- Email: [email protected]
Intact fibula improves fracture healing in a rat tibia osteotomy model
ELSEVIER
Journal of Orthopaed ic Research
Journal of Orthopaedic Research 23 (2005) 489493
www.elsevier.com/locate/orthres
Intact fibula improves fracture healing in a rat tibia osteotomy model Sandra J. Shefelbine *, Peter Augat, Lutz Claes, Alexander Beck Institute for Orthopuedics und Biomechanics, Helmholtzstrasse 14, 89081 Ulm, Germany
Abstract
Rat tibia fractures are often used in fracture healing studies. Usually the fracture is stabilized with an intramedullary pin, which provides bending stiffness, but little torsional stiffness. The objective of this research was to determine the in vitro torsional rigidity of an osteotomized tibia with and without the fibula, and to determine if this difference influences the healing process in vivo. In vitro eleven rat tibias received an osteotomy, were stabilized with an intramedullary pin, and were tested in internal rotation to determine the torsional rigidity. The fibula was then manually broken and the torsional rigidity measured again. In vivo 18 rats received a tibia1 osteotomy, eight of which had an additional fractured fibula. After three weeks, the rats were sacrificed and the tibias were analyzed. Bone density in the fracture callus was measured with qCT. Bending rigidity and maximum breaking moment were determined in three-point bending. In vitro testing demonstrated that the torsional rigidity with an intact fibula was nearly two times higher than when the fibula was fractured. Though the torsional rigidity was still small in comparison with an intact bone, it resulted in a significantly different healing process in vivo. Rats with intact fibulas had significantly higher bone mineral density, bending rigidity, and maximum breaking moment compared to rats with a fractured fibula. These results indicate that torsional stability considerably affects the healing process. In a fracture model, it is critical to characterize the mechanical environment of the fracture. 0 2004 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved. Kewords: Rat; Tibia; Fracture; Torsional rigidity; Stability
Introduction Fracture healing studies often use a rat fracture model to examine the effects of certain treatments, such as growth factors [ 16,24,28-301, altered diets [9,12,13,15], or stabilization devices [I 7,18,23,27] on the healing progress. Rat fracture models are inexpensive in comparison with larger animal models, and fracture is possible with a relatively simple, minimally invasive surgical procedure. It is, however, difficult to control and characterize the mechanical environment in a rat fracture. Rat Corresponding author. Tel.: +49 731 500 33754; fax: +49 731 500 23498. E-mail address: [email protected] (S.J. Shefelbine).
fracture models vary in the type of surgery (fracture or osteotomy) and mechanical environment (stability of the fixation device), making it difficult to compare results among studies. Bonnarens and Einhorn [8] modified the method of Jackson et al. [19] to create a reproducible fracture in the rat femur using a guillotine apparatus. In this method the femur is first stabilized with an intramedullary pin. A blunt weight is then dropped onto the femur producing a consistent fracture with minimal soft tissue damage. An et al. modified this method to produce a standard fracture in the rat tibia and fibula [l]. The rat tibia is not surrounded by musculature like the femur and therefore is more accessible for pin insertion and positioning in the guillotine, minimizing soft tissue damage. Others have designed three point bending devices or
0736-0266/$ - see front matter 0 2004 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.orthres.2004.08.007
490
S.J. Shejelbine et al. I Journal of Orthopaedic Research 23 (2005) 489493
Fig. I. Typical stabilization of a rat tibia osteotomy or fracture
special pliers to produce a standard transverse tibial fracture [6,14,18,21,25]. All of these methods are minimally invasive, requiring only a small incision for insertion of the intramedullary pin. Others have used a more invasive osteotomy method, which requires open surgery and disturbs of the musculature and periosteum [11,17,22,23]. However, an osteotomy allows for a more controlled positioning and geometry of the bone defect and makes it possible to leave the fibula intact. In these experimental models the rat osteotomies or fractures are commonly stabilized with an intramedullary pin. Pins can be inserted before or after the fracture or after the osteotomy (Fig. 1). The intramedullary pin provides bending stability to the tibia as it is healing. Previous studies have shown that the rigidity of the pin influences the healing process [18,22,25,27]. In general tibias with flexible pins heal slower and produce more callus than tibias with rigid pins. The effect of rotational fixation stability has been examined in the rat femur, where it was shown that rotational instability delays fracture healing and reduces strength of the fracture callus [17,23]. In the lower limb of the rat the fibula provides some rotational stability. Complete fractures of both the tibia and fibula, therefore, are likely to decrease the rotational stability compared to isolated tibial fractures. The torsional stability in a pinned tibial fracture or osteotomy, however, has not been studied. We hypothesize that the presence of an intact fibula provides more torsional stability than with a fractured fibula and therefore improves fracture healing in a rat tibia osteotomy. The torsional stability of a tibia with and without the fibula was first determined in vitro. The effects of the fibula on fracture healing were then determined in vivo. Methods In vitro In order to determine the difference in initial torsional stability with and without the fibula, osteotomized rat tibias were tested in vitro. Eleven right tibias from male Wistar rats (weight 250-3OOg)
were explanted, leaving the surrounding musculature intact. Each tibia was osteotomized with a triple-trianon coated miniflex dental grinding wheel (thickness 0.15mm, radius 8Omm. grain size S o p . Orthodontics, Germersheim, Germany) proximal to the junction of the tibia and fibula, leaving the fibula intact. A 0.7mm stainless steel Kirshner pin (Mizuho Medical Co., Tokyo, Japan) was inserted anterior to the tibial plateau through the intramedullary canal and osteotomy gap, mimicking the in vivo surgery. Both ends of the bone were potted in cylinders of polymethylmethacrylate (PMMA, Kulzer, Darmstadt, Germany). The bones were tested in internal rotation using a torsion apparatus in a materials testing machine (Zwick 1454. Ulm, Germany). A vertical displacement was applied to a lever arm and measured by a displacement actuator (MT25, Heidenhain, Traunreut, Germany). The required force to displace the lever arm was measured by a 50N load cell (System-Technik GmbH, Germany). The stiffness was measured between 1 and 1.5N at a quasi-static displacement rate of 20mdmin. The torsional rigidity, defined as the product of shear modulus ( c ) and polar moment of inertia (J). was determined by:
