Introduction

Growth retardation is a common and significant clinical problem that is not adequately managed in children with chronic kidney disease (CKD). The etiology of growth retardation in these patients is multifactorial. Age at CKD onset, primary kidney disease, tubular defects, metabolic acidosis, kidney osteodystrophy, undernourishment, and impairment of the somatotrophic axis [growth hormone (GH), insulin-like growth factor (IGF), IGF binding proteins] are among the main factors identified to cause growth retardation. Successful renal transplantation (RTx) does not in itself correct growth retardation, especially when RTx is performed after 6 years of age, the graft does not achieve optimal function, and high doses of glucocorticosteroids are administered for immunosuppressive therapy [1]. Recombinant human growth hormone (rhGH) therapy has been used to improve growth velocity and final height in children with CKD [2–5] after RTx [6–9]. Increased growth velocity in post-RTx CKD children on rhGH has been well documented; however, the effect of this treatment on final height (FH) has not been clearly determined.

Haffner et al. [10] reported that long-term rhGH treatment induces persistently improved growth with attainment of target adult height in the majority of patients. Studies from the North American Pediatric Renal Transplant Cooperative Study (NAPRTCS) have reported that rhGH treatment is effective in improving growth velocity and final height in pediatric renal allograft recipients [11, 12]. In this study, we assessed the efficacy of rhGH treatment in a group of RTx patients and compared it with the spontaneous growth of a RTx group who were not treated with rhGH but who were followed with the same protocol and in the same center. Our findings show that rhGH treatment prevented further height loss and improved FH standard deviation scores (SDS) without affecting kidney function.

Materials and methods

Thirty-three patients (27 male, six female) aged 13.24 ± 3 years who had undergone RTx because of CKD at Garrahan Pediatric Hospital of Buenos Aires during the last 15 years and who received rhGH treatment for ≥36 months, reaching FH, were included in the study (rhGH group). Inclusion criteria were the following: 18 months post-RTx, stable renal function over the 6 months prior to treatment onset, and growth velocity <25th percentile during the year previous to treatment onset or a height ≤−2 SDS. Exclusion criteria were all patients with kidney rejection within the first 2 years after RTx, cumulative hydrocortisone-equivalent dose therapy >25 mg/m²/day, serum hemoglobin <10 g/dl, serum bicarbonate <22 mEq/l, and serum parathyroid hormone (PTH) >200 pg/ml. All patients received rhGH at a dose of 10 mg/m2 body surface/week, subcutaneously, divided into seven doses until reaching FH. Mean duration of rhGH treatment was 3.53 ± 1 year (range 3.0–5.6 years). Patients were examined at the onset of rhGH treatment and at 3-month intervals thereafter. Evaluation of auxological parameters was carried out every 6 months. After discontinuing rhGH therapy, patients were evaluated every 6 months until FH. At treatment onset, 15 patients were prepubertal, and the remaining 18 were at Tanner’s stage 2–3 of sexual development. Mean ± SDS chronological age was 13.24 ± 3 years, bone age was 8.6 ± 2.6 years, height (H) SDS was −3.28 ± 1.2, and difference between initial height (IH) and target height (TH) was −3.02 ± 1.3 (Table 1). The etiology of CKD was as follows: obstructive uropathy, reflux or renal dysplasia/ hypoplasia (n = 20), hemolytic uremic syndrome (n = 5), hereditary nephropathy (n = 3), and glomerulopathy (n = 5).

Table 1 Clinical material at onset of recombinant human growth hormone (rhGH) treatment. Comparison of clinical features of control and rhGH-treated groups
Full size table

Fourteen patients (eight males and six females; eight prepubertal and six pubertal), who had undergone RTx refused rhGH therapy but could be followed until FH (controls). Inclusion and exclusion criteria were the same as in the treatment group. Chronological age, age at RTx, bone age, TH SDS, IH SDS – TH SDS, creatinine clearance (CrCl), and cumulative hydrocortisone equivalent doses (CD) were similar in both groups (Table 1). In controls, etiologies of CKD were also similar: obstructive uropathy, reflux or renal dysplasia/ hypoplasia (n = 7), hemolytic uremic syndrome (n = 1), hereditary nephropathy (n = 3), and glomerulopathy (n = 3). The patients were assessed according to the same protocol as those in the rhGH-treated group.

