REGULAR ARTICLE

Low-dose thiazide diuretics in children with idiopathic renal hypercalciuriaJi Na Choi, Jae Seung Lee, Jae Il Shin (shinji@yuhs.ac)Department of Pediatrics, The Institute of Kidney Disease, Yonsei University College of Medicine, Severance Children’s Hospital, Seoul, Korea

KeywordsChildren, Haematuria, Hydrochlorothiazide,Idiopathic renal hypercalciuria

CorrespondenceJae Il Shin, M.D., Department of Pediatrics, YonseiUniversity College of Medicine, Sungsan-Ro 250,Seodaemun-Ku, 120-752, CPO Box 8044, Seoul,Korea.Tel: +82-2-2228-2050 |Fax. +82-2-393-9118 |Email: shinji@yuhs.ac

Received21 November 2010; revised 14 January 2011;accepted 24 January 2011.

DOI:10.1111/j.1651-2227.2011.02191.x

ABTRACTAim: To evaluate the therapeutic effect of hydrochlorothiazide in idiopathic renal

hypercalciuria.Methods: We retrospectively analysed the data of 28 children (6.0 ± 4.1 years,

M:F = 19:9) diagnosed as having idiopathic renal hypercalciuria from the years 1991 to

2008. The dose of hydrochlorothiazide was initially 0.5 mg ⁄ kg ⁄ day and gradually increased

to achieve the appropriate hypocalciuric effect (urinary calcium ⁄ creatinine <0.2 mg ⁄ mg) in

some unresponsive patients.Results: Twenty-two patients (79%) had gross haematuria, 6 (21%) microscopic

haematuria, 2 left flank pain, 6 (21%) urolithiasis and 9 (32%) urinary tract infection at the

diagnosis of hypercalciuria. The low doses (0.5 mg ⁄ kg ⁄ day) of hydrochlorothiazide

reduced urinary calcium excretion in 25 patients (89%) and 3 (11%) required the

increased doses (1–2 mg ⁄ kg ⁄ day). Haematuria and urolithiasis gradually resolved in

accordance with the improvement of hypercalciuria. Nineteen patients (68%) maintaining

hypocalciuria during hydrochlorothiazide therapy were discontinued after 12.5 ± 5.3

months of treatment. Eleven of the 19 patients maintained normocalciuria, while 8

showed increased urinary calcium excretion at 2.9 ± 2.3 months after treatment was

stopped, requiring thiazide retreatment.Conclusion: Our results suggest that low dose (0.5 mg ⁄ kg ⁄ day) of hydrochlorothia-

zide may be safe and effective in controlling renal hypercalciuria in children.

INTRODUCTIONIdiopathic hypercalciuria (IH) was first described byAlbright and colleagues in 1953, reporting patients withrecurrent urolithiasis who had excessive urinary calciumexcretion without concomitant hypercalcaemia (1). It is themost common metabolic disorder detected in 3–9% of thechildren (2), causing gross or microscopic haematuria, void-ing dysfunction, urinary tract infection (UTI) and calciumurolithiasis (3).

The two most common subtypes of IH are absorptive andrenal (3–5). The prognosis is relatively good in children ifappropriate interventions, such as high fluid, low-salt andoxalate intake, are performed early. However, the develop-ment of gross haematuria or urolithiasis is possible, particu-larly in those patients who are resistant to the conservativemanagement or those with renal hypercalciuria. In thesecases, thiazide therapy can be indicated (3).

Thiazide diuretics have been widely used to treat renalhypercalciuria (3,6,7) and it may also reduce the risk of uro-lithiasis with a family history and persistent gross haematu-ria associated with a markedly increased urinary calciumexcretion (4). However, it still remains controversial howlong should thiazide be continued or what dosage would beappropriate.

The aim of this study was to evaluate the effective doseand appropriate duration of thiazide treatment in

controlling the urinary calcium excretion in children withidiopathic renal hypercalciuria.

