Piso, Rein Jan MD*; Rothen, Madeleine MD†; Rothen, Jean Pierre MD‡; Stahl, Matthias MD*
From the *Department of Medicine, Kantonsspital, Olten, Switzerland; †Department of Medicine, Regionalspital, Biel, Switzerland; and ‡Medical Laboratories Olten, Olten, Switzerland.
Received for publication September 6, 2010; accepted December 22, 2010.
The authors have no funding or conflicts of interest to disclose.
Correspondence to: Rein Jan Piso, MD, Kantonsspital Olten, Department of Medicine, Baslerstrasse 150, CH- 4600 Olten, Switzerland (e-mail: firstname.lastname@example.org).
Objectives: Osteoporosis and bone fractures are correlated to antiretroviral treatment. It is not clear whether some substances comprise greater risks of bone loss than others.
Methods: We measured pyridinoline, deoxypyridinoline crosslinks, and bone-specific alkaline phosphatase in 113 HIV-positive patients. We compared patients with and without antiretroviral treatment. We then compared patients with versus without tenofovir and patients with protease inhibitor versus nonnucleoside reverse transcriptase inhibitor use.
Results: Bone-specific alkaline phosphatase, pyridinoline, and deoxypyridinoline crosslinks were significantly higher in patients with antiretroviral treatment compared with patients without antiretroviral treatment: 24.5 versus 13.04 pg/L (P < 0.001), 82.73 versus 51.93 nmol/mmol (P < 0.001), and 16.56 versus 9.94 nmol/mmol (P < 0.001), respectively. In contrast, no difference was found between patients with and without tenofovir: 25.38 versus 20.02 pg/l (P = 0.1); 79.85 versus 83.95 nmol/mmol (P = 0.64), and 19.12 versus 14.00 nmol/mmol (P = 0.14), respectively. Comparison between patients with protease inhibitor versus nonnucleoside reverse transcriptase inhibitor yielded no difference either: 23.07 versus 27.18 pg/L (P = 0.24), 92.96 versus 80.73 nmol/mmol (P = 0.36), and 18.22 versus 16.39 nmol/mmol (P = 0.55).
Conclusion: Markers for bone turnover are higher in treated compared with untreated patients. No difference concerning tenofovir use or protease inhibitor versus nonnucleoside reverse transcriptase inhibitor use could be found.
Important advantages in treatment of HIV-infected persons have been made after introduction of antiretroviral treatment, resulting in a massive drop in morbidity and mortality.1 Consequently, the life expectancy of these patients has prolonged remarkably.2 In this newer era in which HIV infection is considered a chronic, treatable disease, the focus on long-term side effects of antiretroviral treatment as well as natural problems of an aging population, partly aggravated by the HIV infection itself, will become more and more important.
Known as a growing epidemic in the older non-HIV-infected population, osteoporosis has also become of particular interest to physicians treating HIV-infected patients.3 Osteoporosis is characterized through progressive loss in bone mineral density leading to fractures in the spine, hip, and wrists as main clinical manifestation.4 It seems that osteoporosis as well as bone fracture rate are higher in HIV-infected patients compared with non-HIV-infected patients.5 It is not clear what contributes to the higher bone loss, the disease itself or the antiretroviral treatment.6-8 Tenofovir (TDF) as well as protease inhibitors (PIs) have been postulated as risk factors for premature bone loss.9,10
Levels of alkaline phosphatase have been shown to be higher in patients treated with TDF compared with other nucleoside backbone treatment.11
Bone stability is determined by the density of crosslinks.12,13 Bone resorption releases components of these crosslinks, which can be measured in urine. While pyridinoline (PYD) can be found in bone as well as in cartilage, deoxypyridinoline (DPD) is only located in bone. Bone-specific alkaline phosphatase (BSAP) and PYD/DPD are markers of bone turnover and not elevated in low turnover osteoporosis. Although dual-energy radiographic absorptiometry quantifies bone density, crosslinks are indicators of the osteoclastic process of bone resorption and elevated in high turnover osteoporosis.12 It has been shown that in HIV-negative women, elevated crosslinks in urine correlated with a doubled risk of hip fracture independently of the bone mineral density of the proximal femur.14 BSAP is produced by osteoblasts and therefore a marker of bone formation. We evaluated the urine levels in a collective of patients with HIV in two outpatient clinics in Switzerland.
