Clinical value of bacterial culture in sinus tract in the diagnosis of fracture-related infection : IJS Global Health

Secondary Logo

Journal Logo

Diagnostic Study

Clinical value of bacterial culture in sinus tract in the diagnosis of fracture-related infection

Ma, Xianzhi MDa; Huang, Kunkun MScb; Ma, Sichao MSca; Lv, Ruiqiang MSca; Qi, Wanle MSca; Wang, Kai BSca; Chao, Shilan BScc,

Author Information
International Journal of Surgery: Global Health 5(5):p e81, September 2022. | DOI: 10.1097/GH9.0000000000000081
  • Open


Key points

  • This paper collected and analyzed clinical data on bacterial culture of sinus tract, to provide help for clinicians in the future diagnosis of fracture-related infection (FRI).
  • In the study, we found that identifying the pathogen and treating it with sensitive antibiotics are the most important treatment for FRI.
  • The sensitivity and specificity of bacterial culture in sinus tract in determining pathogenic bacteria are not high, so it cannot be used as the main means of examination.


FRI has been one of the common diseases in clinical practice for orthopedic surgeons. Among many clinical diagnostic examinations, bacterial culture of the sinus tract, as an auxiliary examination, is increasingly valued by orthopedic surgeons for the diagnosis of FRI. It has been reported in previous literature1 that preoperative bacterial culture of the sinus tract is a routine examination with a high diagnostic rate. However, the study of Mckowicck et al2 has denied this opinion, believing that intraoperative tissue bacterial culture was the main examination method for diagnosis of FRI, which is still widely accepted by clinicians today. While more researches have been done, some scholars insist that the results of sinus culture are basically consistent with the results of intraoperative bacterial culture, and the two have a high degree of consistency3. However, most still believe that intraoperative bacterial culture is still the most important diagnostic method for FRI at present. To address this controversy, this paper collected and analyzed clinical data on bacterial culture of the sinus tract, to provide help for clinicians in the future diagnosis of FRI.

Data and methods

Patients with FRI and sinus tract from June 2016 to June 2021 (n=53) in Qinghai Provincial People’s Hospital were selected for a retrospective analysis. Patients included 46 males and 7 females, from 16 to 74 years old (mean±SD: 37±2.26). The most common infection sites were femur (18 cases), tibiofibular (17 cases), hip joint (7 cases), ankle (5 cases), radius and ulna (3 cases), patella (2 cases), and pelvis (1 case).

All antibiotics were stopped immediately in 53 patients with sinus tract after admission, and the sample tissues were taken for bacterial culture 48 hours after the withdrawal and during the operation.

The Ethics Committee of our hospital approved this study and any related procedures were performed in accordance with relevant guidelines and regulations. The work has been reported in line with the STROCSS criteria4. The current study has been registered in system with a registration unique identifying number (UIN) researchregistry8106

Preoperative sample collection

According to the principles of aseptic operation, the skin around the sinus tract was disinfected twice with 3% iodophor. Two sterile cotton swabs were taken, respectively, inserted into the sinus tract and stirred to obtain secretions, and then placed into a sterile test tube filled with a small amount of normal saline. The cotton plug closed the test tube, and bacteria culture was conducted in time.

Intraoperative sample collection

After the exposure of the nidus, the degenerated necrotic granulate tissue and broken dead bone were removed with curets or bone biters, and placed into a sterile test tube filled with a small amount of normal saline. The test tube was closed with cotton plug, and bacteria culture was conducted in time.

In order to improve the detection rate of bacteria and identify pathogenic bacteria, we stopped antibiotics 48 hours before surgery, or sometimes even 2 weeks. During the operation, samples should be cut with a sharp knife rather than an electric knife, because the electric knife generates huge heat in the process of cutting tissue, which easily leads to tissue necrosis and interferes with the separation of pathogenic bacteria. Tissue samples obtained should be clamped with clean instruments and transferred directly into culture containers, without contacting with gloves or towels, as this principle reduces the incidence of false positives. Culture samples or joint fluid obtained should be transferred immediately. The use of blood media can improve the detection rate of pathogenic microorganisms and reduce the potential contamination of other media. The bacteria detection rate of joint fluid in the blood medium is higher, and the specificity and sensitivity of diagnosis are better. If fungal or mycobacterium infection is suspected, special isolation media should be used. Another way to increase the detection rate is to extend the incubation period to 14 or 21 days. Finally, at least 3–5 samples should be sent for examination.