MATERIALS AND METHODSSubjectsWe retrospectively reviewed the hospital records from theyears 1991 to 2008 and found 54 children with hypercalci-uria, of which 28 met the criteria for renal hypercalciuria,25 absorptive hypercalciuria, and one resorptive hyper-calciuria. The study population with renal hypercalciuriaincluded 19 boys and 9 girls with a mean age of 6.0 ±4.1 years (range 0.4–12.8 years). Six children presentedwith microscopic haematuria and 22 gross haematuria.Two patients were accompanied by left flank pain, whoshowed a renal stone on renal ultrasound examinations. Ofthe 28 children, 3 children were treated with high-dosehydrochlorothiazide (HCT) and 3 were lost to follow-up,

Key notes• Low dose (0.5 mg ⁄ kg ⁄ day) of hydrochlorothiazide may

be safe and effective in controlling renal hypercalciuriain children.

Acta Pædiatrica ISSN 0803–5253

ª2011 The Author(s)/Acta Pædiatrica ª2011 Foundation Acta Pædiatrica 2011 100, pp. e71–e74 e71

and hence were excluded. We analysed the data of the22 children on whom we have adequate follow-up. Theclinical and laboratory findings of these 22 children withidiopathic renal hypercalciuria are presented in Table S1(Supporting Information).

We excluded those patients who had recent treatmentwith calcium or vitamin D, endocrinopathies, nephro-pathies, other metabolic diseases that could predispose tohypercalcaemia and secondary hypercalciuria by historyand various laboratory examinations.

MethodsThe diagnosis of IH was made by an increased urinarycalcium to creatinine ratio (UCa ⁄ Cr) >0.21 mg ⁄ mg in atleast three random morning urine samples to minimizethe daily variation and the absence of hypercalcaemiaor other known causes of hypercalciuria. In childrenwith increased UCa ⁄ Cr, hypercalciuria was confirmed bya 24-h urinary calcium excretion over 4 mg ⁄ kg ⁄ day.Renal ultrasound was performed for the detection ofstructural renal abnormality or nephrolithiasis in allchildren.

Calcium loading testAn oral calcium loading test was performed to determinethe subtypes of IH according to the modified protocols ofStapleton et al. and Pak et al. (4,5). Patients were main-tained on a low-calcium diet (<400 mg ⁄ day) for 5 days andthen fasted from 9 PM of the preceding evening to 9 AM ofthe test day except for 300 mL ⁄ 1.73 m2 of distilled water at9 PM, midnight, and 7 AM. Timed urine was collected for2 h from 7 to 9 AM. At 9 AM, after a standardized lightbreakfast, containing 100 mg calcium, 25 mEq sodium and100 mg phosphate, 1 g ⁄ 1.73 m2 of elemental calcium asoral calcium gluconate was administered. Timed urine wasalso collected from 9 AM to 1 PM under the condition ofproviding 300 mL ⁄ 1.73 m2 of distilled water.

With the results of timed urine samples, renal hypercalci-uria was defined as a fasting UCa ⁄ Cr > 0.20, and absorptivehypercalciuria as a fasting UCa ⁄ Cr £ 0.20 and UCa ⁄ Cr> 0.20 after oral calcium administration.

Thiazide treatment and dose titrationThe patients with renal hypercalciuria were commenced onHCT (Dichlozid) with the initial dose of 0.5 mg ⁄ kg ⁄ dayexcept three patients. After starting therapy, the patientswere examined periodically at the outpatient clinic for uri-nalysis and UCa ⁄ Cr. If hypercalciuria persists despite theinitial dose of HCT (0.5 mg ⁄ kg ⁄ day), the dose of HCT grad-ually increased to achieve the appropriate hypocalciuriceffect (UCa ⁄ Cr £ 0.2). The patients did not receive addi-tional treatment such as sodium-restricted diet or anycitrate applications.

Statistical analysisThe data were expressed as mean ± standard deviation(SD). Statistics were done with SPSS version 18.0 (SPSSInc., Chicago, IL, USA) for Windows, and urinary calcium

excretion before and after HCT were compared using pairedt-test. All differences were considered significant at p < 0.05.

RESULTSAll 28 patients exhibited UCa ⁄ Cr > 0.20 mg ⁄ mg before(mean ± SD, 0.35 ± 0.18) and after calcium loading test(mean ± SD, 0.49 ± 0.22) and increased urinary calciumexcretion rates of more than 4.0 mg ⁄ kg ⁄ day (mean ± SD,6.91 ± 3.02).

