This technique is based on the differential resistance to a low-intensity electrical current by the fat and lean compartments of the body. Measured values of resistance and reactance obtained by bioimpedance analysis (BIA) can be used in regression equations (eg, see Kotler et al [83]) to estimate fat, LBM, and total body water. There is less consensus regarding the ability of BIA to estimate body cell mass and intracellular and extracellular water with accuracy. The equipment is relatively inexpensive and portable. A measurement takes only approximately 10 minutes and involves no pain or discomfort for the patient. However, accurate estimation of body composition using this technique depends to a large extent on accurate measurement of weight and height, as well as correct and consistent positioning of electrodes. Because there are no population standards for fat or lean tissue, BIA cannot be used to diagnose wasting. However, careful and consistent measurements with BIA can be useful for monitoring changes over time.

Sequential measurements of midarm circumference, using a tape measure, and triceps skinfold, using calipers, can be used as surrogate indicators of change in arm muscle circumference, which has been shown to predict survival in other catabolic illnesses.(84) Although these and other anthropometric measurements can also be used to estimate whole-body composition, the equations used in such calculations have not been validated in patients with HIV-associated wasting. Anthropometric measurements are highly technique dependent. Duplicate or triplicate measurements should be obtained by a trained clinician, and serial measurements should be performed by the same individual whenever possible.

There are a variety of other research methods for measuring whole body composition, including dual-energy x-ray absorptiometry (DEXA), dilution techniques, underwater weighing, magnetic resonance imaging (MRI), computed tomography (CT), total body electrical conductivity (TOBEC), and whole-body counting of potassium, nitrogen, and other elements. Regional body composition analysis can be performed using DEXA, MRI, or CT. All of the techniques listed provide more accurate estimates of body composition than BIA or anthropometry but generally involve more costly equipment and a require a greater degree of patient cooperation and time. Although useful in research, they are not considered to be necessary for standard nutritional assessment.

Increases in net daily energy intake can be achieved with the use of oral supplements, despite some compensatory decrease in self-selected food consumption. Such supplements can be very useful in individuals for whom an inability or unwillingness to prepare or consume meals becomes an impediment to oral intake. A variety of liquid and solid oral supplements are available, including conventional preparations and specialized formulas for patients with specific intolerances (eg, fat malabsorption or lactose intolerance). Elemental formulas provide another option for individuals with malabsorptive disorders. Some small studies suggest an increased benefit from special oral preparations containing specific amino acids and proprietary agents for people with HIV infection.(91-95) However, until further data become available, the primary criteria for selection of a specific supplement should be tolerability and cost.

In the short term, repletion or maintenance of weight by enteral or parenteral routes might be considered in individuals who are unable to meet nutritional goals with oral intake because of profound anorexia, nausea, oral or esophageal lesions, diarrhea, or neurologic disorders, but in whom there is potential for stabilization or improvement. Although it may be common for patients receiving enteral or parenteral feeding to gain weight,(96-103) increases in LBM are a less consistent finding.(96,97,99,100) Two studies suggest that survival may be increased in patients receiving enteral or parenteral nutritional support, when compared with those receiving standard nutritional counseling (104) or those who decline supplemental feeding.(98) A major factor in choosing a specific nonvolitional feeding technique is the importance of using the gastrointestinal tract to the greatest extent possible.

Individuals with full or mildly impaired gut function might be candidates for short-term nasogastric tube feeding or, for longer periods, percutaneous endoscopic gastrostomy (PEG) or percutaneous endoscopic jejunostomy (PEJ). In some cases these approaches can use standard or elemental enteral formulas, so the cost of the nutrition itself is considerably less than for parenteral feeding. Placement of the PEG tube is a relatively simple procedure, and routine care and maintenance can be performed by the patients at home. The most common complication is superficial skin infection; other potential complications include aspiration, necrotizing fasciitis, and colocutaneous fistulas.

Provision of central or peripheral parenteral nutrition may nutritionally stabilize and maintain hydration in patients who experience a loss of gastrointestinal function. The costs and risks of this therapeutic maneuver are greater than for enteral approaches, and there is no widespread consensus regarding the appropriate use of this technique in individuals with advanced HIV infection.

A synthetic progestational agent, megestrol acetate, has been shown to be a potent appetite stimulant and is approved by the U.S. Food and Drug Administration (FDA) for the treatment of HIV-associated anorexia. In two randomized, double-blind, placebo-controlled trials of patients with HIV-associated weight loss who were studied before the availability of effective ART, treatment with megestrol acetate significantly increased food intake, weight, self-reported appetite, and sense of well-being.(105,106) In one of these trials, doses of 100, 400, and 800 mg/day were compared, and energy intake, weight, and sense of well-being increased in a dose-dependent manner.(105) In both trials, the majority of weight gained was fat. For example, patients treated with 800 mg megestrol acetate per day for 12 weeks gained an average of 3.5 kg, but only 1.1 kg was LBM.(105) In another trial, treatment with the same dose of megestrol acetate resulted in a 4.5-kg increase in fat with no change in LBM over a comparable treatment period.(106 Weight gain of an even greater magnitude was reported in an open-label pharmacokinetic study of megestrol acetate, but body composition was not measured.(107) It should be noted that healthy, HIV-negative subjects who lose weight due to voluntary or involuntary restriction of energy intake also tend to regain a disproportionate amount of fat during the early stages of weight recovery.(108) However, the failure of HIV-infected patients treated with megestrol acetate to gain LBM may also derive from the hypogonadism that has been shown to occur in men treated with this agent.(60,109) To determine whether testosterone replacement could improve the proportion of weight gained as LBM in patients receiving megestrol acetate, a multicenter study was performed in which men with wasting were treated with megestrol acetate (800 mg/day) and randomly assigned to receive a replacement dose of testosterone enanthate (200 mg every other week by intramuscular injection) or placebo.(110) Most of the participants in this study received effective ART. Preliminary data suggest that, consistent with earlier studies, there was significant weight gain with megestrol acetate, but, in contrast to previous studies, approximately half of the weight gained was lean tissue. Coadministration of testosterone in replacement doses was not associated with this increase in lean tissue accrual. However, sexual functioning was retained in those who received testosterone and diminished significantly in those on placebo during megestrol acetate treatment.

In addition to hypogonadism, adrenal suppression,(110,111) deep venous thrombosis,(112) and avascular necrosis (113) have been reported in patients receiving megestrol acetate.

A synthetic form of delta-9 tetrahydrocannabinol (an active ingredient in marijuana), dronabinol is approved by the FDA for treatment of HIV-associated anorexia. Although open-label studies of dronabinol in patients with cancer (114,115) and HIV infection (116) demonstrated a consistent improvement in self-reported appetite, weight did not increase consistently. A double-blind, placebo-controlled, crossover study of dronabinol in a small number of HIV-infected men with weight loss produced no significant increases in weight or self-reported appetite or energy intake.(117) There was also no significant increase in weight in a larger, randomized, double-blind, placebo-controlled, multicenter trial of dronabinol in patients with HIV-associated weight loss.(118) Average weight change in patients randomized to receive dronabinol (2.5 mg two times per day; n = 50) was +0.1 kg over the 6-week treatment period, as compared with -0.4 kg in those who received placebo (n = 38; p = .14). Although there was no change in weight, self-reported appetite, nausea, and mood improved significantly over the treatment period. In a pharmacokinetic study, patients treated with dronabinol alone failed to gain weight, and those who received dronabinol in combination with megestrol acetate experienced no greater weight gain than those who received megestrol acetate alone.(107)

Some patients have suggested that smoked marijuana may be more effective than dronabinol because the dose can be more effectively titrated to the desired effect. Recently, the effects of smoked marijuana, oral dronabinol, and oral placebo on virologic and immunologic parameters were evaluated in a 3-week, randomized, partially blinded, metabolic ward study.(119) Subjects were allowed to eat ad libitum during the study. Significant weight gain occurred in all three groups, with greater weight gain seen in the two groups randomized to treatment with cannabinoids. However, in all three groups, the composition of the weight gained was predominantly (~80%) fat, and there was a tendency toward greater fat accrual in the central, rather than peripheral regions.(120) The most frequently reported adverse effects of dronabinol are euphoria, dizziness, and thinking abnormalities.(118)

The antihistamine cyproheptadine increased energy intake and weight in a small open-label study in patients with HIV-associated weight loss.(121) Glucocorticoids have also been used to stimulate appetite, but these agents have specific protein catabolic effects and are therefore not recommended as a therapy for HIV-associated wasting.

Administered in pharmacologic doses, recombinant human growth hormone (GH) caused significant weight gain and retention of nitrogen and potassium during a short-term metabolic ward study,(45) and increases in weight and LBM in a 3-month open-label study.(122) In a randomized, double-blind, placebo-controlled trial in 178 patients with HIV-associated wasting, treatment with GH (0.1 mg/kg/day) for 3 months produced sustained and significant (p < .001 vs placebo) increases in weight (+1.6 kg) and LBM (+3.0 kg) that were accompanied by a decrease in fat (-1.7 kg).(123) Treadmill work output at volitional exhaustion increased significantly in patients treated with GH, and changes in work output and time to exhaustion were significantly and positively correlated with changes in LBM. Adverse effects included arthralgias, myalgia, puffiness, and diarrhea, which generally responded to dose reduction. All patients were required to be maintained on ART throughout the study, and there was no acceleration of viral replication with GH. Preliminary results from a recently completed study in 757 HIV-infected patients with 10% weight loss indicate that GH (6 mg daily or every other day) increased LBM, decreased fat, and improved performance on a cycle ergometer in patients in the current treatment era,(124) thus confirming the results of prior studies treatment.

