Equivalent osteopenia in HIV-infected individuals studied before and during the era of highly active antiretroviral therapy

Advances in the treatment of HIV infection, especially the use of highly active antiretroviral agents (HAART), have dramatically reduced the morbidity and mortality rates from HIV infection [1,2]. However, a substantial proportion of patients develop fat redistribution and metabolic alterations [3,4]. In addition, several reports of osteopenia in patients on HAART have been published or presented [5,6], and an association with protease inhibitor therapy has been suggested. In contrast, a study from the pre-HAART era [7] did not demonstrate alterations in bone mineral content in HIV-infected patients. We compared the baseline bone mineral content and density in two groups of HIV-infected individuals, studied before or during the era of HAART. Pre-HAART subjects were 36 malnourished, HIV-infected men who initiated a trial of the anabolic steroid, oxandrolone, in 1993–1994, and HAART subjects were 22 HIV-infected men (n = 19) and women (n = 3) with fat redistribution, who initiated a trial of recombinant human growth hormone for excess visceral adipose tissue in 1998–1999. Each subject was matched with one to six controls by sex, race, age (± 5 years), and height (± 10 cm). Controls were taken from a database of dual X-ray absorptiometry (DXA) studies in healthy adults. All studies were approved by the Institutional Review Board of St Luke's-Roosevelt Hospital Center (NIH DK 42618) and subjects gave informed consent. The total bone density, calcium and mineral content were evaluated by DXA (Lunar DPX, Lunar Inc., Madison, WI, USA), as described previously [8]. Differences between HIV-infected and control subjects were analysed by paired t-test, using the SAS statistical software package (version 7.0, 1998), and P < 0.05 was considered statistically significant. Subjects in the pre-HAART and HAART studies were similar with regard to age and height. HAART subjects were on average 15 kg heavier and had more fat-free mass and fat (all P < 0.001) than pre-HAART subjects (Table 1). Mean bone mineral content, bone calcium and bone density were similar in the two HIV-infected groups. However, when compared with their matched controls, both groups of HIV-infected subjects averaged lower values for all three measures of bone mineralization (P < 0.001 to 0.004).Table 1: Body weight and dual X-ray absorptiometry results in pre-highly active antiretroviral therapy and highly active antiretroviral therapy subjects and matched controls. The etiology and pathogenesis of osteopenia in HIV infection is uncertain. The results of this study suggest that osteopenia has been a feature of HIV infection before as well as during treatment with HAART. These results differ from a previous DXA study in HIV infection before the era of HAART [7]. A growing number of authors have suggested that osteopenia is related to HAART. However, in a report of osteoporosis with vertebral fracture in two young HIV-infected African women, the duration of HAART was short in one patient, whereas the other had amenorrhea [9], suggesting that factors other than antiretroviral agents might affect bone mineral content. Although HIV-infected subjects from both eras had similar degrees of demineralization compared with matched controls, the specific etiologies may differ in the two groups. Osteopenia is multifactorial in development in the absence of HIV infection, and factors both related and unrelated to HIV may play a role in the osteopenia seen in these subjects. The persistent activation of proinflammatory cytokines, in particular tumor necrosis factor alpha (TNF-α), might play an important role in the pathogenesis of osteopenia in HIV infection. A deficiency of 1,25-dihydroxy vitamin D in HIV patients has been associated with proinflammatory cytokine activities, wherease increased serum levels of C-teleopeptide, decreased serum 1,25-dihydroxy vitamin D, and markedly depressed osteocalcin levels normalized during HAART [10–12]. Of note is the fact that Ledru and colleagues [13] reported that cytokine dysregulation may persist during HAART, with excess expression of TNF-α in circulating mononuclear cells. The effect of chronic illness on the hypothalamic–pituitary–adrenal axis may also contribute to bone demineralization, as decreased insulin-like growth factor (IGF-I) [14] and increased corticosteroid [15] levels have been found in the past. HIV might also have direct effects on osteogenic cells, which could affect the bone mineral content [16–18]. Growth hormone levels are elevated, and serum IGF-I responses to growth hormone decreased in men with AIDS wasting [19,20], suggesting a pattern of acquired growth hormone resistance similar to that seen in non-HIV-infected patients with protein–calorie malnutrition [21]. Other possible causes of bone demineralization that may play a role in HIV include the inhibitory effects of TNF-α on parathyroid hormone secretion [14,15], vitamin D or calcium malabsorption, excess calcium losses, or decreased physical activity as a result of debilitation. In conclusion, osteopenia has been a common, although newly identified, problem in HIV infection. Our results indicate that factors in addition to specific antiretroviral agents may promote osteopenia. Adeyemi Lawal Ellen S. Engelson Jack Wang Steven B. Heymsfield Donald P. Kotler