Increased mechanical strength of healing rat tibial fractures treated with biosynthetic human growth hormone

Bone, 11, 233-239 (1990)

8756-3282/90 $3 .00 + .00 Copyright © 1990 Pergamon Press plc

Printed in the USA . All rights reserved .

Increased Mechanical Strength of Healing Rat Tibial Fractures Treated with Biosynthetic Human Growth Hormone B . BAK, P . H . JORGENSEN and T . T . ANDREASSEN Department of Connective Tissue Biology, Institute of Anatomy, University of Aarhus, Aarhus, Denmark Address for correspondence and reprints : Bue Bak, Dept . of Connective Tissue Biology, Institute of Anatomy, University of Aarhus, DK-8000

Aarhus C, Denmark . healing tibial fractures in rats . However, Hsu and Robinson (1969) reported that the healing of tibia) fractures in mice with hereditary pituitary insufficiency did not occur within 13 weeks after fracture, whereas the fractures in the ri unaffected siblings healed routinely . Low GH levels have been reported in delayed fracture union (Misol et al ., 1971) . Harris et al . (1975) studied the biomechanical properties of rabbit tibial fractures but found no effect of GH administration . Northmore-Ball et al . (1980) investigated the effect of bovine GH (5 mg per day, five days a week) on the biomechanical properties of healing femoral fractures in the rat, but found no evidence that GH given alone could stimulate the fracture repair during the first 28 days of healing . Zadek (1961) investigated standardized bone defects in dogs and found that GH administration induced healing in dogs with closed epiphysis in which the bone defects usually do not heal . In a histological study, Tylkowski et al . (1976) found that GH stimulated the healing of standardized bone defects in hypophysectomized, immature rats . In contrast, Herold et al . (1971) did not find any effect of exogenous GH on the healing of bone defects in rabbits and rats when evaluated by radiographs . Thus, the results of in vivo investigations of the effects of GH administration on fracture repair are rather h eterogeneous . GH accelerates longitudinal bone growth in rats through a direct stimulation of the growth plate cartilage (Isaksson et al . 1982 ; Schlechter et al ., 1986; Nilsson et al ., 1987), and from in vitro investigations it is known that GH stimulates the proliferation of cultured chondrocytes (Madsen et al . 1983) and osteoblasts (Ernst and Froesch 1988 ; Slootweg and Van Buul-Offers 1988) . If micro-movements occur between the fractured bone ends, cartilage develops in the fracture callus (Mindell et al . 1971 ; Ashhurst 1986) . The cartilaginous matrix is subsequently replaced by new woven bone by endochondral ossification (Frost 1986 ; Henricson et al . 1987) . It therefore seems likely that GH plays a part in the fracture healing process . The aim of the present experiment was to investigate the influence of GH on the biochemical strength of diaphyseal fractures in the rat healing under non-rigid mechanical conditions .

Abstract The effects of biosynthetic human growth hormone on the biomechanical properties of healing tibial fractures and intact bones in the rat were studied after 20 and 40 days of healing. Growth hormone, 2 .0 mg per kg per day, was given subcutaneously in two daily doses . Control animals were injected with a corresponding volume of saline . After 20 days of fracture healing, there were no differences In mechanical properties between the healing fractures and intact bones . After 40 days, the ultimate load and maximum stiffness of the fractures of the b-hGH injected animals had increased to more than 400% of the corresponding values of the saline injected animals, and ultimate stress and energy absorption at ultimate load had increased to 270% compared with the saline injected animals . Ultimate load, stiffness, and energy absorption of the intact bones increased in the b-hGH injected animals, but no differences were found in ultimate stress values or normalized energy, indicating that the changes in the intact bones were quantitative phenomena . Key Words : Growth hormone-Mechanical strength-FracturesRat

Introduction Conflicting results have been published during the last 50 years concerning the influence of exogenous growth hormone (GH) administration on healing fractures and bone defects (Table I) . Silberberg (1935) found increased healing rates in guinea pigs injected with an acid extract of cattle pituitaries, evaluated by histological and clinical examination . Shepanek (1953) injected mice with GH but did not find any effect on fracture healing . Cordebar (1956) reported a positive effect of GH treatment on the healing of fractures in a series of case stories . Koskinen (1959) demonstrated increased callus formation and faster stabilization in rat tibial fractures when GH was given daily during the healing period . In subsequent clinical studies, Koskinen (1963, 1975) found that GH treatment stimulated the healing of fresh fractures and pseudoarthroses, judged by radiographs and clinical examination . Wray and Goldstein (1966) found no effect of hypophysectomy or GH administration on the mechanical strength in

