. The reduction in SHBG levels was
the oxandrolone side-effect.
There were 14 participants with positive HBsAg, 21 partici-
pants with positive anti-HCV and only one participant with
positive HCV RNA in this study. SHBG levels were grouped
according to these hepatitis proles. In analysis, the authors
found no correlation between SHBG levels and hepatitis pro-
les. The present study also assessed possible predictors of
high SHBG levels in HIV-infected men and found that a low
body mass index (BMI) was the only signicant predictive
factor. Further studies should be required for identication of
the potential associated factors of high SHBG levels in
HIV-infected men.
In the general population, the most widely accepted par-
ameter to establish the presence of hypogonadism is the
measurement of serum total testosterone (TT). But serum TT
level would not be adequate in the diagnosis of hypogonadism
in HIV-infected men. This is because the present study reported
a signicant association between high SHBG levels and
hypogonadism. The median (range) of SHBG levels of
all participants (42.0 [29.467.1] nmol/L) was signicantly
(P , 0.001) higher than the median (range) of SHBG levels
of the laboratory reference range of healthy men (29.9
[14.548.4] nmol/L). The rise in SHBG levels among these
men made it impossible to identify a decrease in TT level,
resulting in an underestimation of hypogonadism diagnosis.
This study also assessed prevalence and associated factors of
hypogonadism using a TT level of less than 10.4 nmol/L as
the reference method. The median (range) of SHBG levels in
hypogonadal men (30.3 [20.237.8] nmol/L) was signicantly
(P , 0.001) lower than the median (range) of SHBG levels of
eugonadal men (45.8 [31.771.0] nmol/L). When grouping
the patients with SHBG levels of more than 70 nmol/L, we
found six of 58 cases or 10.3% of the patients had hypogonad-
ism and 108 of 433 cases (24.9%) had eugonadism. We also
found that SHBG levels increase signicantly in association
with rising TT levels. These ndings caused a reduction in hypo-
gonadal cases from 123 cases to 58 cases and the prevalence of
hypogonadism decreased from 25% to 12%. In multivariate
analysis, SHBG levels were the only signicant protective
factors of hypogonadism. (OR 0.766, P , 0.001). Then if TT
levels were used in the diagnosis of hypogonadism among
HIV-infected men, physicians should be cautious about under-
estimation of the prevalence of hypogonadism and unusual
SHBG hypogonadism correlation. In 2010, Moreno-Perez
et al.
27
found that the median SHBG level among
HIV-infected men was signicantly higher than the reference
population of eugonadal healthy men. They also validated the
determination of serum TT level of less than 10.4 nmol/L for
hypogonadism diagnosis in HIV-infected men using calcu-
lation of free testosterone (cFT) as the reference method, and
TT levels had a sensitivity of 25% in the diagnosis of hypogo-
nadism. They concluded that TT level was not useful in the
diagnosis of hypogonadism in HIV-infected men. The present
study also validated the determination of serum TT level of
less than 10.4 nmol/L for hypogonadism diagnosis in similar
patients using cFT as the reference method and found a sensi-
tivity of 43.9%. According to the result, TT levels might not
be useful in the diagnosis of hypogonadism in HIV-infected
men. Dube et al.
28
studied effects of potent ART on free testos-
terone levels and fat-free mass in HIV-infected men. Their data
support obtainment of free testosterone levels when hypo-
gonadism is suspected in patients with conditions such as
HIV infection, which SHBG levels are commonly elevated.
There was no association between CDC clinical category,
CD4 cell count or HIV-related illness and hypogonadism.
This result is similar to the other studies in the HAART
era.
3,5,6
The advent of HAART might result in the reduction
of cases associated with end-stage AIDS such as low CD4 cell
count, HIV-related illness and advanced CDC clinical category.
Table 2 Multivariate analyses of HIV-infected Thai men with or without hypogonadism
Factors
Univariate analysis Multivariate analysis
COR 95% CI P value AOR 95% CI P value
SHBG level .70 nmol/L 2.248 1.4313.534 ,0.001 1.922 1.2003.080 0.007
Methadone use 5.732 1.88317.449 0.002 2.971 0.88010.026 0.079
Currently receiving ART 0.349 0.1690.721 0.005 0.472 0.2121.054 0.067
Hypertension 5.155 1.21421.897 0.026 3.855 0.82318.065 0.087
DM 9.175 0.94589.034 0.050 7.469 0.70279.495 0.096
COR crude odds ratio; AOR adjusted odds ratio; CI condence interval; SHBG sex hormone-binding globulin; ART antiretroviral therapy; DM diabetes mellitus
................................................................................................................................................