Kantonspital Olten and Regionalspital Biel are both university-affiliated secondary care hospitals with an outpatient clinic caring for HIV-infected patients. Morning urine was used to measure PYD and DPD. Serum parameters were obtained during regular blood sampling on routine visits. No specific clinical assessment for osteoporosis and no densitometry was performed. For people who started antiretroviral treatment, a second collection was done 3 to 5 months later. None of our patients received calcium supplementation or biphosphonate therapy.
The patients were selected in six groups: Group 1: no combination antiretroviral treatment (cART); Group 2: on any cART; Group 3: on TDF containing cART; Group 4: cART but no TDF; Group 5 on nonnucleoside reverse transcriptase inhibitor (NNRTI) containing cART; and Group 6: on PI containing cART. Patients with a combination of NNRTI and PI or NNRTI and a PI-sparing regimen were excluded for this calculation. Patients without ART were either recently diagnosed and collection was done before start of ART (eight), in less advanced state of disease (13), or they refused treatment despite an advanced state of disease (six).
Measurements of Crosslinks
We measured PYD and DPD by high-performance liquid chromatography using a commercially available assay from Chromsystems Instruments & Chemicals (Munich, Germany) as previously shown.15
To avoid multiple thawing, all urine samples were separated into aliquots with no added preservatives. Sample aliquots were stored at -20°C. All data obtained were corrected according to the urine creatinine concentration measured by use of a standard calorimetric method on an automatic analyzer (Hitachi 911; Roche Diagnostics, Indianapolis, IN).15 Corrected serum calcium and phosphate levels were obtained from all patients.
Results are expressed in relation to urinary creatinine. The reference values for a healthy population are 25 to 88 nmol/mmol creatinine for PYD, 2 to 23 nmol/mmol creatinine for DPD, and less than 20.1 μg/L for BSAP.
Descriptive statistic was made for all patients at baseline with the continuous data expressed as mean ± standard deviation and categoric data expressed as counts. Baseline characteristics of the groups were compared with Student test and chi-square test. Comparison for quantitative data between the groups was made with Fisher exact and Student t test.
One hundred thirteen patients were included in the study. Eighty-six patients were on ART, whereas 27 patients were not. Sixty-four were men and 49 women. Mean age of the patients was 43 ± 10 years. Eight of 49 women and 13 of 64 men were older than 50 years old. TDF was used by 57, whereas 29 had another backbone treatment. Forty-five patients received NNRTI; 38 were treated by PI. Nineteen patients had CD4 count less than 200 cells/μL, 33 patients 200 to 350 cell/μL, and 61 patients greater than 350 cell/μL.
There were slightly more women on NNRTI than men (25 versus 20, P = 0.05) and on TDF use (30 versus 27, P = 0.03); otherwise, the patients were equally distributed with regard to sex, hepatitis C infection, opiate consumption/substitution, and diabetes mellitus (Table 1).
There was no difference BSAP (18.09 ± 12.15 versus 20.44 ± 11.92 versus 24.35 ± 15.15 pg/L), PYD (84.42 ± 84.93 versus 78.11 ± 33.95 versus 81.23 ± 44.78 nmol/mmol creatinine), and DPD (17.34 ± 15.94 versus 1495 ± 5.63 versus 15.97 ± 10.08 nmol/mmol creatinine) among patients with low, medium, or high CD4 count.
