0021-972X/98/$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1998 by The Endocrine Society
Vol. 83, No. 1 Printed in U.S.A.
Effect of Therapy with Recombinant Human Growth Hormone on Insulin-Like Growth Factor System Components and Serum Levels of Biochemical Markers of Bone Formation in Children After Severe Burn Injury* GORDON L. KLEIN, STEVEN E. WOLF, CRAIG B. LANGMAN, CLIFFORD J. ROSEN, SUBBURAMAN MOHAN, BRUCE S. KEENAN, SINA MATIN, CHRISTOPHER STEFFEN, MARC NICOLAI, DAWN E. SAILER, AND DAVID N. HERNDON Departments of Pediatrics (G.L.K., B.S.K.) and Surgery (S.E.W., Si.M., D.N.H.), University of Texas Medical Branch and the Shriners Burns Institute (G.L.K., S.E.W., Si.M., M.N., D.N.H.), Galveston, Texas 77555; Nephrology Division, Children’s Memorial Hospital and the Northwestern University Medical School (C.B.L., D.E.S.), Chicago, Illinois 60614; the Maine Center for Osteoporosis Research, St. Joseph Hospital (C.J.R., C.S.), Bangor, Maine 04401; and the Jerry L. Pettis VA Medical Center and Loma Linda University School of Medicine (Su.M.), Loma Linda, California 92357 ABSTRACT Burn injury in children is associated with low bone formation and long-term bone loss. Because recombinant human GH (rHGH) may accelerate burn wound healing, and because rHGH increases bone formation and density in GH-deficient patients, we studied the shortterm effects of rHGH on bone formation, reflected by osteocalcin and type I procollagen propeptide levels in a randomized, double-blind, placebo-controlled study. Nineteen patients were enrolled and received either rHGH (0.2 mg/kgzday) or an equal volume of saline. Mean burn size and age were not different between the groups, and test substances were given from admission to time of wound healing (mean: 43 6 22 days). At wound healing, serum levels of insulin-like growth factor (IGF)-1 and IGF binding protein (IGFBP)-3 in the
rHGH group rose to mean values of 229% and 187% of the respective means of the placebo group (P , 0.025). Serum osteocalcin concentrations remained below normal in both groups, and type I procollagen propeptide levels achieved a low normal level. IGFBP-4 levels were twice that of normal on admission and doubled further at wound healing; IGFBP-5 levels were low on admission but rose to normal at wound healing. We conclude that large doses of rHGH were ineffective in improving disordered bone formation despite increasing serum IGF-1 and IGFBP-3. The rHGH-independent rise in serum levels of the inhibitory binding protein IGFBP-4 suggests a mechanism by which improved bone formation is prevented despite successful elevation of IGF-1 and IGFBP-3 in the burned child. (J Clin Endocrinol Metab 83: 21–24, 1998)
B
URN INJURY is associated with reduction in bone formation (1, 2). In children, markedly diminished bone density (3) and linear growth velocity (4) also have been described. In previous attempts to improve growth velocity in burned children, recombinant human GH (rHGH) was administered during the initial hospitalization. Acceleration of wound healing was previously reported with this therapy (5, 6), as well as an increase in insulin-like growth factor (IGF)-1 levels in the blood from low to normal (7). rHGH increases IGF-1 (8), osteocalcin (8 –10), type I procollagen
propeptide (PICP) (9), and bone density (9), when administered to children with GH deficiency. Because it also seemed to increase bone formation in corticosteroid-dependent children, as evaluated by histomorphometric analysis (11), we hypothesized that rHGH would similarly improve the low bone turnover state associated with severe burns. Therefore, we performed a randomized double-blind, placebo-controlled trial of rHGH to evaluate its effect on bone formation and bone density in burned children. Furthermore, because the characterization of IGF binding protein (IGFBP) response to rHGH administration in burned patients had not been previously studied, measurements of IGFBP-3, -4, and -5 also were undertaken. Based on the study presented, we conclude that short-term treatment of this population with rHGH is not effective in reversing the defects in bone metabolism.
