Introduction
Osteoporosis, defined by a bone mineral density (BMD) 2.5 standard deviations below the young adult reference mean (T score) is well established as a major risk factor for fractures in postmenopausal women and men over 50 years of age [1]. In contrast, low bone density in premenopausal women and men under 50 is defined by a Z score 2 standard deviations below age-matched norms and is referred to as “bone density below the expected range for age” [1]. This condition is typically secondary to underlying causes.
However, there is currently no consensus on the management of low BMD in younger individuals.
Osteoporosis is increasingly recognized as a complication in transplant recipients [2], yet data on its prevalence, management, and clinical consequences, particularly in younger patients, remain limited. Reported post-lung transplantation fracture rates range from 5% to 37% within the first year and up to 53% within 5 years [3–5]. Bone loss in this population is influenced by a combination of pretransplant factors, such as malnutrition, physical inactivity, underlying diseases like cystic fibrosis (CF)-related bone disease, and most notably chronic glucocorticoid therapy, both pre- and posttransplant [2, 6, 7].
Glucocorticoid-induced osteoporosis (GIO), the most common form of secondary osteoporosis, significantly elevates fracture risk, especially within the first 3–6 months of treatment [8, 9]. Vertebral fractures are most frequent early on, while non-vertebral fractures, including hip fractures, are also prevalent and correlate with steroid dose and duration.
Evidence-based guidance on managing bone health in young lung transplant recipients (LTRs) remains scarce. Sithamparanathan et al. [10] followed 21 long-term survivors (mean transplant age 31.8 ± 9.9 years). After 20 years, 10 of the 21 developed osteoporosis, and one experienced an osteoporotic fracture.
Wang et al. [11] reported pretransplant osteopenia and osteoporosis rates of 36% and 31%, respectively, among recipients aged 46.7 ± 12 years, with approximately one-third having sustained prior fractures. Seventy-three percent of patients required bisphosphonate therapy over a 2.5-year follow-up. In another study comparing pamidronate to standard care in young patients with CF, pamidronate improved BMD but did not reduce fracture incidence [12].
Given the limited evidence surrounding the diagnosis, management, and outcomes of low BMD and osteoporotic fractures in young LTRs, this study aimed to evaluate bone health outcomes in a large cohort of this population who were followed at our tertiary transplant center.
Methods
We performed a retrospective analysis of electronic and paper medical records from a pre-defined registry of LTRs followed at the Pulmonary Institute at Rabin Medical Center, the largest lung transplant center in Israel. The local Institutional Review Board approved the study (approval number 0141-21-RMC).
Eligible participants were young adults aged 18–50 at the time of lung transplantation, with follow-up from 2000 until death or data extraction (November 1, 2023). Epidemiologic, anthropometric, clinical history, and laboratory data were collected for each patient within 3 months before (baseline status) and 6 months after transplantation. As the cohort included pre-menopausal women and men under 50 years of age, the BMD assessed by dual-energy X-ray absorptiometry (DXA) scans was evaluated using Z scores.
A Z score < −2 at any site was classified as “lower than expected for age,” while Z scores ≥ −2 across all sites were considered “appropriate for age.” Fracture events were identified from clinical notes and imaging reports (e.g., high-resolution computed tomography, plain radiography). Fractures mentioned in imaging reports were included even if not discussed in subsequent medical records. Imaging studies were not re-evaluated by a radiologist.
Because imaging was performed for clinical indications rather than systematic screening, some asymptomatic fractures may have been missed. While death dates were available, the cause of death was not recorded in the registry.
Statistical Analysis
Statistical analyses were performed using SPSS version 28 (IBM, Chicago, IL) and Microsoft Excel version 2502 (Microsoft, Redmond, WA). Continuous variables with normal distribution are presented as mean and standard deviation, otherwise as median and range. Categorical variables are presented as numbers and percentages.
T test or Mann-Whitney (as appropriate) U test was used to compare continuous variables and chi-square or Fisher’s exact test to compare categorical variables. Variables significantly associated with posttransplant fractures in univariable analyses (p < 0.05) were included in a multivariate Cox regression model, together with age and sex as predefined covariates.