Materials and Methods Fifty-nine three-month old female Wistar rats (Mollegaard, Lille 233

Guinea pig Mouse Man Rat Dog Man Rat Mouse Rabbit Rat Man' Rabbit Man Ra(4 Rat

? M M,F F ? F,M ? ? F ? M F F,M M M'

Sex Fracture Fracture Fracture Fracture Bone defect Fracture Fracture Fracture Bone defect Bone defect Fracture Fracture Non-union Bone defect Fracture

Model

Dose per injection

Bovine ? ? 0 .05 µg Human 100 Evans units Human 30 tibia units Bovine I mg/kg Human t0 USP' Bovine 50µg Growth hormone deficiency Bovine 4 USP Bovine I USP Low GH levels, no treatment 7 0 .46IU/kg Human 16IU Bovine 3 mg Bovine 5 mg

GH preparation

10-50 days 5 weeks 2-6 weeks 2-4 weeks

3 weeks 16 days

2 per week 1 every 2 days Daily 1 every 2 days Twice a day 5 per week

6-21 days 5-20 days 6-7 weeks 6-22 days 4 weeks 3-10 weeks 28 days

Treatment period

Daily Daily 3 per week Daily 3 per week I every 2 days Daily

Injection frequency

H, C M, H, R C, R H, C, R ? C, P, R M, R H, C, R H, R H C, P M C, P, R H M

Method of evaluation

Stimulation No effect Stimulation Stimulation Stimulation Stimulation No effect Delayed union No effect No effect Delayed union No effect Stimulation Stimulation No effect

Effect

Methods of evaluation : M) Mechanical testing, H) Histological examination, C) Clinical examination, P) Metabolic parameters, R) Radiographic examination . ' 10 USP units of GH and 2 USP units of thyrotropine ; Hereditary pituitary insufficient mice versus normal siblings 2 ; 'Case story . Low growth hormone levels observed in a patient with delayed union ; 4Hypophysectomized rats supplemented with GH versus intact rats ; 'Deduced from body weights .

Misol et al . (1971) Harris et al . (1975) Koskinen et al . (1975) Tylkowski et al . (1976) Nortlmmre-Ball et al . (1980)

Silberberg et al . (1935) Shepanek (1953) Cordebar et al . (1956) Koskinen (1959) 7Adek et al . (1961) Koskinen (1963) Wray and Goldstein (1966) Hsu and Robinson (1969) Herold and Hurvitz (1971)

Species

Table I . Prior investigations of the effect of growth hormone (GH) on fracture heating and bone defects .



B . Bak et al . : Fracture healing and growth hormone

Skensved, Denmark) were used for the experiment . In this breed of Wistar rats, the animals are sexually mature after approximately 50 days. The animals were randomized into two groups for injection with saline or biosynthetic human growth hormone (b-hGH, Nordisk Gentofte AIS, Gentofte, Denmark ; specificity : I mg = 3 LU .) . The dose of b-hGH injected was 2 .0 mg per kg body weight (BM per day . The isotonic b-hGH solution was injected subcutaneously into the nape of the neck in two daily doses (between 8 and 10 a .m . and between 4 and 6 p .m .), and the control animals were injected with a corresponding volume of saline . In both groups of animals the injections were commenced one week before the bones were fractured . The animals were weighed and the doses of b-hGH and saline adjusted once a week, The animals had free access to tap water and pellet food (Altromin diet 1314, Chr . Pedersen Ltd ., Ringsted, Denmark) and were caged with a cycle of 12 hours light and 12 hours darkness .