where F was the measured vertical force, a was the length of the lever arm (50mm), x was its vertical displacement, and L was the length of the bone. Because the length of the lever arm was much larger than the vertical displacement, errors made by approximating A+ with x1L were less than 1%. After a pre-load cycle, the torsional rigidity was measured three times for each bone, and an average torsional rigidity was calculated. The fibula was fractured manually adjacent to the osteotomy gap, and the test was repeated. Mean and standard deviations were calculated and a paired Wilcoxon test was used to determine significance. In viuo
Surgery was performed on 18 male Wistar rats (weight 300-3508) under halothane inhalation anaesthesia. (A sample size calculation from the in vitro testing revealed that eight rats per group were necessary.) The left leg was shaved, and a longitudinal incision was made on the anterior edge of the tibia. A 0.7mm Kirshner pin (Mizuho Medical Co., Tokyo, Japan) was inserted in the medullary canal without reaming by insertion of the pin slightly anterior to the tibial plateau. After removing the pin, an osteotomy was made 3mm proximal to the tibiafibula junction with a dental grinding wheel. In eight rats the fibula was manually broken adjacent to the osteotomy location; in 10 control rats the fibula was left intact. The pin was reinserted, and the wound was closed with absorbable sutures. Rats were allowed free movement about their cages and were given unrestricted access to food. All animals were sacrificed after 21 days of healing. All procedures were approved by the local animal welfare committee (Regierungsprasidium, Tubingen No. 648). Contact radiographs (43805NX-Ray System, Hewlett Packard, Palo Alto, CA, USA) of the explanted legs were taken at 4SkV, 3mA to verify placement of the pin. After the pin was removed from the tibias, the fractured region was scanned by computed tomography (pQCT 960; Stratec, Pforzheim, Germany) with a voxel size of 0.15 x 0.15mm and a slice thickness of 1.Omm. Five slices were taken: 3 and 6mm distal the osteotomy, at the osteotomy, and 3 and 6mm proximal to the osteotomy. Bone density and area were calculated with software provided by the manufacturer. To assess the mechanical quality of the healed bone, the bending rigidity of the fracture callus was determined in a three-point bending test in the medio-lateral direction in a material testing machine (Zwick 1445, Ulm, Germany). Load was applied quasi-statically at a rate of 1 m d m i n . Deflection of the bone was measured with a displacement actuator (MT25, Heidenhain, Traunreut, Germany). Load was applied until fracture at the osteotomy sight occurred. Bending rigidity, defined as the product of the elastic modulus ( E ) and the moment of inerwas determined by: tia (0, MI-' El=-,
48d
S.J. Shefelbine et al. I Journal of Orthopaedic Research 23 (2005) 489493 where I was the distance between supports (15mm),and d was the deflection. In addition to the bending rigidity, the maximum moment needed to break the fracture callus was measured (MmaX).Mean and standard deviations were calculated and a Wilcoxon test was used to determine significance.
Results
in vitro The in vitro torsional rigidity (GJ) provided a measure of torsional stability of the initial intramedullary pin stabilization with and without the fibula. For osteotomized tibias with the fibula intact, the average torsional rigidity was 1430Nmm2 (SD 406Nmm2). With an additional fibula fracture the torsional rigidity decreased to 689Nmm2 (SD 377Nmm2),which was significantly lower than the intact group (p = 0.001) (Fig. 2). Both groups, however, had very low torsional rigidity in comparison with an intact tibia and fibula (16363Nmm2, SD 4405mm2).
16 $0N mm2 2000.00 1800.00
“E E
1600.00
Z
1400.00
c .-
7
1200.00
(3 .-5
1000.00
n ._ ?
2o .-
5
800.00 600.00
400.00 2w.w 0.00
!LY p = 0.001
fibula
fibula
intact
fractured
intact tibia
Fig. 2. In vitro torsional rigidity for osteotomized tibias with and without the fibula.
a
49 1
in vivo Animals with an additional fibula fracture had delayed healing in comparison to animals with an intact fibula. After 21 days of healing the average bone mineral density in the fracture gaps was 30% lower in the animals with a fibula fracture (394mg/ccm, SD 59) than without (577mg/ccm, SD 53). This difference in bone mineral density measured from CT scans was significantly different. In animals with fibula fractures the bending rigidity (5718Nmm2, SD 2848) was 25% that of animals with intact fibulas (22853Nmm2, SD 507). The maximum moment needed to break the fracture callus was 59% lower in animals with a fibula fracture (65.7Nmm, SD 22.1) than with an intact fibula (159.ONmm, SD 53.2) (Fig. 3). Discussion
This study has demonstrated the importance of the fibula in stabilizing an osteotomized tibia. An intramedullary pin may provide sufficient bending stability during healing, however provides little torsional stability. The in vitro results of this study demonstrated that the fibula provides small, but significant torsional stability to an osteotomized tibia. Whereas the human fibula is a completely separate bone, the rat fibula joins the tibia distally, and therefore may be even more critical to the mechanical environment. Though the torsional rigidity provided by the fibula is minimal compared to an intact tibia, it is enough to affect the healing process. In vivo, rats with intact fibulas had higher bone density in the fracture callus, a higher bending rigidity, and higher maximum breaking moment than those with fractured fibulas, indicating that the healing progressed faster when the bone was torsionally stabilized by the fibula. There is ample evidence that the mechanical environment affects the fracture healing outcome. The mechanical environment can be characterized by the fracture gap size and by the interfragmentary movement. Several
p = 0.0004
T
g
p = 0.0006
20000
‘C
p
15000
5 10000 n 5000
fibula fractured
fibula intact
fibula fractured
Fig. 3. In vivo results. (A) Average callus mineral density. (B) Bending rigidity measured with three-point bending. (C) Bending moment required to break the callus. In all measurements the difference between the two groups was statistically significant.
