| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Pediatrics (R.N., D.H.), University Children’s Hospital, D-18057 Rostock, Germany; KIGS Outcomes Research (A.L.), Pfizer Endocrine Care, Pfizer, SE-112 87 Stockholm, Sweden; and Division of Pediatric Nephrology (O.M.), University Children’s Hospital, D-69115 Heidelberg, Germany
Address all correspondence and requests for reprints to: Richard Nissel, M.D., Department of Pediatrics, University Children’s Hospital, Rembrandtstrasse 16/17, 18057 Rostock, Germany. E-mail: richard.nissel{at}med.uni-rostock.de.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: This study set out to analyze near-FH in a cohort of GH-treated CKD patients.
Design, Settings, and Patients: Of 240 evaluable patients in the Pfizer International Growth Database (KIGS) with CKD, 39% were prepubertal and 61% were pubertal at baseline; 45% were on conservative treatment for CKD, 28% were on dialysis, and 27% were in the period after renal transplantation.
Main Outcome Measures: Near-FH, relation to pubertal stage, and factors predictive of growth response were the main outcome measures.
Results: Mean height SD scores increased continuously during GH treatment until near-FH by 1.2 and 1.6 in boys and girls, respectively. Mean near-FH differed significantly from prepubertal patients showing severely delayed puberty (–3.6), late pubertal patients (–2.9), early pubertal patients (–2.2), and prepubertal patients with normal onset of puberty (–2.0). The initial degree of stunting, degree of bone age retardation, duration of GH therapy, time spent on conservative treatment/dialysis, pubertal delay (>2 SD), gender, and age at start of GH treatment were significant predictors of growth response to GH therapy, explaining between 33 and 61% of the overall variability.
Conclusions: Long-term GH therapy of CKD patients in prepubertal and pubertal age results in an increased adult height, but response is diminished in patients on dialysis and/or with severely delayed puberty.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
GH therapy is an accepted measure to increase final height (FH) in young prepubertal CKD patients, but the impact of GH treatment on adult height in patients of pubertal age is unclear, and growth response in prepubertal patients shows substantial interpatient variability (2, 3, 4, 5). In prepubertal CKD patients, growth response is positively associated with duration of GH therapy and initial degree of growth failure. It was negatively associated with the duration of dialysis (3, 6).
Previous analyses on the effect of GH treatment on adult height in CKD patients were compromised by small sample sizes, restriction to prepubertal patients, and/or missing differentiation for pubertal status (2, 3, 4, 5, 7, 8, 9). Therefore, we analyzed the effect of GH treatment on near-FH in a large cohort of prepubertal and pubertal patients suffering from CKD and receiving either conservative treatment or dialysis or had undergone renal transplantation (RTx).
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The Pfizer International Growth Database (KIGS) database was founded in 1987 to record long-term outcome and safety of Somatonorm and Genotropin GH products (Pfizer Inc., New York, NY). Patients are enrolled according to the recommendations of the 18th World Medical Assembly (held in Helsinki, Finland, in 1964) and any subsequent revisions directed to guide physicians in biomedical research on humans. At the time of founding, enrollment was independent from informed consent in many countries, but currently data are enrolled only after informed consent has been obtained, and data are analyzed anonymously. Regardless of the time of enrollment, every individual has the right to withdraw at any time. Data from 1710 patients suffering from growth failure due to CKD were enrolled in the KIGS registry by July 1, 2007. Uremic growth failure was diagnosed according to the KIGS etiology Classification List (no. 3.8.5), based on a height SD score (SDS) less than – 2.0 and/or a height velocity less than the 25th percentile over the preceding year in CKD patients on conservative treatment (CKD stage III/IV, i.e. glomerular filtration rate less than 60 ml/min per 1.73 m2), on hemo- or peritoneal dialysis or after RTx.
