Hypogonadism: a neglected comorbidity in young and middle-aged HIV-positive men on effective combination antiretroviral therapy : AIDS

Secondary Logo

Journal Logo


Hypogonadism: a neglected comorbidity in young and middle-aged HIV-positive men on effective combination antiretroviral therapy

Lachâtre, Mariea; Pasquet, Armellec,b; Ajana, Faïzab; Soudan, Benoitd; Quertainmont, Yanne; Lion, Georgesf; Durand, Emmanuelg; Bocket, Laurenceh; Mole, Martinee; Cornavin, Paulineb; Catalan, Pilartxoe; Senneville, Éricb; Goujard, Cécilee; Boufassa, Faroudyi; Cheret, Antoinee,j

Author Information
doi: 10.1097/QAD.0000000000003176



Male hypogonadism is common in people with HIV (PWH) despite the widespread use of combined antiretroviral therapy (cART). Before the widespread use of cART, male hypogonadism was the most common endocrine abnormality detected in PWH [1]. Male hypogonadism was often associated with advanced-stage of HIV-infection and especially with low CD4+ cell count, weight loss, and AIDS wasting [2–5]. Male hypogonadism is also correlated with aging and a number of comorbid conditions (chronic hepatitis C virus infection, use of psychotropic and intravenous drugs) [6–8]. It is mostly caused by impairment of the hypothalamic-pituitary axis (HPA) [9]. It may affect overall quality of life and can lead to osteopenia/osteoporosis, erectile dysfunction/low libido, metabolic syndrome, a decrease in lean body mass/muscle strength and depression; its diagnosis is therefore important [10–13]. Recent studies have estimated its prevalence at 9.3–70% in PWH [4,14–18].

Total testosterone levels are affected by changes in sex-hormone-binding globulin (SHBG) levels in PWH, in whom free testosterone assays are recommended for the diagnosis of male hypogonadism [10,12,13,16]. These assays should be performed according to a well defined method using a reliable technique [Vermeulen equation or equilibrium dialysis (available in only a few specialized laboratories)] [10,12,13,19]. Only limited clinical data obtained with a reliable free testosterone assay are available for analyses of the prevalence of male hypogonadism and factors associated with this condition in PWH [15,16]. Virological suppression is now generally achieved due to the widespread use of cART in PWH in industrialized countries [20], but none of the published studies was conducted in a population of AIDS history-free PWH on effective cART.

Otherwise, osteopenia/osteoporosis [21–23], erectile dysfunction/low libido [24], metabolic syndrome [25], decrease in lean body mass/muscle strength, and depression [26] are widely described in PWH. These disorders may affect the quality of life of patients, which remains today one of the major issues in the overall care of patients. They are classically multifactorial in PWH (e.g. osteoporosis associated with length of time on HIV infection, weight loss, aging, and some cART [27], or erectile dysfunction usually associated with aging, depression, and length of time on cART [28]).

We therefore used a reliable free testosterone assay to determine the prevalence ofmale hypogonadism (and its origin), and to investigate the predictive factors potentially associated with male hypogonadism, including osteoporosis, erectile dysfunction, metabolic syndrome and body composition, in young-to-middle-aged, AIDS history-free, HIV-infected men on effective cART.

Materials and methods

Study design and participants

We conducted a cross-sectional study in PWH attending consultations at two French university hospitals (Tourcoing Hospital and Bicetre Hospital/AP-HP). The study was approved by the Nord-Pas-De-Calais Ethic Committee, in accordance with the 2008 Declaration of Helsinki. We identified potential participants with Nadis computer software. Recruitment began in January 2013, for a planned period of 42 months. The study design is shown in Fig. 1.

Fig. 1:
Study design.

HIV-1-infected men aged between 18 and 50 years were eligible for the study if they displayed a stable suppression of HIV replication (HIV-RNA ≤ 50 copies/ml) on cART for at least 6 months at the time of screening. Key exclusion criteria included HIV-2 infection, hepatitis B and C, cirrhosis, a glomerular filtration rate estimated by the Cockcroft-Gault method ofless than 30 ml/min, past or ongoing opportunistic infection, current cancer, treatment with testosterone or other anabolic agents, antiandrogens, a GnRH (gonadotropin-releasing hormone) analog, estrogens, glucocorticoids, growth hormones or IGF- 1 (insulin-like growth gactor-1) in the previous three months, and a pituitary, testicular, or adrenal disease with or without treatment. Full inclusion and exclusion criteria are provided in Appendix 1, https://links.lww.com/QAD/C448. All participants gave written informed consent before the study began.


The main outcome variable was biochemical hypogo-nadism, defined as a mean serum free testosterone concentration less than 70 pg/ml, in accordance with guidelines [11,29]. Sociodemographic, anthropometric, bone-densitometry, hormonal, immunological, virologi-cal, metabolic, and therapeutic parameters were recorded. The International Index of Erectile Function 5 (IIEF-5) score, the Hamilton Depression Rating scale (HAM-D) and the Aging Male Symptoms (AMS) scale were used to evaluate erectile function, depression, and quality of life, respectively [30,31].


The main parameters, IIEF-5 score, HAM-D scale, AMS scale, and the first venous blood sample were collected at screening. A second venous blood sample was obtained on day 8, for the determination of mean serum free testosterone concentration and for the total testosterone assay. Venous blood samples were collected in the morning, between 0700 and 0900 h, fasting, as recommended [12]. The reference thresholds used for most biochemical parameters were those of the corresponding laboratories. However, the thresholds used for total testosterone and free testosterone were determined in accordance with European guidelines [11], based on the results of the study by Bhasin et al.[29] published in 2011, which was performed on a large cohort of young men (19–40 years of age) with a reliable method. Dual-energy X-ray absorptiometry (DXA) was carried out before week 2, locally, at each site, together with determinations of HIV-RNA, CD4+ and CD8+ T-cell counts, and vitamin D. HIV-DNA, hormonal, and metabolic parameter determinations were performed centrally, in specialized laboratories at Lille Hospital (France).