235

corresponding right tibia using exactly the same procedure, including resection of the fibula . The load and deflection was recorded continuously by transducers coupled to measuring bridges and the signals fed to an x-y recorder. The loaddeflection curves obtained were read by a digitizer into a calculator, and the following parameters were calculated: ultimate load, deflection at ultimate load, maximum stiffness, energy absorption at ultimate load, ultimate stress, elastic modulus and normalized energy absorption . The transverse and antero-posterior diameters of the bone at the point of loading (the fracture line and the corresponding level of the intact bone) were measured by a sliding caliper, and the cross-sectional area was calculated by approximating the outer circumference to an elliptical configuration from which the area of the bone marrow (which is equal to the cross-sectional area of the Kirschner wire) was subtracted, Stress values could then be calculated from the maximum bending moment and the second moment of area (Kenedi 1980) :

Fracture technique The animals were anesthetized with pentobarbital (50 mg per kg, i .p .) . A standardized, closed fracture was produced using the technique described by Bak and Andreassen (1988) . A 1 cm incision was made on the anteromedial aspect of the proximal part of the right tibia . The incision was carried through to the bone, and a 1 nun hole was drilled 4 mm proximal to the distal aspect of the tibial tuberosity and 2 nun medial to the anterior ridge . No reaming of the medullary canal was performed . A closed fracture was performed 2 stun above the tibiofibular junction by three-point bending using specially designed, adjustable forceps with blunt jaws . To minimize soft tissue damage, care was taken not to displace the fracture . Closed medullary nailing was performed with a 0 .73 mm Kirschner wire (ultimate load 35 .5 N, maximum stiffness 62 .2 N/mm, tested under the same conditions as the bones) . The skin was closed with 4-0 monofilament nylon sutures (Prolene®) . The operations were performed under sterile conditions . Radiograms of all fractures were taken immediately after nailing, and animals with fractures outside the specified diaphyseal area or with displaced nails were excluded . The remaining animals in the treatment group and in the control group were randomly divided into two subgroups for testing after 20 and 40 days of fracture healing . Unprotected weight bearing was allowed, and the animals resumed normal activity immediately after recovering from anesthesia . Biomechanical analysis After the healing period, the animals were anesthetized with pentobarbital and killed by exsanguination . Both tibiofibular bones were removed by dissection, and all soft tissues including the periosteum were stripped . The bones were stored in a buffered Ringer's solution (4 °C, pH 7 .4) until testing, which was carried out within three hours . Contact radiographs were taken of the fractures . The fibula was resected and the intramedullary nail removed . The mechanical properties of the healing fracture were analyzed by a destructive three-point bending procedure . The bone was placed on two rounded bars with a distance of 15 mm in a materials testing machine (Alwetron 250, Lorentzen & Wettre, Stockholm, Sweden) and deflected halfway between the supporting bars by another rounded bar on the fracture at the opposite side of the bone with a constant speed of 2 mm per minute . All the bones were oriented alike in the testing machine : the concave facet of the lateral tibial condyle was resting on one of the supporting bars, in order that the load was applied from the medial side of the bone . The left, nonfractured tibia was tested at the same level of the bone as the fracture in the

PL„

N= 4

(1)

M: Maximum bending moment P : Ultimate load Lo : Distance between the supporting bars = length of section

36 - d ) I = 64 (a

(2)

1: Second moment of area a : Height of section b : Width of section d: Diameter of bone marrow

am =

Ma (3)

om : Ultimate stress = stress at the surface of the section . The elastic modulus can he calculated from maximum stiffness, distance between the supporting bars in the bending procedure, and the second moment of area, if it is assumed that a) the cross-sectional area of the bone is constant during loading, b) the shape and area of the cross-section is constant between the supporting bars, c) that the extent of deflection is small, and d) the composition of the bone is homogeneous (Nielsen 1974 ; Kenedi 1980) : L„3K

E= 49T

(4)

E : Elastic modulus (or Young's modulus) K: Stiffness of the bone = maximum slope of load-deflection curve . Normalized energy was calculated as :

W°L„3 241

(5)

W'_ : Normalized energy at ultimate load W: Energy absorbtion at ultimate load . Assumption d) is not valid for the fractures, because the exact geometry and composition of the tissue in the fracture gap are not



236

B . Bak et al . ; Fracture healing and growth hormone

Table H . The effect of biosynthetic human growth hormone (b-hGH ; 2.0 mg/kg BW/day given subcutaneously in two daily doses) on the biomechanical properties of healing tibia) fractures in the rat . Experimental group