Eighty-two samples collected from 53 patients with FRI were sent for bacterial culture. The detected bacteria were listed in Table 1. The 3 most frequently detected were Staphylococcus aureus (20 cases/32.26%), Staphylococcus epidermidis (8 cases/12.9%), and Pseudomonas aeruginosa (5 cases/8.06%). No bacteria were detected in 20 cases.

Table 1 - Detected bacteria in sinus tract.
Bacteria from sinus tract bacterial culture Sample number (%)
Staphylococcus aureus 20 (32.26)
Staphylococcus epidermidis 8 (12.90)
Pseudomonas aeruginosa 5 (8.06)
Bacillus levans 3 (4.84)
Escherichia coli 3 (4.84)
Staphylococcus haemolyticus 2 (3.23)
Enterococcus faecalis 2 (3.23)
Klebsiella pneumoniae 2 (3.23)
Acinetobacter baumannii complex 2 (3.23)
Candida albicans 2 (3.23)
Rhodococcus 2 (3.23)
Peptostreptococcus anaerobius 2 (3.23)
Staphylococcus warneri 1 (1.61)
Enterobacter polymeris 1 (1.61)
Enterobacter lang 1 (1.61)
Providencia rettgeri 1 (1.61)
Staphylococcus hominis 1 (1.61)
Citrobacter braakii 1 (1.61)
Staphylococcus saprophyticus 1 (1.61)
Proteus mirabilis 1 (1.61)
Klebsiella oxytoca 1 (1.61)
In total 62 (100.00)
No growth 20

The results of bacterial cultures were sorted into gram-positive (G+) (40 times/64.5%) and gram-negative (G−) (20 times/20.32%) bacteria, and fungus (2 times/3.2%). The data are shown in the Figure 1.

Figure 1:
The distribution of gram bacteria in sinus tract. G+ indicates gram-positive; G−, gram-negative.

Among the 62 sets of G+ bacteria from the bacterial culture in sinus tract, there are 56 (90.30%) single strains and 6 (9.70%) multiple strains. Among the G+ bacteria in the intraoperative bacterial culture, 9 strains (12.9%) and 53 strains (87.10%) were multistrain and single-strain, respectively.

Since 54 patients underwent intraoperative bacterial cultures, the number of samples in sinus tract should be consistent with the former. Statistical results showed that the agreement rate is 67.30%. Meanwhile, the agreement rate between the bacterial culture results of sinus tract and the pathogenic bacteria was 51.50% (to determine the pathogenic bacteria, at least 2 samples should be sent for intraoperative examination, and the same positive bacteria were cultured from at least 2 samples) (Table 2).

Table 2 - The agreement rate between positive bacteria in sinus tract, in intraoperative bacterial culture and pathogenic bacteria.
Bacterial species from sinus tract bacterial culture Sample number Number of cases sent to intraoperative examination Same with intraoperative culture Agreement rate (%) Determined pathogens Agreement rate (%)
Staphylococcus aureus 20 16 10 62.50 7 7/16(43.75)
Staphylococcus epidermidis 8 7 4 57.13 3 3/7(42.86)
Pseudomonas aeruginosa 5 4 3 75 0 0/4 (0)
Bacillus levans 3 1 1 100 1 1/1 (100)
Escherichia coli 3 3 3 100 3 3/3 (100)
Staphylococcus haemolyticus 2 1 1 100 1 1/1 (100)
Enterococcus faecalis 2 1 1 100 1 1/1 (100)
Klebsiella pneumoniae 4 3 3 100 2 23 (66.67)
Acinetobacter baumannii complex 2 2 2 100 2 2/2 (100)
Candida albicans 2 1 0 0 0 0
No growth 20 13 7 53.85
Total 71 52 35 35/52 (67.30) 20 20/39 (51.50)

The sensitivity and specificity of bacterial culture in sinus tract is 76.92% and 26.90%, respectively (Table 3).

Table 3 - Comparison between the 2 results.
Results of intraoperative bacterial culture
Sinus tract Positive Negative Total
Positive 20 19 39
Negative 6 7 13
Total 26 26
Sensitivity 76.92%
Specificity 26.90%
Positive predictive value 51.30%
Negative predictive value 53.58%


FRI is one of the most challenging complications of fracture surgical treatment4, bringing heavy burden and economic cost to patients and society5–7. The incidence of FRI is 1%–2% in closed fractures and 30% in open fractures5. To date, amputation rates and recurrent infection rates due to FRI remain high5,6. Once FRI occurs, it will lead to an increase in the number of operations for patients, an increase in antibacterial treatment, a prolonged period of healing, and a great impact on prognosis5–7. In addition to causing physical and mental damage to patients, the medical costs of infected patients are 6.5 times higher than those of uninfected patients6–11.

