Bone accrual and structural changes over one year in youth with cystic fibrosis

Abstract

Background

Pediatric bone accrual governs peak bone mass and strength. Longitudinal studies of bone health in youth with cystic fibrosis (CF) may provide insight into CF-related bone disease (CFBD), a prevalent co-morbidity in adults with CF.

Methods

This one-year longitudinal study of youth with pancreatic insufficient CF, enrolled in a nutrition intervention study [n = 62 (36 M/26F)] 1) examined dual-energy x-ray absorptiometry (DXA)-defined lumbar spine (LS) and total body less head (TBLH) bone accrual and 2) compared their changes in peripheral quantitative computed tomography (pQCT) cortical and trabecular tibial bone density and geometry to those of a healthy reference group [n = 143 (68 M/75F)].

Main outcome measures were 1) DXA: lumbar spine areal bone mineral density (LSaBMD) and total body less head bone mineral content (TBLH-BMC), sex- and pubertal status-specific, height velocity (HV)-adjusted or HV and lean body mass velocity (HV-LBMV)-adjusted annualized velocity-Z scores and 2) pQCT: age, sex, pubertal status and, when appropriate, tibial length adjusted Z-scores for bone architecture measures.

DXA velocity-Z were compared to expected mean of 0 and correlations with clinical parameters (age, BMI-Z and FEV₁%-predicted) tested. Within-subject comparisons of HV-adjusted and LBMV-HV-adjusted DXA velocity-Z were conducted in CF.

pQCT Z-scores were compared between the two groups over one year using longitudinal models. Longitudinal relationships between measures of bone health and clinical parameters (age, BMI-Z and FEV₁%-predicted) were examined in individuals with CF.

Results

DXA velocity-Z were higher than normal in females (p < 0.05) but not males with CF. HV-adjusted and LBMV-HV-adjusted velocity-Z did not differ for LSaBMD or TBLH-BMC.

In males with CF, both HV-adjusted and LBMV-HV-adjusted LSaBMD velocity-Z scores correlated negatively with age (HV rho: −0.35; p = 0.045 and LBMV-HV rho: −0.47; p = 0.0046). In males with CF BMI-Z correlated positively with HV-adjusted LSaBMD velocity-Z (rho: 0.37; p = 0.034), but this relationship did not persist for LBMV-HV (rho: 0.14; p = 0.42). In females with CF, no correlations between LSaBMD velocity-Z scores and age or BMI-Z were found (all p > 0.05). No correlations between LSaBMD velocity-Z scores and FEV₁%-predicted were seen in either sex (all p > 0.12). TBLH-BMC velocity Z-scores were not correlated with clinical parameters in either sex (all p > 0.1).

At baseline, multiple pQCT parameters were lower in CF (p < 0.05). pQCT Z-scores did not differ between baseline and one-year in either CF or reference group. In a longitudinal model comparing pQCT-Z changes in CF and reference, multiple pQCT-Z outcomes remained lower in CF, but the changes in parameters did not differ in CF vs reference (all p > 0.26). Lower pQCT outcomes in CF were largely restricted to males (CF group*female sex interaction beta coefficients > 0). In this combined longitudinal model, of both CF and reference, BMI-Z was positively associated with pQCT-Z parameters(p < 0.001).

Multiple pQCT-Z outcomes positively correlated with both BMI-Z and FEV₁%-predicted in males with CF, and with FEV₁%-predicted in females with CF (p < 0.05). Age was negatively associated with section modulus (p = 0.001) in males and with cortical density-Z in females (p < 0.001).

Conclusions

With improved longevity, bone health in CF is of increasing importance. On average, bone accrual was preserved in youth with CF, and while deficits in bone geometry and strength were found, these deficits did not worsen over the one-year study. Lower LS bone accrual with increasing age suggests emerging adulthood is a period of vulnerability in CF while the role of LBM in bone health is underscored by the lack of relationship between LBMV-adjusted accrual and BMI. These findings may be useful in targeting screening practices and interventions.