Does Antibiotic Treatment Before Bone Biopsy Affect the Identification of Bacterial Pathogens From Bone Culture?

There is a common belief that any exposure to antibiotics prior to bone biopsy renders culture results unreliable. In the presence of a negative culture, some physicians disregard the results even if there has been only a single dose of antibiotics, hypothesizing that antibiotic penetration of bone could interfere with recovery and identification of bacteria. Clinically, this is problematic when clinical findings such as “probe to bone” or MRI contradict a negative bone culture. In these cases, the gold standard may be disregarded in favor of less-reliable tests.

The majority of studies that discuss antibiotics and bone culture yield are from patients with hematogenous VO.¹ The results of these studies have been generalized to other types of bone infections that occur from direct extension, such as DFO. The aim of the current study was to evaluate the effect of antibiotic exposure and bacterial pathogen identification in patients with DFO. The null hypothesis was that bacterial culture identification would be the same in patients who did not receive pre-biopsy antibiotics and patients who received pre-biopsy antibiotics.

This study was approved by the Institutional Review Board at the University of Texas Southwestern Medical Center and Parkland Hospital. A total of 114 patients with DFO were identified based on positive histology. Histologic criteria for osteomyelitis included the presence of bacteria, lymphocytes, polymorphonuclear neutrophils, plasma cells, and fibrinoid necrosis.² A diagnosis of diabetes was based on criteria from the American Diabetes Association.³ The primary outcome measurement was the proportion of bone biopsies with positive bacterial cultures. The authors identified demographic, medical, and social history from inpatient and outpatient electronic medical records, documenting the comorbidities of neuropathy, peripheral vascular disease, retinopathy, and CKD, which was defined as an estimated glomerular filtration rate <60 mL per minute. Neuropathy was defined as abnormal vibration sensation or abnormal sensation with 10-gram Semmes-Weinstein monofilaments, and peripheral arterial disease as an ankle-brachial index of less than 0.9 or greater than 1.3. Intraoperative bone biopsies were obtained under direct visualization. The bone was divided and sent for culture and histologic evaluation. 

The data were analyzed using SPSS software version 24 for Macintosh (IBM). A chi-square test was used to compare dichotomous variables, and ANOVA was used for continuous variables with an alpha of 0.05. 

Patient characteristics and demographic data are documented in the Table. There were no differences in patients who did and did not receive antibiotics prior to bone biopsy. When comparing the proportion of patients with a positive bone culture, there was no difference in bacterial culture yield in patients who did not receive antibiotics prior to bone biopsy (n = 22, 84.6%) and patients who received antibiotics (n = 70, 79.5%, P = .57). There was no difference in bacterial culture yield between patients who did not receive antibiotics (n = 22, 84.6%), patients who received antibiotics for 1 to 3 days (n = 47, 85.1%), and patients who received antibiotics greater than or equal to 4 days (n = 41, 73.2%, P = .31). Polymicrobial bone infections were common in both groups (no antibiotics, 30.8%; antibiotics, 45.5%, P = .18).

This study failed to reject the null hypothesis that there were no differences in bacterial yield whether antibiotics were withheld or not before bone cultures were obtained, and the duration of antibiotics prior to bone biopsy did not change the bacterial yield. A review of the literature was only able to identify 1 paper that evaluated the impact of antibiotics on bacterial yield in patients with DFO. A prospective study by Cecillia-Matilla et al⁴ found significantly lower bone biopsy culture yield in patients who did not receive pre-biopsy antibiotics compared to patients who received pre-biopsy antibiotics (n = 165, 61.8% vs 87.8%, P < .001). This is the opposite of what would be expected based on established practice patterns. Cecillia-Matilla et al⁴ and the current study are 2 of the largest studies that address the role of pre-biopsy antibiotics. Neither support the traditional belief to withhold antibiotics.

The gold standard to diagnose DFO based on the IDSA Diabetic Foot Infection Guidelines is bone biopsy with evaluation of both culture and histology.⁵ Guidelines from the IDSA and the International Working Group on the Diabetic Foot and the Société de Pathologie Infectieuse de Langue Française⁶ regarding DFO endorse withholding antibiotics for 2 weeks before obtaining bone cultures, even though there is no evidence in DFO to support these recommendations. The rationale for withholding antibiotics is based on studies in patients with VO, which has distinctly different pathogenesis from DFO.^1,7-9

The evidence in VO is inconsistent and features important methodologic flaws. In a previous work, several of the authors of the current study identified 9 papers that evaluated antibiotic exposure and bone culture yield in patients with VO. Four studies supported withholding antibiotics, and 5 studies demonstrated no significant effect.¹ 

There are serious methodical deficits in the VO literature.¹ Inconsistent and subjective criteria have been used as the reference standard for diagnosing osteomyelitis in most studies. The gold standard of bone biopsy for culture and histology was not the sole criteria to define VO in any of the studies identified in the 2019 meta-analysis.¹ None of the studies used the same reference standard. The majority of identified VO studies were small (median, 46 patients enrolled; range, 20-124), with low culture yields among patients who received antibiotics (median, 25%; range, 0%-72%) and who did not receive antibiotics (median, 56%; range, 29%-88%). This could be explained by the poor reference standards that allowed patients without VO into studies. Based on the small sample size, low culture yields, and flawed criteria used to define VO, physicians should question the traditional belief that withholding antibiotics prior to bone biopsy impacts culture results in patients with VO. 

The results of the current study may have an impact on the diagnosis and therapeutic effectiveness of DFO treatment. The criterion used to determine if all the infected bone has been removed is a clean margin with negative bone culture and histology. If physicians believe that antibiotic exposure renders bone culture results unreliable, many patients will be treated unnecessarily.

Cognitive and cultural bias may affect physician decision-making in the diagnosis and treatment of osteomyelitis.^10,11 Physicians can be affected by cognitive bias regarding how they make diagnosis and plan treatments. To simplify decision-making, physicians use common “rules of thumb” or conventions that are often generalized beyond the scope of the available medical evidence, or they continue despite contradictory evidence. There are other potential limitations of the current study. This was a retrospective study and has inherent measurement and selection bias. The strength of this study was that the authors used histology as the reference standard to define osteomyelitis. Future studies should include larger cohorts from multiple centers. As practice varies by region and this was a single-center study, these results may not be generalizable to practices worldwide.

Administration of pre-biopsy antibiotics did not affect the proportion of positive cultures. Antibiotic exposure and the duration of antibiotic exposure did not change the bacterial yield. Based on these findings, the authors do not recommend withholding antibiotics prior to bone biopsy in patients with suspected DFO.