Renal ultrasounds were performed in all patients, whichshowed normal in 17 (61%), renal stones in 6 (21%), hydro-nephrosis in 4 (14%) and an isolated cyst in one (4%). Atthe follow-up ultrasound performed 6–12 months later,renal stones subsided in 4 (67%) of the 6 patients and thesize of renal stone decreased in two (33%). In one patient, aleft renal stone was newly developed on the follow-up renalultrasound. UTI was documented in nine children (32%) atpresentation.

Haematuria gradually resolved in 25 patients (89%) inaccordance with the improvement of hypercalciuria. How-ever, microscopic haematuria associated with hypercalciu-ria recurred in two patients at 1 and 5 months, respectively,after discontinuation of HCT therapy, requiring HCTretreatment. However, there were no gross haematuria orstone formation in these patients.

Figure 1 shows the clinical response to HCT treatment in28 children. Overall, random UCa ⁄ Cr significantly decreasedfrom 0.44 ± 0.15 to 0.09 ± 0.05 after HCT therapy (p <0.0001). Although the treatment of three patients were initi-ated with higher doses of HCT (1–2 mg ⁄ kg ⁄ day), the lowdoses (0.5 mg ⁄ kg ⁄ day) of HCT reduced urinary calciumexcretion (UCa ⁄ Cr < 0.2) in 22 (88%) of the 25 patients. Inthese patients, hypercalciuria resolved within 2 monthsafter initial HCT treatment. However, three patients (12%)required the increased doses (1–2 mg ⁄ kg ⁄ day) after themedian 12 months and hypercalciuria disappeared in twoof them after discontinuing HCT. Initial urinary calciumexcretion of three patients who required higher doses ofHCT was not different compared to that of 19 patients whowere responsive to lower doses (p = 0.237).

In 19 patients excluding 3 patients who were lost atfollow-up, HCT therapy was discontinued after 12.5 ±5.3 months of treatment (range 4–29 months). Of these,11 patients maintained normocalciuria, while 8 showedincreased urinary calcium excretion at 2.9 ± 2.3 monthsafter treatment was stopped, requiring HCT retreatment.

There were no differences in age, gender, the incidence ofgross or microscopic haematuria, Ca ⁄ Cr ratios before andafter calcium loading between children who did not relapsefollowing HCT discontinuation (11 ⁄ 19) and those whoneeded reinitiation of HCT therapy (8 ⁄ 19; p > 0.05). How-ever, initial 24-h urinary calcium excretion was significantlyhigher in patients who needed reinitiation of HCT than thatin those who did not relapse following HCT discontinuation(5.62 ± 1.54 vs. 8.23 ± 3.16, p = 0.028). We did not findany complications of HCT therapy including a decrease ofblood pressure in our patients.

Low-dose thiazide diuretics in children Choi et al.

e72 ª2011 The Author(s)/Acta Pædiatrica ª2011 Foundation Acta Pædiatrica 2011 100, pp. e71–e74

DISCUSSIONThere have been some reports on the classification of IH(3,4,8–12). Pak et al. (4) introduced the classification ofabsorptive, renal and resorptive hypercalciuria for IH.However, some authors reported that IH is a complex poly-genic trait and always a single disorder characterized byaltered calcium transport in the intestine, kidney and bone,because of various different combinations of multiplegenetic and dietary players (13).

Although oral calcium loading and deprivation test hasbeen used in differentiating absorptive and renal hypercalci-uria (5), it recently faced a challenge because of limitationsin clinical usefulness, such as a treatment strategy (conser-vative or HCT therapy) according to the symptoms (grosshaematuria or dysuria) regardless of IH subtypes (3).

Children with hypercalciuria have been reported to haveoften lower urinary tract signs and symptoms, whichinclude microscopic and gross haematuria, frequency,urgency, dysuria, incontinence and suprapubic pain (14).Also, in our study, all patients presented with microscopicor gross haematuria, which gradually resolved in 25 patients(89%) in accordance with the improvement of hypercalciu-ria after thiazide therapy, indicating that HCT might beeffective in resolving haematuria associated with renalhypercalciuria.

Idiopathic hypercalciuria is also the most common meta-bolic disorder found in patients with urolithiasis. In ourstudy, renal stone was found as a high incidence (21%) inchildren with renal hypercalciuria, compared to 5–10% ofthe general population (13). Also, renal stones disappearedor decreased of the size after thiazide therapy in mostpatients, suggesting it might be very effective in the treat-ment or prophylaxis of the renal stones.

