Interlocking nail treatment of diaphyseal long-bone fractures in dogs

R. Tass Dueland, DVM , MS ; Kenneth A Johnson, MVSc, PhD; Simon C. Roe, BVSc, PhD; Mark H. Engen, DVM; Arnold S. Lesser, VMD

Objective - To determine results of using interlocking nails (IN) for fixation of diaphyseal bone fractures in dogs.

Design - Multi-center prospective clinical trial.

Animals - 134 dogs with diaphyseal fractures of the femur (n = 92), tibia (23), or humerus (19); 11 had previous unsuccessful treatments, and 103 had comminuted fractures of which 70 were classified as unstable.

Procedure -- All fractures were stabilized with 6- or 8-mm-diameter IN with 3.5- or 4.5-mm screws, respectively. Cerclage wires and an autogenous bone graft were used at the surgeon's discretion. Participating surgeons provided information on age, sex, weight, and breed of the dog, details of the surgery, details of any intra- or postoperative complications, fracture healing time, and limb function.

Results - Eight dogs were lost to follow-up evaluation. In 105 of the remaining 126 dogs (83%), fractures healed without complications. For these 105 dogs, limb function was excellent (n = 90), good (12), fair (2), and poor (1). Complications developed for 21 dogs (17%); limb fun ction after additional treatment was excellent (n = 10), good (2), fair (5), poor (1), or unreported (3). Interlocking nails broke in 9 dogs; breakeage was attributed to fatigue failure because of use of too small an IN or because of insertion of the IN so that a screw hole was positioned at the fracture site.

Clinical Implications - The high success rate and low complication rate suggest that IN can be used to stabilize diaphyseal fractures in dogs. Good technique is necessary for optimal results. (J Am Vet Med Assoc 1999;214:59-66).

Intramedullary fixation for internal fixation of fractures has been known since the 16 th century, and interlocking nails (IN) were first developed during the early 1950s. Since then, various improved IN have been developed and used, and use of IN for treatment of long-bone fractures in dogs has been studied experimentally, theoretically, and clinically.

Interlocking nails are inserted in the medullary canal and locked in place by screws or bolts inserted in the proximal and distal fracture fragments (static mode) or in only the proximal or distal fracture fragment (dynamic mode). Fixation of diaphyseal fractures with IN provides stability against axial, being, and torsional loads, and use of IN overcomes many of the limitations associated with use of intramedullary pins for fracture fixation, including collapse of comminuted fractures during weight-bearing, rotational instability, and pin migration.

The purposed of the study reported here were to determine healing of diaphyseal fractures in a large number of dogs in which IN fixation was used, critically evaluate and appropriately modify the design of the IN system, report and analyze complications, and evaluate limb function after fracture healing.

Material and Methods

Dogs - To ensure that a sufficient number of dogs would be included in the study, it was designed as a multi-center trial. Between 1991 and 1995, participating veterinarians at 7 university and 12 private veterinary practices provided information on dogs in which an IN was used for fixation of a fracture of the diaphysis of the femur, humerus, or tibia. Special training or expertise was not required of participating veterinarians; however, most were board-certified surgeons, surgical residens, or individuals who had received training in use of IN through continuing education courses.

Dogs were eligible for inclusion in the study if they had a fracture of the diaphysis of the humerus, femur, or tibia; if it appeared on preoperative radiographs that the proximal and distal fracture fragments were long enough to allow insertion of an IN and placement of at least 1, and preferably 2, locking screws in each fragment; and if bone quality appeared to be sufficient to allow the screws to hold firmly. Dogs with open fractures were included if proper debridement was performed. Because only 6- and 8-mm-diameter IN were available, dogs in which the medullary canal was too small to allow placement of a 6-mm-diameter IN were not included. Although for most dogs in the study fractures had not been treated previously, dogs with a nonunion or malunion and dogs in which previous fracture fixation had failed because of breakage of the implant or development of osteomyelitis were eligible for inclusion if they met the other criteria. Information for 4 of these dogs has been reported.

Participants in the study co mpleted a standard data collection form for each dog included in the study. Information provided included age, sex, weight, breed of the dog, details of the surgery (ie, size of IN used, whether bone graft was placed, whether cerclage wires were used), details of any intra- or postoperative complications, and follow-up data.

Interlocking nails - During the study, 3 generations of In and instrumentation, all manufactured to our design by a single company, were used. The second- and third-gerneration IN encompassed improvements suggested by the primary and collaborating investigators on the basis of experience with the first-generation IN.

First-generation IN were 6 or 8 mm in diameter and 300 mm in length. They had a trocar point at each end and transverse screw holes at 22-mm intervals throughout their length. Nails could be inserted in a normograde or retrograde direction, using a Jacobs chuck. Once an IN had been inserted, a right-angle jig was attached to the proximal end of the IN, which protruded from the skin. The jig had a locking pin that engaged 1 of the open holes in the end of the IN, so that the jig was positioned parallel to the 60 mm away from the IN. the jig had 14-mm-diameter holes through which trocars, guide sleeves, drill bits, and taps could be placed directly in alignment with the holes in the IN. Locking screws (3.5-mm cortical screws for the 6-mm IN and 4.5-mm cortical screws for the 8-mm IN) were inserted in the proximal and distal fragments. Position and number of screws were determined by length of the bone and fracture location and configuration. After all screws were inserted, the jig was removed, and the protruding portion of the IN was cut as short as possible with bolt cutters.

Second-generation IN were also 6 or 8 mm in diameter, but various lengths (140 to 230 mm) with 5 to 7 screw holes were available. These IN had a trocar point on only one end and a machined groove on the other to facilitate locking of a special chuck that attached to the end of the IN. The drill-guide jig was then attached to the chuck. Otherwise, second-generation IN were similar to first-generation ones.

