- Email: [email protected]
Optimal dietary calcium intake in HIV treated patients: No femoral osteoporosis but higher cardiovascular risk
Clinical Nutrition 33 (2014) 363e366
Contents lists available at SciVerse ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Short communication
Optimal dietary calcium intake in HIV treated patients: No femoral osteoporosis but higher cardiovascular riskq Laura Galli a, *, Alessandro Rubinacci b, Deborah Cocorullo c, Stefania Salpietro a, Vincenzo Spagnuolo a, Nicola Gianotti a, Alba Bigoloni a, Concetta Vinci a, Giovanna Mignogna b, Marcella Sirtori b, Adriano Lazzarin a, c, Antonella Castagna a a b c
Department of Infectious Diseases, San Raffaele Scientific Institute, Via Stamira d’Ancona 20, 20127 Milan, Italy Bone Metabolism Unit, San Raffaele Scientific Institute, Milan, Italy Università Vita-Salute San Raffaele, Milan, Italy
a r t i c l e i n f o
s u m m a r y
Article history: Received 21 February 2013 Accepted 6 July 2013
Background & aims: We performed a cross-sectional study on adult HIV-infected patients, on HAART, without calcium or vitamin D supplementation to evaluate if the cardiovascular risk or the presence of osteoporosis may be predictive factors of an optimal daily calcium intake (DCI>1000 mg/day). Methods: Patients underwent a dual-energy X-ray absorptiometry, measured biochemical parameters and compiled a validated questionnaire for the assessment of DCI. Osteoporosis (OP) was defined according to the WHO classification at either the vertebral spine or femoral neck. Cardiovascular risk was assessed by the 10-year Framingham cardiovascular risk score. Results: 200 HIV-infected patients evaluated: 171 (86%) males with a median age of 48.1 (42.3e53.8) years and 10.6 (4.3e13.6) years of HAART exposure. DCI was 889 (589e1308) mg/day and 79 (40%) patients had an optimal DCI. Framingham risk>20% was found in 13 (6.7%) patients and femoral OP was diagnosed in 12 (6%) pts. By multivariate analysis, optimal DCI was more likely in patients with a Framingham risk>20% [OR ¼ 5.547, 95% CI:1.337, p ¼ 0.025] and less likely in patients with femoral osteoporosis [OR ¼ 0.159, 95% CI: 0.018e0.790, p ¼ 0.047]. Conclusions: We found that an optimal dietary calcium intake was more likely in patients with high cardiovascular risk and no femoral osteoporosis. Ó 2013 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Dietary calcium intake Osteoporosis Cardiovascular risk HIV-infected patients
1. Introduction Due to the success of combination antiretroviral therapy (cART), patients with HIV infection live longer and are exposed to ageingrelated complications such as cardiovascular diseases (CVD) and bone mineral disorders. Previous studies, conducted in the general population, have associated calcium supplements which are widely used in the prevention or treatment of osteoporosis, with a possible increase in the risk of adverse cardiovascular outcomes.1 One cohort study identified no adverse effects associated with dietary calcium intake, although a near doubling risk of myocardial infarction (MI) was found in calcium supplements users.2
q Conference presentation: Presented in part at the 4th ICAR, Napoli 2012. * Corresponding author. Tel.: þ39 0226437934; fax: þ39 0226437030. E-mail address: [email protected] (L. Galli).
Limited data on dietary calcium intake are available among HIVinfected subjects. Although one study has recently identified a protective effect of calcium intake on the osteoporosis among HIVinfected subjects,3 it remains unclear whether such advantage might enhance the cardiovascular risk. Thus, the aim of our study was to evaluate if the cardiovascular risk or the presence of osteoporosis may be predictive factors of an optimal daily calcium intake (DCI). 2. Materials and methods We conducted a cross-sectional study on alive HIVþ patients, aged>18 years, on HAART, without calcium or vitamin D supplementation, attending the Infectious Diseases Department of the San Raffaele Scientific Institute between MayeJune 2011. Patients, evaluated on the same day, underwent a dual-energy X-ray absorptiometry (DEXA), measured biochemical parameters and compiled a validated questionnaire for the assessment of DCI.4 Briefly, the questionnaire is composed of 15 food items: selection of the items
0261-5614/$ e see front matter Ó 2013 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. http://dx.doi.org/10.1016/j.clnu.2013.07.004
364
L. Galli et al. / Clinical Nutrition 33 (2014) 363e366
with a p-value <0.20 at the univariate logistic regression models were included into two multivariate models (Model1 considered among the covariates the diagnosis of osteoporosis at femoral neck and alternatively Model 2 considered the diagnosis of osteoporosis at vertebral spine). Odds ratios (OR) and the corresponding 95% confidence intervals (95% CI) were estimated and reported for each variable included in the model. A two-sided p value <0.05 was considered statistically significant. All statistical analyses were performed with SAS software, version 9.2 (SAS Institute, Cary, North Carolina, United States).
was based on the data of the Italian Institute of nutrition relative to the food composition of the Italian diet, frequency of use and relative importance as a calcium food source; additional subgroup items were added for high calcium content food such as milk and derivatives. Servings were listed in three quantitative groups identified as small, medium, large and with grams of food. Subjects were asked to indicate the relative weekly frequency of use of each item. The calcium content estimate was obtained by food tables (attached to the questionnaire) quantifying the calcium content of each food according to the serving and weekly frequency. The daily calcium intake was obtained dividing the weekly calcium intake by 7. Optimal DCI was defined as a DCI>1000 mg/day (recommended threshold in adults according to the IOM5). Osteopenia and Osteoporosis (OP) were defined according to the WHO classification at either the vertebral spine or femoral neck: osteoporosis was defined by T-scores<2.5 for subjects older than 50 years of age or by z-score<2.5 for subjects younger than 50 years of age; osteopenia was defined by T-scoresranging between >2.5/1.0 for subjects older than 50 years of age or by z-score ranging between >2.5/1.0 for subjects younger than 50 years of age Cardiovascular risk was assessed by the 10-year Framingham cardiovascular risk score and used in the analysis as a categorical variable according to the threshold of 20% (an arbitrary cut-off value often used in the clinical setting and in the clinical studies to identify patients with a high 10-year risk of cardiovascular events).