Anthropometry was performed on wall- and table-mounted stadiometers to determine height and sitting height. SDS and differences (∆) between scores were calculated. Optimization of FH was assessed by comparing it with TH. TH (in cm) was determined according to the following method: [paternal height + (maternal height + 12.5)]/2 for boys, and [(paternal height – 12.5) + maternal height]/2 for girls. Decreasing growth velocity to <1 cm/year, and/or a bone age >16 years in boys and >15 years in girls was considered as evidences of FH . H was expressed as SDS according to chronological age and gender based on Tanner’s criteria for healthy children [13]. Bone age was assessed by the same observer using the Greulich and Pyle method [14]. Glomerular filtration rate (GFR) was calculated with the Schwartz formula based on plasma creatinine concentration [15]. The protocol was approved by the Institutional Review Board of the Hospital de Pediatría Garrahan of Buenos Aires.

Statistical analysis

Statistical analysis was performed using Statistix 7 (Analytical Software, Tallahassee, FI, USA). For comparison of study variables of both groups as a function of time, one-way analysis of variance (ANOVA) was used, and multiple comparisons were performed with the Bonferroni method. Comparisons between treatment and control groups were carried out with Student’s t test for unpaired samples. To determine the best predictor parameter of FH, the best subset regression was used, and multiple regression analysis of different variables was carried out. Values are expressed as mean ± standard deviation (SD). A p ≤ 0.05 α level was considered to be statistically significant.

Results

FH in the rhGH patients was significantly greater than in controls [−1.88 ± 1.14 vs -3.48 ± 1.19 SDS, respectively (p = 0.017)]. In Fig. 1, height at baseline and final height in both groups are shown. Baseline height was below the lower quartile in 77.8% of boys and 100% of girls in the rhGH group and in 75% and 83.3% of boys and girls, respectively, in controls.

Fig. 1
figure 1

Growth curves (height): boys; girls. In red, initial height (IH) and in green final height (FH) for the recombinant human growth hormone (rhGH)-treated group. In black, IH and in blue FH for the control group

Full size image

However, final height was below the lower quartile in only 37% of boys and 33.3% of girls in the rhGH group but was 87.5% in boys and 100% of girls in controls. In the rhGH group, a significant H SDS gain (∆H SDS: + 1.18 ± 0.7, p = 0.001) was observed, whereas in controls, a significant loss of H SDS (∆H SDS: −0.52 ± 0.22, p = 0.033) was found (Fig. 2).

Fig. 2
figure 2

Evaluation of final height (FH) in the recombinant human growth hormone (rhGH)-treated (gray bars) and control (black bars) groups. a Comparison between initial height (IH) and FH. Height is expressed as X ± standard error (SE) *p= 0.000, **p = 0.017. b Comparison between IH minus target height (TH) and between FH minus TH. *p = 0.001, **p = 0.02

Full size image

IH–TH SDS was similar in both groups. FH–TH SDS in comparison with IH–TH SDS was significantly less in the rhGH group only (−1.8 ± 1.24 vs. −3.0 ± 1.3, p = 0.001, respectively), whereas in the control group, a tendency to increased difference between IH–TH SDS and FH–TH SDS (−2.41 ± 1 and -2.9 ± 1.5; ns) was observed (Fig. 1). In addition, in the rhGH group, FH–TH SDS was significantly lower than in controls; p = 0.02 (Fig. 2).

