Areal and Volumetric Bone Mineral Density Changes After Renal Transplantation in Children: A Longitudinal Study

Areal and Volumetric Bone Mineral Density Changes After Renal Transplantation in Children: A Longitudinal Study

Areal and Volumetric Bone Mineral Density Changes After Renal Transplantation in Children: A Longitudinal Study L.A. Hamiwka, M. Hanna, J.P. Midgley, ...

73KB Sizes 0 Downloads 15 Views

Areal and Volumetric Bone Mineral Density Changes After Renal Transplantation in Children: A Longitudinal Study L.A. Hamiwka, M. Hanna, J.P. Midgley, A.W. Wade, and S. Grisaru ABSTRACT Introduction. The effect of renal transplantation on areal bone mineral density (aBMD) in children has previously been studied. However, most previous reports did not include estimation of volumetric bone mineral density (vBMD) or analyze longitudinal data in these patients. In addition, updated reference standards for aBMD in children have recently been made available. Methods. This retrospective study describes the longitudinal effect of renal transplantation on aBMD and vBMD in a cohort of 40 pediatric kidney transplant recipients. Lumbar spine aBMD measurements were obtained using dual-energy X-ray absorptiometry prior to transplant and yearly thereafter. vBMD values and z-scores were estimated as described in the most recently published references. Results. A significant decrease in average aBMD and vBMD z-scores was observed within 1 year posttransplant, which did not recover during follow-up. The negative effect of transplantation on vBMD was blunted and vBMD z-scores were higher compared to aBMD. Linear mixed-effects model analysis demonstrated that lumbar spine aBMD and vBMD z-scores were inversely related to yearly prednisone dose (g/m2) but this effect was diminished as glomerular filtration rate was increased. Conclusions. Bone mineral density was negatively affected by renal transplantation in this cohort of pediatric patients. Estimation of vBMD appears to be appropriate for interpretation of the BMD changes occurring after renal transplant in children. The inverse relation between BMD z-scores and yearly prednisone dose suggests that ongoing posttransplant corticosteroid therapy may be responsible for the negative effect of transplantation on bone mineral density in this cohort.


ORTICOSTEROIDS CONTINUE to be widely used as part of immunosuppression protocols despite their known negative effect on bone health. Induction immunosuppression for renal transplantation frequently includes high-dose steroid therapy, followed by continuous longterm maintenance doses. Children who undergo transplantation are likely to have already suffered from bone disease as part of chronic renal failure and growth delay.1,2 Previous studies have reported mostly on posttransplant changes in areal bone mineral density (aBMD) z-scores related to an age-matched standardized control population.1–3 However, aBMD has been shown to be less accurate in pediatric renal transplant recipients who frequently have short stature and delayed bone age.1 Estimation of volumetric bone mineral density (vBMD), known also as bone mineral apparent density, accounts for bone mass and

geometry correlated to height and is thus considered a better measure of bone density in these patients. In healthy children, aBMD increases with age until a peak aBMD is achieved in early adulthood. vBMD is more constant across ages, although it is still directly related to age during childhood.1 In children, the best way to interpret bone mineral density values is by determination of a z-score related to an age-matched reference population. Updated reference

From the Department of Pediatrics, University of Calgary, Calgary, Alberta, Canada. Address reprint requests to Silviu Grisaru, MD, Pediatric Nephrologist, Alberta Children’s Hospital University of Calgary, 2888 Shaganappi Trail, NW, Calgary AB, Canada.

0041-1345/08/$–see front matter doi:10.1016/j.transproceed.2008.03.098

© 2008 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710


Transplantation Proceedings, 40, 1404 –1406 (2008)



