Bone 46 (2010) 852859

Contents lists available at ScienceDirect

Bone
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b o n e

Intermittent PTH(134) does not increase union rates in open rat femoral fractures and exhibits attenuated anabolic effects compared to closed fractures
Magnus Tgil a,c,, Michelle M. McDonald a, Alyson Morse a, Lauren Peacock a, Kathy Mikulec a, Negin Amanat a,b, Craig Godfrey a, David G. Little a,d
a

Orthopaedic Research and Biotechnology, The Children's Hospital at Westmead, Westmead, Australia School of Aerospace, Mechanical and Mechatronic Engineering, University of Sydney, Australia University of Lund, Lund, Sweden d Discipline of Paediatrics and Child Health, Faculty of Medicine, University of Sydney, Australia
b c

a r t i c l e

i n f o

a b s t r a c t
Intermittent Parathyroid Hormone (PTH)(134) has an established place in osteoporosis treatment, but also shows promising results in models of bone repair. Previous studies have been dominated by closed fracture models, where union is certain. One of the major clinical needs for anabolic therapies is the treatment of open and high energy fractures at risk of non-union. In the present study we therefore compared PTH(134) treatment in models of both open and closed fractures. 108 male Wistar rats were randomly assigned to undergo standardized closed fractures or open osteotomies with periosteal stripping. 27 rats in each group were treated s.c. with PTH(134) at a dose of 50 g/kg 5 days a week, the other 27 receiving saline. Specimens were harvested at 6 weeks for mechanical testing (n = 17) or histological analysis (n = 10). In closed fractures, union by any denition was 100% in both PTH(134) and saline groups at 6 weeks. In open fractures, the union rate was signicantly lower (p b 0.05), regardless of treatment. In open fractures the mechanically dened union rate was 10/16 (63%) in saline and 11/17 (65%) in PTH(134) treated fractures. By histology, the union rate was 3/9 (33%) with saline and 5/10 (50%) with PTH(134). Radiological union was seen in 13/25 (52%) for saline and 15/26 (58%) with PTH(134). Open fractures were associated with decreases in bone mineral content (BMC) and volumetric bone mineral density (vBMD) on quantitative computerized tomography (QCT) analysis compared to closed fractures. PTH(134) treatment in both models led to signicant increases in callus BMC and volume as well as trabecular bone volume/total volume (BV/TV), as assessed histologically (p b 0.01). In closed fractures, PTH(134) had a robust effect on callus size and strength, with a 60% increase in peak torque (p b 0.05). In the open fractures that united and could be tested, PTH(134) treatment also increased peak torque by 49% compared to saline (p b 0.05). In conclusion, intermittent PTH(134) produced signicant increases in callus size and strength in closed fractures, but failed to increase the rate of union in an open fracture model. In the open fractures that did unite, a muted response to PTH was seen compared to closed fractures. Further research is required to determine if PTH(134) is an appropriate anabolic treatment for open fractures. 2009 Elsevier Inc. All rights reserved.

Article history: Received 13 August 2009 Revised 9 November 2009 Accepted 10 November 2009 Available online 14 November 2009 Edited by: T. Einhorn Keywords: Fracture healing Non-union Strength Open fracture QCT

Introduction In clinical practice, an open high energy fracture is less likely to go on to uncomplicated union than a closed, less traumatic situation [1,2]. The anabolic response in these cases may be impeded due to tissue disruption around the fracture with avascularity and low oxygen tension. During these situations the non-specic wound repair anabolic phenomena, such as; angiogenesis, cellular recruit Corresponding author. Department of Orthopedics, Lund University Hospital, S-221 85 Lund, Sweden. E-mail address: magnus.tagil@med.lu.se (M. Tgil). 8756-3282/$ see front matter 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2009.11.009

ment and proliferation, are delayed along with the more bone-specic anabolic events such as osteogenic differentiation and bone matrix production. As potent therapies become increasingly available in osteoporosis and metabolic bone diseases, there is growing interest in the pharmacological manipulation of fracture repair. To date such fractures remain problematic, however trials examining Bone Morphogenetic Protein (BMP) treatment have suggested improved outcomes in open tibial fractures [3,4], illustrating the plausibility of pharmaceutical manipulation. A number of bone active agents currently exist for the treatment of osteoporosis and other bone pathologies, many with potential as adjunctive therapies for bone

