Table 2.
PHIV1 | PHEU1 | ||||
---|---|---|---|---|---|
| |||||
Outcome | Exposure | Adjusted difference3 (95%CI) |
P value |
Adjusted difference3 (95%CI) |
P value |
25(OH)D (ng/mL) | |||||
BMD z | |||||
Total body | ≤ 20 vs. >20 | −0.38 (−0.60, −0.16) | <0.001 | −0.34 (−0.64, −0.03) | 0.03 |
Lumbar spine | ≤ 20 vs. >20 | −0.21 (−0.42,0.00) | 0.05 | 0.10 (−0.30,0.51) | 0.62 |
BMC (g) | |||||
Total body | ≤ 20 vs. >20 | −59.1 (−108.3, −9.8) | 0.02 | −0.01 (−73.5,73.5) | 1.00 |
Total body less head | ≤ 20 vs. >20 | −52.3 (−96.7, −7.8) | 0.02 | −0.30 (−69.3,68.7) | 0.99 |
Lumbar spine | ≤ 20 vs. >20 | −0.89 (−2.2,0.47) | 0.20 | 0.57 (−1.2,2.4) | 0.54 |
PTH (pg/mL)2 | |||||
BMD z | |||||
Total body | >65 vs. ≤65 | −0.02 (−0.37,0.42) | 0.91 | - | - |
Lumbar spine | >65 vs. ≤65 | −0.06 (−0.49,0.36) | 0.77 | - | - |
BMC (g) | |||||
Total body | >65 vs. ≤65 | 14.6 (−79.7,108.9) | 0.76 | - | - |
Total body less head | >65 vs. ≤65 | 16.9 (−62.9,96.7) | 0.68 | - | - |
Lumbar spine | >65 vs. ≤65 | −0.54 (−3.1,2.0) | 0.68 | - | - |
Abbreviations: PHIV-Perinatally HIV-infected; PHEU- Perinatally HIV-exposed uninfected; 95%CI-95% confidence interval; 25(OH)D – 25 hydroxy-vitamin D; PTH-parathyroid hormone; Ref-reference group; BMD z-Bone mineral density z score; BMC-bone mineral content.
The number of children included in the model for PHIV and PHEU was as follows: total body z-scores N=388, N=198; spine z-scores N=382, N=196; total body BMC N=387, N= 197; spine BMC N=381, N=395.
No models were fit for high PTH among HEU children because only 5 had high PTH.
Adjusted models of bone outcomes included age at vitamin D measurement, black race, ancestral markers (African, Europe/CSW Asia, other), height z, extremity lean mass, vigorous activity > 75th percentile and CD4 count. When 25(OH)D status was the exposure, we additionally adjusted for percent body fat. When the outcome was Total body-BMC, Total body less head-BMC or Lumbar spine-BMC, we additionally adjusted for Tanner stage and sex.