Pediatric Nephrology

, Volume 23, Issue 12, pp 2215–2220

Bisphosphonates in children with hypercalciuria and reduced bone mineral density

Original Article

DOI: 10.1007/s00467-008-0940-9

Cite this article as:
Freundlich, M. & Alon, U.S. Pediatr Nephrol (2008) 23: 2215. doi:10.1007/s00467-008-0940-9


Previous studies have demonstrated reduced bone mineral density (BMD) and biochemical changes of excessive bone resorption in some patients with idiopathic hypercalciuria (IH). Consequently, bisphosphonates have been successfully employed in research animals and adults with IH and reduced BMD. We evaluated the effect of treatment with bisphosphonates in seven patients ages 10–16 years with persistent IH and reduced BMD. In five children, preceding traditional therapy failed. All children received oral alendronate and one also IV Zoledronic acid for 6–18 (median 9.0, mean 10.7) months. With treatment, BMD Z scores in the lumbar spine improved from −2.0 ± 0.3 to −0.8 ± 0.8 (p = 0.002) and in the femoral neck from −1.8 ± 0.4 to −0.7 ± 0.9 (p = 0.01); urine N-telopeptides/creatinine decreased from 372 ± 289 to 72 ± 39 nmol/mmol (p = 0.05) and calcium/creatinine from 0.29 ± 0.12 to 0.13 ± 0.06 mg/mg (p = 0.009). Height Z scores, normal at baseline in all, remained unaffected, and no new stones or fractures were documented throughout the treatment period. Serum creatinine, electrolytes, calcium, phosphorus and parathyroid hormone remained normal as well. In summary, in children with IH and decreased BMD, treatment with bisphosphonates normalized urine calcium excretion, eliminated urinary symptoms, and significantly improved reduced BMD. These short-term beneficial effects indicate the need for larger prospective studies on the potential of bisphosphonates to serve as a new tool in treating children with IH and reduced BMD.


Bisphosphonates Bone mineral density Bone resorption Calciuria Osteopenia 


The etiology of idiopathic hypercalciuria (IH) remains unknown in many cases. The consensus, though, is that the IH might be due either to increased calcium (Ca) absorption from the gut, increased losses in the kidney, increased bone resorption, or any combination of the above [1, 2]. In the majority of children with IH, including those with stones, treatment is prescribed without further investigation into the source of the hypercalciuria. Treatment includes high fluid intake, dietary modification, and the use of potassium (K) citrate preparations and thiazide diuretics [3]. Whereas treatment is effective in most patients, a minority of children continue to display persistent hypercalciuria.

In adults, the association of hypercalciuria with osteopenia is well established [4, 5, 6, 7]. In those with severe osteopenia, Jaeger et al. [8] found significantly higher incidence of fractures. In children, Garcia-Nieto et al. [9] found 30% of the 72 hypercalciuric children investigated to have decreased bone mineral density (BMD). The association of hypercalciuria or hypercalciuric stone disease with osteopenia in children was confirmed by Freundlich et al. [10], Penido et al. [11], and Schwaderer et al. [12], who reported, respectively, that 38%, 37%, and 47% of hypercalciuric children studied had reduced BMD. Skalova et al. [13] found an inverse correlation between urine Ca and BMD in children with IH. Reduced BMD in children and adolescents is a risk factor for diminished peak bone mass, for life-long low BMD, and eventual bone fractures [14].

Assuming that at least in some patients with IH and decreased BMD the hypercalciuria is due to excessive bone resorption, treatment with bisphosphonates, which act to inhibit bone resorption, have resulted in a significant decrease in urine Ca excretion and improved BMD in experimental animals and adults [4, 15, 16, 17]. In our report, we describe our experience with this novel therapy in a group of children and adolescents with IH and reduced BMD.


We collected retrospective data of all cases followed in our clinics with IH and reduced BMD that received treatment with bisphosphonates. Data collected included age, gender, symptoms and signs at presentation, and history of previous anticalciuric treatment. Height and weight, BMD Z scores of the spine and femoral neck (see below), blood work for creatinine (Cr), Ca, phosphorous (P), and urine for Ca/Cr and N-telopeptides (NTX)/Cr ratios at presentation and treatment end. In four children who had 24-h urine collection, the results were converted to Ca/Cr ratio, employed in the other three, for the sake of consistency and ability to conduct meaningful statistical analysis.

