Description of the condition
Concerns regarding short stature in childhood and adolescence are a frequent reason for consultation with a paediatric endocrinologist. Short stature is defined as height below the third percentile or approximately two standard deviations (SDs) or more below the mean height for children of a given age, sex and population group. The Utah growth study (Lindsay 1994) assessed height and growth velocity in approximately 115,000 American children. A total of 555 children had short stature as defined previously and diminished growth rate (defined as growth velocity less than five cm annually); however only 5% had an endocrine disorder. The most frequent causes of short stature were constitutional delay of growth and pubertal development (CDGP) in addition to familial short stature. Idiopathic short stature (ISS) is a clinical description for a heterogeneous group of short children in whom no chronic disease, nutritional deficiency, psychosocial adversity, endocrinopathy, or other process is responsible for their height outcome (Cohen 2008; Ranke 1996). Children with ISS fall into two main groups: familial short stature and CDGP. The characteristics of familial short stature include bone age appropriate for chronological age, normal growth velocity, and predicted adult height appropriate to the familial pattern. In contrast, constitutional growth delay is characterised by delayed bone age, normal growth velocity, and predicted adult height appropriate to the familial pattern.
A number of interventions are currently available to increase adult height in an attempt to reduce the psychological burden attributable to short stature in childhood and adult life. Growth hormone (GH) therapy is approved for the treatment of short children with reduced adult height expectation. For individuals with constitutional delay, the use of physiologic doses of sex steroids accelerates linear growth and the onset of pubertal changes but does not improve predicted adult height (Richmond 2007).
Available evidence suggests only modest efficacy for GH treatment in children and adolescents with ISS (Bryant 2007). Treatment costs are substantial and represent an important consideration regarding allocation of medical resources. In the context of current costs of GH treatment in the United States, this corresponds to more than USD 20,000 (approx. EUR 14,774, Nov 2013 conversion rate) per centimetre in adult height gained (Lee 2006). Studies indicate that final adult height is increased with early therapy, thus the earlier the diagnosis, the better the prognosis for height. Nonetheless, most children treated with GH remain short when compared to their peers of normal stature. In GH deficiency, it has been shown that the effect of the GH dose on total pubertal growth appears to be smaller than the relative effect on growth during prepubertal years (Ranke 2003). Consequently, it is anticipated that GH treatment started during puberty has only limited effect on final height, a fact compounded by continuing skeletal maturation and progression towards epiphyseal closure.
Gonadotropin-releasing hormone analogues (GnRHas) are the standard of care for treatment of central precocious puberty (CPP) and their efficacy in increasing adult height is undisputed in early-onset CPP. Given the aforementioned limitations of GH therapy started during puberty, interruption of normal puberty in short children with GnRHa has been considered in an attempt to improve adult height. A significant number of non-randomised studies suggest that GnRHas are ineffective for this purpose (Balducci 1995; Carel 1996; Job 1994; Lanes 1998; Lindner 1993). Conversely, other non-randomised studies (Mul 2001; Pasquino 2000) and two randomised trials (Kamp 2001; Mericq 2000) suggested treatment was beneficial, although a follow-up report of the Kamp study (Van Gool 2007) at final height suggested that approximately 50% of the increase in predicted adult height obtained during the trial was lost with discontinuation of treatment, such that final height was not significantly different. A recent consensus statement proposed that routine use of GnRHa, an expensive and invasive treatment, to improve final height in short children with normally-timed puberty, could not be recommended (Carel 2009). Bone mineral density (BMD) may decrease during GnRHa therapy, although bone mass accrual returns to normal when treatment is discontinued (Pasquino 2008). Furthermore, one of the principal indications to offer treatment is because of the adverse psychosocial impact of short stature. One might argue that the psychosocial consequences of delaying puberty do not represent an acceptable trade-off in this regard.
Description of the intervention
Aromatase, encoded by the CYP19A1 gene, catalyses the rate-limiting step in the conversion of testosterone to oestradiol and androstenedione to oestrone. Use of aromatase inhibitors has been reported for an assortment of conditions in children and adolescents, including peripheral precocious puberty (Feuillan 2007; Reiter 2010), congenital adrenal hyperplasia (CAH) (Merke 2000), pubertal gynaecomastia (Mauras 2009) and short stature (Mauras 2004).
The fundamental role of oestrogen in skeletal maturation was initially recognised in the mid-1990s following reports of two young adult males in their 20s, one with an oestrogen receptor defect and another with an inactivating mutation of the aromatase gene (Morishima 1995; Smith 1994). Both were described as having tall stature (adult height up to 204 cm) and unfused epiphyses despite normal pubertal development. These observations led to use of therapies specifically targeting oestrogen to counteract its effects at the growth plate, thus prolonging growth potential. Third-generation aromatase inhibitors including anastrozole and letrozole and selective oestrogen receptor modulators (SERMs) such as tamoxifen have consequently been used as an intervention to improve height outcomes in males as outlined in a number of reports (Kreher 2005; Kumar 2009; Zhou 2005). Duration of therapy appears to influence final height, with a recent report that five-year letrozole monotherapy resulted in improvements up to 15cm on pre-treatment predicted adult height (Krebs 2012).
Oral administration and the ability to allow puberty to continue at a physiologically appropriate time make aromatase inhibitors an attractive intervention to delay skeletal maturation while augmenting height potential.
Adverse effects of the intervention
The clinical phenotype of aromatase deficiency in males, in addition to tall stature, includes osteoporosis, dyslipidaemia (elevated low-density lipoprotein (LDL), reduced high-density lipoprotein (HDL)) and insulin resistance (Carani 1997; Maffei 2004; Morishima 1995). These effects appear reversible following oestrogen supplementation. The possible association of aromatase deficiency with abnormal testicular size, low sperm count and motility has not yet been conclusively resolved. To date, seven adult males with aromatase deficiency have been described (De Ronde 2011), with varying testicular sizes ranging from micro- to macro-orchidism. Four men were infertile, but in one younger male fertility was not described. Two aromatase-deficient men had a brother who also suffered from infertility despite a normal aromatase genotype, suggesting an unrelated cause may also have contributed. Conversely, aromatase inhibitors have been used with success in adults to treat male infertility (Raman 2002; Ramasamy 2009).
Other reported side effects of aromatase inhibitors include nausea, vomiting, abdominal pain, diarrhoea, headache, erythrocytosis, arthralgia, bone pain and skin rashes. They are recommended to be avoided in people with hepatic and renal impairment.
How the intervention might work
The rate of linear growth accelerates during puberty, followed by deceleration and cessation of growth. Clinical findings in people with oestrogen insensitivity or aromatase deficiency suggest that oestrogen is essential for this change in height at puberty. Prolonging the period of growth by diminishing the biological action of oestrogen and, ultimately, by delaying the senescence of the growth plate can theoretically increase adult height. Aromatase inhibitors have been used to achieve this effect.
Why it is important to do this review
Although there have been some reviews regarding use of aromatase inhibitors in paediatrics (Cernich 2004; Shulman 2008; Wit 2012), none has used detailed systematic methodology. Current strategies for increasing adult height are expensive and require parenteral administration. Several years of therapy are required to obtain modest height increases and are of limited benefit to children who are already pubertal. Furthermore, interruption of normal puberty using GnRHa substitutes the psychological impact of short stature with that of delayed puberty. The emergence of aromatase inhibitors as a possible alternative means of delaying epiphyseal fusion and prolonging linear growth has therefore generated immense interest. Aromatase inhibitors are possibly more effective than GnRHa in promoting increased adult height in children with short stature and, unlike GnRHa, they do not delay pubertal development in males.
To assess the effects of aromatase inhibitors on both short-term growth and final height in male children and adolescents with short stature.
Criteria for considering studies for this review
Types of studies
We will include randomised controlled trials (RCTs) or quasi-RCTs which compare an aromatase inhibitor with placebo or no treatment.
Types of participants
Males with a chronological age less than or equal to 16 years with idiopathic short stature (ISS), growth hormone (GH) deficiency or constitutional delay of growth and pubertal development (CDGP) will be eligible for inclusion.
GH deficiency will be defined as evidence of peak GH less than 10 ng/mL following provocative testing using either insulin, glucagon, arginine, clonidine, growth hormone releasing hormone (GHRH) or L-dopa and propranolol.
ISS will be defined as a height two standard deviations (SDs) or more below the mean for age, sex and population group, in whom no chronic disease, nutritional deficiency, psychosocial adversity, endocrinopathy, or other condition is responsible for their height outcome.
CDGP will be defined as testicular volume less than 4 mL at age 14 years or two SDs older than the mean age of pubertal onset in the reference population, where no chronic disease, nutritional deficiency, psychosocial adversity, endocrinopathy, or other condition is responsible.
Types of interventions
Studies where an aromatase inhibitor is used as an adjunctive treatment to existing growth promoting therapies, including but not limited to growth hormone (GH) therapy, will also be included. Trials will be eligible for inclusion provided the duration of aromatase inhibitor therapy is a minimum of 12 months.
We plan to investigate the following comparisons of intervention versus control/comparator where the same letters indicate direct comparisons.
(a) Aromatase inhibitor.
(b) GH plus aromatase inhibitor.
(c) Androgen therapy plus aromatase inhibitor.
(a2) No treatment.
(b1) GH plus placebo.
(c1) Androgen therapy plus placebo.
Concomitant therapies will have to be the same in the intervention and comparator groups.
Types of outcome measures
- Final or near-final height
- Health-related quality of life
- Adverse events
- All-cause mortality
- Cognitive outcomes
- Height measures other than final or near-final height
- Bone density and morphology
- Testosterone levels
- Lipid abnormalities
- Insulin sensitivity
- Socioeconomic effects
Method and timing of outcome measurement
- Final height: defined as height attained at a bone age of greater than 16 years, when height velocity is less than one cm per year or both.
- Near-final height: defined as height attained when bone age is greater than 15 years and height velocity is less than two cm per year.
- Height measures other than final or near-final height: change in height and/or height standard deviation score (SDS), change in predicted adult height, final height SDS minus predicted adult height SDS, final height SDS minus target (mid-parental) height SDS, change in height velocity (cm/year), difference in height velocity and measured at the end of treatment.
- Health-related quality of life: measured by a validated instrument such as SF-36 or EQ-5D and measured before the study and at the end of treatment or when final height is achieved.
- All-cause mortality: defined as death from any cause during the study and measured at the end of treatment.
- Cognitive outcomes: measured by a validated instrument such as the 'Wechsler Intelligence Scale for Children' (WISC) and measured at baseline, during the study and/or at the end of treatment.
- Adverse events: includes headache, myalgia, arthralgia or erythrocytosis and measured at any point during the study or at the end of treatment.
- Bone density, bone turnover and morphology: change in BMD z-score, markers of bone turnover such as urinary collagen type 1 cross-linked N-telopeptide (NTX), morphologic assessment using magnetic resonance imaging (MRI) and measured at baseline and the end of treatment.
- Testosterone levels or testosterone:oestradiol ratio and measured at baseline, during treatment or at study completion.
- Lipid abnormalities: measurement of LDL-cholesterol, HDL-cholesterol, and triglycerides and measured at baseline, during treatment or at study completion.
- Insulin sensitivity: fasting insulin, homeostasis model of assessment-insulin resistance (HOMA-IR), euglycaemic clamp technique and measured at baseline, during treatment and/or at study completion.
- Socioeconomic effects: including cost of treatment per cm of height gained measured at final or near-final height.
'Summary of findings' table
We will present a 'Summary of findings' table reporting the following outcomes listed according to priority.
- Final or near-final height.
- Cognitive outcomes.
- All-cause mortality.
- Health-related quality of life.
- Adverse events.
- Socioeconomic effects.
Search methods for identification of studies
We will undertake a systematic literature search to identify eligible RCTs. We will search the following sources from inception to the present.
- The Cochrane Library.
We will also search trial registers including ClinicalTrials.gov (clinicaltrials.gov/), metaRegister of Controlled Trials (www.controlled-trials.com/mrct/), the EU Clinical Trials register (www.clinicaltrialsregister.eu/) and the WHO International Clinical Trials Registry Platform Search Portal (apps.who.int/trialsearch/).
For detailed search strategies see Appendix 1. We will continuously apply PubMed's 'My NCBI' (National Center for Biotechnology Information) email alert service for identification of newly published studies using a basic search strategy (see Appendix 1). Four weeks before we submit the final review draft to the Cochrane Metabolic and Endocrine Disorders Group (CMED) for editorial approval, we will perform a complete update search on all specified databases. Should we detect new studies for inclusion we will evaluate these and incorporate findings in our review before submission of the final review draft.
