Interventions for metabolic bone disease in children with chronic kidney disease

Deirdre Hahn; Elisabeth M Hodson; Jonathan C Craig

doi:10.1002/14651858.CD008327.pub2

. 2015 Nov 12;2015(11):CD008327. doi: 10.1002/14651858.CD008327.pub2

Interventions for metabolic bone disease in children with chronic kidney disease

Deirdre Hahn ^1,^✉, Elisabeth M Hodson ^2,³, Jonathan C Craig ^2,³

Editor: Cochrane Kidney and Transplant Group

PMCID: PMC7180137 PMID: 26561037

Abstract

Background

Bone disease is common in children with chronic kidney disease (CKD) and when untreated may result in bone deformities, bone pain, fractures and reduced growth rates. This is an update of a review first published in 2010.

Objectives

This review aimed to examine the benefits (improved growth rates, reduced risk of bone fractures and deformities, reduction in PTH levels) and harms (hypercalcaemia, blood vessel calcification, deterioration in kidney function) of interventions (including vitamin D preparations and phosphate binders) for the prevention and treatment of metabolic bone disease in children with CKD.

Search methods

We searched the Cochrane Kidney and Transplant Specialised Register to 8 September 2015 through contact with the Trial's Search Co‐ordinator using search terms relevant for this review.

Selection criteria

We included randomised controlled trials (RCTs) comparing different interventions used to prevent or treat bone disease in children with CKD stages 2 to 5D.

Data collection and analysis

Data were assessed for study eligibility, risk of bias and extracted independently by two authors. Results were reported as risk ratios (RR) or risk differences (RD) with 95% confidence intervals (CI) for dichotomous outcomes. For continuous outcomes the mean difference (MD) or standardised mean difference (SMD) with 95% confidence intervals (CI) was used. Statistical analyses were performed using the random‐effects model.

Main results

This review included 18 studies (576 children); three new studies were added for this update. Adequate sequence generation and allocation concealment were reported in 12 and 11 studies respectively. Only four studies reported blinding of children, investigators or outcome assessors. Nine studies were at low risk of attrition bias and 12 studies were at low risk of selective reporting bias.

Eight different interventions were compared. Two studies compared intraperitoneal (IP) with oral calcitriol. PTH levels were significantly lower with IP compared with oral calcitriol (1 study: MD ‐501.00 pg/mL, 95% CI ‐721.54 to ‐280.46) but the number of children with abnormal bone histology did not differ between treatments. Three studies compared intermittent with daily oral calcitriol. The change in mean height SDS (1 study: MD 0.13, 95% CI ‐0.22 to 0.48) and the percentage fall in parathyroid hormone (PTH) levels at eight weeks (1 study: MD ‐5.50%, 95% CI ‐32.37 to 21.37) and 12 months (1 study: MD ‐6.00% 95% CI ‐25.27 to 13.27) did not differ between treatments.

Four studies compared active vitamin D preparations (calcitriol, paricalcitol, 1α‐hydroxyvitamin D) with placebo or no specific treatment. One study reported vitamin D preparations significantly reduced PTH levels (‐55.00 pmol/L, 95% CI ‐83.03 to ‐26.97). There was no significant difference in hypercalcaemia risk with vitamin D preparations compared with placebo or no specific treatment (4 studies, 103 children: RD 0.08 mg/dL, 95% CI ‐0.08 to 0.24). However, there was heterogeneity (I² = 55%) with one study showing a significantly greater risk of hypercalcaemia with intravenous (IV) calcitriol administration. Two studies (97 children) compared calcitriol with other vitamin D preparations and both found no significant differences in growth between preparations.

Two studies compared ergocalciferol in patients with CKD and vitamin D deficiency. Elevated PTH levels developed significantly later in ergocalciferol treated children (1 study: hazard ratio 0.30, 95% CI 0.09 to 0.93) though the number with elevated PTH levels did not differ between groups (1 study, 40 children: RR 0.33, 95% CI 0.11 to 1.05).

Two studies compared calcium carbonate with aluminium hydroxide as phosphate binders. One study (17 children: MD ‐0.86 SDS, 95% CI ‐2.24 to 0.52) reported no significant difference in mean final height SDS between treatments. Three studies compared sevelamer with calcium‐containing phosphate binders. There were no significant differences in the final calcium, phosphorus or PTH levels between binders. More episodes of hypercalcaemia occurred with calcium‐containing binders. One study reported no significant differences between calcitriol and doxercalciferol in bone histology or biochemical parameters.

Authors' conclusions

Bone disease, assessed by changes in PTH levels, is improved by all vitamin D preparations. However, no consistent differences between routes of administration, frequencies of dosing or vitamin D preparations were demonstrated. Although fewer episodes of high calcium levels occurred with the non‐calcium‐containing phosphate binder, sevelamer, compared with calcium‐containing binders, there were no differences in serum phosphorus and calcium overall and phosphorus values were reduced to similar extents. All studies were small with few data available on patient‐centred outcomes (growth, bone deformities) and limited data on biochemical parameters or bone histology resulting in considerable imprecision of results thus limiting the applicability to the care of children with CKD.

Plain language summary

Interventions for metabolic bone disease in children with chronic kidney disease

Chronic kidney disease (CKD) resulting in reduced kidney function and the need for dialysis and kidney transplant is associated with abnormalities in serum calcium and phosphorus levels leading to high levels of the parathyroid hormone (PTH) and to bone disease. This may result in bone deformities, bone pain, fractures and reduced growth rates. Commonly used treatments (vitamin D compounds and phosphate binders) aim to prevent or correct these outcomes. However, these treatments may raise levels of blood calcium, allow calcium and phosphorus deposition in blood vessels and lead to early cardiovascular disease, which is known to be a problem in adults with CKD.

This review identified 18 small randomised studies involving 576 children comparing different vitamin D compounds administered via different routes and frequencies and different phosphate binders. Only five studies reported growth rates and no differences were detected between treatments. Bone disease, as assessed by changes in PTH levels, was improved by all vitamin D preparations regardless of preparation or route or frequency of administration. Fewer episodes of high blood calcium levels occurred with the non‐calcium‐containing binder, sevelamer, compared with calcium‐containing binders. As newer treatments for renal bone disease are developed, comparisons with the current standard therapies will be required in well‐designed randomised studies in children using outcome measures including those of direct clinical relevance to children and their families such as rates of growth, reduction in bone fractures and bone pain and reduction in calcification in blood vessels.

Background

Description of the condition

Chronic kidney disease (CKD) causes disordered regulation of mineral metabolism (Wesseling‐Perry 2013). Because this disorder results in renal osteodystrophy and vascular and/or soft tissue calcification, the manifestations of the disorder are now known as chronic kidney disease mineral and bone disorder (CKD‐MBD) (Moe 2006). CKD‐MBD is defined as a systemic disorder of bone and mineral metabolism due to CKD and manifested by one or a combination of:

abnormalities of calcium. phosphorus, parathyroid hormone (PTH) or vitamin D metabolism;
abnormalities in bone turnover, mineralization, volume, linear growth or strength; and
vascular or other soft tissue calcification.

In children CKD‐MBD may be associated with increased fracture rates, reduced linear growth, bony deformities and chronic pain. In a review of 890 children on peritoneal dialysis, 5% had limb deformities, 1.4% had bone pain and 1.5% had vascular calcification (Borzych 2010). Abnormalities of bone turnover, mineralization and volume in CKD‐MBD can be quantitated using bone histomorphometry. The predominant lesion noted on bone biopsy in children on dialysis is one of high bone turnover (in 57% to 100% of patients) with low turnover bone disease much less common (4% of patients) (Bakkaloglu 2010; Hodson 1982; Salusky 2005a). Abnormal skeletal mineralization is commonly associated with both high and low turnover bone disease in dialysis patients. Among children with CKD stages 2‐4, high turnover bone disease was seen in 29% of children with CKD stage 4 but was not seen in children with stage 2 disease and was uncommon in children with stage 3 disease (Wesseling‐Perry 2012; Hodson 1982). In contrast mineralization abnormalities occurred in 43% children with stage 2 CKD and in 86% of children with stage 4 CKD (Wesseling‐Perry 2012); these findings confirm previous findings in early stages of CKD (Hodson 1982). Low turnover bone disease is rare in children not on dialysis.

Although bone disease may not be evident on bone histology in early CKD and plasma levels of calcium and phosphorus are normal, increased levels of the hormone fibroblast growth factor 23 (FGF23) (Wesseling‐Perry 2013) increase renal phosphate excretion and inhibit 1α‐hydroxylase activity thus suppressing circulating levels of 1, 25 (OH)₂D leading to increased levels of parathyroid hormone (PTH).

Bone biopsy is an invasive procedure and is now generally limited to research studies so that radiological and biochemical abnormalities are used as surrogate measures of bone disease in CKD‐MBD. Radiological diagnosis is insensitive and cannot distinguish low‐turnover or adynamic bone disease from the high turnover state of secondary hyperparathyroidism. Biochemical abnormalities of parathyroid hormone, serum calcium and phosphate levels are frequently used as markers of bone disease if outside of the recommended KDOQI or European guideline parameters (KDOQI 2005; Klaus 2006). Abnormalities of these values, suggestive of histological changes, have been demonstrated in 28% to 81% of children with CKD stages 2‐5 (Blaszak 2005; Seikaly 2003). These biochemical abnormalities have also been used to specifically diagnose low turnover bone disease. In 41 dialysed children (31 peritoneal dialysis), low turnover bone disease was diagnosed in 48% based on the presence of elevated serum calcium with parathyroid hormone (PTH) values below recommended levels (Avila‐Diaz 2006). PTH levels are most commonly used to monitor the effectiveness of therapy. However optimal target ranges are unclear in part because earlier PTH assays measured active and inactive PTH fragments and newer assays still show considerable variation between PTH assays.