Free testosterone concentration was calculated according to the Vermeulen equation, from total testosterone, SHBG, and albumin concentrations [19]. Serum total testosterone, SHBG, estradiol, prolactin, luteinizing hormone (LH), and follicle-stimulating hormone (FSH) levels were determined. Total testosterone concentration was determined by two different methods: a radioimmunoassay (RIA) (with the ‘Coat-A-Count Total Testosterone’ RIA kit; Siemens Healthcare Diagnostics Inc., Los Angeles, California, USA) with an interassay precision of less than 11%; and a chemiluminescent microparticle immunoassay (CMIA) on an Architect i2000 immunoassay analyser (“ARCHITECT 2nd Generation Testosterone assay”; Abbott Diagnostics, Abbott Park, Illinois, USA) with an interassay precision of less than 10%. We switched from the RIA to the CMIA assay due to Siemens’ decision to stop producing the radioimmunoassay kits in April 2015. The correlation between the two testosterone assays was evaluated on a sample of 34 serum samples, by Fleiss intraclass correlation evaluation. The intraclass correlation coefficient (ICC) was 0.94, with a 95% confidence interval of [0.88;0.97]. Estradiol was assayed by RIA, andLH, FSH, prolactin, and SHBG were determined by chemilumines-cence immunoassays.

Duration of HIV infection, HIV-DNA load, CD4+ and CD8+ T-cell counts, CD4+ T-cell nadir, and CD4+/ CD8+ T-cell ratio were determined. HIV-RNA load was analyzed on site with an ultrasensitive real-time PCR method. HIV-DNA load was also determined with an ultrasensitive real-time PCR method (Generic HIV-DNA assay; Biocentric, France). CD4+/CD8+ T-cell count ratios were determined by flow cytometry. The duration of exposure to ART and the current ART regimen were recorded. Drugs were classified into the following ART classes: nucleoside reverse transcriptase inhibitors (NRTI), nonnucleoside reverse transcriptase inhibitors (NNRTI), protease inhibitors, and integrase inhibitors.

Age, lifestyle, BMI, waist/hip circumference ratio, and serum 25-OH vitamin D concentration were noted. Lumbar spine (L1 –L4) and hip (left total femur) bone mineral densitometry (BMD) Z-score, and total body composition, including total fat content expressed as a percentage of total mass, appendicular lean mass (kg), and appendicular lean mass index (kg/m2) were evaluated by DEXA (dual-energy X-ray absorptiometry [32]). Sarco-penia was defined as follows: appendicular lean mass < 20 kgs or appendicular lean mass index less than 7 kg/m2[33]. For patients attending of Tourcoing Hospital, DEXA was performed in the Department of Nuclear Medicine, Lille Hospital (France), with a Hologic Discovery A device (Hologic, USA). For patients attending Bicêtre Hospital/AP-HP, DEXA was performed in the Department of Nuclear Medicine, Bicêtre Hospital/AP-HP (France), with a Prodigy Advance GE Lunar device (GE Healthcare, UK). Given the differences in calibration between these two machines, conversion formulas were obtained from GE Healthcare, UK, to harmonize the measurements. Patients were considered to have a normal BMD, osteopenia, or osteoporosis if the lumbar or femoral Z-score was at least –1, between –1 and –2, or –2 or less, respectively [32]. 25-OH vitamin D levels were determined by an electrochemilumines-cence assay (ECLIA).

Metabolic syndrome was defined as the combination of at least three of the following five criteria: waist circumference more than 102 cm, plasma triglyceride concentration at least 1.5 g/l, serum high-density lipoprotein (HDL)-cholesterol concentration less than 0.4 g/l, blood pressure at least 130/85 mmHg, and fasting blood glucose concentration at least 1.1 g/l [34]. HOMA-IR, used to evaluate insulin resistance, was defined as follows: fasting insulin (mIU/l) × fasting blood glucose concentration (mmol/l)/22.5 [35]. Total cholesterol, HDL-cholesterol and LDL-cholesterol, triglycerides, glucose, and insulin were determined by colorimetry.

International Index of Erectile Function (IIEF)-5 score was classified as follows: uninterpretable (1–4), erectile dysfunction (5–20), normal (21–25) [30]. Hamilton Rating Scale for Depression (HAM-D) scores were classified as follows: depression (≥ 8), normal (0–7) [31]. Aging Male Symptoms scale (AMS scale) scores were classified as follows: deterioration in quality of life (> 27), no deterioration in quality of life (≤ 26).

Statistical analysis

We first described the baseline characteristics of patients enrolled in the study. We described quantitative variables using medians and interquartile ranges (IQRs), and compared the results from the two study centers in a nonparametric Wilcoxon rank-sum test. We summarized qualitative variables using percentages and compared the results of the two centers in a chi-squared test.

Factors associated with male hypogonadism were then identified with univariable and multivariable logistic regression models. The ‘log-linear’ hypothesis was verified for each quantitative variable. Variables for which the ‘log-linear’ hypothesis was rejected were introduced as class in the univariable and multivariable models. Variables were added to the multivariable model if found to be associated with male hypogonadism in univariable analysis (P-value <0.25). We developed the multivariable model by backward stepwise regression analysis. In the multivariable analysis, P values less than 0.05 were considered to be significant. The adequacy of the model was assessed with the Hosmer–Lemeshow test.

