WCJC #10 - Bone Biopsy and DFU
Admission Time Deep Swab Specimens Compared With Surgical Bone Sampling in Hospitalized Individuals With Diabetic Foot Osteomyelitis and Soft Tissue Infection
Ana Belen Manas, MD, Surabhi Taori, FRCPath PhD, Raju Ahluwalia, BSc
FRCS, Hani Slim, MD MRCS, C. Manu, BSc MRCP, Hisham Rashid, MSc
FRCS, Venu Kavarthapu, FRCS, Michael Edmonds, MD FRCP, and
Prashanth R. J. Vas, MRCP PhD
The International Journal of Lower Extremity Wounds, May 5, 2020 https://doi.org/10.1177%2F1534734620916386
I selected this article for review because it grapples with question of how does a bone biopsy compare to a deep wound swab?
This study was a retrospective review of consecutive diabetic foot patients who were admitted to King’s College Hospital in London, UK over the calendar year 2014. All the patients included in the study were admitted with infection of a previously known diabetic foot ulcer. They all had a clinical diagnosis of diabetic foot osteomyelitis based on a combination of probe to bone test and typical radiological abnormalities.
On admission, a deep wound swab was performed on each patient after cleaning the ulcer with sterile gauze soaked in sterile saline and after removal of superficial slough by sharp debridement. When possible, the culture swab was rotated over the bone. Surgical bone specimens were obtained under sterile condition after the patient received either regional or general anesthesia and after the debridement of all visibly infected tissue.
There was a total of 63 patients who had clinical and radiographic evidence of diabetic foot osteomyelitis. Of the 63, 80% had type two diabetes. Chronic diabetic foot ulceration, defined as present for more than eight weeks, was present in 68% of the patients. As well, 70% of the patients had received prior antibiotic therapy in the previous 30 days, of which 57% had received antibiotics for more than 30 days. The common sites of ulceration were over the head of the metatarsals at 36%, the first metatarsophalangeal joint at 18%, the toes in 12%, and the remaining 34% at either the midfoot or hindfoot. All ulcerations tracked to the bone.
For the deep wound swabs, 11% did not have any pathogen isolated. The mean number of isolates per patient was 1.21. Gram-positive were found in 49% of isolates, gram negatives were found in 60% of isolates, anaerobes were found in 4% of isolates. Among the gram-positive isolates, S aureus was found 50% of the time. Among the gram-negatives, prevalent was E. coli but it only 21%. Up to 75% of all the specimens isolated some kind of gram-negative organisms.
For the surgical bone specimens, 13% had no growth. The overall mean number of isolates was 1.47. Gram-positive organisms were noted in 52% of patients and gram-negative organisms were found in 60% of patients. Proportions were similar to those of the deep wound swabs except that vancomycin resistant enterococcus was found in 6% of the surgical bone specimen patients whereas none were seen in the deep wound swab specimens.
Isolate were strictly identical in only 16% of patients. At least one isolate was identical in both techniques in only 40% of patients. The overall value for concordance where at least one pathogen was similar was fair at 0.302. The concordance for the presence of any gram-negative isolate was moderate with = 0.417. The best concordance was observed for S. aureus with = 0.571 and specifically MRSA with = 0.644. The presence of anaerobes was poorly concordance at = 0.022.
Previous antibiotic therapy (p = .03) and ulcer duration of less than 8 weeks (p = .025) were found to be a risk factor for negative growth in the bone specimens.
The authors concluded that deep wound swab and surgical bone specimen show a poor concordance, however the swabs were more likely to concur for S. aureus or MRSA. They concluded that contrary to their expectations, the results highlighted that the pathogens involved in bone infection may not necessarily be represented on the surface where the milieu constantly evolves.
The purpose of the study was to compare the results of the two methods of obtaining culture, not to determine which method provides better information for determining antibiotic therapy. However, it does seem reasonable to assume that the bone biopsy culture is a more accurate representation of the actual infecting organism for the osteomyelitis. With that in mind, I think the results support that if S. aureus, and more specifically MRSA, are found on a deep wound swab of a diabetic foot ulcer with clinical and radiographic osteomyelitis, that the results can be relied upon to determine antibiotic therapy in most cases.
With other organisms, if an initial bone biopsy has not been obtained for culture, the clinician needs to be concerned that the isolated organism may not be the infecting organism.
Since the study did not provide outcome data and was not designed to compare making treatment decisions based on deep wound swab versus surgical bone specimen, it is not able to answer the question of if all patients with suspected diabetic foot osteomyelitis require surgical bone specimen. Not withstanding that, I think it would be reasonable to attempt to obtain a bone biopsy for culture in all patients with diabetic foot osteomyelitis and certainly should be considered in patients that fail initial antibiotic therapy regardless of the method of obtaining the original culture.