Bone mineral density in pediatric heart transplanted patients: A retrospective single‐center study at Skåne University Hospital in Lund 1988–2016

Impaired bone mineral density (BMD) and osteoporosis are commonly found in patients who have undergone heart transplantation (HT), which increases the risk for bone fractures which is associated with increased morbidity and mortality in adults. However, the long‐term evolution of BMD after HT in pediatric patients has not been thoroughly investigated.


| INTRODUC TI ON
Patients who undergo heart transplantation (HT) have increased risk for impaired bone mineral density (BMD) and osteoporosis. 1 The pathophysiology behind impaired BMD in HT patients is multifactorial and includes side effects of the immunosuppressive regimen, given to prevent allograft rejection. 2 Out of the immunosuppressive agents, corticosteroids (CS) have the greatest negative impact on BMD, but calcineurin inhibitors, such as tacrolimus, also impair BMD. [3][4][5][6] Osteoporosis furthermore increases the risk of fracture-related morbidity and mortality in adults. 7,8 However, in children and adolescents, the association between BMD and fractures is not clear. Thus, it is not possible to diagnose osteoporosis based on BMD alone. 9 In pediatric patients, the BMD is related to an age-, race-, and gender-matched control, resulting in a Z score, which may indicate low BMD for chronological age. 10 Previous studies have shown that pediatric solid organ recipients have impaired bone health post-operatively, and that patients who underwent HT in adolescence have increased prevalence of osteoporosis in adulthood, mainly due to mild renal impairment, secondary hyperparathyroidism, and altered bone turnover. [11][12][13] Pediatric HT has also been found to be a significant risk factor for scoliosis. 14 However, the long-term evolution of BMD after HT in pediatric patients has not been thoroughly investigated. Our objective was therefore to investigate the BMD as evaluated by the Z score before and up to 10 years after HT in pediatric patients, as well as to investigate the BMD in adults who underwent HT during childhood.

| Study design and patient selection
This retrospective cohort study was conducted at Skåne University Hospital in Lund, Sweden, as part of establishing Lund Heart Transplantation Research Register (LHTRR). Included subjects underwent HT between January 1988 and June 2016, at an age younger than 20 years, and were followed at Skåne University Hospital in Lund until 31st of December 2019. Patients who underwent re-HT were excluded. Part of the patient cohort has previously been described in a separate report on acute cellular rejection (ACR). 15 The

| Data collection
Data were collected from clinical records from the transplant assessment (TA) before HT and annually up to 10 years after HT. BMD was obtained using Dual-energy X-ray absorptiometry (DXA) at the lumbar spine and whole body before HT and up to 10 years after HT. The term "low BMD for chronological age" was defined as a Z score of at least 2 standard deviations (SD) below the age-, race-, and gender-matched control, according to the official positions of the International Society for Clinical Densitometry. 16 In addition to BMD, data on age, gender, body mass index (BMI), maintenance immunosuppressive treatment, time on waiting list, urine albumin/ creatinine ratio, frequency and grade of ACR, osteoporosis preventive treatment, as well as serum levels of creatinine, urea, calcium, alkaline phosphatase (ALP), and parathyroid hormone (PTH) were collected. Glomerular filtration rate (GFR) was measured through plasma clearance of iohexol.
To investigate the BMD in adults who underwent HT at an age younger than 20 years, a comparison was performed between patients who had undergone HT at an age younger than 20 years and patients who underwent HT at an age of at least 20 years. In this analysis, BMD was obtained through DXA at the lumbar spine and femoral neck at the 10 th annual check-up, using T score. The adult patient cohort has previously been described in a separate report on HT and chronic kidney disease. 17 Osteoporosis was defined as a T score of ≥−2.5 SD from normal BMD of young and healthy adults, in accordance with the recommendation from the World Health Organization. 18

| Statistical analysis
Statistical analyses were performed using IBM SPSS for Windows (version 26.0, IBM Corp). All statistical analyses were two-tailed with a 5% level of significance. The median and interquartile range (IQR) were calculated for continuous variables.
One-sample Wilcoxon signed rank tests were performed in order to investigate whether the observed medians of Z scores Conclusions: Patients who undergo HT at an age of <20 years have low BMD for chronological age already before HT, but BMD may recover completely within the first 4 years after HT. The results indicate no difference in BMD at 10 years after HT between pediatric and adult patients.

K E Y W O R D S
bone mineral density, heart transplantation, immunosuppression, pediatric differed significantly from the limit which define low BMD for chronologic age.
In order to compare the median T score at 10 years post-HT in patients who had undergone HT at an age younger than 20 years versus patients who were transplanted at an age of at least 20 years, independent samples Mann-Whitney U-tests were performed.
In an attempt to control for missing data at TA, a binary logistic regression was implemented to investigate whether primary indication for HT, urgency of HT, and need of mechanical support were predictors for missing BMD data before HT. Baseline characteristics, with stratification based on age group, are displayed in Table 1.

| Study population
Dual-energy X-ray absorptiometry was performed once pre-HT and annually after HT according to the local routine. However, BMD was missing for the younger patients since DXA was only performed in patients with an age of at least 10 years. However, as the cohort aged during the follow-up, BMD measurements were subsequently more frequent. Treatment with vitamin D or calcium supplementation, or antiresorptive therapies, was given to patients with abnormal DXA results after discussion with an endocrinologist, according to the local routine.