492
S.J. Shefelbine et al. 1 Journal of Orthopaedic Research 23 (2005) 489493
studies have demonstrated that an increasing gap site results in a delay of healing [4,10]. Also excessive motion at the fracture size induces callus formation but retards the consolidation process of the callus [5,20,27]. Moreover the type of interfragmentary motion affects the healing process. Because torsional instability results in shear stresses in the healing region, the effect of shear on the healing outcome is of particular interest for the results of this study. While axial motion at the fracture site is thought to stimulate the healing process, results are contradictory and inconclusive as to whether shear promotes or delays healing. It was shown that shear motion in a transverse tibial sheep osteotomy decreased the callus size, delayed bone healing, and reduced flexural rigidity in comparison with axial motion [3]. It was also found that oblique osteotomies in the canine tibia had lower bending and torsional stiffness during healing than transverse osteotomies [2]. Delayed healing was attributed to shear motion present in the oblique osteotomy. Other studies have concluded that shear stimulates healing. In a rabbit tibial fracture model, oblique fractures with a sliding fixator developed more callus and had a larger torsional rigidity than transverse fractures with an axial motion fixator [26]. However, it is difficult to determine if the outcome is due to the difference in strain magnitude or type (axiahhear). Similarly, it was found that torsional shear resulted in a larger cartilaginous callus and a higher torsional rigidity than axial compression of equal strain magnitudes in a sheep tibial osteotomy model [7]. The tissue strains in this study were constant during the healing process and imposed manually by the fixator and not by normal activity of the animal. The tissue stresses, therefore, most likely increased dramatically as the tissue calcified. In vitro measurements of stability do not provide exact measurements of the initial conditions of an osteotomy in vivo. All muscles surrounding the tibia were left intact during the in vitro testing. However, in vivo, the muscles are active and may provide additional stability. In vitro measurements, however, provide an estimation of the initial mechanical environment during fracture healing. Osteotomized tibias with an additional fibula fracture had very low torsional stability. In these specimens, the stabilization was provided mainly by the muscles. The torsional rigidity measured was primarily that of the surrounding soft tissue, which often demonstrates creep and non-linear behavior. We minimized the error caused by these factors by examining only the linear part of the stiffness curve and by measuring only the instantaneous response. A reproducible surgical procedure is critical in order to standardize the conditions for all animals. An osteotomy model was used in this study instead of a fracture model to control the geometry and location of the bone defect. The fibula was fractured manually with the fin-
gers, making it difficult to control the location of the break. However, all fractures were transverse and the location of the fibula fracture is not likely to have a significant influence on the torsional stability or healing process. The location of the intramedullary pin may influence the stability of the bone. In all in vitro and in vivo tests, pin placement was verified by radiograph. In all bones, the pin extended at least lOmm past the osteotomy into the distal part of the tibia. The diameter of the pin may also affect torsional stability with larger diameter pins giving a tighter and more stable fit. In preliminary studies pins with diameters of 0.7mm, 0.8mm, and 0.9mm were tested. The 0.9mm diameter pins were too thick to insert into the distal portion of the tibia. There was no difference in torsional stability between ~ the 0.7mm and 0 . 8 pins. Results from various studies can only be compared when the mechanical environment (i.e. stabilization technique) is the same. Most studies assume that an intramedullary pin is sufficient stabilization in a rat femur or tibia fracture. This study shows that an intramedullary pin provides very little torsional stability. In particular if the fibula is also fractured, the bone is extremely torsionally instable. An osteotomized tibia with an intact fibula had only 8% the torsional rigidity of an intact tibia in vitro. When the fibula was fractured the torsional rigidity reduced to 4% that of an intact tibia. The minimal stability that the fibula provided was enough to influence the healing and resulted in bending stiffness of the callus nearly five times higher than with a fractured fibula. These findings indicate the sensitivity of the healing process to the mechanical environment. Because the mechanical environment is critical to the healing process, it is difficult to compare results from studies using different stabilization techniques and fracture methods. We recommend, therefore, that the mechanical conditions be well characterized when using animal models to study various aspects of fracture healing. References [I] An Y, Friedman RJ, Parent T, Draughn RA. Production of a standard closed fracture in the rat tibia. J Orthop Trauma 1994 8:111-5. [2] Aro HT, Wahner HT, Chao EY. Healing patterns of transverse and oblique osteotomies in the canine tibia under external fixation. J Orthop Trauma 1991;5:351-64. [3] Augat P, Burger J, Schorlemmer S, Henke T, Peraus M, Claes L. Shear movement at the fracture site delays healing in a diaphyseal fracture model. J Orthop Res 2003;21:1011-7. [4] Augat P, Margevicius K, Simon J, Wolf S, Suger G. Claes L. Local tissue properties in bone healing: influence of size and stability of the osteotomy gap. J Orthop Res 1998;16:475-81. [5] Augat P, Merk J, Genant HK, Claes L. Quantitative assessment of experimental fracture repair by peripheral computed tomography. Calcif Tissue Int 1997;60:19+9. [6] Bak B, Jensen KS. Standardization of tibial fractures in the rat. Bone 1992;13:289-95.
S.J. Shefelhine et al. I Journal of Orthopaedic Research 23 (2005) 489493
[7] Bishop N, Tami I, van Rhijn M, Corveleijn R, Schneider E, Ito K. Effects of volumetric vs. shear deformation on tissue differentiation during secondary bone healing. In: Transactions of 49th orthopaedic research society 2003;23: 114. New Orleans: Orthopaedic Research Society. [8] Bonnarens F, Einhorn TA. Production of a standard closed fracture in laboratory animal bone. J Orthop Res 1984;297-101. [9] Chakkalakal DA, Novak JR, Fritz ED, Mollner TJ, McVicker DL, Lybarger DL, et al. Chronic ethanol consumption results in deficient bone repair in rats. Alcohol Alcoh 2OO2;37: 13-20. [lo] Claes L, Augat P, Suger G, Wilke HJ. Influence of size and stability of the osteotomy gap on the success of fracture healing. J Orthop Res 1997;15:577-84. [ I I] David R, Nissan M, Cohen I, Soudry M. Effect of low-power HeNe laser on fracture healing in rats. Lasers Surg Med 1996;19: 458-64. [12] Day SM, DeHeer DH. Reversal of the detrimental effects of chronic protein malnutrition on long bone fracture healing. J Orthop Trauma 2001;15:47-53. [13] Einhorn TA, Bonnarens F, Burstein AH. The contributions of dietary protein and mineral to the healing of experimental fractures. A biOmeChdnicdl study. J Bone Joint Surg Am 1986; 68:1389-95. [I41 Ekeland A, Engesaeter LB, Langeland N. Mechanical properties of fractured and intact rat femora evaluated by bending, torsional and tensile tests. Acta Orthop Scand 1981;52:605-13. [I51 Elmali N, Ertem K, Ozen S, Inan M, Baysal T, Guner G, et al. Fracture healing and bone mass in rats fed on liquid diet containing ethanol. Alcohol Clin Exp Res 2002;26:509-13. [I61 Evans BA, Warner JT, Elford C, Evans SL, Laib A, Bains RK, et al. Morphological determinants of femoral strength in growth hormone-deficient transgenic growth-retarded (Tgr) rats. J Bone Miner Res 2003;18:1308-16. [I71 Grundnes 0, Reikeras 0. Effects of instability on bone healing. Femoral osteotomies studied in rats. Acta Orthop Scand 1993; M55-8. [18] Hankemeier S, Grassel S, Plenz G, Spiegel HU, Bruckner P, Probst A. Alteration of fracture stability influences chondrogenesis, osteogenesis and immigration of macrophages. J Orthop Res 2001;19:531-8.
493
[I91 Jackson RW, Reed CA, Israel JA, Abou-Keer FK, Garside H. Production of a standard experimental fracture. Can J Surg 1970; 13:415-20. [20] Kenwright J, Goodship AE. Controlled mechanical stimulation in the treatment of tibia1 fractures. Clin Orthop Re1 Res 1989;241:3&47. [21] Madsen JE, Hukkanen M, Aune AK, Basran I, Moller JF, Polak JM, et al. Fracture healing and callus innervation after peripheral nerve resection in rats. Clin Orthop 1998:230-40. [22] Molster A, Gjerdet NR, Raugstad TS, Hvidsten K, Alho A, Bang G. Effect of instability of experimental fracture healing. Acta Orthop Scand 1982;53:5214. [23] Molster AO. Effects of rotational instability on healing of femoral osteotomies in the rat. Acta Orthop Scand 1984;55:6324. [24] Nakajima A, Shimizu S, Moriya H, Yamazaki M. Expression of fibroblast growth factor receptor-3 (FGFR3), signal transducer and activator of transcription- 1, and cyclin-dependent kinase inhibitor p21 during endochondral ossification: differential role of FGFR3 in skeletal development and fracture repair. Endocrinology 2003;144:465948. [25] Otto TE, Patka P, Haarman HJ. Closed fracture healing: a rat model. Eur Surg Res 1995;27:277-84. [26] Park SH, OConnor K, Sung R, McKellop H, Sarmiento A. Comparison of healing process in open osteotomy model and closed fracture model. J Orthop Trauma 1999;13:11420. [27l Probst A, Jansen H, Ladas A, Spiegel HU. Callus formation and fixation rigidity: a fracture model in rats. J Orthop Res 1999;17: 25640. [28] Rundle CH, Miyakoshi N, Ramirez E, Wergedal JE, Lau KH, Baylink DJ. Expression of the fibroblast growth factor receptor genes in fracture repair. Clin Orthop 2002:253-63. [29] Wildemann B, Schmidmaier G, Brenner N, Huning M, Stange R, Haas NP. Quantification, localization, and expression of IGF-I and TGF-beta1 during growth factor-stimulated fracture healing. Calcif Tissue Int; 2004. [30] Wildemann B, Schmidmaier G, Ordel S, Stange R, Haas NP, Raschke M. Cell proliferation and differentiation during fracture healing are influenced by locally applied IGF-I and TGF-betal: comparison of two proliferation markers, PCNA and BrdU. J Biomed Mater Res 2003;65B: 1 5 M .