In the absence of clinical symptoms and signs, it has been unclear until recently what impact CKD‐MBD has on the outcome for children with CKD. Recent data have demonstrated an increased risk of coronary arterial calcification in young adults on dialysis (Goodman 2000) while elevated levels of PTH and phosphate are independent risk factors for left ventricular hypertrophy (Bakkaloglu 2011). These factors have been associated with increased mortality in children and young adults with CKD.

Description of the intervention

Treatment of CKD‐MBD aims to normalise mineral metabolism and minimise progression of extraskeletal calcification by maintaining blood levels of calcium and phosphorus close to the normal range for age and maintaining PTH levels at levels considered to be appropriate for the stage of CKD. The mainstays of treatment are with phosphate binders (calcium or non‐calcium containing) and vitamin D metabolites (calcitriol, 1α‐hydroxyvitamin D, newer vitamin D analogues). Dietary measures are instituted to reduce phosphate intake while maintaining adequate calcium and vitamin D intake. Also calcium levels in the dialysis fluid can be manipulated to maintain normocalcaemia. New agents include calcimimetic agents (cinacalcet), which control secondary hyperparathyroidism. For medically unresponsive patients, parathyroidectomy may be required.

How the intervention might work

Early in the development of CKD, circulating levels of 1,25 (OH)₂ vitamin D falls following suppression of the renal enzyme 1‐ɑ‐hydoxylase by FGF23, resulting in reduced calcium absorption from the gut and increased PTH levels (Wesseling‐Perry 2013). Increased PTH levels initially maintain serum calcium levels by increasing bone resorption and by stimulating 1‐ɑ‐hydoxylase activity. With further decline in glomerular filtration rate (GFR), phosphate levels rise. These lower calcium levels and further suppress renal enzyme 1‐ɑ‐hydoxylase levels so PTH levels rise further. PTH increases bone turnover leading to renal osteodystrophy. Therefore therapies which increase gut absorption of calcium, reduce phosphate levels and increase circulating levels of 1,25 (OH)₂ vitamin D will reduce PTH levels. However both calcium‐containing phosphate binders and vitamin D metabolites may cause hypercalcaemia and elevated calcium‐phosphorus product and predispose to vascular and soft tissue calcification. Calcimimetic agents modulate the calcium sensitive receptor in parathyroid glands, increase intracellular calcium and decrease PTH release.

Why it is important to do this review

There are a large number of studies reporting the efficacy of various medications and dietary manipulations to prevent and treat CKD‐MBD in children and there is extensive clinical experience confirming that current treatment of CKD‐MBD has reduced the severity of bony deformities and fractures over the past few decades. However there remains considerable uncertainty in children about the vascular outcomes related to treatment or non‐treatment of CKD‐MBD. In addition because of the recognised severity of potential side‐effects and the uncertain efficacy of some of the therapeutic agents used to treat CKD‐MBD, it is appropriate to critically review the treatment options.

Objectives

Methods

Criteria for considering studies for this review

Types of studies

All randomised controlled trials (RCTs) and quasi‐RCTs (studies in which allocation to treatment was obtained by alternation, use of alternate medical records, date of birth or other predictable methods) examining treatments for the prevention and treatment of CKD‐MBD in children and adolescents were included.

Types of participants

Inclusion criteria

Studies involving children with CKD stages 2 to 5D (glomerular filtration rate < 90 mL/min/1.73 m²)
Childhood was defined according to the definitions applied in the included studies, but did not exceed 21 years of age.

Exclusion criteria

Studies of children with CKD secondary to primary tubulopathies, e.g. cystinosis, or with diseases known to directly affect bones e.g. primary hyperoxaluria, and studies in children following kidney transplant were excluded. However it is possible that individual children with the above disorders might be included within an eligible study but not specifically specified. Studies of recombinant human growth hormone in children with CKD were excluded as these are included in a separate Cochrane Review (Growth Hormone in children with chronic kidney disease) (Hodson 2012).

Types of interventions

Interventions considered for inclusion were as follows.

Dietary
Pharmacological ‐ specifically vitamin D or metabolites, calcimimetic and phosphate binding agents
Surgical
Herbal or alternative treatments
Changes in dialysis prescription.

For each of these interventions the following comparisons were considered.

Intervention versus placebo
Intervention A versus intervention B
Frequency and mode of administration (e.g. oral or intravenous (IV))
Dose and duration of treatment.

Types of outcome measures

Primary outcomes

Patient‐centred outcome measures

Growth
Bone fractures
Bone deformities
Symptoms related to hypercalcaemia
Parathyroidectomy

Secondary outcomes

Patient‐centred outcome measures

Commencing dialysis treatment
Dialysis‐related clinical events
Parathyroidectomy.

Surrogate outcomes

Change in bone histology
Changes in radiological abnormalities
Changes in PTH levels
Changes in alkaline phosphatase (ALP) levels
Changes in serum calcium, phosphorus and calcium‐phosphorus product
Changes in FGF23 levels

Adverse events

Vascular or extraosseous calcifications
Deterioration of kidney function
Hyperphosphataemia
Hypercalcaemia
Elevation of calcium‐phosphorus product
Radiological deterioration of CKD‐MBD
Development of adynamic bone disease on bone histomorphometry
Hypertension or hypotension
Aggravation of anaemia.

Search methods for identification of studies

Electronic searches

We searched the Cochrane Kidney and Transplant Specialised Register to 8 September 2015 through contact with the Trials Search Co‐ordinator using search terms relevant to this review. The Specialised Register contains studies identified from the following sources.

Monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL)
Weekly searches of MEDLINE OVID SP
Handsearching of kidney‐related journals and the proceedings of major kidney conferences
Searching of the current year of EMBASE OVID SP
Weekly current awareness alerts for selected kidney journals
Searches of the International Clinical Trials Register (ICTRP) Search Portal and ClinicalTrials.gov.

Studies contained in the Specialised Register are identified through search strategies for CENTRAL, MEDLINE, and EMBASE based on the scope of Cochrane Kidney and Transplant. Details of these strategies, as well as a list of handsearched journals, conference proceedings and current awareness alerts, are available in the Specialised Register section of information about the Cochrane Kidney and Transplant.

See Appendix 1 for search terms used in strategies for this review.

Searching other resources

Reference lists of clinical practice guidelines review articles and relevant studies.
Letters seeking information about unpublished or incomplete studies to investigators known to be involved in previous studies.

Data collection and analysis

Selection of studies

In the previous version of this review (Geary 2009) all electronically derived abstracts and study titles were assessed for subject relevance and methodological quality. All possible RCTs or quasi‐RCTs which were relevant were assigned specific topic keywords in Reference Manager and the full published paper was obtained for full assessment. The review was undertaken by four authors. The search strategy described was used to obtain titles and abstracts of studies that were relevant to the review. The titles and abstracts were screened independently by DG, EH and DH, who discarded studies that were not applicable. However studies and reviews that might include relevant data or information on studies were retained initially. Three authors independently assessed retrieved abstracts, and if necessary, the full text of these studies to determine which studies satisfied the inclusion criteria.

In this update, study titles and abstracts were reviewed by two authors. Full text articles of studies considered relevant were obtained and assessed for eligibility by the same authors.

Data extraction and management

A data abstraction form was devised to record details of data elements such as outcome measures, participants and intervention of each included study for the 2009 review. Only comparisons and outcomes which were pre‐specified in the protocol were included. For this review, data were abstracted by a single assessor and a sample was double checked.

For this update, data extraction and assessment of risk of bias were performed by two authors using standardised data abstraction forms. Disagreements not resolved by discussion between authors were referred to a third author. Studies reported in languages other than English were to be translated before data extraction, but no foreign language reports were identified. Where more than one report of a study was identified, data were extracted from all reports. Where there were discrepancies between reports, data from the primary source were used. Study authors were contacted for additional information about studies.

Assessment of risk of bias in included studies

Hard copies of studies were independently assessed for the methodological quality by two assessors for the 2010 review. Quality assessments were made for allocation concealment, blinding, description of withdrawals and drop‐outs, numbers lost to follow‐up, and whether intention‐to‐treat (ITT) analysis was possible.

In the 2014 update, the following terms were assessed using the risk of bias assessment tool (Higgins 2011) (see Appendix 2).

Was there adequate sequence generation (selection bias) ?
Was allocation adequately concealed (selection bias)?
Was knowledge of the allocated interventions adequately prevented during the study?
- Participants and personnel (performance bias)
- Outcome assessors (detection bias)
Were incomplete outcome data adequately addressed (attrition bias)?
Are reports of the study free of suggestion of selective outcome reporting (reporting bias)?
Was the study apparently free of other problems that could put it at a risk of bias?

Measures of treatment effect

For dichotomous outcomes (number of children with improved growth, radiological bone changes, improved bone histology) results were expressed as risk ratio (RR) with 95% confidence intervals (CI). For numbers of children experiencing adverse effects, risk difference (RD) with 95% CI was used. For continuous outcomes scales of measurement were used to assess the effects of treatment (levels of PTH, serum levels of calcium, phosphorus and calcium‐phosphorus product, and creatinine clearance (CrCl), the mean difference (MD) with 95% CI were calculated unless the scales were different; in this instance, standard mean difference (SMD) was used.

Unit of analysis issues

Data from the first phase of cross‐over RCTs could not be separated so results from cross‐over studies were reported qualitatively.

Dealing with missing data

We aimed to analyse available data in meta‐analyses using intention‐to‐treat (ITT) data. However, where ITT data were only available graphically or not provided and additional information could not be obtained from study authors, per‐protocol (PP) data were used in analyses.

Assessment of heterogeneity

Heterogeneity was analysed using a Chi² test on N‐1 degrees of freedom, with an alpha of 0.05 used for statistical significance and with the I² test (Higgins 2003). I² values of 25%, 50% and 75% correspond to low, medium and high levels of heterogeneity respectively.

Assessment of reporting biases

Because of the few studies available for each intervention, it was not possible to use funnel plots to assess for the potential existence of small study bias (Higgins 2011).

Data synthesis

Data were pooled using a random‐effects model for dichotomous and continuous data.

Subgroup analysis and investigation of heterogeneity

We had planned subgroup analyses to examine certain between‐study differences in participants (age, stage of CKD, type of dialysis), interventions (agent, dose and duration of treatment) and risk of bias hypothesised to explain any observed heterogeneity of treatment effects but there were insufficient studies for these to be performed.

Sensitivity analysis

We wished to perform sensitivity analyses in order to explore the influence of the following factors on effect size.

Repeating the analysis excluding unpublished studies
Repeating the analysis taking account of risk of bias, as specified
Repeating the analysis excluding any very long or large studies to establish how much they dominate the results
Repeating the analysis excluding studies using the following filters: diagnostic criteria, language of publication, source of funding (industry versus other), and country.

However except for one analysis, the maximum number of studies included in any analysis was two so we were not able to carry out any sensitivity analyses.

Results

Description of studies

Results of the search

2009 review

In the 2009 review, of the 1137 titles and abstracts identified, 19 studies were identified for full text review. Of these 15 studies (26 reports) (Ardissino 2000; Eke 1983; GFRD Study 1990; Greenbaum 2005; Greenbaum 2007; Hodson 1985; Jones 1994; Klaus 1995; Mak 1985; Pieper 2006; Salusky 1991; Salusky 1998; Salusky 2005; Schmitt 2003; Watson 1988) involved the defined populations and addressed relevant interventions and were included in the review. A copy of the completed manuscript was provided before publication by Greenbaum 2007. Three studies were cross‐over studies (Jones 1994; Mak 1985; Pieper 2006), and one study (Klaus 1995) was available in abstract form only. Four studies (nine reports) were excluded for ineligible intervention (Ardissino 2000a), ineligible population (El Husseini 2004; Ferraris 2000) and uncertainty as to whether the study was an RCT (Bettinelli 1986). There was no disagreement between authors regarding inclusion of studies. The 2009 review included 15 studies (26 reports) with 472 children.

2015 update

For the 2015 update 47 new reports were identified (Figure 1). Of these, four reports were of three new studies (Gulati 2010; Rianthavorn 2013; Shroff 2012) and 38 reports were of seven previously included studies (Eke 1983; GFRD Study 1990; Greenbaum 2005; Salusky 1991; Salusky 1998; Salusky 2005; Watson 1988). The remaining five new reports were excluded. Two studies were excluded as the populations were ineligible (Choudhary 2014; Kim 2006b), and in one study (Witmer 1976) randomisation was unclear. The remaining three reports were from two previously excluded studies (Ardissino 2000a; Ferraris 2000).

Re‐evaluation of Ardissino 2000 and Schmitt 2003 indicated that the data in Schmitt 2003 represented a 12 month follow‐up of 29 prepubertal children treated in Ardissino 2000. For ease of identification, we have continued to report these studies as two studies.

To allow analyses of comparisons between calcitriol and doxercalciferol and for separate reporting of the comparison between sevelamer and calcium carbonate in this 2 x 2 longitudinal factorial study, we have divided Salusky 2005 into three studies (Salusky 2005; Salusky 2005a; Salusky 2005b). Salusky 2005 includes the groups treated with sevelamer or calcium carbonate irrespective of vitamin D preparation used. Salusky 2005a includes the two groups treated with doxercalciferol while Salusky 2005b includes the two groups treated with calcitriol.

Therefore the 2015 update included 18 studies (68 reports) with 576 enrolled children.

Included studies

The 18 included studies were divided into eight treatment comparisons (Figure 1).

Intraperitoneal calcitriol versus oral calcitriol

Two studies (40 enrolled; 40 evaluated) compared intraperitoneal compared with oral calcitriol. Children receiving continuous cycling or continuous ambulatory peritoneal dialysis were treated for 12 months (Salusky 1998) or 3 months (Jones 1994). The study by Jones 1994 was a cross‐over study and data could not be meta‐analysed.

Intermittent oral calcitriol versus daily oral calcitriol

Three studies (109 enrolled; 104 evaluated) compared intermittent oral administration of calcitriol with daily oral administration in children with CKD stages 2 to 5 for 8 to 10 weeks (Ardissino 2000), 12 months (Schmitt 2003), or from 2 to 36 weeks (Klaus 1995).

Different vitamin D preparations versus placebo/no specific treatment

Four studies (104 enrolled; 103 evaluated) compared different vitamin D preparations administered orally or IV with placebo/no specific treatment in CKD patients stages 3 and 4 (Eke 1983) and in patients receiving peritoneal or haemodialysis (Greenbaum 2005; Greenbaum 2007; Watson 1988).

Different vitamin D preparations

Two studies (106 enrolled; 97 evaluated) compared different vitamin D preparations. GFRD Study 1990 compared oral calcitriol with oral dihydrotachysterol in children with CKD stages 3 and 4 treated for 12 months. Hodson 1985 compared oral calcitriol with oral ergocalciferol in children on dialysis or CKD stages 2 to 4.

One study (60 enrolled; 51 evaluated) compared doxercalciferol (Salusky 2005a) and calcitriol (Salusky 2005b) in combination with either sevelamer or calcium carbonate. Comparisons were reported for vitamin D preparations irrespective of the phosphate binder given.

Ergocalciferol versus placebo/control

Two studies (67 enrolled; 60 evaluated) compared ergocalciferol with placebo/no specific treatment in patients with CKD stages 2 to 5D (Rianthavorn 2013; Shroff 2012). In Rianthavorn 2013 the primary outcome was reduction in the dose of erythrocyte‐stimulating agent (ESA).

Calcium carbonate versus aluminium hydroxide

Two studies (34 enrolled; 29 evaluated) compared calcium hydroxide with calcium carbonate in pre‐dialysis children (Mak 1985) or those receiving peritoneal dialysis (Salusky 1991).

Sevelamer versus calcium‐containing phosphate binders

Three studies (98 enrolled; 66 evaluated) compared sevelamer with calcium‐containing phosphate binders in children with CKD stages 2 to 4 (Gulati 2010) or receiving dialysis (Pieper 2006; Salusky 2005) compared the non‐calcium‐containing sevelamer with calcium carbonate (Salusky 2005) or calcium acetate (Gulati 2010; Pieper 2006). In Salusky 2005, children were also randomised to receive doxercalciferol or calcitriol. Factorial analysis provided no evidence of treatment interaction between the two sterols so comparisons were reported for phosphate binders irrespective of vitamin D sterol given.

No RCTs examining interventions with dietary changes, surgery, alterations in dialysis prescription or calcimimetic agents were identified.

Excluded studies

Seven studies were excluded (Ardissino 2000a; Bettinelli 1986; Choudhary 2014; El Husseini 2004; Ferraris 2000; Kim 2006b; Witmer 1976). One cross‐over study (Ardissino 2000a) in children with CKD examined calcium absorption only after oral or IV calcitriol. Two studies examined bone mineral density in kidney transplant patients treated with calcitonin, alendronate or 1α‐hydroxyvitamin D (El Husseini 2004) or with methylprednisolone or deflazacort (Ferraris 2000). Randomisation was unclear and data was no longer available in Witmer 1976.

Risk of bias in included studies

Figure 2; Figure 3

Methodological quality graph: review authors' judgements about each methodological quality item presented as percentages across all included studies

Risk of bias summary: review authors' judgements about each risk of bias item for each included study

Allocation

Sequence generation was deemed to be at low risk of bias in 12 studies (Ardissino 2000; GFRD Study 1990; Greenbaum 2005; Greenbaum 2007; Gulati 2010; Hodson 1985; Pieper 2006; Salusky 1998; Salusky 2005; Schmitt 2003; Shroff 2012; Watson 1988). One study was at high risk of bias as patients were randomised sequentially (Rianthavorn 2013). In the remaining five studies, the sequence generation methodology was unclear.

Allocation concealment was at low risk of bias in 11 studies (Ardissino 2000; GFRD Study 1990; Greenbaum 2005; Greenbaum 2007; Gulati 2010; Pieper 2006; Salusky 1998; Salusky 2005; Schmitt 2003; Shroff 2012; Watson 1988). Two studies (Hodson 1985; Rianthavorn 2013) were considered at high risk of bias. Allocation concealment methodology was unclear in the remaining five studies (Eke 1983; Jones 1994; Klaus 1995; Mak 1985; Salusky 1991).