3. Results Two hundred HIV-infected patients were included: 171 (86%) were males; age was 48.1 (42.3e53.8) years; they were infected with HIV since 14.5 (7.1e19.5) years, had 10.6 (4.3e13.6) years of HAART. DCI was 889 (589e1308) mg/day and 79 (40%) patients had an optimal DCI. As shown in Table 1, no significant differences were observed between the demographic, clinical, immuno-virological and metabolic characteristics between subjects with or without an optimal DCI. Framingham risk >20% was found in 13 (6.7%) patients; 8/13 (61.5%) subjects had a DCI>1000 mg/day as compared to 71/187 (38.0%) of patients with a CVD risk<20% (p ¼ 0.141). Subjects with a high CVD risk were older (median: 61 vs 47 years, p < 0.0001), had a slightly higher body mass index (median 25.6 vs 23.5, p ¼ 0.086), had higher lipid values (median total cholesterol: 207 vs 188 mg/dL, p ¼ 0.068; median LDL-cholesterol: 153 vs 120, p ¼ 0.004; median triglycerides: 148 vs 120, p ¼ 0.056), systolic and diastolic blood pressure (median systolic: 149 vs 120 mmHg, p ¼ 0.0001; median diastolic: 84 vs 77 mmHg, p ¼ 0.028) and fasting glucose values (median: 100 vs 87, p ¼ 0.0003) as compared to those with a CVD risk<20%. Osteopenia was diagnosed in 96 (48%) patients: 18 (9%) subjects had a diagnosis only at vertebral spine, 37 (18.5%) patients only at
2.1. Statistical analysis To analyze whether the demographic and clinical characteristics of the subjects included in this study were similar, the chi-square test or the Wilcoxon rank sum tests were applied when dealing with categorical or continuous variables, respectively. At multivariate analysis, logistic regression was applied to determine predictive factors of a DCI>1000 mg/day. Characteristics
Table 1 Patients’ characteristics according to dietary calcium intake at the time of study evaluation. Characteristica
Overall (N ¼ 200)
DCI 1000 mg/day (N ¼ 121)
DCI >1000 mg/day (N ¼ 79)
P-value
Age (years) Males Smokers Overweight (Body mass index 25 kg/m2) Years since first HIV positive test Years of exposure to HAART Nadir Cd4þ (cells/mL) Positive HCVAb Type of antiretroviral regimen NRTI-based NNRTI-based PI-based Otherb Use of tenofovir CD4þ (cells/mL) CD4þ <500 cells/mL HIV-RNA <50 copies/mL Total cholesterol (mg/dL) HDL-cholesterol (mg/dL) LDL-cholesterol (mg/dL) Triglycerides (mg/dL) Fasting glucose (mg/dL) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) 25 (OH) Vitamin D (ng/mL)
48.1 171 94 62 14.5 10.6 229 30
(42.3e53.8) (86%) (47%) (31%) (7.1e19.5) (4.3e13.6) (102e329) (15%)
48 104 55 35 13.9 9.7 233 19
(42.7e53.8) (86%) (46%) (29%) (6.1e19.4) (4.1e13.5) (100e328) (16%)
48.2 67 39 27 16.4 11.6 216 11
(42.2e53.9) (85%) (49%) (34%) (7.1e20.2) (4.8e13.6) (108e329) (14%)
0.496d 0.840c 0.563c 0.365c 0.413d 0.782d 0.579d 0.475c 0.754c
4 27 79 74 89 540 85 164 188 44 123 124 88 121 79 19.8
(2%) (15%) (43%) (40%) (45%) (384e777) (45%) (86%) (164e218) (36e54) (103e151) (89e194) (80e97) (112e133) (70e87) (13.8e27.2)
3 18 45 45 52 542 52 99 189 46 125 120 88 122 77 20
(3%) (16%) (41%) (41%) (43%) (376e801) (45%) (86%) (164e217) (36e55) (99e152) (89e178) (79e97) (113e132) (70e85) (13.2e27.2)
1 9 34 29 37 531 33 65 186 42 121 129 87 120 79 19
(1%) (13%) (47%) (40%) (47%) (406e751) (43%) (86%) (164e218) (36e50) (107e151) (87e203) (80e97) (111e133) (71e89) (15.2e27.3)
0.663c 0.946d 0.882c 0.999c 0.904d 0.311d 0.760d 0.395d 0.630d 0.886d 0.144d 0.726d
Abbreviations: DCI, dietary calcium intake; NRTI, Nucleoside Reverse Transcriptase Inhibitors; NNRTI, Non-nucleoside Reverse Transcriptase Inhibitors; PI, Protease Inhibitors. a Median (first quartile- third quartile) or frequency (%) used to describe results of continuous or categorical variables, respectively. b Other included the following drugs: maraviroc, raltegravir, etravirine. c By chis-quare or Fisher exact test, as appropriate. d By Wilcoxon rank sum test.
L. Galli et al. / Clinical Nutrition 33 (2014) 363e366
365
observed only for a Framingham risk>20% [OR (>20% vs 20%) ¼ 3.439, 95% CI: 0.930e14.442, p ¼ 0.071].
