Use of topical vancomycin powder in total joint arthroplasty: Why the current literature is inconsistent?

Abstract

Periprosthetic joint infection (PJI) is a rare but terrible complication in hip and knee arthroplasty, and the use of topical vancomycin powder (VP) has been investigated as a tool to potentially reduce its incidence. However, there remains no consensus on its efficacy. Therefore, the aim of this review is to provide an overview on the application of topical vancomycin in orthopaedic surgery focusing on the recent evidence and results in total joint arthroplasty. Several systematic reviews and meta-analyses on topical VP in hip and knee arthroplasty have been recently published reporting sometimes conflicting results. Apart from all being limited by the quality of the included studies (mostly level III and IV), confounding variables are often included potentially leading to biased conclusions. If taken into consideration the exclusive use of VP in isolation, the available data, although very limited, suggest that it does not reduce the infection rate in routine primary hip and knee arthroplasty. Therefore, we still cannot advise for a routinary application. A properly powered randomized-controlled trial would be necessary to clarify the role of VP in hip and knee arthroplasty. Based on the analysis of the current evidence, the use of topical VP appears to be safe when used locally in terms of systemic adverse reactions, hence, if proven to be effective, it could bring great benefits due to its low cost and accessibility.

Core Tip: Vancomycin powder is widely used in orthopaedic surgery and it has been recently investigated in total joint arthroplasty (TJA), however, results are often conflicting. The aim of this study was to report on the use of vancomycin powder in orthopaedic surgery focusing on its application in TJA.

INTRODUCTION

Periprosthetic joint infection (PJI) is one of the leading causes of revision in total joint arthroplasty (TJA) and its incidence has been reported between 1% to 4% after primary total knee arthroplasty (TKA) and 1% to 2% after primary total hip arthroplasty (THA) [1,2]. According to the available projections, the number of revisions is expected to grow proportionally to the number of primary implants performed every year[3] showing and increase of revision for PJI by 176% between 2014 and 2030 in THA, and by 170% in TKA[4]. Economic-based studies have reported that the yearly cost associated with PJI in the United States was approximately one billion United States dollars in 2017, and projected to reach almost two billion United States dollars by 2030[5].

Multiple strategies have been pursued to try to reduce the PJI rate in TJA, including preoperative screening, patient optimization, modified intraoperative techniques, and enhanced postoperative surveillance[6]. Vancomycin is a widely adopted and effective antibiotic in orthopaedic surgery, and its topical application has been investigated in different fields including spine surgery, trauma, and sport medicine to reduce the incidence of infection by providing a high concentration of antibiotic in a specific surgical site. Therefore, it has also been studied to reduce the PJI rate in TJA, reporting however conflicting conclusions.

The aim of this review is to provide an overview on the applications of topical vancomycin in orthopaedic surgery focusing on the use in TJA summarizing the results reported in the literature in order to clarify the current evidence for the use of topical VP.

The United States National Library of Medicine (PubMed/MEDLINE), EMBASE, and the Cochrane Database of Systematic Reviews were queried for publications utilizing various combinations of the search terms “VP”, “vancomycin powder”, “orthopaedic surgery”, “orthopedic surgery” “arthroplasty”, in combination with the Boolean operators (AND, OR, *) since January 2020 to December 2022. Two authors (Fabio Mancino and Christopher W Jones) independently conducted all the searches and screened the titles and abstracts to identify relevant studies. Differences were resolved by consulting a third senior reviewer (Piers J Yates). Only abstracts that evaluated the outcomes of VP in orthopaedic surgery were reviewed. If the title and abstract of each study contained insufficient information, the full manuscript was reviewed. An additional search was conducted by screening the references list of each selected article. Inclusion criteria were any systematic review and/or meta-analysis that pooled the results on the application of VP in orthopaedic surgery and TJA, analyzing the outcomes in terms of infection rate. Exclusion criteria were cohort studies, clinical trials, case reports, surgical technique reports, expert opinions, letters to editors, biomechanical reports, instructional course lectures, studies on animals, cadaver or in vitro investigations, book chapters, abstracts from scientific meetings, unpublished reports, and studies written in a non-English language. Two independent reviewers (Fabio Mancino and Christopher W Jones) separately examined all the identified studies and extracted data. During the initial review of the data, the following information was collected for each study: Title, first author, year of publication, study design, number of studies included, number of patients included, type of surgery, methods of application of VP, complications related to VP, superficial and deep infection rates.