SAS software version 9.3 (SAS Institute Inc., Cary, North Carolina, USA) was used for all analysis.

This trial is registered with ClinicalTrials.gov, under number NCT02665559.


As shown in the study flow chart (Fig. 2), 2388 PWH, including 1758 men, followed up in the University department of Infectious Diseases of Tourcoing Hospital and 2184 PWH, including 1384 men, followed up in the University department of Internal Medicine of Bicêtre Hospital/APHP were screened between January 1, 2013, and June 30, 2016. Of these patients, 421 men infected with HIV were meeting the inclusion/noninclusion criteria. Finally, 240 patients were enrolled and 231 patients were analyzed. Testosterone data (D0 or D8) were missing for the other nine patients.

Fig. 2:
Flow chart.

Most patients [187 (77.9%)] were MSM, and their median age at baseline (Table 1) was 43 years (IQR: 36– 47years). The patients with free testosterone deficiency were older than the eugonadal patients (45.5 versus 43 years; P < 0.01). Most patients had CD4+ T-cell counts above 500 cells/μl, with a median CD4+ T-cell count of 632 cells/μl (IQR: 500–766 cells/ml) and a median duration of undetectable viral load of almost 3 years [33.8 months (IQR:15.6–66.6 months)]. HIV infection was diagnosed a median of 8 years before the study (IQR: 4–13 years). Hypogonadal patients had a longer duration of viral load suppression (58.8 versus 33.5 months; P = 0.02) and a lower CD4+ T-cell nadir (256 versus 302 cells/μl; P < 0.01) than eugonadal patients. No differences in CD4+ T-cell counts or time since the diagnosis of HIV infection were observed between these two groups. The median duration of exposure to cART was 47.5 months (IQR: 20.3– 100 months) and 200 (95%), 113 (47.1%), 92 (38.3%), and 52 (21.7%) patients were treated with combinations containing an NRTI, NNRTI, protease inhibitor, or integrase inhibitor, respectively. The patients with free testosterone deficiency had been receiving global cART (78.2 versus 46.2months; P = 0.04), NRTI (76 versus 45 months; P = 0.04), protease inhibitor (99.7 versus 24.3; P = 0.01), and II (55.9 versus 17.9months; P= 0.01) for longer than the eugonadal patients. High SHBG levels were detected in 112 patients (47%), and were more frequent in the hypogonadal group [15 patients (75%) versus 91 patients (43.5%); P < 0.01]. Median LH, FSH, prolactin, and estradiol levels were normal with no significant difference between groups for the first three. Analyzing the worst BMD Z-scores at any two sites, 75 patients (32.6%) met the definition of osteopenia and 36 (15.7%) met the definition of osteoporosis. Most patients were neither overweight nor obese, with a median BMI of23 (IQR: 21 –25) and only 22 patients (9.2%) had metabolic syndrome. Appendicular lean mass and appendicular lean mass index were normal in most patients, with median levels of 25.9 kg (IQR: 23.9–28.1) and 8.2 kg/m2 (IQR: 7.6– 8.8), respectively. More than half the patients reported having erectile dysfunction and deterioration in quality of life (133 patients (55.4%) in both cases), 80 patients (33.3%) had depressive symptoms.