femoral neck and 41 (20.5%) patients had a diagnosis at both vertebral spine and femoral neck. OP was diagnosed in 25 (12.5%) patients: 13 (6.5%) patients had a diagnosis at vertebral spine, 7 (3.5%) patients at femoral neck and 5 (2.5%) patients had a diagnosis at both vertebral spine and femoral neck. An optimal DCI tended to be less frequent among patients with femoral OP [patients with femoral OP: 2/12 (16.7%); patients without femoral OP: 77/188 (41%), p ¼ 0.130]; optimal and optimal DCI was similarly distributed among patients with or without vertebral OP [patients with vertebral OP: 7/18 (38.9%); patients without vertebral OP: 72/182 (39.6%), p ¼ 0.999]. Among patients with a CVD risk >20%, 3/13 (25%) had a femoral OP diagnosis as compared to 9/187 (4.8%) of patients with a CVD risk less than 20% (p ¼ 0.037); no association between CVD risk and vertebral OP was observed (Table 1). Results from univariate and multivariate logistic regression are described in Table 2; only few covariates were found to be predictive of an optimal DCI and were entered into the multivariate models. In the first model, after adjustment for body mass index, current use of tenofovir and fasting glucose, an optimal DCI resulted to be more likely in patients with a Framingham risk>20% [OR (>20% vs 20%) ¼ 5.547, 95% CI: 1.337e29.185, p ¼ 0.025] and less likely in patients with the presence of femoral osteoporosis [OR (OP vs no-OP) ¼ 0.159, 95% CI: 0.018e0.790, p ¼ 0.047]. In the second multivariate model which considered the diagnosis of vertebral osteoporosis among the covariates, no significant predictors of an optimal DCI were found; a marginal effect was
4. Discussion To our knowledge, this is the only study to date evaluating if optimal DCI was predicted by OP and CVR among HIV-infected subjects. We found that optimal DCI does seem to be more likely in subjects with no femoral OP, though not with vertebral OP, and in subjects with an increased CVR. Among HIV-positive subjects, osteoporosis may be related to ageing and life-style risk factors (physical activity) but also to patient’s long-term immuno-virological course, type of antiretroviral drugs, antiretroviral treatment exposure and HIV-infection. As shown in prior studies, ART initiation was demonstrated to lead to a large initial decrease in BMD which thereafter reduces in its extent with substantial stability in BMD over time.6 High bone turnover states, such as HIV-induced osteoporosis, involve trabecular bone (spine) earlier and to a greater extent, compromising cortical bone (hip) only much later.7 In line with previous studies,3 the finding that optimal DCI is only associated with less late (femoral) osteoporosis may identify subjects at early rather than late stages of HIV-infection for whom nutrition does not compensate the effects of risk factors, other than those we considered in the analysis, which are associated with bone loss. In fact, subjects with femoral osteoporosis were older (median value: 60 vs 48 years), had a longer time of exposure to HIV infection
Table 2 Univariate and multivariate logistic regression to estimate predictive factors of an optimal calcium intake. Characteristics
Age (per 5-years older) Gender (Males vs Females) Smoker (Yes vs No) Body mass index (25 vs <25 kg/m2) Years since first HIV positive test (per year longer) Exposure to HAART (per year longer) Nadir CD4þ (per 100-cells/mL higher) HCVAb Positive vs Unknown Negative vs Unknown Type of antiretroviral regimen NRTI-based vs Otherc NNRTI-based vs Otherc PI-based vs Otherc Use of tenofovir (Yes vs No) CD4þ (per 100-cells/mL higher) CD4þ (<500 v s 500 cells/mL) HIV-RNA (50 vs <50 copies/mL) Total cholesterol (per 50-mg/dL higher) HDL-cholesterol (per 20-mg/dL higher) LDL-cholesterol (per 50-mg/dL higher) Triglycerides (per 50-mg/dL higher) Fasting glucose (per 20-mg/dL higher) 25-OH Vitamin D (per 10-ng/mL higher) Systolic blood pressure (per 10-mmHg higher) Diastolic blood pressure (per 10-mmHg higher) Framingham cardiovascular risk score (>20 vs 20%) Lumbar osteoporosisd (Present vs Absent) Femoral neck osteoporosisd (Present vs Absent)
Univariate analysis
Multivariate analysisa (model 1)
Multivariate analysisa (model 2)
ORb
95% CI (OR)
p-value
AORb
95% CI (OR)
p-value
AORb
95% CI (OR)
p-value
0.991 1.096 2.079 0.756 1.021 1.020 1.006
0.857e1.246 0.492e2.439 0.995e4.348 0.410e1.396 0.983e1.061 0.961e1.082 0.848e1.193
e e e 1.265 e e e e
e e e 0.609e2.709 e e e e
e e e 0.535 e e e e
e e e 1.193 e e e e
e e e 0.583e2.505 e e e e
e e e 0.634 e e e e
1.406 1.743
0.444e4.447 0.681e4.459
e
e
e
e
e
e
0.467 0.700 0.956 0.855 0.998 0.930 1.047 1.073 0.817 1.104 1.070 1.331 0.927 1.041 1.230 2.597 0.972 0.288
0.047e4.677 0.282e1.739 0.500e1.827 0.484e1.513 0.908e1.096 0.519e1.667 0.457e2.398 0.732e1.573 0.525e1.271 0.660e1.848 0.917e1.248 1.000e1.770 0.661e1.300 0.868e1.247 0.986e1.535 0.817e8.258 0.360e2.625 0.061e1.353
0.903 0.823 0.052 0.372 0.279 0.517 0.945 0.479 0.888 0.255 0.811 0.591 0.878 0.496 0.191 0.960 0.807 0.913 0.718 0.371 0.706 0.392 0.050 0.659 0.665 0.067 0.106 0.956 0.115
0.578 e e e e e e e 1.250 e e e 5.547 e 0.159
0.291e1.132 e e e e e e e 0.936e1.780 e e e 1.337e29.185 e 0.018e0.790
0.112 e e e e e e e 0.156 e e e 0.025 e 0.047
0.638 e e e e e e e 1.229 e e e 3.439 0.935 e
0.326e1.237 e e e e e e e 0.926e1.723 e e e 0.930e14.442 0.274e2.849 e
0.186 e e e e e e e 0.174 e e e 0.071 0.909 e
Abbreviation: HAART, highly active antiretroviral therapy; HCVAb, antibody anti-HCV; NRTIs ¼ nucleoside reverse transcriptase inhibitors; NNRTIs ¼ non-nucleoside reverse transcriptase inhibitors; PIs ¼ protease inhibitors; L1eL4, lumbar vertebrae 1e4. a Variables (other than CVD risk score or osteoporosis) with a P-value < 0.20 at the univariate analysis were entered into the multivariate model; Model 1 and 2 alternatively included femoral or lumbar osteoporosis. b Odds ratio of an optimal daily calcium intake defined as >1000 mg/day which is the intake recommended by the Institute of Medicine (9). c The category “Other” included new-drugs-based regimens with at least one of the following drugs: maraviroc, raltegravir, etravirine or darunavir/r. d Osteoporosis was defined according to the WHO classification by T-scores <2.5 for subjects older than 50 years of age or by Z-score <2.5 for subjects younger than 50 years of age.