A multiple predictor model (Best Subset Regression analysis) of FH was used to identify the best parameters associated with response to rhGH treatment (Table 2). We analyzed the following variables: TH, time spent on dialysis before RTX, duration of rhGH treatment, chronological age at RTx, CrCl, cumulative hydrocortisone equivalent doses, chronological age, and H SDS at the onset of rhGH treatment. H SDS at the onset of rhGH treatment was the best parameter associated with good response to treatment (cp = -0.4, R² = 0.36, p = 0.0001).

Table 2 Parameters of multiple regression analysis of selected independent variables for final height
Full size table

At FH, a similar drop in GFR was found in both groups (rhGH from 75.7 ± 17.2 to 67 ± 14.7; controls from 72.6 ± 19.2 to 56.9 ± 9.3 ml/min/m²surface area). In addition, in both groups, similar CrCls were observed during the entire follow-up.

Discussion

Despite advances in the management of children with chronic renal insufficiency (CRI), low stature continues to be a condition that is difficult to manage and that causes significant comorbidity in these patients. Multiple studies have demonstrated the efficacy of rhGH treatment in RTx recipients. However, rhGH is still not used frequently in this pathology, and reports on the impact on FH are scarce. This study shows that rhGH treatment resulted in an increase in FH of 1.18 SDS in these pediatric RTx recipients. Similar results were found in other published series [10, 11, 16]. Haffner et al. [10] described the impact of rhGH treatment for 5 years in a group of CRI patients compared with patients who did not receive the treatment. Although this study concludes that rhGH induces a persistent increase of growth velocity resulting in normal adult height in the majority of patients, the study mixed patients receiving conservative treatment, dialysis, or renal transplantation, making it difficult to assess the effect of rhGH treatment on growth in RTx patients. In the 2005, in a NAPRTCS report, Fine et al. [11] described that rhGH was effective in improving FH in RTx patients. In this study, results of a data base including several centers in the United States were shown. The large size of the series (n = 669) supported the relevance of the results obtained in our study, as our patients were followed up in a single center using the same protocol. In 2009, Seikaly et al. [12] reported an update on this data base and clearly demonstrated the benefits of rhGH in terms of increased growth velocity compared with a nontreated age-matched control group. Nissel et al. [17] described persistently increased growth velocity with rhGH treatment in prepubertal as well as pubertal patients. H SDS gain in RTx recipients was similar to that observed in our patients.

According to our results, the variable that best predicted FH was H at onset of treatment, and a similar finding was reported by Nissel et al. [17]. In contrast with our study, Nissel et al. [16] also observed that rhGH treatment duration was correlated with FH gain. However, Berard et al. [18] and Haffner et al. [10] described additional factors that influence FH, such as duration of treatment and growth velocity previous to treatment onset [18], as well as greater initial height deficit, male gender, longer duration of the prepubertal and pubertal observation periods, and a lower percentage of time spent on dialysis [10]. Again, as in both studies patients receiving conservative treatment, dialysis, and had received RTx were included, it is difficult to specifically analyze growth outcome of RTx recipients only.

According to our results, although different predictive factors of FH have been reported [10, 17, 18], IH was a common determining factor for predicting improved FH. Even though it has been described in an animal model that rhGH through increased IGF-1 production [16] produces renal hyperfiltration followed by glomerulosclerosis and a decreased GFR, our study showed that the long-term use of rhGH in RTx recipients does not affect GFR. Indeed, the fall in GFR was similar in treatment and control groups. Similar findings have been reported in other series [6–10, 14].

Conclusion

In this study, efficacy of rhGH treatment was analyzed in a cohort of RTx recipients followed at a single center with the same protocol. Our findings confirm that rhGH is effective in improving FH in RTx recipients without affecting kidney function. Finally, rhGH treatment proved to be more effective in improving FH when the loss of IH SDS was less marked at onset of treatment. Yearly growth velocity checking is important, as it alerts physicians not to delay initiation of rhGH therapy.