Blinding

Four studies were blinded and considered to be at low risk of bias for performance bias ( GFRD Study 1990; Greenbaum 2005; Greenbaum 2007; Shroff 2012). Blinding was unclear in one study, which was reported to be double‐blinded but did not clarify how this was achieved (Eke 1983). The remaining thirteen studies were not blinded and were considered at high risk of performance bias.

Since the primary outcomes (PTH level, bone mineralization) in all studies were based on laboratory assessment, and unlikely to be influenced by blinding, all included studies were considered to be at low risk of detection bias.

Incomplete outcome data

Nine studies were considered at low risk of attrition bias (Eke 1983; Greenbaum 2005; Greenbaum 2007; Gulati 2010; Jones 1994; Mak 1985; Rianthavorn 2013; Shroff 2012; Watson 1988). Seven studies were considered at high risk of attrition bias with more than 15% loss to follow‐up or exclusion from analysis (Ardissino 2000; GFRD Study 1990; Hodson 1985; Pieper 2006; Salusky 1991; Salusky 2005; Schmitt 2003). In the remaining two studies attrition bias was considered unclear (Klaus 1995; Salusky 1998). Loss to follow‐up or exclusion resulted commonly when children on dialysis underwent kidney transplant. Other reasons for exclusion were non‐adherence to treatment, protocol violation or withdrawal by families or physicians.

Selective reporting

Studies were considered to be at high risk of reporting bias if they did not provide data on final or change in PTH, calcium, phosphorus, calcium‐phosphorus product or ALP levels, bone histology, patient centred outcomes such as fractures or growth, adverse events such as hypercalcaemia or all‐cause mortality. Seven studies were considered at high risk of reporting bias (Eke 1983; GFRD Study 1990; Jones 1994; Klaus 1995; Mak 1985; Pieper 2006; Salusky 2005). The remaining 11 studies were considered to be at low risk of reporting bias.

Other potential sources of bias

Eleven studies appeared to be free of other potential sources of bias (Ardissino 2000; Eke 1983; GFRD Study 1990; Gulati 2010; Hodson 1985; Mak 1985; Salusky 1991; Salusky 1998; Salusky 2005; Schmitt 2003; Shroff 2012). Five studies were funded by industry and considered at high risk of bias (Greenbaum 2005; Greenbaum 2007; Jones 1994; Pieper 2006; Watson 1988). In the remaining two studies, it was unclear whether the study was free of other potential sources of bias (Klaus 1995; Rianthavorn 2013).

Effects of interventions

Intraperitoneal versus oral calcitriol

Two studies investigated this comparison (Salusky 1998; Jones 1994) (Table 1). The first study (Jones 1994) compared IP with oral calcitriol in a cross‐over study including seven children and reported that mean height standard deviation score (SDS) did not differ between groups at the end of the study. The data for each part of the cross‐over could not be separated.

1. Intraperitoneal versus oral calcitriol in children on peritoneal dialysis.

Outcome	Jones 1994^a	Salusky 1998	Summary estimate^b
Abnormal bone biopsy	Not reported	12/16^c versus 12/17^d	RR 1.06 (0.70 to 1.61)
Adynamic bone disease	Not reported	7/16 versus 5/17	RR 1.49 (0.59 to 3.74)
Osteitis fibrosa	Not reported	2/16 versus 6/17	RR 0.35 (0.08 to 1.51)
Mixed/mild disease	Not reported	3/16 versus 7/17	RR 0.46 (0.14 to 1.46)
Normal/reduced bone formation rate	Not reported	11/16 versus 11/17	RR 1.06 (0.66 to 1.72)
Bone formation rate (µm²/mm²/d)	Not reported	297 ± 472 versus 586 ± 973	MD ‐289.00 (‐806.13 to 228.13)
PTH (pg/mL)	No significant difference	169 ± 228 versus 670 ± 400	MD ‐501.00 (‐721.54 to ‐280.46)
Maximum serum calcium (mg/dL)	Not reported	10.4 ± 2 versus 9.7 ± 1.64	MD 0.70 (‐0.55 to 1.95)
Hypercalcaemic patients	0/7 versus 0/7	8/16 versus 5/17	RD 0.21 (‐0.12 to 0.53)
Hypophosphataemic patients	Not reported	2/16 versus 3/17	RD ‐0.05 (‐0.29 to 0.19)
Peritonitis episodes/patient‐month	1/ 11 versus 1/11	1/12 versus 1/10	RD 0.01 (‐0.24 to 0.26)
Height SDS	‐2.26 (‐4.5 to ‐1.61) versus ‐2.33 (‐4.27 to ‐1.43) (end)	Not reported	Not calculated

Outcome	Ardissino 2000	Klaus 1995	Schmitt 2003	Summary estimate ^a
% fall PTH	13.7 ± 46.7^b versus 19.2 ± 57.8^c	Not reported	58 ± 26 versus 64 ± 22	MD ‐5.83 (‐21.49 to 9.83)
Number with fall in PTH	21/30 versus 23/29	"median PTH fell in both groups"	Not reported	RR 0.88 (0.65 to 1.19)
Mean integrated PTH (pg/mL)	Not reported	Not reported	285 ± 213 versus 343 ± 171	MD ‐58.00 (‐212.55 to 96.55)
CrCl (mL/min/1.73 m²)	22 ± 13 versus 22 ± 11(end)	Not reported	‐1.6 ± 4.0 versus 3.1± 4.8 (change)	MD 0.50 (‐5.72 to 6.72)
Change in height SDS	Not reported	Not reported	‐0.05 ± 0.52 versus ‐0.18 ± 0.34	MD 0.13 (‐0.22 to 0.48)
Hypercalcaemic patients	1/30 versus 1/29	2/12 versus 3/9	No difference in number of episodes	RD ‐0.02 (‐0.17 to 0.13)
Hypophosphataemic patients	0/30 versus 1/29	Not reported	No difference in number of episodes	RD 0.03 (‐0.06 to 0.12)

Outcomes	Eke 1983	Greenbaum 2005	Greenbaum 2007	Watson 1988	Summary estimate^a
Number with abnormal bone histology	7/8^b versus 2/7^c	Not reported	Not reported	Reduced osteoid, bone formation and  resorption in 1αOHD versus control	RR 0.17 (0.03 to 1.16)
PTH (pmol/L)	No difference	Not reported	Not reported	18 ± 15 versus 73 ± 32	MD ‐55.00 (‐83.03 to ‐26.97)
Elevated PTH	Not reported	Not reported	Not reported	1/6 versus 6/6	RR 0.17 (0.03 to 1.00)
Number with 30% fall in PTH	Not reported	11/21 versus 5/26	9/15 versus 3/14	Not reported	RR 2.75 (1.39 to 5.47)
Change in PTH (pg/mL)	Not reported	‐193 ± 637 versus 10 ± 347	‐164 versus +238	Not reported	MD ‐203.00 (‐506.34 to 100.34)
Hypercalcaemic patients	0/8 versus 0/7	5/21 versus 0/26	0/15 versus 0/14	3/6 versus 2/6	RD 0.08 (‐0.08 to 0.24)
Ca x P > 7.5 mg²/dL²	Not reported	8/21 versus 1/26	Not reported	Not reported	RD 0.34 (0.12 to 0.56)
Phosphorus > 6.5 mg/dL x 2	Not reported	15/21 versus 12/26	Not reported	Not reported	RD 0.25 (‐0.02 to 0.52)
Change in serum calcium (mg/dL)	No difference	0.38 ± 0.6 versus 0.08 ± 0.66	0.00 ± 1.01 versus 0.29 ± 1.01	No difference	MD ‐0.10 (‐0.45 to 0.65)
Change in Ca x P (mg²/dL²)	Not reported	4.7 ± 12.4 versus 0.6 ± 14.8	‐1.35 ± 14.1 versus 3.21 ± 14.1	Not reported	MD ‐0.45 (‐7.94 to 8.83)
Change in serum phosphorus (mg/dL)	No difference	0.23 ± 1.33 versus 0.00 ± 1.68	‐0.15 ± 1.35 versus 0.18 ± 1.35	No difference	MD ‐0.01 (‐0.66 to 0.63)
Change in bone ALP (U/L)	No difference	‐4.5 ± 67 versus 43.2 ± 66	Not reported	Not reported	MD ‐47.70 (‐88.54 to ‐6.86)

Outcome	GFRD Study 1990	Hodson 1985	Salusky 2005a Doxercalciferol treatment	Salusky 2005b Calcitriol treatment	Summary estimate^a
PTH (pg/mL)	Not reported	0.87 ± 0.44^b versus 1.35 ± 0.89^c	Graphical data only	Graphical data only	MD ‐0.48 (‐1.23 to 0.27)
Number with Improved histology	Not reported	7/8 versus 4/7	Not reported	Not reported	RR 1.53 (0.77 to 3.06)
Height velocity ≥ expected	Positive growth scores with both treatments	1/8 versus 4/7	Not reported	Not reported	RR 0.22 (0.03 to 1.53)
Hypercalcaemic patients	No difference in number of episodes	6/8 versus 3/7	17 in sevelamer treated^b versus  48 episodes in CaCO₃ treated^c	17 in sevelamer treated^b versus  48 episodes in CaCO₃ treated^c	RR 1.75 (0.68 to 4.50)
GFR (mL/min/1.73 m²)	Fell in both groups	Not reported	Peritoneal dialysis patients	Peritoneal dialysis patients	Not calculated
Calcium (mmol/L)	Not reported	2.63 ± 0.10 versus 2.45 ± 0.20	Graphical data only	Graphical data only	MD 0.18 (0.01 to 0.35)
Phosphorus (mmol/L)	Not reported	1.59 ± 0.30 versus 1.93 ± 0.47	Graphical data only	Graphical data only	MD ‐0.34 (‐0.76 to 0.08)
ALP (U/L)	Not reported	169 ± 58 versus 208 ± 84	Graphical data only	Graphical data only	MD ‐39.00 (‐116.63 to 38.63)
Bone formation rate (µm³/µm²/d)	Not reported	Not reported	60 ± 97 versus 53 ± 100	62 ± 201 versus 45 ± 73	MD 10.29 (‐53.07 to 73.65)
Osteoid volume (%)	Not reported	Not reported	11.1 ± 19.8 versus 10.8 ± 10.4	8.2 ± 20.1 versus 8.9 ± 14.3	MD ‐0.16 (‐9.16 to 8.83)
Eroded bone surface (%)	Not reported	Not reported	7.9 ± 9.7 versus 7.3 ± 15.6	8.6 ± 12.8 versus 7.7 ± 11.6	MD 0.76 (‐6.16 to 7.71)
Bone volume (%)	Not reported	Not reported	29.9 ± 17.3 versus 28.2 ± 15.9	30.3 ± 44.5 versus 26 ± 24.9	MD 2.19 (‐9.12 to 13.91)