Our analysis is restricted to 240 patients who had been treated with GH (target dose 0.33 mg/kg body weight per week) for at least 1 yr and had reached their near-FH by July 1, 2007 (Table 1
). Near-FH was defined as a height velocity less than 1 cm/yr, advanced clinical signs of puberty (Tanner stage 
4), in boys with an age of at least 16 yr and in girls of at least 14 yr. At baseline, 38% of patients were prepubertal, 47% were in early (Tanner II/III), and 15% in late (Tanner IV/V) puberty. At baseline, 108 patients (45%) received conservative treatment, 67 (28%) underwent dialysis (continuous peritoneal dialysis in 22% and hemodialysis in 78%), and 65 (27%) had functioning renal allografts. During the study period, 38 children (35%) switched from conservative treatment to dialysis, and of these children, 15 (39%) subsequently received RTx. In addition, eight of the patients on conservative treatment (7%) were transplanted preemptively and 12 patients initially treated by dialysis (18%) received a renal allograft. In total, 35 patients being initially on conservative or dialysis treatment received a renal allograft during the observation period, and in eight of these patients, dialysis treatment had to be restarted. In 15 patients with functional renal allografts at baseline (38%), dialysis treatment had to be started, and three of these patients received a second allograft during further follow-up. The mean prednisolone dosage in RTx patients was 5.5 ± 3.0 mg/d. Baseline characteristics of patients are given in Fig. 1, A–C.
|
|
Clinical studies
Standard anthropometric measures, sexual maturation (Tanner stage), serum creatinine, treatment modalities, prednisolone dosage, use of cytotoxic drugs, and bone age were assessed at 3- up to 12-month intervals. Bone age was determined according to Greulich and Pyle (10) and Tanner et al. (11) in 71 and 29% of the patients, respectively. Reference data for growth analysis and pubertal development were taken from the first Zurich longitudinal study (12, 13, 14). Genetic target height was calculated as midparental height + 10 cm for boys and – 2.6 cm for girls (15). The estimated glomerular filtration rate (GFR) was assessed by the Schwartz formula (16).
Statistical analyses
Results are expressed as means ± SD. Comparisons between groups were performed by ANOVA, followed by Duncan’s multiple range test. The longitudinal changes in anthropometric measurements were evaluated by repeated-measure ANOVA including a within-subject factor (time) and a between-subject factor (e.g. prepubertal vs. pubertal patients, boys vs. girls, conservative treatment vs. dialysis and transplantation, respectively). Pair-wise comparisons between different time points were performed using the CONTRAST option of the general linear models procedure in the SAS/STAT software, version 8 (SAS Institute, Cary, NC).
Possible predictors of catch-up growth, i.e. change in height SDS from start of GH therapy until near-FH and near-FH SDS were identified by univariate linear regression analysis. Age, gender, pubertal stage at baseline, clinical onset of puberty (normal vs. delayed (> 2 SD) onset of puberty in patients being prepubertal at baseline), degree of bone age retardation, height SDS, genetic target height deficit at baseline, duration and dosage of GH treatment, initial treatment modality, percentage of time on conservative treatment, dialysis and post-RTx, GFR, underlying renal disease, use of cytotoxic drugs, and prednisolone dosage were examined. Multiple regression analysis was performed starting with a model including all potential predictors (see above), followed by stepwise elimination of those variables not significantly contributing to the overall variance. All P values are two sided.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. The mean age of the patients at start of GH treatment was 13.7 ± 3.0 yr, their bone age was 10.4 ± 2.9 yr, and their mean height SDS was –3.6 ± 1.2. For all patients, mean height SDS significantly decreased by –0.2 (P < 0.0001) during the year before onset of GH treatment. The cumulative change in height SDS during the observation period was significantly higher in girls (+1.6), compared with boys (+1.2; each P < 0.0001 vs. baseline; P < 0.01 girls vs. boys) (Fig. 2A). Despite the lower growth response in boys the mean near-FH SDS was significantly higher (–2.3), compared with girls (–2.9; P < 0.01). This was mainly due to the markedly higher initial degree of stunting in the latter group (girls –4.5 SD, boys –3.4 SD; P < 0.0001). Overall, 40% of these children (girls, 28%; boys, 43%) reached an adult height within the normal range (i.e. within 2 SD of normal height). However, the mean near-FH was below the genetic target height by –16.2 cm (–2.7 SD) and –15.1 cm (–2.0 SD) in girls and boys, respectively (each P < 0.0001).