Treated versus Nontreated Patients
Patients on cART had significantly higher values for BSAP (24.5 ± 15.06 versus 13.04 ± 5.38 pg/L, P < 0.001), PYD (82.73 ± 37.33 versus 51.93 ± 24.86 nmol/mmol creatinine, P < 0.001), and DPD (16.56 ± 8.41 versus 9.94 ± 4.75 nmol/mmol creatinine, P < 0.001). No difference in corrected calcium levels could be detected (Fig. 1).
Tenofovir versus Tenofovir-Sparing Regimes
The difference between patients treated with TDF was not significant compared with patients treated with another nucleoside backbone: 25.38 ± 13.53 versus 20.02 ± 12.04 pg/L (P = 0.1 for BSAP), 79.85 ± 38.91 versus 85.95 ± 30.05 nmol/mmol creatinine (P = 0.64 for PYD), and 19.12 ± 17.00 versus 14.00 ± 5.00 nmol/mmol creatinine (P = 0.13 for DPD) crosslinks (Fig. 2).
Protease Inhibitor versus Nonnucleoside Reverse Transcriptase Inhibitor
We found no difference for patients treated with PI compared with patients treated with NNRTI: 23.08 ± 16.27 versus 27.18 ± 12.02 pg/L (P = 0.24 for BSAP), 92.96 ± 66.55 versus 80.73 ± 34.6 nmol/mmol creatinine (P = 0.36 for PYD), and 18.22 ± 8.07 versus 16.39 ± 9.21 (P =0.55 for DPD) (Fig. 3).
Short- versus Long-Term Treatment
Thirteen patients had a recent introduction of cART. Patients with a total duration of ART shorter than 6 month had similar values for BSAP (21.55 ± 11.91 versus 24.63 ± 13.87 pg/L, P = 0.48), PYD (95.96 ± 108.18 versus 83.18 ± 38.39 nmol/mmol creatinine, P = 0.52), and DPD (19.37 ± 19.8 versus 16.44 ± 8.61 nmol/mmol creatinine, P = 0.36) than patients with longer treatment (Fig. 4).
Correlations Between Bone Turnover Markers
There was a significant correlation between BSAP and PYD (r = 0.3; 95% confidence interval [CI], 0.11-0.47; P = 0.003) and BSAP and DPD (r = 0.31; 95% CI, 0.12-0.48; P = 0.002). Also, 14 of 29 patients in the lowest quarter for BSAP were also in the lowest quarter for PYD and/or DPD (P = 0.05), and 18 of 29 patients in the highest quarter for BSAP were also in the highest quarter for PYD and/or DPD (P < 0.001).
HIV-infected adults have a higher prevalence in osteopenia (T-score -1 to -2.5) and osteoporosis (T-score less than -2.5)16 as well as higher fracture rates compared with non-HIV-infected individuals.5 Because HIV-infected adults tend to have higher prevalence of traditional risk factors, data on HIV-related risk factors as a pivotal role for osteoporosis in this cohort are conflicting.7,8,17 We focused on ART as part of the complex resulting in higher risk for osteoporosis. TDF has been made responsible for elevated serum alkaline phosphatase11 and bone mineral density seems to be lower in patients treated with PIs.17 Compared with non-treated HIV-infected patients, we found an increase in bone resorption markers as well as in bone formation markers. A higher level of serum alkaline phosphatase was found in patients treated with TDF compared with TDF-sparing cART11 and a more pronounced decline in hip and lumbar spine bone mineral density could be found in patients treated with TDF/FTC compared with ABC/3TC,18 but we did not find a difference, neither in bone formation nor in bone resorption markers, even as the proportion of women taking TDF was higher than men. Our study was much smaller than the Swiss HIV cohort investigation and the Assessment of Safety and Efficacy with Abacavir or Tenofovir (ASSERT) Study, and albeit we measured the more specific BSAP, we cannot completely rule out that a larger collective may had shown a significant difference. However, we could not show a trend in DPD and PYD values in patients treated with TDF compared with TDF-sparing regimens. This is in contrast to the highly significant difference we found in patients treated with any cART compared with treatment-naïve patients (P < 0.001). Even if there might be a contribution of TDF on bone metabolism, the relative contribution compared with the overall effect of cART may be low. Decreases in bone mineral density independent of the antiretroviral regimen was found in a recent study. It was shown that lopinavir- and efavirenz-containing regimens resulted in a similar decrease in bone mineral density.6 No difference in bone mineral density was found in women