Clinical Application of Tobramycin-Impregnated Calcium Sulfate Beads in Six Dogs (2002–2004)

Introduction

Eradication of infectious agents causing osteomyelitis following joint replacement, open-fracture repair, elective orthopedic procedures, and allografting remains a challenge in small animal surgery. Often, removal of the implant and/or amputation is necessary to completely control the infection.^1–3 Complex orthopedic procedures such as limb sparing and tibial plateau leveling osteotomies (TPLOs) have led to higher infection rates.^4–6 Osteomyelitis has been reported as the most common complication following TPLO, occurring in 7.3% of cases.⁵ Traumatic injury to bone may create areas of bony destruction, bone loss, and/or significant soft-tissue damage—all of which predispose to infection.⁷ Additionally, repair of these injuries often requires the use of metallic implants, cements, and bone allografts, which can also contribute to the pathogenesis of osteomyelitis.^1,7

Difficulties in managing osteomyelitis stem from inability to achieve bactericidal levels of antibiotics at the infection site while providing adequate debridement and drainage. Local antimicrobial therapy to achieve higher tissue levels while avoiding systemic toxicity can be crucial in the elimination of orthopedic infections.^8–11 Antibiotic-impregnated polymethylmethacrylate (PMMA) beads have been used for this purpose. Disadvantages of using PMMA beads include the exothermic reaction during curing that can release toxic substances and damage tissues. Polymethylmethacrylate may also inhibit bone formation and serve as a surface for bacterial colonization. It is nonresorbable, thereby requiring removal at a second surgery.^9,12 These limitations have prompted the development of bioresorbable carriers.

Calcium sulfate is a bioresorbable material that has been used to treat bony defects since the late 1800s. Medical-grade pellets of calcium sulfate have consequently been modified to optimize their rate of resorption while maintaining their biocompatibility and osteoconductive properties. These pellets can be impregnated with antibiotics prior to implantation and be used to prevent or treat osteomyelitis.^{8–10,13–16} Although substantial literature supports the use of tobramycin-impregnated calcium sulfate beads in research, clinical use in animals has not been reported. The purpose of this case series is to document the clinical applications and outcomes of treatment with tobramycin-impregnated calcium sulfate beads in six dogs.

Materials and Methods

Dogs were identified for inclusion in this study if they were treated with tobramycin-impregnated calcium sulfate beads between 2002 and 2004 and if they had sufficient radiographic or verbal follow-up to determine outcome. The medical records of these dogs were evaluated to determine the indication for bead implantation, duration of clinical signs prior to bead implantation, clinical signs at the time of implantation, number of previous treatments, organisms cultured from the surgical sites, quantity of beads implanted, and concurrent treatments. Radiographs obtained prior to bead implantation were reviewed to confirm signs of osteomyelitis, including bone lysis, periosteal response, presence of a sequestrum, implant integrity, fracture stability, and soft-tissue swelling. The number of beads implanted in each case varied with the extent of disease and the amount of dead space.

Following implant revision, adequate lavage, and debridement, the beads were placed into bone defects and soft tissue surrounding the bone and implant [Figure 1]. Outcomes were measured by improvement in clinical signs such as eradication of draining tracts and improvement of lameness. Follow-up radiographs were evaluated for bony union, evidence of bead resorption, and resolution of signs of osteomyelitis.

Calcium sulfate beads were prepared according to the manufacturer’s recommendations for the fast cure Osteoset (Wright Medical Technology) kit, with the addition of 1.2 grams of tobramycin powder. The paste was spread into molds that were pressed together for 3 to 5 minutes during the curing process [Figure 2]. The beads are approved by the Food and Drug Administration for use as bone filler, but the addition of antibiotics remains an off-label use in the United States.¹⁷ Each kit provides 60 beads, each 7 mm in diameter, for a total cost (kit and antibiotics) of about $1300. The beads were packaged into sets of six beads, resterilized with ethylene oxide, aerated for 24 hours, and stored at room temperature.

Results

Tobramycin-impregnated calcium sulfate beads were used to treat osteomyelitis in five dogs and were implanted preemptively in one dog with an open fracture [see Table]. The five dogs in which beads were used to treat osteomyelitis included one dog with an infected total hip replacement (THR) and subsequent implant removal, two dogs with infected TPLOs, one dog with an infected fracture repair, and one dog with an infected ulnar ostectomy. The histories, signalments, diagnostic findings, treatments, and outcomes of the six dogs treated with tobramycin-impregnated calcium sulfate beads are summarized in the Table.