BMD was determined using a Lunar Prodigy dual-energy X-ray absorptiometry (DEXA) scanner (Madison, WI, USA) with coefficient of variation values of <1%, utilizing age- and gender-based pediatric software, providing Z scores for the lumbar spine (L2–4). Areal BMD measurements (g/cm2) of the femoral neck were reported as Z scores [standard deviation (SD) scores compared with age- and gender-matched control means] based on reference values obtained by the manufacturer in normal female (879) and male (736) individuals 5–19 years of age (Bone Growth in Children. Lunar News, October, 1993). As all subjects were within normal height percentiles for age; no correction for height was made [18]. In adults, osteopenia (low bone mass) is defined as a BMD T score between −1 and −2.5 [19]. In children, the diagnostic thresholds to define osteopenia have not been clearly established [20], and low bone density for chronologic age has been the terminology recommended for a Z score below −2.0 [21].

Details of the seven children included in this report are presented in Table 1. The decision to treat with bisphosphonates and method of administration, dosage (weight-adjusted according to the standard adult dose), and length of treatment were tailored individually for each patient. We also recorded what treatment the patients were given in addition to bisphosphonates and what side effects to the latter, if any, were reported.
Table 1

Demographic, clinical, and radiological data in seven children with hypercalciuria and decreased bone mineral density (BMD)

Patient Number

Age (years)



Imaging (urinary tract ultrasound)

Prior treatment (months)

Treatment with bisphosphonates (months)






T + C (33)*

A (6) **




Fractures (history of a stone)



A (15)







Z + A (9)






T + C (38)

A (18)**






C (6)

A (6)**






T (48)*

A (15)






T (36)*

A (6)**

A alendronate, C citrate preparations, T thiazides ± amiloride preparations, Z Zoledronic acid

* BMD values decreased, ** continued with previous treatment

Statistical analysis included two-tailed Student’s paired t test or regression analysis (Microsoft Office Excel 2003). P ≤ 0.05 was considered statistically significant. Results are expressed as mean ± SD. The study was approved by our institutional review boards.


The salient clinical and radiological data of the seven patients are presented in Table 1. Children’s ages ranged between 10 and 16 years; three were male patients. Five presented with urinary symptoms (stones, hematuria, dysuria, frequency) and the other two (patients 2 and 3) with fractures due to idiopathic juvenile osteoporosis. In one patient presenting with stones (patient 5), history of fractures was elucidated, and diagnosis of osteopenia was made after 6 months of follow-up. In the two patients who presented with fractures, hypercalciuria was detected due to the fact that urine Ca analysis is part of our protocol for osteopenic patients. All five children who presented with hypercalciuria were previously treated with a low sodium, high K, recommended daily allowance of Ca and protein diet, and were given traditional pharmacologic agents consisting of thiazides and K citrate. In three of them (patients 1, 6, and 7), longitudinal bone density studies showed worsening Z scores concomitant with persistent hypercalciuria. In patient 1, over 2.9 years, spinal Z score decreased from −0.9 to −1.7 and that of the femoral neck from −0.5 to −1.7. In patient 6, over 3.8 years, the corresponding numbers were −0.36 to −2.37 and −1.04 to −2.47, and in patient 7, 0.18 to −1.69 and 0.13 to −2.20, respectively over 2.7 years.

The length of treatment with bisphosphonates ranged from 6 to 18 months (median 9.0, mean 10.7). Six children were treated with oral alendronate only. Patient 3 was treated initially with IV Zoledronic acid but was subsequently switched to oral alendronate due to one short-lived episode of hypocalcemic tetany. In four patients (1, 4, 5, 7), previous pharmacologic anticalciuric treatment was continued. No adverse effects were recorded during the treatment with alendronate.