Received August 18, 1997. Revision received September 25, 1997. Accepted October 6, 1997. Address all correspondence and requests for reprints to: Gordon L. Klein, M.D., Pediatric Gastroenterology Division, Room 3.240B, Children’s Hospital, University of Texas Medical Branch, 301 University Boulevard, Galveston, Texas 77555-0352. E-mail:
[email protected]. * This work was presented, in part, at the 19th Annual Meeting of The American Society for Bone and Mineral Research, Cincinnati, OH, September 10 –14, 1997. Funding for this work was obtained, in part, from Shriners Hospitals for Children Grants 8770 and 8680 (to G.L.K.); NIH Grant AR-31062 (to S.M.); NIH, NIA Grant AG-1094 – 01 (to C.J.R.); and NIH Grant RR-00078 (to C.B.L.).
Subjects and Methods We studied 19 children [9 males and 10 females, ages 5.3 6 3.6 yr (sd)], randomly assigned to receive rHGH (0.2 mg/kg day) sc (GenotropinR, generously provided by Pharmacia-Upjohn, Kalamazoo, MI and Stockholm, Sweden) or a saline placebo of equivalent volume. Study drug was administered at 0600 h daily from time of admission, within 72 h of burn,
21
22
KLEIN ET AL.
to the time when wounds were considered 95% healed (indicating that no further grafting was needed and that the patient was ready for hospital discharge). All patients received iv fluids according to a standard formula (12). Enteral nutrition support was instituted within 24 h of admission, provided as formula, Vivonex TENR or Vivonex PediatricR (Sandoz Nutrition, Minneapolis, MN), providing daily: 1200 kcal/m2 as maintenance and an additional 1800 kcal/m2 open-wound area, and 30 –50 g protein/m2. Calcium intake was 0.7–1.2 g/m2 and an additional 0.9 –1.8 g/m2 body surface area burned; phosphate intake was 0.7–1.0 g/m2, plus 0.9 to 1.44 g/m2 surface area burned; magnesium intake was 0.3 g/m2, and 0.36 g/m2, depending on which formula was used. An age-appropriate regular diet was routinely added as tolerated. Blood was obtained on admission and again at wound healing, as defined above, for the following levels: IGF-1; IGFBP-3, -4, and -5; osteocalcin; and PICP. Bone mineral density (BMD) of the lumbar spine was obtained at the time of wound healing. We determined IGF-1 and IGFBP-3, respectively, by immunoradiometric assay and RIA kits, purchased from Nichols Laboratories, San Juan Capistrano, CA (13, 14). For IGF-1, the mean intraassay and interassay coefficients of variation were 5% and 15%, respectively, for 20 assays. For IGFBP-3, the mean intra- and interassay coefficients of variation were #8% and ,6.4%, respectively, for 20 assays. IGFBP-4 was analyzed first by Western blot and gel densitometry (15) and then by RIA (16). Intra- and interassay coefficients of variation, previously published (16), are ,5% and ,8.1%, respectively. IGFBP-5 levels also were determined by RIA (17) with intra- and interassay coefficients of variation, previously published (17), of ,4% and ,8%, respectively. Results of the IGFBP-4 and -5 assays were compared with age-related controls being treated for hypothyroidism and whose thyroid function was normal at the time of blood sampling. Osteocalcin and PICP levels were determined by enzyme-linked immunosorbent assay and compared with the age-related normal range, as determined in the same laboratory (2, 18). Intra- and interassay coefficients of variation have been previously published, being 2% and 6% (2), and 4% and 7% (2), respectively. BMD was determined by dual-energy x-ray absorptiometry using a QDR1000W absorptiometer (Hologic, Waltham, MA). Results are reported as z-scores, compared with age and gender-related normals (3). Sample size was chosen: 1) to exceed that necessary to detect a 40% difference in serum osteocalcin between rHGH and placebo groups at time of wound healing, a difference that would place osteocalcin in the normal range for the rHGH group, with 95% probability at the a 5 0.05 level; and 2) to exceed the number needed to detect the 3-fold rise in serum levels of IGF-1 previously reported in burned children (7) who received rHGH therapy. Statistical methods used for data analysis included linear regression, paired, and unpaired t tests, as appropriate. Results are expressed as means and sd unless otherwise noted. This study was reviewed and approved by the Institutional Review Board of the University of Texas Medical Branch, Galveston, and by the Shriners Hospitals for Children, Tampa, FL. Informed consent was obtained from parents of all study participants.