Ultimate load (N)

Maximum stiffness (N/mm)

Ultimate stress (Nlmm2)

Energy absorption (N/mm)

Deflection at ultimate load (mm)

20 days healing control (n= 11) b-hGH (n=9)

12 .1 ± 2 .1 14 .8 ± 2 .6

23 .7 ± 5 .2 58 .1 ± 20 .1

3 .6 ± 0 .7 5 .1±2 .0

6 .0±1 .8 4 .3±0 .6

0 .98±0 .15 0 .75±0 .13

40 days healing control (n=8) b-hGH (n= 8)

21 .9 ± 5 .7 98 .3 ± 8 .3***

7 .1±2 .4 19 .4 ± 3 .0**

1 .00±0 .32 0 .51 ± 0 .08

86 365

± 39 t 36***

15 .3 ± 7 .8 41 .3 ± 8 .6**

Mean values ± S .E .M. ; **2p < 0 .01 ; ***2p < 0 .001 ; b-hGH injected animals compared with corresponding controls . N = newton .

known, and therefore the elastic modulus of the fractures cannot be calculated . For the same reason, energy values of the fractures cannot be normalized . At the end of the experiment, the total length of the nonfractured, left tibia was measured from the contact radiogram using a sliding caliper . Of the 59 three-month old animals used, 23 were excludedfour because of technical failure in the osteosynthesis, two because of infection around the intramedullary nail, nine because of bending of the nail and eight animals died during the experiment . Statistical analyses The groups were tested for normal distribution and homogeneity of variances (GI, G2, Kolmogorov-Smirnov's test and F-test), followed by a Student's t-test. In case of heterogeneity of variances, Wilcoxon's nonpaired two sample test was used . 2p < 0 .05 was considered statistically significant . Results Fractures After 20 days of fracture healing, no differences in the biomechanical properties of the healing fractures were found between

the b-hGH injected animals and the controls (Table II) . But after 40 days of fracture healing, ultimate load and maximum stiffness of the fractures in the b-hGH injected animals had increased to values of more than 400% of the corresponding values in the control group (Table II, Figs . 1 and 2, and Fig. 5) . Ultimate stress and energy absorption at ultimate load were increased by approximately 270% in the b-hGH injected animals (Table II, Fig . 4 and Fig . 5) .

Intact bones Forty days after fracture, ultimate load had increased by 22%, maximum stiffness by 19%, and energy absorption at ultimate load by 15% in the nonfractured left tibia in the b-hGH injected animals compared with the controls (Table III, Figs . 2 and 3, and Fig . 5) . But when the values were normalized, no differences in stress values and elastic modulus were found (Table III, Fig . 4 and Fig. 5) . Body weights and tibial lengths are given in Table IV . During the experiment the body weight of the saline injected animals increased by 4% in the 20 days group and 9% in the 40 days group . The body weight of the b-hGH injected animals had increased by 36% in the 20 days group and by 49% in the 40 days group . At termination of the experiment, the total length of the intact CONTROL GROUP/40 DAYS OF HEALING LOAD (newton)

120 (mean values*S .E .M .)

60

03

Fig. 1 . Examples of radiographs of the fractures A) immediately after fracture, B) after 40 days of healing and saline injections, and C) after 40 days of heating and h-hGH injection .

06

09

12

Fig. 2 . Load-deflection curve for fractured and intact tibiae of the saline injected controls . Mean values ± S .E .M. are shown for each deflection increment of 0 .05 mm . The dotted part of the curves represents values where failure has4aken place in some of the healing or intact bones .

B . Bak et al . : Fracture healing and growth hormone

237

Table MI . Effect of biosynthetic human growth hormone (bhGH ; 2 .0 mg/kg B W/day) on biomechanical properties of intact tibial bones in the rat .