never treated compared with treated patients.8 However, this was a cross-sectional study with 83% women of African descent and a mean duration of cART of 3.5 years. Because loss of bone mineral density occurs slowly, the duration of cART might have been too short for a cross-sectional study with patients with a lower risk of osteoporosis compared with white women. PI has been made responsible for a decrease in bone loss of HIV-infected patients.9 We did not find a difference in bone mineral density between patients treated with NNRTI or PI. Similar to TDF, as a result of the small sample size, we cannot completely rule out that differences might have occurred in a larger study. Because there was a highly significant difference between treated and naïve patients, the difference between NNRTI and PI treatment might be small compared with the difference between treated and nontreated patients. In the Strategies for the Management of Anti-Retroviral Therapy (SMART) trial, it could be found that patients with continuous antiretroviral treatment showed a more pronounced decline in bone mineral density compared with the intermittent treated group, but no single agent or drug class could be made responsible for this effect.19 It has been postulated that a decrease in bone mineral density occurs mainly in the beginning of an ART and will be spontaneously corrected while continuing the treatment.20 The Assessment of Safety and Efficacy with Abacavir or Tenofovir (ASSERT) study demonstrated that after an initial decline, bone mineral density stabilized over a 48-week period.18 The mechanism of this regulation is not clear. With bone turnover markers, we can only indicate differences in the metabolic process and are not able to quantify bone loss. We could only observe a limited number of patients with newly introduced ART, but we found no difference in bone turnover markers between patients with treatment duration of less than 6 months compared with the group of a longer duration of treatment. It has been postulated that a remodeling of bone formation occurs after initial bone loss.7 We could also observe a correlation between bone resorption and bone formation markers but do not call it synchronization or remodeling. We rather think that in this relatively young population in which a substantial number of patients have not reached peak bone mass, an accelerated resorption of bone can be compensated by bone formation. However, we postulate that the antiretroviral treatment itself contributes to an increased bone resorption. As this metabolic process continues, the capacity of compensation will decline as patients become older, and new bone formation might not keep pace with bone resorption resulting in overall bone loss.
Our study has some limitations. Bone mineral density reflects the underlining metabolic process and is not either a clinical end point nor does it reflect the extent of osteoporosis in a given patient. We did not include body mass index, smoking status, or gonadal function in our analysis. We do believe that the size of our study was too small to include further subgroup analysis. Low 25-hydroxyvitamin D [25(OH)D] levels have been found in a substantial proportion of HIV-infected patients.21 Because these results were published after data collection of our study, this parameter was not measured. For measuring the effect of HIV and ART on the risk of bone fractures, we propose larger, longitudinal studies with ART-naive as well as treated patients and including multiple confounders in the analysis.
1. Vergidis PI, Falagas ME, Hamer DH. Meta-analytical studies on the epidemiology, prevention, and treatment of human immunodeficiency virus infection. Infect Dis Clin North Am. 2009;23:295-308.
2. Lohse N, Hansen AB, Pedersen G, et al. Survival of persons with and without HIV infection in Denmark, 1995-2005. Ann Intern Med. 2007;146:87-95.
3. Paccou J, Viget N, Legrout-Gerot I, et al. Bone loss in patients with HIV infection. Joint Bone Spine. 2009;76:637-641.
4. Post TM, Cremers SC, Kerbusch T, et al. Bone physiology, disease and treatment: towards disease system analysis in osteoporosis. Clin Pharmacokinet. 2010;49:89-118.