Case no. 1, a 2-year-old, castrated male Irish setter, had a humeral fracture repaired with a 4.5-mm, limited-contact dynamic compression (LC-DCP) plate and cerclage wires, which required revision 8 weeks after initial repair. Eleven months after the second surgery, the dog returned with a chronic draining tract originating from the distal aspect of the plate and a periosteal response noted along the humerus. All implants were removed, with the exception of a cerclage wire that was embedded in callus [Figure 3A]. Six tobramycin-impregnated calcium sulfate beads were placed around the bone at the level of the draining tract, and systemic antimicrobials were initiated (cephalexin^a 22 mg/kg per os [PO] q 8 hours for 2 weeks).

Case nos. 2 (1-year-old, spayed female rottweiler) and 3 (10-year-old, spayed female rottweiler) were presented for elective TPLOs. The first dog (case no. 2) returned with implant fracture and loosening and a periosteal response seen radiographically 56 days after surgery. The surgical site was débrided; the screws were tightened; the fractured screw was replaced; six tobramycin-impregnated calcium sulfate beads were implanted; and systemic antimicrobials (clavamox 13.25 mg/kg PO q 12 hours for 10 days) were initiated.

The second dog (case no. 3) returned with an avulsion fracture of the tibial tuberosity, which was repaired with a pin and tension band. Pseudomonas aeruginosa was cultured from an intraoperative sample, and clavamox (13.25 mg/kg PO q 12 hours) was initiated. Six days later, the fracture was determined to be unstable, as the Kirschner wires had migrated. A subsequent surgery was required. The screws were tightened; the pin and tension band was replaced; an autograft was applied; tobramycin-impregnated calcium sulfate beads were placed along the fracture site; and a Penrose drain was maintained for 3 days [Figure 4A].

Case no. 4, a 4-year-old, intact female Labrador retriever, was presented for an elective THR. Eight weeks later, radiographic signs of osteomyelitis included soft-tissue swelling, periosteal reaction, and osteolysis of the proximal femur and along the bone of the acetabular cement bone interface. Staphylococcus aureus was cultured from discharge at the incision site, and several courses of systemic antimicrobials (clavamox 13.25 mg/kg PO q 12 hours) were prescribed for the dog over the next few months. Eighteen weeks postoperatively, the implants were removed, and systemic antimicrobials (cephalexin 22 mg/kg PO q 8 hours) were continued based on sensitivity results. Six weeks following removal of the THR, radiographic evidence of osteomyelitis was noted as a periosteal response on the femur and acetabulum, along with soft-tissue and gas opacities surrounding the joint. Multiple cutaneous fistulas had developed over the medial, craniolateral, and caudal aspects of the thigh by 10 weeks after implant removal. Treatment then included surgical debridement of draining tracts, placement of 24 tobramycin-impregnated beads, a closed suction drain, and administration of baytril^c (7.5 mg/kg PO q 24 hours for 45 days).

Case no. 5, a 3-month-old, intact male Labrador retriever, was presented with a malunion of the radius resulting in antebrachial valgus and elbow incongruity. The puppy was treated with an ulnar ostectomy, and synostosis of the radius and ulna developed. The fused bones were surgically separated, and an osteotomy of the proximal ulna was performed. Four weeks postoperatively, a draining tract was present, and cephalexin (22 mg/kg PO q 8 hours for 28 days) was initiated. Eight weeks postoperatively, a draining tract was present with radiographic changes of a periosteal response and possible sequestration of the piece of ulna between the ostectomy and osteotomy. Surgical treatment included debridement of the draining tract, sequestrum, and osteotomy site prior to placement of six tobramycin-impregnated calcium sulfate beads in and around the bone defect. Oral antimicrobials with cephalexin were continued for an additional 4 weeks.

Case no. 6, a 4-year-old, spayed female Maltese, was presented with an open grade II, displaced transverse comminuted fracture of the distal one-third diaphysis of the radius and ulna caused by a dog bite. The fracture was treated with a hybrid external fixator, application of tobramycin-impregnated beads, and clavamox therapy. Staphylococcus sp. was found with intraoperative culture. Six weeks postoperatively, the wound had healed, but bone healing appeared delayed on radiographs. The external fixator was replaced with an eight-hole, 1.5/2.0 cuttable bone plate, and the fracture site was grafted with autogenous cancellous bone.

