Can a fluorescent imaging device (MolecuLight) function as an adjunct to surgical debridement during revision knee arthroplasty for periprosthetic infection in a tertiary centre: a feasibility study

Abstract

Objectives

Radical surgical debridement is central to the eradication of prosthetic joint infection. Surgeons are taught that the adequacy of debridement is critical to the success of revision procedures. We aimed to assess the feasibility of using a handheld fluorescent imaging device as an adjunct to tissue debridement in the management of periprosthetic joint infection (PJI) after standard radical debridement.

Design

Prospective feasibility trial.

Setting

This was performed at a tertiary prosthetic joint infection unit (The Royal Orthopaedic Hospital, Birmingham, UK), multidisciplinary team (MDT).

Participants

10 patients with established PJI managed through a tertiary prosthetic joint infection MDT were recruited between January 2023 and December 2023. All patients underwent standard management, including radical debridement.

Intervention

After completion of standard radical debridement, the device was used to identify any remaining areas of fluorescence. These areas underwent excision for further intraoperative microbiological and histological sampling to analyse if the represented areas had residual bacterial load.

Primary outcome

To assess whether the fluorescence imaging device is able to detect additional areas of fluorescence which led to positive tissue sampling.

Results

In all cases, the device was able to identify areas of fluorescence, indicating further debridement. In eight cases, an organism was isolated after additional microbiological sampling and culture. In all eight cases, the same isolate was identified with the standard sampling. Additional histological sampling was performed in eight cases and confirmed acute infection in three cases. The remaining five cases demonstrated histologically inflammatory tissue consistent with chronic infection.

Conclusion

This study is encouraging for the feasibility of using this device as an adjunct for debridement in the surgical management of PJIs; further evaluation is underway.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This is a small feasibility study, only recruiting 10 patients.

• The fluorescence device allows for ‘real-time’ fluorescence guidance of tissue debridement intraoperatively.

• The device is feasible for the assessment of the adequacy of debridement.

Introduction

Periprosthetic joint infection (PJI) is a devastating complication for patients undergoing knee arthroplasty and has a huge cost implication for treatment.¹ Appreciation of the pathophysiology of PJI is essential for understanding the clinical presentation and management. The European Bone and Joint Infection Society² and Musculoskeletal Infection Society (MSIS)³ have published guidance on the difficulties of diagnosing PJI and highlight the multimodal approach required. In particular, radical surgical debridement to remove infected tissue is likely to be the most important intervention for infection eradication; however, the adequacy of debridement cannot effectively be measured intraoperatively.⁴ Therefore, any adjunct that can help to identify infected tissue and aid in debridement is likely to have a significant impact on infection eradication.

MolecuLight is a novel point-of-care fluorescent imaging device used to help identify wounds with a high bacterial burden (>10⁴ colony forming unit/g).⁵ It detects endogenous fluorescence from commonly found wound bacteria such as Staphylococcus, Proteus, Klebsiella and Pseudomonas species without the use of contrast agents. High predictive values of these signals have been reported in locating regions of high bacterial load.^{6 7} Therefore, this may provide an adjunct that can help in the detection of areas of clinically significant bacterial load, which are not visible or missed at the time of surgery. If this device can identify such areas and permit more efficacious debridement, with the intention of reducing the risk of recurrent PJI, it could potentially be beneficial to the management of PJI.

In this study, we aimed to assess the feasibility of the use of a fluorescence device as an adjunct to a radical debridement in the setting of revision knee arthroplasty for PJI.

Methods

Design

There was no patient or public involvement in the design, conduct, reporting or dissemination plans of this research. The protocol and the patient information sheet are available as onlinesupplemental materials 1 2. We aimed to prospectively recruit 10 patients with proven chronic PJI of the knee undergoing revision surgery for eradication of infection. Recruitment began in January 2023 and ended in December 2023. All patients would be managed through a tertiary prosthetic joint infection unit (The Royal Orthopaedic Hospital, Birmingham, UK) multidisciplinary team (MDT), and the diagnosis of PJI was confirmed using the Musculoskeletal Infection Society criteria. All patients had to have been off antibiotics for at least 2 weeks prior to surgery. Identified patients were provided with a patient information sheet (included in online supplemental material 1) on the study and consented to participate in the study. Our only exclusion criteria were patients not able to provide informed consent for the study.