Sunchatawirul et al. Hypogonadism in HIV-infected Thai men 879
by guest on May 15, 2013 std.sagepub.com Downloaded from
In 2007, Crum et al. found that increasing age and BMI were
risk factors and smoking was protective factor of hypogonad-
ism in HIV-infected men.
19
These results were different from
the present study and cannot be comparable because Crum
et al. used total testosterone level for diagnosing hypogonad-
ism. In contrast with the report of Moreno-Perez et al.,
27
increas-
ing age in this study had no correlation with hypogonadism.
This might be affected by the range of age (3444 years) in
our study which was narrower than the range of age (2568
years) in the study of Moreno-Perez et al.
27
The AMS questionnaire used in the present study was not
helpful for hypogonadism diagnosis because there was no cor-
relation between hypogonadism and positive hypogonadism
symptoms. Furthermore, this questionnaire yielded quite low
sensitivity (78.0%) and low specicity (29.1%) that was compar-
able with the nding of Moreno-Perez et al.
27
which showed
quite low sensitivity (66.7%) and low specicity (37.7%).
These ndings suggest that the AMS questionnaire can only
be used as a self-evaluation screening questionnaire but not
diagnostic test for hypogonadism in HIV-infected men. Even
in the general population, the AMS questionnaire is not rec-
ommended for making the diagnosis of hypogonadism due
to its low specicity.
There was no correlation between hypogonadism and ED in
the present study; this is similar to other studies in the HAART
era.
19,20
Crum et al.
19
found increasing age to be a signicant
predictor of both ED and hypogonadism. These results
suggest a non-hormonal aetiology of ED in HIV-infected men.
On the one hand, ED is not very useful as a sign of hypogonad-
ism in the setting of HIV-infected men whose symptoms are
different from those of HIV-negative men. On the other hand,
ED is frequent in HIV-infected men due to other factors that
are different from hypogonadism. In addition, depression was
also evaluated in this study and we found that it had no corre-
lation with hypogonadism, similar to a previous study.
19
The present study has a number of limitations. Firstly, the
study was an observational cross-sectional descriptive study
so the authors could only observe the prevalence and tem-
porally associated factors. Approximately 35% of persons
with hypogonadism in this study had an increased SHBG
level, suggesting that other mechanisms such as DM, hyper-
tension, methadone use and lack of ART may be at play.
Further prospective analytical study is required for determi-
nation of the potential pathogenesis of hypogonadism among
HIV-infected men. More analytical or experimental studies
are also warranted for determination of therapeutic or
adverse effects of testosterone supplementation. Secondly, the
present study collected just single serum samples of cFT
without assessment of luteinizing hormone (LH), prolactin
and follicle stimulating hormone (FSH). The lack of results on
serum gonadotropins limited the ability of our study to estab-
lish the cause of hypogonadism. ISSAM recommends that if tes-
tosterone levels are at the lower limit of or below the accepted
normal values, it is sensible to conrm the results with a second
serum sample together with prolactin, FSH and LH. Thirdly,
signs and symptoms related to hypogonadism and testicular
pathology were assessed by different physicians and the
reports were returned to the authors. This might limit the
strength of the results. Fourthly, The AMS questionnaire was
developed to assess severity of symptoms of hypogonadism
in ageing men. The questionnaires used were not validated
for young/middle-aged men. This limitation might account
for underestimating the signs and symptoms of hypogonadism
in this study. Fifthly, this study used the SHIM questionnaire
because it has been proved to possess favourable statistical
properties in diagnosing the presence and severity of ED in
many prevalence and intervention studies.
29
It is a convenient
and reliable tool to rapidly identify patients with ED who
should be further assessed. This might limit the strength of
the results. The original 15-item version of the IIEF was devel-
oped for use in determining efcacy of treatment in controlled
clinical trials. The IIEF-15 has high sensitivity for detecting real
treatment effects and has been adopted as the gold standard
treatment outcome measure for clinical trials in ED.