The duration of clinical signs from initial presentation until bead implantation ranged from 0 to 505 days, with a median of 194 days. The number of surgical procedures performed prior to bead implantation ranged from zero to three, with a median of two. Previous treatments included oral antimicrobial therapy, implant revisions and removal, and debridement [see Table].

The diagnosis of osteomyelitis was based on a combination of clinical signs (e.g., discharge, draining tracts, or lameness; n=4), radiographic signs (e.g., bone lysis, periosteal reactions, implant loosening; n=4), and/or positive cultures (n=5) [see Table]. Staphylococcus sp. was the most common organism isolated from cases with positive cultures (n=4), and Pseudomonas sp. was isolated in one case. The only dog (case no. 2) with a negative culture of the surgical site was presented 16 weeks after TPLO with a draining tract and implant loosening. This dog was diagnosed with cystitis, and Escherichia coli was isolated from a urine sample [Figures 4A, 4B, 4C].

Six beads were implanted into the surgical site in four dogs; nine beads were implanted into the surgical site in one dog [Figure 3A], and 24 beads were implanted into the surgical site of the dog with the THR. Concurrent treatments varied but included surgical debridement, implant removal or revision, bone grafts, drain placement, and oral antibiotics [see Table]. The different treatments varied based on indications such as implant loosening, healed fractures or delayed healing, and excessive drainage.

Radiographs were used to verify the presence of the beads postoperatively in all cases, and follow-up radiographs were obtained between 2 and 28 weeks after implantation [Figures 3A, 4A]. Beads lost most of their radio-opacity by 4 weeks and were no longer visible radiographically by 6 and 8 weeks [Figures 3B, 4B]. The outcome was considered successful if clinical and radiographic signs of infection resolved [Figures 3B, 4C]. Follow-up was available for the five dogs with osteomyelitis, and clinical signs resolved in all of these dogs [see Table]. Complete healing of osteotomies and fractures was confirmed radiographically in three of four dogs by 15±10 weeks after treatment. One dog did not return for radiographs after surgery, but communication with the owner 8 weeks following revision confirmed resolution of clinical signs. In case no. 6, no signs of infection developed after prophylactic implantation of beads and hybrid fixation of the open fracture. Radiographs made 6 weeks later revealed a lack of bone healing; this was attributed to the excessive rigidity of the external fixator, which justified revision of the repair.

Discussion

To the authors’ knowledge, this is the first report of the clinical application of antibiotic-impregnated calcium sulfate beads used to successfully treat osteomyelitis in dogs. The value of this study is limited by the small size of its population. Indications and follow-up times varied between cases, but these limitations are inherent to clinical studies. The tobramycin-impregnated calcium sulfate beads were used as an adjunct to surgical therapy for osteomyelitis. Therefore, outcomes were results of the beads and the other treatments provided. Resolution of clinical signs and resorption of the beads without side effects occurred in all cases in this study.

Surgical-grade calcium sulfate occurs naturally as a relatively pure alpha hemihydrate crystal. Calcium sulfate offers advantages over other antibiotic-loaded implants, because it is osteoconductive and provides a scaffold for new bone formation while occupying dead space, and it is biocompatible, completely resorbing without adverse reactions.^{8,10,13,15,16,18–20} Local antibiotic delivery with calcium sulfate has successfully resolved experimentally induced infections.^19,21,22 Medical literature also includes clinical reports of using antibiotic-impregnated calcium sulfate beads to treat osteomyelitis in humans.^8,10,20

Tobramycin-impregnated calcium sulfate beads (Osteoset-T) have recently become available in Canada and Europe for use in people, and they are approved in these countries for the treatment of osteomyelitis. They are not commercialized in the United States, and the use of calcium sulfate beads for local antibiotherapy remains off-label. In human patients, this kit is intended for single use. Repackaging the beads into smaller sets will improve the accessibility of this therapy by bringing the cost to a level similar to that of bone graft substitutes currently commercialized for veterinary applications (such as demineralized bone matrix).