|
To elucidate the influence of the pubertal status on growth response to GH treatment, CKD patients were classified accordingly as being prepubertal at start of GH treatment showing 1) normal or 2) delayed (>2 SD) onset of puberty during follow-up, and 3) patients being in early (Tanner II/III) or 4) late puberty (Tanner IV/V) (Fig. 2B). The mean chronological age (bone age) at baseline was: 1) 10.0 ± 2.2 (7.3 ± 1.8) yr; 2) 14.2 ± 1.9 (10.3 ± 2.8) yr; 3) 14.6 ± 1.8 (11.2 ± 2.0) yr; and 4) 16.4 ± 1.3 (13.4 ± 1.8) yr. Bone age advancement during the first treatment year was: 1) 1.2 ± 0.7 yr; 2) 0.6 ± 0.9 yr; 3) 1.2 ± 1.1 yr; and 4) 1.1 ± 0.7 yr (each P > 0.05 vs. increment in chronological age). The increment in height SDS during the first year of GH treatment was higher in prepubertal patients with normal onset of puberty and late pubertal patients (each +0.5), compared with prepubertal patients with delayed onset of puberty (+0.2) and early pubertal patients (+0.36; each P < 0.01). The cumulative increase in height SDS tended to be higher in prepubertal patients with normal onset of puberty (+1.3) and early pubertal patients (+1.3), compared with prepubertal patients with delayed puberty (+0.9) and late pubertal patients (+1.0; each P > 0.05; Fig. 2B). Regardless of gender bias, the mean absolute height increment during the pubertal growth period (i.e. time from Tanner stage II up to final height) was higher in prepubertal patients with normal pubertal onset, compared with patients with delayed puberty (prepubertal boys with normal/delayed puberty, 23.7 cm/10.5 cm; prepubertal girls with normal/delayed puberty, 15.4 cm/6.9 cm; each P < 0.005). Delayed onset of puberty was significantly associated with the proportion of time spent on dialysis treatment (delayed puberty, 74%; normal onset of puberty, 30%; P < 0.0001). In summary, mean near-FH SDS was significantly higher in prepubertal patients with normal pubertal onset (–2.0) and early pubertal patients (–2.2), compared with late pubertal patients (–2.9) and prepubertal patients with delayed puberty (–3.6; each P < 0.001).
Influence of treatment modality for CKD on near-FH
A total of 166 patients were on a constant treatment modality for CKD throughout the observation period (Fig. 3A). The cumulative increase in mean height SDS as well as near-FH SDS was significantly higher in patients on conservative treatment (+1.5/–1.7), compared with patients on dialysis (+1.1/–3.0) or after RTx (+1.1/–2.4) (each P < 0.05). During the observation period mean GFR in RTx patients was about twice that in patients on conservative treatment (71 ± 28 ml/min per 1.73m2 vs. 39 ± 19 ml/min per 1.73m2; P < 0.0001).
|
The mean cumulative increase in height SDS did not differ significantly with respect to the underlying renal disease (Fig. 3B). However, mean near-FH SDS was significantly lower in patients with nephropathic cystinosis (–3.4), compared with the other groups (congenital uropathies, –2.3; glomerulopathies, –2.2; hereditary nephropathies, –2.2; interstitial nephritis, – 2.1; each P < 0.01). This was mainly due to the markedly higher initial degree of stunting and the fact, that the relative proportion of time spent on conservative treatment was significantly lower in cystinosis patients (14 vs. 38%; P = 0.03).