M. Tgil et al. / Bone 46 (2010) 852859

853

repair. Among these, the best characterized is Parathyroid Hormone (PTH) and in particular the active peptide PTH(134). Parathyroid Hormone is an important factor in bone homeostasis with the major overall effect of conserving calcium. Both anabolic and catabolic actions have been ascribed to PTH. While hyperparathyroidism is associated with bone loss, where resorption dominates, intermittently administered PTH(134) and PTH(184) are known to increase both cortical and trabecular bone mass, with the anabolic effect determining outcome [5]. However, after withdrawal of the drug, the bone mass and strength rapidly return to the pre-treatment values [6]. Clinical applications of PTH(134) have successfully improved bone architecture and quality in patients suffering from osteoporosis related bone loss [7,8]. Due to its potent anabolic effects, an alternate use for PTH(134) might be as a shorter term treatment option for fracture repair, the aim of which is to enhance callus formation and strength. In closed animal fracture models, where non-unions are unlikely, intermittent PTH(134) [912], the application of BMPs [13], and the bisphosphonate zoledronic acid [14,15] have all shown a positive effect on both callus size and strength. Furthermore, in models examining closed fractures with moderately impaired bone healing responses, such as using aged or estrogen decient rats, high doses of PTH still enhanced callus size and strength, although not to the same magnitude [16,17]. We hypothesized that when both the non-specic and specic anabolic responses were blunted, as in open fracture, PTH(134) treatment would not improve union rate. However, in a closed fracture model, where non-specic responses are not impaired, PTH(134) should enhance callus size and strength as previously documented. Methods Experimental design 108 male Wistar rats at 9 weeks of age, mean weight 300 g, were randomly divided into 2 groups with 54 rats each. One group was operated using an open fracture model and in the other group closed fractures were induced. Within each group 27 rats were randomly selected for treatment with PTH(134) and 27 for saline. Thus, in the four treatment groups with 27 animals per group, 17 rats were randomly selected for mechanical testing and 10 for histology. Institutional animal ethics approval was obtained for the study. Anesthetic-related death or post-surgical complications resulted in 2 exclusions. Errors during post harvest sample handling resulted in another 2 exclusions from union analysis and grading and an additional sample was damaged prior to QCT scans. Finally one further sample was damaged prior to processing for undecalcied histology. Final numbers for each set of analysis are outlined in each respective table. Open fracture model The rats were anaesthetized with an intraperitoneal injection of ketamine HCl (75 mg/mL, Parnell Laboratories, Roseberry, Australia) and xylazine (10 mg/mL, Ilium, Smitheld, Australia). An incision was made over the lateral right femur from the level of the lateral condyle distally, to the greater trochanter proximally. The gluteus supercialis and biceps femoris muscles were separated to expose the femur. The periosteum was stripped circumferentially 15 mm proximal and distal to the planned osteotomy. A mid-femoral transverse osteotomy was made with a sagittal saw (Stryker, Kalamazoo, MI, USA) without cooling. An intramedullary 1.1 mm Kirschner wire was inserted at the osteotomy and passed antegrade through the knee before the osteotomy was reduced and the fracture was xed in a retrograde fashion. The wire was cut at the level of the articular surface of the knee. The incision was closed in two layers with an absorbable suture.

Closed fracture model Under the same anesthetic protocol as for open fractures, a 1.1 mm K wire was introduced through the skin and the knee joint into the distal femur and driven up to the greater trochanter in a retrograde manner. A closed fracture was produced by dropping a weight on the femur in a specially designed apparatus [18]. Fracture quality was assessed radiologically using a Digital Faxitron Specimen X-ray System (Buffalo Grove, IL, USA), and animals were excluded and immediately replaced if the fracture did not meet the criteria of a transverse, non-comminuted mid-diaphyseal fracture. 4 fractures were excluded and replaced. In both groups the rats received uid therapy in the form of subcutaneous physiologic saline and buprenorphine (Temgesic, Reckitt and Coleman, Hull, UK) as required as post-surgical analgesia at 0.05 mg/kg 2 times daily. The animals were followed for 6 weeks to harvest for analysis using the following parameters. PTH administration Recombinant PTH(134) was supplied by Auspep (Parkville, Australia) and dosed subcutaneously 5 out of 7 days a week [19] at 50 g/kg/day. An in vitro culture system was used to conrm bioactivity of the PTH(134) stock. Primary murine calvarial osteoblasts were treated for 1 week with 10 nM PTH(134), and this led to a 2-fold increase in alkaline phosphatase activity compared to untreated cells. Evaluation Harvests and radiographs After 6 weeks the rats were euthanized by CO2 inhalation. The right and left femora were dissected free, being careful to maintain the integrity of the callus, and radiographed in a Digital Faxitron Specimen X-ray System (Buffalo Grove, IL, USA). The intramedullary pins were removed. The radiographs were analyzed by a blinded examiner to determine radiological union and each sample was categorized as being completely united (bridged on both sides in a frontal view), partially united (bridged on one side) or not united (Fig. 1). Samples allocated to mechanical testing were wrapped in saline-soaked gauze and stored at 80 C until testing. The samples for histological analysis were xed in 4% paraformaldehyde at room

Fig. 1. Representative X-rays of fracture calluses showing complete union, partial union and non-union.

854

M. Tgil et al. / Bone 46 (2010) 852859

temperature for 5 h and then overnight at 4 C before being transferred to 70% alcohol and stored at 4 C. QCT Quantitative computerized tomography (QCT) analysis was performed as previous on all right femora in the study using a Stratec Research SA machine (Birkenfeld, Germany) with 9 1 mm slices at a slice distance of 1 mm and a voxel size of 70 m through the fracture region [15]. Similarly, left femora were scanned with 5 1 mm slices 2 mm apart in a representative mid-diaphyseal region and data was extrapolated for a comparable region to that scanned on right femora. QCT scans provided accurate data for callus bone volume, volumetric bone mineral density (vBMD) and bone mineral content (BMC). Mechanical testing The left and right femora assigned to torsional mechanical testing were thawed while being kept moist in saline. Each sample was assessed by manual manipulation to assess union. Obvious nonunions were/could not be mechanically tested. The bone ends were embedded in steel hexagonal nuts using a two-part epoxy resin (Ebalta SG 130/PUR 11, Ebalta Kunststoff GmbH, Germany). Both right and left femora were tested in torsion, using previously published methods [13,14,18], in a materials tester (ELF 3400, EnduraTEC, USA) in external rotation at an angular displacement of 6/s to failure. The angular displacement and torque were recorded and maximum failure torque was compared between groups. The stiffness of the bone was calculated by the slope of the linear region of the torque vs angular displacement curve. The energy absorbed to failure was calculated as the area under this curve. Histological analysis Fracture samples were bisected in a caudal plane into medial and lateral halves. The medial halves were decalcied in 0.34 M EDTA at room temperature for 46 weeks and processed and embedded in parafn wax. The lateral halves were processed through graded acetone solutions and embedded undecalcied in resin. Central 5 m

