Before the introduction of highly active antiretroviral therapy,
patients with HIV were commonly diagnosed with hypogonadism — likely an
effect of “wasting” and chronic illness associated with their
Now, years later, data continue to reveal an increased prevalence of
hypogonadism among patients with HIV, although the rates are not as high since
the introduction of highly active antiretroviral therapy (HAART).
Todd T. Brown, MD, PhD, said studies defining the frequency of
hypogonadism in the era of HAART are needed to determine who is at highest
Reprinted with permission from:
The cause of hypogonadism in HIV is unknown. Researchers continue to
investigate the reasons behind the increased incidence of hypogonadism, as well
as ways to improve and standardize diagnosis and treatment and to determine
which patients with HIV are at highest risk for hypogonadism.
“Hypogonadism among men infected with HIV is nothing new and was
very common before there was effective ART,” Todd T. Brown, MD,
PhD, associate professor of medicine in the division of endocrinology
and metabolism at Johns Hopkins University, told Infectious Disease
News. “Testosterone replacement was an important treatment for
AIDS-related wasting before there was effective ART. Now we’re seeing sort
of a different ‘flavor’ because most people are on effective ART, but
the question remains about what’s happening to their testosterone
In an interview with Infectious Disease News,
Steven K. Grinspoon, MD, professor of medicine at Harvard
Medical School and director of the program in nutritional metabolism and
clinical director of the neuroendocrine clinical center at Massachusetts
General Hospital, said in the early days of HIV, before ART, patients with HIV
were often chronically ill with opportunistic infections and low weight, and
were often hypogonadal.
“It was unclear if the hypogonadism was related to HIV per se or
associated illness, but the prevalence was fairly high — typically around
50%,” Grinspoon said in an interview. “More recent data suggest that
the prevalence is about 20%, but we’re not 100% sure what the number
In a 1988 study published in the American Journal of
Medicine, Adrian Dobs, MD, MHS, of Johns Hopkins School
of Medicine, and colleagues observed a 50% incidence of hypogonadism among men
with HIV. A 2011 study published in PLoS One, conducted by
researchers in Italy, suggested that the current prevalence of hypogonadism is
“The frequency of hypogonadism has changed over the years,”
Dobs told Infectious Disease News. “For AIDS patients, it may
be as high as 50%, and in patients without AIDS who are HIV-positive, the
numbers are much lower, probably about 15% to 20%.”
AIDS wasting was long-thought to be the culprit behind hypogonadism in
these patients. AIDS wasting refers to when patients with AIDS lose 10% or more
of their body weight. It is often accompanied by diarrhea, extreme weakness and
fever that are not related to an infection. In severe cases, including AIDS
wasting, the illness suppresses hypothalamic gonadotropin-releasing hormone and
pituitary secretion of luteinizing hormone and follicle-stimulating hormone,
which in turn results in reduced testosterone production and hypogonadism.
However, the incidence of AIDS-related wasting has subsided, and the
essential question is why hypogonadism persists in current patients, even with
ART, according to Marc Hellerstein, MD, PhD, professor of
human nutrition at the University of California at Berkeley and professor of
endocrinology, metabolism and nutrition at the University of California at San
“My best guess is that it is related to the chronic ongoing
inflammation due to immune activation,” Hellerstein said. “Starvation
or infection/inflammation, for example, are well understood to be associated
with hypogonadism. This might be a reflection of ongoing inflammation, but we
really don’t know.”
According to Grinspoon, the reasons that hypogonadism is common in
people with HIV are multifactorial. Chronic disease itself is associated with
hypogonadism, especially if associated with weight loss and severe illness.
Grinspoon also said there may be some effects of excess inflammatory cytokines
on gonadal function, and that HIV medications may also have an effect on
There are several other mechanisms that may be related to hypogonadism,
according to Dobs. There could be an infection related to AIDS that directly
affects either the pituitary gland or the testes. In addition, HIV and AIDS are
related to a number of inflammatory factors, and these are associated with
decreased production of testosterone due, in part, to interference with several
of the enzymes in the pathway to testosterone production, Dobs said.
Since the introduction of HAART, studies have shown varying numbers for
the incidence of hypogonadism in men. For example, in a 2000 article published
in Clinical Infectious Diseases, Grinspoon and colleagues found
that 21% of men with HIV receiving HAART had low free testosterone levels.
Based on these data, the researchers concluded that hypogonadism remains
relatively common in men with AIDS wasting despite treatment with HAART.
Wunder and colleagues in 2007 reported that 70% of treatment-naive men
with HIV in their study had subnormal free testosterone levels, which did not
improve after 2 years on combination ART.