Results
There were 10 patients enrolled in the rHGH group and 9 enrolled in the placebo group. Samples were obtained from children 5 6 7 days post burn at baseline, range: 1–23 days (n 5 19); and 43 6 22 days post burn at time of wound healing, range: 12 to 78 days (n 5 16). Burn size was comparable between the two groups, 62 6 15% total body surface area for the rHGH group and 65 6 19% total body surface area in the placebo group. Serum IGF-1 levels in the serum were not significantly different at admission between the rHGH and placebo groups, although they were low in one third of the children. However, at wound healing, IGF-1 levels were significantly higher in the rHGH group (Fig. 1). Serum concentrations of IGFBP-3 were not different between the groups on admission but were low in 37% of the children. However, serum
JCE & M • 1998 Vol 83 • No 1
FIG. 1. Serum concentrations of IGF-1 in the rHGH and placebo groups at baseline (admission) and at time of wound healing. Data are given as mean 6 SD. Numbers of patients in each group are shown at the top of each bar. Normal range from ages 2 months to 5 yr is 17–248 ng/mL; and from 6 –12 yr, from 88 –1096 ng/mL.
FIG. 2. Serum concentrations of IGFBP-3 in the rHGH and placebo groups at baseline (admission) and at time of wound healing. Data are given as mean 6 SD. Numbers of patients in each group are shown at the top of each bar. Normal range in children up to age 4 yr is 0.66 –3.77 mg/mL; and from ages 4 –12 yr, 1.16 –5.46 mg/mL.
IGFBP-3 levels rose to be significantly higher in the rHGH group by time of wound healing (Fig. 2). Western ligand blot analysis revealed the presence of IGFBP-4 in nearly all sera. Scanning densitometry of the gels, reported as the ratio of percent area of IGFBP-4 in study subjects over the control value was 2.4 6 0.8 for the rHGH group (n 5 6) and 2.17 6 0.2 for the placebo group (n 5 8). RIA results (Fig. 3) revealed elevated serum levels of IGFBP-4 on admission, when compared with age-matched controls. These levels increased significantly, irrespective of rHGH administration, from admission to wound healing, thus further exceeding the normal range. Serum concentrations of IGFBP-5 (Fig. 4) were uniformly low at admission but normal at time of wound healing. There were no significant differences between the GH and control groups. Serum concentrations of osteocalcin (Fig. 5) and PICP did not differ between the two groups either at admission or at wound healing. Serum levels of both markers of bone formation rose at the time of healing. Baseline serum levels of PICP were low in the rHGH group (n 5 3, 115 6 28 ng/mL) and low normal in the placebo group (n 5 4, 233 6 183 ng/mL). Serum PICP levels rose in both groups at time of
GH AND BONE FORMATION
FIG. 3. Serum concentrations of IGFBP-4 are shown at baseline (admission) and at wound healing for individual members of both rHGH and placebo groups. Because there were no statistically significant differences between the patients receiving rHGH (n 5 7) and the patients receiving placebo (n 5 8), the observations were combined. Lines connect the dots for each patient from baseline to healed periods. Data obtained from 9 age-matched control children are given as mean 6 SD. Normal range was 177–368 ng/mL. The dotted line extends the mean value of the normal group across the length of the horizontal axis.
FIG. 4. Serum concentrations of IGFBP-5 are shown at baseline (admission) and at wound healing for individual members of the rHGH and placebo groups. Because there were no statistically significant differences between the patients receiving rHGH (n 5 7) and the patients receiving placebo (n 5 8), the observations were combined. Lines connect the dots for each patient from baseline to healed periods. Data obtained from 9 age-matched control children are given as mean 6 SD. Normal range was 184 –248 ng/mL. The dotted line extends the mean value of the normal group across the length of the horizontal axis.
healing to 296 6 79 ng/mL (rHGH) and 283 6 74ng/mL (placebo). Normal range is 200 –700 ng/mL. Lumbar spine BMD was obtained in four of the patients in the rHGH group and five patients in the placebo group. Z-scores were 20.5 6 1.0 (range: 21.26 to 1 0.87) in the rHGH group and 20.7 6 1.2 (range: 21.75 to 11.20) in the placebo group, respectively (P 5 not significant). Discussion
Our data, reported herein, show that rHGH does not increase bone formation or bone density in the short term for burned children. This failure is not caused by a dose-response effect, because we administered a daily dose of rHGH that is four times that given for GH deficiency, in which measurable increases in serum levels of PICP and osteocalcin occur (8 –10). The efficacy of rHGH therapy is generally monitored biochemically by demonstration of a rise in serum levels of IGF-1, and in our study, we observed increases in both IGF-1
23
FIG. 5. Serum concentrations of osteocalcin are shown for the rHGH and placebo groups at baseline (admission) and at wound healing. Data are given as mean 6 SD. The lower limit of normal is designated by a dotted line at the top of the figure. Numbers of patients in each group are given at the top of the bars.
