| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH COMUNICATION |
1 From the Johns Hopkins University Center for Human Nutrition, School of Hygiene and Public Health, Baltimore; the Department of Pediatric Gastroenterology and Nutrition, Johns Hopkins Hospital, Baltimore; and the US Department of Agriculture/Agricultural Research Service Children's Nutrition Research Center, Houston.
2 Supported by an NIH General Clinical Research Center grant (M01RR000520347) and the Johns Hopkins University School of Hygiene and Public Health Gilbert Award. 3 Address reprint requests to K O'Brien, Johns Hopkins University School of Hygiene and Public Health, 615 North Wolfe Street, Room 2041, Baltimore, MD 21205-2179. E-mail: kobrien{at}jhsph.edu.
ABSTRACT
Background: Early diagnostic efforts and advances in multidrug therapy have considerably prolonged the survival time of children infected perinatally with HIV. Despite these advances, few studies have addressed calcium status and bone growth in HIV-infected children.
Objective: Our objective was to examine the effect of HIV infection on calcium status and bone growth in children.
Design: We measured calcitropic hormones, urinary calcium excretion, bone mineral content, and body composition in 19 young girls aged 9.2 ± 2.6 y (range: 5.9–15.2 y) who were infected perinatally with HIV.
Results: Serum concentrations of 1,25-dihydroxyvitamin D [1,25(OH)2D] and parathyroid hormone concentrations were elevated above normal ranges in 25% and 12% of these girls, respectively. Urinary calcium excretion normalized for creatinine excretion was also elevated (Ca/Cr >0.18) in 17% of these children despite suboptimal calcium intakes (679 ± 437 mg/d). Total-body bone mineral content, measured with the use of dual-energy X-ray absorptiometry, averaged 845.1 ± 279.0 g and was on average 2.7 z scores below age- and race-matched values reported in non-HIV-infected healthy girls. Significant positive correlations were found between an indirect marker of bone resorption in urine (N-telopeptide) and 1,25(OH)2D (P < 0.02, r2 = 0.586, n = 9), and between serum N-telopeptide and total alkaline phosphatase (P < 0.001, r2 = 0.541, n = 17), suggesting that calcium insufficiency may be increasing bone resorption in this group.
Conclusions: Young girls with HIV infection had low bone mass and evidence of calcium insufficiency. Nutritional counseling of children with HIV infection should emphasize adequate calcium intakes because of the importance of this age period in bone mineral acquisition.
Key Words: Girls • HIV infection • bone density • calcium • calcitropic hormones • body composition • nutrition • 1,25-dihydroxyvitamin D
INTRODUCTION
Little information is currently available regarding the nutritional requirements of HIV-infected children. Growth retardation is common in HIV-infected children (1–3) and is associated with a significantly shorter survival time than in noninfected children (4). Furthermore, poor growth is frequently a presenting symptom of HIV infection and may also be predictive of disease progression (5). The mechanisms by which HIV infection influences growth are not known with certainty, but are likely to be related to both increased metabolic requirements and inadequate nutrient intake. Predictors of growth failure have been difficult to identify in HIV-infected children because growth failure occurs over a wide range of CD4 cell numbers and in all Centers for Disease Control and Prevention (CDC) classifications (6).
To our knowledge, no data are currently available on the effect of the impaired growth frequently observed in HIV-infected children on calcium status and bone mineral content (BMC). This is alarming given that the calcium requirement of HIV-infected children may exceed that of the general population because of an increased risk of nutrient malabsorption (7, 8). Furthermore, studies in HIV-infected adults have abnormal bone turnover (9) and alterations in calcitropic hormones (10, 11).
To date, limited data are available on HIV infection and bone mineral density (BMD) in adult populations. Three reports in HIV-seropositive adults (12–14) showed either no evidence of osteopenia, or minor, nonsignificant reductions in bone mass compared with control populations (14). Furthermore, the rate of bone loss observed in HIV-infected men was not significantly different from that observed in a control group over a 15-mo study interval (14).
HIV infection, however, may have disparate effects on BMD in pediatric compared with adult populations because of the rapid bone acquisition that occurs during childhood. The implication of HIV infection on bone growth may be of particular concern in females because 37% of bone mass is accumulated between pubertal stage 2 and pubertal stage 5 in girls (15). Because females have an earlier acquisition of peak bone mass and a markedly high risk of developing osteoporosis later in life, the potential effect of HIV infection on bone acquisition may be most apparent in females.
Early identification and treatment of nutritional deficiencies in HIV-infected children may assist in decreasing morbidity and improving the quality of life for this group. The current study was designed to begin examining calcium status, calcitropic hormones, and bone development in HIV-infected girls.
SUBJECTS AND METHODS
Subject recruitment
Young African American (n = 17) and white females (n = 2) between the ages of 5 and 15 y with perinatally acquired HIV infection were recruited from a patient population of HIV-infected children attending the Intensive Primary Care Clinic at Johns Hopkins Hospital. All HIV-infected girls attending this clinic for routine health care were eligible to participate in the study if they were afebrile and free of acute infections at the time of the study and if they were not immobile or had no other illnesses that affected their mobility. More than 90% of the HIV-infected children seen at this clinic are African American and the racial distribution of the subjects recruited for this study reflects of this distribution. Children were receiving well-child care, medical care for acute infections or other health problems, and nutritional intervention as clinically indicated. They were recruited sequentially as they came into the clinic for their prescheduled health checks. Study participants were recruited in the fall of 1997 and all were receiving antiretroviral therapy. HIV infection and CDC classification were established according to the 1994 revised classification system (16). Informed, written consent and assent was obtained from all children and their parents or guardians. The study was approved by the Joint Committee of Clinical Investigations at Johns Hopkins University.
Study design
Studies were undertaken on the same day that children had a scheduled clinic visit for a routine health checkup. On this day, a spot urine sample was collected and an additional blood sample (15 mL) was collected for analysis of a standard blood chemistry profile, calcitropic hormone concentration, HIV RNA concentration, and CD4 cell number and percentage. Serum and urinary measurements were compared with normal ranges reported from each reference laboratory.
On the day of the study, each girl's weight and height were obtained and these values were converted to z scores with the use of the National Center for Health Statistics reference data according to standard procedures (17). On the day of the study, bone mineral mass and body composition were also measured. These values were compared with values for a reference Houston population of 483 non-HIV-infected children (18). A 24-h dietary recall was also obtained and the nutrient content determined by using the NUTRITIONIST IV software package (version 4.1, 1996; N-Squared Computing, Salem, OR).
Bone density and body-composition assessment
Total-body BMC, total-body BMD, and body composition [including percentage fat mass and lean tissue mass (LTM)] were measured using fan-beam dual-energy X-ray absorptiometry (DXA) (QDR 4500A; Hologic Inc, Waltham, MA). This procedure is rapid (
10 min), noninvasive, and requires minimal radiation exposure (equivalent to 1 d of natural exposure to the background concentration of radiation).
All body-composition and bone mineral data were compared with a database of healthy, non-HIV-infected African American (n = 187) and white (n = 296) girls who had participated in research studies at the Children's Nutrition Research Center at Baylor College of Medicine (18). This database was acquired with the use of both a Hologic QDR 2000W, software version 5.57, and more recently, a Hologic QDR 4500A fan-beam DXA instrument (Hologic Inc). To ensure that the measurements were standardized between the 2 instruments, the results for the older DXA instrument were adjusted by a cross-calibrations between the pencil-beam and fan-beam techniques (19).
An age-matched, ethnic-specific z score for BMC was obtained for each HIV-infected child. This value was defined as
 |
where the BMCage and SDage values were derived from the appropriate age and ethnic reference group. A similar z score was derived for LTM:
 |
where the LTMage and SDage values were obtained from the appropriate age and ethnic reference group.
The age-matched z score does not take into consideration differences in body size between HIV and reference groups. To match the 2 groups for body size, prediction models for whole body BMC were derived by using age, height, and ethnicity as the independent parameters. The multiple regression analyses were performed by using the reference population, with and without log transformations, and a dummy variable for ethnicity (1 = African American, 0 = White). The best prediction model (r2 = 0.93, P < 0.0001) was
 |
where height is in cm, age is in y, and BMC is in g. The measured BMC was compared with the predicted value and expressed as the BMC ratio (BMCmeasured/BMCpredicted). The mean (±SD) BMC ratio for the reference population was 1.000 ± 0.125. A similar prediction model was derived for LTM and used to calculate a LTM ratio. The mean (±SD) LTM ratio for the reference population was 1.000 ± 0.095.