BURDEN OF PERIPROSTHETIC JOINT INFECTION

PJI is a relatively rare complication. However, it is associated with a significantly greater morality when compared with patients undergoing aseptic revisions, up to five times higher at one year[7,8,9]. In addition, after the first case of PJI, the reinfection rate is up to 8.5% in THA and up to 16% in TKA[9], showing that the long-term consequences can be devastating. Kapadia et al[10], reported that patients with PJI had a significantly higher number of readmissions (3.6 vs 1.2; P < 0.001), length of hospitalization, clinic visits and sum-total episode cost than patients who had a non-infected primary implant (US$96,166 vs US$21,654; P < 0.001). When considering the economic burden, the cost of a revision for PJI is up to five times higher than a primary TJA ($116,382 vs $28,249)[11]. Moreover, managing this complication often requires a two-stage revision strategy, costing approximately US$60,000 more than revisions for mechanical failure and/or aseptic loosening[12].

Currently, the only consensus recommendation for the use of antibiotics in TJA by international authorities is systemic perioperative administration[13].

VANCOMYCIN POWDER IN ORTHOPAEDIC SURGERY AND TJA

Gram-positive bacteria, particularly staphylococcal species, are the most common pathogenic organisms involved in post-operative orthopaedic infections[14]. Vancomycin is a tricyclic glycopeptide antibiotic with activity against gram-positive bacteria initially derived in 1953 from a compound produced by Amycolatopsis orientalis, a soil bacterium discovered within mud collected from a Borneo forest. The compound nicknamed “Mississippi mud” because of its appearance prior to purification became vancomycin (after the word “vanquish”) and nearly 70 years later still retains antimicrobial activity against the majority of gram-positive organisms and remains the most commonly used antibiotic in the United States for the treatment of serious gram-positive infections, including those caused by methicillin-resistant Staphylococcus aureus (MRSA)[15].

The topical application of this antibiotic has been widely adopted in different fields of orthopaedic surgery with promising results. Sweet et al[16], demonstrated a significant reduction in postoperative deep wound infection rates (0.2% vs 2.6%; P < 0.0001) in posterior instrumented thoracolumbar spinal fusions with the adjunctive application of 2 g of VP before wound closure. Similar findings were reported by O'Neill et al[17], when analyzing 110 patients that underwent posterior spinal stabilization of traumatic injuries. The authors noted that the infection rate was significantly reduced (13% vs 0%; P = 0.02) when 1 g of vancomycin was applied before wound closure. Moreover, similarly reduced infection rates were reported both by Molinari et al[18] and by Bakhsheshian et al[19] when studying the effect of topical VP in instrumented and uninstrumented spine surgery.

The use of VP has been also investigated in tibial fractures, considered to be at high risk of infection, in an open-label multicentre randomized clinical trial reporting that the application of 1 g of VP was associated with a reduced risk of deep surgical site infection due to gram-positive organisms (risk difference, -3.7%; 95%CI, -6.7% to -0.8%; P = 0.02), in line with the activity of the antibiotic[20].

In addition, when VP was used in 422 shoulder arthroplasty, it has been associated with a significant reduction in PJI with no increased rate of aseptic wound complications, however, literature on shoulder surgery is limited and results are mostly based on retrospective analysis[21].

Similarly, studies on the application of topical VP in foot and ankle surgery and in total ankle arthroplasty (TAA) are limited, however, the economic viability has been investigated by Nam et al[22]. At their institutional cost of UD$3.06 per gram and a TAA PJI rate of 3%, VP would be cost-effective for TAA revision costs with an absolute risk reduction of 0.02% (number needed to treat = 5304). In addition, the authors showed that VP, when considered at their institutional price, would remain cost-effective even if the initial PJI rate was as low as 0.05%, and that if the PJI rate was held constant at 4%, VP would remain cost effective even within a range of price from US$2.50 to US$100.00 per gram. Nevertheless, the power analysis performed by the authors to confirm such results in a clinical trial shows the main limit of the investigations on VP.