The radiographic disappearance of beads (4 to 5 weeks) was slightly faster than the dissolution rate of 6 to 8 weeks previously published for calcium sulfate beads used in experimental animals.¹⁵ Possible explanations for the difference may include: differing rates of resorption when beads are implanted in the medullary canal versus in extraosseous soft tissues; the amount of inflammation present in the soft tissues; and the quantity of beads implanted in a site. Only six beads were implanted in the bone defects and surrounding soft tissues in four of the dogs in the authors’ study. In contrast, much larger numbers of beads are inserted into the medullary canal in experimental models and clinical cases.^9,15,21 The number of beads implanted is dependent on the amount of dead space, and most of the applications involved extremities with limited soft tissue and minimal bone defects. Therefore, the quantity of beads applied in the dogs of this study was much smaller than the quantity used experimentally and clinically in people.

The application of beads in clinical studies evaluating the efficacy of tobramycin calcium sulfate beads in people is determined by the size of the bone defect following adequate debridement. When using the Osteoset kit, the maximum number of antibiotic-impregnated beads implanted is one bead per kilogram of body weight.⁸ This maximum number is determined to prevent toxicity from the aminoglycoside absorption, and additional nonmedicated calcium sulfate beads can be used if additional dead space is present.⁸ The actual number implanted is subject to the surgeon, but it should be maximized to fill all bone defects and soft-tissue gaps. The maximum prescribed dose of tobramycin delivered in calcium sulfate in dogs has been calculated as 20 mg/kg; therefore, care should be taken to avoid systemic toxicity.²¹

Prophylactic local antibiotic delivery may be indicated in select cases of severely contaminated fractures, although when choosing the antibiotic, care should be taken to consider methicillin- and vancomycin-resistant bacterial species. Bacteria sensitive to tobramycin were cultured from each case in this series. Tobramycin, which exhibits bactericidal properties by inhibiting protein synthesis, was selected for its spectrum of activity against aerobic gram-negative and some aerobic gram-positive bacteria, including most Staphylococcus and Pseudomonas spp. Local antibiotic therapy is advantageous by providing delivery of high local doses and avoiding systemic concentrations.

Elution characteristics of tobramycin impregnated calcium sulfate beads have previously been evaluated.^15,16,21,23 The systemic concentration following implantation of tobramycin beads reached a peak at 1 hour, but it was undetectable after 24 hours.¹⁶ Systemic concentrations achieved even after experimentally implanting 1.8 times the maximum prescribed human dose to dogs were undetectable after 24 hours, and no changes in serum biochemical and hematological parameters were noted.²¹ No signs of systemic toxicity were seen in any of the dogs in this case series.

Bactericidal levels of tobramycin have been found to be maintained locally, and the antibiotic is released for up to 28 days.¹⁵ Additionally, previous studies have determined that calcium sulfate beads release bactericidal levels of aminoglycosides following ethylene oxide sterilization and storage at room temperature for at least up to 5 months.²³ Aminoglycosides are dose-dependent bactericidal antibiotics, and the local concentration of tobramycin is many times the minimum inhibitory concentration of most isolates found in osteomyelitis.^13,19,21,23 Experimentally, the elution concentrations of tobramycin locally are 2000- to 10,000-fold greater than the minimum inhibitory concentration for most Staphylococcus spp.¹⁰ Antibiotic release may persist beyond radiographic loss of the beads (28 days in this study), thereby maintaining local effect beyond 5 weeks.¹⁵

The high prevalence of infected TPLOs (two out of five cases of osteomyelitis) in this study is in accordance with previous reports of osteomyelitis as the most common complication associated with the TPLO procedure, occurring at a rate of 7.3%.^4,5 This incidence is high compared to the overall postoperative incidence of osteomyelitis (0.5% to 1.8%) in human patients.⁴ Although the cause for this predisposition to infection remains unclear, several factors may contribute to this issue. The extent of soft-tissue dissection around the proximal tibia may increase the likelihood of postoperative seroma or hematoma, thereby creating an environment that enhances bacterial colonization. In addition, the plate placement on the medial aspect of the proximal tibia is in an area with very little soft-tissue coverage.

Conclusion

Osteomyelitis is a serious condition that can lead to further complications such as nonunion, amputation, multiple invasive procedures, loss of function, and death. The lack of complications and the positive outcomes seen in this case series of dogs with osteomyelitis support the clinical application of tobramycin-impregnated calcium sulfate beads as an adjunct in the treatment of osteomyelitis.