In adult women, osteopenia is typically defined when BMD is between 1 SD and 2.5 SD below the mean of young women (20). Using an analogy to this definition, children with a BMC ratio >1 SD below the mean observed in a non-HIV infected reference populations would be defined as osteopenic (BMC ratio 
0.88). Our use of a BMC ratio may be a more stringent and accurate approach to define osteopenia in this study population because it measures total body BMC (as opposed to BMC at only one particular site) and also adjusts for the height, age, and ethnicity of each subject.
Serum measurements
Serum 1,25-dihydroxyvitamin D [1,25(OH)2D] and serum parathyroid hormone concentrations were measured by Smith Kline Laboratories using radioimmunoassays. Serum alkaline phosphatase was measured with the use of an autoanalyzer procedure in the Johns Hopkins Hospital reference laboratory (Hitachi 917; Boehring Mannheim, Indianapolis). All data that fell outside of the expected reference range were indicated individually by the reference laboratory. The expected reference range for the 1,25(OH)2D analysis was between 15 and 60 ng/L, whereas that for parathyroid hormone was between 10 and 65 ng/L.
Plasma HIV RNA was measured by reverse transcriptase polymerase chain reaction by the Johns Hopkins Hospital reference laboratory. The percentage of total lymphocytes and CD4 cell number were analyzed with the use of flow cytometry in the pathology department at the Johns Hopkins Hospital. An indirect marker of bone resorption (N-telopeptide) was measured in serum samples with the use of a commercially available ELISA method (Ostex, Seattle).
Urinary analyses
Spot urine samples were collected for routine urinalysis and urinary calcium excretion was determined with the use of flame atomic absorption spectrophotometry (Model 3300; Perkin Elmer, Norwalk, CT). Urinary creatinine concentrations were determined with the use of a standard colorimetric procedure. Mineral excretion was reported in relation to the creatinine content of the urine sample. N-Telopeptide was also measured in spot urine samples with the use of a commercially available ELISA method (Ostex).
Statistical analysis
BMC and body-composition values obtained in the HIV-infected girls were compared with a prediction model obtained in non-HIV-infected girls matched for age, race, and height. Statistical differences between these groups were examined by using the STATVIEW statistical package (version 5.0.1; SAS Institute Inc, Cary, NC). Stepwise and multiple regression analyses were used to examine relations between hormone concentrations, viral load, and other clinically measured variables and BMC. Differences were considered significant if P < 0.05.
RESULTS
The physical characteristics of the study population are shown in Table 1
. In general, these girls were below average in height as evidenced by an average height z score 1 SD below the median National Center for Health Statistics (NCHS) value. Despite their shorter stature, the girls' average weight-to-height z score was 0.30 ± 1.29. Dietary intakes of calcium in this group averaged 679 ± 437 mg/d.
|
). Predicted BMC for both age and height was used to calculate the BMC ratios shown in Table 2. BMC ratios, defined as measured BMC to predicted BMC of the HIV-infected girls, remained below what was expected in healthy children after adjustment for both age and height (Figure 2). The ratio of measured to predicted LTM exhibited a different pattern from BMC in that most of the HIV-infected children had values within 1 SD of the predicted value (Figure 3).
|
|
|
|
Evidence of hypercalciuria (Ca/Cr >0.18) was found in 17.6% (3/17) of the children in whom this measurement was available. Urinary Ca/Cr values were not significantly related to calcitropic hormone concentrations or to indirect markers of bone resorption (N-telopeptide) measured in urine. However, urinary N-telopeptide excretion was significantly related to serum concentrations of 1,25(OH)2D (y = -549.632 + 16.351x; r2 = 0.586, P < 0.02, n = 9). N-Telopeptide concentrations measured in serum were also significantly related to serum total alkaline phosphatase (y = 20.225 + 0.159x; r2 = 0.541, P < 0.001, n = 17).