Independent predictors of growth
The cumulative increase in height SDS during the observation period was positively associated with the genetic target height deficit, duration of GH therapy, bone age retardation at baseline, and relative proportion of time spent on conservative treatment. It was negatively associated with the height velocity before GH therapy (Table 2). Near-FH SDS was positively associated with height SDS at baseline and duration of GH therapy and was negatively associated with the relative proportion of time spent on dialysis treatment, delayed puberty (>2 SD), female gender, and age at onset of GH treatment.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Prepubertal and pubertal CKD patients showed continuous catch-up growth during GH treatment. The cumulative increase in mean height SDS was significantly higher in girls vs. boys (+1.6 vs. 1.2) and in patients being on conservative treatment (+1.5) vs. patients on dialysis or after RTx (each +1.1). Forty percent of the patients reached a normal adult height. The initial degree of stunting, bone age retardation, duration of GH therapy, time spent on conservative treatment/dialysis, severe pubertal delay, gender, and age at start of GH treatment were significant predictors of growth response to GH therapy, explaining between 33 and 61% of the overall variability.
In prepubertal patients a mean cumulative increase in height SDS of +1.3 was seen during GH therapy. This is in line with previously reported total height gains ranging from 1.0 up to 1.4 SD in GH-treated prepubertal patients on conservative treatment or dialysis or after RTx (2, 3, 4, 5). However, these data are summarized from four different studies representing a total of 57 patients only. Our study is unique in that the large patient cohort compromising prepubertal and pubertal patients with variable treatment modalities for CKD enables us to characterize the main factors affecting catch-up growth and FH achieved by GH treatment.
Target height deficit at baseline and duration of GH therapy were the most important predictors of cumulative height gain. The former factor has also been demonstrated as an important predictor of longitudinal-growth in children with idiopathic short stature or Turner’s syndrome treated by GH (17, 18). The positive association between cumulative height gain and the initial degree of growth retardation suggests a relation between the efficacy of GH treatment and the biological demand for catch-up growth. The mean duration of GH treatment in the present study was 87% of the total observation period. GH therapy was stopped before attainment of near-FH in 97 of 240 patients secondary to RTx or attainment of genetic target height. Although after cessation of GH treatment standardized height in these patients did slightly increase until near-FH (+0.15 SD; P < 0.001), the cumulative change in height SDS during the whole observation period was significantly correlated with the duration of GH therapy.
Dialysis was clearly associated with poor growth response. In patients on long-term dialysis, the mean cumulative increase in height SDS amounted to +1.1, whereas patients on conservative treatment showed a mean increase of +1.5. Although the initial age at onset of GH treatment turned out to be of only minor importance for growth response, the slightly higher age at onset of GH therapy in the dialysis group might, at least partly, have contributed to the diminished growth response. The substantially higher initial target height deficit in dialysis patients and the fact that this is the most important positive predictor of catch-up growth indicates that the overall negative effect of dialysis treatment on growth response must be even estimated higher. This is in agreement with earlier studies in prepubertal CKD patients showing reduced growth rates during GH treatment in dialyzed, compared with conservatively treated patients (6, 19, 20). The reduced number of cellular GH receptors and higher levels of circulating IGF inhibitors point to a reduced GH sensitivity in patients with end-stage CKD, compared with those with pre-end-stage CKD (21). Although evidence was obtained that intensified dialysis (i.e. prolonged thrice weekly or short daily hemodialysis sessions) may improve longitudinal growth in GH-treated patients, this warrants further investigations (22, 23).