Outcome	Mak 1985^a	Salusky 1991	Summary estimate^b
Abnormal bone biopsy	4 improved, 5 no change, 2 worse	3/10 versus 6/7	RR 0.35 (0.13 to 0.95)
Bone aluminium (mg/kg)	Not reported	9 ± 9.5 versus 10 ± 13	MD ‐1.00 (‐12.29 to 10.29)
PTH levels (mL‐eq/L)	Normalised with both treatments	731 ± 1062^c versus 918 ± 833^d	MD ‐‐187.00 (‐1089.25 to 715.25)
ALP (IU/L)	Not reported	259 ± 231 versus 328 ± 256	MD 21.00 (‐216.62 to 258.62)
Hypercalcaemic patients	Not reported	6/10 versus 2/7	RD 0.31 (‐0.14 to 0.77)
GFR (mL/min/1.73 m²)	21 (SE 4; beginning) versus 23 (SE 5; end)	Peritoneal dialysis patients	Not calculated
Height SDS	Not reported	‐2.73 ± 1.55 versus ‐1.92 ± 1.33 (end)	MD ‐0.86 (‐2.44 to 0.52)
Growth velocity SDS for chronological age	‐0.82 (SE 0.31) increased to +0.31 (SE 0.39) P < 0.01	Not reported	Not calculated
Plasma aluminium (µ/dL)	3.0 ± 2.1 versus 9.0 ± 4.5	Not reported	Not calculated

Outcome	Gulati 2010	Pieper 2006^a	Salusky 2005	Summary estimate^b
PTH (pg/mL)	165 ± 45 versus 144 ± 34 (final)	‐7 ± 196 versus ‐43 ± 270 (change)	562 ± 403^c versus 369 ± 344^d (final)	MD 51.92 (‐77.53 to 181.36)
ALP (U/L)	1352 ± 500 versus1018 ± 557 (final)	70 ± 108 versus 34 ± 127 (change)	254 ± 155 versus 220 ± 157 (final)	MD 90.48 (‐139.38 to 320.35)
Hypercalcaemic patients	0/10 versus 0/9	1/32 versus 6/30 periods of calcitriol	5 episodes versus 22 episodes	Not calculated
Ca x P (mg²/dL²)	51.2 ± 8.4 versus 49.4 ± 12.9	‐1.37 ± 1.4 versus ‐1.12 ± 1.25 (change)	51 ± 7.4 versus 53 ± 7.5	MD ‐1.12 (‐5.9 to 3.64)
Calcium (mg/dL)	8.6 ± 1 versus 8.5 ± 1	‐0.2 ± 0.7 versus 0.4 ± 1.0 (change)	8.9 ± 0.8 versus 9.6 ± 0.4	MD ‐0.4 (‐1.16 to ‐0.36)
Phosphorus (mg/dL)	6.2 ± 0.4 versus 5.8 ± 1.7	‐1.5 ± 1.6 versus ‐1.7 ± 1.7 (change)	5.6 ± 1.2 versus 5.5 ± 0.4	MD 0.17 (‐0.37 to 0.71)

Date	Event	Description
2 November 2015	New citation required but conclusions have not changed	No change to conclusions
2 November 2015	New search has been performed	New search; 3 new studies included
8 September 2015	Amended	Search strategies revised

Database	Search terms
CENTRAL	(kidney disease or kidney failure or renal disease or renal failure):ti,ab,kw renal insufficiency:ti,ab,kw (CKD or CKF or CRD or CRF or CRI or ESKD or ESRD or ESKF or ESRF):ti,ab renal next replacement next therap:ti,ab,kw (predialysis or dialysis):ti,ab,kw (hemodialysis or haemodialysis):ti,ab,kw (CAPD or CCPD or APD):ti,ab,kw {or #1‐#7} bone disease:ti,ab,kw (osteo* or hyperparathyroid* or hyperphosphatm):ti,ab,kw MeSH descriptor: [Bone Diseases] this term only MeSH descriptor: [Bone Diseases, Metabolic] explode all trees MeSH descriptor: [Hyperparathyroidism, Secondary] explode all trees {or #9‐#13} {and #8, #14}
MEDLINE	Bone Diseases/ exp Bone Diseases, Metabolic/ exp Hyperparathyroidism, Secondary/ bone disease$.tw. (osteo$ or hyperparathyroid$ or hyperphosphat$).tw. or/1‐5 Renal Replacement Therapy/ exp Renal Dialysis/ (hemodialysis or haemodialysis).tw. (hemofiltration or haemofiltration).tw. (hemodiafiltration or haemodiafiltration).tw. dialysis.tw. (CAPD or CCPD or APD).tw. Renal Insufficiency/ exp Renal Insufficiency, Chronic/ Kidney Diseases/ (end‐stage renal or end‐stage kidney or endstage renal or endstage kidney).tw. (ESRF or ESKF or ESRD or ESKD).tw. (chronic kidney or chronic renal).tw. (CKF or CKD or CRF or CRD).tw. (predialysis or pre‐dialysis).tw. or/7‐21 and/6,22
EMBASE	exp metabolic bone disease/ bone disease/ secondary hyperparathyroidism/ (osteo$ or hyperparathyroid$ or hyperphosphat$).tw. bone disease$.tw. or/1‐5 exp Renal Replacement Therapy/ (hemodialysis or haemodialysis).tw. (hemofiltration or haemofiltration).tw. (hemodiafiltration or haemodiafiltration).tw. dialysis.tw. (CAPD or CCPD or APD).tw. Kidney Disease/ Chronic Kidney Disease/ Kidney Failure/ Chronic Kidney Failure/ (chronic kidney or chronic renal).tw. (CKF or CKD or CRF or CRD).tw. (end‐stage renal or end‐stage kidney or endstage renal or endstage kidney).tw. (ESRF or ESKF or ESRD or ESKD).tw. (predialysis or pre‐dialysis).tw. or/7‐21 exp Child/ exp newborn/ Adolescent/ child$.tw. (pediatr$ or paediatr$).tw. infant$.tw. adolescen$.tw. or/23‐29 and/6,22,30

Potential source of bias	Assessment criteria
Was there adequate sequence generation?	Yes (low risk of bias): Random number table; computer random number generator; coin tossing; shuffling cards or envelopes; throwing dice; drawing of lots; minimization (minimization may be implemented without a random element, and this is considered to be equivalent to being random).
	No (high risk of bias): Sequence generated by odd or even date of birth; date (or day) of admission; sequence generated by hospital or clinic record number; allocation by judgement of the clinician; by preference of the participant; based on the results of a laboratory test or a series of tests; by availability of the intervention.
	Unclear: Insufficient information about the sequence generation process to permit judgement.
Was allocation adequately concealed?	Yes (low risk of bias): Randomisation method described that would not allow investigator/participant to know or influence intervention group before eligible participant entered in the study (e.g. central allocation, including telephone, web‐based, and pharmacy‐controlled, randomisation; sequentially numbered drug containers of identical appearance; sequentially numbered, opaque, sealed envelopes).
	No (high risk of bias): Using an open random allocation schedule (e.g. a list of random numbers); assignment envelopes were used without appropriate safeguards (e.g. if envelopes were unsealed or non‐opaque or not sequentially numbered); alternation or rotation; date of birth; case record number; any other explicitly unconcealed procedure.
	Unclear: Randomisation stated but no information on method used is available.
Was knowledge of the allocated interventions adequately prevented during the study?	Yes (low risk of bias): No blinding, but the review authors judge that the outcome and the outcome measurement are not likely to be influenced by lack of blinding; blinding of participants and key study personnel ensured, and unlikely that the blinding could have been broken; either participants or some key study personnel were not blinded, but outcome assessment was blinded and the non‐blinding of others unlikely to introduce bias.
	No (high risk of bias): No blinding or incomplete blinding, and the outcome or outcome measurement is likely to be influenced by lack of blinding; blinding of key study participants and personnel attempted, but likely that the blinding could have been broken; either participants or some key study personnel were not blinded, and the non‐blinding of others likely to introduce bias.
	Unclear: Insufficient information to permit judgement of ‘Yes’ or ‘No'
Were incomplete outcome data adequately addressed?	Yes (low risk of bias): No missing outcome data; reasons for missing outcome data unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias); missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes not enough to have a clinically relevant impact on observed effect size; missing data have been imputed using appropriate methods.
	No (high risk of bias): Reason for missing outcome data likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk enough to induce clinically relevant bias in intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes enough to induce clinically relevant bias in observed effect size; ‘as‐treated’ analysis done with substantial departure of the intervention received from that assigned at randomisation; potentially inappropriate application of simple imputation.
	Unclear: Insufficient information to permit judgement of ‘Yes’ or ‘No'.
Are reports of the study free of suggestion of selective outcome reporting?	Yes (low risk of bias): The study protocol is available and all of the study’s pre‐specified (primary and secondary) outcomes that are of interest in the review have been reported in the pre‐specified way; the study protocol is not available but it is clear that the published reports include all expected outcomes, including those that were pre‐specified (convincing text of this nature may be uncommon).
	No (high risk of bias): Not all of the study’s pre‐specified primary outcomes have been reported; one or more primary outcomes is reported using measurements, analysis methods or subsets of the data (e.g. subscales) that were not pre‐specified; one or more reported primary outcomes were not pre‐specified (unless clear justification for their reporting is provided, such as an unexpected adverse effect); one or more outcomes of interest in the review are reported incompletely so that they cannot be entered in a meta‐analysis; the study report fails to include results for a key outcome that would be expected to have been reported for such a study.
	Unclear: Insufficient information to permit judgement of ‘Yes’ or ‘No'.
Was the study apparently free of other problems that could put it at a risk of bias?	Yes (low risk of bias): The study appears to be free of other sources of bias.
	No (high risk of bias): Had a potential source of bias related to the specific study design used; stopped early due to some data‐dependent process (including a formal‐stopping rule); had extreme baseline imbalance; has been claimed to have been fraudulent; had some other problem.
	Unclear: Insufficient information to permit judgement of ‘Yes’ or ‘No'.