Primary outcome

The study aimed to assess if the fluorescence-guided device (MolecuLight) could detect areas of high fluorescence after radical debridement. If areas could be detected, then tissue sampling would be performed and sent for microbiological and histological analysis to evaluate if infection was truly present, allowing proof of the feasibility of using this device as an adjunct in debridement.

Standard care

Patients underwent standard care intraoperatively, which is to completely debride the effective joint space of all contaminated tissues, including bone, to healthy bleeding tissue and to send five representative microbiology samples from around the joint (one of which is synovial fluid in broth) and two representative tissue samples for histopathology, that is, tissue exposed to the effective joint space. All debridement was performed by one experienced periprosthetic joint infection revision surgeon. Systemic antibiotics were administered routinely at the time of anaesthesia, guided by sensitivities to preoperative microbial sampling and MDT recommendations from specialist infectious diseases and microbiology consultants.

Intervention

After completion of the radical debridement, the fluorescence-guided device (MolecuLight) was used to identify any ‘missed’ fluorescent areas that potentially indicated high bacterial load and therefore required further debridement. Fluorescence interpretation was subjective and not standardised, and did not involve any blinding. The additional samples taken with the guidance of fluorescent imaging were harvested using fresh instruments to avoid contamination, divided in half and sent to the microbiology laboratory for culture and the pathology laboratory for histological analysis, labelled separately from the routine tissue sampling as tissue for the study. Histological confirmation of infection was defined using the MSIS criteria³ (greater than five neutrophils per high-power field in five high-power fields observed from histologic analysis of periprosthetic tissue at ×400 magnification).

All tissue for microbiology was sent in Ballotini bottles (containing saline 0.9% NaCl with glass beads). All other aspects of care were delivered according to the local standard operating procedures, including preoperative diagnostics, MDT discussion and treatment planning, perioperative antibiotics (administered within 1 hour prior to skin incision) and 0.9% sodium chloride irrigation only prior to use of fluorescence-guided debridement.

Statistical interpretation

No statistical interpretation was planned due to the feasibility study design and small sample size.

Results

10 patients were prospectively recruited into the study; there were six male patients and four females. Ages ranged from 55 to 79 years of age (median age 65.5 years). All operations were performed in 2023. All cases were deemed to be chronic PJI (greater than 3 months as defined by the MSIS), rather than acute presentations of PJI. In all cases, the use of fluorescence-guided debridement identified additional fluorescent areas for sampling and debridement. If only one area fluoresced, this was preferentially sent for microbiology; if multiple areas were identified for sampling and further debridement, these were divided equally between microbiological and histological analysis.

Figure 1 demonstrates an intraoperative image of the tissues after completion of a radical debridement with no visible areas of concern or remaining synovium. Figure 2 demonstrates the same area under the fluorescent imaging delivered by the device. This demonstrates the areas that may contain high bacterial load and provides guidance for areas of tissue sampling and further debridement.

Figure 1. A photograph of the tissues after completion of radical debridement and washing.

Figure 2. The same photograph as seen in figure 1, but with the fluorescent areas highlighted by the device.

Figures3 4 demonstrate another postdebridement wound as seen to the naked eye and as seen with the fluorescent imaging.

Figure 3. Another photograph of the tissues after radical debridement and washing.

Figure 4. The same photograph as seen in figure 3, but with the fluorescent areas highlighted by the device.

In two cases, no isolate was identified on any of the routine microbiology sampling, nor extended sampling guided by the fluorescence device. Both of these patients were undergoing repeat first-stage procedures. The first of these two cases had undergone debridement and implant retention at another hospital for Escherichia coli PJI. In the second case, previous sampling had demonstrated Enterococcus faecalis, and the patient had already undergone a first-stage procedure at another hospital, but required soft tissue reconstruction; hence, a repeat first stage with gastrocnemius flap was performed. These previous interventions are likely to have contributed to the failure to culture any organisms in the routine and extended samples in these cases. The histology sampling demonstrated inflammation and reactive changes, consistent with chronic infection, but not achieving the criteria defined by the MSIS.

In all the remaining eight cases, the same organism was identified after additional fluorescent-guided tissue sampling. A wide range of organisms were identified with standard sampling, as shown in online supplemental table 1. This is representative of the typical range of causative organisms seen by the PJI MDT. Identification of the organism is an essential cornerstone for guiding management, especially with regard to choice and duration of antibiotic treatment.