sections were cut from both resin and parafn embedded samples and stained with haematoxylin and eosin, Saffranin O light green, von Kossa and cleared for uorescent label analysis. Histologically, osseous healing was dened using Saffranin O stained sections as both cortices being connected by a bridging bony callus. If the callus was predominantly brous or cartilage tissue on one side it was deemed not united. A more detailed qualitative histological fracture callus grading was performed, on the Saffranin O stained sections, using a modied version of the method of Allen and colleagues [20]. Both anterior and posterior sides of each callus were considered and each sample was graded as either 1 brous nonunion (Allen's grading groups 0 and 1), 2 cartilaginous delayed union (Allen's grading groups 2 and 3) or 3 completely bridged bony union (Allen's grading group 4) (Fig. 2). Callus trabecular bone volume of total volume ratio (Tb. BV/TV) was determined by measuring the trabecular bone area within the neo-cortex of the ossied regions of each callus in von Kossa stained sections. At 10 and 3 days prior to harvest, calcein was administered s.c. at 10 mg/kg to allow for dynamic histomorphometric analysis. The area of calcein labeled bone within each callus was measured including trabecular and neo-cortical bone and the percentage labeled bone area of each callus was determined. To assess the systemic anabolic effects of PTH(134) treatment, the contra-lateral tibia and femora were analyzed. The ratio of tb.BV/TV was calculated in the proximal left tibial metaphysis of all samples. Statistical analyses and power Previous data from rat fracture experiments were used as the basis for the power analysis calculation. For proportions of healed fractures, assuming a non-union rate of 50% for controls, union rates of 95% or higher can be detected with n = 20 with power N0.8 with p b 0.05. Differences in union rates were analyzed by Fishers exact test. For differences in callus volume a sample size of 20 gave a power of N0.9 to detect a difference of 1 standard deviation (SD) with p b 0.05. 15 samples per group were required for mechanical testing analysis based on previous experience of variability in this model [10, 11]. Statistical analysis for mean values was performed with a one-way ANOVA and post-hoc t-tests using the LSD method (SPSS ver 17). The radiographic, histological and mechanical non-union rates were not identical, thus for the primary outcome we chose mechanical union, i.e. the ability to withstand load in the mechanical test. Results Union rate In the closed fractures 100% were united by 6 weeks regardless of method dening union. In contrast, in the open fractures only 13/25 (52%) of the saline and 15/26 (58%) of the PTH(134) treated fractures were united radiographically at this stage (Table 1). The remaining open fractures were graded as either partial unions or non-unions as shown in Fig. 1. All closed fractures were also healed mechanically and could be mechanically tested (Table 1). In open fractures there was no

Table 1 Union rates as dened by radiographic, mechanical and histological methods. Group Radiographic N Saline open PTH(134) open Saline closed PTH(134) closed 25 26 26 26 Unions 13 15 26 26 (52%) (58%) (100%)a (100%)a Mechanical N 16 16 16 17 Unions 10 11 16 17 (62.5%) (68.75%) (100%)a (100%)a Histological N 9 10 10 9 Unions 3 (33%) 5 (50%) 10 (100%)a 9 (100%)a

Fig. 2. Representative sections of fracture sites demonstrating bony union (modied Allen's grading 3), cartilaginous delayed union (modied Allen's grading 2) and brous non-union (modied Allen's grading 1). Saffranin O/light green stained sections. Red staining is specic for cartilage matrix, green bone and brous tissue. Original magnication 2.5.

a Signicant p b 0.01 compared to saline open and PTH(134) open by Fishers exact test.

M. Tgil et al. / Bone 46 (2010) 852859

855

Table 2 Mineralised bone properties as analyzed by QCT. An increased callus volume and bone mineral content were found in the PTH treatment groups both in the open and the closed fracture models. Closed fractures showed increased BMC and BMD compared to open fractures when treated with saline only. PTH treatment resulted in increases in all parameters in closed fractures compared to open. Right femur Model Open fracture Treatment N BMC (mcg) vBMD (mcg/mm3) 915.1 976.3 1005.4 1108.6 (67.1) (75.2) (53.1) (46.0), Volume (mm3) 123.3 165.6 131.9 202.6 (30.1) (42.6) (23.4) (47.8),

Saline 25 111.2 (23.6) PTH 25 159.6 (40.1) Open fracture Saline 26 132.2 (22.8)# PTH 26 222.5 (47.7), p b 0.01 vs saline open, #p b 0.05 vs saline open, p b 0.01 vs saline closed, p b 0.01 vs PTH open. Left femur Model Open fracture Closed fracture Treatment Saline PTH Saline PTH N 25 25 26 26 BMC (mcg) 98.0 107.9 104.7 109.1 (6.0) (7.9) (7.8), (8.7) vBMD (mcg/mm3) 1207.2 1215.7 1222.4 1229.0 (20.4) (30.5) (21.4)# (19.1)