Average serum concentrations of 1,25(OH)2 D and parathyroid hormone are shown in Table 3. In this group of children, 2 of 17 girls had parathyroid hormone concentrations >65 ng/L and 3 of 12 girls had calcitriol concentrations >60 ng/L.
|
Suboptimal bone mineralization and evidence of calcium insufficiency was found in this population of HIV-infected girls. The mechanisms responsible for these alterations are almost certainly multifactorial, involving both nutritional and disease related processes.
Calcium intakes in this study population were 20–50% below recommended calcium intakes for girls in this age range (21). Because calcium intake was determined with the use of only a single 24-h dietary recall, this value is only an estimate and larger studies are clearly needed to better characterize habitual calcium intake in this population. However, a previous study in 38 children from this same population showed that a 24-h recall provides data similar to that obtained with the use of a 24-h weighed-diet record (22). Although the estimate of calcium intake in this group is limited, the group's mean calcium intake was substantially lower than the calcium intake threshold (1500 mg/d) necessary for maximal calcium retention in healthy children of this age (23). Optimal calcium intakes may be even higher for HIV-infected children than for uninfected children because of the likelihood of nutrient malabsorption in this group.
To date no data on calcium absorption or balance have been reported in HIV-infected individuals. Fat malabsorption has been reported in HIV-infected adults (24). This, coupled with the gastrointestinal dysfunction and disaccharide intolerance that have been reported in HIV-infected children (7, 8), may reduce calcium absorption, thus decreasing the net amount of calcium available for bone mineralization.
During childhood, urinary calcium excretion does not significantly increase in response to increases in dietary calcium intake or net calcium absorption, presumably because the additional available calcium is used for bone accretion (25, 26). In these HIV-infected girls, hypercalciuria (Ca/Cr >0.18) was shown in nearly 20% of this group. This occurred even though our study population was predominantly black, although several studies reported lower urinary calcium excretion (27, 28) and a lower incidence of hypercalciuria in black than in white children (26).
The observed increases in urinary calcium excretion might be indicative of alterations in bone resorption. To indirectly examine bone resorption, we monitored N-telopeptide, a specific cross-linking domain of bone type 1 collagen, in this population. This marker is released into the circulation during bone resorption and concentrations, measured in either urine or serum, can be used as an indication of the degree of bone resorption (29, 30). Concentrations of urinary N-telopeptide were variable in these girls, but were significantly related to serum 1,25(OH)2 D concentrations. Serum N-telopeptide concentrations were also significantly related to serum alkaline phosphatase, which suggests increased bone resorption due to calcium insufficiency. Several studies have indicated that both osteogenic cells and bone are targets for HIV infection (31–33). Studies in adult populations showed low osteocalcin concentrations and reduced rates of bone turnover by tetracycline labeling (9, 12). Moreover, cytokines known to activate osteoclasts (IL-6 and tumor necrosis factor) were also shown to be elevated in HIV-positive individuals (34, 35).
Suboptimal calcium intakes, excessive urinary or fecal losses, and imbalances in bone turnover may all lead to impaired bone mineral acquisition. This population of HIV-infected girls was markedly osteopenic as evidenced by a mean z score of -2.7 for total body BMC in comparison with age-matched, non-HIV-infected girls. Even though the control population comprised healthy African American and white children from the Houston area, we have no reason to believe that the observed differences in bone mass were solely a result of geographical, social, or economic differences between these populations. The degree to which the low mean height of these HIV-infected girls was responsible for the low BMC is more difficult to quantify. However, to account for the low height of our subjects, a predicted total body BMC was determined after adjustment for both age and height. When we used this approach, BMC was still substantially lower than the norm, with the majority of HIV-positive children having BMC ratios lower than age- and height-matched, non-HIV-infected children. Unlike the finding with BMC, LTM did not decrease below predicted height- and age-adjusted values as markedly. The average LTM in this group of girls (21.1 ± 6.0 kg) was comparable with that reported in another DXA study of 20 prepubertal HIV-infected children (18.4 ± 5.5 kg) (36).