Table 1 - Baseline characteristics of the entire cohort and according to the threshold testosterone concentration.
Normal range Entire cohort Patients (n = 240) Hypogonadism FT < 70 pg/ml Patients (n = 20) Eugonadism FT > 70 pg/ml Patients (n = 211) P
Sociodemographic parameters (n = 231)
 Age (years) 18–50 43 [36–47] 45.5 [43.5–50] 43 [36–47] <0.01
Mode of contamination/HIV
 MSM 187 (77.9)
Lifestyle (n = 231)
 Active smoking 98 (40.8) 9 (45) 85 (40) 0.81
 Alcohol > 20g/day 21 (8.8) 1 (5) 20 (9) 0.82
 Psychoactive drugs 36 (15) 2 (10) 34 (16) 0.75
Infectious parameters (n = 231)
 Duration of HIV infection (years) 8 [4–13] 9 [6–17] 7 [3–13] 0.14
 Duration of HIV load undetectability (months) 33.8 [15.6–66.6] 58.8 [41.4–80.6] 33.5 [15.5–66.1] 0.02
Immunological parameters (n = 231)
 CD4+ (cells/μl) 700–1100 632 [500–766] 580 [553–696] 633 [508–769] 0.28
 CD4+/CD8+ ratio > 1 0.9 [0.7–1.2] 0.9 [0.7–1.1] 1 [0.7–1.2] 0.68
 CD4+ nadir (cells/μl) 298 [218–404] 256 [152–299] 302 [221–414] <0.01
Therapeutic parameters
 Duration of exposure to cART (months) (n = 231) 47.5 [20.3–100] 78.2 [51.3–135] 46.2 [19.3–101] 0.04
 Duration of exposure to NRTI (months) (n = 226) 45.9 [18.9–90.8] 76 [51.3–91.6] 45 [17.9–97.1] 0.04
 Duration of exposure to NNRTI (months) (n = 113) 43.5 [19–76.2] 69.8 [46.3–80.4] 42.5 [18.7– 74.3] 0.06
 Duration of exposure to PI (months) (n = 92) 29.7 [12.9–85.2] 99.7 [28.2–152.4] 24.3 [12.8–74.6] 0.01
 Duration of exposure to II (months) (n = 52) 20.2 [9–49.8] 55.9 [55.3–68.6] 17.9 [8.2–36.9] 0.01
Ongoing cART (n = 231)
 NRTI 200 (95) 17 (85) 199 (95) 0.1
 NNRTI 113 (47.1) 10 (50) 96 (45) 0.81
 PI 92 (38.3) 10 (50) 80 (38) 0.34
 II 52 (21.7) 5 (25) 47 (22.3) 0.78
Hormonal parameters (n = 231)
 Free testosterone (pg/ml) ≥ 70 103.9 [89.2–126.9]
 Total testosterone (ng/dl) ≥ 348 585 [482–693.5]
 SHBG > 42.2nmol/la 112 (47) 15 (75) 91 (43.5) <0.01
 LH (Ul/I) 0.6–12 3.7 [2.8–4.9] 3.9 [1.9–5.5] 3.7 [2.8–4.9] 0.6
 FSH (Ul/I) 0.9–12 4.1 [2.9–5.1] 3.9 [2.7–7.1] 4.1 [2.9–5.8] 0.8
 Prolactin (ng/ml) 3.5–19.4 7 [5–10] 6 [5–7.5] 7 [5–10] 0.4
 Estradiol (pg/ml) 9–62 20 [16–26]
 Anthropometric/bone-densitometry parameters BMI (kg/m2) 18.5–24.9 23 [21–25] 24 [22–26] 23 [21–25] 0.41
Bone mineral density (BMD) (n = 230)
 Worst BMD Z-score 0.18
 Normal ≥ - 1 119 (51.7) 13 (65) 104 (50)
 Osteopenia - 1 > et > - 2 75 (32.6) 3 (15) 72 (35)
 Osteoporosis ≤ - 2 36 (15.7) 4 (20) 29 (14)
Body composition (n = 230) % Fat mass (/total mass)
 Trunk (%) 18 [14–26] 21 [15–25] 18 [14–26] 0.4
 Total (%) 18.9 [15.1–23.2] 21 [17–25] 18 [15–23] 0.2
 Appendicular lean mass (kg) > 20 25.9 [23.9–28.1] 25 [23.8–28.1] 25.9 [23.9–28.1] 0.61
 Appendicular lean mass index (kg/m2) > 7 8.2 [7.6–8.8] 7.9 [7.7–8.6] 8.2 [7.6–8.9] 0.42
Metabolic parameters (n = 231)
 Metabolic syndrome 22 (9.2) 2 (10) 19 (9) 0.7
 IIEF-5 score (n = 231) Pathological (5–20) 133 (55.4) 13 (65) 115 (55) 0.48
 HAM-D scale Pathological (≥ 8) 80 (33.3) 3 (15) 74 (35) 0.05
 AMS-scale Pathological О 27) 133 (55.4) 12 (60) 113 (53.5) 0.78
The data shown are medians [IQR] and numbers (%). IQR, interquartile range; n, number of patients. BMD, bone mineral density; cART, combined antiretroviral therapy; FT, free testosterone; II, integrase inhibitor; IIEF-5, international index of erectile function-5; NNRTI, nonnucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; PI, protease inhibitor; SHBG, sex hormone-binding globulin.
aThreshold between normal and high.

Mean serum free testosterone concentration was low (< 70 pg/ml) in 20 patients (8.7%), and mean serum total testosterone concentration was low (< 348 ng/dl) in nine patients (3.9%). Male hypogonadism was exclusively of secondary origin, as indicated by the normal or low levels of LH.

In multivariable analysis (Table 2), the parameters independently associated with male hypogonadism were age more than 43 years [adjusted odds ratio (aOR) 3.17, 95% CI 1.02–9.86; P = 0.04], total fat percentage more than 19% (aOR3.5, 95% CI 1.18–10.37; P = 0.02), and a treatment regimen including efavirenz (aOR 3.77, 95% CI 1.29–10.98; P = 0.02), all of which were identified as risk factors. A CD4+ cell count nadir more than 200cells/μl (aOR 0.22, 95% CI 0.07–0.65; P < 0.01) was found to be protective. Like IIEF-5 score, none of the infectious or metabolic parameters were associated with male hypogonadism.