366
L. Galli et al. / Clinical Nutrition 33 (2014) 363e366
(median value: 19 vs 16 years), a longer time to ART initiation (median value: 3.5 vs 2.1 years) and lower nadir CD4 (median value:169 vs 233 cells/mm3) than patients with vertebral osteoporosis. A potential explanation for the absence of relationship between optimal DCI and vertebral osteoporosis could be that these subjects might have been affected by vertebral arthrosis whose presence led to an overestimation of bone mineral density at this level. In the general population, many studies debated about the effect of calcium supplementation on the risk of cardiovascular events1,2: a possible explanation is related to the acute increase in serum calcium, which has been observed after ingestion of calcium supplements, but not after eating calcium-rich foods. High-normal serum calcium levels are associated with an increased risk of vascular calcification, carotid artery atheroma, cardiovascular events and mortality.8 Dietary calcium instead is taken in small amounts spread throughout the day; thus, it is absorbed slowly, causing little change in serum calcium levels.9 One cohort study,2 conducted on local German residents aged 35e64 years, analysed also the effect of dietary calcium intake: the authors concluded that increasing calcium intake from diet might not increase the risk of cardiovascular events or mortality. On the contrary, we found that, among HIV infected patients, an optimal DCI was predicted by a high CVR. This could be explained by the multifactorial causes of CVD in HIV infection, which imply complex interactions between traditional risk factors (confirmed also by our data), lifestyle, HIV-related and cART-related factors. Indeed, traditional risk factors are involved in the development of calcified lesions, considered stable, whereas inflammation and immune dysfunction caused by HIV may contribute to non-calcified plaque, with a greater risk of rupture.10 Among the limitations of the study, we point out the crosssectional nature of the study which does not allow to assess cause-and-effect relationships, the use of cardiovascular risk which gives less strength to our findings being a surrogate outcome of the cardiovascular events and finally the small number of patients with a high cardiovascular risk or with osteoporosis that did not allow to obtain accurate estimates of the magnitude of the effect of cardiovascular risk score and osteoporosis on DCI. In conclusion, we found that an optimal dietary calcium intake was more likely in patients with high cardiovascular risk and no femoral osteoporosis Our results may help in the clinical management of HIV-infected patients: interventions aimed at preventing osteoporosis should be adopted although they would also need to balance the cardiovascular risk; in chronically HIV-infected subjects, considering the lipid content of dairy products, which might contribute to a high CVR, behavioural changes in food intake should be promoted, in order to preserve calcium intake selecting nutrients (such as vegetables and water) that may balance the risk of cardiovascular events and osteoporosis.
Source of funding None. Statement of authorship LG conceived and designed the study, performed the statistical analyses and interpretations, and contributed to writing the manuscript. AB, CV and SS collected and updated data. AR, GM, MS, NG, VS and AL contributed to the interpretation of the results and to writing the manuscript. DC contributed to the writing of the manuscript. AC conceived the study and contributed to writing of the manuscript. Conflict of interest statement All the authors have no conflicts of interest. Acknowledgements We thank Dr. Giuseppe Balconi for his help in planning and executing dual-energy X-ray absorptiometry on the patients included in this study. References 1. Bolland MJ, Avenell A, Baron JA, Grey A, MacLennan GS, Gamble GD, et al. Effect of calcium supplements on risk of myocardial infarction and cardiovascular events: meta-analysis. BMJ 2010;341:c3691. 2. Li K, Kaaks R, Linseisen J, Rohrmann S. Associations of dietary calcium intake and calcium supplementation with myocardial infarction and stroke risk and overall cardiovascular mortality in the Heidelberg cohort of the European Prospective Investigation into Cancer and Nutrition study (EPIC-Heidelberg). Heart 2012;98:920ee5e. 3. Li Vecchi V, Soresi M, Giannitrapani L, Mazzola G, La Sala S, Tramuto F, et al. Dairy calcium intake and lifestyle risk factors for bone loss in hiv-infected and uninfected mediterranean subjects. BMC Infect Dis 2012;12:192. 4. Montomoli M, Gonnelli S, Giacchi M, Mattei R, Cuda C, Rossi S, et al. Validation of a food frequency questionnaire for nutritional calcium intake assessment in Italian women. Eur J Clin Nutr. 2002;56(1):21e30. 5. Ross AC, Taylor CL, Yaktine AL, Del Valle HB, Breiner H, Bandy A, et al. Committee to Review Dietary Reference Intakes for Vitamin D and Calcium. Dietary Reference Intakes for Calcium and Vitamin D e Institute of Medicine. National Academies of Sciences 2010; Department of Nutritional Sciences, The Pennsylvania State University, 110 Chandlee Laboratory, University Park, PA 16802, USA. 6. McComsey GA, Kitch D, Daar ES, Tierney C, Jahed NC, Tebas P, et al. Bone mineral density and fractures in antiretroviral-naïve persons randomized to receive abacavir-lamivudine or tenofovir disoproxil fumarate-emtricitabine along with efavirenz or atazanavir-ritonavir: AIDS Clinical Trials Group A5224s, a substudy of ACTG A5202. JID 2011;203:1791e801. 7. Borderi M, Gibellini D, Vescini F, De Crignis E, Cimatti L, Biagetti C, et al. Metabolic bone disease in HIV infection. AIDS 2009;23:1297e310. 8. Reid IR, Bolland MJ, Grey A. Does calcium supplementation increase cardiovascular risk? Clin Endocrinol 2010;73:689e95. 9. Reid IR, Bolland MJ. Calcium supplements: bad for the heart? Heart 2012;98(12):895e6. 10. Fitch K, Lo J, Abbara S, Ghoshhajra B, Shturman L, Soni A, et al. Increased coronary artery calcium score and noncalcified plaque among HIV-infected men: relationship to metabolic syndrome and cardiac risk parameters. J Acquir Immune Defic Syndr 2010;55:495e9.