Results
Baseline Characteristics
A total of 150 young transplant recipients participated in the study, which included 146 lung transplants and six combined heart-lung transplants; 45% were female. The mean age at transplantation was 38 ± 9 years. The most common indication for surgery was interstitial lung disease (n = 65, 45%), consistent with its status as the leading indication for lung transplantation overall.
CF also appeared frequently as an indication, reflecting its greater prevalence among younger recipients. In contrast, chronic obstructive pulmonary disease and emphysema were less common in this age-group. The median follow-up period was 6.2 years (interquartile range, 0.0–11.8).
All patients received a standard immunosuppressive regimen that included a calcineurin inhibitor and glucocorticoids. Mycophenolate mofetil was also routinely prescribed; however, some patients could not tolerate mycophenolate mofetil, most commonly due to leukopenia.
Clinical Characteristics
The baseline, pretransplant characteristics of the study cohort are detailed in Table 1. Patients were normocalcemic and normophosphatemic. However, vitamin D deficiency was detected with an average level of 35 ± 22 nmol/L.
Diabetes mellitus (CF-related diabetes, n = 18; type 2 diabetes mellitus, n = 13) was noticed in 20% pretransplant, while other metabolic syndrome-associated diseases were uncommon. None had celiac disease or thyrotoxicosis. The study collection of data did not capture the development of posttransplantation diabetes mellitus, hypertension, or dyslipidemia.
DXA Scans
Thirty-two patients (21%) underwent DXA scans before the transplant surgery (shown in Fig. 1). Details of DXA charts were available in 14 cases; 8 out of 14 patients (57%) had baseline bone density lower than expected for age (lumbar spine, n = 6; femoral neck, n = 5; total hip, n = 5 patients).
Fractures
Ten patients (7%) had fractures prior to transplantation (shown in online suppl. Table S1; for all online suppl. material, see https://doi.org/10.1159/000550456); 3 patients experienced a single fracture, which was at the femoral neck in all cases. None had received osteoporotic treatment before transplant.
Seven patients experienced multiple fractures; only two of them received osteoporosis therapy prior to the transplant surgery.
Osteoporosis Therapy
Overall, 12 patients (8%) were treated for osteoporosis prior to transplantation (two who had experienced a fracture and 10 who had not). All received first-line treatment with oral bisphosphonates. Vitamin D3 supplementation was given to 23% of the cohort and calcium supplementation to 9% of the cohort at baseline.
None of the patients received testosterone medications before the transplantation. Only 1 patient received oral contraceptives.
Posttransplant BMD
Timing and Prevalence of DXA Scans after Surgery
Overall, 103 patients (69%) had DXA scans after the transplant surgery (shown in Fig. 1). Among them, 82 (80%) patients had the test during the first year after the surgery and 21 patients over the next 10 years.
All patients assessed for bone density more than 10 years posttransplant had previously undergone at least one DXA scan.
Of the 32 who had bone density assessment before the transplant 23 received a follow-up scan within 3 years after the surgery (shown in Fig. 1) and nine did not (six died within the first year, one died during the second year after the transplantation, and two did not have a scan over 2.4 and 2.7 years of follow-up until the date of the study’s data extraction). Eighty patients of the 118 who did not have bone density assessment before the transplant were assessed with a DXA scan after the surgery (shown in Fig.
1): 63 during the first year after the surgery and the remaining within one to 10 years following transplantation.
Clinical Characteristics according to Posttransplant DXA Evaluation
Patients who underwent DXA were similar to those without DXA across most baseline characteristics (Table 2). Mortality was substantially higher in the no-DXA group (79% vs. 33%, p < 0.001).
Comorbidities, use of vitamin D or calcium supplements, proton-pump inhibitors, and rates of pretransplant fractures or osteoporosis therapy were comparable between groups.
Baseline laboratory values were largely similar as well. Posttransplant fasting glucose was lower among DXA patients (100 ± 35 vs. 124 ± 51 mg/dL, p = 0.007), and alkaline phosphatase levels were consistently lower in the DXA group at both baseline and follow-up (p = 0.037 and p = 0.033, respectively).
