Intermittent Parathyroid Hormone (1–34) Treatment Increases Callus Formation and Mechanical Strength of Healing Rat Fractures

Authors

  • Troels T. Andreassen,

    Corresponding author
    1. Department of Connective Tissue Biology, Institute of Anatomy, University of Aarhus, Aarhus, Denmark
    • Department of Connective Tissue Biology, Institute of Anatomy, University of Aarhus, DK-8000 Aarhus C, Denmark
    Search for more papers by this author
  • Charlotte Ejersted,

    1. Department of Connective Tissue Biology, Institute of Anatomy, University of Aarhus, Aarhus, Denmark
    Search for more papers by this author
  • Hans Oxlund

    1. Department of Connective Tissue Biology, Institute of Anatomy, University of Aarhus, Aarhus, Denmark
    Search for more papers by this author

Abstract

The influence of intermittent parathyroid hormone (PTH(1–34)) administration on callus formation and mechanical strength of tibial fractures in rats was investigated after 20 and 40 days of healing. A dose of 60 μg of PTH(1–34)/kg/day and 200 μg of PTH(1–34)/kg/day, respectively, was administered during the entire periods of healing, and control animals with fractures were given vehicle. The dose of 200 μg of PTH(1–34)/kg/day increased the ultimate load and the external callus volume of the fractures by 75% and 99%, respectively, after 20 days of healing and by 175% and 72%, respectively, after 40 days of healing. The dose of 60 μg of PTH(1–34)/kg/day did not influence either ultimate load or external callus volume of the fractures after 20 days of healing, but the ultimate load was increased by 132% and the external callus volume was increased by 42% after 40 days of healing. During the healing period, the callus bone mineral content (BMC) increased in all groups. After 40 days of healing, the callus BMC was increased by 108% in the 200 μg of PTH(1–34)/kg/day group and by 76% in the 60 μg of PTH(1–34)/kg/day group. Both doses of PTH(1–34) steadily augmented the contralateral intact tibia BMC (20 days and 40 days: 60 μg of PTH (1–34)/kg/day 9% and 19%, respectively; 200 μg of PTH (1–34)/kg/day 12% and 27%, respectively) and bone mineral density (20 days and 40 days: 60 μg of PTH(1–34)/kg/day 11% and 12%, respectively; 200 μg of PTH(1–34)/kg/day 11% and 15%, respectively).

INTRODUCTION

During recent years, a number of animal studies have demonstrated that intermittent parathyroid hormone (PTH) administration induces anabolic effects on both cancellous and cortical bone.1-14 In parallel with the enhanced bone mass, increased mechanical strength of the bones has been found.15-18

However, little information is available about the effects of PTH treatment on fracture healing.19, 20 We have therefore studied the influence of PTH(1–34) administration on callus formation and mechanical strength development in tibial fractures in rats after 20 and 40 days of healing. PTH(1–34) was administered in doses of 60 and 200 μg/kg/day, respectively. The former dose is within the dose-range normally used in rat experiments, and the latter dose is known to be the highest dose inducing new bone with normal morphology on surfaces of intact bone.10, 21 Higher doses of PTH also result in formation of woven bone on the surfaces.10, 21

MATERIALS AND METHODS

Animals and hormone administration

Three-month-old female Wistar rats (Møllegaard, Lll., Skensved, Denmark) were randomly divided into six groups. The fractures were tested after 20 and 40 days of healing, and at each healing period three groups were tested: rats injected daily with vehicle, rats injected daily with 60 μg of PTH (1–34)/kg, and rats injected daily with 200 μg of PTH(1–34)/kg. Human parathyroid hormone (1–34) (hPTH(1–34); Bachem, Bubendorf, Switzerland) was dissolved in a vehicle consisting of 0.15 M saline with 2% heat-inactivated rat serum. The injections were given subcutaneously from the day of fracturing, and each weekthe rats were weighed and the dose adjusted to the actual body weight. The rats were housed with a cycle of 12 h of light and 12 h of darkness and had free access to tap water and pellet food (Altromin diet 1324; Chr. Petersen, Ltd., Ringsted, Denmark; 0.9% calcium, 0.7% phosphorus).