Moreover, topical vancomycin is frequently used in anterior cruciate ligament reconstruction (ACLR) by wrapping the graft in a swab saturated with 5mg/mL vancomycin solution[23] and it has been associated with reduced incidence of postoperative septic arthritis[24]. In fact, Xiao et al[25], reported in a survey on the ACL Study Group members that 37.9% of the members pre-soak their ACL graft in vancomycin prior to implantation. In addition, Naendrup et al[24], pooled the results on 5075 ACLR showing a significant reduction in septic arthritis with no differences in clinical outcomes, biomechanical tendon properties, or cartilage integrity. Despite having many clinicians concerns regarding the potential toxicity on chondroblasts and osteoblasts, it has been proven in-vitro that when used at concentrations up to 5mg/mL, the vancomycin levels reached within the first 24-hours remain below the toxicity threshold for chondroblasts and osteoblasts[26].

Recently, vancomycin application has also been investigated in intraosseous (IO) infusion in THA at the concentration of 500mg/100cc of normal saline showing increased local tissue and decreased systemic concentrations when compared with standard prophylactic intravenous (IV) administration[27]. Similar findings have also been reported in a high body mass index (BMI) population that underwent TKA showing local concentrations up to 9-times higher than systemic administration[28].

Considering these promising results, VP is used in TJA with the hope of significantly reducing the risk of PJI (Figure 1). Weight-based (15 mg/Kg) IV vancomycin is already widely adopted as a second-line prophylaxis instead of first- or second-generation cephalosporin in case of allergies to penicillin, history of MRSA, or positive preoperative MRSA nasal-swab colture[29]. However, considering the better results associated with cephalosporins, the International Consensus on PJI recommended that these antibiotics can be safely used in case of non-anaphylactic penicillin allergy[30] since the cross-reactivity risk has been proven to be as low as 1%[31].

Figure 1.

Vancomycin powder used in 1-st stage revision total knee arthroplasty.

Topical application of VP allows higher concentrations in the surgical area while minimizing the systemic adverse effects[32]. In a rat model, the use of intra-articular VP combined with IV antibiotics resulted in the complete eradication of MRSA bacteria from contaminated implants[33]. Johnson et al[32] studied vancomycin concentration both locally and systemically after the administration of 1 g of intra-articular VP and 1 g after closure of the fascia in the superficial tissues in 34 THA reporting the different serum levels at 90 min, 3 h, 12 h, and 24 h, and the local levels at 3 h, 12 h, and 24 h. The authors reported that the mean serum concentration peaked at 12 h (4.7 mcg/mL; max observed 12.7 mcg/mL at 3 h) while the systemic therapeutic levels of 15-20 mcg/mL were never reached in any of the time-points. In addition, the intra-wound half-life was estimated to be 7.2 h with mean wound levels > 900 mcg/mL at 3 h while maintaining the local concentration over 200 mcg/mL for 24 h. Finally, the authors estimated that it would take up to 64 h for intrawound levels to drop below the minimum inhibitory concentration for S. aureus of 2 mcg/mL (Table 1).

Table 1.

Serum and local levels of vancomycin at different post-wound closure collection times

Procedure	Serum levels after wound closure of VP intrawound administration (g/mL)
	1.5 h (mean ± SD; max)	3 h (mean ± SD; max)	12 h (mean ± SD; max)	24 h (mean ± SD; max)	Highest level observed across the 24-h period
THA (n = 15)	3.8 ± 3.9; 9.5	4.9 ± 4.5; 12.7	5.1 ± 3.3; 8.4	3.5 ± 3.5; 8.0	6.6 ± 3.8; 12.7
TKA (n = 19)	1.0 ± 2.5; 8.7	1.8 ± 3.2; 9.8	4.4 ± 3.1; 7.3	3.5 ± 3.6; 10.4	5.2 ± 3.4; 10.4
THA + TKA (n = 34)	2.2 ± 3.4; 9.5	3.2 ± 4.1; 12.7	4.7 ± 3.2; 8.4	3.5 ± 3.5; 10.4	5.8 ± 3.6; 12.7
	Local levels after wound closure of VP intrawound administration, n (g/mL)
	-	3 h (mean ± SD)	12 h (mean ± SD)	24 h (mean ± SD)	-
THA	-	988 ± 628 (12)	769 ± 1059 (11)	280 ± 436 (11)	-
TKA	-	877 ± 455 (18)	288 ± 203 (16)	163 ± 220 (18)	-
THA + TKA	-	922 ± 523 (30)	484 ± 716 (27)	207 ± 317 (29)	-