In two separate placebo-controlled trials, combinations of pharmacologic doses of insulinlike growth factor-1 (total 10 mg/day) with smaller doses of GH (total 1.4 mg/day [125] or 0.7 mg/day[126,127]) produced increases in LBM that were comparable to or less than those obtained with GH alone.

Because weight loss in patients with HIV infection occurs most rapidly during episodes of acute secondary infection, it has been hypothesized that treatment during these periods might prevent or mitigate this weight loss.(41) Two randomized, double-blind, placebo-controlled studies of short-term treatment with GH in HIV-infected subjects with newly diagnosed opportunistic infections have shown that treatment with GH can induce weight gain and accrual of LBM.(128,129) However, it is unclear whether there is any long-term advantage to such short-term treatment. Because treatment with GH in non-HIV-infected patients hospitalized in an intensive care unit has been associated with greater mortality rates,(130) GH treatment is contraindicated in HIV-infected patients during critical illness. However, it should be noted that no such deleterious effects were observed in the two aforementioned studies of GH treatment in HIV-infected patients with mild to moderately severe secondary infections.

The FDA granted accelerated approval for GH in HIV-associated wasting at a dose of 6 mg/day for patients with weight >55 kg and lower doses for weight <55 kg. The cost of this recombinant agent limits its accessibility, and the optimal therapeutic and maintenance dosing regimens have not yet been identified. Data are limited regarding whether weight is maintained following discontinuation of GH treatment.

In HIV-negative hypogonadal men, testosterone replacement produced striking increases in weight and LBM.(131,132) In randomized, double-blind, placebo-controlled studies of HIV-positive men with reduced serum testosterone levels, weight and LBM have also increased after testosterone treatment, but to a more modest extent than in hypogonadal HIV-negative men.(133-136) These less robust responses may be a result of the fact that in all of the studies in men with HIV infection, the average baseline serum testosterone levels were not as low as those in the seronegative men.

Two randomized, placebo-controlled studies of injectable testosterone, alone or in combination with resistance exercise, have also been performed in HIV-infected men with wasting, one using a replacement dose (100 mg/week) in men with low serum testosterone levels (137) and the other using a supraphysiologic dose (200 mg/week) in eugonadal men.(138) In both studies, significant increases in LBM occurred with either testosterone or exercise, and in the latter study these effects tended to be additive. Taken together, these studies suggest that testosterone can increase LBM, muscle mass, and function in men with wasting and low serum testosterone levels, and a pharmacologic dose can similarly improve lean tissue in eugonadal men with wasting. In some cases, these changes have been associated with improvements in self-assessed quality of life and indices of depression.(139)

Based on evidence suggesting that HIV-infected women with wasting may have lower free testosterone levels than seronegative controls,(65) a randomized, double-blind, placebo-controlled study was performed in 53 women with HIV-associated wasting.(140) A specially prepared transdermal system was developed to deliver 0, 150, or 300 (µg/day in the three study groups. A significant increase in weight (+1.9 kg [4%]) was seen in the women receiving physiologic testosterone (150 (µg/day), but the weight gain was primarily fat.

Physiologic testosterone replacement has not been associated with significant adverse effects in any of the aforementioned studies in men. However, testosterone in pharmacologic doses decreases high-density lipoprotein (HDL) cholesterol levels, and thus might increase risk of cardiovascular disease. In addition, the effects of pharmacologic testosterone on the risk of prostate cancer in the setting of HIV infection are unknown. Testosterone enanthate or cypionate is administered by intramuscular injection. Transdermal preparations (both by patch or gel) for men are also commercially available.