Fracture technique

The animals were anesthetized with halothane (Halocarbon Laboratories, River Edge, NJ, U.S.A.), and unilateral, standardized closed fractures were produced above the tibiofibular junction in the right tibia by three-point bending. Closed medullary nailing was performed using a 0.8 mm Kirschner wire (Extron, Copenhagen, Denmark), and the skin was closed with monofilament sutures.22, 23 The operations were performed under sterile conditions. Postoperatively, contact X-rays were taken to secure correct localization of the fracture and fixation with the wire. Unprotected weight-bearing was allowed, and the animals resumed normal activity after recovery from the anesthesia. Out of 120 animals operated on, 107 were included in the subsequent experiment (5 animals died as a result of anesthesia and 8 animals were killed as the fractured tibia shattered during fixation with the Kirschner wire).

The rats were killed with pentobarbital (150 mg/kg intraperitoneally; Mebumal, SA, Copenhagen, Denmark). In the 40-day groups, blood was collected from the vena cava inferior at the time of killing. Serum calcium was measured using a Cobas Integra 700 autoanalyzer with Cobas Integra calcium kit (Roche Diagnostics, Basel, Switzerland). The tibial bones were dissected free, and contact X-rays of the fractured and opposite intact bone were taken. Then the bones were stored in buffered Ringer's solution (4°C, pH 7.4). The experiment was approved by the Danish Animal Experiment Inspectorate.

Callus dimensions

In the fractured tibia, the external medial–lateral and anterior–posterior diameters were measured at the fracture line by a sliding caliper. In the opposite intact tibia, the corresponding diameters were measured. Total volumes of both fractured and intact tibiae were gauged using Archimedes' principle,24 and external callus volume was subsequently calculated as the volume of the fractured tibia minus the volume of the intact tibia.

Mechanical testing

The mechanical strength of the healing fractures was measured by a destructive three-point bending procedure using a materials testing machine (Alwetron 250; Lorentzen and Wettre, Stockholm, Sweden). The intramedullary nail was removed, and the fractured bone was placed on two rounded bars (spaced 15 mm apart) with the fracture line between the bars. Deflection was performed by lowering another bar onto the fracture line, using a constant speed of 2 mm/minute. The intact left tibia was tested at a point corresponding to the fracture line of the right tibia, using exactly the same procedure. All bones were placed in the same position with the concave facet of the lateral tibial condyle resting on the supporting bar and the load was therefore applied from the medial side. Load and deflection were recorded continuously by transducers coupled to measuring bridges. The signal was fed to an xy recorder, and the load-deflection curves obtained were read by a graphic tablet into a computer (HP 9178 A and HP 9816S; Hewlett-Packard, Fort Collins, CO, U.S.A.). The following parameters were calculated: ultimate load, ultimate stiffness, and deflection at ultimate load.22

Dual-energy X-ray absorptiometry

All fractures broke at the fracture line and the intact bones broke at the loading point, showing fractured surfaces without loose fragments. The two ends of bones tested were realigned at their original position and fixed with surgical tape (Tegaderm; 3M Health Care, St. Paul, MN, U.S.A.), then frozen in Ringer's solution for later measurement of the bone mineral content (BMC) by dual-energy X-ray absorptiometry (DXA). The tibiae were placed in Ringer's solution with the anteromedial surface downward and scanned by DXA using a line spacing of 0.508 mm, a point resolution of 0.127 mm, and the regional high resolution analysis program for small animals (QDR-2000; Hologic, Inc., Waltham, MA, U.S.A.). BMC of the whole bone was measured, both in the fractured and the intact tibiae, and bone mineral density (BMD) of the intact tibia was calculated as BMC/tibia volume. The amount of bone around the fracture line was measured using regional DXA. BMC was measured in a 7.6-mm-high diaphysial segment (3.8 mm proximal and 3.8 mm distal to the fracture line). In the opposite intact tibia, BMC of the corresponding diaphysial segment was measured. BMC of the callus was calculated as BMC of the fractured segment minus BMC of the intact segment.

Histologic appearance of intact mid-diaphyseal bone

To evaluate whether the PTH(1–34) treatments induced woven bone formation on intact periosteal and endocortical surfaces, a 200-μm transverse section from the left mid-diaphysial femur was cut out and analyzed as previously described.24 The morphology of the bone sections was examined using light microscopy and polarization microscopy with a lambda filter, which identified the direction of the collagen fibers in the bone. Bone sections from all the animals in the experiment were investigated.