Irrespectively of pubertal stage, the magnitude of total height gain in RTx patients (+1.1 SD) is in between that of patients with pre-end-stage CKD and those on dialysis. Results of previous studies comparing growth response in RTx and pre-end-stage CKD patients of prepubertal age are conflicting in that comparable as well as reduced growth rates were seen (24, 25, 26). Interestingly, in the present study, mean GFR in RTx patients was about twice that in patients on conservative treatment. No relationship between glucocorticoid dosages and response to GH in our RTx patients was seen. This is probably due to the small dose ranges of glucocorticoids and the well-known large pharmacokinetic and pharmacodynamic variability (27, 28). However, several previous studies have shown that even low-dose glucocorticoid treatment (<4 mg/m2·d) results in growth suppression in RTx patients not treated by GH (29, 30). Thus, we conclude, that the growth-suppressing effects of glucocorticoids can be only partly overcome by GH therapy in RTx patients.
Although, the use of cytotoxic agents in patients with glomerular diseases has been shown to adversely effect pubertal development and growth, this was not observed in our study (31).
The growth response to GH was significantly associated with the stage of pubertal development. In line with previous observations, about one quarter of the prepubertal patients showed severe delayed onset of puberty (>2 SD) during GH treatment (3, 32, 33). In prepubertal patients the change in height SDS during the first year and the cumulative height increase until near-FH as well as total pubertal height gain were almost 2-fold higher in patients with normal onset, compared with those with delayed onset of puberty. Consequently, mean near-FH SDS in patients with delayed onset of puberty was about 1.6 lower, compared with patients with normal onset of puberty. This finding was at least partly related to the higher degree of stunting at baseline. The diminished growth rate in late prepubertal CKD patients despite GH treatment was striking. This might be an accentuation of the physiological phenomenon known as the prepubertal dip due to the severely delayed pubertal growth spurt (4, 30). Delayed onset of puberty was highly associated with dialysis treatment, resulting in a markedly shortened pubertal growth period and reduced final height. This is thought to be mainly due to insufficient activation of the hypothalamic GnRH pulse generator, likely mediated via circulating inhibitors associated with the uremic state and/or concomitant glucocorticoid treatment in RTx patients (34).
The strikingly lower growth response to GH and reduced adult height in patients with delayed onset of puberty raise the question whether puberty should be induced exogenously. Experience with sex hormone therapy in CKD patients is limited and restricted to patients not being on concomitant GH therapy. In boys on hemodialysis with extreme delay of puberty low-dose testosterone administration resulted in rapid increase in height velocity and development of secondary sexual characteristics (35, 36). However, bone maturation accelerated disproportionately, at rates of 2–4 yr of bone age per year of chronological age, resulting in a dramatic loss of predicted final height. This finding suggests that CKD patients might be hypersensitive to stimulation by sex steroids.
In both sexes the epiphyseal fusion is mediated by an estrogen triggered cascade of proliferation, differentiation, and apoptosis of chondrocytes (37). Inhibition of estrogen synthesis by aromatase inhibitors appears to decelerate the process of growth plate fusion and thus may be used therapeutically to increase adult height in male patients. This concept has recently been proven to be effective in boys with idiopathic short stature and warrants future adequately designed studies in CKD patients with and without concomitant GH therapy (38).
Interestingly, the mean total height gain in early pubertal patients and prepubertal patients (each +1.3 SD) was comparable. By contrast, in late pubertal patients cumulative height gain (+0.9 SD) as well as near-FH was markedly lower. Despite the advanced age (boys, 15.1 yr; girls, 14.7 yr) the mean absolute height gain in pubertal patients (treatment period: 3.5 yr; 3.2 yr) amounted to 19.7 ± 10.2 cm and 14.2 ± 7.6 cm in boys and girls, respectively. This is comparable with the total pubertal height gain in CKD patients without GH therapy being on conservative treatment or dialysis or after RTx (30, 33). This also is in line with preliminary observations obtained in a small cohort of early and late pubertal patients with functioning renal allografts showing a mean increase in height of 19 cm within 2 yr of GH treatment (39). Thus, we conclude that GH treatment significantly improved FH in early and late pubertal CKD patients.