Volume (mm3) 81.2 89.0 85.7 88.8 (5.5) (8.8) (6.8)# (7.5)

Cortical thickness (mm) 1.0 (0.1) 1.1 (0.1), 1.0 (0.1) 1.1 (0.1),

p b 0.01 vs saline open, #p b 0.05 vs saline open, p b 0.05 vs saline closed.

improvement in the union rate as dened mechanically with 10/16 (63%) of the saline treated fractures united at 6 weeks, vs 11/16 (69%) in the PTH(134) group. Histologically, while bony union was again seen in all fractures in the closed group, in the open fracture group only 3/9 (33%) of saline treated and 5/10 (50%) of PTH(134) treated open fractures were united. QCT analysis Open fractures demonstrated altered bone mineral properties. BMC and vBMD were decreased by a mean of 16% (p b 0.05) and 9% (p b 0.01) respectively in saline treated open fractures compared closed fractures treated with saline. In closed fractures, PTH(134) treatment increased callus volume was increased by mean 54% and callus BMC by 68% over saline treatment (p b 0.01). In open fractures, PTH(134) treatment resulted in a mean 34% increase in callus volume and a mean 44% increase in callus BMC compared to saline (p b 0.01, Table 2). PTH(134) treatment led to more pronounced differences in closed fractures with a mean 39% increase in callus BMC, a 14% increase in callus vBMD and a 22% increase in callus volume compared to PTH(134) treated open fracture specimens (p b 0.01). Mechanical testing Only those fractures that had united could be mechanically tested. For closed fractures, all were united and therefore tested. Data for all treatment groups is outlined in Table 3. A 60% mean increase in peak torque and a 36% increase in stiffness were found with PTH(134) treatment over saline (p b 0.05). In united open fractures, PTH(134) treatment (11/16) produced a mean 49% increase in peak load and a

48% increase in stiffness compared to united saline treated open fractures (10/16) (p b 0.05). Mechanical properties were also inuenced by the open fracture injury. In united open fractures peak torque was reduced by mean 50% compared to closed fractures, both in the PTH(134) and saline treated animals (p b 0.05). Importantly, PTH(134) treated open and united fractures did not reach the strength of a saline treated closed fracture, with a mean 25% lower peak torque observed (p b 0.05). Also the energy to failure was 8183% lower in open fractures compared to closed fractures, in both the saline and PTH(134) treated groups (p b 0.05). The energy to failure was increased ve-fold in closed over open fractures for both PTH(134) and saline treatment (p b 0.05, Table 3). Histological analysis Fracture callus. The modied Allen's grading of all histological samples revealed extensive differences between the two fracture models with 100% (10/10) of the saline treated, closed fractures given a grading of 3 (complete union) whereas only 33% (3/9) of the saline treated, open fractures were graded 3 (p b 0.01 Fishers exact test, Fig. 3). As expected PTH(134) treatment also led to 100% of the closed fracture samples being graded 3 (9/9), in contrast only 50% (5/10) of PTH(134) treated open fractures were graded 3 (p b 0.05 Fishers exact test). Within open fractures, the number of completely united fractures was not different between PTH(134) and saline treated groups, with 50% and 33% unions respectively. When considering those fractures that were not united (grades 1 and 2), all remaining saline treated open fractures 66% (6/9) were graded 1, brous non-

Table 3 Mechanical properties of both right (fractured) and left (un-fractured) femora. Increase in callus peak torque, stiffness and energy to failure were noted with PTH treatment in both open and closed fracture models. Differences between these models were also demonstrated with open fractures showing a 50% reduction in callus load compared to closed. This difference was evident with and without PTH treatment. Model Right femur Open fracture Closed fracture Treatment Saline PTH Saline PTH N 10 11 16 17 Peak torque (Nm) 0.30 0.45 0.60 0.96 (0.10) (0.09) (0.10) (0.20), Stiffness (Nm/deg) 0.042 0.062 0.067 0.092 (0.007) (0.014) (0.012) (0.027), Energy to failure (Nm deg) 2.96 3.61 17.33 19.10 (1.02) (0.99) (2.21) (2.22),

Left femur Open fracture Closed fracture

Saline PTH Saline PTH

10 11 16 15

0.50 0.54 0.49 0.53

(0.08) (0.15) (0.21) (0.18)

0.047 0.055 0.049 0.048

(0.007) (0.008) (0.013) (0.015)

4.84 5.77 6.82 6.85

(0.86) (0.72) (0.75) (0.96)

p b 0.05 vs saline open, p b 0.05 vs saline closed, p b 0.05 vs PTH open.

