The management of long-bone nonunion remains challenging.1 Typically in the setting of active infection, a staged protocol is usually employed with emphasis on initial infection eradication followed by definitive treatment of the nonunion.2,3 A particular diagnostic and therapeutic dilemma exists in patients that seem to be free of infection at their nonunion site but reveal positive intraoperative cultures taken at the time of nonunion surgery. Often in this scenario, preoperative laboratory assessment such erythrocyte sedimentation rate, C-reactive protein, and a white blood cell count was not able to indicate an indolent infection.4–6 A recent multicenter study showed a union rate of 95% in patients with “surprise” intraoperative positive cultures at the expense of increased reoperation rates.7 Similarly, in an earlier study, we reported on our experience of 25 presumed aseptic diaphyseal long-bone nonunions with positive cultures.8 Using a standardized single-stage surgical approach, a 28% reoperation rate was noted due to persistent nonunion with previous positive intraoperative cultures; eventually, all 25 patients went on to union. We have revisited our early experience to include long-bone nonunions of the proximal, middle, and distal third (including the metaphysis) thought to be aseptic. We employed a single-stage surgical protocol for all nonunion cases with no obvious evidence of infection preoperatively.
The objectives of our study were to (1) to assess the rate of radiographic union after index nonunion surgery, (2) to report the final rate of union after subsequent procedures, and (3) to analyze the need for additional surgeries in patients with long-bone nonunions and positive cultures obtained at the time of index nonunion surgery.
MATERIAL AND METHODS
Between 1992 and 2014, 922 patients underwent operative treatment for a nonunion. All patients were treated by fellowship-trained Orthopaedic traumatologists (D.L.H. and D.S.W.). Nonunion was defined as lack of any progressive healing on radiographs over a 3-month period or failure to heal 9 months after the initial injury.9,10 Patients with a remote history of wound infections, sinuses, osteomyelitis, previous treatment with antibiotic-impregnated cement beads, and history of pathologic fractures were excluded (n = 24). Patients with a diagnosis of aseptic nonunion confirmed through negative tissue culture results from surgery were excluded. This left 77 patients for inclusion; all with a presumed aseptic long-bone nonunion that grew positive cultures obtained at the time of nonunion surgery. Preoperative assessment of all 77 patients exhibited no signs and/or symptoms of infection. This included a clinical examination to rule out the presence of fever, draining wound, or sinus and a laboratory work up with normal values of erythrocyte sedimentation rate, C-reactive protein, and white blood cell count with differential.
The medical charts of all 77 patients were reviewed and following data were recorded: age, sex, anatomic region of the nonunion, high- versus low-energy mechanism, history of open versus closed fracture, number of previous surgeries, time from injury to index nonunion surgery, time from index nonunion surgery to healing, and the incidence of secondary revision surgeries. Additional data included the organism(s) that grew from intraoperative cultures, the antibiotic(s) prescribed, and the duration of administration.
In total, there were 35 women and 42 men with an average age of 48 years (range, 18–96). Average length of time between initial injury and index nonunion surgery was 30 months (range, 3–336 months). High-energy mechanism was the most common cause injury (n = 39, 51%), and 22 (29%) patients had a history of open fractures. Patients had undergone an average of 2 surgeries (range, 1–9) before our index nonunion surgery (Table 1).
Based on anatomic region, there were 10 clavicle nonunions, 8 humeral nonunions, 4 radius/ulnar nonunions, 27 femoral nonunions, 27 tibial nonunions, and 1 fibular nonunion (Tables 2 and 3).
A wide spectrum of bacterial species was isolated from cultures (Tables 2 and 3). The most common bacteria was coagulase-negative staphylococcus species (32 patients) followed by propionibacterium acnes (21 patients). Thirteen patients had multiple organisms. All patients were followed until healing occurred, resulting in an average follow-up of 26 months (range, 3–136 months).
Preoperative planning included obtaining previous operative reports with implant information to facilitate its removal. Antibiotics were held in each case of nonunion surgery until at least 5 cultures were obtained. Fifty (65%) patients underwent open debridement of the nonunion site followed by surgical stabilization through plates and screws. Twenty-seven (35%) patients underwent exchange nailing with canal reamings used for cultures. Standard postoperative antibiotics were administered between 24 and 48 hours per surgeon preference. After cultures returned positive, an infectious disease specialist was consulted. After multidisciplinary consultation, 51 (66%) patients were diagnosed with an infected nonunion, whereas positive cultures in the remainder of cases (n = 26) were deemed contaminates not necessitating tailored and prolonged antibiotic therapy in the postoperative period. In general, patients deemed infected were placed on intravenous antibiotic therapy for a minimum of 6 weeks after surgery followed by an oral antibiotic agent until healing had occurred. A total of 76 (99%) patients received some sort of biologic augmentation at the time of index nonunion surgery (Table 4).
Statistical analysis was performed using SAS software Version 9.2 (SAS Institute Inc, Cary, NC). Descriptive statistics were reported using mean (range) and frequency (percentage). A 2-sample t test was used to compare differences in continuous variables between the infected and contaminated groups. Chi square tests were used when comparing categorical variables between both groups.
Fisher exact test was used to compare the rate of union after index nonunion surgery, the rate of revision procedures, and the final union rate between groups at the latest follow-up. The α level was set at 0.05 for statistical significance.
Osseous union after index nonunion surgery was achieved in 84% of the patients (65 of 77) (Table 5).
After index nonunion surgery, 78% (40 of 51 patients) healed in the infected group and 88% (23 of 26 patients) in the contaminated group (88%) (P = 0.303) (Table 5). For patients that healed after their index nonunion surgery, time to clinical union was 6.3 months (range, 1–24 months) and time to radiographic union was 7.4 months (range, 2–24 months).
Eighteen percent (14 of 77 patients) did not heal after index nonunion surgery, and therefore, required additional surgeries and details are listed in Table 6. In the infected group, 22% (11 of 51 patients) needed additional procedures for a persistent nonunion (Table 5). At final follow-up, all patients healed.
In the contaminated group, 12% (3 of 26 patients) needed additional surgery to treat a persistent nonunion (Table 5). One patient with a bisphosphonate-associated subtrochanteric femur nonunion failed to unite despite 2 revision surgeries and went on to undergo total hip arthroplasty. Furthermore, 2 patients that had previously undergone exchange nailing for a tibial shaft and a femoral shaft nonunion underwent proximal interlocking screw removal for dynamization purposes (Table 6). In total, 96% (25 of 26 patients) in contaminated group healed eventually and the final union rate for all patients was 99% (76 of 77) at the latest follow-up (Table 5).
Nine patients (18%) in the infected group underwent removal of symptomatic hardware after the fracture had healed. Four of these patients (44%) had positive cultures obtained at the time of hardware removal and were placed on intravenous antibiotics regimens for 6 weeks. In the contaminated group, 2 patients (8%) underwent removal of symptomatic interlocking screws in healed tibial shaft nonunions.
Eradication and reconstruction of infected long bones remain a challenge for the treating surgeon. When there is obvious infection, most advocate for a staged protocol that entail removal of all hardware, aggressive debridement, temporary external fixation, and revision internal fixation when soft tissues are amenable and the patient is cured of infection.11–13
In the absence of clinical and laboratory findings that suggest infection, our treatment strategy uses single-stage reconstruction of long-bone nonunions with removal of all implants, at least 5 cultures, followed by thorough debridement and definitive surgical fixation.8 The purpose of our study was to evaluate the outcomes of this protocol in patients with positive intraoperative cultures obtained at the time of index nonunion surgery.
This single-stage protocol of long-bone nonunions with positive intraoperative cultures had an overall success rate of 84%. Olszewski et al7 noted in their retrospective multicenter study a slightly lower rate (78%) of bony union after 1 nonunion surgery. Patients with at least 1 positive culture had an increased reoperation rate (22%) compared with patients with aseptic nonunions (4%). In our study, 11 patients (22%) in the infected group required revision procedure for a persistent nonunion, 9 of these patients achieved union after only 1 additional surgery. The other 2 patients underwent complex reconstructive surgery with a Taylor spatial frame and bone transport, and final union rate for the entire infected nonunion cohort was 100% at the latest follow-up. This echoes our early experience in presumed aseptic diaphyseal nonunions with positive cultures, in which we noted an increased incidence of reoperation rates, but eventually achieved successful union rates.8
The significance of a positive culture needs to be evaluated in the context of the clinical scenario and not always translates into inferior results.14 In our study, when deemed contaminated, patients had excellent results with a union rate of 88% after initial nonunion surgery. This increased to 96% union rate after minor additional procedures; this aligns with results for aseptic nonunions in other studies.1,15,16 For the infected patient, repeat salvage surgery has a high probability of success, although this study showed the importance of working in a multidisciplinary fashion by collaborating with infectious disease specialists to optimize a patient's treatment.13
The study has several limitations. It is retrospective in its nature. Second, patients differed in their initial injury severity and fracture management. The exact classification details in open fractures were not readily accessible to us because all patients were initially treated at outside institutions. In addition, differentiating a subclinical infected nonunion from a contaminated nonunion proves to be difficult. Our typical practice is to consult with an infectious disease specialist, and in the setting of 1 of 5 positive cultures growing a low virulent organism, no prolonged antibiotics are given. This strategy achieved an 88% single-stage success rate in the contaminated group. Three or more positive cultures of 5 were considered infected. Those with 2 positive cultures were treated as “infected” or “contaminated” after a thorough review of the particular bacteria isolated and the patient's history in conjunction with the infectious disease specialist.
