Session III - Polytrauma Fracture Healing

Is Human Fracture Hemotoma Inherently Angiogenic?

John T. Street, FRCSI; Jaing H. Wang, PhD; Abel Wakai, FRCSI; Anthony McGuinness, FRSCE; H. Paul Redmond, FRCSI, Cork University Hospital, Cork, Ireland

Introduction: Angiogenesis is an essential component of normal tissue repair and precedes osteogenesis during bone formation¹. Vascular endothelial growth factor (VEGF) is a potent mediator of hypoxia induced angiogenesis and may be released by cultured osteoblast-like cell lines in response to insulin-like growth factor 1 and prostaglandins². To date VEGF has not been identified in fracture healing. Fracture hematoma (FH) induces endochondral ossification in non-skeletal sites³, and its early removal significantly attenuates fracture repair⁴. While it has been demonstrated as a reservoir for interleukins⁵, its role during the angiogenic phase of fracture healing is unknown. This study tested the hypothesis that human fracture hematoma would contain VEGF and thus promote angiogenesis during bone repair.

Materials and Methods: Fracture hematoma and serum samples (hereafter referred to as subject serum) were obtained from 32 patients undergoing open reduction and internal fixation of isolated closed long bone fractures. Fully informed consent was obtained from all subjects. Duplicate control serum samples were obtained from age- and sex-matched volunteers. All specimens were centrifuged, microfiltered, aliquotted under sterile conditions and stored at - 85^oC for assay.

Experiment 1: Concentrations of VEGF in control serum (CS), subject serum (SS) and fracture hematoma supernatant (FHS) were measured in pg /ml using a solid phase enzyme linked immunosorbant assay (R&D Systems, MN).

Experiment 2: One of the control serum samples was allowed to hemolyze at body temperature (37^oC) to mimic FH and comparative biochemical analysis was performed.

Experiment 3: Primary isolates of human umbilical vein endothelial cells (HUVEC) were cultured to adherence under standard conditions. These were then incubated for 24 hours in specimens of CS, FHS and HCS. Resultant angiogenesis was determined by assessing HUVEC proliferation and tubule formation.

HUVEC proliferation was measured by a 5-bromo-2 deoxyuridine labeling colorimetric assay (Boehringer Mannheim, Mannheim, Germany) and tubule formation was determined on Matrigel with Rapi-Diff II staining (Diachem, Lancashire, UK).

Data is expressed as mean +/- SEM and is representative of 32 subjects. Statistical analysis was performed using analysis of variance (ANOVA) and a post hoc Scheffe F-test with a P value < 0.01 * was considered significant.

Results: All patients with isolated long bone fractures had significantly raised circulating VEGF levels compared to controls (310.4 +/- 45.1* versus 68.4 +/- 2.68 pg /ml) while FHS contained markedly higher concentrations of VEGF (5335.4 +/- 1020.6 pg./ml*) than subject serum. While serum >from injured patients induced angiogenesis in vitro, FHS significantly attenuated both endothelial cell proliferation and tubule formation compared to complete medium M199.

This inhibition was mirrored by that of hemolyzed control serum (Figure 1)

of almost identical biochemical profile (Table 1).

Table 1.

	Potassium (mmol /l)	LDH (mmol /l)	pH (index)	Hemolysis (index)
Fracture Hematoma	9.37+/-0.03	24.9+/-0.2 7.	38+/-0.01	22+/-2
Hemolyzed Control	9.26+/-0.06	22.5+/-0.7	7.39+/-0.02	21+/-2.5

Conclusions: This study quantifies the potent angiogenic cytokine vascular endothelial growth factor in human fracture hematoma in concentrations known to induce endothelial cell proliferation and tubule formation. Paradoxically, however, fracture hematoma significantly inhibits angiogenesis in vitro. This is due in part at least to non-specific biochemical toxicity and possibly in part to a local accumulation of anti-angiogenic cytokines as yet undetermined. It is intriguing indeed that while trauma induces a systemic state of pro-angiogenesis, local factors may mitigate against revascularization at the site of bony injury.

Acknowledgements: The technical assistance of Qiong Di Wu BSc is greatly appreciated. This work is supported by the Health Research Board Ireland.

References: (1) Am J Orthod Dentofac Orthop 111(4):382-390,1997; (2) Curr Top Microbiol Immunol 237:1-30,1999;(3) J Bone Joint Surg [Br] 72(5):822-9,1990; (4) Acta Orthop Scand 64(3):340-2,1993; (5) J Trauma 42(5):895-904, 1997.