An injectable derivative of 19-nortestosterone, nandrolone decanoate has a relatively long half-life. Although nandrolone is approved only for the treatment of anemia associated with chronic renal failure, there has been considerable off-label use of this agent in patients with HIV infection. Significant increases in weight and LBM (generally 3-4 kg) have been seen in open-label studies in men using doses ranging from 100 mg every 2 weeks to 600 mg/week for treatment periods of 12-16 weeks.(141-144) Studies using higher doses have not produced proportionally greater increases in LBM. In one study, subjects who were randomly assigned to undergo supervised progressive resistance exercise during nandrolone treatment had greater increases in weight, LBM, and strength than those receiving nandrolone with no exercise, demonstrating that the protein anabolic effects of nandrolone can be augmented by concurrent resistance exercise.(144)

In preliminary reports from separate randomized, double-blind, placebo-controlled studies in men (145) and women (146) with a documented weight loss of >=5%, treatment with nandrolone (200 mg/week in men and 100 mg every other week in women) for 12 weeks was associated with significant increases in LBM with no changes in fat. Men who were randomized to nandrolone had significant decreases in total testosterone, sex hormone-binding globulin, follicle-stimulating hormone, luteinizing hormone, and HDL cholesterol levels, and increases in hemoglobin and hematocrit. In women, some hoarseness and hirsutism were reported. Nandrolone had no significant effect on liver enzymes in either study.

An oral testosterone derivative, oxandrolone is approved by the FDA as a short-term treatment for weight loss incurred in conjunction with surgery, chronic infection, trauma, or prolonged use of corticosteroids. There is no HIV-specific indication for this agent. At the approved dosing level (5-20 mg/day), there appears to be less potential for virilizing effects and hepatic toxicity than has been seen with other oral agents. In a placebo-controlled study in 63 HIV-infected men with >10% weight loss, those randomized to receive oxandrolone in a dose of 15 mg/day gained an average of 0.6 kg at the end of the 16-week study period; weight was unchanged in those who received 5 mg/day and decreased in those on placebo (-1.1 kg).(147) The composition of the weight change was not measured. Oxandrolone treatment was associated with improvements in self-reported appetite and activity, and there were no reported toxicities. In HIV-infected men with weight loss, a 20-mg/day dose of oxandrolone in combination with resistance exercise and testosterone replacement produced striking increases in LBM (+6.9 kg) and indices of strength.(148) As with pharmacologic testosterone and nandrolone, oxandrolone treatment in this latter study produced significant decreases in HDL cholesterol. Higher doses of oxandrolone have been studied in HIV-associated wasting, but no data are available on either safety or efficacy.

Another oral agent, oxymetholone, was reported to produce weight gain (mean 5.7 kg) and improvements in Karnofsky score in an early open-label study in patients with HIV-associated weight loss.(149) More recently, a double-blind, placebo-controlled study was performed in which 92 men and women with wasting received oxymetholone in total daily doses of 100 or 150 mg or placebo.(150) Preliminary data indicate that both dosing levels of oxymetholone produced significant increases in weight and LBM (+3.7 and +2.7 kg for weight and LBM, respectively, in the group that received 100 mg/day). However, elevations in liver enzymes occurred in 14% of patients receiving oxymetholone. Compared to the 150-mg/day dose of oxymetholone, 100 mg/day appeared to have equivalent efficacy with less toxicity. Oxymetholone is currently approved as a treatment for anemia, but not for wasting.

The effects of thalidomide in patients with HIV-associated weight loss have been evaluated in three randomized, double-blind, placebo-controlled studies.(152-154) In the largest study of the three, 100 patients were randomized to receive thalidomide in doses of either 100 or 200 mg (nightly) or placebo for 8 weeks.(154) Patients randomized to either dosing level of thalidomide experienced modest but significant increases in weight (~4%). Treatment with a higher dose (200 mg) did not result in a better rate of weight gain and was associated with increased frequency of adverse effects and higher dropout rate. Other studies of thalidomide in patients with HIV infection have shown dramatic reversal of oral aphthous ulcers in a placebo-controlled clinical trial,(155) and reductions in stool frequency in patients with chronic diarrhea.(156) In each case, increases in weight accompanied improvement in symptoms. The most prevalent adverse effects of thalidomide in patients with HIV infection have been somnolence, peripheral neuropathy, hypersensitivity, and neutropenia.

Despite evidence that thalidomide decreases HIV replication and TNF-alpha production in vitro, plasma levels of HIV RNA and TNF-alpha were found to increase modestly (0.3-0.4 log for HIV RNA) in two recent clinical trials of this agent.(154,155) Although the durability and clinical significance of these increases are not known at this time, they have created uncertainty about the potential role for thalidomide in the treatment of wasting in patients with HIV infection.

Thalidomide is available as a treatment for erythema nodosum leprosum. Because of the well-documented and tragic teratogenic effects in infants whose mothers used this drug during pregnancy, thalidomide can be obtained only through specially licensed prescribers and pharmacists, and patients treated with this agent must adhere to strict guidelines for prevention of conception.

Small studies of a variety of other weak cytokine suppressors, including pentoxifylline,(157-159) omega-3 fatty acids (fish oil [160,161], and ketotifen [either alone (162) or in combination with oxymetholone (149)]) have produced results that are modest at best and provide no compelling rationale to pursue these agents as treatments for HIV-associated wasting.