In summary, 84% of presumed aseptic nonunions of long-bone fractures with positive intraoperative cultures fully healed after a single-stage surgical protocol and long-term antibiotic when appropriate. When patients are diagnosed with a subclinical infected nonunion, they should be counseled about the higher likelihood of reoperation, but most cases can expect excellent union rates after 1 additional surgery.
1. Hak DJ. Management of aseptic tibial nonunion. J Am Acad Orthop Surg. 2011;19:563–573.
2. Cierny G III. Surgical treatment of osteomyelitis. Plast Reconstr Surg. 2011;127(suppl 1):190S–204S.
3. Struijs PAA, Poolman RW, Bhandari M. Infected nonunion of the long bones
. J Orthop Trauma. 2007;21:507–511.
4. Stucken C, Olszewski DC, Creevy WR, et al. Preoperative diagnosis of infection in patients with nonunions. J Bone Joint Surg Am. 2013;95:1409–1412.
5. Bourguignat A, Férard G, Jenny JY, et al. Diagnostic value of C-reactive protein and transthyretin in bone infections of the lower limb. Clin Chim Acta. 1996;255:27–38.
6. Wright EH, Khan U. Serum complement-reactive protein (CRP) trends following local and free-tissue reconstructions for traumatic injuries or chronic wounds of the lower limb. J Plast Reconstr Aesthet Surg. 2010;63:1519–1522.
7. Olszewski D, Streubel PN, Stucken C, et al. Fate of patients with a “surprise” positive culture after nonunion surgery. J Orthop Trauma. 2016;30:e19–e23.
8. Amorosa LF, Buirs LD, Bexkens R, et al. A single-stage treatment protocol for presumptive aseptic diaphyseal nonunions: a review of outcomes. J Orthop Trauma. 2013;27:582–586.
9. Bhandari M, Guyatt GH, Swiontkowski MF, et al. A lack of consensus in the assessment of fracture healing among orthopaedic surgeons. J Orthop Trauma. 2002;16:562–566.
10. Bhandari M, Fong K, Sprague S, et al. Variability in the definition and perceived causes of delayed unions and nonunions: a cross-sectional, multinational survey of orthopaedic surgeons. J Bone Joint Surg Am. 2012;94:e1091–e1096.
11. Sadek AF, Laklok MA, Fouly EH, et al. Two stage reconstruction versus bone transport in management of resistant infected tibial diaphyseal nonunion with a gap. Arch Orthop Trauma Surg. 2016;136:1233–1241.
12. Conway J, Mansour J, Kotze K, et al. Antibiotic cement-coated rods: an effective treatment for infected long bones
and prosthetic joint nonunions. Bone Joint J. 2014;96-B:1349–1354.
13. Bose D, Kugan R, Stubbs D, et al. Management of infected nonunion of the long bones
by a multidisciplinary team. Bone Joint J. 2015;97-B:814–817.
14. Moussa FW, Anglen JO, Gehrke JC, et al. The significance of positive cultures from orthopedic fixation devices in the absence of clinical infection. Am J Orthop. 1997;26:617–620.
15. Wu CC, Shih CH, Chen WJ, et al. High success rate with exchange nailing to treat a tibial shaft aseptic nonunion. J Orthop Trauma. 1999;13:33–38.
16. Oh JK, Bae JH, Oh CW, et al. Treatment of femoral and tibial diaphyseal nonunions using reamed intramedullary nailing without bone graft. Injury. 2008;39:952–959.