Contents lists available at SciVerse ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Short communication
Optimal dietary calcium intake in HIV treated patients: No femoral osteoporosis but higher cardiovascular riskq Laura Galli a, *, Alessandro Rubinacci b, Deborah Cocorullo c, Stefania Salpietro a, Vincenzo Spagnuolo a, Nicola Gianotti a, Alba Bigoloni a, Concetta Vinci a, Giovanna Mignogna b, Marcella Sirtori b, Adriano Lazzarin a, c, Antonella Castagna a a b c
Department of Infectious Diseases, San Raffaele Scientific Institute, Via Stamira d’Ancona 20, 20127 Milan, Italy Bone Metabolism Unit, San Raffaele Scientific Institute, Milan, Italy Università Vita-Salute San Raffaele, Milan, Italy
a r t i c l e i n f o
s u m m a r y
Article history: Received 21 February 2013 Accepted 6 July 2013
Background & aims: We performed a cross-sectional study on adult HIV-infected patients, on HAART, without calcium or vitamin D supplementation to evaluate if the cardiovascular risk or the presence of osteoporosis may be predictive factors of an optimal daily calcium intake (DCI>1000 mg/day). Methods: Patients underwent a dual-energy X-ray absorptiometry, measured biochemical parameters and compiled a validated questionnaire for the assessment of DCI. Osteoporosis (OP) was defined according to the WHO classification at either the vertebral spine or femoral neck. Cardiovascular risk was assessed by the 10-year Framingham cardiovascular risk score. Results: 200 HIV-infected patients evaluated: 171 (86%) males with a median age of 48.1 (42.3e53.8) years and 10.6 (4.3e13.6) years of HAART exposure. DCI was 889 (589e1308) mg/day and 79 (40%) patients had an optimal DCI. Framingham risk>20% was found in 13 (6.7%) patients and femoral OP was diagnosed in 12 (6%) pts. By multivariate analysis, optimal DCI was more likely in patients with a Framingham risk>20% [OR ¼ 5.547, 95% CI:1.337, p ¼ 0.025] and less likely in patients with femoral osteoporosis [OR ¼ 0.159, 95% CI: 0.018e0.790, p ¼ 0.047]. Conclusions: We found that an optimal dietary calcium intake was more likely in patients with high cardiovascular risk and no femoral osteoporosis. Ó 2013 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Dietary calcium intake Osteoporosis Cardiovascular risk HIV-infected patients
1. Introduction Due to the success of combination antiretroviral therapy (cART), patients with HIV infection live longer and are exposed to ageingrelated complications such as cardiovascular diseases (CVD) and bone mineral disorders. Previous studies, conducted in the general population, have associated calcium supplements which are widely used in the prevention or treatment of osteoporosis, with a possible increase in the risk of adverse cardiovascular outcomes.1 One cohort study identified no adverse effects associated with dietary calcium intake, although a near doubling risk of myocardial infarction (MI) was found in calcium supplements users.2
q Conference presentation: Presented in part at the 4th ICAR, Napoli 2012. * Corresponding author. Tel.: þ39 0226437934; fax: þ39 0226437030. E-mail address: [email protected] (L. Galli).
Limited data on dietary calcium intake are available among HIVinfected subjects. Although one study has recently identified a protective effect of calcium intake on the osteoporosis among HIVinfected subjects,3 it remains unclear whether such advantage might enhance the cardiovascular risk. Thus, the aim of our study was to evaluate if the cardiovascular risk or the presence of osteoporosis may be predictive factors of an optimal daily calcium intake (DCI). 2. Materials and methods We conducted a cross-sectional study on alive HIVþ patients, aged>18 years, on HAART, without calcium or vitamin D supplementation, attending the Infectious Diseases Department of the San Raffaele Scientific Institute between MayeJune 2011. Patients, evaluated on the same day, underwent a dual-energy X-ray absorptiometry (DEXA), measured biochemical parameters and compiled a validated questionnaire for the assessment of DCI.4 Briefly, the questionnaire is composed of 15 food items: selection of the items
0261-5614/$ e see front matter Ó 2013 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. http://dx.doi.org/10.1016/j.clnu.2013.07.004
364
L. Galli et al. / Clinical Nutrition 33 (2014) 363e366
with a p-value <0.20 at the univariate logistic regression models were included into two multivariate models (Model1 considered among the covariates the diagnosis of osteoporosis at femoral neck and alternatively Model 2 considered the diagnosis of osteoporosis at vertebral spine). Odds ratios (OR) and the corresponding 95% confidence intervals (95% CI) were estimated and reported for each variable included in the model. A two-sided p value <0.05 was considered statistically significant. All statistical analyses were performed with SAS software, version 9.2 (SAS Institute, Cary, North Carolina, United States).
was based on the data of the Italian Institute of nutrition relative to the food composition of the Italian diet, frequency of use and relative importance as a calcium food source; additional subgroup items were added for high calcium content food such as milk and derivatives. Servings were listed in three quantitative groups identified as small, medium, large and with grams of food. Subjects were asked to indicate the relative weekly frequency of use of each item. The calcium content estimate was obtained by food tables (attached to the questionnaire) quantifying the calcium content of each food according to the serving and weekly frequency. The daily calcium intake was obtained dividing the weekly calcium intake by 7. Optimal DCI was defined as a DCI>1000 mg/day (recommended threshold in adults according to the IOM5). Osteopenia and Osteoporosis (OP) were defined according to the WHO classification at either the vertebral spine or femoral neck: osteoporosis was defined by T-scores<2.5 for subjects older than 50 years of age or by z-score<2.5 for subjects younger than 50 years of age; osteopenia was defined by T-scoresranging between >2.5/1.0 for subjects older than 50 years of age or by z-score ranging between >2.5/1.0 for subjects younger than 50 years of age Cardiovascular risk was assessed by the 10-year Framingham cardiovascular risk score and used in the analysis as a categorical variable according to the threshold of 20% (an arbitrary cut-off value often used in the clinical setting and in the clinical studies to identify patients with a high 10-year risk of cardiovascular events).