As shown in Fig. 1, following bisphosphonate treatment, reduced BMD measurements at baseline improved significantly at the lumbar spine (−2.0 ± 0.3 to −0.8 ± 0.8, p = 0.002) and femoral neck (−1.8 ± 0.4 to −0.7 ± 0.9, p = 0.01). Figure 2 depicts the decrease in urine Ca/Cr from 0.29 ± 0.12 to 0.13 ± 0.06 mg/mg (p = 0.009), and the decrease in urine NTX/Cr, measured only in five patients due to technical reasons, from 372 ± 289 to 72 ± 39 nmol/mmol (p = 0.05). In the four children who completed 24-h urine collection (confirmed by urine Cr/kg), Ca excretion decreased from greater than to less than 4.0 mg/kg.
Fig. 1

Bone mineral density, expressed as Z score for age and gender, in seven hypercalciuric children at the start and end of treatment with bisphosphonates. a Lumbar spine L24, showing improvement from −2.0 ± 0.3 to −0.8 ± 0.8 (p = 0.002). b Femoral neck, showing an increase from −1.8 ± 0.4 to −0.7 ± 0.9 (p = 0.02)

Fig. 2

Urine chemistries in seven hypercalciuric children with decreased bone mineral density (BMD) at the start and end of treatment with bisphosphonates. a Urine calcium/creatinine showing a decrease from 0.29 ± 0.12 to 0.13 ± 0.06 mg/mg (p = 0.009). b Urine N-telopeptides/creatinine (measured in five subjects) showing a decrease from 372 ± 289 to 72 ± 39 nmol/mmol (p = 0.05)

BMD Z score in the femoral neck trended to be inversely correlated with urine Ca/Cr ratio (r = −0.37, p = 0.06). No such correlation was found between urine Ca/Cr and BMD of the lumbar spine. There was a correlation between urine NTX/Cr and urine Ca/Cr (r = 0.41, p = 0.04), but no correlation was found between urine NTX/Cr and BMD Z scores of either the femoral neck or the lumbar spine.

Height and weight Z scores, normal at baseline in all, remained unaffected throughout the treatment period, as were height and weight velocities. Similarly, serum Cr, electrolytes, Ca, P, and parathyroid hormone (PTH) remained normal. As documented clinically and by imaging studies, no new stones developed, and no patient experienced hypercalciuria-related symptoms during the treatment period. Similarly, no patient developed fractures.


As in other studies, we found that some hypercalciuric children can have decreased BMD (Fig. 1). Skalova et al. [13] reported a negative correlation between 24-h urine Ca and spinal BMD. In our patients, we found a strong and almost significant inverse correlation between urine Ca and BMD of the femoral neck, but no such trend was appreciated between urine Ca and spinal BMD. Interestingly, in adult IH stone formers, Asplin et al. [22] also found a strong inverse correlation between urine Ca and femoral BMD but only a marginal one with spinal BMD. It is possible that with a larger sample size, we would have achieved similar statistical significance. We also found a correlation between urine NTX/Cr and urine Ca/Cr, namely, with the decrease in bone resorption, urine Ca decreased. In any event, none of the previous studies or our own findings answer the question as to whether hypercalciuria is the cause or the consequence of osteopenia or whether they have a common etiology [1, 5, 6]. An example of the complexity of the interaction among absorptive, renal, and resorptive hypercalciuria is demonstrated by the recent study by Heller et al. [16], who showed evidence of increased bone resorption by histomorphometric analysis of iliac crest biopsies in adults with absorptive hypercalciuria and negative Ca balance. In children, Stapleton et al. [23], using single-photon absorptiometry, found bone density to be similar in children with renal and absorptive hypercalciuria, but this methodology is not as precise as DEXA employed in our study.

Irrespective of the precise mechanisms resulting in the association of hypercalciuria and osteopenia, it seems to respond well to antiresorptive treatment. Weisinger et al. [4, 24] were the first to report a significant reduction in the rate of hypercalciuria in a large group of adult stone formers with idiopathic hypercalciuria treated with alendronate. Bushinsky et al. [15] showed a decrease in urine Ca in response to administration of alendronate in genetically hypercalciuric rats. In the aforementioned study by Heller et al. [16], short-term treatment with alendronate not only normalized urine Ca excretion but also converted the negative Ca balance to a positive one. Recently, in 25 adults with recurrent Ca lithiasis associated with bone loss, Arrabal et al. [17] initiated alendronate treatment; 76% achieved remission and the other 24% reduction in litogenesis, and all showed an increase in bone mineralization. The findings in our study of significant reduction in urine Ca excretion (Fig. 2a) and in hypercalciuria-related morbidity, as well as the concomitant improvement in BMD (Fig. 1), concur with the above studies.