bone mineral density values of a large healthy pediatric population have recently been published.4 No studies have yet reported vBMD longitudinally or used the new reference standards in a cohort of pediatric renal transplant recipients. We herein report the results of a longitudinal follow-up of aBMD and vBMD in a cohort of pediatric renal transplant recipients. METHODS At our center, all pediatric renal transplant recipients undergo dual-energy X-ray absorptiometry (DXA) studies prior to transplantation and yearly thereafter as part of routine monitoring. Data for this descriptive, retrospective chart review were obtained from 40 pediatric kidney recipients followed in the nephrology clinic at the Alberta Children’s Hospital. Children enrolled were identified through the nephrology clinic database, and data were obtained from their clinic charts or health records after consent was obtained. aBMD was determined by DXA using a Hologic QDR 4500W scanner. Results of lumbar spine (LS) aBMD expressed as grams per square centimeter were included in this analysis. Lumbar spine vBMD was calculated using the following formulas: vBMDLS ⫽ spine BMC ⫹ spine bone area1.5, where BMC ⫽ bone mineral content.5 aBMD and vBMD z-scores were determined as described in the most recent reference data publication.4 The following additional data were collected during the chart review: age; type of renal disease; dialysis type; date and type of transplant (living related vs deceased donor); immunosuppressive medications (type and doses) posttransplant; yearly measurements of height, weight, body mass index; bone age; rejection episodes; and kidney function (glomerular filtration rate [GFR]). Collected data were entered into an SPSS 14.0 database (Release 14.0.1; SPSS Inc, Chicago, Ill, USA). Data were presented as a mean ⫾ SE. Student t tests or Mann-Whitney rank sum test were used to compare pretransplant average z-scores with average z-scores at years 1 to 5 posttransplant. Correlation coefficients were obtained using Spearman’s method. A linear mixed-effects model was performed to determine the relationship between aBMD or vBMD z-scores and the demographic and clinical variables collected from the patients’ charts. P values less than .05 were considered significant.


Forty renal transplant patients who had 197 DXA studies from 1999 to 2007 were included. Average age at transplant was 11.4 ⫾ 0.7 years, male/female ratio 19/21, LRD/DD ratio 17/23, and average follow-up duration was 4.5 ⫾ 0.3 years. Primary renal disease in this cohort was as follows: 18 congenital malformations, 9 nephronophthisis, 2 FSGS, 2 ARPKD, 1 SLE, 1 ANCA-associated vasculitis, 1 immunoglobulin A nephropathy, and 6 other. All patients were treated with an immunosuppression protocol that included corticosteroids. During the study period, average height SDS z-score increased from ⫺1.4 ⫾ 0.2 (21 ⫾ 4 percentile) at the time of transplant to ⫺0.4 ⫾ 0.3 (37 ⫾ 6 percentile) at last follow-up. Average bone age progressed from 10.5 ⫾ 1.1 to 14.0 ⫾ 0.7 years and average GFR declined from 77.4 ⫾ 4.6 to 72.4 ⫾ 4.3 mL/min/1.73 m2.

Fig 1. Average aBMD and vBMD z-scores prior to and after renal transplantation in a cohort of pediatric patients.

Average pretransplant lumbar spine aBMD z-score for this cohort was 0.201 ⫾ 0.35 and average vBMD z-score was 0.528 ⫾ 0.24. Figure 1 shows longitudinal values of average aBMD and vBMD lumbar spine z-scores up to 5 years posttransplant. One year posttransplant, both aBMD and vBMD average z-scores were lower by 1.15 and 0.74, respectively. The drop in lumbar spine aBMD and vBMD average z-scores was found to be statistically significant (P ⬍ .05) and remained so up to 5 years posttransplant. A linear mixed-effects model was applied to detect correlation between posttransplant aBMD and all other patient parameters that were collected as detailed in the Methods section. No correlations were found except the following equations describing an inverse proportional relationship between lumbar spine aBMD or vBMD z-scores and steroid dose. Lumbar spine aBMD z-score ⫽ ⫺0.94 ⫺ 0.13 · steroid dose (g/m2) ⫺ 0.01 (GFR) ⫹ 0.002 (steroid dose · GFR). Lumbar spine vBMD z-score ⫽ [⫺1.67 ⫺ 0.25 · steroid dose (g/m2) ⫺ 0.01 (GFR) ⫹ 0.003 (steroid dose · GFR)]. These equations indicate that an additional interaction exists between steroid dose and GFR. Lumbar spine aBMD and vBMD z-scores are inversely porportional to annual steroid dose (g/m2); however, the interaction with the GFR suggests that better kidney function is associated with a diminished effect of steroid dose on bone mass. DISCUSSION