Radical debridement, using intravenous antibiotics and implantation of local antibiotics sustained-release carriers are the main methods for the treatment of FRI at the present stage. The most important factor affecting the surgical efficacy is whether the correct pathogenic bacteria can be identified. In the past, bacterial culture of sinus tract was considered to have good reliability1. However, the study of Mackowiak et al2 published in JAMA in 1978 has completely changed the understanding of bacterial culture of sinus tract, pointing out that in order to make a correct bacteriological diagnosis, dead bone, pus, implants and other materials must be taken for bacteriological culture during the operation, and bacterial culture of sinus tract cannot truly reflect pathogenic bacteria. Several subsequent articles reported on the importance of intraoperative bacteriological culture, which is still the main test for diagnosing chronic osteomyelitis and identifying pathogens.12–16

According to the results, in the bacterial culture of sinus tract and intraoperative culture, the positive rate of S. aureus was the highest (34%), followed by S. epidermidis and P. aeruginosa. This result is consistent with that of related studies17,18. In the sinus tract bacterial culture, we found 2 cases of Candida albicans infection, which belonged to anaerobe infection. Anaerobe infection was rarely found in patients with FRI, and no anaerobe infection was detected in the intraoperative bacterial culture, which may be the reason for the absence of anaerobe culture or infection. In case no pathogens were detected repeatedly during operations, the possibility of fungi and divergent Tuberculosis bacilli should be considered.

The agreement rate between sinus tract bacterial culture and intraoperative bacterial culture is 67.30%, while the former one has 51.50% agreement rate with the pathogenic bacteria. In clinical studies, Mackowiak et al2 and Patzakis et al19 reported that the agreement rates of sinus culture and intraoperative culture were 44% and 47%. On the contrary, some studies reported higher agreement rates, such as the studies Mousa20 and Perry et al3, who reported the 88.7% and 70% agreement rates, respectively.

Patients stopped using antibiotics 48 hours or even 2 weeks before the bacterial culture of sinus tract and the first debridement. Multiple samples were taken during the operation, and at least 3–5 samples were sent for culture. Through these measures, the detection rate can be improved. But the agreement rate of bacterial culture in the sinus was not high, which may be because <2 samples were sent during the operation, or there was no growth of the same bacteria in at least 2 samples.

According to the results of bacterial culture in sinus tract and during operation, more than 87% of the patients were infected with single strain, and about 10% of the patients were infected with multiple strains. Most of the patients with FRI were caused by single strain infection, and multistrain infection was likely to be contaminated bacteria rather than pathogenic bacteria. The reason may be that bacterial culture of the sinus tract contaminated local tissues and affected bone and nonbone tissues. Long-term use of sensitive antibiotics can basically kill pathogenic bacteria, and pollution bacteria are inhibited and localized.

In dealing with patients with FRI, the sensitivity and specificity of sinus tract bacterial culture in determining pathogenic bacteria are not high. Therefore, we should carefully use the results of this examination, and more attention should be paid to taking multiple samples during operation sent to bacterial culture, especially dead bone, pus, and around implants. In addition, aseptic technique during the operation, timely inspection and long-time bacterial culture are conducive to improve the detection rate of pathogenic bacteria.

Several limitations of our study should be pointed out. The sample size in our study, as a retrospective research, was relatively small. Since patients came from only a local hospital, results cannot be generalized to a larger area. Hopefully some related studies in other national areas can be supplementary data. Carrying out a multicenter study, refining parameters, increasing sample size, and incorporating big data comparison should be done in the next research. Gene sequencing to identify gene mutations to confirm the pathogenic bacteria is also worthy further research.


Identifying the pathogen and treating it with sensitive antibiotics are the most important treatment for FRI. The sensitivity and specificity of bacterial culture in sinus tract in determining pathogenic bacteria are not high, so it cannot be used as the main means of examination. We suggest that taking multiple samples at different sites to improve its detection rate.

Ethical approval

The Ethics Committee of our hospital approved this study and any related procedures were performed in accordance with relevant guidelines and regulations.

Sources of funding

This work was supported by Grants from Natural Science Foundation of Qinghai Province (2019-ZJ-927).