Endocrinology follow-up was far more common in the DXA group (74% vs. 15%, p < 0.001). Among those evaluated in endocrinology, 28 patients underwent DXA after the visit, 48 had DXA beforehand, and 7 were followed without a DXA scan (4 died within 1 year of the endocrinology visit, 2 within 2 years, and 1 remained alive through 13 years of follow-up).
DXA Scan Results
At 1 year posttransplantation, 19 of the 82 patients (23%) with bone scan assessment had lower bone density than expected for age (shown in Fig. 2). Nine of these patients initiated anti-osteoporotic therapy; two sustained fractures while on treatment.
Nine of these 19 patients also underwent a follow-up DXA scan 2 years later; in all but 2 patients, the bone density remained lower than expected for age.
One year posttransplant, 20 of 63 patients with bone density appropriate for age received anti-osteoporosis therapy (shown in Fig. 2), likely due to other risk factors (e.g., prior fracture). Eight of them experienced an osteoporotic fracture on treatment.
Seventeen patients had a follow-up DXA scan after 2 years; in four, the bone density decreased to lower than expected for age. Fifteen years posttransplant, 18 of the 83 surviving patients (22%) underwent a DXA scan, with three showing values lower than expected for age.
Posttransplant Fractures
Thirty-four out of the 150 patients (23%) experienced a nontraumatic fracture after transplantation. The initial fracture site was vertebral in 10 (29%), femoral neck in 11 (32%), and other sites in 13 (38%) patients. Eleven of these 34 patients (32%) experienced multiple fractures posttransplantation (shown in Fig.
3a).
Notably, 71 patients died during the study period (47% of the cohort, 35 women). Fifteen of them (21%) had fractures after the transplantation.
Fracture Timing
The timing of the first posttransplant fracture varied; the highest absolute number and rate of fractures occurred during the second to fifth year posttransplant (shown in Fig. 3b).
Vertebral Fractures
Of the 34 patients who experienced posttransplant fractures, 11 (32%) sustained vertebral fractures. As shown in Figure 3c, vertebral fractures occurred as early as the first year following transplantation, with their incidence declining over time.
Out of 11 patients with vertebral fractures, only one received osteoporosis therapy prior to the transplant surgery. Seven patients received treatment posttransplant, including the 1 patient who had started treatment pretransplant. Four out of 11 patients (36%) did not receive any posttransplant treatment, even after sustaining a fracture.
For those treated posttransplant, all 7 patients initiated treatment prior to their fractures; the choice of treatment in all cases was bisphosphonates, specifically alendronate, as the first-line therapy.
Femoral Neck Fractures
Femoral neck fractures occurred in 11 (32%) out of the 34 young LTRs who experienced fractures posttransplantation. The rate of femoral neck fractures increased significantly 2–5 years after the transplant surgery (shown in Fig. 3d).
None of these patients received osteoporosis treatment before their transplant. Posttransplant, 8 out of the 11 patients (73%) received osteoporosis treatment. Of these, five initiated treatment prior to their fracture, 1 patient started treatment 12 years after the fracture, and for 2 patients, the timing of treatment initiation was unknown.
Six of 8 patients (75%) received bisphosphonates (3 alendronate, two risedronate, and one zoledronic acid); two were treated with denosumab.
Fracture in Patients with a History of Pretransplant Fractures
Out of 10 patients with a pretransplant history of fractures, six also sustained fractures posttransplantation; one received treatment only prior to transplantation, one both before and after transplantation, and four received treatments exclusively after transplantation. Notably, 4 of the 5 patients treated posttransplant received treatment before sustaining posttransplant fractures, while the treatment status of the remaining patient is unknown. For the 4 patients with a pretransplant history of fractures but no posttransplant fractures, two received treatment posttransplantation, while the other two died within 1.5 years posttransplant.
Predictors for Fracture
The baseline characteristics of patients who experienced fractures after the transplantation (n = 34, 50% females) compared to patients who did not sustain fractures after the transplantation (n = 116, 43% females) are presented in Table 3. The study groups did not have significant differences in most demographic parameters, transplant indications, medical background, or baseline laboratory findings prior to transplantation and within 6 months following transplant. Alkaline phosphatase was higher at baseline among patients who did not experience posttransplant fracture (87 ± 38 versus 68 ± 21 U/L, p = 0.008) but was similar in both groups when measured 6 months after the transplantation.