Outcome or subgroup title	No. of studies	Statistical method	Effect size
1 Bone disease on bone histology	1	Risk Ratio (M‐H, Random, 95% CI)	Totals not selected
1.1 Abnormal bone histology	1	Risk Ratio (M‐H, Random, 95% CI)	0.0 [0.0, 0.0]
1.2 Adynamic bone disease	1	Risk Ratio (M‐H, Random, 95% CI)	0.0 [0.0, 0.0]
1.3 Osteitis fibrosa	1	Risk Ratio (M‐H, Random, 95% CI)	0.0 [0.0, 0.0]
1.4 Mixed or mild bone disease	1	Risk Ratio (M‐H, Random, 95% CI)	0.0 [0.0, 0.0]
1.5 Normal or reduced bone formation	1	Risk Ratio (M‐H, Random, 95% CI)	0.0 [0.0, 0.0]
2 Bone formation rate	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
3 PTH	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
4 Maximum calcium level	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
5 Adverse events	1	Risk Difference (M‐H, Random, 95% CI)	Totals not selected
5.1 Calcium > 11 mg/dL	1	Risk Difference (M‐H, Random, 95% CI)	0.0 [0.0, 0.0]
5.2 Phosphate > 7.0 mg/dL	1	Risk Difference (M‐H, Random, 95% CI)	0.0 [0.0, 0.0]
6 Peritonitis episodes/patient‐months	1	Risk Difference (M‐H, Random, 95% CI)	Totals not selected

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Change in height SDS at 12 months	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
2 Fall in PTH	2		Mean Difference (IV, Random, 95% CI)	Totals not selected
2.1 8 weeks	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
2.2 12 months	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
3 Number with fall in PTH at 8 weeks	1		Risk Ratio (M‐H, Random, 95% CI)	Totals not selected
4 Mean integrated PTH at 12 months	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
5 Kidney function	2		Mean Difference (IV, Random, 95% CI)	Totals not selected
5.1 Final CrCl at 8 weeks [mL/min/1.73 m²]	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
5.2 Change in CrCl at 12 months [mL/min/1.73 m²]	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6 Adverse events	2		Risk Difference (M‐H, Random, 95% CI)	Subtotals only
6.1 Calcium > 11.5 mg/dL	2	80	Risk Difference (M‐H, Random, 95% CI)	‐0.02 [‐0.17, 0.13]
6.2 Phosphorus > 7.5 mg/dL	1	59	Risk Difference (M‐H, Random, 95% CI)	0.03 [‐0.06, 0.12]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Abnormal bone histology	1		Risk Ratio (M‐H, Random, 95% CI)	Totals not selected
2 Elevated PTH	1		Risk Ratio (M‐H, Random, 95% CI)	Totals not selected
3 PTH	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
4 Number with 30% fall in PTH on two occasions	2	76	Risk Ratio (M‐H, Random, 95% CI)	2.75 [1.39, 5.47]
5 Change in PTH with IV calcitriol	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
6 Hypercalcaemia	4	103	Risk Difference (M‐H, Random, 95% CI)	0.08 [‐0.08, 0.24]
7 Calcium‐phosphorus product > 7.5 mg²/dL²	1		Risk Difference (M‐H, Random, 95% CI)	Totals not selected
8 Phosphorus > 6.5 mg/dL	1		Risk Difference (M‐H, Random, 95% CI)	Totals not selected
9 Change in biochemical values	2		Mean Difference (IV, Random, 95% CI)	Subtotals only
9.1 Calcium [mg/dL]	2	76	Mean Difference (IV, Random, 95% CI)	0.10 [‐0.45, 0.65]
9.2 Calcium‐phosphorus product [mg²/dL²]	2	76	Mean Difference (IV, Random, 95% CI)	0.45 [‐7.94, 8.83]
9.3 Phosphorus [mg/dL]	2	76	Mean Difference (IV, Random, 95% CI)	‐0.01 [‐0.66, 0.63]
9.4 Bone ALP [µg/L]	1	41	Mean Difference (IV, Random, 95% CI)	‐47.7 [‐88.54, ‐6.86]

Outcome or subgroup title	No. of studies	Statistical method	Effect size
1 Height velocity ≥ expected	1	Risk Ratio (M‐H, Random, 95% CI)	Totals not selected
2 Improved bone histology	1	Risk Ratio (M‐H, Random, 95% CI)	Totals not selected
3 PTH	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
4 Hypercalcaemia	1	Risk Ratio (M‐H, Random, 95% CI)	Totals not selected
5 Biochemical values	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
5.1 Calcium [mmol/L]	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
5.2 Phosphorus [mmol/L]	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
5.3 ALP [U/L]	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Number developing hyperparathyroidism	1		Risk Ratio (M‐H, Random, 95% CI)	Totals not selected
2 PTH	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
3 Calcium	2	60	Mean Difference (IV, Random, 95% CI)	0.26 [‐0.28, 0.81]
4 Phosphorus	2	60	Mean Difference (IV, Random, 95% CI)	‐0.29 [‐0.96, 0.39]
5 End 1.25(OH)D3 levels	1		Mean Difference (IV, Random, 95% CI)	Totals not selected

Outcome or subgroup title	No. of studies	Statistical method	Effect size
1 Height SDS	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
2 Abnormal bone biopsy	1	Risk Ratio (M‐H, Random, 95% CI)	Totals not selected
3 Bone aluminium	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
4 PTH	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
5 ALP	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
6 Hypercalcaemia	1	Risk Difference (M‐H, Random, 95% CI)	Totals not selected

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 PTH	2	48	Mean Difference (IV, Random, 95% CI)	51.92 [‐77.53, 181.36]
2 ALP	2	48	Mean Difference (IV, Random, 95% CI)	90.48 [‐139.38, 320.35]
3 Calcium‐phosphorus product	2	48	Mean Difference (IV, Random, 95% CI)	‐1.12 [‐5.88, 3.64]
4 Calcium	2	48	Mean Difference (IV, Random, 95% CI)	‐0.40 [‐1.16, 0.36]
5 Phosphorus	2	48	Mean Difference (IV, Random, 95% CI)	0.17 [‐0.37, 0.71]
6 Bone histomorphometry	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
6.1 Bone formation rate [µm³/mm²/d]	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6.2 Fall in bone formation rate [%]	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6.3 Eroded perimeter [%]	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6.4 Osteoid area [%]	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6.5 Osteoid perimeter [%]	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6.6 Osteoid seam width [µm]	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6.7 Bone area [%]	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Bone histomorphometry	2		Mean Difference (IV, Random, 95% CI)	Subtotals only
1.1 Bone formation rate [µm³/µm²/d]	2	51	Mean Difference (IV, Random, 95% CI)	10.29 [‐53.07, 73.65]
1.2 Eroded bone surface [%]	2	51	Mean Difference (IV, Random, 95% CI)	0.76 [‐6.19, 7.71]
1.3 Osteoid volume [%]	2	51	Mean Difference (IV, Random, 95% CI)	‐0.16 [‐9.16, 8.83]
1.4 Osteoid surface [%]	2	51	Mean Difference (IV, Random, 95% CI)	1.04 [‐29.63, 31.71]
1.5 Osteoid maturation time [days]	2	51	Mean Difference (IV, Random, 95% CI)	0.82 [‐14.41, 16.05]
1.6 Bone volume [%]	2	51	Mean Difference (IV, Random, 95% CI)	2.19 [‐9.52, 13.91]