3. Results Two hundred HIV-infected patients were included: 171 (86%) were males; age was 48.1 (42.3e53.8) years; they were infected with HIV since 14.5 (7.1e19.5) years, had 10.6 (4.3e13.6) years of HAART. DCI was 889 (589e1308) mg/day and 79 (40%) patients had an optimal DCI. As shown in Table 1, no significant differences were observed between the demographic, clinical, immuno-virological and metabolic characteristics between subjects with or without an optimal DCI. Framingham risk >20% was found in 13 (6.7%) patients; 8/13 (61.5%) subjects had a DCI>1000 mg/day as compared to 71/187 (38.0%) of patients with a CVD risk<20% (p ¼ 0.141). Subjects with a high CVD risk were older (median: 61 vs 47 years, p < 0.0001), had a slightly higher body mass index (median 25.6 vs 23.5, p ¼ 0.086), had higher lipid values (median total cholesterol: 207 vs 188 mg/dL, p ¼ 0.068; median LDL-cholesterol: 153 vs 120, p ¼ 0.004; median triglycerides: 148 vs 120, p ¼ 0.056), systolic and diastolic blood pressure (median systolic: 149 vs 120 mmHg, p ¼ 0.0001; median diastolic: 84 vs 77 mmHg, p ¼ 0.028) and fasting glucose values (median: 100 vs 87, p ¼ 0.0003) as compared to those with a CVD risk<20%. Osteopenia was diagnosed in 96 (48%) patients: 18 (9%) subjects had a diagnosis only at vertebral spine, 37 (18.5%) patients only at
2.1. Statistical analysis To analyze whether the demographic and clinical characteristics of the subjects included in this study were similar, the chi-square test or the Wilcoxon rank sum tests were applied when dealing with categorical or continuous variables, respectively. At multivariate analysis, logistic regression was applied to determine predictive factors of a DCI>1000 mg/day. Characteristics
Table 1 Patients’ characteristics according to dietary calcium intake at the time of study evaluation. Characteristica
Overall (N ¼ 200)
DCI 1000 mg/day (N ¼ 121)
DCI >1000 mg/day (N ¼ 79)
P-value
Age (years) Males Smokers Overweight (Body mass index 25 kg/m2) Years since first HIV positive test Years of exposure to HAART Nadir Cd4þ (cells/mL) Positive HCVAb Type of antiretroviral regimen NRTI-based NNRTI-based PI-based Otherb Use of tenofovir CD4þ (cells/mL) CD4þ <500 cells/mL HIV-RNA <50 copies/mL Total cholesterol (mg/dL) HDL-cholesterol (mg/dL) LDL-cholesterol (mg/dL) Triglycerides (mg/dL) Fasting glucose (mg/dL) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) 25 (OH) Vitamin D (ng/mL)
48.1 171 94 62 14.5 10.6 229 30
(42.3e53.8) (86%) (47%) (31%) (7.1e19.5) (4.3e13.6) (102e329) (15%)
48 104 55 35 13.9 9.7 233 19
(42.7e53.8) (86%) (46%) (29%) (6.1e19.4) (4.1e13.5) (100e328) (16%)
48.2 67 39 27 16.4 11.6 216 11
(42.2e53.9) (85%) (49%) (34%) (7.1e20.2) (4.8e13.6) (108e329) (14%)
0.496d 0.840c 0.563c 0.365c 0.413d 0.782d 0.579d 0.475c 0.754c
4 27 79 74 89 540 85 164 188 44 123 124 88 121 79 19.8
(2%) (15%) (43%) (40%) (45%) (384e777) (45%) (86%) (164e218) (36e54) (103e151) (89e194) (80e97) (112e133) (70e87) (13.8e27.2)
3 18 45 45 52 542 52 99 189 46 125 120 88 122 77 20
(3%) (16%) (41%) (41%) (43%) (376e801) (45%) (86%) (164e217) (36e55) (99e152) (89e178) (79e97) (113e132) (70e85) (13.2e27.2)
1 9 34 29 37 531 33 65 186 42 121 129 87 120 79 19
(1%) (13%) (47%) (40%) (47%) (406e751) (43%) (86%) (164e218) (36e50) (107e151) (87e203) (80e97) (111e133) (71e89) (15.2e27.3)
0.663c 0.946d 0.882c 0.999c 0.904d 0.311d 0.760d 0.395d 0.630d 0.886d 0.144d 0.726d
Abbreviations: DCI, dietary calcium intake; NRTI, Nucleoside Reverse Transcriptase Inhibitors; NNRTI, Non-nucleoside Reverse Transcriptase Inhibitors; PI, Protease Inhibitors. a Median (first quartile- third quartile) or frequency (%) used to describe results of continuous or categorical variables, respectively. b Other included the following drugs: maraviroc, raltegravir, etravirine. c By chis-quare or Fisher exact test, as appropriate. d By Wilcoxon rank sum test.
L. Galli et al. / Clinical Nutrition 33 (2014) 363e366
365
observed only for a Framingham risk>20% [OR (>20% vs 20%) ¼ 3.439, 95% CI: 0.930e14.442, p ¼ 0.071].