Pretransplant fractures were more prevalent among patients who experienced posttransplant fractures (18% versus 3%, p = 0.010). However, rates of osteoporotic therapy were similar in both groups.
Multivariate regression analysis was conducted to explore independent predictors for posttransplant fracture (shown in Table 4). Pretransplant fractures were a strong risk factor (HR 6.720, 95% CI: 1.572–28.724). Alkaline phosphatase was also an independent risk factor as lower baseline values were associated with higher risk for fracture.
Osteoporosis Therapy
Posttransplant osteoporosis treatment was administered to 27% of patients (41/150), with bisphosphonates serving as the primary therapy in most cases (38/41, 93%). The sequence of therapies is summarized as follows:First-line therapy: 38 patients received bisphosphonates (27 alendronate, nine risedronate, two zoledronic acid), and 3 received denosumab. Teriparatide was not used as a first-line agent.Second-line therapy: 9 patients received bisphosphonates (2 risedronate, seven zoledronic acid), 1 received denosumab, and 2 received teriparatide.Third-line therapy: 1 patient was treated with denosumab and 3 with teriparatide.
Among those treated, 21 patients (51%) sustained fractures. Of these, 12 (57%) had initiated osteoporosis therapy prior to the fracture, while 3 (14%) began treatment only after the fracture occurred. Four patients (19%) started therapy within the same calendar year as their fracture, and in two cases (10%), treatment was documented, but the timing relative to the fracture could not be determined.
Endocrinology Visits
A total of 83 patients (55% of the cohort) had at least one endocrinology consultation following transplantation. Among those who sustained a posttransplant fracture, 26 patients (77%) were evaluated by an endocrinologist: 15 were seen before the fracture occurred, while 11 were seen only afterward: 7 within the first year post-fracture, two between 3 and 5 years, and two between 7 and 13 years after the fracture. Specifically focusing on femoral neck fractures, 9 out of 11 patients had an endocrinologist visit, 8 of these occurring before the fracture.
Of all patients evaluated by an endocrinologist, 35 (42%) received osteoporosis treatment. Additionally, 6 patients received treatment without endocrinologist involvement.
Discussion
Posttransplant bone health and related complications are poorly documented among young LTRs, and our study adds significant data to this understudied complication. We found a substantial burden of low BMD and fractures, including a high rate of multiple fractures and a notable incidence of femoral neck fractures. These findings challenge the assumption that younger ages confer protection against bone fragility and highlight a clinical blind spot in posttransplant management.
Although the incidence of posttransplant fractures tends to be lower in younger transplant recipients compared to older individuals [13], the long-term impact on quality of life and fracture risk reduction can be significant [14]. The high incidence of femoral neck fractures in this young population is particularly striking as these fractures are typically associated with older populations and carry significant morbidity and mortality [15, 16]. The occurrence of hip fractures in younger patients suggests accelerated skeletal aging, likely due to chronic illness and immunosuppressive therapy.
Although osteoporosis guidelines tend to focus on older adults, high-risk patients, such as those on long-term glucocorticoids, should also be prioritized for screening regardless of age, as recommended by the International Society for Clinical Densitometry [17].
Current candidate selection guidelines identify osteoporosis as a relevant risk factor for adverse posttransplant outcomes and suggest BMD assessment prior to transplant [18]. In our cohort, only 21% of patients underwent BMD evaluation prior to transplantation, highlighting a gap between recommended practices and real-world implementation. This likely reflects the complexity of caring for patients with end-stage lung disease, many of whom experience significant physical debilitation, chronic hypoxemia, and progressive respiratory failure prior to transplant.
In this context, bone health assessments may be deprioritized in favor of more urgent pretransplant evaluations and stabilization efforts.
