If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
PURPOSE: Weight loss is a strong predictor of morbidity and mortality in human immunodeficiency virus (HIV)-infected patients. Men with acquired immunodeficiency syndrome (AIDS) lose body cell mass. Hypogonadism is also common. This study tested the efficacy of a testosterone transscrotal patch (6 mg/day) in improving body cell mass and treating hypogonadism in these patients.
SUBJECTS AND METHODS: This multicenter, randomized, double-blinded, placebo-controlled trial was conducted from August 1995 to October 1996 in 133 men, 18 years of age and older, who had AIDS, 5% to 20% weight loss, and either a low morning serum total testosterone level (<400 ng/dL) or a low free testosterone level (<16 pg/mL). Outcomes included weight, body cell mass as measured using bioelectrical impedance analysis, quality of life, and morning measurements of serum testosterone and dihydrotestosterone levels, lymphocyte subsets, and HIV quantification.
RESULTS: There were no significant differences in baseline weight, CD4 cell counts, or HIV serum viral quantification between treatment arms. Morning total and free testosterone levels increased in those treated with testosterone, but not with placebo. Following 12 weeks of treatment there were no differences (testosterone—placebo) in mean weight change (−0.3 kg [95% confidence interval (CI): −1.4 to 0.8]) or body cell mass (−0.2 kg [95% CI: −1.0 to 0.6]) in the two groups. There were also no changes in quality of life in either group.
CONCLUSION: Hypogonadal men with AIDS and weight loss can achieve adequate morning serum sex hormone levels using a transscrotal testosterone patch. However, this system of replacement does not improve weight, body cell mass, or quality of life.
Weight loss and wasting in persons with advanced human immunodeficiency virus (HIV) infection are strong predictors of morbidity and mortality (
) have suggested that the time of death from wasting is related to the magnitude of body cell mass depletion: There is a critical body cell mass for survival, and maintaining this mass could prolong survival. Men with acquired immunodeficiency syndrome (AIDS) lose muscle with little loss of fat (
) demonstrated that intramuscular testosterone replacement led to significant increases in fat-free, lean body, and muscle mass, as well as improved quality of life.
In this study we evaluated the effects of testosterone replacement using a transscrotal testosterone patch (6 mg/day) in men with hypogonadism and HIV-related wasting.
This multicenter, randomized, double-blinded, placebo-controlled trial compared the effects of transdermal scrotal testosterone patches (Testoderm, ALZA Corporation, Palo Alto, California) with identical placebo patches. Each testosterone patch system contained 15 mg of testosterone and was designed to release controlled amounts of testosterone continuously upon application to the scrotal skin. In hypogonadal men without HIV infection, this system provides delivery of 6 mg testosterone per day. The patch has two layers: a soft flexible backing of polyethylene terephthalate and a drug-containing film of ethylene-vinyl acetate copolymer impregnated with testosterone (
After we explained the protocol and obtained signed consent, each subject underwent a baseline evaluation to determine eligibility, including HIV testing, verification of weight loss, and measurement of morning serum testosterone level. Enrolled participants were seen at weeks 2, 4, 8, and 12 after randomization. Weight was measured at all visits using the same instrument at each site. Serum hormone levels, bioelectrical impedance assessment of body composition, and quality of life questionnaires were measured at baseline and weeks 4, 8, and 12. There was no structured exercise program.
The testosterone and placebo patch systems were identically packaged in foil-lined pouches. Patches were to be applied each morning and worn for 22 to 24 hours each day. Men were instructed to dry shave their scrotum weekly and use a hair dryer to warm the patch before application for better adherence.
A private nutritionist provided nutritional evaluations based on 24-hour dietary recall at weeks 0, 4, 8, and 12. Each site faxed intake information to the study nutritionist after the patient visit. Total caloric and protein intake were estimated, and an assessment of their adequacy was faxed back to the site. If intake was not adequate, the site was instructed to counsel the patient on proper nutrition. The protocol was reviewed and approved by the human subjects committee at each participating institution.