and IGFBP-3 levels in the serum of the burned patients. IGF-1 is reported to be anabolic to bone (19), and serum levels of IGFBP-3 are closely associated with increased lumbar spine BMD in adult males (20). Although rHGH administration in GH deficiency has been reported to increase serum levels of IGFBP-5, a binding protein that may fix IGF-1 to bone (21), no such effect of rHGH was seen in the burn patients, perhaps because of other, as yet unspecified, metabolic abnormalities. However, serum IGFBP-5 levels rose to normal in these children, regardless of GH administration, at the time of wound healing. One possible explanation for the apparent resistance of bone to increased circulating levels of IGF-1, as well as for the previous finding of retardation in growth velocity post burn, may be insufficient length of treatment. Data from Saggese et al. (9) in GH-deficient patients demonstrated that rHGH produces a significant rise in serum levels of PICP after 1 week of treatment with rHGH, and a significant increase occurred in serum osteocalcin concentration by 3 months. Our patients exhibited little change in PICP levels and osteocalcin levels in serum during the study period, which lasted a mean of 6 weeks from time of burn injury, with an upper range of 11 weeks, using doses that were four times those given to GH-deficient children. Though the possibility exists that treatment length was insufficient, it would not be unreasonable to expect increases in serum PICP or osteocalcin levels in at least some of the patients. To explain resistance to rHGH in bone, an alternative to insufficient time of therapy is the rise in serum concentration of IGFBP-4, present on admission and rising further by an apparently growth-hormone-independent mechanism by the time of wound healing. IGFBP-4 is reported to inhibit the anabolic effects of IGF-1 on bone and other tissues, perhaps by decreasing the bioavailability of IGF-1 to the local tissue IGF-1 receptor (16). IGFBP-4 is produced by bone cells (16) and by keratinocytes (22). Its production is stimulated by both 1,25-dihydroxyvitamin D (calcitriol) (15) and by PTH (16, 23). However, serum levels of calcitriol have been reported as either low or normal in burn patients (1). Similarly,
24
KLEIN ET AL.
burned children are hypoparathyroid, as evidenced by subnormal PTH response to the commonly observed low serum concentrations of ionized calcium in these patients (24). Therefore, neither can be considered a stimulus for increased IGFBP-4 production in our patients. We speculate that IGFBP-4 levels are increased by proliferating keratinocytes in the burned child during the process of wound healing. Such increases in IGFBP-4 may divert the anabolic activities of IGF-1 from bone to promote the cutaneous wound healing effects of rHGH in the burned patient (5–7) or other functions, as yet unspecified. Our data demonstrate further that the long-lasting effects of the burn injury on the metabolic aspects of bone function are not easily reversed. The exact mechanism remains uncertain but now can be extended to include the GH-IGF-1 axis, including an important role for IGF-1 binding proteins.
10. 11. 12. 13. 14. 15. 16.
17.
References 1. Klein GL, Herndon DN, Rutan TC, et al. 1993 Bone disease in burn patients. J Bone Miner Res. 8:337–345. 2. Klein GL, Herndon DN, Goodman WG, et al. 1995 Histomorphometric and biochemical characterization of bone following acute severe burns in children. Bone. 17:455– 460. 3. Klein GL, Herndon DN, Langman CB, et al. 1995 Long-term reduction in bone mass following severe burn injury in children. J Pediatr. 126:252–256. 4. Rutan RL, Herndon DN. 1990 Growth delay in postburn patients. Arch Surg. 125:392–395. 5. Herndon DN, Barrow RE, Kunkel KR, Broemeling L, Rutan RL. 1990 Effects of recombinant human growth hormone on donor-site healing in severely burned children. Ann Surg. 212:424 – 431. 6. Gilpin DA, Barrow RE, Rutan RL, Broemeling L, Herndon DN. 1994 Recombinant human growth hormone accelerates wound healing in children with large cutaneous burns. Ann Surg. 220:19 –24. 7. Fleming RYD, Rutan RL, Jahoor F, Barrow RE, Wolfe RR, Herndon DN. 1992 Effect of recombinant human growth hormone on catabolic hormones and free fatty acids following thermal injury. J Trauma. 32:698 –703. 8. Zamboni G, Antoniazzi F, Radetti G, Musumesi C, Tato L. 1991 Effects of two different regimens of recombinant growth hormone therapy on the bone mineral density of patients with growth hormone deficiency. J Pediatr. 119:483– 485. 9. Saggese G, Baroncelli GI, Bertelloni S, Cinquanta L, DiNero G. 1993 Effects
18. 19. 20.