In line with previous studies on GH treatment, a sustained catch-up growth (+1.0 SD) in cystinosis patients was observed (40). Although the relative proportion of time spent on conservative treatment during the observation period was significantly lower in cystinosis patients, compared with patients suffering from other renal diseases, catch-up growth in these patients was comparable. Thus, in cystinosis patients the substantially higher initial degree of stunting translated into significantly lower near-FH.
Girls showed significantly better growth response as boys with respect to the cumulative increase in standardized height. This discrepancy might be, at least partly, due to the higher degree of growth retardation in girls before onset of GH therapy and the fact that the initial degree of stunting is the most important predictor of growth response. However, assessing gender differences via changes in height SDS in patients of pubertal age may be significantly biased by the individual and gender-related variability of pubertal development. In the present as well as several previous studies on GH treatment in CKD, boys represent the majority of patients (2, 3, 4, 5). The higher degree of growth retardation in girls points to a bias in selection criteria made by the families and physicians. It might well be that girls were enrolled only when severe stunting has been noted, whereas boys are enrolled already with signs of less severe stunting.
In conclusion, GH treatment in severely growth retarded prepubertal and pubertal CKD patients results in continuous catch-up growth and increased near-FH. Growth response to GH is significantly associated with the initial degree of stunting, bone age retardation, duration of GH therapy, time spent on conservative treatment/dialysis, pubertal delay, gender, and age at start of GH treatment. Although GH treatment is effective in increasing near-FH in CKD patients of pubertal age, response is clearly reduced in patients on dialysis and/or with severely delayed puberty. Efforts to improve the early recognition of growth failure, thereby preserving GH sensitivity and warranting normal pubertal development alone or in combination with intensified dialysis, exogenous induction of puberty, and steroid-free immunosuppression are necessary.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
This study was presented as an oral presentation at the 14th Congress of the International Society of Pediatric Nephrology, September 2007, Budapest, Hungary.
Disclosure Statement: The KIGS registry is sponsored by Pfizer Inc., New York, NY. A.L. is employed by Pfizer (KIGS outcome research, Pfizer Endocrine Care, Stockholm, Sweden). Pfizer employees identified the participating centers and monitored the collected data to ensure adherence to good clinical practices. Data interpretation and statistical analyses were performed by R.N., O.M., and D.H. The manuscript was prepared, reviewed, and revised by all authors, and revisions were then rejected or approved by Pfizer. R.N., O.M., and D.H. received lecture fees and/or grant support from Pfizer.
First Published Online January 15, 2008
Abbreviations: CKD, Chronic kidney disease; FH, final height; KIGS, Pfizer International Growth Database; RTx, renal transplantation; SDS, SD score.
Received October 15, 2007.
Accepted January 3, 2008.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
; n = 47; aged 8.1–18.0 yr) with CKD in the year before start of therapy until attainment of near-FH (data are given as mean ± SEM; #, P < 0.01 boys vs. girls; *, P < 0.01 vs. previous time point). B, Mean height SDS of prepubertal CKD patients with normal (
, n = 25; aged 10.1–17.1 yr) onset of puberty and patients in early (
, n = 112; aged 10.1–19.7 yr) and late (
, n = 35; aged 13.8–19.5 yr) puberty in the year before start of GH therapy until attainment of near-FH (data are given as mean ± SEM; *, P < 0.003 vs. previous time point).
, transplantation, n = 49, aged 5.1–19.0 yr) (data are given as mean ± SEM; #, P < 0.006 conservative treatment vs. dialysis; ##, P < 0.004 conservative treatment vs. other groups; *, P < 0.002 vs. previous time point). B, Mean height SDS of GH-treated patients with uremic growth failure according to the underlying renal disease (data are given as mean ± SEM; •, uropathy, n = 103, aged 4.7–19.5 yr;