We observed a high incidence of UTI (32%) in patientswith renal hypercalciuria, implicating IH might be a majorcontributing factor to UTI in children. In rat models, hyper-calciuria significantly affected on the cell architecture of theuroepithelium and disruption of the epithelial barrier of the

bladder, ureter and all kidney structures, especially on theproximal epithelial cells (15). Therefore, this complicationshould also be considered when managing patients with IHin clinical practice.

Thiazide diuretics have been widely used in correctingrenal hypercalciuria not only in children but also in adults(7). Lamberg et al. (6) first observed the effect of thiazideon calcium reabsorption in 1959. In TRPV5 knockoutmice, in which active calcium reabsorption in the distalconvoluted tubule was completely abolished (16), previousdata demonstrated that thiazide-induced hypocalciuriaoccurred from extracellular volume contraction (ECV),independent of active calcium reabsorption as inhibition ofsodium reabsorption in distal convoluted tubule. Conse-quently, ECV contraction resulted in a compensatoryincrease in renal proximal sodium reabsorption, therebyenhancing the electrochemical driving force for passiveparacellular calcium reabsorption (17). Furthermore, Janget al. (18) recently suggested that the hypocalciuric effectof HCT might be associated with increased protein abun-dance of TRPV5 in high salt or calcium diet–inducedhypercalciuric rats.

There have been few studies on the appropriate initialdosage of thiazide in controlling hypercalciuria. It wasreported that administration of low doses (0.75 mg ⁄ kg ⁄ day)of HCT for 3 months were ineffective in children with renalhypercalciuria and 1 mg ⁄ kg ⁄ day of HCT resulted in a rapidand long-lasting correction of renal hypercalciuria for aperiod of 2.5 ± 0.95 years (7). In our study, however, therelatively low dose (0.5 mg ⁄ kg ⁄ day) of HCT reduced uri-nary calcium excretion in 25 of the 28 patients, but threerequired increased doses (1–2 mg ⁄ kg ⁄ day) of HCT becauseof persistent hypercalciuria. Also, 42% (8 ⁄ 19) of ourpatients who had stopped HCT treatment increased thelevel of calciuria again, leading to low-dose HCT retreat-ment. Nevertheless, there has been no study regarding howlong HCT therapy should be continued in children withrenal hypercalciuria.

Total number of patients : 28

Low dose(0.5mg/kg/day) of HCT : 25 1–2 mg/kg/day of HCT : 3

F/U loss : 3

Uca/cr ≥ 0.2Uca/cr < 0.2 HCT D/C HCT continuation

19 3Uca/cr < 0.2 : 2 Uca/cr < 0.2 : 1

1–2 mg/kg/day of HCT

Uca/cr < 0.2 : 11

HCT D/C

Uca/cr < 0.2 : 2

HCT D/C

Uca/cr ≥ 0.2 : 8 Uca/cr ≥ 0.2 : 1

Figure 1 Clinical course of children with renal hypercalciuria after treatment with hydrochlorothiazide (HCT, hydrochlorothiazide; UCa ⁄ Cr, random urine calcium tocreatinine ratio; D ⁄ C, discontinuation).

Choi et al. Low-dose thiazide diuretics in children

ª2011 The Author(s)/Acta Pædiatrica ª2011 Foundation Acta Pædiatrica 2011 100, pp. e71–e74 e73

Thiazide can produce some complications, such as hypo-kalaemia, hyperglycaemia, metabolic alkalosis, hyperurica-emia and hyperlipidaemia. Although we did not measurethese metabolic parameters because of low doses of HCT,those should be closely monitored when high doses of HCTare used.

Our study also has some limitations: (1) small numbers ofpatients were studied, (2) a retrospective nature of our study,and (3) calcium loading test was performed only once.

Nevertheless, our results suggest that low dose (0.5 mg ⁄kg ⁄ day) of HCT may be safe and effective in controllingrenal hypercalciuria in children. HCT could be discontin-ued in most children with well-controlled hypercalciuria,while some required increased doses of HCT or HCTretreatment after discontinuation.