5. Dao C, Young B, Buchacz K, et al; the Outpatients Study Investigators. Higher and increasing rates of FRacture among HIV-infected persons in the HIV outpatient study compared to the general US population, 1994 to 2008. 2010. 17th conference on retroviruses and opportunistic infections (CROI).
6. Brown TT, McComsey GA, King MS, et al. Loss of bone mineral density after antiretroviral therapy initiation, independent of antiretroviral regimen. J Acquir Immune Defic Syndr. 2009;51:554-561.
7. Aukrust P, Haug CJ, Ueland T, et al. Decreased bone formative and enhanced resorptive markers in human immunodeficiency virus infection: indication of normalization of the bone-remodeling process during highly active antiretroviral therapy. J Clin Endocrinol Metab. 1999;84:145-150.
8. Libois A, Clumeck N, Kabeya K, et al. Risk factors of osteopenia in HIV-infected women: no role of antiretroviral therapy. Maturitas. 2010;65:51-54.
9. Duvivier C, Kolta S, Assoumou L, et al. Greater decrease in bone mineral density with protease inhibitor regimens compared with nonnucleoside reverse transcriptase inhibitor regimens in HIV-1 infected naive patients. AIDS. 2009;23:817-824.
10. Gafni RI, Hazra R, Reynolds JC, et al. Tenofovir disoproxil fumarate and an optimized background regimen of antiretroviral agents as salvage therapy: impact on bone mineral density in HIV-infected children. Pediatrics. 2006;118:e711-e718.
11. Fux CA, Rauch A, Simcock M, et al. Tenofovir use is associated with an increase in serum alkaline phosphatase in the Swiss HIV cohort study. Antivir Ther. 2008;13:1077-1082.
12. Garnero P. Bone markers in osteoporosis. Curr Osteoporos Rep. 2009;7:84-90.
13. Knott L, Bailey AJ. Collagen cross-links in mineralizing tissues: a review of their chemistry, function, and clinical relevance. Bone. 1998;22:181-187.
14. Robbins JA, Schott AM, Garnero P, et al. Risk factors for hip fracture in women with high BMD: EPIDOS study. Osteoporos Int. 2005;16:149-154.
15. Kraenzlin ME, Kraenzlin CA, Meier C, et al. Automated HPLC assay for urinary collagen cross-links: effect of age, menopause, and metabolic bone diseases. Clin Chem. 2008;54:1546-1553.
16. Cazanave C, Dupon M, Lavignolle-Aurillac V, et al. Reduced bone mineral density in HIV-infected patients: prevalence and associated factors. AIDS. 2008;22:395-402.
17. Anastos K, Lu D, Shi O, et al. The association of bone mineral density with HIV infection and antiretroviral treatment in women. Antivir Ther. 2007;12:1049-1058.
18. Stellbrink HJ, Orkin C, Arribas JR, et al. Comparison of changes in bone density and turnover with abacavir-lamivudine versus tenofovir-emtricitabine in HIV- infected adults: 48-week results from the ASSERT study. Clin Infect Dis. 2010;51:963-972.
19. Grund B, Peng G, Gibert CL, et al. Continuous antiretroviral therapy decreases bone mineral density. AIDS. 2009;23:1519-1529.
20. McComsey G, Kitch D, Daar E, et al. Bone and limb fat outcomes of ACTG A5224s, a substudy of ACTG A5202: a prospective, randomized, partially blinded phase III trial of ABC/3TC or TDF/FTC with EFV or ATV/r for initial treatment of HIV-1 infection. 2010. 17th conference on retroviruses and opportunistic infections (CROI).
21. Mueller NJ, Fux CA, Ledergerber B, et al. High prevalence of severe vitamin D deficiency in combined antiretroviral therapy-naive and successfully treated Swiss HIV patients. AIDS. 2010;24:1127-1134.
© 2011 Lippincott Williams & Wilkins, Inc.