VP: Vancomycin powder; THA: Total hip arthroplasty; TKA: Total knee arthroplasty. Adapted from: Johnson JD, Nessler JM, Horazdovsky RD, Vang S, Thomas AJ, Marston SB. Serum and Wound Vancomycin Levels After Intrawound Administration in Primary Total Joint Arthroplasty. J Arthroplasty 2017 Mar; 32(3): 924-928. Copyright © 2015 Elsevier Inc. All rights reserved.

Despite the potential benefits, there are also theoretical drawbacks. Firstly, the low systemic concentration of vancomycin may induce the development of resistant species of gram-positive bacteria colonizing the body. The Infectious Disease Society of America recommended serum levels > 10 mcg/mL to avoid the potential development of resistance[34]. Given the short half-life of the antibiotic when administered parenterally (4-6 h), this is not problematic when administered as a single dose of prophylactic IV antibiotic providing coverage for the first 24 h, but maybe a factor during ongoing and prolonged systemic absorption of intra-articular antibiotics. Secondly, a potential third body wear mechanism has been hypothesized between crystalline vancomycin and implant components since the solubility of vancomycin may vary in an intra-articular environment compared to saline solution. Nevertheless, Qadir et al[35] reported no appreciable difference in wear rates after 10 million simulated cycles between ultra-high-molecular-weight-polyethylene and Cobalt-Chrome alloy with the addition of VP. Lastly, vancomycin may have negative effects on the proliferation of viable cells including osteoblasts. Braun et al[36], reported the in-vitro effect of vancomycin on osteoblasts, endothelial cells, fibroblasts and skeletal muscle cells showing that the toxic effects were time (from day-3) and concentration-dependent (> 0.01 mg/mL). However, such results are yet to be proven in-vivo, and as shown by Johnson, no such concentrations have been reported at the 3-d mark. Therefore, based on the aforementioned studies, topical administration of VP can reasonably be considered clinically safe when used in TJA. Finally, if proven to significantly impact the PJI rate, VP would be highly cost-effective as its price has been reported from $2.50 to the highest of $44.00 per gram[37].

CURRENT LITERATURE FOR VP IN TKA AND THA

Overall, seven systematic reviews and/or meta-analyses were identified and analyzed[38-44] (Table 2).

Table 2.

Main characteristics and results of the recent literature on the topic vancomycin powder

Ref.	Type of study	No. of studies	No. of cases (control/intervention)	PJI Rate/RR (control vs intervention)	SSI/Aseptic wound complications (control vs intervention)	Authors’ conclusions
Martin et al[36], 2022	Systematic review and meta-analysis	7/7	144724/8029	RR 0.39 (95%CI 0.27-0.56, P < 0.001)	6.48% vs 3.79%	VP ± PI lavage reduced PJI rate in primary and revision THA/TKA. Associated with reduced aseptic wound complications
Liao et al[35], 2022	Systematic review and meta-analysis	14	7720/1292	RR 0.41 (95%CI 0.29-0.58, P < 0.001)	-	VP recommended in primary TKA
Movassaghi et al[30], 2022	Systematic review and meta-analysis	16	3731/17164	1.65% vs 0.87% (P < 0.05)	-	Local VP may reduce the risk of PJI in primary and revision TJA
Wong et al[31], 2021	Systematic review	9	6255/3371	-	No difference	Recommend the surgeons not to use VP in routine THA and TKA
Peng et al[32], 2021	Systematic review and meta-analysis	9	4512/2354	RR 0.37 (95%CI 0.23- 0.60, P < 0.001)	RR = 0.40, 95%CI 0.27-0.61 (P < 0.001)	Local VP could significantly decrease the rate of SSI and PJI in primary TJA
Saidahmed et al[33], 2021	Systematic review and meta-analysis	9	3714/1985	3.5% vs 1.6%, RR 0.53 (95%CI 0.35-0.79, P = 0.002, I2 = 0.0%)	No difference 1.6% vs 0.7%, RR = 0.61, 95%CI 0.17-2.12, (P = 0.43, I2 = 0.0%)	Local antibiotic application results in a moderate reduction in deep infection rates in primary TJA, with no significant impact on SSI rate
Xu et al[34], 2020	Systematic review and meta-analysis	9	4607/2497	2.75% vs 1.20% (OR 0.44, 95%CI 0.28-0.69, I2 = 0.0%)	No difference 1.60% vs 0.67% (OR 0.60, 95%CI 0.17-2.12, I2 = 0.0%)	VP used in primary hip and knee arthroplasty may reduce the incidence of PJI but it may increase the risk of aseptic wound complications