References

1. Albright F, Henneman P, Benedict PH, Forbes AP. Idiopathichypercalciuria: a preliminary report. Proc R Soc Med 1953; 46:1077–81.

2. Moore ES, Coe FL, McMann BJ, Favus MJ. Idiopathic hyper-calciuria in children: prevalence and metabolic characteristics.J Pediatr 1978; 92: 906–10.

3. Srivastava T, Schwaderer A. Diagnosis and management ofhypercalciuria in children. Curr Opin Pediatr 2009; 21: 214–19.

4. Pak CY, Kaplan R, Bone H, Townsend J, Waters O. A simpletest for the diagnosis of absorptive, resorptive and renal hyper-calciurias. N Engl J Med 1975; 292: 497–500.

5. Stapleton FB, Roy S 3rd, Noe HN, Jerkins G. Hypercalciuria inchildren with hematuria. N Engl J Med 1984; 310: 1345–8.

6. Lamberg BA, Kuhlback B. Effect of chlorothiazide andhydrochlorothiazide on the excretion of calcium in urine.Scand J Clin Lab Invest 1959; 11: 351–7.

7. Voskaki I, al Qadreh A, Mengreli C, Sbyrakis S. Effect of hydro-chlorothiazide on renal hypercalciuria. Child Nephrol Urol1992; 12: 6–9.

8. Bataille P, Achard JM, Fournier A, Boudailliez B, Westeel PF,el Esper N, et al. Diet, vitamin D and vertebral mineral densityin hypercalciuric calcium stone formers. Kidney Int 1991; 39:1193–205.

9. Coe FL, Canterbury JM, Firpo JJ, Reiss E. Evidence for second-ary hyperparathyroidism in idiopathic hypercalciuria. J ClinInvest 1973; 52: 134–42.

10. Frick KK, Bushinsky DA. Molecular mechanisms of primaryhypercalciuria. J Am Soc Nephrol 2003; 14: 1082–95.

11. Srivastava T, Alon US. Pathophysiology of hypercalciuria inchildren. Pediatr Nephrol 2007; 22: 1659–73.

12. Stapleton FB, Noe HN, Jerkins G, Roy S 3rd. Urinary excretionof calcium following an oral calcium loading test in healthychildren. Pediatrics 1982; 69: 594–7.

13. Vezzoli G, Soldati L, Gambaro G. Hypercalciuria revisited: oneor many conditions? Pediatr Nephrol 2008; 23: 503–6.

14. Heiliczer JD, Canonigo BB, Bishof NA, Moore ES. Noncalculiurinary tract disorders secondary to idiopathic hypercalciuriain children. Pediatr Clin North Am 1987; 34: 711–8.

15. Akil I, Kavukcu S, Inan S, Yilmaz O, Atilla P, Islekel H, et al.Evaluation of histologic changes in the urinary tract of hyper-calciuric rats. Pediatr Nephrol 2006; 21: 1681–9.

16. Hoenderop JG, van Leeuwen JP, van der Eerden BC, KerstenFF, van der Kemp AW, Merillat AM, et al. Renal Ca2+ wasting,hyperabsorption, and reduced bone thickness in mice lackingTRPV5. J Clin Invest 2003; 112: 1906–14.

17. Mensenkamp AR, Hoenderop JG, Bindels RJ. Recent advancesin renal tubular calcium reabsorption. Curr Opin NephrolHypertens 2006; 15: 524–9.

18. Jang HR, Kim S, Heo NJ, Lee JH, Kim HS, Nielsen S, et al.Effects of thiazide on the expression of TRPV5, calbindin-D28K, and sodium transporters in hypercalciuric rats. J KoreanMed Sci 2009; 24(Suppl): S161–9.

SUPPORTING INFORMATIONAdditional Supporting Information may be found in theonline version of this article:

Table S1 Clinical data of 22 patients with idiopathic renalhypercalciuria.

Please note: Wiley-Blackwell is not responsible for the con-tent or functionality of any supporting materials supplied bythe authors. Any queries (other than missing material)should be directed to the corresponding author for thearticle.

Low-dose thiazide diuretics in children Choi et al.

e74 ª2011 The Author(s)/Acta Pædiatrica ª2011 Foundation Acta Pædiatrica 2011 100, pp. e71–e74