1. Title of abstract

Treatment of an infected atrophic non-union and a non-union with implant failure by means of the distraction/compression principle according to Ilizarov

A. Short Abstract

The purpose of this study was to evaluate the feasibility of the dynamic distraction-compression technique without bone grafting for the treatment of non-unions. Implants (bent 3.5mm reconstruction plate in case one and broken 4.5mm plate in case two) were removed and a pre-assembled hinged circular external fixator frame (Small Bone Fixator - Hofmann company) was mounted closed. The defect was curetted only. No bone graft was needed. In case 1 (one year old female Greyhound, 16kg body weight) the infected radius and ulna non-union with a deformity in two planes was dynamically corrected. In case 2 (five year old female St. Bernhard, 42kg) the tibio-tarsal non-union after a tarsal panarthrodesis was dynamically compressed from day three to five at a rate of one millimeter per day. In both cases a solid bony union was achieved and the circular fixators were removed in 48 (case 1) respectively 47 days (case 2). In case one the deformity was corrected and the infection cleared.

B. Long Abstract

Introduction
The standard for the treatment of non-unions with implant failure in small animal surgery has been the use of autogenous cancellous bone graft in conjunction with plates, linear external fixaters or intramedullary implants such as interlocking nails. The aim of this study was to evaluate the feasibility of the dynamic distraction-compression technique according to Ilizarov without bone grafting for the treatment of an infected atrophic non-union with deviation in two planes (case 1) and a non-union (case 2).

Materials and methods

Case history 1
A one year old female Greyhound, 16kg body weight was referred with the diagnosis of a malalignment, swelling and fistulous tract of the left distal forearm. The dog had been operated some months ago in Spain. The clinical examination revealed a third degree lameness, swelling, higher local temperature and a purulent fistulous tract in the distal and dorsal radial area. The left carpal joint had a reduced range of motion of 50¡ in flexion compared to 160¡ on the right side. On radiographs a bent 5-hole 3.5mm reconstruction plate, bony atrophy at the former fracture site, but some scattered bone densities at the caudal ulna aspect and a varus and cranio-caudal deviation were evident.
Pasteurella multocida was cultured from the drainage tract, sensitive to imipenem and enrofloxacin only, but resistant to 29 tested antimicrobial drugs. Based on this result the dog received enrofloxacin at a dosage of 5mg/kg body weight BID for 6 weeks.
In preoperative planning schematic drawings of the dorso-palmar and medio-lateral view radiographs were made. A graphical method was used to determine the plane of deformity and apex of deformity from lateral and dorso-palmar radiographs.
The apex of the deformity was defined and the varus/valgus and craniocaudal components of the deformity were measured from their respective orthogonal view radiograph. The resulting two vector components were plotted on an x (varus/valgus) and y (cranio-caudal) grid. The resultant vector defined the plane of deformity.

Hinges were placed at the apex of the deformity. A line drawn between hinges was perpendicular to the plane of deformity. The angular motor was placed in parallel to this plane. In surgery the plate was exposed and, after closed montage of the pre-assembled Ilizarov frame, removed. The defect was curetted and osteostixis performed. The 1.2mm wires of the _, 1, 1 60mm rings were tensioned with 60kg according to the recommendations of Ferretti. The post was tensioned with 30kg.

The dog had been weight bearing from the first postoperative day. Exercise was limited to leash walk. After a latency period of two days the distraction was started at a rate of 1mm and a twice daily rhythm (0.5mm BID). Physical therapy consisting of passive flexion and extension of the carpal joint for 5 to 10 minutes twice daily was started by the owner. Due to an excellent callus response the rate was increased on day 13 to 1.2mm. On day 35 the deformity was corrected and distraction was ceased. The frame was axially dynamised by releasing the tension of the wires. On day 48 the frame was removed. Range of motion of the carpal joint of 50¡ equalled the preoperative situation.

Case history 2
A 5 year old female St. Bernhard, 42kg body weight was admitted with the history of a non weight bearing lameness. A left tarsal panarthrodesis has been performed at another place after the dog was hit by car. Five weeks ago pins which migrated had been removed. On clinical examination the dog showed a non weight bearing lameness and swelling of the left tarsal joint. On radiographs breakage of an 8-hole 4.5mm plate, implant migration of a K-wire, a remnant of a broken K-wire, non-union of the talocrural joint but fusion of the intertarsal- and tarso-metatarsal joints were detected.

As in case one the frame was pre-assembled. In the surgery all implants were removed and the pre-assembled frame was applied. The hinged frame consisted of _, 1, 1 80mm rings and a distal post. The non-union was curetted from a medial limited open approach. The 1.5mm wires were tensioned with 60kg and the post with 30kg.

As in case 1 the dog had been weight bearing from day one.
After a latency of two days the non-union was dynamically compressed at a rate of 1mm per day divided BID for three days. The follow up radiograph on day 30 showed bony union of the defect. The frame was removed on day 47.

Results

In both cases a solid bony union was achieved and the circular fixators were removed in 48 (case 1) respectively 47 days (case 2). In case one the deformity was corrected and the infection cleared.

Discussion / Conclusion

The method has several advantages : Closed montage with no interference with local blood supply and no implants at the site of non-union using the natural healing potential to its maximum.
Possibility to dynamically correct deformities (case 1) or to dynamically compress non-unions (case 2) at a rate and for the duration chosen by the surgeon.
No need for an autogenous bone graft, resulting in less morbidity for the patient. Possibility of axial dynamisation by reducing tension on the wires. It has been shown that controlled axial micromotion induces stress generated potentials, which stimulate callus formation.
Relatively short healing times compared to other techniques.
Easy implant removal at the end of treatment, leaving no implants in the patient.