Men, 18 years of age and older, with a diagnosis of AIDS (based on 1993 Center for Disease Control criteria) were eligible for enrollment. Subjects were required to have lost between 5% and 20% of their baseline weight, as noted in their primary provider’s written records. Additional enrollment criteria included either a morning serum total testosterone level ≤400 ng/dL or a morning serum free testosterone level ≤16 pg/mL. Subjects could not have received testosterone or anabolic steroids within 6 months before entry into the study. Laboratory enrollment criteria included a prostate specific antigen (PSA) level <4 ng/mL, serum aminotransferase levels <5 times the upper limit of reference range, hemoglobin level >8.5 g/dL, and a hematocrit <54%. There were no restrictions on antiretroviral regimen or therapy or prophylaxis for pneumocystis carinii or Mycobacterium avium complex infections. Subjects were excluded if they had an active or recent opportunistic infection requiring systemic anti-infective agents, any predisposing condition affecting oral ingestion or food absorption, or concurrent therapy with medications known to affect testosterone metabolism or alter its serum levels (ie, spironlactone, megestrol acetate, dronabinol, or ketoconazole).
Subjects were randomly assigned in equal proportions to either the active treatment or placebo group.
Morning serum was collected at weeks 0, 2, 4, 8, and 12 for measurement of total and free testosterone levels. All hormones were measured using standard radioimmunoassay techniques at one laboratory (Scicor). Corning Hazleton Laboratories assayed dihydrotestosterone levels after the study’s completion, using serum from day 0 and week 12 of the study.
Body cell mass was measured by single frequency bioelectrical impedance analysis using the BIA 101Q instrument (RJL Systems, Inc., Clinton Township, Michigan) and the “Fluid and Nutrition Analysis Version 3.1” software, provided by the manufacturer. Bioelectrical impedance analysis is a validated, noninvasive method to determine body cell mass in a three-compartment model (
). Measured resistance, reactance plus standard anthropometric measurements, and the patient’s age were used to determine body cell mass using prediction equations (Operations and Instructions Manual, RJL Systems, Inc.).
To measure the effect of the intervention on HIV disease progression, serum specimens were collected at weeks 0 and 12 for measurement of lymphocyte subsets and plasma HIV-1 quantification. Plasma HIV-1 quantification was assayed by the Chiron Reference Testing Laboratory (Emeryville, California) using signal amplification technology for performing ultrasensitive branched-chain DNA assays, with a limit of detection of 500 equivalents/mL.
Several self-assessment quality of life scales were administered at baseline and at weeks 4, 8, and 12 to evaluate overall mental health, pain, health perceptions, energy/fatigue, health distress, quality of life, health transitions, and cognitive/physical/social/sexual functioning. The validated instruments employed included the Rand HIV-Medical Outcomes Study short form instrument (
Sample size calculations were done to provide 90% power to detect a difference of 2.0 kg in the change in body cell mass from baseline to 12 weeks after treatment between the two groups, assuming a standard deviation of 3.0 kg and a two-sided alpha of 0.05. To allow for 20% attrition, the anticipated sample size was 124 subjects. The primary outcome variable was mean change in body cell mass evaluated by bioelectrical impedance analysis from baseline to week 12. Secondary efficacy parameters were mean change in body weight from baseline to week 12 and quality of life parameters. Two-sample t tests, chi-square tests, or two-way analysis of variance (ANOVA) were used as appropriate. All analyses were performed in SAS. For subjects who terminated the study before 12 weeks, the last “on-treatment” value was carried forward for analysis. This value had to be obtained within 5 days of termination of study treatment. Statistical significance was set at P <0.05 (two-sided). Continuous data are reported as mean ± SD.
Of the 133 patients enrolled in the study, 67 received testosterone and 66 received placebo. A total of 110 men had at least one follow-up evaluation and were included in the analysis. Fifteen (23%) of the testosterone-treated subjects and 20 (30%) of the placebo-treated subjects did not complete the study. Reasons for premature termination were not significantly different between the treatment groups.