Methods	Study design: parallel RCT Time frame: not reported Follow‐up period: 8 weeks Loss to follow‐up/excluded: unclear; possibly 31%. 85 enrolled (ASN 1998 abstract) but only 59 included in 8 week analysis
Participants	Country: Europe Setting: international eGFR < 75 mL/min/1.73 m²; PTH > 70 ng/mL Number (treatment/control): 30/29 Mean age ± SD (years): treatment group (8.3 ± 4.8); control group (8.6 ± 4.7) Sex (M/F): treatment group (21/9); control group (24/5) Exclusion criteria: dialysis patients; patients on steroids; rhGH or with other disease; Ca < 8.5 mg/dL or > 11.5 mg/dL or K < 3.5 mg/dL or > 7.5 mg/dL
Interventions	Treatment group Oral calcitriol: 35 ng/kg twice a week (70 ng/kg/wk) for 8 weeks Control group Daily oral calcitriol: 10 ng/kg/d (70 ng/kg/wk) for 8 weeks Co‐interventions Phosphate binders
Outcomes	Median (range) in PTH values at end of treatment % change from baseline in PTH levels Number with fall in PTH levels Mean CrCl at end of treatment Number with hypercalcaemia, hyperphosphataemia Serum Ca x P
Notes	Diagnosis of renal bone disease: secondary hyperparathyroidism with PTH level > 70 pg/mL ALP and PTH levels presented as median and range Data on serum Ca x P graphical only
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Central randomisation stratified for PTH levels 70 to 399 ng/mL and > 399 ng/mL
Allocation concealment (selection bias)	Low risk	Central randomisation stratified for PTH levels 70 to 399 ng/mL and > 399 ng/mL
Blinding of participants and personnel (performance bias)   All outcomes	High risk	No blinding. Lack of blinding could influence management
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Primary outcome was laboratory based and unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias)   All outcomes	High risk	In the full publication, all 59 reported patients completed study in full publication. However as abstract reports that 85 patients were enrolled in the study.
Selective reporting (reporting bias)	Low risk	Data available on PTH, GFR & hypercalcaemia for meta‐analysis
Other bias	Low risk	No evidence of other bias

Methods	Study design: open‐label RCT Time frame: April 2006 to July 2007 Follow‐up period: 12 weeks after 2 to 6 weeks washout Loss to follow‐up: Three (14%) of 22 lost to follow‐up
Participants	Country: India Setting: Single centre Ages 2 to 18 years, with CKD stages 3 and 4 (eGFR between 30 to 59 and 15 to 29 mL/min/1.73 m²); Serum phosphate > 5.5 mg/dL Number (randomised/analysed): treatment group (11/10); control group (11/9) Mean age ± SD (years): treatment group (9.6 ± 4.8); control group (10.6 ± 4.5) Sex (M/F): treatment group (6/5); control group (5/6) Exclusion criteria: corrected Ca < 7 mg/dL or > 11 mg/dL; haemoglobin < 6 g/dL; prolonged prothrombin time (INR > 1.5); residing > 100 km from hospital and patients not willing to come for monthly follow‐up visits
Interventions	Treatment group Sevelamer: 400 mg 3 times/d Dose titrated at 4th and 8th week according to serum phosphate levels Control group Calcium acetate: 667 mg 3 times/d Dose titrated at 4th and 8th week according to serum phosphate levels Co‐interventions Patients with hypocalcaemia ≤ 7 mg/dL were treated with oral calcium carbonate
Outcomes	Decrease in blood phosphate level Changes in Ca, Ca x P, ALP and PTH
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer generated randomisation
Allocation concealment (selection bias)	Low risk	Computer generated randomisation and opaque sealed envelopes
Blinding of participants and personnel (performance bias)   All outcomes	High risk	No blinding and lack of blinding may influence management
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Open label but primary outcome was laboratory based
Incomplete outcome data (attrition bias)   All outcomes	Low risk	3 (14%) of 22 lost to follow‐up after randomisation
Selective reporting (reporting bias)	Low risk	All relevant laboratory outcomes and adverse events included
Other bias	Low risk	Emcure Pharmaceuticals Limited provided medication

Methods	Study design: cross‐over RCT Time frame: 1993 to 1994 Follow‐up period: 7 months including 1 month run‐in Loss to follow‐up/exclusions: 0% (0/7)
Participants	Country: Canada Setting: Single centre CCPD or CAPD Number: 7 Mean age ± SD: 7.2 ± 5.2 years Sex (M/F): 5/2 Exclusion criteria: not reported
Interventions	Treatment/control group IP or oral calcitriol 0.01‐0.02 µg/kg/d for 3 months, then crossed over for 3 months Co‐interventions CaCO₃ as phosphate binder, Dietary phosphorus restriction
Outcomes	Height SDS (mean and range) Bone X‐rays PTH levels Number with hypercalcaemia Number with peritonitis
Notes	Two phases of cross‐over combined in results Most data graphical only for individual patients
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not reported
Allocation concealment (selection bias)	Unclear risk	Said to be random assignment
Blinding of participants and personnel (performance bias)   All outcomes	High risk	No blinding and lack of blinding could influence patient management
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Outcomes laboratory based
Incomplete outcome data (attrition bias)   All outcomes	Low risk	All patients completed study
Selective reporting (reporting bias)	High risk	No data available in form that can be included in meta‐analyses as cross over study
Other bias	High risk	Supported by Abbott Laboratories

Methods	Study design: cross‐over study Time frame: 1983 to 1984 Follow‐up period: 15 months including 3 months run‐in Loss to follow‐up/excluded: 0% (0/12)
Participants	Country: UK Setting: Single centre Moderate‐severe CKD (GFR 8 to 45 mL/min/1.73 m²) Number: 12 Mean age ± SD: 10.4 ± 3.1 years Sex (M/F): 7/5 Exclusion criteria: not reported
Interventions	Treatment/control group CaCO₃ or AlOH₃ for 6 months as sole phosphate binder. Crossed over after 6 months Dihydrotachysterol continued at same dose throughout Co‐interventions Dietary restriction of phosphorus
Outcomes	Bone biopsy pre‐study and 12 months Change GFR over 12 months PTH values Ca, P, % theoretical reabsorption of phosphate values Growth (HVSDS) pre‐study and 12 months
Notes	Little comparative data available Most data graphical only
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information provided
Allocation concealment (selection bias)	Unclear risk	Patients said to be randomised
Blinding of participants and personnel (performance bias)   All outcomes	High risk	No blinding and lack of blinding could influence patient management
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Outcomes are laboratory based and unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias)   All outcomes	Low risk	All patients appear to be included
Selective reporting (reporting bias)	High risk	Only combined data provided for both treatment groups so no data available for meta‐analyses
Other bias	Low risk	National Medical Research Fund

PERMALINK

Interventions for metabolic bone disease in children with chronic kidney disease

Deirdre Hahn

Elisabeth M Hodson

Jonathan C Craig

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Background

Description of the condition

Description of the intervention

How the intervention might work

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Inclusion criteria

Exclusion criteria

Types of interventions

Types of outcome measures

Primary outcomes

Patient‐centred outcome measures

Secondary outcomes

Patient‐centred outcome measures

Surrogate outcomes

Adverse events

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Selection of studies

Data extraction and management

Assessment of risk of bias in included studies

Measures of treatment effect

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

Results

Description of studies

Results of the search

2009 review

2015 update

1.

Included studies

Intraperitoneal calcitriol versus oral calcitriol

Intermittent oral calcitriol versus daily oral calcitriol

Different vitamin D preparations versus placebo/no specific treatment

Different vitamin D preparations

Ergocalciferol versus placebo/control

Calcium carbonate versus aluminium hydroxide

Sevelamer versus calcium‐containing phosphate binders

Excluded studies

Risk of bias in included studies

2.

3.

Allocation

Blinding

Incomplete outcome data

Selective reporting

Other potential sources of bias

Effects of interventions

Intraperitoneal versus oral calcitriol

1. Intraperitoneal versus oral calcitriol in children on peritoneal dialysis.

1.1. Analysis.

1.2. Analysis.

1.3. Analysis.

1.4. Analysis.

1.5. Analysis.

Methods	Study design: RCT Time frame: not reported Follow‐up period: not reported Loss to follow‐up/excluded: none reported
Participants	Country: Thailand Setting: university teaching hospital Aged < 18 years. CKD stage 5/5D, vitamin D deficiency with levels < 30 ng/mL, Hb level 10.0 to 12.5 g/dL, serum phosphorus, 6.5 mg/dL, corrected Ca < 10.5 mg/dL, Ca x P, 65mg²/dL² one month prior to recruitment Number: treatment group (10); control group (10) Mean age ± SD (years): treatment group (7.1 ± 5.4); control group (9.3 ± 5.3) Sex (M/F): treatment group (7/3); control group (6/4) Exclusion criteria: thalassaemia; chronic liver disease; gastrointestinal malabsorption; significant blood loss; PTH > 800 pg/mL; proteinuria > 2 mg/mg of urine creatinine; blood transfusion; chronic anticonvulsant therapy; prior ergocalciferol supplementation and kidney transplantation
Interventions	Treatment group Severe vitamin D deficiency (serum 25D level < 5 ng/mL) 40,000 IU ergocalciferol weekly for 4 weeks, then 40,000 IU every second week for 8 weeks (total 320,000 IU ergocalciferol) Mild deficiency (serum 25D level 5 to 15 ng/mL) 40,000 IU ergocalciferol every second week for 12 weeks (total 240,000 IU ergocalciferol) For 25D insufficiency (serum 25D levels 16 to 30 ng/mL) 40,000 IU ergocalciferol every 4 weeks for 12 weeks (total 120,000 IU ergocalciferol) Control group No specific therapy Co‐interventions Calcium carbonate for phosphate binding with phosphorus level > 5.5 mg/dL Alfacalcidol for secondary hyperparathyroidism Anaemia management: epoetin alfa administered subcutaneously
Outcomes	Ca levels PTH levels Phosphorous levels
Notes	Primary outcome was the effect of ergocalciferol on ESA dose
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Simple randomisation, randomisation was sequentially done (information from authors)
Allocation concealment (selection bias)	High risk	Simple randomisation, randomisation was sequentially done (information from authors)
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Not blinded and lack of blinding could affect patient management
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Laboratory outcome and unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias)   All outcomes	Low risk	All patients completed 12 week therapy
Selective reporting (reporting bias)	Low risk	All prespecified outcomes mentioned
Other bias	Unclear risk	Insufficient information to permit judgement