Conversely, also in 2007, Michael P. Dubé, MD,
and colleagues reported a subnormal free testosterone level among 6% of
treatment-naive patients at baseline before receiving ART. Dubé,
professor of medicine in the division of infectious diseases at Keck School of
Medicine at the University of Southern California, said this number is not
significantly higher than in the general population.
The varied numbers, according to Grinspoon, may be due to differences in
screening techniques. In Wunder and colleagues’ study, for example,
Grinspoon said the researchers did not strictly assess free testosterone levels
in the morning, which is a recommended technique.
“The conclusion by Wunder and colleagues of a 70% incidence may be
somewhat overstated,” he said. “But, the data showing that there was
no increase in testosterone levels after combination ART are interesting and
suggest factors intrinsic to HIV disease may contribute to an increased
prevalence of hypogonadism.”
Dubé said he and colleagues in their 2007 paper measured free
testosterone by dialysis.
“When studies select out people such as individuals with wasting,
who are more likely to display abnormalities, you’re going to find a
higher frequency for that group just by definition,” he said.
Assessing hypogonadism among patients with HIV is often difficult for a
number of reasons. First, according to Grinspoon, patients with HIV often have
increased sex hormone-binding globulin (SHBG) levels, which can affect the
total testosterone level. The Endocrine Society recommends generally assessing
morning total testosterone, except for in men in whom SHBG abnormality is
“If you assess the total testosterone, you may get slightly
elevated or falsely reassuring results, so most experts agree that assessing
the free or bioavailable testosterone is more valid in the HIV
population,” Grinspoon said.
Dobs also said the evaluation should include two separate blood samples,
both drawn early in the morning. In addition, Brown said the algorithm for
calculating testosterone levels must be validated.
“This algorithm involves obtaining total testosterone from a
reliable methodology and then calculating the free testosterone using an
equation that’s well established in the general population,” he said.
“Whether or not this equation holds up in HIV-infected men who have very
high levels of SHBG hasn’t been determined yet.”
Another issue is the assay used to assess the levels.
“The equilibrium dialysis method is the most robust, but not
everyone uses that assay or assesses the bioavailable testosterone,”
Grinspoon said. “To add an additional wrinkle because of a diurnal rhythm,
in which testosterone is highest in the morning and then low in the afternoon
and evening, the morning level should be checked and confirmed with a second
assessment. There’s some precision that’s needed to make this
diagnosis, and the real prevalence in a very large population using this
algorithm is unknown.”
Dubé said the variability in prevalence results is related to
differences in screening methods.
“There’s no good definition of what defines a serum
testosterone level that requires intervention,” he said. “It depends
on who you talk to, but it’s still not clear what the best screening test
is for low testosterone levels in patients with HIV.”
Determining whether a patient with HIV has symptoms that warrant
assessment for hypogonadism is also difficult. Because of the association of
hypogonadism with sarcopenia and low bone density, however, making a diagnosis
“The symptoms and signs of hypogonadism are varied, while some of
them are more specific,” Brown said, adding that sexual-type symptoms such
as erectile dysfunction and loss of libido should prompt screening. Changes in
body composition that accompany hypogonadism, such as increased fat, lower
muscle mass and lower BMD, should also serve as red flags.
“Some of these problems are completely unrelated to hypogonadism,
but it’s important to keep these things in mind,” Brown said.
Unexplained fatigue and depression may also warrant screening, although
Grinspoon said he would not test patients for hypogonadism based on just these
two symptoms alone, as they are non-specific.
Steven K. Grinspoon
Grinspoon recommended testing patients who present with low bone density
and to conversely test for low bone density among men who have been diagnosed
“Most important from a medical standpoint is to test those with
advanced HIV disease who have wasting because those are the individuals for
whom testosterone therapy is likely to be of significant benefit,”
Dubé said. “That’s something I think is underutilized —
testing for hypogonadism in that particular group. Once you get beyond that,
symptoms have not been a reliable indicator of whether or not there are low
serum testosterone levels.”
Dobs said it is important to obtain a good medical history and to
perform a physical to determine whether the patients are symptomatic for
hypogonadism. Some of the questions to be asked include: Does the patient have
decreased libido, mild depression or decreased BMD? On a physical exam, what
are the size and the consistency of the testes?
“A [DXA] scan may be helpful in this situation because men who have
hypogonadism have low BMD, and also, some of the medications that are used for
HIV disease are associated with thin bones,” Dobs said.