Following transplantation, the intensive demands of postoperative monitoring, immunosuppressive management, and the need to address acute complications may further limit patients’ and healthcare providers’ focus on bone health. Notably, among those who were assessed, half were found to have bone density below the age-expected norms. These findings underscore a missed opportunity to optimize skeletal health in a high-risk population and suggest that more integrated, multidisciplinary approaches may be needed to ensure early identification and management of bone fragility in the lung transplant setting.
Previous data had shown that the most pronounced decline in BMD typically occurs within the first 6–12 months following transplantation, which correlates with a substantial increase in fracture risk [7, 13]. Consistently, significant bone loss and number of fractures in our cohort occurred in the first posttransplant year, with vertebral fractures predominating early and non-vertebral fractures, particularly femoral neck, emerging later. This shift likely reflects the effects of chronic steroid use as vertebral fractures are common in the initial months of glucocorticoid exposure, while hip fracture risk rises with cumulative steroid burden [8, 9].
Fracture history emerged as a strong predictor of future fractures, with a hazard ratio of 6.720. This finding is in line with previous studies showing exponential increases in risk with each additional fracture [19]. These data reinforce the need for early fracture recognition and aggressive primary and secondary prevention in LTRs.
Interestingly, low baseline ALP levels were associated with an increased risk of fractures in our cohort; bone turnover markers have inconsistent accuracy in fracture prediction and are not currently recommended for decision on initiation of therapy for osteoporosis or included in prediction tools such as the Fracture Risk Assessment tool (FRAX) [1, 20]. Low ALP has been implicated in specific conditions associated with impaired bone turnover, such as hypophosphatasia [21] and adynamic bone disease [22]. In our setting, reduced ALP may signal steroid-induced suppression of bone turnover before transplant [20], potentially compounded by further bone deterioration following transplantation.
Since we measured total rather than bone-specific ALP, interpretation remains limited, highlighting the need for studies clarifying the contribution of ALP to fracture risk in transplant recipients.
Childbirth history was more frequent among patients who did not sustain posttransplant fractures. This finding may indirectly reflect preserved gonadal function and premenopausal status, both of which exert protective effects on bone metabolism. Conversely, breastfeeding is known to transiently decrease BMD, though this effect is typically reversible.
Given the small number of female participants and incomplete reproductive data, we did not include this parameter in the multivariate analysis, and this observation should be interpreted cautiously, yet it may suggest a role for gonadal function preservation in mitigating bone fragility among young female transplant recipients.
Currently, no osteoporosis treatment guidelines are specifically tailored to young LTRs, and clinical management is often extrapolated from GIO protocols [7]. However, even GIO guidelines do not provide definitive recommendations for young adults, instead advising referral to a bone specialist for individualized assessment and treatment planning [9]. In our cohort, 27% of patients received osteoporosis treatment following transplantation, one-third of patients who experienced posttransplant fractures had not received any osteoporosis medication, and therapy was frequently delayed.
Multiple factors may contribute to these care gaps. Some fractures were incidentally identified on routine CT imaging and may have been unrecognized clinically. The substantial posttransplant care burden, including intensive follow-up, complex immunosuppressive regimens, and frequent medical complications, can lead to patient fatigue and limit both patient and provider capacity to address bone health.
Physician barriers to bisphosphonate prescribing might include fear of adverse events and concerns regarding renal impairment as many transplant recipients develop calcineurin-associated nephropathy. The paucity of fracture-endpoint data in transplant settings might also reduce prescriber confidence. Limited access and referral rates to bone health specialists may further contribute.
Proactive osteoporosis management by a dedicated specialist, such as an endocrinologist, has been shown to improve treatment delivery and may help prevent future fractures [23, 24], yet only 55% of our cohort had endocrinology follow-up after transplantation, and 24% of those who experienced fractures had no endocrinology evaluation at all.
From a programmatic standpoint, these findings highlight the need to better integrate bone health into routine transplant care. Current expert recommendations support baseline DXA and vitamin D assessment before transplantation, followed by close monitoring during the first posttransplant years when bone loss is most pronounced [2]. Yet in our cohort, only a minority of patients underwent pretransplant DXA, indicating a gap between guideline recommendations and real-world practice.