21.
22. 23. 24.
JCE & M • 1998 Vol 83 • No 1
of long-term treatment with growth hormone on bone and mineral metabolism in children with growth hormone deficiency. J Pediatr. 122:37– 45. Delmas PD, Chatelain P, Malaval L, Bonne G. 1986 Serum bone Gla-protein in growth hormone-deficient children. J Bone Miner Res. 1:333–338. Sanchez CF, Goodman WG, Brandli D, et al. 1995 Skeletal response to recombinant human growth hormone (rHGH) in children treated with long-term corticosteroids. J Bone Miner Res. 10:2– 6. Herndon DN, Rutan RL, Alison Jr WG, Cox Jr CC. 1993 Management of burn injuries. In: Eichelberger MR, ed. Pediatric trauma: prevention, acute care, and rehabilitation. St. Louis: Mosby-Yearbook; 568 –590. Nichols Institute Diagnostics. 1997 Directional insert. IGF-1 by IRMA. San Juan Capistrano, CA: Nichols Institute; 1–14. Nichols Institute Diagnostics. 1995 Directional insert for the quantitative determination of IGFBP-3 in human serum. San Juan Capistrano, CA: Nichols Institute; 1–26. Scharla SH, Strong DD, Rosen CJ, et al. 1993 1,25-dihydroxyvitamin D increases expression of BP-4 in human OB like cells in vitro and elevates IGFBP-4 serum levels in vivo. J Clin Endocrinol Metab. 77:1190 –1197. Honda Y, Lansdale EC, Strong DD, Baylink DJ, Mohan S. 1996 Recombinant synthesis of insulin-like growth factor binding protein-4 (IGFBP-4). Development, validation, and application of a radioimmunoassay for IGFBP-4 in human serum and other biological fluids. J Clin Endocrinol Metab. 81:1389 –1396. Mohan S, Libanati C, Dony C, Lang K, Srinavasan N, Baylink DJ. 1995 Development, validation, and application of a radioimmunoassay for insulinlike growth factor binding protein-5 in human serum and other biological fluids. J Clin Endocrinol Metrab. 80:2638 –2645. Reed A, Haugen M, Pachman LM, Langman CB. 1990 Abnormalities of serum osteocalcin in children with chronic rheumatic diseases. J Pediatr. 116:574 –580. McCarthy TL, Centrella M, Canalis E. 1989 Regulatory effects of insulin-like growth factors I and II on bone collagen synthesis in rat calvarial cultures. Endocrinology. 124:301–309. Johansson AG, Forslund A, Hambraeus L, Blum WF, Ljunghall S. 1994 Growth hormone-dependent insulin-like growth factor binding protein is a major determinant of bone mineral density in healthy men. J Bone Miner Res. 9:915–921. Ono T, Kanzaki S, Seino Y, Baylink DJ, Mohan S. 1996 Growth hormone (GH) treatment of GH-deficient children increases serum levels of insulin-like growth factors (IGFs), IGF binding proteins-3 and-5, and bone alkaline phosphatase isoenzyme. J Clin Endocrinol Metab. 81:2111–2116. Wraight CJ, Murashita MM, Russo VC, Werther GA. A keratinocyte cell line synthesizes a predominant IGFBP-4 that modulates IGF-1 action. J Invest Dermatol. 103:627– 631. LaTour D, Mohan S, Linkhart TA, Baylink DJ, Strong DD. 1990 Inhibitory IGFBP: cloning complete sequence and physiological regulation. Mol Endocrinol. 4:1806 –1814. Klein GL, Nicolai M, Langman CB, Cuneo BF, Sailer DE, Herndon DN. 1997 Dysregulation of calcium homeostasis after severe burn injury in children: possible role of magnesium depletion. J Pediatr. 131:246 –251.