Table 2 - Factors associated with male hypogonadism.
Univariable analysis Multivariable analysis
Patients (n = 231) Patients (n = 215)
OR 95% CI P aOR 95% CI P
Sociodemographic parameters Age (≥43 years) 3.66 1.29–10.44 0.01 3.17 1.02–9.86 0.04
Active smoking 1.21 0.48–3.05 0.68
 Alcohol > 20g/day 0.86 0.21–3.48 0.83
 Psychoactive drugs 0.57 0.12–2.60 0.47
Infectious parameters
 Duration of HIV infection (years) (> 10) 0.88 0.34–2.30 0.79
 Duration of HIV nondetectability (months) (> 33) 4.36 1.41–13.48 0.01
Immunological parameters
 CD4+ (cells/μl) (> 500) 0.59 0.22–1.56 0.29
 CD4+/CD8+ ratio (> 1) 0.48 0.18–1.29 0.14
 CD4+ nadir (cells/μl) (> 200) 0.25 0.09–0.66 0.01 0.22 0.07–0.65 <0.01
Therapeutic parameters
 Duration of exposure to cART (months) (> 52) 3.39 1.19–9.67 0.02
 Duration of exposure to NRTI (months) (> 48) 3.29 1.19–9.76 0.02
 Duration of exposure to NNRTI (months) (> 48) 2.32 0.91 –5.90 0.08
 Duration of exposure to PI (months) (> 33) 1.29 0.49–3.40 0.60
 Duration of exposure to II (months) (> 20) 2.37 0.80–7.07 0.12
Ongoing cart
 NRTI 0.31 0.09–1.22 0.09
 NNRTI 1.19 0.48–3.00 0.69
 Efavirenz 2.68 1.03 –6.98 0.04 3.77 1.29–10.98 0.02
 Protease inhibitors 1.64 0.65–4.10 0.29
 Integrase inhibitors 1.16 0.4–3.37 0.78
Anthropometric and bone-densitometry parameters
 BMI (kg/m2)(≥ 25) 1.67 0.61–4.61 0.32
Bone mineral density (BMD) Worst BMD Z-score (versus normal)
 Osteopenia 0.33 0.09–1.21 0.09
 Osteoporosis 1.10 0.33–3.64 0.87
Body composition % Fat mass (/total mass) (> 19)
 Total (%) 2.60 0.96–7.02 0.06 3.5 1.18 –10.37 0.02
Metabolic parameters
 Metabolic syndrome 1.12 0.24–5.21 0.88
 IIEF-5 score (pathological versus normal) 1.55 0.60–4.04 0.37
Univariable analysis: data are shown as odds ratios (ORs) with 95% confidence intervals (95% CI). Multivariable analysis: data are shown as adjusted odds ratios (aORs) with 95% confidence intervals (95% CI). BMD, bone mineral density; cART, combined antiretroviral therapy;FT, free testosterone; II, integrase inhibitor; IIEF-5, international index of erectile function-5; n, number of patients; NNRTI, nonnucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; PI, protease inhibitor; SHBG, sex hormone-binding globulin.


Male hypogonadism, which is widely associated with advanced stages ofHIV infection, used to affect up to 50% of men with AIDS [3,5]. Despite the widespread use of cART, it remains common among HIV-infected men, but its prevalence is unclear, with a wide range of estimates, from 9.3 to 70%, reported in recent studies [4,14–18,36]. The male hypogonadism prevalence of 8.7% reported here is twice that for the general population of the same age [37]. The prevalence of male hypogonadism in the general population is well established, and can therefore be compared indirectly with that of our cohort, despite being based on total testosterone levels. The use of a reliable free testosterone assay in young-to-middle-aged AIDS-free men in whom HIV replication is stably suppressed may account for the lower, but more accurate prevalence rate reported here than in previous studies [4,14–18]. Recent studies and guidelines recommend the use of free testosterone assays for male hypogonadism diagnosis in some physiological or pathological situations, including HIV infection, due to changes in SHBG levels [10,12,13,16]. This is particularly relevant in our study, in which half the patients had high SHBG levels and more than half the patients with male hypogonadism would have not been diagnosed if only total testosterone assays had been used. Recent studies have shown that low free testosterone concentration is associated with androgen deficiency-related symptoms, even in the presence of normal total testosterone concentrations [38]. Moreover, the prevalence reported here is similar to the values reported for the general population of older men [37].

Male hypogonadism is rare in young-to-middle-aged individuals in the general population, in which it is mostly attributed to a congenital pituitary or gonadal dysfunction [10,11]. Male hypogonadism may be of testicular origin, but is mostly of hypothalamic-pituitary origin in PWH [9]. Our data are consistent with published studies identifying hypothalamic-pituitary axis impairment as the exclusive cause of male hypogonadism. Compared with other systemic diseases [39], HIV infection is itself a specific cause of hypogonadism [40,41]. One of the first reasons for this is the occurrence of significant immune disorders (inappropriate secretion of cytokines) [42]. Several studies have also demonstrated a direct cytopathic effect of the virus, which may cause gonadal dysfunction [43]. Also, testosterone deficiency is strongly associated with the decline of CD4+ T-cell counts [5]. Despite the widespread use of cART, CD4+ T-cell counts remain associated with male hypogonadism [4,5,14,36]. Consistent with published findings, the CD4+ T-cell count nadir was found to be associated with male hypogonadism in our study, with values greater than 200 cells/μl considered protective. Moreover, we report for the first time a correlation between male hypogonadism and any class of antiretroviral drugs. Lipodystrophy in PWH may manifest as breast lipohypertrophy also known as gynecomastia, and may be promoted by cART regimens including efavirenz [44,45]. The aromatization of testosterone to estradiol in patients with gynecomastia could explain the association between male hypogonad-ism and efavirenz found in this study, this especially as gynecomastia has been reported to be associated with male hypogonadism [46]. However, the association between male hypogonadism and efavirenz does not disappear when estradiol is added to the multivariable model. To our knowledge, there is no other mechanism to date to explain this correlation. Moreover, several publications now report the risk of weight gain on effective cART including integrase inhibitors [47,48]. By analogy and taking into account the current widespread prescription of integrase inhibitors, it seems necessary to remain vigilant on the possible occurrence of male hypogonadism secondary to this weight gain in PWH treated with integrase inhibitors. On the contrary, as SHBG levels increase with age [10], HIV infection seems to lead to an early decline in circulating testosterone levels in PWH [17], even in AIDS history-free HIV-infected men on effective cART, as indicated in our study. As reported for other comorbid conditions observed during HIV infection [49], the early onset of male hypogonadism may reflect the premature aging of PWH [17].