| Bone mineral density evolution
At baseline (TA), a striking 86% had low BMD for chronologic age in the lumbar spine, whereas only 13% were affected in the whole-body measurements. At TA, 100% of patients aged 15-20 years at the HT had low BMD for chronologic age in the lumbar spine, but 100% of the same age group had normal BMD in the whole-body measurements.
The median Z score in the lumbar spine at the TA, and at 1, 5, and 10 years after HT was −2.9 SD (−3. The median T score in the lumbar spine and femoral neck 10 years after HT did not differ between the two groups based on age at the time of HT (p = .779 in the lumbar spine and p = .388 in the femoral neck, respectively).

| Kidney function and serum calcium, ALP, and PTH
Kidney function, measured as GFR, as well as serum concentrations of calcium, ALP, and PTH from TA up to 10 years after HT are displayed in Figure 3. GFR, as well as serum concentrations of calcium, ALP, and PTH remained relatively stable throughout the follow-up.
None of the patients suffered from end-stage renal disease at any time during follow-up.

| Supplementation and antiresorptive therapy
The number of patients who received vitamin D or calcium supplementation, or bisphosphonates, is displayed in Table 3.

| DISCUSS ION
It is well known that osteoporosis frequently affects adult HT patients which increases morbidity and mortality, but BMD and bone fracture risk in pediatric HT patients remain to be thoroughly investigated. 19 This single-center retrospective cohort study aimed to describe the BMD evolution in pediatric patients up to 10 years after HT, as well as to analyze the BMD in adults who underwent HT during childhood.
The results of the present study suggest that patients who underwent HT at an age of <20 years had significantly reduced BMD in the lumbar spine compared with the limit for low BMD for chronological age of −2.0 SD at the TA before HT, Figure 1A   Patients aged 15-20 years at the time of the HT were more likely to suffer from low BMD for chronological age in the lumbar spine than in the whole-body measurements at baseline. This may be a reflection of the heart failure preceding HT and its different impact on trabecular and cortical bone. In a cross-sectional study, using quantitative CT to assess bone density in young healthy girls, it was reported that trabecular bone may be influenced by sexual development to a greater extent than cortical bone, which may be more correlated to weight-bearing or mechanical stress. 24 In another cross-sectional multicenter study which included subjects with Fontan patients, both boys (58%) and girls (58%) had a delay of 1.5-2 years in ≥1 Tanner stage parameter compared with a control population. 25 Although the mechanisms behind BMD regulation in growing subjects in relation to puberty are complex, the findings may suggest that trabecular bone, which is more abundantly found in the lumbar spine, may be more vulnerable to hormonal changes in relation to heart failure per se.
The main limitation of the present study was the limited amount of included patients which was further limited by the fact that a substantial amount of DXA measurements were missing at baseline. Since bone densitometry could have been down-prioritized in more urgent cases of HT and therefore missing in these cases, it was hypothesized that this may bias the results. However, in a regression analysis, no evidence was found on correlations between No difference was found between BMD in patients who had undergone HT at an age of <20 years and patients who had undergone HT at an age of ≥20 years, Figure 3. Also, as previously described, BMD normalized at the fourth annual check-up in the latter group, Bone health and renal disease are known to be closely related. 26,27 In a previously published study on an adult cohort from the same center, the issue of chronic kidney disease in relation to DXA, particularly in the lumbar spine, was discussed. 28 The DXA results were found to be a potential result of vascular disease and aortic calcification which is associated with renal disease. 29,30 In the F I G U R E 2 A comparison of BMD expressed as T score in the (A) lumbar spine and (B) femoral neck 10 years after HT in patients who underwent HT at an age younger than 20 years versus patients who underwent HT at an age of at least 20 years (p = .779 in the lumbar spine and p = .388 in the femoral neck). The dashed line represents the limit under which osteoporosis can be diagnosed (−2.5 SD and lower). * Outlier. BMD, bone mineral density; HT, heart transplantation HT at age 20 years or older HT at age < 20 years Restrictive cardiomyopathy 2 7 Total 30 100 Abbreviations: ARVD, arrhythmogenic right ventricular dysplasia; HT, heart transplantation.
present study, GFR remained stable throughout the follow-up and none of the included subjects suffered from end-stage renal disease at any time, suggesting that kidney function did not significantly influence the results.
In conclusion, the present study showed that patients who underwent HT at an age of <20 years had low BMD for chronological age in the lumbar spine already before HT. BMD then recovered at the fourth-year after HT in both lumbar spine and whole-body measurements. No difference in BMD at 10 years after HT was found between patients who underwent HT at an age of <20 years versus patients who underwent HT at an age of ≥20 years. The results suggest that pediatric HT patients may recover completely in measures of skeletal health within the first years after HT. For the future, it would be desirable to investigate the skeletal health using advanced imaging techniques, such as quantitative magnetic resonance, and measurement of sex hormones, using a prospective study design.

ACK N OWLED G EM ENTS
We acknowledge the support of the staff at the Section for Heart Heart Center, Skåne University hospital, Lund, Sweden.

CO N FLI C T O F I NTE R E S T
None.

AUTH O R CO NTR I B UTI O N S
EL involved in study design, data collection, data analysis, and writing of the article. KTL involved in study design, data acquisition, F I G U R E 3 Boxplot of (A) GFR measured by iohexol clearance, serum levels of (B) S-Calcium, (C) S-ALP, and (D) S-PTH up to 10 years after HT. ALP, alkaline phosphatase; GFR, glomerular filtration rate; HT, heart transplantation; PTH, parathyroid hormone; TA, transplant assessment