Renal transplantation is the treatment of choice for children with end-stage renal disease. Prior to transplantation, many children suffer from delayed growth and bone age.1,3,6 Catch-up growth has been shown to occur posttransplant,


but there is less evidence to support the same effect on aBMD. One probable reason is the frequent use of steroids as part of immunosuppression protocols. In children, assessment of aBMD is determined best by estimation of z-scores. However, accurate z-score calculation requires reference to appropriate standardized healthy populations. Unfortunately, there has been a lack of large studies defining aBMD in healthy children until recently when Kalkwarf et al published the largest study to date, including 1554 healthy subjects of different ages and from various ethnic backgrounds.4 We were able to use this updated new reference for estimation of aBMD and vBMD z-scores in this study. Other confounding factors related to the normal variability in growth and development during childhood and adolescence create major difficulties for interpreting BMD in children. Changing bone size, height, weight, sexual maturation, and bone age complicate comparisons of patients based solely on age. Estimation of vBMD is an attempt to partially account for patient’s height.7,8 Thus, while major fluctuations may be observed in aBMD, vBMD is more constant over time during childhood.9 Pretransplant, our cohort demonstrated a decreased average height SDS zscore (⫺1.4) and therefore conversion of aBMD to vBMD in this population resulted in higher pretransplant lumbar spine z-scores. Previous reports have demonstrated that in both adults and children, there is loss of aBMD, mostly in the immediate period after transplantation, likely as a result of corticosteroid use.6 – 8 Our data show similar outcomes: average aBMD and vBMD z-scores decreased in the first year after transplant, the decline was statistically significant, and the z-scores remained low compared to baseline up to 5 years posttransplant. However, the posttransplant decline of average vBMD z-score was blunted compared to average aBMD z-score. Since vBMD has been shown to better represent bone mass in growing children, this indicates that the negative effect of transplant on bone mass may be overestimated by aBMD and partially explained by patients’ posttransplant growth catch-up, which was observed in our cohort. Induction and maintenance steroid immunosuppression are most likely responsible for the effect of renal transplant


on BMD in our cohort. This is supported by the application of the linear mixed-effects model to our data, which demonstrated that lumbar spine aBMD and vBMD z-scores were inversely proportional to annual steroid dose (g/m2). The interaction of this relationship with the GFR indicates that better graft function is associated with diminished negative effect of steroids on BMD. With the increased use of steroid-free immunosuppression protocols in children, the negative effect of renal transplant on BMD may be significantly reduced. In summary, using updated reference data, we demonstrated an early decline in BMD z-scores in a cohort of children after renal transplantation, which seems to be an adverse effect of steroid immunosuppression. In addition, we have shown that the effect of renal transplantation on vBMD is less prominent then on aBMD z-scores. Future studies will need to compare these findings to longitudinal BMD data in pediatric renal transplant recipients subjected to newer steroid-free transplant immunosuppression protocols. REFERENCES 1. Göks¸en D, Darcan S, Kara P, et al: Bone mineral density in pediatric and adolescent renal transplant patients: How to evaluate. Pediatr Transplant 9:464, 2005 2. Klaus G, Paschen C, Wüster C, et al: Weight/height-related bone mineral density is not reduced after renal transplantation. Pediatr Nephrol 12:343, 1998 3. Kodras K, Haas M: Effect of kidney transplantation on bone. Eur J Clin Invest 36(suppl 2):63, 2006 4. Kalkwarf HJ, Zemel BS, Gilsanz V, et al: The bone mineral density in childhood study: bone mineral content and density according to age, sex, and race. J Clin Endocrinol Metab 92:2087, 2007 5. Bachrach LK, Hastie T, Wang MC, et al: Bone mineral acquisition in healthy Asian, Hispanic, black, and Caucasian youth: a longitudinal study. J Clin Endocrinol Metab 84:4702, 1999 6. Acott PD, Crocker JF, Wong JA: Decreased bone mineral density in the pediatric renal transplant population. Pediatr Transplant 7:358, 2003 7. Grotz WH, Mundinger FA, Rasenack J, et al: Bone loss after kidney transplantation: a longitudinal study in 115 graft recipients. Nephrol Dial Transplant 10:2096, 1995 8. Feber J, Cochat P, Braillon P: Bone mineral density in children after renal transplantation. Pediatr Nephrol 14:654, 2000 9. Wren T, Gilsanz V: Assessing bone mass in children and adolescents. Curr Osteoporos Rep 4:153, 2006