29
Finally,
the present study could not accurately determine the free testo-
sterone level by the equilibrium dialysis method, which is the
ideal method,
14
because it is not available in a laboratory in
Thailand and is very costly.
However, the present study had several benets. Firstly, the
optimum sample size by statistical calculation enabled accurate
determination of the prevalence of hypogonadism in
HIV-infected Thai men. Secondly, the authors diagnosed hypo-
gonadism by using cFT level that is more accepted by the
ISSAM than using the total testosterone level.
14
Thirdly,
widely available information technology, the authors demon-
strated the procedure to obtain the cFT for diagnosis of hypogo-
nadism in a way that was practical for most Thai provincial
hospitals. Finally, the authors did not request for free testoster-
one level determination by the equilibrium dialysis method
from abroad but instead opted for the use of the cFT level for
identication of hypogonadism due to its cost-effectiveness.
In the USA, the costs of measuring and reporting free testoster-
one level by equilibrium dialysis, cFT level and total testoster-
one level are estimated at $30.58, $5.93 and $3, respectively.
30
However, the costs are proportionately much higher in
Thailand (cFT level measurement costs $20 and total testoster-
one level measurement costs $10).
In conclusion, the prevalence of hypogonadism among
HIV-infected Thai men in the present study was comparable
to other HAART era studies
3,5,6,19,20
but still remained higher
than the average prevalence for men in general. The only
important signicant risk factor for hypogonadism was a
SHBG level .70 nmol/L, indicating a high tendency towards
with hypogonadism. As a result, the authors recommend the
use of cFT for the diagnosis of hypogonadism in HIV-infected
Thai men.
ACKNOWLEDGEMENTS
The authors wish to express our thanks for the proofreading by
Dr Supara Kapasuwan, Mahidol University International
College, the printing by Wattana Sanchiem, the data manage-
ment by Sumrauy Nilkamhang, the funding permission by
Dr Preecha Tunthanathip (Director of Bamrasnaradura
Infectious Diseases Institute) and kind support from Innaree
Jirawatkhanaporn.
REFERENCES
1 Crum NF, Furtek KJ, Olson PE, Amling CI, Wallace MR. A review of
hypogonadism and erectile dysfunction among HIV-infected during the
pre- and post-HAART eras: diagnosis, pathogenesis, and management.
AIDS Patient Care STDs 2005;19:65571
2 Grinspoon SK, Bilezikian JP. HIV disease and the endocrine system. N Engl
J Med 1992;327:13605
................................................................................................................................................
880 International Journal of STD & AIDS Volume 23 December 2012
by guest on May 15, 2013 std.sagepub.com Downloaded from
3 Rietschel P, Corcoran C, Stanley T, et al. Prevalence of hypogonadism among
men with weight loss related to human immunodeciency virus infection who
were receiving highly active antiretroviral therapy. Clin Infect Dis
2000;31:12404
4 Dobs AS, Dempsey MA, Ladenson PW, et al. Endocrine disorder in men
infected with human immunodeciency virus. Am J Med 1988;84:6116
5 Berger D, Muurahainen N, Wittert H, et al. Hypogonadism and wasting in the
era of HAART in HIV-infected patients [Abstract 32174]. Int Conf AIDS
1998;12:555
6 Fisher A, Desyatnik M, Baaj A. The prevalence of hypogonadism in HIV
infected patients receiving HAART [Abstract no. TuPeB3180]. Int Conf AIDS
2000;13
7 Cofrancesso J Jr, Whalen JJ III, Dob AS. Testosterone replacement treatment
options for HIV-infected men. J Acquir Immune Dec Syndr Hum Retrovirol
1997;16:25465
8 Mylonakis E, Koutkia P, Grinspoon S. Diagnosis and treatment of androgen
deciency in human immunodeciency virus-infected men and women. Clin
Infect Dis 2001;33:85764
9 Heineman LAJ, Saad F, Thicle K, Wood-Dauphinee S. The aging males
symptoms (AMS) rating scale cultural and linguistic validation into English.