856

M. Tgil et al. / Bone 46 (2010) 852859

(p b 0.01) and in open fractures from 42% in saline treated to 74% with PTH(134) treatment (p b 0.01, Fig. 4a). Large differences were seen in uorochrome labeling between open and closed fractures. In saline treated closed fracture calluses, there was a 134% increase in uorochrome labeled area, i.e. sites of active bone formation, compared to saline treated open fractures (p b 0.01) and a 92% increase compared to PTH(134) open treated fractures (p b 0.05). Furthermore, PTH(134) treatment in closed fractures showed a 87% increase in labeled callus area compared to PTH(134) treated open fractures (p b 0.05). When comparing PTH(134) and saline treatment, uorochrome labeled area was not signicantly increased within either open or closed fracture groups. When the uorochrome labeled area was normalised to bone tissue area (LBA/TBA), no differences in labeled bone area were noted between saline and PTH(134) within open fractures or closed fractures. However PTH(134) treated open fractures showed reductions in LBA/TBA of 41% and 55% compared to both PTH and saline treated closed fractures respectively (p N 0.05, Fig. 4b).
Fig. 3. Union rates as dened histologically. As outlined in Fig. 2 histology samples were graded morphologically as unions, delayed unions or non-unions. All fractures healed in the closed model, however only 33% of saline treated and 50% of PTH treated open fractures showed bony union. Note the larger proportion of delayed unions (grade 2) in the PTH open fracture group compared to saline open fractures.

unions. Interestingly, PTH(134) treatment led to only 10% (1/10) of samples graded as a 1, the remaining 40% (4/10) of samples being graded as cartilaginous delayed unions (2) (p b 0.02 Fishers exact test). The trabecular bone volume ratio of the total volume (tb.BV/TV) within the newly formed calluses was signicantly increased in closed fractures from 31% in saline treated to 92% with PTH(134) treatment

Systemic effects of PTH(134). The metaphyseal tb.BV/TV in the left proximal tibia was increased 72% and 116% by PTH(134) over saline treatment in closed and open fracture samples respectively (p b 0.01). These increases were due to increases in trabecular thickness (p b 0.05) and not trabecular number (Figs. 4c and d). As well as increasing trabecular bone tissue, PTH(134) treatment led to significant changes in cortical bone parameters as shown in Table 2. A 10% increase in cortical bone BMC and a 9% increase in cortical bone volume of the non-operated femoral diaphyses were produced with PTH(134) treatment compared to saline in open fractures (p b 0.01). Whereas PTH(134) treatment only produced a 4% increase in cortical

Fig. 4. Bar charts of mean values for (a) callus trabecular BV/TV, (b) labeled bone area of total bone area, (c) proximal tibial trabecular BV/TV and (d) trabecular thickness. Error bars represent 1 standard deviation.

M. Tgil et al. / Bone 46 (2010) 852859

857

BMC compared to saline in closed fractures (p b 0.05). Cortical thickness was also increased 8% with PTH(134) treatment compared to saline in the open fracture groups (p b 0.01) and 7% compared to saline in closed fracture groups (p b 0.01). These increases in cortical thickness were a result of a mean 3%, (p = 0.053) increase in periosteal circumference as well as a mean 9% (p b 0.05) decrease in endosteal circumference with PTH(134) treatment compared to saline. As a result of these changes in cortical bone mass, non-operated limbs showed a 17% increase in stiffness and a 19% increase in energy to failure in PTH(134) treated open fracture samples compared to saline (p b 0.01). Discussion In animal models of fracture repair, exogenously administered PTH or its analogues has produced increases in bone mass locally at the fracture site. These increases are commonly associated with enhanced callus mechanical properties with PTH treatment and these results have been replicated in rats [10,11,12,2123] mice [24] and nonhuman primates [25]. In a recent study using a closed rat fracture model, 8 weeks of PTH treatment increased the strength of the fractured bone compared to vehicle by approximately 50% in rats [26], similar to the ndings of the current investigation. Withdrawal from PTH treatment for another 8 weeks led to normal restoration of remodeling but still increased bone mineral content and strength compared to untreated controls. Delayed hard callus remodeling has been raised as a complication of long acting anti-resorptive therapies such as bisphosphonates during bone repair. These agents are capable of producing the same approximate 50% increase in strength at the time of healing as PTH [14,15], however long term analysis shows extensive delays in remodeling with these agents, even if BP treatment is ceased at the time of union [15]. Hence, short term PTH(134) treatment has become an attractive option in the eld of orthopaedics to hasten the healing process or increase the union rate, without interfering with the later stage of remodeling of the hard callus. All of these previous studies however examined the anabolic potential of PTH(134) in relatively unchallenged models of bone repair, in which even control treated samples heal in a timely manner. One recent study of PTH(134) in mice who had undergone tibial diaphyseal osteotomy fractures showed no increase in mechanical strength but some enhancement in callus osteoid area, BMC and bone fraction volume at 18 days post fracture [27]. Union as determined by bridged cortices on micro CT images was not improved with PTH(134) treatment in this model. In the present study, fracture union rate was assessed radiologically, histologically and by mechanical manipulation. None of the methods demonstrated any improvements in union rates for PTH(134) over saline treated open fractures. Although union rates were not improved, PTH(134) treatment did produce increases in callus volume, BMC and BV/TV compared to saline treatment. The increases in callus mineral properties realised with PTH(134) treatment correlated with enhanced mechanical strength in both open and closed models. Peak torque of the calluses was increased by a similar percentage with PTH(134) treatment in both closed and open fractures compared to their respective saline controls. It is important to note however that in the open fracture groups only those samples that had united were tested for mechanical outcomes, so that these increases only applied to fractures where a sufcient anabolic response led to union. The true mean values for strength and stiffness in open fractures would therefore be considerably less than the gures reported here for united fractures. Even though PTH(134) treatment did improve callus size and strength in both open and closed fractures, differences in callus healing and formation response to PTH(134) in the two models of repair were evident. There was signicant attenuation of the response to PTH(134), with signicant differences in the percentage increase of