Characteristics of the subjects were similar in the active treatment and control groups (Table 1). About half had a history of prior opportunistic infections. There were no significant between-group differences in weight, body cell mass, total and free testosterone levels (Figure 1, Figure 2 ) , and CD4 cell counts.
Table 1Baseline Characteristics of All Enrolled Subjects
At baseline, 44% of subjects had serum total testosterone levels ≤400 mg/dL, and 75% had free testosterone levels ≤16 pg/mL. Testosterone patches were effective in increasing serum testosterone levels (Figure 1, Figure 2). The mean total testosterone level in the subjects treated with testosterone patches at week 12 was 697 ± 327 ng/dL, a mean increase of 277 ± 43 ng/dL. Subjects treated with placebo had no significant change in total testosterone levels at week 12. Similarly, changes in free testosterone levels were apparent in testosterone-treated subjects by week 4 and maintained through the study, with a mean level of 24 ± 11 pg/dL at week 12 (a mean change of 11 ± 2 pg/mL). No changes were observed in the placebo-treated subjects.
Serum dihydrotestosterone levels were similar in the two groups at baseline and increased only in the treatment group (to 183 ng/dL; 95% confidence interval [CI]: 197 to 220 ng/dL). No significant changes were seen in the placebo group.
Use of testosterone or placebo patches did not significantly affect body cell mass or weight. After up to 12 weeks, the difference in weight change between groups (testosterone—placebo) was −0.3 kg (95% CI: −1.4 to 0.8 kg, P = 0.6). The difference in change in body cell mass, −0.2 kg (CI: −0.3 to 0.6 kg, P = 0.6), was not statistically significant. Subgroup analysis by baseline total testosterone levels (<400, ≥400 ng/dL), free testosterone levels (<16 pg/mL, ≥16 pg/mL), and CD4 counts (<200, ≥200 cells/mm3) showed no significant testosterone-induced changes. Testosterone replacement failed to correct the pretreatment weight loss.
The majority of patients in both groups consumed adequate calories and protein as determined by 24-hour recall: at week 12, 76% of testosterone-treated and 80% of placebo-treated subjects consumed adequate calories, and 64% of testosterone-treated and 61% of placebo-treated subjects had adequate protein intake. There were no statistically significant differences in caloric or protein consumption between the two groups.
There were no significant differences between the testosterone and placebo groups in the changes in either CD lymphocyte subset (as absolute numbers or percentages), or plasma quantification of HIV-1. CD4 lymphocyte counts increased by 18 to a mean of 166 ± 171 cells/mm3 in the testosterone group, and by 24 to a mean of 193 ± 180 cells/mm3 in the placebo group (P = 0.74). The mean difference between treatment groups in change in HIV serum quantification during the course of the study was −0.1 log mEq/mL (95% CI: −0.4 to 0.1 log mEq/mL).
Quality of life measures
No statistically significant differences were observed between the testosterone and placebo groups in changes in overall health, pain, functioning, role or social functioning, mental health, energy/fatigue, health distress, cognitive functioning, quality of life, or health transition at baseline using the Rand HIV-Medical Outcomes Study short form instrument. A trend toward improvement with testosterone was noted in two areas: “overall health” (P = 0.05) and “cognitive functioning” (P = 0.08).
Using the EUROQOL Feeling Thermometer, the testosterone and placebo groups were comparable at baseline, and neither group had statistically significant changes in mood at any time points measured. The testosterone group had a mean score of 73 ± 24 at week 12; the placebo group had a mean score of 74 ± 21. No differences were noted between the testosterone and placebo groups in sexual function and satisfaction at baseline, nor were any changes noted at 4, 8, or 12 weeks in either group. The patches were well-tolerated. Adverse event rates, including any local reaction to the patch, were comparable in the testosterone (1.4%) and placebo (1.5%) groups.