RR: Relative risk; SSI: Superficial site infection; VP: Vancomycin powder; PI: Povidone iodine; PJI: Periprosthetic joint infection; THA: Total hip arthroplasty; TKA: Total knee arthroplasty; TJA: Total joint arthroplasty; OR: Odds ratio.

Movassaghi et al[38], reported that intrawound VP may reduce the risk of PJI in primary and revision TJA while not leading to systemic complications. The authors included in their analysis 16 studies and 17164 TJA that received intrawound VP reporting an overall decreased rate of PJI (OR 0.46, P < 0.05), a decreased rate when considering TKA and THA separately (OR 0.41, P < 0.05 and OR 0.45, P < 0.05, respectively), and a decreased rate when considering primary implants only (OR 0.44, P < 0.05). Most of their results came from the outcomes of 14262 primary TKA (of 17164 joints, 83%) and that among them, 9884 cases (69% of primary knees) came from a study[44] where the so-called “VIP protocol” was used by mixing VP and 0.35% povidone-iodine (PI) solution (17.5 mL in 500 mL saline).

Regarding PI lavage, Kim et al[46], reported in a systematic review on 7 studies and 8861 TJA no difference between PI and saline in reducing the PJI rate. However, more recent studies showed efficacy in revision TJA reducing the PJI rate from 3.4% to 0.4% (P = 0.038, 478 revisions)[47], and efficacy in reducing the rate of any infection over 3232 TJA (OR 0.45, P < 0.05) or superficial site infections (SSI, OR 0.3, P < 0.05)[47]. Finally, Shohat et al[49] recently reported on the outcomes of 31331 cases showing a 2.34 times lower rate of PJI when comparing PI lavage with saline in TJA (0.6% vs 1.3%) with an absolute risk reduction of 0.73% and a number needed to treat of 137 patients. Therefore, the positive outcomes reported by Movassaghi et al[38] may have been influenced by the inclusion of iodine lavage.

Similarly, Liao et al[43], published in strong favor of VP suggesting that VP has a clear effect on preventing PJI in primary TKA. The authors reported on 11292 TKA where VP was used with a Risk Ratio (RR) of 0.41 (95%CI 0.29 to 0.58, P < 0.001) when compared to cases where VP was not used. However, as previously mentioned, 46.7% of of the cases analyzed came from studies[45,50] where VP was used in combination with a PI solution, potentially having once again a significant effect on the final results.

Moreover, Peng et al[40], stated that “the local application of VP could significantly decrease the rate of SSI and PJI in primary TJA” recommending its topical administration before wound closure. The meta-analysis included nine studies and three of those[49,51,52], representing a weight on the result of 44%, did not involve only the application of topical VP, therefore, their inclusion could be misleading. One of these[50], reported on the combined application of PI lavage and VP showing that administration of local antibiotics was preventative for PJI only in the primary TKA (OR 0.28, 95%CI 0.09–0.89). The other two[51,53], reported on the application of VP on the surface of cementless implants in THA and TKA and not in the soft tissue deep or superficial to the fascia/capsule, therefore, a completely different way of using VP.