If we detect additional relevant key words during any of the electronic or other searches we will modify the electronic search strategies to incorporate these terms and document the changes. We will place no restrictions on the language of publication when searching the electronic databases or reviewing reference lists in identified studies.
We will send results of electronic searches to the CMED for databases which are not available at the editorial office.
Searching other resources
We will try to identify other potentially eligible trials or ancillary publications by searching the reference lists of retrieved included trials, (systematic) reviews, meta-analyses and health technology assessment reports.
Data collection and analysis
Selection of studies
To determine the studies to be assessed further, two review authors (NMG, MOG) will independently scan the abstract, title, or both sections of every record retrieved by the searches. We will investigate all potentially relevant articles as full text. Where differences in opinion exist, we will resolve them by discussion. If resolving disagreement is not possible, we will add the article to those 'awaiting assessment' and we will contact study authors for clarification. We will present an adapted PRISMA (preferred reporting items for systematic reviews and meta-analyses) flow-chart of study selection (Figure 1) (Liberati 2009).
|Figure 1. Study flow diagram.|
Data extraction and management
For studies that fulfill inclusion criteria, two review authors (NMG, MOG) will independently extract relevant population and intervention characteristics using standard data extraction templates, with any disagreements to be resolved by discussion, or if required by the CMED (for details see Table 1; Appendix 2; Appendix 3; Appendix 4; Appendix 5; Appendix 6; Appendix 7; Appendix 8; Appendix 9; Appendix 10; Appendix 11).
We will provide information including trial identifier about potentially relevant ongoing studies in the table 'Characteristics of ongoing studies' and in Appendix 6 'Matrix of study endpoints (trial documents)'. We will try to find the protocol of each included study, either in databases of ongoing trials, in publications of study designs, or both, and specify data in the appendix 'Matrix of study endpoints (trial documents)'.
We will send an email to all authors of included studies to enquire whether they are willing to answer questions regarding their trials. We will present the results of this survey in Appendix 12. Thereafter, we will seek relevant missing information on the trial from the primary author(s) of the article, if required.
Dealing with duplicate and companion publications
In the event of duplicate publications, companion documents or multiple reports of a primary study, we will maximise yield of information by collating all available data. In case of doubt the publication reporting the longest follow-up associated with our primary or secondary outcomes will be given priority.
Assessment of risk of bias in included studies
Two review authors (NMG, MOG) will assess the risk of bias of each included study independently. We will resolve disagreements by consensus, or by consultation with a third party (the CMED).
- Random sequence generation (selection bias).
- Allocation concealment (selection bias).
- Blinding (performance bias and detection bias), separated for blinding of participants and personnel and blinding of outcome assessment.
- Incomplete outcome data (attrition bias).
- Selective reporting (reporting bias).
- Other potential sources of bias.
We will assess outcome reporting bias (Kirkham 2010) by integrating the results of 'Examination of outcome reporting bias' (Appendix 7), 'Matrix of study endpoints (trial documents)' (Appendix 6) and section 'Outcomes (outcomes reported in abstract of publication)' of the 'Characteristics of included studies' table. This analysis will form the basis for the judgement of selective reporting (reporting bias).
We will judge risk of bias criteria as 'low risk', 'high risk' or 'unclear risk' and will evaluate individual bias items as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a). We will present a 'Risk of bias' figure and a 'Risk of bias summary' figure.
We will assess the impact of individual bias domains on study results at endpoint and study levels.
For performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessors) and attrition bias (incomplete outcome data) we intend to evaluate risk of bias separately for subjective and objective outcomes (Hróbjartsson 2013). We will consider the implications of missing outcome data from individual participants.
We define the following endpoints as subjective outcomes.
- Health-related quality of life.
- Adverse events.
We define the following endpoints as objective outcomes.
- Height measures.
- All-cause mortality.
- Cognitive outcomes.
- Bone density, turnover and morphology.
- Testosterone levels.
- Lipid abnormalities.
- Insulin sensitivity.