Experimental group

Ultimate load (N)

Stress (N/run2)

Stiffness (N/mm)

Elastic modulus (10 1 N/mm)

Energy (Norm)

Normalized energy (103 N)

Deflection at ultimate load (mm)

20 days healing control (n=11) b-hGH (n=9)

98 3 3 107 3 4

245 ± 12 268 3 10

332 ± 7 348 ± 15

13 .2 3 0 .9 13 .630 .7

22 .2 3 1 .6 25 .931 .5

1 .74 3 0 .10 1 .98 *30 .09

0 .40 - 0 .01 0 .43-0 .01

40 days healing control (n=8) b-hGH (n=8)

98 t 2 119 ± 6**

253 t 15 257 3 11

322 ± 9 384 - 21*

13 .011 .1 12 .5-0 .7

23 .1 30 .6 26 .631 .5*

1 .88 1 .71 ± 0 .12

0 .42±0 .01 0 .41-0.01

Mean values 3 S .E.M . ; *2p < 0 .05 ; **p < 0 .01 ; b-hGH injected animals compared with corresponding controls . N = newton .

tibiae in the 20 days group of b-hGH injected animals had increased by 3 .7% compared with the controls, whereas the length of the intact tibiae of the b-hGH injected animals in the 40 days group had increased by 4 .9% compared with the controls .

Discussion In the present experiment, administration of 2 .0 mg of b-hGH per kg BW per day given in two daily injections produced a striking increase in the mechanical strength of healing tibia] fractures in the rat when evaluated by a three-point bending procedure after 40 days of fracture healing . Growth hormone administration also increased the maximum stiffness and the load sustained by the intact bones, but there were no differences between the intact bones of the b-hGH injected animals and the controls in terms of stress values or elastic modulus, indicating that the differences in tetras of ultimate load values and stiffness observed between the nonfractured bones from the b-hGH injected animals and the controls were quantitative rather than qualitative . Some essential factors might explain the results of the fracture healing experiment . In previous investigations of the influence of GH on fracture repair, different types and doses of GH were used, as well as different experimental subjects and

b-hGH TREATED GROUP/40 DAYS OF HEALING

0.0

0.6

02

fracture models. Pituitary derived GH was injected once a day, three to seven days a week . The endogenous GH secretion is highly pulsatile in nature, and in the rat the secretory bursts occur at approximately 3-hour intervals (Tannenbaum and Martin 1976 ; Eden 1979 ; Clark et al . 1987) . Jansson et al . (1982) found that the longitudinal bone growth and weight gain in hypophysectomized rats increased significantly when equal daily doses of GH were given in 2 to 8 daily injections rather than in one single injection . Thus b-hGH was given in 2 daily injections in the present experiment . A total dose of 2 mg per kg BW per day was chosen because this dose has been found to give an optimal increase in mechanical strength and collagen formation in subcutaneously implanted cellulose sponges in rats in a dose-response study (Jorgensen and Andreassen 1987) . The different shape of the load-deflection curves of the fractures after 40 days of fracture healing indicates that the calluses of the b-hGH injected animals have mineralized, whereas the repair tissue of the controls is still mainly fibrous (Fig . 2 and Fig . 3) .

Conclusion Biosynthetic human growth hormone can induce increased mechanical strength development in healing tibial fractures after 40

40 DAYS OF HEALING

12

Fig. 3 . Load-deflection curve for fractured and intact tibiae of the b-hGH injected animals . Mean values 3 S .E .M . are shown for each deflection increment of 0 .05 mm . The dotted part of the curves represents values where failure has taken place in some of the healing or intact bones .

FIg. 4 . Stress-deflection curve for fractured and intact tibiae of the saline injected and the b-hGH injected animals . Mean values S,E .M . are shown for each deflection increment of 0 .05 mm . The dotted part of the curves represents values where failure has taken place in some of the healing or intact bones .



248

B . Bak et al . : Fracture healing and growth hormone

LOAD (N)

STRESS N/rams) Intact

A

B Intact

Intact 250

Intact

Fracture

n'rr

200

100 it 75

150

50

100

Fracture

25

50

L

i 20 days

El

20 days

40 days

NaCL injected

STIFFNESS (N/mm)

40 days

® b hGH injected (2mg/kg/day)

ENERGI (N mm)

C

D

Intact

Intact i Intact

Fracture

I

L

I

Fracture 20

400 L 300

15

200

10

Fracture

Fracture 100

I

5

20 days

40days

20 days

40days

Fig. 5 . Maximum load (A) . maximum stress (B), maximum stiffness (C), and energy absorption at maximum load (D) for intact tibiae and corresponding fractured tibiae after 20 and 40 days of fracture healing .