In hypercalciuric osteopenic adults, both thiazide diuretics and K citrate were previously demonstrated to be effective in simultaneously reversing hypercalciuria and improving reduced BMD [25, 26, 27, 28]. Reusz et al. [29] studied 18 children with IH and showed the beneficial effect of thiazide therapy on BMD. This observation could not be reproduced in nonambulatory children [30]. Our five patients who presented with hypercalciuria (Table 1, patients 1, 4–7) were previously treated with appropriate diet, thiazide diuretics, and K citrate. However, they continued to be hypercalciuric, and in the three patients observed longitudinally, BMD values actually decreased (patients 1, 6, 7). Although reduced BMD may not necessarily always indicate accelerated bone loss and could represent failure to gain the expected BMD, the continuous reduction of BMD Z scores in the patients studied longitudinally prior to the institution of bisphosphonate therapy and the elevated urine NTX/Cr strongly suggest accelerated bone resorption. Indeed, under this assumption, whether presented by bone-related or urinary-tract-related symptoms, we addressed all seven patients as having IH due to enhanced bone resorption.

The reason for the lack of response to traditional treatment is unclear. Although there are no good tools to measure compliance to treatment with thiazide diuretics and K citrate, these patients did respond well to treatment with bisphosphonates. Monitoring compliance with the latter can be inferred by measuring urine NTX, a surrogate noninvasive diagnostic tool of bone resorption [18], which, as expected, decreased consistently in all five tested patients (Fig. 2b). Furthermore, oral alendronate has the advantage of once-weekly administration, thus potentially improving compliance [31].

Although not completely free from potential adverse effects, the long-term safety records of thiazides and K citrate are excellent [28, 32]. In contrast, the long-term safety of bisphosphonates and in particular the effect of their long-term accumulation in the growing bone remains unknown [33, 34, 35, 36]. We therefore recommend that until they establish their efficacy and safety, bisphosphonates should be reserved for use only after nonpharmacological and other pharmacological interventions prove unsuccessful. Furthermore, bisphosphonates should be prescribed only by experts in the field and for the shortest treatment period possible.

The major weaknesses of our study are the small number of patients and its retrospective nature. As in the study by Arrabal et al. [17], some of our patients continued to receive anticalciuric agents while receiving bisphosphonates. Nevertheless, other studies showed, as was the case in some of our other patients, that bisphosphonates alone can achieve similar results [1, 16, 17, 24].

Finally, although the focus of this report is on the beneficial effect of bisphosphonate treatment on stubborn hypercalciuria, one cannot overlook its salutary effect on bone. Without its correction, reduced BMD, as observed in our patients, can become a risk factor for diminished peak bone mass, for life-long decreased BMD, and eventually fragility fractures and height loss [14].

We conclude that in a select number of children with hypercalciuria and decreased BMD, short-term treatment with bisphosphonates is effective in correcting both hypercalciuria and bone mineral status. Additional large-scale studies are required to establish the long-term safety of bisphosphonates and the indications for their use.


This work was supported in part by The Sam and Helen Kaplan Research Fund in Pediatric Nephrology and The Eric McClure Research Fund in Pediatric Bone and Mineral Diseases. We thank Regina Johnson, Connie Haney, RN, and Michal Alon, RN, for their valuable assistance.

Copyright information

© IPNA 2008

Authors and Affiliations

  1. 1.Pediatric NephrologyUniversity of MiamiMiamiUSA
  2. 2.Bone and Mineral Disorders Clinic, Section of Pediatric Nephrology, Children’s Mercy Hospital and ClinicsUniversity of Missouri–Kansas City, School of MedicineKansas CityUSA
  3. 3.Pediatric Nephrology, Childrens Mercy HospitalKansas CityUSA

Personalised recommendations