Aging Male 2001;3:1422
10 Leungwattanakij S, Ratana-Olarn K, Lertsittichai P. The aging males
symptoms rating scale. cultural and linguistic validation into Thai. J Med Assoc
Thai 2003;86:110615
11 Rosen RC, Cappelleri JC, Gendrano N. The international index of erectile
function (IIEF): a state-of-the-science review. Int J Impot Res 2002;14:22644
12 Kongkanand A. Prevalence of erectile dysfunction in Thailand. Thai erectile
dysfunction epidemiological study group. Int J Androl 2000;23:7780
13 Nilchaikovit T, Lortrakul M, Phisansuthideth U. Development of Thai version
of hospital anxiety and depression scale in cancer patients. J Psychiatr Assoc
Thailand 1996;41:1830
14 Wang C, Nieschlag E, Swerdloff RS, et al. Investigation, treatment and
monitoring of late-onset hypogonadism in males. ISA, ISSAM, EAU, EAA and
ASA recommendations. Eur J Endocrinol 2008;159:50714
15 Vermeulen A, Kaufman JM. Diagnosis of hypogonadism in the aging male.
Aging Male 2002;5:1706
16 Bhasin S, Cunningham GR, Hayes FJ, et al. Testosterone therapy in men with
androgen deciency syndromes: an endocrine society clinical practice
guideline. J Clin Endocrinol Metab 2010;95:253659
17 Raf F, Brisseau JM, Planchon B, Remi JP, Barrier JH, Grolleau JY. Endocrine
function in 98 HIV-infected patients: a prospective study. AIDS 1991;5:72933
18 Grinspoon S, Corcoran C, Lee K. Loss of lean body and muscle mass correlates
with androgen levels in hypogonadal men with acquired immunodeciency
syndrome and wasting. J Clin Endocrinol Metab 1996;81:40518
19 Crum NF, Bavaro M, Hale B, et al. Erectile dysfunction and hypogonadism
among men with HIV. AIDS Patient Care STDs 2007;21:919
20 Rochira V, Zirilli L, Orlando G, et al. Premature decline of serum total
testosterone in HIV-infected men in the HAART-era. PLoS One 2011;6:e28512
21 Kaufman JM, Vermeulen A. Declining gonadal function in elderly men.
Baillieres Clin Endocrinol Metab 1997;11:289309
22 Araujo AB, ODonnell AB, Brambilla DJ, et al. Prevalence and incidence of
androgen deciency in middle-aged and older men: estimate from
Massachusetts male aging study. J Clin Endocrinol Metab 2004;89:59206
23 Morales A, Morley J, Heaton JPW. Androgen deciency in the aging male.
In: Wein AJ, Kavoussi LR, Novick AC, et al., eds. Campbells Urology. 9th edn.
Philadelphia: WB Saunders, 2007:85062
24 Kalyani RR, Gavini S, Dob AS. Male hypogonadism in systemic disease.
Endocrinol Metab Clin N Am 2007;36:33348
25 Martin ME, Benassayag C, Amiel C, Canton P, Nunez EA. Alterations in the
concentrations and binding properties of sex steroid binding protein and
corticosteroid-binding globulin in HIV patients. J Endocrinol Invest
1992;15:597603
26 Wasserman P, Segal-Maurer S, Rubin D. Low sex hormone-binding globulin
and testosterone levels in association with erectile dysfunction among human
immunodeciency virus-infected men receiving testosterone and
oxandrolone. J Sex Med 2008;5:2417
27 Moreno-Perez O, Escoin C, Serna-Candel C, et al. The determination of total
testosterone (RIA) are not applicable to the evaluation of gonadal function in
HIV-infected males. J Sex Med 2010;7:287383
28 Dube MP, Parker RA, Mulligan K, et al. Effects of potent antiretroviral therapy
on free testosterone levels and fat-free mass in men in a prospective,
randomized trial: a 5005s, a substudy of AIDS Clinical Trial Group Study 384.
Clin Infect Dis 2007;45:1206
29 Cappelleri JC, Rosen RC. The Sexual Health Inventory for Men (SHIM): a
5-year review of research and clinical experience. Int J Impot Res
2005;17:30719
30 Devan ML, Bankson DD, Abadie JM. To what extent are free testosterone (FT)
values reproducible between the two Washingtons, and can calculated FT be
used in lieu of expensive direct measurement? Am J Clin Pathol
2008;129:45963
(Accepted 31 March 2012)
................................................................................................................................................
Sunchatawirul et al. Hypogonadism in HIV-infected Thai men 881
by guest on May 15, 2013 std.sagepub.com Downloaded from