Journal of Orthopaed ic Research
Journal of Orthopaedic Research 23 (2005) 489493
www.elsevier.com/locate/orthres
Intact fibula improves fracture healing in a rat tibia osteotomy model Sandra J. Shefelbine *, Peter Augat, Lutz Claes, Alexander Beck Institute for Orthopuedics und Biomechanics, Helmholtzstrasse 14, 89081 Ulm, Germany
Abstract
Rat tibia fractures are often used in fracture healing studies. Usually the fracture is stabilized with an intramedullary pin, which provides bending stiffness, but little torsional stiffness. The objective of this research was to determine the in vitro torsional rigidity of an osteotomized tibia with and without the fibula, and to determine if this difference influences the healing process in vivo. In vitro eleven rat tibias received an osteotomy, were stabilized with an intramedullary pin, and were tested in internal rotation to determine the torsional rigidity. The fibula was then manually broken and the torsional rigidity measured again. In vivo 18 rats received a tibia1 osteotomy, eight of which had an additional fractured fibula. After three weeks, the rats were sacrificed and the tibias were analyzed. Bone density in the fracture callus was measured with qCT. Bending rigidity and maximum breaking moment were determined in three-point bending. In vitro testing demonstrated that the torsional rigidity with an intact fibula was nearly two times higher than when the fibula was fractured. Though the torsional rigidity was still small in comparison with an intact bone, it resulted in a significantly different healing process in vivo. Rats with intact fibulas had significantly higher bone mineral density, bending rigidity, and maximum breaking moment compared to rats with a fractured fibula. These results indicate that torsional stability considerably affects the healing process. In a fracture model, it is critical to characterize the mechanical environment of the fracture. 0 2004 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved. Kewords: Rat; Tibia; Fracture; Torsional rigidity; Stability
Introduction Fracture healing studies often use a rat fracture model to examine the effects of certain treatments, such as growth factors [ 16,24,28-301, altered diets [9,12,13,15], or stabilization devices [I 7,18,23,27] on the healing progress. Rat fracture models are inexpensive in comparison with larger animal models, and fracture is possible with a relatively simple, minimally invasive surgical procedure. It is, however, difficult to control and characterize the mechanical environment in a rat fracture. Rat Corresponding author. Tel.: +49 731 500 33754; fax: +49 731 500 23498. E-mail address: [email protected] (S.J. Shefelbine).
fracture models vary in the type of surgery (fracture or osteotomy) and mechanical environment (stability of the fixation device), making it difficult to compare results among studies. Bonnarens and Einhorn [8] modified the method of Jackson et al. [19] to create a reproducible fracture in the rat femur using a guillotine apparatus. In this method the femur is first stabilized with an intramedullary pin. A blunt weight is then dropped onto the femur producing a consistent fracture with minimal soft tissue damage. An et al. modified this method to produce a standard fracture in the rat tibia and fibula [l]. The rat tibia is not surrounded by musculature like the femur and therefore is more accessible for pin insertion and positioning in the guillotine, minimizing soft tissue damage. Others have designed three point bending devices or
0736-0266/$ - see front matter 0 2004 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.orthres.2004.08.007
490
S.J. Shejelbine et al. I Journal of Orthopaedic Research 23 (2005) 489493
Fig. I. Typical stabilization of a rat tibia osteotomy or fracture
special pliers to produce a standard transverse tibial fracture [6,14,18,21,25]. All of these methods are minimally invasive, requiring only a small incision for insertion of the intramedullary pin. Others have used a more invasive osteotomy method, which requires open surgery and disturbs of the musculature and periosteum [11,17,22,23]. However, an osteotomy allows for a more controlled positioning and geometry of the bone defect and makes it possible to leave the fibula intact. In these experimental models the rat osteotomies or fractures are commonly stabilized with an intramedullary pin. Pins can be inserted before or after the fracture or after the osteotomy (Fig. 1). The intramedullary pin provides bending stability to the tibia as it is healing. Previous studies have shown that the rigidity of the pin influences the healing process [18,22,25,27]. In general tibias with flexible pins heal slower and produce more callus than tibias with rigid pins. The effect of rotational fixation stability has been examined in the rat femur, where it was shown that rotational instability delays fracture healing and reduces strength of the fracture callus [17,23]. In the lower limb of the rat the fibula provides some rotational stability. Complete fractures of both the tibia and fibula, therefore, are likely to decrease the rotational stability compared to isolated tibial fractures. The torsional stability in a pinned tibial fracture or osteotomy, however, has not been studied. We hypothesize that the presence of an intact fibula provides more torsional stability than with a fractured fibula and therefore improves fracture healing in a rat tibia osteotomy. The torsional stability of a tibia with and without the fibula was first determined in vitro. The effects of the fibula on fracture healing were then determined in vivo. Methods In vitro In order to determine the difference in initial torsional stability with and without the fibula, osteotomized rat tibias were tested in vitro. Eleven right tibias from male Wistar rats (weight 250-3OOg)
were explanted, leaving the surrounding musculature intact. Each tibia was osteotomized with a triple-trianon coated miniflex dental grinding wheel (thickness 0.15mm, radius 8Omm. grain size S o p . Orthodontics, Germersheim, Germany) proximal to the junction of the tibia and fibula, leaving the fibula intact. A 0.7mm stainless steel Kirshner pin (Mizuho Medical Co., Tokyo, Japan) was inserted anterior to the tibial plateau through the intramedullary canal and osteotomy gap, mimicking the in vivo surgery. Both ends of the bone were potted in cylinders of polymethylmethacrylate (PMMA, Kulzer, Darmstadt, Germany). The bones were tested in internal rotation using a torsion apparatus in a materials testing machine (Zwick 1454. Ulm, Germany). A vertical displacement was applied to a lever arm and measured by a displacement actuator (MT25, Heidenhain, Traunreut, Germany). The required force to displace the lever arm was measured by a 50N load cell (System-Technik GmbH, Germany). The stiffness was measured between 1 and 1.5N at a quasi-static displacement rate of 20mdmin. The torsional rigidity, defined as the product of shear modulus ( c ) and polar moment of inertia (J). was determined by:
where F was the measured vertical force, a was the length of the lever arm (50mm), x was its vertical displacement, and L was the length of the bone. Because the length of the lever arm was much larger than the vertical displacement, errors made by approximating A+ with x1L were less than 1%. After a pre-load cycle, the torsional rigidity was measured three times for each bone, and an average torsional rigidity was calculated. The fibula was fractured manually adjacent to the osteotomy gap, and the test was repeated. Mean and standard deviations were calculated and a paired Wilcoxon test was used to determine significance. In viuo
Surgery was performed on 18 male Wistar rats (weight 300-3508) under halothane inhalation anaesthesia. (A sample size calculation from the in vitro testing revealed that eight rats per group were necessary.) The left leg was shaved, and a longitudinal incision was made on the anterior edge of the tibia. A 0.7mm Kirshner pin (Mizuho Medical Co., Tokyo, Japan) was inserted in the medullary canal without reaming by insertion of the pin slightly anterior to the tibial plateau. After removing the pin, an osteotomy was made 3mm proximal to the tibiafibula junction with a dental grinding wheel. In eight rats the fibula was manually broken adjacent to the osteotomy location; in 10 control rats the fibula was left intact. The pin was reinserted, and the wound was closed with absorbable sutures. Rats were allowed free movement about their cages and were given unrestricted access to food. All animals were sacrificed after 21 days of healing. All procedures were approved by the local animal welfare committee (Regierungsprasidium, Tubingen No. 648). Contact radiographs (43805NX-Ray System, Hewlett Packard, Palo Alto, CA, USA) of the explanted legs were taken at 4SkV, 3mA to verify placement of the pin. After the pin was removed from the tibias, the fractured region was scanned by computed tomography (pQCT 960; Stratec, Pforzheim, Germany) with a voxel size of 0.15 x 0.15mm and a slice thickness of 1.Omm. Five slices were taken: 3 and 6mm distal the osteotomy, at the osteotomy, and 3 and 6mm proximal to the osteotomy. Bone density and area were calculated with software provided by the manufacturer. To assess the mechanical quality of the healed bone, the bending rigidity of the fracture callus was determined in a three-point bending test in the medio-lateral direction in a material testing machine (Zwick 1445, Ulm, Germany). Load was applied quasi-statically at a rate of 1 m d m i n . Deflection of the bone was measured with a displacement actuator (MT25, Heidenhain, Traunreut, Germany). Load was applied until fracture at the osteotomy sight occurred. Bending rigidity, defined as the product of the elastic modulus ( E ) and the moment of inerwas determined by: tia (0, MI-' El=-,
48d
S.J. Shefelbine et al. I Journal of Orthopaedic Research 23 (2005) 489493 where I was the distance between supports (15mm),and d was the deflection. In addition to the bending rigidity, the maximum moment needed to break the fracture callus was measured (MmaX).Mean and standard deviations were calculated and a Wilcoxon test was used to determine significance.
Results
in vitro The in vitro torsional rigidity (GJ) provided a measure of torsional stability of the initial intramedullary pin stabilization with and without the fibula. For osteotomized tibias with the fibula intact, the average torsional rigidity was 1430Nmm2 (SD 406Nmm2). With an additional fibula fracture the torsional rigidity decreased to 689Nmm2 (SD 377Nmm2),which was significantly lower than the intact group (p = 0.001) (Fig. 2). Both groups, however, had very low torsional rigidity in comparison with an intact tibia and fibula (16363Nmm2, SD 4405mm2).
16 $0N mm2 2000.00 1800.00
“E E
1600.00
Z
1400.00
c .-
7
1200.00
(3 .-5
1000.00
n ._ ?
2o .-
5
800.00 600.00
400.00 2w.w 0.00
!LY p = 0.001
fibula
fibula
intact
fractured
intact tibia
Fig. 2. In vitro torsional rigidity for osteotomized tibias with and without the fibula.
a
49 1
in vivo Animals with an additional fibula fracture had delayed healing in comparison to animals with an intact fibula. After 21 days of healing the average bone mineral density in the fracture gaps was 30% lower in the animals with a fibula fracture (394mg/ccm, SD 59) than without (577mg/ccm, SD 53). This difference in bone mineral density measured from CT scans was significantly different. In animals with fibula fractures the bending rigidity (5718Nmm2, SD 2848) was 25% that of animals with intact fibulas (22853Nmm2, SD 507). The maximum moment needed to break the fracture callus was 59% lower in animals with a fibula fracture (65.7Nmm, SD 22.1) than with an intact fibula (159.ONmm, SD 53.2) (Fig. 3). Discussion
This study has demonstrated the importance of the fibula in stabilizing an osteotomized tibia. An intramedullary pin may provide sufficient bending stability during healing, however provides little torsional stability. The in vitro results of this study demonstrated that the fibula provides small, but significant torsional stability to an osteotomized tibia. Whereas the human fibula is a completely separate bone, the rat fibula joins the tibia distally, and therefore may be even more critical to the mechanical environment. Though the torsional rigidity provided by the fibula is minimal compared to an intact tibia, it is enough to affect the healing process. In vivo, rats with intact fibulas had higher bone density in the fracture callus, a higher bending rigidity, and higher maximum breaking moment than those with fractured fibulas, indicating that the healing progressed faster when the bone was torsionally stabilized by the fibula. There is ample evidence that the mechanical environment affects the fracture healing outcome. The mechanical environment can be characterized by the fracture gap size and by the interfragmentary movement. Several
p = 0.0004
T
g
p = 0.0006
20000
‘C
p
15000
5 10000 n 5000
fibula fractured
fibula intact
fibula fractured
Fig. 3. In vivo results. (A) Average callus mineral density. (B) Bending rigidity measured with three-point bending. (C) Bending moment required to break the callus. In all measurements the difference between the two groups was statistically significant.