femoral neck and 41 (20.5%) patients had a diagnosis at both vertebral spine and femoral neck. OP was diagnosed in 25 (12.5%) patients: 13 (6.5%) patients had a diagnosis at vertebral spine, 7 (3.5%) patients at femoral neck and 5 (2.5%) patients had a diagnosis at both vertebral spine and femoral neck. An optimal DCI tended to be less frequent among patients with femoral OP [patients with femoral OP: 2/12 (16.7%); patients without femoral OP: 77/188 (41%), p ¼ 0.130]; optimal and optimal DCI was similarly distributed among patients with or without vertebral OP [patients with vertebral OP: 7/18 (38.9%); patients without vertebral OP: 72/182 (39.6%), p ¼ 0.999]. Among patients with a CVD risk >20%, 3/13 (25%) had a femoral OP diagnosis as compared to 9/187 (4.8%) of patients with a CVD risk less than 20% (p ¼ 0.037); no association between CVD risk and vertebral OP was observed (Table 1). Results from univariate and multivariate logistic regression are described in Table 2; only few covariates were found to be predictive of an optimal DCI and were entered into the multivariate models. In the first model, after adjustment for body mass index, current use of tenofovir and fasting glucose, an optimal DCI resulted to be more likely in patients with a Framingham risk>20% [OR (>20% vs 20%) ¼ 5.547, 95% CI: 1.337e29.185, p ¼ 0.025] and less likely in patients with the presence of femoral osteoporosis [OR (OP vs no-OP) ¼ 0.159, 95% CI: 0.018e0.790, p ¼ 0.047]. In the second multivariate model which considered the diagnosis of vertebral osteoporosis among the covariates, no significant predictors of an optimal DCI were found; a marginal effect was
4. Discussion To our knowledge, this is the only study to date evaluating if optimal DCI was predicted by OP and CVR among HIV-infected subjects. We found that optimal DCI does seem to be more likely in subjects with no femoral OP, though not with vertebral OP, and in subjects with an increased CVR. Among HIV-positive subjects, osteoporosis may be related to ageing and life-style risk factors (physical activity) but also to patient’s long-term immuno-virological course, type of antiretroviral drugs, antiretroviral treatment exposure and HIV-infection. As shown in prior studies, ART initiation was demonstrated to lead to a large initial decrease in BMD which thereafter reduces in its extent with substantial stability in BMD over time.6 High bone turnover states, such as HIV-induced osteoporosis, involve trabecular bone (spine) earlier and to a greater extent, compromising cortical bone (hip) only much later.7 In line with previous studies,3 the finding that optimal DCI is only associated with less late (femoral) osteoporosis may identify subjects at early rather than late stages of HIV-infection for whom nutrition does not compensate the effects of risk factors, other than those we considered in the analysis, which are associated with bone loss. In fact, subjects with femoral osteoporosis were older (median value: 60 vs 48 years), had a longer time of exposure to HIV infection
Table 2 Univariate and multivariate logistic regression to estimate predictive factors of an optimal calcium intake. Characteristics
Age (per 5-years older) Gender (Males vs Females) Smoker (Yes vs No) Body mass index (25 vs <25 kg/m2) Years since first HIV positive test (per year longer) Exposure to HAART (per year longer) Nadir CD4þ (per 100-cells/mL higher) HCVAb Positive vs Unknown Negative vs Unknown Type of antiretroviral regimen NRTI-based vs Otherc NNRTI-based vs Otherc PI-based vs Otherc Use of tenofovir (Yes vs No) CD4þ (per 100-cells/mL higher) CD4þ (<500 v s 500 cells/mL) HIV-RNA (50 vs <50 copies/mL) Total cholesterol (per 50-mg/dL higher) HDL-cholesterol (per 20-mg/dL higher) LDL-cholesterol (per 50-mg/dL higher) Triglycerides (per 50-mg/dL higher) Fasting glucose (per 20-mg/dL higher) 25-OH Vitamin D (per 10-ng/mL higher) Systolic blood pressure (per 10-mmHg higher) Diastolic blood pressure (per 10-mmHg higher) Framingham cardiovascular risk score (>20 vs 20%) Lumbar osteoporosisd (Present vs Absent) Femoral neck osteoporosisd (Present vs Absent)
Univariate analysis
Multivariate analysisa (model 1)
Multivariate analysisa (model 2)
ORb
95% CI (OR)
p-value
AORb
95% CI (OR)
p-value
AORb
95% CI (OR)
p-value
0.991 1.096 2.079 0.756 1.021 1.020 1.006
0.857e1.246 0.492e2.439 0.995e4.348 0.410e1.396 0.983e1.061 0.961e1.082 0.848e1.193
e e e 1.265 e e e e
e e e 0.609e2.709 e e e e
e e e 0.535 e e e e
e e e 1.193 e e e e
e e e 0.583e2.505 e e e e
e e e 0.634 e e e e
1.406 1.743
0.444e4.447 0.681e4.459
e
e
e
e
e
e
0.467 0.700 0.956 0.855 0.998 0.930 1.047 1.073 0.817 1.104 1.070 1.331 0.927 1.041 1.230 2.597 0.972 0.288
0.047e4.677 0.282e1.739 0.500e1.827 0.484e1.513 0.908e1.096 0.519e1.667 0.457e2.398 0.732e1.573 0.525e1.271 0.660e1.848 0.917e1.248 1.000e1.770 0.661e1.300 0.868e1.247 0.986e1.535 0.817e8.258 0.360e2.625 0.061e1.353
0.903 0.823 0.052 0.372 0.279 0.517 0.945 0.479 0.888 0.255 0.811 0.591 0.878 0.496 0.191 0.960 0.807 0.913 0.718 0.371 0.706 0.392 0.050 0.659 0.665 0.067 0.106 0.956 0.115
0.578 e e e e e e e 1.250 e e e 5.547 e 0.159
0.291e1.132 e e e e e e e 0.936e1.780 e e e 1.337e29.185 e 0.018e0.790
0.112 e e e e e e e 0.156 e e e 0.025 e 0.047
0.638 e e e e e e e 1.229 e e e 3.439 0.935 e
0.326e1.237 e e e e e e e 0.926e1.723 e e e 0.930e14.442 0.274e2.849 e
0.186 e e e e e e e 0.174 e e e 0.071 0.909 e
Abbreviation: HAART, highly active antiretroviral therapy; HCVAb, antibody anti-HCV; NRTIs ¼ nucleoside reverse transcriptase inhibitors; NNRTIs ¼ non-nucleoside reverse transcriptase inhibitors; PIs ¼ protease inhibitors; L1eL4, lumbar vertebrae 1e4. a Variables (other than CVD risk score or osteoporosis) with a P-value < 0.20 at the univariate analysis were entered into the multivariate model; Model 1 and 2 alternatively included femoral or lumbar osteoporosis. b Odds ratio of an optimal daily calcium intake defined as >1000 mg/day which is the intake recommended by the Institute of Medicine (9). c The category “Other” included new-drugs-based regimens with at least one of the following drugs: maraviroc, raltegravir, etravirine or darunavir/r. d Osteoporosis was defined according to the WHO classification by T-scores <2.5 for subjects older than 50 years of age or by Z-score <2.5 for subjects younger than 50 years of age.