Third-generation IN were also 6 or 8 in diameter; 5 lengths (140, 160, 185, 205, and 230 mm) were available (Fig. 1). All third-generation IN had only 2 holes at each end for locking screws. The proximal end of the IN had an internally threaded hole and 2 alignment tabs to which an extension piece 80 or 120 mm in length could be attached (Fig. 2). The drill-guide jig was then attached to the extension piece (Fig. 3). The longer extension piece was used when inserting IN into the tibia, and prevented the drill-guide jig from interfering with the femoral condyles or patella. The shorter extension piece was used when inserting IN into the femur or humerus. An inserter-extractor handle was attached to the extension piece, and the IN was inserted in a normograde direction. The end of each extension piece was marked with 5 marker rings at 2-mm intervals, which allowed the surgeon to determine how deeply the IN had been inserted.

Surgical technique - IN most instance, open fracture reduction was performed through a standard, but limited, surgical approach. The medullary canal was initially opened by inserting a series of Steinmann pins of increasing diameter in succession. In some instances, a 7- or 9-mm medullary hand reamer (for 6- and 8-mm IN, respectively) was als used. Reaming removed mainly medullary and metaphyseal cancellous bone; substantial reamin of endosteal bone, as is done in people, was not performed. AN IN of the proper length was chosen by taking measurements from radiographs. Whenever possible, an 8-mm-diameter IN was used in preference to a 6-mm-diameter IN. An attempt was made to seat the IN in metaphyseal cancellous bone, but precaustions were taken to prevent penetrating the articular surface. Because IN were locked in place by the screws, it was not essential to seta the IN completely into the distal metaphyseal bone for stabilization of very proximal fractures. After the IN was adequately seated, the drill-guide jig was attached, and the locking screws were inserted. In most instances, screws were first inswerted in the distal fracture fragment. Rotational alignment and axial length were checked, and screws were inserted in the proximal fracture fragment. IN instances when fracture reduction alone was sufficient to guarantee axial length and rotational alignment, screws were inserted in the proximal fracture fragment first. An autogenous cancellous bone graft was placed at the fracture site at the discretion of the attending surgeon. Radiographs were obtained immediately after surgery to assess adequacy of fracture reduction and positioning of the IN and screws.

When placing IN in the femur, the caudal curvature of the distal portion of the bone had to be taken into consideration. Radiographs of the contralateral bone were used to determine the degree of curvature and to select the length of IN to be used. Generally, the IN was placed so that its distal tip was not distal to the proximal pole of the patella. If necessary, the fracture was slightly over-reduced in a cranial direction so that the IN pin could be inserted distally enough to allow placement of the locking screws. Proximally, the end of the IN did not protrude dorsal to the greater trochanter.

Interlocking nails were placed in the tibia in a normograde direction to prevent impingement on structures in the stifle joint. A medial parapatellar approach was used, the fat pad was reflected, the stifle joint was placed in 90 degree of flexion, and the IN was inserted via a slight depression in the craniomedial aspect of the tibial plateau, just cranial to the tibial insertion of the cranial cruciate ligament.

Because most fractures of the humerus involved the distal third of the bone, there often was room for placement of only 1 locking screw in the distral fracture fragment. When using third-generation IN with 2 screw holes in each end, the distal end of the IN was sawed off so that a screw hole would not be located at the fracture site. The IN was directed so that it engaged as much of the medial aspect of the distal portion of the humerus as possible but did not penetrate the olecranon fossa. The location of the radial nerve was considered during fracture manipulation and IN and screw placement. If possible, a short enough IN was used so that he proximal end was within the cortex to prevent tissue irritation associated with protrusion of the IN.

Outcome assessment - Fractures were classified as closed or open, comminuted or noncomminuted, and fresh or preexisting (ie, malunion or nonunion). Severity of comminution was graded on the bsis of modified Winquist-Hansen criteria.

Information on fracture healing and limb function was obtained from participating surgeons or referring veterinarians or through telephone conversations with owners. Limb function was categorized as poor (non-weight-bearing or toe-touching only), fair (persistent marked limp or persistent slight limp), good (slight limp only after exercise), or excellent (no limp after exercise).

Radiographs were generally obtained every 4 to 8 weeks after surgery until radiographic signs of fracture union were evident (ie, evidence of bridging callus on at least 3 sides of the fracture) and signs of pain were not evident during palpation of the fracture site. After the fracture was healed, the IN and screws were removed at the discretion of the attending veterinarian.

Complications associated with the use of IN, including difficulties encountered during placement, problems with fracture healing or maintenance of stability, and abnormalities in limb function, were recorded. If possible, complications were classified as resulting from a design deficiency, technical errors, or errors in decision-making during surgery. A diagnosis of nonunion was made if the fracture was not healed by 7 months after the surgery. Failure of IN fixation was defined as breakage of the IN prior to fracture union or removal of the IN with replacement by another type of fracture fixation.

Results

Dogs-One hundred thirty-four dogs weighing between 11.4 and 91 kg (25 and 200 lb; mean, 32.5 kg {72 lb}) were included in the study. A 6-mm-diameter IN was used in 56 dogs (mean weight, 27.7 kg {61 lb}; range, 11 to 50 kg [24 to 110 lb]); an 8-mm-diameter IN was used in the remaining 78 dogs (mean weight, 35.5 kg {78 lb]; range 21 to 91 kg [46 to 200 lb]).

FOR MORE INFORMATION AND THE REMAINDER OF THE ARTICLE AND PICTURES, PLEASE CLICK HERE