492
S.J. Shefelbine et al. 1 Journal of Orthopaedic Research 23 (2005) 489493
studies have demonstrated that an increasing gap site results in a delay of healing [4,10]. Also excessive motion at the fracture size induces callus formation but retards the consolidation process of the callus [5,20,27]. Moreover the type of interfragmentary motion affects the healing process. Because torsional instability results in shear stresses in the healing region, the effect of shear on the healing outcome is of particular interest for the results of this study. While axial motion at the fracture site is thought to stimulate the healing process, results are contradictory and inconclusive as to whether shear promotes or delays healing. It was shown that shear motion in a transverse tibial sheep osteotomy decreased the callus size, delayed bone healing, and reduced flexural rigidity in comparison with axial motion [3]. It was also found that oblique osteotomies in the canine tibia had lower bending and torsional stiffness during healing than transverse osteotomies [2]. Delayed healing was attributed to shear motion present in the oblique osteotomy. Other studies have concluded that shear stimulates healing. In a rabbit tibial fracture model, oblique fractures with a sliding fixator developed more callus and had a larger torsional rigidity than transverse fractures with an axial motion fixator [26]. However, it is difficult to determine if the outcome is due to the difference in strain magnitude or type (axiahhear). Similarly, it was found that torsional shear resulted in a larger cartilaginous callus and a higher torsional rigidity than axial compression of equal strain magnitudes in a sheep tibial osteotomy model [7]. The tissue strains in this study were constant during the healing process and imposed manually by the fixator and not by normal activity of the animal. The tissue stresses, therefore, most likely increased dramatically as the tissue calcified. In vitro measurements of stability do not provide exact measurements of the initial conditions of an osteotomy in vivo. All muscles surrounding the tibia were left intact during the in vitro testing. However, in vivo, the muscles are active and may provide additional stability. In vitro measurements, however, provide an estimation of the initial mechanical environment during fracture healing. Osteotomized tibias with an additional fibula fracture had very low torsional stability. In these specimens, the stabilization was provided mainly by the muscles. The torsional rigidity measured was primarily that of the surrounding soft tissue, which often demonstrates creep and non-linear behavior. We minimized the error caused by these factors by examining only the linear part of the stiffness curve and by measuring only the instantaneous response. A reproducible surgical procedure is critical in order to standardize the conditions for all animals. An osteotomy model was used in this study instead of a fracture model to control the geometry and location of the bone defect. The fibula was fractured manually with the fin-
gers, making it difficult to control the location of the break. However, all fractures were transverse and the location of the fibula fracture is not likely to have a significant influence on the torsional stability or healing process. The location of the intramedullary pin may influence the stability of the bone. In all in vitro and in vivo tests, pin placement was verified by radiograph. In all bones, the pin extended at least lOmm past the osteotomy into the distal part of the tibia. The diameter of the pin may also affect torsional stability with larger diameter pins giving a tighter and more stable fit. In preliminary studies pins with diameters of 0.7mm, 0.8mm, and 0.9mm were tested. The 0.9mm diameter pins were too thick to insert into the distal portion of the tibia. There was no difference in torsional stability between ~ the 0.7mm and 0 . 8 pins. Results from various studies can only be compared when the mechanical environment (i.e. stabilization technique) is the same. Most studies assume that an intramedullary pin is sufficient stabilization in a rat femur or tibia fracture. This study shows that an intramedullary pin provides very little torsional stability. In particular if the fibula is also fractured, the bone is extremely torsionally instable. An osteotomized tibia with an intact fibula had only 8% the torsional rigidity of an intact tibia in vitro. When the fibula was fractured the torsional rigidity reduced to 4% that of an intact tibia. The minimal stability that the fibula provided was enough to influence the healing and resulted in bending stiffness of the callus nearly five times higher than with a fractured fibula. These findings indicate the sensitivity of the healing process to the mechanical environment. Because the mechanical environment is critical to the healing process, it is difficult to compare results from studies using different stabilization techniques and fracture methods. We recommend, therefore, that the mechanical conditions be well characterized when using animal models to study various aspects of fracture healing. References [I] An Y, Friedman RJ, Parent T, Draughn RA. Production of a standard closed fracture in the rat tibia. J Orthop Trauma 1994 8:111-5. [2] Aro HT, Wahner HT, Chao EY. Healing patterns of transverse and oblique osteotomies in the canine tibia under external fixation. J Orthop Trauma 1991;5:351-64. [3] Augat P, Burger J, Schorlemmer S, Henke T, Peraus M, Claes L. Shear movement at the fracture site delays healing in a diaphyseal fracture model. J Orthop Res 2003;21:1011-7. [4] Augat P, Margevicius K, Simon J, Wolf S, Suger G. Claes L. Local tissue properties in bone healing: influence of size and stability of the osteotomy gap. J Orthop Res 1998;16:475-81. [5] Augat P, Merk J, Genant HK, Claes L. Quantitative assessment of experimental fracture repair by peripheral computed tomography. Calcif Tissue Int 1997;60:19+9. [6] Bak B, Jensen KS. Standardization of tibial fractures in the rat. Bone 1992;13:289-95.