This multicenter, placebo-controlled study demonstrated that hypogonadal men with AIDS and documented 5% to 20% weight loss could achieve adequate morning serum testosterone levels using a transscrotal testosterone delivery system applied in the morning. However, there was no effect on weight, body cell mass, measured quality of life parameters, or sexual function. There were no differences in adverse events between subjects treated with testosterone or placebo, and there were no differences in changes in CD4 lymphocyte counts and percentages, or plasma HIV-1 RNA quantification. Serum dihydrotestosterone levels were greater than normal in testosterone-treated subjects, likely due to the high level of 5 alpha-reductase activity in scrotal skin. The clinical significance of this increase is unknown.
There are several possible explanations for why replacement with testosterone failed to reverse weight loss. Progressive malnutrition and body cell mass depletion in AIDS patients can be resistant to therapy, even during periods of clinical stability, and this may include resistance to the anabolic effects of testosterone replacement (
) reported growth hormone deficiency along with hypogonadism in patients with AIDS and suggested that the hypogonadism was a result of the pituitary dysfunction associated with growth hormone deficiency. In that case, testosterone replacement alone would not be expected to be sufficient. The role of cytokines in HIV-related weight loss also needs further consideration (
) influence the effects of exogenous androgens. There was no systematic exercise component to the intervention, and lack of exercise may have limited the efficacy of testosterone.
Bioelectrical impedance analysis may have been unable to detect changes in body cell mass, although this is unlikely. Body composition measurements by bioelectrical impedance analysis in AIDS patients are strongly correlated with the results of deuterium and sodium chloride (22NaCl) dilution (
). Among HIV-infected men, there are no significant differences between mean lean-body mass estimates obtained by total body electrical conductivity using a prediction equation based on body mass index, or by bioelectrical impedance analyses (
). Bioelectrical impedance is less accurate in estimating body cell mass if there has been a change in the normally constant relation between body fluid and lean body mass, as may occur in patients with AIDS who have changes in total body water and its distribution, opportunistic infections, and abnormal hydration states (
). Subjects in this study did not have any opportunistic infections that required systemic therapy, nor did they have intractable diarrhea or marked dehydration. Moreover, although testosterone may increase sodium absorption, body weight did not change significantly during the study. Consequently, there are no reasons to suspect any major changes in the relation between lean body mass and body fluid content during the study.
The lack of improvement in several measures of quality of life with testosterone replacement therapy is somewhat surprising. This suggests that adequate levels of hormone were not maintained throughout the day, or that the subjects were not truly hypogonadal at baseline. A recent study demonstrated that intramuscular testosterone replacement increased fat-free, lean body, and muscle mass in HIV-infected hypogonadal men (
). Perhaps the transscrotal delivery of testosterone was ineffective at maintaining hormone levels or did not achieve peak serum levels necessary for an anabolic effect. Given the lack of effect on quality of life and sexual function, we believe the transscrotal patch did not maintain adequate serum testosterone levels throughout the day.
Our sample may have been too small to detect small changes in quality of life. Baseline quality of life was relatively high and offered limited opportunity for improvement. Our instruments may have been inadequate, or the 12-week duration of study may have been too short, to detect relatively small changes. Finally, there are conflicting reports of the prevalence of hyperprolactinemia in HIV-infected men (
). Correction of hyperprolactinemia, if present, may be required before testosterone hormone replacement is effective in correcting sexual dysfunction.
Because AIDS-related weight loss is a complex problem, patients require evaluation of their nutritional needs, gastrointestinal absorptive and digestive capabilities, metabolic demands, exercise capabilities, psychosocial concerns, cytokine levels, and serum hormone levels. Treatment of HIV-associated weight loss will likely require a multifaceted approach. The role of anabolic steroids requires further investigation. Although it is reassuring that testosterone replacement did not adversely affect immune status, this study suggests that testosterone replacement administered via a scrotal patch does not reverse the loss of body cell mass or improve quality of life.
Weight loss prior to clinical AIDS as predictor of survival. Mulitcenter AIDS Cohort Study Investigators.