Right and left intact tibiae in rats

Because callus volume and callus DXA-BMC were calculated by comparison of the right fractured tibia with the left intact tibia, we compared right and left tibial volume, total DXA-BMC, and DXA-BMC in a 7.6-mm-high diaphysial segment in seven 3-month-old female rats with intact tibiae. No differences were found between right and left intact tibiae (mean ± SEM): volume (right 304 ± 8, left 305 ± 8 [mm3]; p = 0.99); total DXA-BMC (right 242 ± 6, left 242 ± 6 [mg]; p = 0.87); total BMD (right 0.80 ± 1, left 0.80 ± 1 [10–2 mg/mm3]; p = 0.87); diaphysial segment DXA-BMC (right 33.4 ± 0.4, left 33.8 ± 0.4 [mg]; p = 0.18); length (right 36.2 ± 0.2, left 36.2 ± 0.02 [mm]; p = 0.26).

Statistical analyses

For analysis of differences between the groups at each healing period, the Kruskal–Wallis test was used, and in cases where differences occurred, the Mann–Whitney U-test was applied. Differences between intact right and intact left tibiae were analyzed by Wilcoxon's test for matched pairs. p < 0.05 (two-tailed) was considered statistically significant.

RESULTS

The results of the mechanical testing are given in Table 1. After 20 days of healing, the ultimate load of the fractures was increased in the group given 200 μg of PTH(1–34)/kg/day, both in comparison with the 60 μg of PTH(1–34)/kg/day group and the vehicle group (67% and 75%, respectively), whereas no differences were found between the 60 μg of PTH (1–34)/kg/day group and the vehicle-injected group. After 40 days of healing, however, ultimate load and ultimate stiffness of the fractures were enhanced in both PTH(1–34) groups compared with the vehicle-injected group. The ultimate load was increased by 132% (60 μg of PTH(1–34)/kg/day) and 175% (200 μg of PTH (1–34)/kg/day), and ultimate stiffness by 208% (60 μg of PTH(1–34)/kg/day) and 253% (200 μg of PTH(1–34)/kg/day). No differences were found between the 60 μg of PTH(1–34)/kg/day group and the 200 μg of PTH(1–34)/kg/day group.

Table Table 1. Mechanical Properties of Fractured and Contralateral Intact Tibiae After 20 and 40 Days of Healing
original image

The mechanical strength of the contralateral intact tibia was not influenced by PTH(1–34) administration after the first 20 days of treatment. However, after 40 days, the ultimate load was 16% greater in the rats given 200 μg of PTH(1–34)/kg/day than in the vehicle-injected animals. The rats given 60 μg of PTH(1–34)/kg/day showed an ultimate load which was found to be between those of the other two groups, but not significantly different from either.

Dimensions, volume, BMC, and BMD of fractured and contralateral intact tibiae after 20 days of healing are given in Table 2. Both anterior-posterior and medial-lateral external callus dimensions were increased in the group given 200 μg of PTH(1–34)/kg/day compared with both the 60 μg of PTH(1–34)/kg/day group (21% and 31%, respectively) and the vehicle group (30% and 35%, respectively). External callus volume of the 200 μg of PTH(1–34)/kg/day group was also enhanced compared with both the 60 μg of PTH(1–34)/kg/day group and the vehicle group (77% and 99%, respectively). The DXA-BMC was augmented in both fractured and intact tibiae of the animals given 60 μg of PTH(1–34)/kg/day and 200 μg of PTH(1–34)/kg/day compared with the vehicle group. The callus DXA-BMC was enhanced by 62% in the 200 μg of PTH(1–34)/kg/day group compared with the vehicle group. The rats given 60 μg of PTH(1–34)/kg/day showed callus DXA-BMC values between those of the other two groups, but not significantly different from either.

Table Table 2. Dimensions, Volume, BMC, and BMD of Fractured and Contralateral Intact Tibiae After 20 Days of Healing
original image

Dimensions, volume, BMC, and BMD of fractured and contralateral intact tibiae after 40 days of healing are given in Table 3. The anterior-posterior and medial-lateral external callus dimensions were increased in both groups treated with PTH(1–34) compared with the vehicle group (200 μg of PTH(1–34)/kg/day: 19% and 28%, respectively; 60 μg of PTH(1–34)/kg/day: 12% and 22%, respectively). External callus volume of the two PTH(1–34) groups was also enhanced compared with the vehicle group (200 μg of PTH(1–34)/kg/day: 72%; 60 μg of PTH(1–34)/kg/day: 42%). However, after the 40-day healing period no significant difference in external callus volume was found between the 200 μg of PTH(1–34)/kg/day and the 60 μg of PTH(1–34)/kg/day groups (21%, p = 0.07). The callus DXA-BMC was enhanced by 108% in the group given 200 μg of PTH(1–34)/kg/day and by 76% in the group given 60 μg of PTH(1–34)/kg/day compared with the vehicle group, but no significant difference in callus DXA-BMC was found between the 200 μg of PTH(1–34)/kg/day and the 60 μg of PTH(1–34)/kg/day groups (18%, p = 0.25).