Dobs said most HIV doctors are comfortable with treating hypogonadism,
as it is a fairly common issue among their patients. If there are questions
about why the patient has hypogonadism, or how the patient is responding to
treatment, then the patient should be referred to an endocrinologist.
Patients with HIV and hypogonadism are treated similarly to those
without HIV — testosterone replacement therapy received intramuscularly or
through newer types of preparations such as gels and patches.
“The disadvantage of the intramuscular injection is that you get a
spike in the testosterone level right after the injection … whereas with
the gels and the patches, it’s more steady state,” Grinspoon said.
“Some people worry that the high levels achieved in association with IM
injections may do more to stimulate prostate growth or result in increased
testosterone conversion to estrogen through aromatization.”
He said, in any instance, patients must be followed appropriately when
receiving testosterone to ensure that levels are not too high and that there is
no increase in blood counts or prostate-specific antigen or prostate growth.
Grinspoon and colleagues have shown that testosterone treatment has
several benefits. They reported in 1998 that patients who received testosterone
saw an increase in fat-free mass, lean body mass and muscle mass, and they also
improved quality of life. They also reported in 2000 that testosterone
administration led to significant improvements in depression scores.
However, Grinspoon said he advises against the use of anabolic steroids
in treating hypogonadism, as these are associated with significant adverse
health consequences and may further suppress endogenous testosterone
“Anabolic steroids, particularly the oral ones, can affect the
liver and are not particularly healthy,” he said. “There is no
advantage to stacking an anabolic steroid on top of testosterone
Moving forward, researchers continue to examine the effects of
testosterone on bone density and muscle, quality of life, psychiatric symptoms,
sexual function and lipids. Hellerstein said more studies examining quality of
life and whether better, more selective androgens can be developed for treating
hypogonadism are also needed.
The continuing high prevalence of hypogonadism among men with HIV may be
because HAARTs have become so effective that more men with HIV are aging,
according to Brown.
“We know that testosterone levels decline with aging and may have
downstream consequences,” he said. “There’s a question as to
whether or not HIV-infected patients, and in this case men, have accelerated
aging and whether gonadal function is one of these things which declines faster
among men with HIV, compare with HIV-negative men.”
In the non-HIV world, researchers are already examining the harms and
benefits of replacing testosterone due to aging.
“It is necessary to test and demonstrate low testosterone levels
before treatment, rather than presuming hypogonadism and treating empirically,
based on symptoms or presumed accelerated aging,” Grinspoon said.
Brown called for studies that define the prevalence of hypogonadism in
this population of men in the era of HAART to understand which men are at
highest risk. He also calls for studies that examine whether there is change in
testosterone over time in men with HIV vs. those without.
He also said there is a need for studies examining whether testosterone
resolves issues associated with aging that seem to occur more frequently in
HIV-infected men, including osteoporosis, low muscle mass, fat redistribution
and diabetes. Brown and colleagues recently published data on the
cross-sectional association between low testosterone level and insulin
resistance and diabetes in HIV-infected men.
Most importantly, however, Dubé said, “An area of needed
research is to get some standardization of the assays and a better definition
of who is going to benefit from testosterone replacement because it is not
without potential side effects.” – by Tina DiMarcantonio and
- Bhasin S. J Clin Endocrinol Metab. 2010;95:2536-2559.
- Choi HH. J Clin Endocrinol Metab. 2005;90:1531-1541.
- Dobs AS. Am J Med. 1988;84:611-616.
- Dolan Looby SE. AIDS. 2009;23:951-959.
- Dolan S. AIDS. 2009;23:951-959.
- Dolan S. Arch Intern Med. 2004;164:897-904.
- Dubé MP. Clin Infect Dis. 2007;45:120-126.
- Grinspoon S. Ann Intern Med. 1998;129:18-26.
- Grinspoon S. J Clin Endocrinol Metab. 2000;85:60-65.
- Herbst KL. Fertil Steril. 2006;85:1794-1802.
- Miller K. J Clin Endocrinol Metab. 1998;83:2717-2725.
- Rietschel P. Clin Infect Dis. 2000;31:1240-1244.
- Rochira V. PLoS One.
- Wunder DM. Antivir Ther. 2007;12:261-265.
- Drs. Brown, Dobs, Dubé, Grinspoon
and Hellerstein report no relevant financial disclosures.
Should women with HIV ever be given testosterone replacement therapy
The few studies published with small cohorts have shown safety,
Several prior studies suggest that women with HIV experience androgen
deficiency. Low testosterone levels are associated with reduced bone density,
lean muscle and fat mass and impaired muscle function in this population.
Additionally, reduced quality of life, including fatigue, has been observed
among those with low testosterone levels.