Table IV . Body weight (BW) and length of the nonfractured (left) tibiae (TL) in the fracture healing experiment, and cross-sectional area of the nonfractured tibiae and the fractures at the point of testing . Experimental group

BW at fracture (grams)

BW at testing (grams)

20 days of healing control (n = 11) b-bGH (n = 9)

227 ± 2 227 ± 4

236 t 309 r

40 days of healing control (n = 8) b-hGH (n = 8)

230 ± 3 229 - 2

251 ± 3 342 ± 12***

3 4***

TL at testing (mm)

Area, intact bone (rams)

Callus area (mm2)

38 .03 ± 0 .19 39 .42 ± 0 .30**

4 .39 ± 0 .14 4 .33 ± 0 .16

22 .2 ± 2 .7 24 .8 ± 2,2

38 .57 ± 0 .26 40 .47 ! 0 .21***

4 .24 ± 0 .20 4 .92 ± 0 .31

16 .3 ± 2 .1 21 .6 ± 3 .5

Mean values ± S .E.M . ; **2p < 0 .01 ; ***2p < 0.001 ; b-hGH injected animals compared with corresponding controls .

B . Bak et al . : Fracture healing and growth hormone

days of fracture healing when 2 .0 mg per kg B W per day is given in two daily doses . Frequency of administration as well as the high dose used might be important in explaining the results obtained in this rat fracture model .

Acknowledgments : The authors would like to thank Klaus Skovbo Jensen, M .Sc ., Ph-D ., Physical Laboratory Ill, The Technical University of Denmark, for his assistance in the field of theoretical biomechanics, and Lotte Kristsensen and Aase Young for skillful technical assistance . This work was supported by The Danish Medical Research Council (grants No . 12-7699 and 12-8077) and Nordic Insulin Foundation .

References Ashhurst, D . E. The influence of mechanical conditions on the healing of experimental fractures in the rabbit : a microscopical study . Phil. Trans. R . Soc. Load. B313 :271-302 ;1986 . Bak, B. ; Andreassen T. T . Reduced energy absorption of healed fracture in the mt. Acta Orrhop . Stand . 59:548-551 ; 1988. Clark . R . G . ; Carlsson, L . M . S .; Robinson, I . C . A . F . Growth hormone secretary profiles in conscious female rats . J. Endocr. 114 :399-407; 1987 . Cordebar, R . ; Guilleman, P. Unction de 1'hormone somatothrope sur Is consolidation osseuse . Press . Med. 64 :363-365; 1956 . Eden, S . Age- and sex-related differences in episodic growth hormone secretion in the rat . Endocrinology 105:555-560 ; 1979 . Ernst, M . ; Proesch, E . R. Growth hormone dependent stimulation of osteoblastlike cells in serum-free cultures via local synthesis of insulin-like growth factor 1. Biochem . Biophys . Res. Commun . 151:142-147 ; 1988 . Frost, H . M . Intermediary organization of the skeleton . Vol . 1 . Boca Raton: CRC Press Inc, 1986:175 . Harris, J . M . ; Rean, D . A . ; Banks, H . H. Effects of phosphate supplementation, thyrocalcitonin, and growth hormone on strength of fracture healing . Surg . Forum 26:519-521 ; 1975 . Henricson, A . ; Hulth, A . ; Johnell, 0. The cartilaginous fracture callus in rats . Acta Orthop. Scand. 58 :244-248;1987 . Herold, H . Z . ; Hurvitz, A. :: Tadmor, A . The effect of growth hormone on the healing of experimental bone defects . Acta Onhop . Stand. 42:377-384; 1971 . Hsu, J . D . ; Robinson, R . A. Studies on the healing of long-bone fractures in hereditary pituitary insufficient mice . J . Stag . Res . 9 :535-536; 1969 . Isaksson, 0 . G . P . ; Jansson, J . 0. ; Cause, 1. A . M . Growth hormone stimulates longitudinal bone growth directly . Science 216 :1237-1239; 1982 . Jansson, J . 0 . ; Albertsson-Wikland, K . ; Eden, S . ; Thorngren, K . G . ; Isaksson, 0. Effect of frequency of growth hormone administration on longitudinal bone growth and body weight in hypophysectomized rats . Acta Physiol . Stand . 114:261-255 ;1982 . Jorgensen, P . H . ; Andreassen, T. T. A dose-response study of the effects of biosynthetic human growth hormone, on formation and strength of granulation tissue . Endocrinology 121 :1637-1641 ; 1987 . Kenedi, R . M . A textbook of biomedical engineering. Glasgow, New Zealand :