callus volume and BMD in PTH(134) treated open fractures compared to closed fractures. Furthermore, these differences were associated with an approximate 50% reduction in peak torque and an approximate 80% reduction in energy to failure in open fractures compared to closed fractures treated with either PTH(134) or saline. PTH(134) treatment also did not fully rescue the mechanical properties of the callus in open fractures, even in those that united, with the peak torque for PTH(134) treated open fractures signicantly lower than saline treated closed fractures. Histological analysis provided an interesting insight into the potential mechanisms behind these outcomes. Apart from the obvious differences in union rates between the two fracture models, further differences in the response to PTH(134) treatment between the two were seen with detailed histological measures. Callus Tb.BV/TV was increased almost 3-fold (values from 31% to 92%) over saline with PTH(134) treatment in closed fractures, whereas within open fracture calluses, this increase did not reach 2-fold (values from 42% to 74%) (Fig. 4a). Hence, even in ossifying regions of the impaired healing calluses of the open fracture model, the effect of PTH(134) on anabolism was blunted. This was further supported by the labeled bone area of callus bone area being signicantly reduced in PTH(134) open fractures compared to both saline and PTH(134) treated closed fractures (Fig. 4b). Potential alterations in the bio-distribution of the drug to the fracture site could be one reason for these diminished responses, as well as differences in cell recruitment. These are areas for further study that were not able to be examined in this study. Analysis of union histologically revealed that, as expected, all closed fractures showed complete bony bridging by the 6 week end time point. Open fractures on the other hand showed only 33% full bony unions. The remaining samples contained mainly extensive brous tissue or cartilaginous callus (Figs. 2 and 3). PTH(134) treatment had no effect on the total number of fractures showing bony union, however within the non-united fractures, a possible PTH(134) effect was evident. A larger proportion of the non-united fractures exhibited cartilaginous callus with PTH(134) treatment compared to saline, as illustrated by a signicant increase in the number of samples graded 2 compared to saline (p b 0.02). Increases in callus cartilage content have been noted previously with PTH(134) treatment [23,24], and was suggested to be the mechanism through which PTH(134) increases the callus size, by enhancing chondrogenic differentiation and proliferation as well as the subsequent osteogenic response. By speculation, a cartilaginous callus might more likely reach bony union than a brous callus, but at the standard time point examined in the present study, union rate was unchanged. In bone remodeling it has been suggested that PTH(134) increases the number of osteoblasts by recruitment of resting lining cells into active osteoblasts rather than affecting mitosis of the osteoprogenitor cells [28,29]. However, in bone repair, PTH(134) in closed fractures has been associated with early increases in mesenchymal proliferation as well as proliferative effects on cartilage production [11,23,24]. Furthermore, increased expression of osterix (Osx), a marker of osteoblastic cell commitment, in the fracture callus has been shown following PTH treatment in a mouse model, supporting its ability to enhance osteogenic differentiation [30]. In challenged areas such as in an open fracture with necrotic bone ends and periosteal and vascular destruction, less cells are immediately available for recruitment. Our results suggest that PTH(134) treatment may not be capable of stimulating the recruitment of progenitor cells from other potential sources in the surrounding tissues. In contrast, the BMPs, which are also considered pro-anabolic for bone, are capable of inducing bone differentiation in mesenchymal cells in surrounding tissue, thereby contributing to the proliferation of osteoprogenitor cells at the fracture site [31]. BMPs also have effects on the vasculature [32] and in a critical sized defect, where complete absence of bone healing is noted in controls, BMPs have been used to produce bony bridging [33]. Thus, in open fractures BMPs appear to be