Table Table 3. Dimensions, Volume, BMC, and BMD of Fractured and Contralateral Intact Tibiae After 40 Days of Healing
original image

Appearances of the fractures are depicted in Figs. 1 and 2. In all three groups, the X-ray pictures show distinct fracture lines after 20 days of healing, whereas the fracture lines are less perceptible after 40 days of healing. Correspondingly, in all three groups, the callus DXA-BMC increased substantially from day 20 to day 40 of healing (vehicle: 92%, p = 0.001; 60 μg of PTH(1–34)/kg/day: 159%, p = 0.000; 200 μg of PTH(1–34)/kg/day: 148%, p = 0.000).

Figure FIG. 1.

Twenty days of healing. X-ray and corresponding photograph of healing tibia fracture and contralateral intact tibia. (A) Vehicle; (B) 60 μg of PTH(1–34)/kg/day; (C) 200 μg of PTH(1–34)/kg/day. Dimensions are given by the bar (30 mm).

Figure FIG. 2.

Forty days of healing. X-ray and corresponding photograph of healing tibia fracture and contralateral intact tibia. (A) Vehicle; (B) 60 μg of PTH(1–34)/kg/day; (C) 200 μg of PTH(1–34)/kg/day. Dimensions are given by the bar (30 mm).

Body weight during the fracture healing periods are given in Table 4. No differences in body weights were found between the groups either at the day of operation or at the day of killing. However, when changes in weight of each animal from the day of operation to the day of killing were calculated, a modest but significant weight increase was found after 20 days of healing in both groups given PTH(1–34). Table 4 also depicts the concentration of serum calcium after 40 days of healing. No differences in serum calcium were found between the groups.

Table Table 4. Body Weight Changes During Fracture Healing Periods
original image

No woven bone formation was seen at the intact femoral mid-diaphysial surfaces either at the periosteal surface or at the endocortical surface in any of the rats treated with PTH(1–34) for 20 days or for 40 days.

Of the 107 animals included in the experiment, 6 were excluded at the time of killing, 5 because of infection (vehicle, 1; PTH(1–34), 4), and 1 because of a Kirschner wire displacement (PTH(1–34)). After 20 days of healing, the total number of animals examined in each group was: vehicle, 17; PTH(1–34) 60 μg/kg/day, 16; PTH(1–34) 200 μg/kg/day, 17; and after 40 days of healing: vehicle, 16; PTH(1–34) 60 μg/kg/day, 18; PTH(1–34) 200 μg/kg/day, 17.

DISCUSSION

The healing of fractures in intact rats after PTH administration has not been studied before. However, PTH treatment has been used in two studies on the healing of fractures in rats with pathologic bone metabolism and delayed fracture healing.25, 26 Kim et al.20 used ovariectomized rats and showed that the decreased mechanical strength seen in fractures after 28 days of healing could be partly prevented by giving the animals 175 μg of PTH(1–84)/kg/day. This dose corresponds with the 60 μg of PTH(1–34)/kg/day used in our experiment (18 nmol and 15 nmol, respectively), and equal molar of PTH(1–84) and PTH(1–34) induces equal anabolic bone effects when given to rats in the dose-range 1.1–30 nmol.7, 15 Fukuhara and Mizuno19 studied parathyroidectomized rats with and without PTH administration during the first 5 weeks of healing. Applying histologic techniques, they found that PTH administration enhanced both bone formation and resorption in the early stage of healing, whereas bone resorption was decreased later in the healing period. They used 6-week-old male rats and injected PTH(1–34) intramuscularly in a daily dose of 20 U and 100 U per rat, respectively, which corresponds to ∼10–12 and 40–45 μg/kg/day, given the weight of the rats being 150 g.27

In previous experiments using normal rats, we have found that the callus volume reaches its maximum after 20–30 days of healing, and thereafter gradually declines during the subsequent 130 days of healing (unpublished data), even though the mechanical strength of the fracture steadily augments during the entire healing period.28 The vehicle-injected animals in the present experiment follow the same pattern as the strength of the fractures increases by 150%, and the external callus volume declines by 30% from day 20 to day 40 of healing. Compared with the vehicle-injected groups, in the 200 μg of PTH(1–34)/kg/day groups fracture strength and external callus volume are enhanced substantially after both 20 and 40 days of healing. Both parameters, however, follow the same pattern as observed in the vehicle-injected animals, as strength increases by 300% and external callus volume declines by 30% from day 20 to day 40 of healing.