Authors’ contribution

X.M. and K.H. conceived the study. S.M., R.L., W.Q., and K.W. collected the data. R.L., S.H., W.Q. and K.W. contributed to patient follow-up. X.M. and K.H. analyzed the data. X.M., K.H., and S.M. performed surgeries. X.M. and K.H. wrote the paper.

Conflicts of interest disclosure

The authors declare that they have no financial conflict of interest with regard to the content of this report.

Research registration unique identifying number (UIN)





1. Waldvogel FA, MedoV G, Swartz MN. Osteomyelitis: a reviewof clinical features, therapeutic considerations, and unusualaspects. N Engl J Med 1970;282:198–206.
2. McKowiack PA, Jones RS, Smith JW. Diagnostic value ofsinus tract cultures in chronic osteomyelitis. J Am Math Assoc 1978;239:2772–5.
3. Perry CR, Pearson RL, Miller G. A accuracy of cultures of materials from swabbing of the superficial aspect of the wound and needle biopsy in the pre-operative assessment of osteomyelitis. J Bone J Surg 1991;73A:745–9.
4. Mathew G, Agha R. STROCSS 2021: Strengthening the Reporting of cohort, cross-sectional and case-control studies in Surgery. Int J Surg 2021;96:106165.
5. Metsemakers WJ, Morgenstern M, McNally MA, et al. Fracture-related infection: a consensus on definition from an international expert group. Injury 2018;49:505–10.
6. Metsemakers WJ, Smeets B, Nijs S, et al. Infection after fracture fixation of the tibia: analysis of healthcare utilization and related costs. Injury 2017;48:1204–10.
7. Bezstarosti H, Van Lieshout EMM, Voskamp LW, et al. Insights into treatment and outcome of fracture-related infection: a systematic literature review. Arch Orthop Trauma Surg 2019;139:61–72.
8. Yokoyama K, Itoman M, Uchino M, et al. Immediate versus delayed intramedullary nailing for open fractures of the tibial shaft: a multivariate analysis of factors affecting deep infection and fracture healing. Indian J Orthop 2008;42:410–9.
9. Hoekstra H, Smeets B, Metsemakers WJ, et al. Economics of open tibial fractures: the pivotal role of length-of-stay and infection. Health Econ Rev 2017;7:32.
10. van Houten AH, Heesterbeek PJ, van Heerwaarden RJ, et al. Medial open wedge high tibial osteotomy: can delayed or nonunion be predicted? Clin Orthop Relat Res 2014;472:1217–23.
11. Depypere M, Morgenstern M, Kuehl R, et al. Pathogenesis and management of fracture-related infection. Clin Microbiol Infect 2020;26:572–8.
12. Zuluaga AF, Galvis W, Jaimes F, et al. Lack of microbiological concordance between bone and non-bone specimens in chronic osteomyelitis: an observational study. BMC Infect Dis 2002;2:8.
13. Jacobson IV, Sieling WL. Microbiology of secondary osteomyelitis and value of bone biopsy. S Afr Med J 1987;72:476–7.
14. Khatri G, Wagner DK, Sohnle P. EVect of bone biopsy in guiding antimicrobial therapy of osteomyelitis complicating open wounds. Am J Med Sci 2001;321:367–71.
15. Aggarwal VK, Higuera C, Deirmengian G, et al. Swab cultures are not as effective as tissue cultures for diagnosis of periprosthetic joint infection. Clin Orthop Relat Res 2013;471:3196–203.
16. Dy Chua J, Abdul-Karim A, Mawhorter S, et al. The role of swab and tissue culture in the diagnosis of implantable cardiac device infection. Pacing Clin Electrophysiol 2005;28:1276–81.
17. Onuminya JE. A prospective evaluation of the diagnostic value of sinus specimen cultures in chronic osteomyelitis. Trop Doct 2006;36:38–39.
18. Ulug M, Ayaz C, Celen MK. Are sinus-track cultures reliable for identifying the causative agent in chronic osteomyelitis? Arch Orthop Trauma Surg 2009;129:1565–70.
19. Patzakis MJ, Wilkins J, Kumar J, et al. Comparison of the results of bacterial cultures from multiple sites in chronic osteomyelitis of long bones. A prospective study. J Bone Joint Surg Am 1994;76:664–6.
20. Mousa HA. Evaluation of sinus tract cultures in chronic bone infection. J Bone Jt Surg Br 1997;79:567–9.

Fracture-related infections; Bacterial culture; Sinus tract

Copyright © 2022 The Authors. Published by Wolters Kluwer Health, Inc. on behalf of IJS Publishing Group Ltd.