858

M. Tgil et al. / Bone 46 (2010) 852859 type on fracture healing: a review of 104 consecutive tibial shaft fractures. Arch Orthop Trauma Surg 2001;121:3258. Govender S, Csimma C, Genant HK, Valentin-Opran A, Amit Y, Arbel R, et al. BMP-2 Evaluation in Surgery for Tibial Trauma (BESTT) Study Group. Recombinant human bone morphogenetic protein-2 for treatment of open tibial fractures: a prospective, controlled, randomized study of four hundred and fty patients. J Bone Joint Surg 2002;84A:212334. Ristiniemi J, Flinkkil T, Hyvnen P, Lakovaara M, Pakarinen H, Jalovaara P. RhBMP7 accelerates the healing in distal tibial fractures treated by external xation. J Bone Joint Surg 2007;89B:26572. Oxlund H, Ejersted C, Andreassen TT, Torring O, Nilsson MH. Parathyroid hormone (134) and (184) stimulate cortical bone formation both from periosteum and endosteum. Calcif Tissue Int 1993;53:3949. Ejersted C, Oxlund H, Eriksen EF, Andreassen TT. Withdrawal of parathyroid hormone treatment causes rapid resorption of newly formed vertebral cancellous and endocortical bone in old rats. Bone 1998;23:4352. Neer RM, Arnaud CD, Zanchetta JR, Prince R, Gaich GA, Reginster JY, et al. Effect of parathyroid hormone (134) on fractures and bone mineral density in postmenopausal women with osteoporosis. N Engl J Med 2001;344:143441. Cosman F, Nieves JW, Zion M, Barbuto N, Lindsay R. Effect of prior and ongoing raloxifene therapy on response to PTH and maintenance of BMD after PTH therapy. Osteoporosis Int 2008;19:52935. Andreassen TT, Ejersted C, Oxlund H. Intermittent parathyroid hormone (134) treatment increases callus formation and mechanical strength of healing rat fractures. J Bone Miner Res 1999;14:9608. Alkhiary YM, Gerstenfeld LC, Krall E, Westmore M, Sato M, Mitlak BH, et al. Enhancement of experimental fracture-healing by systemic administration of recombinant human parathyroid hormone (PTH 134). J Bone Joint Surg 2005; 87A:73141. Nakajima A, Shimoji N, Shiomi K, Shimizu S, Moriya H, Einhorn TA, Yamazaki M. Mechanisms for the enhancement of fracture healing in rats treated with intermittent low-dose human parathyroid hormone (134). J Bone Miner Res 2002;17:203847. Warden SJ, Komatsu DE, Rydberg J, Bond JL, Hassett SM. Recombinant human parathyroid hormone (PTH 134) and low-intensity pulsed ultrasound have contrasting additive effects during fracture healing. J Bone Min Res 2009;44: 48594. Einhorn TA, Majeska RJ, Mohaideen A, Kagel EM, Bouxsein ML, Turek TJ, et al. A single percutaneous injection of recombinant human bone morphogenetic protein-2 accelerates fracture repair. J Bone Joint Surg 2003;85A:142535. Amanat N, McDonald M, Godfrey C, Bilston L, Little D. Optimal timing of a single dose of zoledronic acid to increase strength in rat fracture repair. J Bone Miner Res 2007;22:86776. McDonald MM, Dulai S, Godfrey C, Amanat N, Sztynda T, Little DG. Bolus or weekly zoledronic acid administration does not delay endochondral fracture repair but weekly dosing enhances delays in hard callus remodeling. Bone 2008;43:65362. Andreassen TT, Fledelius C, Ejerstad C, Oxlund H. Increases in callus formation and mechanical strength of healing fractures in old rats treated with parathyroid hormone. Acta Orthop Scand 2001;72:3047. Jahng JS, Kim HW. Effect of intermittent administration of parathyroid hormone on fracture healing in ovariectomized rats. Orthopaedics 2000;23:108994. Bonnarens F, Einhorn TA. Production of a standard closed fracture in laboratory animal bone. J Orthop Res 1984;2:97101. Armamento-Villareal R, Ziambaras K, Abbasi-Jarhomi SH, Dimarogonas A, Halstead L, Fausto A, et al. An intact N terminus is required for the anabolic action of parathyroid hormone on adult female rats. J Bone Miner Res 1997;3: 38492. Allen HL, Wase A, Bear WT. Indomethacin and aspirin: effect of nonsteroidal antiinammatory agents on fracture repair in the rat. Act Orthop Scand 1980;51:595600. Holzer G, Majeska RJ, Lundy MW, Hartke JR, Einhorn TA. Parathyroid hormone enhances fracture healing. A preliminary report. Clin Orthop Relat Res 1999;366: 25863. Jahng JS, Kim HW. Effect of intermittent administration of parathyroid hormone on fracture healing in ovariectomized rats. Orthopedics 2000;23:108994. Nakazawa T, Nakajima A, Shiomi K, Moriya H, Einhorn TA, Yamazaki M. Effects of lowdose, intermittent treatment with recombinant human parathyroid hormone (134) on chondrogenesis in a model of experimental fracture healing. Bone 2005;37:7119. Kakar S, Einhorn TA, Vora S, Miara LJ, Hon G, Wigner NA, et al. Enhanced chondrogenesis and Wnt signaling in PTH-treated fractures. J Bone Miner Res 2007;22:190312. Manabe T, Mori S, Mashiba T, Kaji Y, Iwata K, Komatsubara S, et al. Human parathyroid hormone (134) accelerates natural fracture healing process in the femoral osteotomy model of cynomolgus monkeys. Bone 2007;40:147582. Andreassen TT, Willick GE, Morley P, Whiteld JF. Treatment with parathyroid hormone hPTH(134), hPTH(131), and monocyclic hPTH(131) enhances fracture strength and callus amount after withdrawal fracture strength and callus mechanical quality continue to increase. Calcif Tissue Int 2004;74:3516. Gardner MJ, Van der Meulen MCH, Carson J, Zelken J, Ricciardi BF, Wright TM, et al. Role of parathyroid hormone in the mechanosensitivity of fracture healing. J Orthop Res 2007;25:147480. Li M, Liang H, Shen Y, Wronski TJ. Parathyroid hormone stimulates cancellous bone formation at skeletal sites regardless of marrow composition in ovariectomized rats. Bone 1999;24:95100. Dobnig H, Turner RT. Evidence that intermittent treatment with parathyroid hormone increases bone formation in adult rats by activation of bone lining cells. Endocrinology 1995;136:36328.