Methods	Study design: RCT Time frame: 1988 to 1991 Follow‐up period: 13 ± 2 months Loss to follow‐up/exclusions: 23% (5/22)
Participants	Country: USA Setting: university teaching hospital Maintenance PD Number (randomised/analysed): treatment group (12/10); control group (10/7) Mean age ± SD (years): treatment group (15.5 ± 3.7); control group (14.1 ± 3.7) Sex (M/F): treatment group (2/8); control group (3/4) Exclusion criteria: aluminium‐related bone disease
Interventions	Treatment group CaCO₃ to maintain phosphate < 2.0 mmol/L (2.5 to 12 g/d) for 1 year Control group AlOH₃ to maintain phosphate < 2.0 mmol/L (maximum dose 30 mg/kg) for 1 year Co‐interventions Calcitriol to maintain Ca 2.6 to 2.8 mmol/L
Outcomes	Bone histology PTH, ALP, Ca, P levels Plasma aluminium levels Number of patients with Ca > 2.8mmol/L
Notes	Graphical data only for Ca and P levels
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Said to be randomly assigned
Allocation concealment (selection bias)	Unclear risk	Said to be randomly assigned
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Not blinded and lack of blinding could affect patient management
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Outcomes are laboratory based and unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias)   All outcomes	High risk	5 withdrawn for transplant, i.e. 23% did not complete study
Selective reporting (reporting bias)	Low risk	All outcomes reported
Other bias	Low risk	Grants from US Public Health Service and Department of Veterans Affairs

Methods	Study design: RCT Time frame: 1996 to 1998 Follow‐up period: 12 months Loss to follow‐up/excluded: 28%, 13/46 did not complete study. Unclear whether exclusions pre‐ or post‐randomisation
Participants	Country: USA Setting: University teaching hospital CCPD for > 2 months; bone biopsy normal or hyperparathyroidism Number: treatment group (16); control group (17) Mean age ± SD (years): treatment group (12.5 ± 1.1); control group (13.2 ± 1.3) Sex (M/F): not reported Exclusion criteria: low turnover bone disease; parathyroidectomy in prior 12 months; immunosuppression; documented non‐compliance
Interventions	Treatment group Intraperitoneal calcitriol 1.0 µg 3 times/week for 12 months Control group Oral calcitriol 1.0 µg 3 times/week for 12 months Co‐interventions CaCO₃ for phosphate binder
Outcomes	Bone histology Lowest PTH levels, maximum Ca levels, P and ALP levels Number with Ca >11 mg/dL, P > 7.0 mg/dL Peritonitis
Notes	P and ALP data only available graphically
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	3 randomisation schedules according to bone histology. Blocks of 4, 6, 8 with block size determined at random
Allocation concealment (selection bias)	Low risk	Independent biostatistician
Blinding of participants and personnel (performance bias)   All outcomes	High risk	No blinding and lack of blinding could influence patient management
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Outcomes laboratory based and unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	13 (28%) of 46 did not complete study and unclear whether these excluded pre‐ or post‐randomisation
Selective reporting (reporting bias)	Low risk	All expected outcomes reported
Other bias	Low risk	Supported by grants from USPHS & Casey Lee Ball Foundation

Methods	Study titles Salusky 2005, Salusky 2005a and Salusky 2005b represent the same RCT. The study (Salusky 2005) presents the data comparing sevelamer with calcium carbonate (Comparison 7) irrespective of vitamin D preparation. Data reported on 29 patients of 42 allocated to study Study design: parallel RCT, 2 x 2 longitudinal factorial study design Time frame: 2003 to 2005 Follow‐up period: 8 months Loss to follow‐up/excluded: 13/42 did not complete the study
Participants	Country: USA Setting: University teaching hospital CCPD, PTH > 400 pg/mL; bone histomorphometry of secondary hyperparathyroidism Number (randomised/analysed): treatment group 1 (21/14); treatment group 2 (21/15) Mean age ± SD (years): treatment group 1 (11 ± 5); treatment group 2 (15 ± 3) Sex (M/F): treatment group 1 (10/4); treatment group 2 (8/7) Exclusion criteria: previous history of poor compliance; parathyroidectomy in last 12 months; immunosuppressive agents; rhGH
Interventions	Treatment group 1 CaCO₃: titrated to keep P at 4 to 6 mg/dL for 8 months Doxercalciferol or calcitriol given 3 times/wk. Initial dose depended on PTH level, then titrated to keep PTH at 300 to 400 pg/mL and Ca 8.4 to 10.2 mg/dL Treatment group 2 Sevelamer: initial dose extrapolated from previous calcium carbonate doses; then titrated to keep P at 4 to 6 mg/dL. Continued for 8 months Doxercalciferol or calcitriol given 3 times/week. Initial dose depended on PTH level, then titrated to keep PTH at 300 to 400 pg/mL and Ca 8.4 to 10.2 mg/dL Co‐interventions 1000 mg oral calcium in sevelamer group if Ca < 8.2 mg/dL
Outcomes	Outcomes for comparison between phosphate binders Primary outcome: Bone formation rate Other bone histomorphometric parameters Final levels of ALP, PTH Average P, Ca x P, Ca
Notes	Factorial analysis provided no evidence of treatment interaction between two sterols so comparisons reported for phosphate binders irrespective of D sterol given 2005 report included 42 allocated patients with data on 29 (14 receiving CaCO₃ and 15 receiving sevelamer)
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomised longitudinal factorial study. Computer generated randomisation
Allocation concealment (selection bias)	Low risk	Computer generated, allocated by statistician
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Participants, care givers, investigators not blinded to interventions
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Outcomes laboratory based and unlikely to be influenced by lack of blinding. Bone histology performed by scientist, who was blinded to treatment group
Incomplete outcome data (attrition bias)   All outcomes	High risk	13/42 (31%) did not complete study. Lack of data on these patients could have influenced results
Selective reporting (reporting bias)	High risk	Primary outcome was bone histology; ; secondary outcomes only reported graphically
Other bias	Low risk	USPH grants and Casey Lee Ball Foundation but Bone Care International provided medications and other unrestricted support for the study

Methods	Study design: RCT; data from a subset of prepubertal participants with GFR < 40 mL/min/1.73 m² included in Ardissino 2000 Time frame: 1998 to 2003 Follow‐up period: 12 months Loss to follow‐up/excluded post randomisation: 17% (5/29)
Participants	Country: Europe Setting: International, multicentre study eGFR < 40 mL/min/1.73 m²; PTH > 70 pg/mL Number (randomised/analysed): treatment group (14/12); control group (15/12) Mean age, range (years): treatment group (5.5, 2.4 to 8.4); control group (5.1, 1.4 to 9.1) Sex (M/F): treatment group (11/1); control group (10/2) Exclusion criteria: Ca < 8.5 mg/dL or > 11.5 mg/dL; P < 3.8 mg/dL or > 7.5 mg/dL; underlying serious disease; rhGH or corticosteroids treatment; dialysis
Interventions	Treatment group Intermittent oral calcitriol 35 ng/kg twice weekly for 12 month. After 1 month dose adjusted for PTH level Control group Daily oral calcitriol 10 ng/kg for 12 months. After 1 month dose adjusted for PTH levels Co‐interventions Phosphate binders and other medications according to clinical requirements
Outcomes	Change in height SDS Average time integrated mean plasma PTH; % maximal fall in mean PTH levels; number with reduced PTH levels Change in estimated CrCl Change in median ALP levels Serum Ca x P; number with hypercalcaemia or hyperphosphataemia
Notes	Serum Ca x P shown graphically only
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Centrally randomised according to PTH levels
Allocation concealment (selection bias)	Low risk	Centrally randomised according to PTH levels
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Not blinded and lack of blinding could influence patient management
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	not blinded but lack of blinding unlikely to influence patient management
Incomplete outcome data (attrition bias)   All outcomes	High risk	5/29 (17%) left study (RRT 4, rhGH 1). Could have influenced results
Selective reporting (reporting bias)	Low risk	Data reported on all expected outcomes
Other bias	Low risk	No evidence of other bias

Study	Reason for exclusion
Ardissino 2000a	Examining calcium absorption only
Bettinelli 1986	Ten children included and 5 ceased calcitriol for second 6 months of study. Unclear as to how patients were chosen for cessation of calcitriol
Choudhary 2014	Ineligible population; children with nephrotic syndrome and normal kidney function
El Husseini 2004	Examining osteopenia in transplant patients
Ferraris 2000	Examining corticosteroids in kidney transplant patients
Kim 2006b	Wrong population
Witmer 1976	Randomisation unclear