366
L. Galli et al. / Clinical Nutrition 33 (2014) 363e366
(median value: 19 vs 16 years), a longer time to ART initiation (median value: 3.5 vs 2.1 years) and lower nadir CD4 (median value:169 vs 233 cells/mm3) than patients with vertebral osteoporosis. A potential explanation for the absence of relationship between optimal DCI and vertebral osteoporosis could be that these subjects might have been affected by vertebral arthrosis whose presence led to an overestimation of bone mineral density at this level. In the general population, many studies debated about the effect of calcium supplementation on the risk of cardiovascular events1,2: a possible explanation is related to the acute increase in serum calcium, which has been observed after ingestion of calcium supplements, but not after eating calcium-rich foods. High-normal serum calcium levels are associated with an increased risk of vascular calcification, carotid artery atheroma, cardiovascular events and mortality.8 Dietary calcium instead is taken in small amounts spread throughout the day; thus, it is absorbed slowly, causing little change in serum calcium levels.9 One cohort study,2 conducted on local German residents aged 35e64 years, analysed also the effect of dietary calcium intake: the authors concluded that increasing calcium intake from diet might not increase the risk of cardiovascular events or mortality. On the contrary, we found that, among HIV infected patients, an optimal DCI was predicted by a high CVR. This could be explained by the multifactorial causes of CVD in HIV infection, which imply complex interactions between traditional risk factors (confirmed also by our data), lifestyle, HIV-related and cART-related factors. Indeed, traditional risk factors are involved in the development of calcified lesions, considered stable, whereas inflammation and immune dysfunction caused by HIV may contribute to non-calcified plaque, with a greater risk of rupture.10 Among the limitations of the study, we point out the crosssectional nature of the study which does not allow to assess cause-and-effect relationships, the use of cardiovascular risk which gives less strength to our findings being a surrogate outcome of the cardiovascular events and finally the small number of patients with a high cardiovascular risk or with osteoporosis that did not allow to obtain accurate estimates of the magnitude of the effect of cardiovascular risk score and osteoporosis on DCI. In conclusion, we found that an optimal dietary calcium intake was more likely in patients with high cardiovascular risk and no femoral osteoporosis Our results may help in the clinical management of HIV-infected patients: interventions aimed at preventing osteoporosis should be adopted although they would also need to balance the cardiovascular risk; in chronically HIV-infected subjects, considering the lipid content of dairy products, which might contribute to a high CVR, behavioural changes in food intake should be promoted, in order to preserve calcium intake selecting nutrients (such as vegetables and water) that may balance the risk of cardiovascular events and osteoporosis.
Source of funding None. Statement of authorship LG conceived and designed the study, performed the statistical analyses and interpretations, and contributed to writing the manuscript. AB, CV and SS collected and updated data. AR, GM, MS, NG, VS and AL contributed to the interpretation of the results and to writing the manuscript. DC contributed to the writing of the manuscript. AC conceived the study and contributed to writing of the manuscript. Conflict of interest statement All the authors have no conflicts of interest. Acknowledgements We thank Dr. Giuseppe Balconi for his help in planning and executing dual-energy X-ray absorptiometry on the patients included in this study. References 1. Bolland MJ, Avenell A, Baron JA, Grey A, MacLennan GS, Gamble GD, et al. Effect of calcium supplements on risk of myocardial infarction and cardiovascular events: meta-analysis. BMJ 2010;341:c3691. 2. Li K, Kaaks R, Linseisen J, Rohrmann S. Associations of dietary calcium intake and calcium supplementation with myocardial infarction and stroke risk and overall cardiovascular mortality in the Heidelberg cohort of the European Prospective Investigation into Cancer and Nutrition study (EPIC-Heidelberg). Heart 2012;98:920ee5e. 3. Li Vecchi V, Soresi M, Giannitrapani L, Mazzola G, La Sala S, Tramuto F, et al. Dairy calcium intake and lifestyle risk factors for bone loss in hiv-infected and uninfected mediterranean subjects. BMC Infect Dis 2012;12:192. 4. Montomoli M, Gonnelli S, Giacchi M, Mattei R, Cuda C, Rossi S, et al. Validation of a food frequency questionnaire for nutritional calcium intake assessment in Italian women. Eur J Clin Nutr. 2002;56(1):21e30. 5. Ross AC, Taylor CL, Yaktine AL, Del Valle HB, Breiner H, Bandy A, et al. Committee to Review Dietary Reference Intakes for Vitamin D and Calcium. Dietary Reference Intakes for Calcium and Vitamin D e Institute of Medicine. National Academies of Sciences 2010; Department of Nutritional Sciences, The Pennsylvania State University, 110 Chandlee Laboratory, University Park, PA 16802, USA. 6. McComsey GA, Kitch D, Daar ES, Tierney C, Jahed NC, Tebas P, et al. Bone mineral density and fractures in antiretroviral-naïve persons randomized to receive abacavir-lamivudine or tenofovir disoproxil fumarate-emtricitabine along with efavirenz or atazanavir-ritonavir: AIDS Clinical Trials Group A5224s, a substudy of ACTG A5202. JID 2011;203:1791e801. 7. Borderi M, Gibellini D, Vescini F, De Crignis E, Cimatti L, Biagetti C, et al. Metabolic bone disease in HIV infection. AIDS 2009;23:1297e310. 8. Reid IR, Bolland MJ, Grey A. Does calcium supplementation increase cardiovascular risk? Clin Endocrinol 2010;73:689e95. 9. Reid IR, Bolland MJ. Calcium supplements: bad for the heart? Heart 2012;98(12):895e6. 10. Fitch K, Lo J, Abbara S, Ghoshhajra B, Shturman L, Soni A, et al. Increased coronary artery calcium score and noncalcified plaque among HIV-infected men: relationship to metabolic syndrome and cardiac risk parameters. J Acquir Immune Defic Syndr 2010;55:495e9.