Histological sampling was performed in eight cases, with two cases not having enough suitable material to send for histological analysis. Acute infection was confirmed histologically in three cases. The histology findings showed inflammatory changes in all cases. However, not all cases achieved the criteria required for an infection (greater than five neutrophils per high-power field in five high-power fields observed from histologic analysis of periprosthetic tissue at ×400 magnification), as defined by the MSIS criteria.³ This may be due to the chronic nature of these infections, as all cases were established PJIs and not acute infections. Therefore, lymphocytes and fibrosis were more likely to be seen than the neutrophils required for a histological diagnosis of infection.

Discussion

Periprosthetic infection is a devastating complication of joint replacement, and eradication of infection can be challenging. This study aimed to assess the feasibility of using fluorescence for detecting areas for further intraoperative debridement.

The Infectious Diseases Society of America has produced the Clinical Practice Guidelines on Diagnosis and Management of Prosthetic Joint Infection, which recommends 3–6 tissue biopsy samples and the presence of at least two positive culture specimens with an identical microorganism to confirm PJI.⁸ Radical debridement, synovectomy and irrigation should be performed in all devitalised and infected or potentially infected tissues intraoperatively to reduce the burden.⁹

However, knowing what to debride and how aggressively to debride can be challenging. Furthermore, quantifying the adequacy of debridement between surgeons to allow comparison has proved difficult.⁹ Infection is not visible to the naked eye, and therefore, intraoperative adjuncts to help guide this debridement would be potentially beneficial. This has therefore led to the guidelines by the European Society of Clinical Microbiology and Infectious Diseases¹⁰ which have strongly and urgently recommended research to improve the methods of detecting organisms in vivo.

Being able to detect areas for debridement using fluorescence has been studied in leg ulcer wounds¹¹ and in the setting of trauma, with early positive results.¹² Within the field of PJI, the use of this technology is in its infancy, and there are minimal studies on the subject.¹³ One approach has been to use a topical tracer, which can be applied to tissues and helps to identify biofilms¹⁴ or to help stain bacteria and aid in their debridement.¹⁵ Any adjunct to help identify and eradicate infection is of interest, and multiple methods, alongside fluorescence imaging, are being explored.¹⁶ At present, no studies have been published on the use of a handheld adjunct that does not require the application of tracers.

Our study uses a handheld device which did not require any topical application. The device was able to identify areas of fluorescence, which led to positive cultures for the same infecting organism in all cases, often with multiple positive results. However, the histological sampling was only positive in three cases and highlights the risks of false positives and false negatives in chronic PJI. While organism concordance does not equate to improved eradication, it does suggest that the positive results are not contaminants. Therefore, the device appears able to aid in identifying remaining areas of tissue with bacterial load left behind after the radical debridement has been performed.

This study is purely a feasibility study performed at a single centre without a control group. Due to the small sample size, lack of clinical outcome data and limitations of the methodological design, statistical analysis or greater interpretation of these results is impossible. Furthermore, due to the potential observer bias of fluorescence interpretation, the heterogeneity of this population and the organisms involved, its generalisability may be limited. However, the results are encouraging and do pave the way for further clinical studies on the utility of this device. Importantly, if it can be shown that identifying additional areas for debridement leads to improved rates of eradication and treatment of infection, this would be an excellent adjunct to pre-existing treatment.

Conclusion

This study was performed to assess the feasibility of using a fluorescent imaging device in the detection of areas of residual high bacterial load after standard debridement of infected revision knee arthroplasty. Patients were recruited from a tertiary bone infection unit and reflect the wide heterogeneity of hosts, pathogens and prostheses; however, it should be emphasised that this was a very small sample of 10 patients.

In all our cases, the device identified areas for further sampling and debridement, which subsequently confirmed infection in 8 of 10 cases. The microbiological findings were consistent with the standard sampling results, which reduces the likelihood of these being contaminants. Histology confirmed an acute infection in three cases, with evidence of inflammation seen in the other five sampled cases, suggestive of chronic infection.

This confirms the feasibility of this handheld device for intraoperative detection of fluorescence. Further investigation of whether fluorescence-guided debridement leads to improvements in clinical outcomes, particularly eradication of infection, is underway.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.