Answers are grounded in this paper and cite their sources.
To determine the prevalence of CD4 lymphocyte levels less than 350 cells/mm3 accompanying viral loads (VL) less than 500 copies/ml (discordant CD4 lymphocyte levels/VL) in HIV-infected men with bone pathologies, we conducted a review of the records of 3512 men. We found discordant CD4 lymphocyte levels/VL in 26 (36.1%) out of 72 with bone pathologies, and in 704 (20.5%) out of 3440 without bone pathologies (P< 0.01), and concluded that HIV sequestered in bone was a possible explanation. Reports describing HIV-infected patients with osteonecrosis [1–5] indicated surprising numbers who have unexpectedly low CD4 lymphocyte levels together with undetectable, or relatively low, viral loads (VL) (discordant CD4 lymphocyte levels/VL). For example, Keruly et al. [1] reported a mean CD4 lymphocyte level of 124 cells/mm3 with VL of 4.1 log10/ml in 17 patients with osteonecrosis, whereas their patients without osteonecrosis had a higher CD4 lymphocyte level of 212 cells/mm3 despite having higher VL at 4.7 log10/ml. We compared the proportions of patients with discordant CD4 lymphocyte levels/VL in HIV-positive men with and without various bone disorders. We examined the CD4 lymphocyte level and VL of HIV-positive male members of the Kaiser Permanente Health Plan (KP) with bone disorders and compared them with those of the HIV-positive male KP members without bone pathologies. We defined discordant CD4 lymphocyte levels/VL as a CD4 cell count of less than 350 cells/mm3 concomitant with a VL of less than 500 copies/ml. Among patients with bone disorders, the CD4 lymphocyte levels and VL used were those made closest to, but within the time frame 6 months before or 3 months after, the date of radiological diagnosis. Among patients without bone pathologies, the CD4 lymphocyte levels and VL used were those closest to the arbitrary date 1 October 1999. The bone disorders examined were osteonecrosis; fractures of the vertebrae, hips, or distal radius/ulna (i.e. associated with osteoporosis); and osteopenia/osteoporosis (defined by bone density data using WHO criteria) of the hip or lumbar spine. For osteonecrosis or osteoporotic fracture, we scanned radiological reports for the period between late 1998 and mid-2000. Osteonecrosis resulting from major trauma was excluded. Bone density reports to 31 December 2001 were examined. To assess the confounding roles of corticosteroids and of other drugs sometimes used to treat patients with HIV infection (amphoterocin, carbamazepine, cyclophosphamide, doxorubicin, flucytosine, foscarnet, ganciclovir, gentamicin, hydroxyurea, interferon, pentamidine, pyrimethamine, sulfamethoxasole, thalidomide, trimethoprim, valganciclovir, vinblastine, vincristine, and zidovudine) pharmacy records were examined for the 5-year period before the CD4 lymphocyte level/VL measurement. A high dosage of prednisone was defined as an aggregate of 500 mg or more. Dexamethasone and hydrocortisone were converted to prednisone equivalence using a dexamethasone : prednisone ratio of 1 : 5 and a hydrocortisone : prednisone ratio of 7.5 : 1. The duration of HIV infection as a confounder was examined by comparing the proportion of patients with discordant VL/CD4 lymphocyte levels in each of four ‘duration of infection’ strata: 0–4 years, 5–9 years, 10–14 years, or over 15 years. Statistical analysis compared men with and without bone disorders with respect to the proportion of patients with discordant CD4 lymphocyte level/VL, using the χ2 and two-tailed test of significance. Among 3512 men, we identified 23 with osteonecrosis, 28 with osteoporotic fractures (including two patients with both osteonecrosis and osteoporotic fractures), and 23 with spinal or hip T scores that were 1.0 or less (excluding those with known osteonecrosis or osteoporotic fractures). Of the above 72 men with bone disease, 26 (36.1%) had discordant CD4 lymphocyte levels/VL, whereas only 704 (20.5%) of the 3440 without bone disease had this discordance (Table 1) (P < 0.01).Table 1: HIV-RNA level less than 2.7 log10/ml with CD4 lymphocyte levels less than 350/mm3 (discordant CD4 lymphocyte levels/viral loads) in HIV-positive men with osteonecrosis, osteoporotic fractures, or osteopenia/osteoporosisa.Medications did not cause the lymphocytopenia. Among the 26 patients with discordant CD4 lymphocyte levels/VL, only two had used high doses of corticosteroids in the previous 5 years, and none had used any corticosteroids during the 8 weeks before the radiological diagnosis. Other drugs that sometimes cause lymphocytopenia had been used by a smaller proportion of the patients with the discordant CD4 lymphocyte levels/VL than by those patients without the discordance. A discordant CD4 lymphocyte level/VL was not associated with a longer duration of infection. Across the four periods of increasing duration of infection that were examined, the proportions with discordant CD4 lymphocyte levels/VL were 22.1, 16.2, 23.5 and 19.5%. In addition, the proportion of men with CD4 lymphocyte level/VL discordance was lower among men with the longest duration of infection (19.5%) than among men with bone pathologies (36.1%). Overall, 20.8% of the 3512 HIV-positive men and 20.5% of the 3440 men without bone pathologies had discordant CD4 lymphocyte levels/VL, a proportion similar to the 17.4% seen by Marimoutou et al. [6] in their 1044 patients receiving antiretroviral therapy. However, among our 72 HIV-positive men with bone pathologies, 36.1% had discordant CD4 lymphocyte levels/VL: those with discordance comprised 43.5% with osteopenia or osteoporosis, 34.8% with osteonecrosis, and 28.6% with osteoporotic fractures (Table 1). One explanation of our findings is that bone might be a sequestered reservoir of HIV, and that HIV in bone might effect the depletion of CD4 cells in bone marrow. The Haversian system gives limited blood flow through cortical bone and, thus, poor exposure to drugs, enabling HIV to survive in osteocytes but undergo restrained release, producing only a low plasma viral load. Low CD4 cell levels in blood would result from gp120, TAT, or cytokines being released from cortical osteons, then reaching the marrow to cause the apoptosis of CD4 cells. HIV has been cultured from bone, human osteoblasts, osteocytes, and osteogenic sarcoma cells [7–10]. In addition, monocytes, readily infected by HIV, are precursors of osteoclasts [11]. Messenger RNA coding for CD4 cells was found in osteoblasts [12], and the receptor for phosphate in osteoblasts also serves as a receptor for retroviruses [13]. Murine, avian, and feline retroviruses infect osteoblasts and osteocytes [14–16]. HIV infection of osteocytes would plausibly explain why osteopenia and osteonecrosis are seen in patients who have not received antiretroviral therapy [5,17–23], and why osteopenia affects cortical bone more than trabecular bone [21,22]. Also consistent with HIV infection of osteoblasts are observations in HIV-infected patients of low serum osteocalcin levels, reflecting low osteoblast activity [21,24–26]. The positive correlation between osteocalcin levels (i.e. osteoblast activity) and CD4 cell counts [24,25] is consistent with our observations of CD4 lymphocyte levels/VL discordance in some patients with bone pathologies. Neither corticosteroids nor other drugs contributed to the CD4 lymphocytopenia. Corticosteroids had not been used during the 8 weeks before the radiological diagnosis by any of our 26 patients with bone pathologies who also had discordant CD4 lymphocyte levels/VL. We could not assess the contributions of alcohol or tobacco use. However, alcohol raises, rather than lowers, lymphocyte levels and smoking raises CD4 cell counts [27]. The same factors causing bone disease in HIV-negative individuals also apply to HIV-positive individuals, but the excess of bone disease in HIV-positive individuals implies additional mechanisms, one of which might be HIV infection of osteocytes.
No takes yet. Share an insight, caveat, or question.
Fessel et al. (2003) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: