Timing of low bone mineral density and predictors of bone mineral density trajectory in children on long-term warfarin: a longitudinal study.

UNLABELLED

We studied bone mineral density (BMD) of children exposed to long-term warfarin. BMD Z-scores ≤ -2.0 were estimated to occur in less than one fifth of the patients after 10 years of warfarin exposure, and BMI and growth hormone deficiency predicted BMD changes over time. These predictors can help identify high-risk patients.

INTRODUCTION

Children with chronic diseases are at increased risk of developing thrombosis, which may require long-term warfarin therapy. Warfarin could further jeopardize the bone health of a population already at risk for bone fragility. Our objective was to investigate the occurrence and timing of low bone mineral density (BMD) and the predictors that influence BMD trajectory in children receiving warfarin for >1 year.

METHODS

We analyzed the results of an institutional protocol that includes dual-energy X-ray absorptiometry, with or without spinal X-rays and laboratory biomarkers, as required.

RESULTS

Low BMD (age, sex, race, and height-for-age-Z-score adjusted BMD Z-score ≤ -2.0) was detected in 13 % (9/70) of the patients at some point during their follow-up; these patients were more likely to have complex underlying medical conditions and low body mass index (BMI) percentile. BMD Z-scores remained within normal range in 87 % of children. Survival analysis showed that the estimated 10-year abnormal BMD-free rate for the entire group was 81 % (95 % confidence interval [CI] 69 to 93 %). Trajectory analysis revealed that BMI percentiles at baseline and growth hormone deficiency (GHD) were associated with lower BMD Z-scores at the first assessment, whereas baseline BMI percentile was the only predictor of BMD Z-score over time.

CONCLUSIONS

Our findings identified BMI and GHD as risk factors influencing BMD in children exposed to long-term warfarin, creating an opportunity for early detection and intervention in these patients.