S.J. Shefelhine et al. I Journal of Orthopaedic Research 23 (2005) 489493
[7] Bishop N, Tami I, van Rhijn M, Corveleijn R, Schneider E, Ito K. Effects of volumetric vs. shear deformation on tissue differentiation during secondary bone healing. In: Transactions of 49th orthopaedic research society 2003;23: 114. New Orleans: Orthopaedic Research Society. [8] Bonnarens F, Einhorn TA. Production of a standard closed fracture in laboratory animal bone. J Orthop Res 1984;297-101. [9] Chakkalakal DA, Novak JR, Fritz ED, Mollner TJ, McVicker DL, Lybarger DL, et al. Chronic ethanol consumption results in deficient bone repair in rats. Alcohol Alcoh 2OO2;37: 13-20. [lo] Claes L, Augat P, Suger G, Wilke HJ. Influence of size and stability of the osteotomy gap on the success of fracture healing. J Orthop Res 1997;15:577-84. [ I I] David R, Nissan M, Cohen I, Soudry M. Effect of low-power HeNe laser on fracture healing in rats. Lasers Surg Med 1996;19: 458-64. [12] Day SM, DeHeer DH. Reversal of the detrimental effects of chronic protein malnutrition on long bone fracture healing. J Orthop Trauma 2001;15:47-53. [13] Einhorn TA, Bonnarens F, Burstein AH. The contributions of dietary protein and mineral to the healing of experimental fractures. A biOmeChdnicdl study. J Bone Joint Surg Am 1986; 68:1389-95. [I41 Ekeland A, Engesaeter LB, Langeland N. Mechanical properties of fractured and intact rat femora evaluated by bending, torsional and tensile tests. Acta Orthop Scand 1981;52:605-13. [I51 Elmali N, Ertem K, Ozen S, Inan M, Baysal T, Guner G, et al. Fracture healing and bone mass in rats fed on liquid diet containing ethanol. Alcohol Clin Exp Res 2002;26:509-13. [I61 Evans BA, Warner JT, Elford C, Evans SL, Laib A, Bains RK, et al. Morphological determinants of femoral strength in growth hormone-deficient transgenic growth-retarded (Tgr) rats. J Bone Miner Res 2003;18:1308-16. [I71 Grundnes 0, Reikeras 0. Effects of instability on bone healing. Femoral osteotomies studied in rats. Acta Orthop Scand 1993; M55-8. [18] Hankemeier S, Grassel S, Plenz G, Spiegel HU, Bruckner P, Probst A. Alteration of fracture stability influences chondrogenesis, osteogenesis and immigration of macrophages. J Orthop Res 2001;19:531-8.
493
[I91 Jackson RW, Reed CA, Israel JA, Abou-Keer FK, Garside H. Production of a standard experimental fracture. Can J Surg 1970; 13:415-20. [20] Kenwright J, Goodship AE. Controlled mechanical stimulation in the treatment of tibia1 fractures. Clin Orthop Re1 Res 1989;241:3&47. [21] Madsen JE, Hukkanen M, Aune AK, Basran I, Moller JF, Polak JM, et al. Fracture healing and callus innervation after peripheral nerve resection in rats. Clin Orthop 1998:230-40. [22] Molster A, Gjerdet NR, Raugstad TS, Hvidsten K, Alho A, Bang G. Effect of instability of experimental fracture healing. Acta Orthop Scand 1982;53:5214. [23] Molster AO. Effects of rotational instability on healing of femoral osteotomies in the rat. Acta Orthop Scand 1984;55:6324. [24] Nakajima A, Shimizu S, Moriya H, Yamazaki M. Expression of fibroblast growth factor receptor-3 (FGFR3), signal transducer and activator of transcription- 1, and cyclin-dependent kinase inhibitor p21 during endochondral ossification: differential role of FGFR3 in skeletal development and fracture repair. Endocrinology 2003;144:465948. [25] Otto TE, Patka P, Haarman HJ. Closed fracture healing: a rat model. Eur Surg Res 1995;27:277-84. [26] Park SH, OConnor K, Sung R, McKellop H, Sarmiento A. Comparison of healing process in open osteotomy model and closed fracture model. J Orthop Trauma 1999;13:11420. [27l Probst A, Jansen H, Ladas A, Spiegel HU. Callus formation and fixation rigidity: a fracture model in rats. J Orthop Res 1999;17: 25640. [28] Rundle CH, Miyakoshi N, Ramirez E, Wergedal JE, Lau KH, Baylink DJ. Expression of the fibroblast growth factor receptor genes in fracture repair. Clin Orthop 2002:253-63. [29] Wildemann B, Schmidmaier G, Brenner N, Huning M, Stange R, Haas NP. Quantification, localization, and expression of IGF-I and TGF-beta1 during growth factor-stimulated fracture healing. Calcif Tissue Int; 2004. [30] Wildemann B, Schmidmaier G, Ordel S, Stange R, Haas NP, Raschke M. Cell proliferation and differentiation during fracture healing are influenced by locally applied IGF-I and TGF-betal: comparison of two proliferation markers, PCNA and BrdU. J Biomed Mater Res 2003;65B: 1 5 M .