239

Blackie & Sao Ltd; 1980:39-54 . Koskineo, E . V . S . The repair of experimental fractures under the action of growth hormone, thyrotropine and cortisone . A tissue analytic, roentgenologic and autoradiographic study . Ann . Chir. Gynaecol. Fe" . suppl . 90; 1959 . Koskinen, E . V. S . The effect of growth hormone and thyrotropin on human fracture healing . A clinical, quantitative radiographic and metabolic study . Anti Onhop . Stand. suppl. 62 ; 1963 . Koskinen, E . V . S . ; Lindholm, R . V . ; Nieminen, R . A . ; Puranen, J . ; Attila, U. Humanes Wachstumshormon bei Frakturen der langen ROhrenknochen min verzugerter Knochenbruchheilung . Med . Welt. 26 :1905-1910; 1975. Madam, K . ; Friberg, U . ; Runs, P . ; Eden, S . ; Isaksson, 0 . Growth hormone stimulates the proliferation of cultured chondrocytes from rabbit ear and mt nb growth cartilage . Nature 304 :545-547 ; 1983 . Mindell, E . R . ; Rodbard, S . ; Kwasman, B . G . Chondrogenesis in bone repair . A study of the healing fracture callus in the rat . Clin . Orthop . 79:187-196 ; 1971 . Misol, S . ; Samara, N . ; Ponseti, 1 . V . Growth hormone in delayed fracture union . Clan . Orehop . 74:206-208 ; 1971 . Nielsen, L . E . Mechanical properties of polymers and composites . Vol . I . St . Louis : Marcel Dekker Inc . ; 1974:39-47 . Nilsson, A . ; Isgaard, J . ; Lindahl, A . ; Peterson, L . : Isaksson, 0 . Effects of unilateral arterial infusion of GH and IGF-1 on tibiat longitudinal bone growth in hypophysectomized rats . Calcif. Tissue Int . 40:91-96 ; 1987 . Nonluoore-Ball, M . D . ; Wood, M . R. ; Meggit, B . F . A biomechanical study of the effects of growth hormone in experimental fracture healing . J . Bone Joint Surg. Br. 62-B:391-396; 1980. Schlechter, N . L . ;Russel, S . M . ; Greenberg, S . ; Spencer, E. M . ; Nicoll, C. S . A direct growth effect of growth hormone in rat hindlimb shown by arterial infusion . Am . J . Physiol. 250:E231-E235 ; 1986 . Shepanek, L. A . The effect of endocrine substances (ACTH and growth hormone) on experimental fractures . Surg. Gynecol . Obster. 96:200-204; 1953 . Silberberg, M . ; Silbetberg, R . Influence of acid extract of cattle pituitary gland on bone repair in young guinea pigs . Proc . Soc . Exp . Riot . Med. 33:177-179; 1935 . Slootweg, M . C . ; Van Buul-Offers, S . C . Direct stimulatory effect of growth hormone on DNA synthesis of fetal chicken osteoblasts in culture . .Acm Endocrinol . 118:294-300 ; 1988 . Tannenbaum, G . S . ; Martin, J. B . Evidence for an endogenous ultradian rhythm governing growth hormone secretion in the rat . Endocrinology 98 :562-570; 1976 . Tylkowski, C . M. ; Wezeman, F. H . ; Ray, R . D . Hormonal effects on the morphology of bone defect heating . C/in . Onhop. 115 :274-285 : 1976 . Wray, J . B . ; Goldstein, J. The effect of the pituitary gland and growth hormone upon the strength of the healing fracture in the mt . J. Bone Joint Surg . 48-A:815-816 ;1966. Zadek, R. E . ; Robinson, R . A . The effect of growth hormone on experimental long-bone defect . J. Bone Joint Surg . 43-A : 1261 ; 1961 .

Received : July 18, 1989 Revised : January 17, 1990 Accepted: March 28, 1990