Implementing structured pathways, such as systematic pretransplant DXA in high-risk candidates, early posttransplant surveillance, and automatic endocrinology referral for patients with low bone density or fractures, may help close this gap. Coordinated models similar to fracture liaison services, shown to improve treatment uptake and secondary fracture prevention in other high-risk populations [25], could be adapted within transplant programs to improve skeletal outcomes in young LTRs.
This study has several limitations. Its retrospective design inherently carries risks of incomplete data capture and potential bias. BMD assessments using DXA scans were not available for all patients, and scanning protocols and equipment varied over the study period.
Fractures were captured from existing records and imaging reports without re-reading by an imaging expert. As vertebral fractures are often asymptomatic, under-ascertainment is likely, and fracture prevalence may therefore be underestimated. Key variables known to influence bone health, such as body mass index, hormonal status, and the use of contraceptives, were not consistently documented.
Data collection was limited to the pretransplant period; laboratory values and comorbidities were recorded only within a narrow window prior to transplantation. Consequently, posttransplant developments, such as disturbances in calcium, phosphate, or vitamin D metabolism, or the onset of conditions like diabetes mellitus or hypertension, were not accounted for in the fracture risk analysis.
Some potentially relevant covariates, including body mass index, vitamin D levels, diabetes, cumulative steroid exposure, and underlying diagnosis, were either unavailable for a substantial proportion of patients or showed no significant differences between study groups in univariable analysis. These variables were therefore not included in the multivariate model to avoid overfitting or introducing bias. This may have resulted in residual confounding that should be considered when interpreting the findings.
The study cohort was heterogeneous, encompassing a spectrum of underlying diseases such as interstitial lung disease, CF, and other rare disorders leading to lung transplantation in young adults. This diversity reflects real-world clinical practice but limits disease-specific interpretation as subgroup analyses were not feasible due to sample size constraints. Moreover, these conditions often involve distinct pretransplant therapies, such as chronic corticosteroid or antifibrotic treatment, that may differentially affect bone metabolism and contribute to residual confounding.
Overall, this study reflects the experience of a single national transplant center, which performs over 90% of lung transplantations in the country. While this monocentric and country-specific context may limit generalizability to other healthcare systems, the centralized, long-term follow-up minimizes selection bias and enhances the representativeness of the cohort within the national population.
Despite these limitations, the study was conducted at our country’s largest lung transplant center and provides centralized, long-term follow-up care. This setting likely minimizes selection bias and enhances the generalizability of our findings to similar populations.
To conclude, this study sheds light on a clinically vulnerable yet underexplored group – young LTRs. Using a large, nationally representative cohort with centralized data collection, we detail fracture frequency, timing, and osteoporosis-treatment patterns within the context of chronic immunosuppression and complex posttransplant care. Despite their age, these recipients demonstrate low BMD and fracture profiles that mirror those of much older adults, indicating accelerated skeletal decline.
Even in the era of available therapies, persistently high rates of fractures underscore the need for earlier screening, individualized interventions, and rigorous follow-up. Our findings therefore support proactive, multidisciplinary strategies to optimize bone-health outcomes in this and other high-risk transplant populations.
Statement of Ethics
This study protocol was reviewed and approved by the local Institutional Review Board of Rabin Medical Center, Israel, Approval No. 0141-21-RMC. Because the investigation relied solely on retrospectively collected data, the requirement for individual informed consent was waived by the reviewing committee.
Conflict of Interest Statement
Gloria Tsvetov has received fees from Medison Pharma for educational lectures. Aviva Lerman, Osnat Shtraichman, Yaron Rudman, Mordechai R. Kramer, Idit Dotan, and Talia Diker Cohen have no conflicts of interest to declare.
Funding Sources
This study was not supported by any sponsor or funder.
Author Contributions
Design: A.L., O.S., and T.D.C. Conduct/data collection: A.L., O.S., and T.D.C., Analysis: A.L., O.S., Y.R., M.R.K., I.D., G.T., and T.D.C. Writing manuscript – first draft: A.L. and T.D.C.
Final version: A.L., O.S., Y.R., M.R.K., I.D., G.T., and T.D.C. During the preparation of this work, the authors used ChatGPT to enhance readability and language. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.