Few studies have investigated the use of testosterone replacement among
women with HIV. Early work from our group established safety and efficacy, with
regard to improving weight, quality of life, and muscle strength at an
effective dose of 150 mcg per day (during a 3- to 4-day application period;
transdermal delivery system).
In a more recent longitudinal investigation, transdermal testosterone
replacement at 300 mcg per day (via transdermal delivery system; patch changed
twice weekly) during 18 months resulted in an increase in bone mineral density
at the hip and greater trochanter, as well as improvement in lean muscle mass
and BMI. Moreover, testosterone use was associated with improvement in
depressive symptoms and problems affecting sexual function in this population.
Of importance, in this 18-month investigation, transdermal testosterone
administration to women with HIV and low baseline free testosterone
concentrations was very well tolerated and did not result in adverse events,
including lipid or liver abnormalities or significant differences in hirsutism
Other studies have demonstrated testosterone administration at a dose of
300 mcg per day for 6 months is well tolerated and effective in increasing
testosterone levels (but not body composition or muscle function) among women
with HIV. Moreover, high-normal range testosterone levels achieved during a
6-month period in studies of 300 mcg per day did not negatively impact insulin
sensitivity or markers of inflammation/thrombolysis and did not significantly
reduce abdominal fat.
In our experience, transdermal testosterone dosing at 300 mcg per day
among women with HIV and reduced basal testosterone concentrations improved
selected quality-of-life indices, including depression and problems affecting
sexual function, in addition to BMD and lean mass. Although these data would
suggest a potential benefit of testosterone administration in women with HIV,
more data among larger study cohorts of women with HIV are necessary to
determine safety and efficacy in this population. Testosterone administration
in this population remains a promising therapy in need of further study.
Sara E. Dolan Looby, PhD, ANP-BC, is an instructor
in medicine at Harvard Medical School and the Program in Nutritional Metabolism
at Massachusetts General Hospital in Boston. Disclosure: Dr. Dolan
Looby reports no relevant financial disclosures.
Data not robust enough and no appropriate preparations are FDA
I would not generally recommend that testosterone be prescribed for
women with HIV. The first reason is that the data on the effectiveness of
replacement dose testosterone in this group of women are not very robust. There
have been several small, randomized, placebo-controlled studies examining the
effects of testosterone replacement therapy in women with HIV, all with modest
results, and not all with consistent findings.
The first short-term study (4-month duration) in women with AIDS wasting
demonstrated a mean increase in weight of 1.3 kg compared with placebo with the
lower (150 mcg daily) but not higher (300 mcg daily) testosterone dose, while a
second short study (6-month duration) demonstrated an improvement in strength
but not weight or lean body mass with the 300 mcg daily dose. A third study, of
6-month duration, demonstrated no increase in weight, lean body mass or muscle
strength in those who received testosterone compared with placebo. The only
long-term published study (18-month duration) demonstrated modest improvements
in lean body mass (by a mean of 1 kg more than the placebo group), BMI (by a
mean of 0.8 kg/m2 more than the placebo group) and in BMD at the
hip, but not spine, as well as an improvement in mood compared with placebo.
Therefore, although there are promising published data on the
effectiveness of testosterone replacement in other groups of women — for
example, in surgically and naturally postmenopausal women for the treatment of
libido and sexual function — the data in support of testosterone use in
women with HIV are not compelling.
Moreover, and importantly, in most countries, including the United
States, there is no government agency-approved preparation available that
raises testosterone within a physiologic range for women, and supraphysiologic
dosing is inadvisable.
Karen K. Miller, MD, is an associate professor
of medicine at Harvard Medical School and in the neuroendocrine unit at
Massachusetts General Hospital in Boston. Disclosure: Dr. Miller reports
no relevant financial disclosures.
Robert W. Rebar
Testosterone replacement therapy is not approved for use in any
women, and the use is still very controversial.
It is true that women who’ve had their ovaries removed and women
with premature ovarian failure have significantly lower testosterone levels
than women of a similar age, or for premature ovarian failure, even than
postmenopausal women. However, efficacy of providing exogenous testosterone has
not been demonstrated for any product to the satisfaction of the FDA.
The risks of therapy are also unclear. If usage is not yet approved for
women who are HIV-negative, then how can it be recommended for women with HIV?
Robert W. Rebar, MD, is an executive director of
the American Society for Reproductive Medicine. Disclosure: Dr. Rebar is
an employee of ASRM and on the editorial board of Journal Watch. He
also serves as chair of the NIH RMN network DSMB and on a DSMB of an estrogen
trial for postmenopausal women.