The mechanisms by which HIV infection might influence bone mass are numerous. A more recent study of HIV-positive hemophilic males with abnormal height or growth velocity showed that this group had delayed pubertal progression and a significantly lower bone age than did HIV-positive hemophilic subjects of the same age without growth abnormalities (37). Furthermore, one-half of the HIV-positive group had abnormal peak growth hormone concentrations after clonidine stimulation and significantly lower age-adjusted testosterone concentrations compared with the HIV-negative hemophiliac males. These investigators concluded that HIV infection reduced growth hormone production or release and decreased androgen secretion (37). The effect of HIV infection on pubertal progression, sex steroid concentrations, and growth hormone secretion in our study population was not addressed.
In conclusion, alterations in calcitropic hormones and deficits in bone mineral mass were observed in girls with perinatally acquired HIV infection. Although most of this deficit can be attributed to the reduced height of this group, the HIV-infected girls still had suboptimal BMC after we accounted for the height differences. Because of the recent advances in multidrug therapy, individuals are now surviving for prolonged periods over which the long-term adverse effects of suboptimal bone acquisition may become apparent. Larger studies are needed to establish the optimal nutrient intake required to maximize growth and bone acquisition in HIV-positive children and to address the ability of treatment interventions to improve bone mass and calcium status in this group.
ACKNOWLEDGMENTS
We thank the girls who volunteered to participate in this study and the staff of the Pediatric Clinical Research Unit and Intensive Primary Care Clinic for their care of these subjects.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  S. M. Arpadi, D. McMahon, E. J. Abrams, M. Bamji, M. Purswani, E. S. Engelson, M. Horlick, and E. Shane Effect of Bimonthly Supplementation With Oral Cholecalciferol on Serum 25-Hydroxyvitamin D Concentrations in HIV-Infected Children and Adolescents Pediatrics, January 1, 2009; 123(1): e121 - e126. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. B Stephensen, G. S Marquis, L. A Kruzich, S. D Douglas, G. M Aldrovandi, and C. M Wilson Vitamin D status in adolescents and young adults with HIV infection Am. J. Clinical Nutrition, May 1, 2006; 83(5): 1135 - 1141. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Hazra, R. I. Gafni, F. Maldarelli, F. M. Balis, A. N. Tullio, E. DeCarlo, C. J. Worrell, S. M. Steinberg, J. Flaherty, K. Yale, et al. Tenofovir Disoproxil Fumarate and an Optimized Background Regimen of Antiretroviral Agents as Salvage Therapy for Pediatric HIV Infection Pediatrics, December 1, 2005; 116(6): e846 - e854. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Schulze, K. O. O'Brien, E. L. Germain-Lee, S. L. Booth, A. Leonard, and B. J. Rosenstein Calcium Kinetics Are Altered in Clinically Stable Girls with Cystic Fibrosis J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3385 - 3391. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Arpadi, M. Horlick, and E. Shane Metabolic Bone Disease in Human Immunodeficiency Virus-Infected Children J. Clin. Endocrinol. Metab., January 1, 2004; 89(1): 21 - 23. [Full Text] [PDF]
|
|
|
|
|
|
S. Mora, I. Zamproni, S. Beccio, R. Bianchi, V. Giacomet, and A. Vigano Longitudinal Changes of Bone Mineral Density and Metabolism in Antiretroviral-Treated Human Immunodeficiency Virus-Infected Children J. Clin. Endocrinol. Metab., January 1, 2004; 89(1): 24 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. O O'Brien, M. S. Nathanson, J. Mancini, and F. R Witter Calcium absorption is significantly higher in adolescents during pregnancy than in the early postpartum period Am. J. Clinical Nutrition, December 1, 2003; 78(6): 1188 - 1193. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Gaughan, L. M. Mofenson, M. D. Hughes, G. R. Seage III, G. L. Ciupak, and J. M. Oleske Osteonecrosis of the Hip (Legg-Calve-Perthes Disease) in Human Immunodeficiency Virus-Infected Children Pediatrics, May 1, 2002; 109(5): e74 - 74. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and white (
) girls in relation to normative curves obtained in a healthy population of African American (n = 187) and white (n = 296) girls. This population of HIV-infected girls was markedly osteopenic as evidenced by a mean z score for total-body BMC of -2.7 in comparison with age-matched non-HIV-infected girls. As age increased, the observed BMC of these girls tended to fall further below the normative curves.