Measures of treatment effect
We will express dichotomous data as odds ratios (ORs) or risk ratios (RRs) with 95% confidence intervals (CIs). We will express continuous data as mean differences (MDs) with 95% CIs.
Unit of analysis issues
We will take into account the level at which randomisation occurred, such as cluster-randomised trials and multiple observations for the same outcome.
Dealing with missing data
We will obtain relevant missing data from authors, if feasible, and carefully evaluate important numerical data such as screened, randomised participants as well as intention-to-treat (ITT), as-treated and per-protocol populations. We will investigate attrition rates, for example drop-outs, losses to follow-up and withdrawals, and will critically appraise issues of missing data and imputation methods (e.g. last observation carried forward (LOCF)).
Where standard deviations for outcomes are not reported we will impute these values by assuming the standard deviation of the missing outcome to be the average of the standard deviations from those studies where this information was reported. We will investigate the impact of imputation on meta-analyses by means of sensitivity analysis.
Assessment of heterogeneity
In the event of substantial clinical, methodological or statistical heterogeneity, we will not report study results as the pooled effect estimate in a meta-analysis.
We will identify heterogeneity (inconsistency) by visual inspection of the forest plots and by using a standard Chi² test with a significance level of α = 0.1. In view of the low power of this test, we will also consider the I² statistic, which quantifies inconsistency across studies to assess the impact of heterogeneity on the meta-analysis (Higgins 2002; Higgins 2003), where an I² statistic of 75% or more indicates a considerable level of heterogeneity (Higgins 2011a).
When we find heterogeneity, we will attempt to determine possible reasons for it by examining individual study and subgroup characteristics.
We expect the following characteristics to introduce clinical heterogeneity.
- Underlying diagnosis.
- Bone age at commencement of treatment.
- Pubertal status at commencement of treatment.
- Duration of treatment.
Assessment of reporting biases
We will use funnel plots if we include 10 or more studies for a particular outcome, in order to assess small study effects. As there could be several explanations for funnel plot asymmetry, we will interpret results carefully (Sterne 2011).
Unless there is good evidence for homogeneous effects across studies, we will summarise primarily low risk of bias data by means of a random-effects model (Wood 2008). We will interpret random-effects meta-analyses with due consideration of the whole distribution of effects, ideally by presenting a prediction interval (Higgins 2009). A prediction interval specifies a predicted range for the true treatment effect in an individual study (Riley 2011). In addition, we will perform statistical analyses according to the statistical guidelines contained in the latest version of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a).
Subgroup analysis and investigation of heterogeneity
We will carry out the following subgroup analyses and plan to investigate interaction.
- Age at start of treatment.
- Pubertal status at start of treatment.
- Underlying diagnosis (ISS, GH deficiency, CDGP).
- Duration of intervention.
We will perform sensitivity analyses in order to explore the influence of the following factors on effect sizes.
- Restricting the analysis to published studies.
- Restricting the analysis taking into account risk of bias, as specified in the section Assessment of risk of bias in included studies.
- Restricting the analysis to very long or large studies to establish how much they dominate the results.
- Restricting the analysis to studies using the following filters: diagnostic criteria, language of publication, source of funding (industry versus other), country.
We will also test the robustness of the results by repeating the analysis using different measures of effect size (RR, OR etc.) and different statistical models (fixed-effect and random-effects models).
Appendix 1. Search strategies
Appendix 2. Description of interventions
Appendix 3. Baseline characteristics (I)
Appendix 4. Baseline characteristics (II)
Appendix 5. Matrix of study endpoints (publications)
Appendix 6. Matrix of study endpoints (trial documents)
Appendix 7. Examination of outcome reporting bias
Appendix 8. Definition of endpoint measurement
Appendix 9. Adverse events (I)
Appendix 10. Adverse events (II)
Appendix 11. Adverse events (III)
Appendix 12. Survey of authors providing information on included trials
Contributions of authors
Niamh McGrath (NMG): protocol draft, search strategy development, acquiring trial reports, trial selection, data extraction, data analysis, data interpretation, review draft and future update draft.
Michael J O'Grady (MOG): protocol draft, search strategy development, acquiring trial reports, trial selection, data extraction, data analysis, data interpretation, review draft and future update draft.
Declarations of interest
Niamh McGrath : none known.
Michael J O'Grady: none known.