Is non-union of tibial shaft fractures due to nonculturable bacterial pathogens? A clinical investigation using PCR and culture techniques

Abstract

Background: Non-union continues to be one of the orthopedist's greatest challenges. Despite effective culture methods, the detection of low-grade infection in patients with non-union following tibial fracture still presents a challenge. We investigated whether "aseptic" tibial non-union can be the result of an unrecognized infection.

Methods: A total of 23 patients with non-union following tibial shaft fractures without clinical signs of infection were investigated. Intraoperative biopsy samples obtained from the non-union site were examined by means of routine culture methods and by polymerase chain reaction (PCR) for the detection of 16 S ribosomal RNA (rRNA). Control subjects included 12 patients with tibial shaft fractures.

Results: 23 patients (8 women and 15 men; mean age: 47.4 years) were included into this study. Preoperative C-reactive protein levels (mean: 20.8 mg/l) and WBC counts (mean: 8,359/μl) in the study group were not significantly higher than in the control group. None of the samples of non-union routine cultures yielded microorganism growth. Bacterial isolates were found by conventional culturing methods in only 1 case of an open fracture from the control group. In this case, PCR yielded negative results. 16 S rRNA was detected in tissue specimens from 2 patients (8.7%) with non-union. The analysis of these variable species-specific sequences enabled the identification of specific microorganisms (1x Methylobacterium species, 1x Staphylococcus species). Both PCR-positive patients were culture-negative.

Conclusions: The combination of microbiological culture and broad-range PCR seems to substantially add to the number of microbiological diagnoses obtained and may improve the clinician's ability to tailor therapy to the individual patient's needs.

Figure 1

Consecutive x-rays of a 30-year-old male with tibial non-union and valgus malalignment following stabilization with external fixator for 6 months (Figures a,b). Postoperative findings after excision of necrotic tissue and re-osteosynthesis with a multidirectional locking plate (tifix®-tibia-plate, Litos, Hamburg, Germany) in combination with reconstruction of skeletal defects by implantation of autologous bone removed from the iliac crest (Figures c,d). Conventional and molecular bacteria detection methods were both negative.

Figure 2

38-year-old male with an open tibial fracture (control group). This figure demonstrates the findings before the initial treatment. The isolated pathogens in culture were Streptococcus suis and Enterococcus species. Tissue from the fracture gap after initial débridement yielded negative results by PCR.