As pointed out by Turner and Burr, the span between the two supporting bars should be sufficiently long to minimize shear at the loading point when using three-point loading.29 The substantial increase in callus volume at 20 days of healing in the group treated with 200 μg of PTH(1–34)/kg/day would most likely have induced increased shear at the loading point. Since increased shear at the loading point leads to a lower ultimate strength,30 the ultimate load of the group treated with 200 μg of PTH(1–34)/kg/day might have been underestimated, even though the ultimate load is significantly increased in this group compared with the two other groups.

A dose of 60 μg of PTH(1–34)/kg/day does not influence external callus volume and mechanical strength of fractures after 20 days of healing compared with vehicle-injected animals. The anabolic effect of the PTH(1–34) treatment on the contralateral intact tibia is, however, measurable as an increase in DXA-BMC. From day 20 to day 40, the 60 μg of PTH(1–34)/kg/day animals increase fracture strength by 450%, whereas the callus volumes are found to be almost identical. A possible explanation of the unchanged callus volumes could be that the 60 μg of PTH(1–34)/kg/day treatment reduces the decline in callus volume normally seen in the healing of fractures in rats during this period. This corresponds with the reported decline in callus resorption seen after 5 weeks of healing in parathyroidectomized rats treated with ∼40–45 μg of PTH(1–34)/kg/day.19 Further static and dynamic histomorphometric investigations are needed to evaluate the mechanisms by which PTH treatment influences callus formation and resorption. Since PTH treatment is known to induce bone formation on intact surfaces, in vivo measurements of the collagen deposition in the healing bone callus might also be useful; a method for such measurements has previously been used when measuring collagen deposition in healing soft tissues.31, 32

The present experiment also shows that the anabolic bone effect of PTH(1–34) on the contralateral intact tibia continues during the healing period. DXA-BMC of the intact tibia is already enhanced after 20 days of healing in both PTH(1–34)-treated groups. The DXA-BMC is further increased in both groups after 40 days of healing compared with the values after 20 days (60 μg of PTH(1–34)/kg/day: 14%, p = 0.000; 200 μg of PTH(1–34)/kg/day: 19%, p = 0.000). During this period, DXA-BMC of the vehicle-treated animals does not increase significantly (5%, p = 0.105).

The well being of the rats did not seem to be negatively influenced by the PTH(1–34) treatment. On the contrary, the PTH(1–34)-treated rats had an increased weight gain during the first 20 days of healing compared with the vehicle-injected rats. After 40 days of healing, no differences in weight gain were found between the groups. No differences in serum calcium concentration were found between the groups after 40 days of healing. These results are in accordance with a previous dose–response experiment on rats in which neither body weight nor serum calcium concentration were influenced by PTH(1–34) treatments.15

The present results indicate that intermittent PTH treatment may positively influence the management of fracture healing. This could be of special interest in situations where the fracture healing is impaired. In very old rats, fracture healing proceeds slowly.23, 33 34 It might therefore be of relevance to consider the possibility of accelerating the healing process, especially because it is a well documented fact that intermittent PTH treatment also induces an anabolic effect on intact bones in old rats.14, 18 35 Experimental rat models for producing nonunion fractures have been developed,36, 37 but the influence of PTH treatment has yet to be investigated. A very recent abstract reports the influence of a PTH-related protein analog (RS-66271; Roche Bioscience, Palo Alto, CA, U.S.A.) administered to rabbits with impaired fracture healing caused by systemic corticosteroid treatment. The authors conclude that treatment with the PTH-related protein analog enhances radiographic healing parameters and mechanical strength of the investigated ulnar osteotomy.38

In conclusion, the present experiment shows that intermittent administration of a high dose of PTH(1–34) is able to enhance callus volume and the mechanical strength of fractures after both 20 and 40 days of healing. A lower PTH(1–34) dose, however, in the dose range normally used when investigating the anabolic effects of PTH on rats, does not influence healing of fractures after the first 20 days; but after 40 days of healing, this dose causes a substantial increase in callus volume and the mechanical strength of the fractures.

Acknowledgements

We are grateful to M. Fischer, J. Utoft and P.K. Nielsen for excellent technical assistance. This work has been supported by The Danish Health Research Council, grant numbers 9503023 and 9600822 (Aarhus University – Novo Nordisk Center for Research in Growth and Regeneration), Aarhus University Research Foundation and Novo Nordisk Foundation.

Ancillary