suitable anabolic agents, increasing the rate of union [3,4]. In our current study, PTH(134) failed to have a major impact on fracture repair in the open model, despite its previously noted positive effects in closed fractures in normal, aged and ovariectomized rats [10 12,16,17,21,22]. In addition, recently the combination of BMP-7 and PTH in a metaphyseal defect model produced additive increases in bone formation compared to BMP-7 or PTH alone [34]. In this study differences in repair sites were also noted between BMP-7 and PTH, implying that these two anabolic stimuli act through different mechanisms. The authors suggest that BMP-7 stimulated stem cell recruitment and osteogenic differentiation, whereas PTH promoted osteoblast differentiation and bone remodeling, leading to better integration of the new and old bone matrix. The interpretation of the outcomes of the current study is limited by the open fracture model and the PTH(134) dose levels used. With variations in union outcomes within this delayed/non-union model, larger sample sizes and an increased number of time points are required to address the mechanistic questions opened up by this investigation. This may aid in determining whether PTH(134) treatment can enhance either mesenchymal proliferation or chondrogenesis in impaired healing sites. Systemic analysis of the effects of PTH(134) in this study demonstrated the efcacy of the 50 g/kg/day dose rate in increasing bone mass in both trabecular and cortical bone, as demonstrated previously [5,6,28]. The dose used in this study is high in comparison to the clinical dose level of between 20 and 40 g/day, but also lower than in some previous fracture experiments [9,21,26]. A lower dose of 10 g/ kg/day has been shown to be effective in rats, with increases in load to failure of 55% [11] while 5 g/kg/day was not effective in changing mechanical properties in closed fractures [10]. We chose the dose based on pilot data in our laboratory showing that PTH(134) 50 g/kg/day 5 of 7 days was highly effective in closed fractures and systemically. Even with this high dose, this study has outlined extensive differences between closed low trauma fractures and traumatic open fractures, and their response to PTH(134) intervention. PTH(134) clearly enhanced the normal response to injury in low trauma injuries, but did not produce union in the challenged open model. Assuming a dose-dependent response, the effects on open fractures with lower doses may be even less impressive than presented here. Other agents that exhibit non-specic and specic bone anabolic effects that may be useful as part of therapeutic manipulation include EP2 agonists [35] and statins [36] or agents that stimulate VEGF production [37]. PTH(134) as an adjunctive therapy in fracture healing may therefore be a possible strategy for use to enhance well vascularized fractures or delayed unions, but may not be effective in the acute open fracture setting. Without an endogenous drive towards non-specic bone anabolism (mesenchymal chemotaxis, and vascularisation), the bone-specic effects of PTH(134) treatment on committed cartilage and bone cells may fail to deliver sufcient stimuli to achieve union. Alternative agents, such as local BMP administration with its potential to initiate non-specic anabolism, complimentary anabolic strategies that enhance non-specic anabolism or novel pharmaceutical strategies may prove to be a better option when such an anabolic deciency is encountered. Acknowledgments This study was funded in part by NHMRC grant 352391 (animal work) and the University of Sydney travelling fellowship fund (stipend for MT). References

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17] [18] [19]

[20] [21]

[22] [23]

[24]

[25]

[26]

[27]

[28] [1] Gaebler C, Berger U, Schandelmaier P, Geitbauer M, Schaumecker HH, Appelgate B, et al. Rate and odds ratios for complications in closed and open tibial fractures treated with unreamed, small diameter tibial nails: a multi-center analysis of 467 cases. J Orthop Trauma 2001;15:243415. [2] Karladani AH, Granhed H, Karrholm J, Styf J. The inuence of fracture etiology and

[29]

M. Tgil et al. / Bone 46 (2010) 852859 [30] Kaback LA, Soung DY, Naik A, Geneau G, Schwarz EM, Rosier RN, et al. Teriparatide (134 human PTH) regulation of osterix during fracture repair. J Cell Biochem 2008;105:21926. [31] Ai-Aql ZS, Alagl AS, Graves DT, Gerstenfeld LC, Einhorn TA. Molecular mechanisms controlling bone formation during fracture healing and distraction osteogenesis. J Dent Res 2008;2:10718. [32] Deckers MM, van Bezooijen RL, van der Horst G, Hoogendam J, van Der Bent C, Papapoulos SE, et al. Bone morphogenetic proteins stimulate angiogenesis through osteoblast-derived vascular endothelial growth factor A. Endocrinology 2002;143:154553. [33] Little DG, McDonald M, Bransford R, Godfrey CB, Amanat N. Manipulation of the anabolic and catabolic responses with OP-1 and zoledronic acid in a rat critical defect model. J Bone Miner Res 2005;20:20445.

859

[34] Morgan EF, Mason ZD, Bishop G, Davis AD, Wigner NA, et al. Combined effects of recombinant human BMP-7 (rhBMP-7) and parathyroid hormone(1 34) in metaphyseal bone healing. Bone 2008;43:10318. [35] Li M, Ke HZ, Qi H, Healy DR, Li Y, Crawford DT, Paralkar VM, et al. A novel, nonprostanoid EP2 receptor-selective prostaglandin E2 agonist stimulates local bone formation and enhances fracture healing. J Bone Miner Res 2003;18: 203342. [36] Shen X, Wan C, Ramaswamy G, Mavalli M, Wang Y, Duvall CL, et al. Prolyl hydroxylase inhibitors increase neoangiogenesis and callus formation following femur fracture in mice. J Orthop Res 2009;27:1298305. [37] Garrett IR, Gutierrez GE, Rossini G, Nyman J, McCluskey B, Flores A, et al. Locally delivered lovastatin nanoparticles enhance fracture healing in rats. J Orthop Res 2007;25:13517.