Not surprisingly, age was identified as a risk factor for male hypogonadism in our study, and the patients with testosterone deficiency were older than the eugonadal patients. The correlation between age and male hypogonadism is well established [6,14,15,17,36], but our findings (based on a reliable free testosterone assay) show for the first time that this correlation is independent of viral control and of total body fat percentage. Indeed, male hypogonadism was also associated with total body fat content, as reported for obese patients with or without HIV infection in whom excess adipose tissue can provoke male hypogonadism by promoting the aromatization of testosterone to estradiol [15,17,50]. In a recent publication, low free testosterone and visceral fat were reported as independently associated to poor health status and frailty, being possible hallmarks of unhealthy conditions in young-to-middle-aged HIV-infected men. The authors concluded to an association between multimorbidity, frailty and body fat mass to each other and to sex steroids (among which free testosterone), concurring together to functional male hypogonadism in HIV [51]. This information is of clinical relevance suggesting that testosterone deficiency in PWH with visceral adiposity may potentially be reversed through weight loss and visceral fat reduction.

No association was found between male hypogonadism and osteopenia/osteoporosis neither erectile dysfunction/ low libido, nor metabolic syndrome, nor decrease in lean body mass/muscle strength, nor depression, nor deterioration in quality oflife. The young age of patients [median age of 43years (IQR: 36–47years)], their short duration of HIV infection [median of 8 years (IQR: 4–13 years)], and short duration of exposure to cART [median exposure time of 47.5 months (IQR: 20.3–100 months)] would all account for our results (Table 1). The relatively small size of the hypogonadal group could also account for our findings and represents one of the limits of this work. However, more than half the patients (entire cohort) reported having erectile dysfunction and deterioration in quality of life and third the patients had depressive symptoms. These results are in line with the data in the literature specifying the multifactorial origin of depressive disorders, erectile disorders, and impairment of quality of life in PWH [24,52,53]. A larger sample of the population may allow us to conclude.

In this study, male hypogonadism was defined biochemically. Each patient diagnosed with male hypogonadism had at least one clinical sign of male hypogonadism and was referred to a specialist and this consultation provided an opportunity to discuss the possibility of introducing testosterone supplementation. Current Endocrine Society clinical guidelines recommend adjuvant treatment with testosterone at least in the short term, in HIV-infected men with weight loss and low testosterone levels, to promote gains in muscle strength and lean body mass [12,13]. Today, the major issue in the overall care of PWH is not to ignore a disorder that could affect their quality of life. Knowing how to recognize male hypogonadism therefore seems essential. Determining the factors associated with male hypogonad-ism could help identify patients at risk and discuss the indication for testosterone supplementation.


The prevalence of male hypogonadism was not negligible in our population of young and middle-aged, AIDS history-free, HIV-infected men on effective cART. The early onset of hypogonadism may reflect the premature aging of HIV-infected patients. Our findings suggest that cART regimens including efavirenz, age over 43 years, and total body fat content exceeding 19% could be used as criteria for the identification of PWH at high risk of hypogonadism. Viral replication is stably suppressed in most patients in industrialized countries. Screening for this endocrine disorder, from venous blood samples collected early in the morning, and with a reliable free testosterone assay, is thus a key element of the overall management of HIV-infected patients.


We thank Aissi E, Alcaraz I, Allienne C, Baclet V, Duracinsky M, Huleux T, Kamenicky P, Melliez H, Meybeck A, Liotier J-Y, Riff B, Robin G, Teicher E, Valette M, Viget N for their valuable contributions. We also thank Biekre R, Choisy P, Cornavin P, Derdour V, Mole M, Vandamme S for data monitoring, and Julie Sappa for editing the manuscript. We thank the entire staff of the Department of Infectious Diseases of Tourcoing Hospital and that of the Department of Internal Medicine, Bicêtre Hospital, AP-HP. We thank the patients warmly for their participation in the study.

We thank COREVIH Nord-Pas-De-Calais (regional and French coordinating committee for the fight against sexually transmitted infections and HIV) and Merck for their financial contribution to the study.

A C and A P were the principal investigators A C, A P, M.L. designed the trial and developed the protocol. B.S. was responsible for the biochemical/hormonal investigations. L.B. was responsible for the virological investigations. P.A. was responsible for the statistical analysis. G.L. and E.D. were responsible for the investigations with Dexa. A.P., F.A., Y.Q., E.S., C.G., and A.C. enrolled patients. P.C., M.M. coordinated data collection and regulatory requirements. C.A., P.A., L.M., and F.B. interpreted the data. A.P., M.L., C.A. generated the tables and figures. M.L., A.C., and A.P. wrote the manuscript, and all authors reviewed, revised, and approved the final manuscript.

Appendix 1: Full inclusion and exclusion criteria of the study.

Full inclusion and exclusion criteria of the study are provided in the Appendix 1.

Conflicts of interest

A.C. reports grants from Viiv, Gilead, and Janssen.

M.L., A.P., F.A., B.S., Y.Q., G.L., E.D., L.B., M.M., P.C., E.S., C.G., and F.B. report no conflicts of interest that could be perceived as prejudicing the impartiality of the research reported.


