Levine Technique Is Inadequate for Bacterial Recovery in the Sampling of Wounds

Introduction: Levine technique is the current gold standard for swab-based wound sampling. Punch biopsies more accurately sample subsurface tissues, but many clinicians prefer Levine to alleviate concerns of invasiveness, delaying healing or causing pain. Alternatively, subsurface tissues can be assessed in a less invasive manner via curettage sampling, which scrapes regions of interest during curettage debridement. This study compared microbial recovery between Levine technique and fluorescence-guided curettage sampling. 

Methods: Fifteen wounds were imaged in real-time for bacterial fluorescence as part of a single-site clinical trial (#NCT03754426) and sampled in the wound centre with Levine technique. Clinicians deemed curettage sampling appropriate in 11/15 wounds; curettage sampling was targeted to regions positive for bacterial fluorescence on images (if present) during routine debridement. The remaining 4 wounds were not debrided (e.g. fresh surgical sites, almost healed wounds). The 26 samples collected were analyzed via semi-quantitative cultures. 

Results: Curettage samples were targeted to regions positive for bacterial fluorescence (red or cyan on images) in 8/11 wounds. Of those eight wounds, curettage cultures revealed higher bacterial loads (relative to Levine cultures) in 6/8 (75%) and higher numbers of bacterial species in 5/8 (63%). Three wounds which Levine technique deemed ‘light growth’ came back as ‘heavy growth’ on curettage-cultures and one wound negative on Levine cultures had ‘moderate growth’ from fluorescence-targeted curettage. In the three wounds negative for bacterial fluorescence all cultures confirmed, at most, only light bacterial growth. 

Discussion: These data suggest that Levine swabs routinely underestimate wound bacterial loads and under-report bacterial species present, causing the culture data to be unreliable. Bacterial fluorescence imaging enabled targeted sampling to regions highly suspect for bacterial presence while curettage sampling of these regions facilitated a surface and subsurface sample without causing any additional pain or damage to the wound.