Xu et al[42], reported that “the current literature suggests that intrawound vancomycin used in primary hip and knee arthroplasty may reduce the incidence of PJI, but it may also increase risk of aseptic wound complications”. Nine studies were included in their final analysis with 4605 TJA, 2497 of which were treated with VP. The authors reported a reduced PJI rate in the VP group (1.20% vs 2.75%) with an OR of 0.44 (95%CI 0.28 to 0.69, I² = 0.0%), a comparable risk of SSI (OR 0.60, 95%CI 0.17 to 2.12), and a higher incidence of aseptic wound complications (2.15% vs 0.96%, OR 2.39, 95%CI 1.09 to 5.23). However, when considering the aseptic wound complications, only four of the nine studies reported on such events (1069 treated cases), and all of them had different methodology protocols in terms of the amount of VP used, placement of the VP (deep to the fascia, superficial, or both), and the application of a drain for up to 48 h post-operative. Therefore, the conclusion that VP is associated with an increased risk of aseptic wound complications, based on such results, may require stronger evidence.

Saidahmed et al[41], stated that topical antibiotics led to a moderate reduction in PJI in primary TJA, with no significant impact on SSI rates but that it may be associated with a moderate increase in aseptic wound complication. However, once again, four of the nine studies reported mixed results considering the combined activity of PI lavage and VP[50], the application on cementless implants[52,53], or did not consider only the application of VP in TJA but more generally the use of topical antibiotics[54].

On the other hand, Wong et al[39] discouraged the application of VP in primary TJA after systematically analysing the outcomes of 9 studies and 3371 TJA in which VP was used compared with 2884 in which it was not. Only studies with similar procedures and those limited to the application of VP were included. The authors reported that only one of the studies included[51] was associated with significant improvement while the remaining eight had OR that broadly bracketed the line of no difference (range, 0.09 to 1.97). In addition, the authors noted insufficient evidence on the question of safety, therefore, their final statement was against the use of topical VP in routine THA and TKA unless adequately powered, multicentre, prospective trials demonstrate clear evidence. However, despite the methodology and the inclusion criteria being well defined to include only studies using topical VP in isolation, no statistical analysis was performed to verify the results.

Lastly, Martin et al[44] recently pooled together the studies using VP alone (7 studies) and in combination with PI lavage (7 studies) reporting a significant reduction of PJI rate (RR 0.39, 95%CI 0.27 to 0.56, P < 0.001) in primary and revision THA and TKA when compared with a control group. However, there remain doubts on the contribution of the PI lavage as we are still missing clear results on the VP alone used with standardized methods and compared with a control group. Interestingly, the authors reported a reduced aseptic wound complication rate in the treatment pool (110/2903, 3.79% vs 98/1512, 6.48%), though, still considering the combined effect of VP and PI lavage.

CONCLUSION

PJI in TJA is certainly one of the biggest challenges that the orthopaedic community is now facing with tremendous impact on the patient, the treating multi-disciplinary team, and the health care system. Despite the topical application of VP appears to be safe in terms of systemic complications, there are potential risks regarding the development of antimicrobial resistance following the administration of VP and most importantly, from the available data, we cannot conclude that when used in isolation it is effective in reducing the PJI rate. Evidence remains lacking with varying methodologies and important technical differences (amount of VP, placement deep or superficial to the fascia, use of drain). In fact, positive outcomes appear only to have been reported when the additional application of PI is considered together with VP. It must also be noted that the use of intraoperative antimicrobial irrigation (e.g. deep or subcutaneous tissues), or the application of antimicrobial agents (e.g. ointments, solutions, or powders) to the surgical incision for the prevention of SSI are not currently recommended by The Centers for Disease Control and Prevention Guideline for the Prevention of Surgical Site Infection[55]. Moreover, evidence supports the perceived increased risk of aseptic wound complications, which should be further investigated.

Therefore, despite the multiple studies recently published, the efficacy of VP in TJA for reducing PJI is still essentially unknown. To overcome this issue, a randomized controlled trial with homogeneous methodology and exclusion of additional confounding variables (such as PI lavage) would be necessary.

ACKNOWLEDGEMENTS

The Orthopaedic Research Foundation of Western Australia (ORFWA) for providing research support.