1. Dolin R, Masur H, SaagM-S. AIDS therapy, third edition. Emerg Infect Dis 2008; 14:867–868.
2. Arver S, Sinha-Hikim I, Beall G, Guerrero M, Shen R, Bhasin S, et al. Serum dihydrotestosterone and testosterone concentrations in human immunodeficiency virus-infected men with and without weight loss. J Androl 1999; 20:611–618.
3. Raffi F, Brisseau J-M, Planchon B, Remi J-P, Barrier J-H, Grolleau J-Y. Endocrine function in 98 HIV-infected patients: a prospective study. AIDS 1991; 5:729–733.
4. Wunder D-M, Bersinger N-A, Fux C-A, Mueller N-J, Hirschel B, Cavassini M, et al. Hypogonadism in HIV-1-infected men is common and does not resolve during antiretroviral therapy. Antivir Ther 2007; 12:261–265.
5. Dobs A-S, Dempsey M-A, Ladenson P-W, Polk B-F. Endocrine disorders in men infected with human immunodeficiency virus. Am J Med 1988; 84:611–616.
6. Slama L, Jacobson L-P, Li X, Palella F-J, Margolick JB, Kingsley LA, et al. Longitudinal changes over 10 years in free testosterone among HIV-infected and HIV-uninfected men. J Acquir Immune Defic Syndr 2016; 71:57–64.
7. Klein R-S, Lo Y, Santoro N, Dobs A-S. Androgen levels in older men who have or who are at risk of acquiring HIV infection. Clin Infect Dis 2005; 41:1794–1803.
8. Cooper O-B, Brown T-T, Dobs A-S. Opiate drug use: a potential contributor to the endocrine and metabolic complications in human immunodeficiency virus disease. Clin Infect Dis 2003; 37:132–136.
9. Wong N, Levy M, Stephenson I. Hypogonadism in the HIV-infected man. Curr Treat Options Infect Dis 2017; 9:104–116.
10. Basaria S. Male hypogonadism. Lancet 2014; 383:1250–1263.
11. Dohle G-R, Arver S, Bettocchi C, Jones T-H, Kliesch S. EAU Guidelines on male hypogonadism European Association of Urology 2018. https://uroweb.org/wp-content/uploads/EAU-Guidelines-on-Male-Hypogonadism-2018-large-text.pdf. [Accessed 26 December 2021].
12. Bhasin S, Cunningham G-R, Hayes F-J, Matsumoto A-M, Snyder P-J, Swerdloff R-S, et al. Testosterone therapy in men with androgen deficiency syndromes: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab 2010; 95:2536–2559.
13. Bhasin S, Brito J-P, Cunningham G-R, Hayes F-J, Hodis H-N, Matsumoto A-M, et al. Testosterone therapy in men with hypogonadism: an Endocrine Society∗ Clinical Practice Guideline. J Clin Endocrinol Metab 2018; 103:1715–1744.
14. Dutta D, Sharma L-K, Sharma N, Gadpayle A-K, Anand A, Gaurav K, et al. Occurrence, patterns & predictors of hypo-gonadism in patients with HIV infection in India. Indian JMed Res 2017; 145:804–814.
15. Gomes A-R, Souteiro P, Silva C-G, Sousa-Pinto B, Almeida F, Sarmento A, et al. Prevalence of testosterone deficiency in HIV-infected men under antiretroviral therapy. BMC Infect Dis 2016; 16:628.
16. Monroe A-K, Dobs A-S, Palella F-J, Kingsley L-A, Witt M-D, Brown T-T. Morning free and total testosterone in HIV-infected men: implications for the assessment of hypogonadism. AIDS Res Ther 2014; 11:6.
17. Rochira V, Zirilli L, Orlando G, Santi D, Brigante G, Diazzi C, et al. Premature decline of serum total testosterone in HIV-infected men in the HAART-era. PLoS One 2011; 6:e28512.
18. Crum-Cianflone N-F, Bavaro M, Hale B, Amling C, Truett A, Brandt C, et al. Erectile dysfunction and hypogonadism among men with HIV. AIDS patient care and STDs 2007; 21:9–19.
19. Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 1999; 84:3666–3672.
20. Mary-Krause M, Grabar S, Lievre L, Abgrall S, Billaud E, Boué F, et al. Cohort profile: French hospital database on HIV (FHDH-ANRS CO4). Int J Epidemiol 2014; 43:1425–1436.
21. Maffezzoni F, Porcelli T, Delbarba A, Pezzaioli L-C, Properzi M, Cappelli C, et al. Hypogonadism and bone health in men with HIV. Lancet HIV 2020; 7:e782–e790.
22. Hileman C-O, Overton E-T, McComsey G-A. Vitamin D and bone loss in HIV. Curr Opin HIV AIDS 2016; 11:277–284.
23. Walker Harris V, Brown TT. Bone loss in the HIV-infected patient: evidence, clinical implications, and treatment strategies. J Infect Dis 2012; 205:S391–398.
24. Luo L, Deng T, Zhao S, Li E, Liu L, Li F, et al. Association between HIV infection and prevalence of erectile dysfunction: a systematic review and meta-analysis. J Sex Med 2017; 14:1125–1132.
25. Naidu S, Ponnampalvanar S, Kamaruzzaman S-B, Kamarulza-man A. Prevalence of metabolic syndrome among people living with HIV in developing countries: a systematic review. AIDS patient care STDs 2017; 31:1–13.
26. Rubin L-H, Maki P-M, Depression HIV. and cognitive impairment in the era of effective antiretroviral therapy. Curr HIV/AIDS Rep 2019; 16:82–95.
27. Cotter A-G, Powderly W-G. Endocrine complications of human immunodeficiency virus infection: hypogonadism, bone disease and tenofovir-related toxicity. Best Prac Res Clin Endo-crinol Metab 2011; 25:501–515.
28. Zona S, Guaraldi G, Luzi K, Beggi M, Santi D, Stentarelli C, et al. Erectile dysfunction is more common in young to middle-aged HIV-infected men than in HIV-uninfected men. J Sex Med 2012; 9:1923–1930.
29. Bhasin S, Pencina M, Jasuja G-K, Travison T-G, Coviello A, Orwoll E, et al. Reference ranges for testosterone in men generated using liquid chromatography tandem mass spectro-metry in a community-based sample of healthy nonobese young men in the Framingham Heart Study and applied to three geographically distinct cohorts. J Clin Endocrinol Metab 2011; 96:2430–2439.
30. Rosen R-C, Cappelleri J-C, Smith M-D, Lipsky J, Pena B-M. Development and evaluation of an abridged, 5-item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. Int J Impot Res 1999; 11:319–326.
31. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23:56–62.
32. World Health Organization. WHO scientific group on the assessment of osteoporosis at primary healthcare level. WHO;2007. Available at: https://www.sheffield.ac.uk/FRAX/pdfs/WHO_Technical_Report.pdf.[Accessed 26 December 2021].
33. Cruz-Jentoft A-J, Bahat G, Bauer J, Boirie Y, Bruyere O, Ceder-holm T, et al. Sarcopenia: revised European consensus on definition and diagnosis. Age Ageing 2018; 48:16–31.
34. Expert panel on detection, evaluation, and treatment of high blood cholesterol in adults. Executive summary of the third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001; 285:2486–2497.
35. Matthews D-R, Hosker J-P, Rudenski A-S, Naylor B-A, Treacher D-F, Turner R-C. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28:412–419.
36. Bajaj S, Pathak Y, Varma S, Verma S. Metabolic status and hypogonadism in human immunodeficiency virus-infected males. Indian J Endocrinol Metab 2017; 21:684–687.
37. Araujo A-B, O’Donnell A-B, Brambilla D-J, Simpson W-B, Longcope C, Matsumoto A-M, et al. Prevalence and incidence of androgen deficiency in middle-aged and older men: estimates from the Massachusetts Male Aging Study. J Clin En-docrinol Metab 2004; 89:5920–5926.
38. Antonio L, Wu F-C-W, O’Neill T-W, Pye S-R, Ahern T-B, Laurent M-R, et al. Low Free testosterone is associated with hypogonadal signs and symptoms in men with normal total testosterone. J Clin Endocrinol Metab 2016; 101:2647–2657.
39. Baker HW. Reproductive effects of nontesticular illness. En-docrinol Metab Clin North Am 1998; 27:831–850.
40. Dobs AS. Androgen therapy in AIDS wasting. Clin Endocrinol Metab 1998; 12:379–390.
41. Poretsky L, Can S, Zumoff B. Testicular dysfunction in human immunodeficiency virus-infected men. Metabolism 1995; 44:946–953.
42. Roubenoff R, Grinspoon S, Skolnik P-R, Tchetgen E, Abad L, Spiegelman D, et al. Role of cytokines and testosterone in regulating lean body mass and resting energy expenditure in HIV-infected men. Am J Physiol Endocrinol Metab 2002; 283:E138–E145.
43. Ferreiro J, Vinters H-V. Pathology of the pituitary gland in patients with the acquired immune deficiency syndrome (AIDS). Pathology 1988; 20:211–215.
44. Lagathu C, Béreziat V, Gorwood J, Fellahi S, Bastard J-P, Vigouroux C, et al. Metabolic complications affecting adipose tissue, lipid and glucose metabolism associated with HIV antiretroviral treatment. Expert Opin Drug Saf 2019; 18:829–840.
45. Tebas P, Sension M, Arribas J, Duiculescu D, Florence E, Hung C-C, et al. Lipid levels and changes in body fat distribution in treatment-naive, HIV-1-infected adults treated with rilpivirine or Efavirenz for 96 weeks in the ECHO and THRIVE trials. Clin Infect Dis 2014; 59:425–434.
46. Biglia A, Blanco J-L, Martinez E, Domingo P, Casamitjana R, Sambeat M, et al. Gynecomastia among HIV-infected patients is associated with hypogonadism: a case-control study. Clin Infect Dis 2004; 39:1514–1519.
47. Sax P-E, Erlandson K-M, Lake J-E, McComsey G-A, Orkin C, Esser S, et al. Weight gain following initiation of antiretroviral therapy: risk factors in randomized comparative clinical trials. Clin Infect Dis 2020; 71:1379–1389.
48. McComsey G-A, Moser C, Currier J, Ribaudo H-J, Paczuski P, Dubée M-P, et al. Body composition changes after initiation of raltegravir or protease inhibitors: ACTG A5260 s. Clin Infect Dis 2016; 62:853–862.
49. Deeks S-G. HIV infection, inflammation, immunosenescence, and aging. Annu Rev Med 2011; 62:141–155.
50. Kalyani R-R, Dobs A-S. Androgen deficiency, diabetes, and the metabolic syndrome in men. Curr Opin Endocrinol Diabetes Obes 2007; 14:226–234.
51. De Vincentis S, Decaroli M-C, Fanelli F, Diazzi C, Mezzullo M, Morini F, et al. Health status is related to testosterone, estrone and body fat: moving to functional hypogonadism in adult men with HIV. Eur J Endocrinol 2021; 184:107–122.
52. Nanni M-G, Caruso R, Mitchell A-J, Meggiolaro E, Grassi L. Depression in HIV infected patients: a review. Curr Psychiatry Rep 2015; 17:530.
53. Jin Y, Liu Z, Wang X, Liu H, Ding G, Su Y, et al. A systematic review of cohort studies of the quality of life in HIV/AIDS patients after antiretroviral therapy. Int J STD AIDS 2014; 25:771–777.

antiretroviral therapy; free testosterone; HIV; hypogonadism; testosterone

Supplemental Digital Content

Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc.