Abstract
Aims
The efficacy and safety of intrawound vancomycin for preventing surgical site infection in primary hip and knee arthroplasty is uncertain.
Methods
A systematic review of the literature was conducted, indexed from inception to March 2020 in PubMed, Web of Science, Cochrane Library, Embase, and Google Scholar databases. All studies evaluating the efficacy and/or safety of intrawound vancomycin in patients who underwent primary hip and knee arthroplasty were included. Incidence of periprosthetic joint infection (PJI), superficial infection, aseptic wound complications, acute kidney injury, anaphylactic reaction, and ototoxicity were meta-analyzed. Results were reported as odds ratios (ORs) and 95% confidence intervals (CIs). The quality of included studies was assessed using the risk of bias in non-randomized studies of interventions (ROBINS-I) assessment tool.
Results
Nine studies involving 4,607 patients were included. Intrawound vancomycin was associated with lower incidence of PJI (30 patients (1.20%) vs 58 control patients (2.75%); OR 0.44, 95% CI 0.28 to 0.69) and simultaneous acute kidney injury (four patients (0.28%) vs four control patients (0.35%), OR 0.71, 95% CI 0.19 to 2.55). However, it did not reduce risk of superficial infection (four patients (0.67%) vs six control patients (1.60%), OR 0.60, 95% CI 0.17 to 2.12) and was associated with higher incidence of aseptic wound complications (23 patients (2.15%) vs eight in control patients (0.96%), OR 2.39, 95% CI 1.09 to 5.23). Four studies reported no anaphylactic reactions and three studies reported no ototoxicity in any patient group.
Conclusion
The current literature suggests that intrawound vancomycin used in primary hip and knee arthroplasty may reduce incidence of PJI, but it may also increase risk of aseptic wound complications.
Cite this article: Bone Joint Res 2020;9(11):778–788.
Article focus
-
Can intrawound vancomycin reduce risk of periprosthetic joint infection (PJI) in patients after primary hip and knee arthroplasty?
-
Can intrawound vancomycin reduce risk of superficial infection?
-
Does intrawound vancomycin increase the risk of aseptic wound complications and other adverse events?
Key messages
-
Intrawound vancomycin used in primary hip and knee arthroplasty may reduce incidence of PJI.
-
Intrawound vancomycin did not reduce the risk of superficial infection.
-
Intrawound vancomycin may increase risk of aseptic wound complications.
Strengths and limitations
-
This is the first systematic review and meta-analysis assessing both the efficacy and safety of intrawound vancomycin treatment in primary hip and knee arthroplasty.
-
The primary studies included in our analyses are of poor quality, which may affect our results.
-
The included studies varied in recruitment period, follow-up time, dosage, sites of application, and vancomycin formulation.
Introduction
Surgical site infection (SSI), including superficial infection and periprosthetic joint infection (PJI), is a catastrophic complication after total hip arthroplasty (THA), total knee arthroplasty (TKA), and unicompartmental knee arthroplasty (UKA). It is a significant challenge for patients, surgeons, and healthcare providers. It not only delays functional rehabilitation, but also prolongs postoperative length of stay to facilitate intravenous antibiotics and is associated with increased morbidity and mortality.1,2 National data from both the UK and the USA showed that the incidence of PJI varies from 0.4% to 2% after THA3 or TKA.4 An estimated 40,000 to 80,000 patients each year will suffer from PJI following hip and knee arthroplasty by 2030 in the USA.5
Administration of intravenous antibiotics, typically cephalosporins such as cefazolin and cefuroxime,6 is initiated before a skin incision is made in order to significantly reduce risk of PJI.7 However, as many as 60% of wound infection isolates cultured from joint arthroplasties in the USA are resistant to cephalosporins.8-10 This includes the two most common bacteria causing deep infection, methicillin-resistant Staphylococcus aureus (MRSA) and coagulase-negative Staphylococcus. Hence, local use of antibiotics that are effective against these bacteria, for example vancomycin, may be beneficial.
Vancomycin kills gram-positive bacteria by inhibiting the production of phospholipids and peptides in cell walls at the trans-glycosylation stage.11 A meta-analysis including 27 studies of 17,321 patients showed that intrawound application of vancomycin can reduce SSI in various spinal surgeries.12 The available studies for primary hip and knee arthroplasty have given inconsistent results.13,14 Although local use of antibiotics is considered safer than intravenous application, local and systematic adverse reactions, such as aseptic wound complications, anaphylactic reaction, ototoxicity, wound healing, and kidney function, must be considered. In addition, the International Consensus on Orthopaedic Infections offers only a limited recommendation for intrawound administration of vancomycin during arthroplasty procedures.15
This systematic review and meta-analysis was conducted to assess the safety and efficacy of intrawound vancomycin in reducing risk of PJI and superficial infection in patients after primary hip and knee arthroplasty.
Methods
This systematic review and meta-analysis followed the methodology proposed by the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (Supplementary Table i). It also had been registered at International Prospective Register of Systematic Reviews (CRD42020175679).
Search strategy and eligibility criteria
We conducted a comprehensive literature search across the electronic databases of PubMed, Web of Science, Cochrane Library, Embase, and Google Scholar, from their inception to March 2020. Three groups of keywords or medical subject headings were used in our queries: ‘vancomycin’ OR ‘vancomycin powder’ OR ‘antibiotic’ AND ‘local’ OR ‘topical’ OR ‘intrawound’ OR ‘intra-wound’ AND ‘arthroplasty’ OR ‘knee arthroplasty’ OR ‘hip arthroplasty’ OR ‘arthroplasty’ OR ‘knee arthroplasty’ OR ‘hip arthroplasty’. The vocabulary and keyword combinations were adjusted for each database. Two reviewers (HX and JY) manually and independently reviewed all the retrieved literature, and any disagreements were resolved by discussion with a third reviewer (JX).
To be eligible for inclusion, studies must have assessed the efficacy and safety outcomes from intrawound administration of vancomycin after primary hip and knee arthroplasty to prevent PJI and superficial infection. Studies were excluded if they did not include controls (who underwent surgery but did not receive local vancomycin), case reports, commentaries, expert opinions, or reviews, or if animals were used instead of humans. Furthermore, all references cited in these included studies and relevant review articles were screened manually. No language restrictions were applied during the literature search.
Data extraction
Two reviewers independently extracted relevant data from the included studies using a standardized form from March 2020. The following data were collected from each study: 1) the characteristics of each included study, including study design, location and name of study site, type of surgery, the dose and site of vancomycin administration, drain usage, details of intravenous antibiotics, and the definitions of PJI, superficial infection, aseptic wound complications, and acute kidney injury; and 2) the characteristics of populations involved in each included study, including their number, sex, age, and body mass index (BMI), recruitment period and follow-up time, and comorbidities such as diabetes, hypertension, chronic obstructive pulmonary disease or pulmonary disease, and chronic kidney disease. Outcomes included the incidence of PJI, superficial infection, aseptic wound complications, acute kidney injury, anaphylactic reaction, and ototoxicity.
Quality assessment
The quality of the included studies was assessed using the risk of bias in non-randomized studies of interventions (ROBINS-I) assessment tool. This tool comprises seven domains: the first and second domains address ‘baseline’ issues, such as confounding factors and selection of participants, which involve comparisons before the start of the intervention. The third domain addresses classification of the interventions, and the other four domains address biases arising after the start of the intervention, such as deviations from intended interventions, missing data, measurement of outcomes, and selection of the reported results. The risk of bias within each domain is determined based on their 'signalling questions'. The overall bias risk of each study is classified as ‘low’, ‘moderate’, ‘serious’, ‘critical’, or ‘no information’, which is determined according to the results for the seven domains.16 In our study, individual domains and overall risk of bias assessment was performed independently by two reviewers for each included study. Any discrepancies were resolved by discussion with a third reviewer.
Statistical analysis
All statistical analyses were performed using Stata 11 (StataCorp, College Station, Texas, USA). Heterogeneity across studies was assessed using the inconsistency index (I2). Briefly, this analysis estimates the percentage of the variability in results from multiple studies that is likely due to true differences in outcomes, study design, patients, and tests, rather than to sampling error.17 Heterogeneity was considered ‘low’ if I2 was 25%, ‘moderate’ if I2 was 50%, or ‘high’ if I2 was 75%.18 All meta-analyses were performed using a fixed-effect model because I2 was less than 25% for all outcomes assessed.19
Pooled results are presented as odds ratios (ORs) and 95% confidence intervals (CIs) where appropriate. Publication bias was assessed using a funnel plot20 and Egger’s test.21 Significant publication bias was considered if the associated p-value was less than 0.05.
Results
Search results
Our systematic search retrieved 721 literature records, of which 516 remained after removing duplicates. Studies were first assessed based on their title and abstract, which led to 12 studies that were then read in full. We excluded three studies for the following reasons: in one study, both the intravenous and local antibiotic regimens among patients were inconsistent,22 one lacked a control group,23 and one lacked specific data of infection patients.24 Finally, nine studies were included in the meta-analysis (Figure 1). Two studies25,26 examined patients who underwent primary or revision THA and TKA, but we excluded the participants who underwent revision procedures due to increased risk of infection.13
Fig. 1
Flow diagram of literature searching and study inclusion. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.
Characteristics of included studies and participants
All included studies were retrospective with the exception of one prospective study (Table I).27 A total of 4,607 patients were included in our analyses, comprising 2,497 patients treated with intrawound vancomycin and 2,110 controls (Table II). All included studies reported the definition of PJI: four13,28-30 of them determined the PJI based on the International Consensus Meeting (2013);31 two,25,26 based on the Musculoskeletal Infection Society (2011) criteria;32 one, based on treatment method;33 one, based on articular fluid culture;27 and one,14 based on Musculoskeletal Infection Society (2011) criteria and culture. Only one study30 included patients who underwent UKA (n = 26). While a majority of studies used vancomycin alone in soft tissue, some studies employed other methods: one study included surgical intervention that coated 1 g of vancomycin onto the acetabular and femoral components immediately before implantation;29 and in another study, 1 g to 2 g of vancomycin powder was combined with 2 g of an absorbable calcium hydroxyapatite, and then the mixture was spread in a thin layer on the articular surface of the implants before implantation.27
Table I.
Characteristics of studies included in the meta-analysis.
| Study | Location | Study design | Surgery | Intervention | Definition of PJI | Definition of superficial infection | Definition of acute kidney injury | Drain use | Intravenous antibiotic |
|---|---|---|---|---|---|---|---|---|---|
| Yavuz et al13 2019 | Univ. of Health Science, Ankara Numune Training and Research Hospital, Ankara, Turkey | R | Primary TKA | 2 g VP / joint capsule | ICM (2013)34 | NR | NR | 24 hrs post-op | 2 g cefazolin / 30 mins before incision |
| Hanada et al28 2019 | Hamamatsu University School of Medicine, Hamamatsu, Japan | R | Primary TKA or UKA | 1 g VP / intracapsular | ICM (2013) | NR | NR | 1 or 2 days post-op | 2 g cefazolin before incision, 1 g once every 6 hrs on the day of surgery |
| Cohen et al29 2019 | Rhode Island Hospital, Providence, Rhode Island, USA | R | Primary THA | 1 g VP / coat the acetabular and femoral components | ICM (2013) | NR | NR | No drain | 1 dose pre-op and 2 doses post-op cefazolin (weight-based) |
| Dial et al30 2018 | Duke University Medical Center, Durham, North Carolina, USA | R | Primary THA | 1 g VP / intracapsular and extracapsular | ICM (2013) | Surgical site infections that resolved with oral antibiotics | An increase in serum creatinine greater than 0.3 mg/dl | NR | Cefazolin (2 g) started prior to skin incision and 1 dose 24 hrs post-op |
| Patel et al14 2018 | Emory University School of Medicine, Atlanta, Georgia, USA | R | Primary THA and TKA | 1 g VP / joint, muscle, fascia, subcutaneous tissues | MSIS or single positive culture | Culture-positive of superficial wound | Increase of > 0.3 mg/dl in serum creatinine post-op | NR | 1 dose cephalosporin/ within 1 hr before incision |
| Winkler et al25 2018 | Texas Tech University Health Sciences Centre, Lubbock, Texas, USA | R | Primary THA and TKA | 2 g VP / joint capsule | MSIS (2011)32 | NR | Identified by blood urea, nitrogen, and creatinine values | Dressed with an incisional wound vacuum | Cefazolin/1 dose pre-op followed by 3 doses post-op |
| Khatri et al33 2017 | Multicentre, India | R | Primary TKA | 1 g VP / subfacial layer | Infection treated by debridement, intravenous antibiotics, even two-stage revision | Infection managed with oral antibiotics | NR | No drain | 1.5 g cefuroxime/ within 1 hr of skin incision and repeated every 12 hrs until drain removal |
| Otte et al26 2017 | Mount Carmel Health Systems, Columbus, Ohio, USA |
R | Primary knee and hip arthroplasty | 1 g VP / intra-articular joint space | MSIS (2011) | NR | NR | NR | Standard systemic prophylaxis |
| Assor27 2010 | University of Marseille, Marseille, France | P | Primary TKA | 1 g to 2 g VP mixed with 2 g hydroxyapatite / articular surface of the implants | Confirmed by articular fluid culture | NR | NR | 3 or 5 days post-op | 0.5 g cefazolin / used during the first 48 hrs post-op (2 g/day) |
-
ICM, International Consensus Meeting; MSIS, Musculoskeletal Infection Society; NR, not reported; P, prospective; PJI, periprosthetic joint infection; post-op, postoperatively; pre-op, preoperatively; R, retrospective; SSI, surgical site infection; THA, total hip arthroplasty; TKA, total knee arthroplasty; UKA, unicompartmental knee arthroplasty; VP, vancomycin.
Table II.
Characteristics of patients from included studies.
| Study | Group | No. of patients (male/female) |
Mean age, yrs (SD) | Mean BMI , kg/m2 (SD) | Recruitment period | Follow-up time | Diabetes n (%) |
Hypertension n (%) | COPD or pulmonary disease, n (%) | CAD or heart disease, n (%) | Chronic kidney disease, n (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Yavuz et al13 2019 | C: | 502 (154/348) | 63.4 (12.1) | 28.9 (5.4) | 2012 to February 2015 | mean: 53.2 mths (24 to 84) for both groups | 129 (25.6) | 296 (58.9) | 51 (10.2) | 54 (10.8) | 39 (7.8) |
| V: | 474 (148/326) | 65.5 (10.7) | 29.0 (5.7) | February 2015 to December 2016 | 102 (21.5) | 291 (61.4) | 53 (11.2) | 41 (8.7) | 27 (5.7) | ||
| All: | 976 | ||||||||||
| Hanada et al28 2019 | C: | 92 (22/70) | 73.3 | 25.7 | 2010 to 2014 | Not less than 1 yr for both groups | 17 (18.5) | NR | 4 (4.3) | 12 (13.0) | 6 (6.5) |
| V: | 110 (27/83) | 74.6 | 26.7 | 2014 to 2017 | 22 (20) | NR | 4 (3.6) | 16 (14.5) | 4 (3.6) | ||
| All: | 202 | ||||||||||
| Cohen et al29 2019 | C: | 246 (109/137) | 67.3 ( 12.6) | 29.2 (5.6) | April 2015 to December 2016 | NR | NR | NR | NR | NR | NR |
| V: | 309 (149/160) | 66.0 ( 10.2) | 29.6 (5.8) | April 2015 to December 2016 | NR | NR | NR | NR | NR | NR | |
| All: | 555 | ||||||||||
| Dial et al30 2018 | C: | 128 (64/64) | 61.5 (10.5) | 29.8 (5.8) | June 2013 to January 2015 | mean: 14.4 mths (SD 7.5) | 19 (14.8) | 83 (64.8) | 11 (8.6) | 18 (14.1) | 10 (7.8) |
| V: | 137 (65/72) | 61.2 (11.1) | 30.0 (6.0) | January 2015 to February 2016 | mean: 8.2 mths (SD 4.5) | 16 (11.7) | 83 (60.6) | 4 (2.9) | 23 (16) | 15 (10.9) | |
| All: | 265 | ||||||||||
| Patel et al14 2018 | C: | 112 (48/64) | 64.8 | 31.1 | April 2016 to September 2016 | mean: 11.3 mths (3.0 to 25.6) | 14 (12.5) | NR | NR | 12 (9.8) | 2 (1.8) |
| V: | 348 (138/210) | 63.6 | 30.6 | October 2016 to October 2017 | mean: 7.7 mths (3.2 to 19.1) | 34 (9.8) | NR | NR | 22 (6.3) | 13 (3.7) | |
| All: | 460 | ||||||||||
| Winkler et al25 2018 | C: TKA: THA: | 152 (48/104) 97 (47/50) |
27 (17.8) 63.5 (17.7) |
NR | 1 January 2012 to 31 December 2013 | At least 6 mths for both groups | 2 (1.3) | NR | 14 (9.2) | 40 (26.3) | NR |
| V: TKA: THA: | 191 (70/121) 133 (61/72) |
80 (41.9) 65 (11.9) |
NR | 1 January 2014 to 31 December 2015 | 11 (5.8) | NR | 24 (12.6) | 50 (26.2) | NR | ||
| All: | 573 | ||||||||||
| Khatri et al33 2017 | C: | 64 (44/20) | NR | NR | February 2014 to January 2016 for two groups | 6 mths for both groups | 23 (35.9) | 25 (39.1) | NR | NR | NR |
| V: | 51 (32/19) | 17 (33.3) | 22 (43.1) | NR | NR | NR | |||||
| All: | 115 | ||||||||||
| Otte et al26 2017 | C: TKA: THA: |
392 252 |
NR | NR | 1 May 2012 to 30 April 2013 | 3 mths for both groups | NR | NR | NR | NR | NR |
| V: TKA: THA: |
400 282 |
NR | NR | 1 May 2013 to 30 April 2014 | NR | NR | NR | NR | NR | ||
| All: | 1,326 | ||||||||||
| Assor27 2010 | C: | 73 (17/56) | 72 (7.6) | NR | 2002 to 2006 for both groups | mean: 5 yrs (3 to 7) for both groups | NR | NR | NR | NR | NR |
| V: | 62 (16/46) | 73 (8.2) | NR | NR | NR | NR | NR | NR | |||
| All: | 135 |
-
BMI, body mass index; C, control group; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; NR, not reported; THA, total hip arthroplasty; TKA, total knee arthroplasty; V, vancomycin group.
Table I presents further study characteristics, including the dose of vancomycin intervention, drain use, details of intravenous antibiotics, and definitions of PJI, superficial infection, and acute kidney injury. The recruitment period differed for patients treated with intrawound vancomycin and controls in all but two studies.27,33 Follow-up time was different between the two groups in two studies,14,30 and one study25 did not report follow-up time. Table II describes patient characteristics including sex, age, BMI, and comorbidities.
Quality assessment and publication bias of included studies
In general, the quality of included studies was unsatisfactory (Table III). Three studies29,33 were considered to have ‘moderate’ risk of bias and another six studies13,14,25,26,28,30 were at ‘serious’ risk of bias. We believe this is due to the retrospective nature and substantial confounding bias. In contrast, according to the funnel plot analysis and the result of Egger’s test (t = -1.550; p = 0.165), we detected no observable publication bias across all included studies (Figure 2).
Fig. 2
Funnel plot of included studies. The dotted lines indicate pseudo 95% confidence limits.
Table III.
Quality assessment of included studies
| Study | Pre-intervention | At intervention | Post-intervention | Total | ||||
|---|---|---|---|---|---|---|---|---|
| Bias due to confounding |
Bias in selection of participants |
Bias in classification of interventions | Bias due to deviations from intended interventions | Bias due to missing data |
Bias in measurement of outcomes | Bias in selection of reported result | Bias across domains |
|
| Yavuz et al13 2019 | Serious | Moderate | Low | Low | Moderate | Moderate | Moderate | Serious |
| Hanada et al28 2019 | Serious | Moderate | Low | Moderate | Low | Moderate | Moderate | Serious |
| Cohen et al29 2019 | Moderate | Moderate | Low | Low | Low | Moderate | Moderate | Moderate |
| Dial et al30 2018 | Serious | Moderate | Low | Moderate | Low | Low | Low | Serious |
| Patel et al14 2018 | Serious | Moderate | Low | Moderate | Low | Low | Low | Serious |
| Winkler et al25 2018 | Serious | Moderate | Low | Low | Low | Moderate | Moderate | Serious |
| Khatri et al33 2017 | Moderate | Low | Low | Low | Moderate | Moderate | Low | Moderate |
| Otte et al26 2017 | Serious | Moderate | Low | Moderate | Low | Low | Moderate | Serious |
| Assor27 2010 | Moderate | Moderate | Moderate | Moderate | Moderate | Moderate | Low | Moderate |
Efficacy of intrawound vancomycin application
PJI was reported for all participants in the included studies, so data were pooled from all studies. A total of 30 (1.20%) patients in the vancomycin group and 58 (2.75%) in the control group developed PJI, respectively; meta-analysis indicated an OR for PJI of 0.44 (95% CI 0.28 to 0.69) when intrawound vancomycin was used, with no study heterogeneity observed (I2 = 0.0%) (Figure 3a). Four studies14,27,30,33 reported superficial infections in 598 patients treated with intrawound vancomycin and 377 in the control group. The incidence of superficial infection between the patients receiving intrawound vancomycin and control group showed no difference: four (0.67%) versus six (1.60%), with meta-analysis giving a pooled OR of 0.60 (95% CI 0.17 to 2.12) with no study heterogeneity (I2 = 0.0%) (Figure 3b).
Fig. 3
Forest plots of meta-analysis of risk of a) periprosthetic joint infection, b) superficial infection, c) aseptic wound complications, and d) acute kidney injury when intrawound vancomycin was used relative to the risk without it. CI, confidence interval; OR, odds ratio.
Safety of intrawound vancomycin administration
We also analyzed the incidence of adverse events associated with intrawound administration of vancomycin. Four studies13,14,28,30 reported that aseptic wound complications occurred in 1,069 treated patients (23 with aseptic wound complications, 2.15%) and 834 control patients (eight with aseptic wound complications, 0.96%). Meta-analysis confirmed that intrawound vancomycin treatment was associated with higher risk of aseptic wound complications (OR 2.39, 95% CI 1.09 to 5.23), with no substantial heterogeneity observed across the studies (I2 = 16.2%) (Figure 3c). Moreover, the study performed by Hanada et al28 reported that three out of five patients who suffered from PJI in the vancomycin group had aseptic wound complications within three months after primary surgery in their subjects. Additionally, they also revealed that intrawound vancomycin treatment may be related to higher risk of prolonged healing of wound (> two weeks): 14 (12.7%) versus three (3.3%) (OR 4.32, 95% CI 1.20 to 15.56; p = 0.016). The definition, number of patients, treatment measures, and outcomes of aseptic wound complication and prolonged healing of wound are shown in Table IV.
Table IV.
The details of aseptic wound complication and prolonged healing of wound of included studies.
| Study | Definition of aseptic wound complication or prolonged healing of wound | No. of patients | Treatment measures | Outcomes |
|---|---|---|---|---|
| Yavuz et al13 2019 |
Wound bleeding during adjustment of coumadin dose | C: 4 V: 3 All: 7 |
Coumadin was stopped and enoxaparin was started | No PJI was observed during the follow-up |
| Hanada et al28 2019 |
Skin erosion and wound dehiscence | C: 4 V: 13 All: 17 |
3 cases in the C and 6 cases in the V required treatment by lavage and resuturing under local anaesthesia | NR |
| Prolonged healing of wound (> 2 wks)* (excessive discharge, subcutaneous oedema, and seroma) | C: 3 V: 14 All: 17 |
3 cases in the C and 6 cases in the V required treatment by lavage and resuturing under local anaesthesia | NR | |
| Dial et al30 2018 |
Wound breakdowns that did not meet the criteria for PJI, and required a return to the operating room for debridement and closure | C: 0 V: 6 All: 6 |
All patients were treated with wound debridement and closure without head or liner exchange and no postoperative antibiotics | The wounds of all 6 patients healed after debridement and closure |
| Patel et al14 2018 |
Stitch abscesses or erythema | C: 0 V: 1 All: 1 |
Irrigation and debridement, and no antibiotic was used | Healed uneventfully, and no further surgical intervention was required |
-
*
Only one study28 reported the information of prolonged healing of wound.
-
C, control group; NR, not reported; PJI, periprosthetic joint infection; V, vancomycin group.
Based on data from six studies,13,14,28–30,33 the incidence of acute kidney injury after surgery occurred in 1,429 treated patients was 4 (0.28%) and 4 (0.35%) in 1,144 controls. Meta-analysis showed that intrawound vancomycin did not significantly alter risk of acute kidney injury (OR 0.71, 95% CI 0.19 to 2.55), with no substantial heterogeneity (I2 = 16.6%) (Figure 3d). Additionally, anaphylactic reactions did not occur in 772 treated patients or 786 controls in studies that reported the outcome,13,28,30,33 nor did ototoxicity occur in 595 treated patients or 332 controls in studies reporting that outcome.14,28,30
Discussion
To our knowledge, this is the first systematic review and meta-analysis assessing both the efficacy and safety of intrawound vancomycin treatment in primary hip and knee arthroplasty. We found that intrawound treatment may reduce the incidence of PJI without increasing risk of acute kidney injury; however, it does not reduce superficial infection but contrarily may increase risk of aseptic wound complications. Other complications, such as anaphylactic reaction and ototoxicity, are rare among patients treated with intrawound vancomycin in primary hip and knee arthroplasty. However, the reliability of the available evidence is limited due to the poor quality of the included studies.
Although intravenous cephalosporins are widely used to prevent infection in primary hip and knee arthroplasty, studies report that they have no effect on more than 60% of Staphylococcus strains, which are the most common causes of SSIs.10 Vancomycin is effective against many strains of pathogens, although some bacterial strains are resistant.35 Intrawound vancomycin used in spine surgery significantly decreases risk of postoperative infection.36 This finding likely inspired local vancomycin use in primary hip and knee arthroplasty.37 Indeed, a previous systematic review and meta-analysis demonstrated that intrawound vancomycin may decrease the risk of PJI in primary (OR 0.44, 95% CI 0.24 to 0.77) and revision (OR 0.28, 95% CI 0.13 to 0.61) TKA and THA; however, that study did not evaluate its safety.37 Safety risks associated with intrawound vancomycin must be taken into account, including aseptic wound complications, acute kidney injury, anaphylactic reaction, and ototoxicity. Updated, comprehensive assessment of the safety and efficacy of intrawound vancomycin in hip and knee arthroplasty is needed in order to provide the latest evidence for this practice, especially since there are no relevant international guidelines.
A small inoculum, such as Staphylococcus aureus present on the skin of 15% to 25% of healthy individuals, may accrue during surgery, especially during procedures in which a foreign body is implanted, giving rise to SSIs although the skin surface is prepared beforehand and sterile technique is carefully performed.11,38 Intrawound vancomycin was originally intended as an adjunct to intravenous prophylactic antibiotics routinely to reduce local contamination.39 Vancomycin administered intravenously may damage liver and kidney functions, whereas local application restricts high drug concentrations to surgical site, thereby reducing systemic absorption and systemic side effects, such as nephrotoxicity, ototoxicity, anaphylactic reaction, and even red man syndrome.40–42 Meta-analyses indicate that intrawound application of vancomycin significantly reduces SSI rate both in spinal surgery (OR 0.31, 95% CI 0.19 to 0.50)12 and non-spinal neurosurgery (OR 0.25, 95% CI 0.12 to 0.52).43 Similarly, our results also associate intrawound application of vancomycin with lower PJI in primary hip and knee arthroplasty (OR 0.44, 95% CI 0.28 to 0.69). However, we note that most of the studies in these three meta-analyses were retrospective, and their quality was poor.
The safety of intrawound application of vancomycin in primary hip and knee arthroplasty is a major concern for surgeons.44 Our study reinforces this concern by showing that intrawound vancomycin (1 g to 2 g) increases the risk of aseptic wound complications, including wound bleeding, skin erosion, wound dehiscence, seroma, and stitch erythema. Correspondingly, this may associate with the higher risk of reoperative including lavage or irrigation, debridement, and re-suturing.28,30 Another study showed that intrawound application of vancomycin can delay wound healing,35 which was in line with the study performed by Hanada et al.28 Hoelen et al45 reported that vancomycin may cause local non-bullous skin necrosis if it leaks into the surrounding tissues from veins due to vancomycin solution having a low pH (2.8 to 4.5), which has a direct irritant effect on the vascular wall. Although the cause of aseptic wound complications is uncertain, these may result, at least in part, from a direct release of histamine by vancomycin or coproducts, local inflammatory response triggered by impurities in the vancomycin itself, or by the body’s response to vancomycin and/or impurities.29
The risk of positively selecting drug-resistant organisms is another concern that has prevented broader application of intrawound vancomycin in arthroplasty.39 Johnson et al46 reported that vancomycin could be present in treated tissue at concentrations over 200 μg/ml for 24 hours postoperatively after local administration of 2 g of vancomycin, much higher than the concentrations of 15 μg/ml to 20 μg/ml recommended for life-threatening infections.47 However, the serum level of vancomycin in that study was below 4.7 ± 3.2 μg/ml, which is subtherapeutic and may lead to emergence of drug-resistance,48 although an association between local administration of antibiotic and emergence of resistant bacteria has been difficult to demonstrate. In addition, the potential for third body wear from crystalline vancomycin applied to the implant interface may promote degradation of the implant. Nevertheless, simulations by Qadir et al49 suggest that local crystalline of vancomycin does not alter wear rates of cobalt-chromium (Co-Cr) on ultra-high molecular weight polyethylene, which may reflect the short half-life of vancomycin (7.2 hours) in the intra-articular space.46 These considerations make clear that intrawound vancomycin, while offering clinical benefits, should be further optimized.
While we found no systemic adverse events from intrawound application of vancomycin, such as acute kidney injury, ototoxicity, or anaphylactic reaction, only a subset of included studies reported these outcomes. This lack of adverse events may reflect the benefits of local administration, although the vancomycin is ultimately excreted through the kidneys.50 In addition, some surgeons are concerned that high local vancomycin levels may affect bone healing.51 Edin et al52 suggested that local vancomycin of therapeutic concentration has no obvious impact on osteoblast replication. In fact, that study concluded that vancomycin may be even less toxic to osteoblasts than cefazolin. On the other hand, Chu et al53 reported that vancomycin has dose-dependent toxic effects on mesenchymal stem cells (MSCs) in vitro: incubation for 24 hours in medium containing 400 μg/ml or 1,600 μg/ml vancomycin led to MSC death rates of 9.43% and 13.79%, respectively. Consequently, higher doses of local vancomycin may not bring additional benefits, and optimal dosing needs to be further explored.
Our meta-analysis should be interpreted with caution given the limitations of the available literature. The studies included in our analyses being of poor quality will impact the results. In addition, the included studies varied in recruitment period, follow-up time, dosage, sites of application, and vancomycin formulation, and these variations may create biases and affect our results. Finally, future prospective high-quality studies are needed to verify our results, especially for the impact of intrawound vancomycin on aseptic wound complication including its dose, site of use, and the size of the wound. However, due to the low incidence of PJI, about 1,000 patients undergoing primary hip and knee arthroplasty are required to conform a large decrease in PJI from 2.74% to 1.00% in control versus vancomycin groups, with a power of 80% and at a 5% significance level in a randomized controlled trial (RCT).
In conclusion, intrawound application of vancomycin may decrease the risk of PJI but concurrently increase the risk of aseptic wound complications in primary hip and knee arthroplasty. More high-quality RCTs should be performed to clarify the advantages or disadvantages of intrawound vancomycin due to the poor quality of the included studies.
References
1. Parvizi J , Shohat N , Gehrke T . Prevention of periprosthetic joint infection: new guidelines . Bone Joint J . 2017 ; 99-B ( 4 Supple B ): 3 – 10 . Crossref PubMed Google Scholar
2. Bozic KJ , Ong K , Lau E , et al. . Estimating risk in Medicare patients with THA: an electronic risk calculator for periprosthetic joint infection and mortality . Clin Orthop Relat Res . 2013 ; 471 ( 2 ): 574 – 583 . Crossref PubMed Google Scholar
3. Lenguerrand E , Whitehouse MR , Beswick AD , et al. . Revision for prosthetic joint infection following hip arthroplasty: evidence from the National joint registry . Bone Joint Res . 2017 ; 6 ( 6 ): 391 – 398 . Crossref PubMed Google Scholar
4. Gold P , Garbarino LJ , Sodhi N , et al. . Annual nationwide infection trends for revision total knee arthroplasty . J Knee Surg . 2019 . (Epub ahead of print) PMID: 31491795. Crossref PubMed Google Scholar
5. Bosco JA , Bookman J , Slover J , Edusei E , Levine B . Principles of antibiotic prophylaxis in total joint arthroplasty: current concepts . J Am Acad Orthop Surg . 2015 ; 23 ( 8 ): e27 – e35 . PubMed Google Scholar
6. Johnson DP . Antibiotic prophylaxis with cefuroxime in arthroplasty of the knee . J Bone Joint Surg Br . 1987 ; 69-B ( 5 ): 787 – 789 . Crossref PubMed Google Scholar
7. Milstone AM , Maragakis LL , Townsend T , et al. . Timing of preoperative antibiotic prophylaxis: a modifiable risk factor for deep surgical site infections after pediatric spinal fusion . Pediatr Infect Dis J . 2008 ; 27 ( 8 ): 704 – 708 . Crossref PubMed Google Scholar
8. Murphy E , Spencer SJ , Young D , et al. . MRSA colonisation and subsequent risk of infection despite effective eradication in orthopaedic elective surgery . J Bone Joint Surg Br . 2011 ; 93-B ( 4 ): 548 – 551 . Crossref PubMed Google Scholar
9. Norton TD , Skeete F , Dubrovskaya Y , et al. . Orthopedic surgical site infections: analysis of causative bacteria and implications for antibiotic stewardship . Am J Orthop (Belle Mead NJ) . 2014 ; 43 ( 5 ): E89 – E92 . PubMed Google Scholar
10. Klevens RM , Morrison MA , Nadle J , et al. . Invasive methicillin-resistant Staphylococcus aureus infections in the United States . JAMA . 2007 ; 298 ( 15 ): 1763 – 1771 . Crossref PubMed Google Scholar
11. Sadigursky D , Sousa MD , Cajaíba YGL , Martins RR , Lobão DMV . Infectious prophylaxis with Intrawound vancomycin powder in orthopedic surgeries: systematic review with meta-analysis . Rev Bras Ortop . 2019 ; 54 ( 6 ): 617 – 626 . Crossref PubMed Google Scholar
12. Tailaiti A , Shang J , Shan S , Muheremu A . Effect of intrawound vancomycin application in spinal surgery on the incidence of surgical site infection: a meta-analysis . Ther Clin Risk Manag . 2018 ; 14 : 2149 – 2159 . Crossref PubMed Google Scholar
13. Yavuz IA , Oken OF , Yildirim AO , Inci F , Ceyhan E , Gurhan U . No effect of vancomycin powder to prevent infection in primary total knee arthroplasty: a retrospective review of 976 cases . Knee Surg Sports Traumatol Arthrosc . 2019 ; 28 ( 9 ): 3055 – 3060 . Crossref PubMed Google Scholar
14. Patel NN , Guild GN , Kumar AR . Intrawound vancomycin in primary hip and knee arthroplasty: a safe and cost-effective means to decrease early periprosthetic joint infection . Arthroplast Today . 2018 ; 4 ( 4 ): 479 – 483 . Crossref PubMed Google Scholar
15. Baeza J , Cury MB , Fleischman A , et al. . General assembly, prevention, local antimicrobials: proceedings of international consensus on orthopedic infections . J Arthroplasty . 2019 ; 34 ( 2S ): S75 – S84 . Crossref PubMed Google Scholar
16. Sterne JA , Hernán MA , Reeves BC , et al. . ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions . BMJ . 2016 ; 355 : i4919 . Crossref PubMed Google Scholar
17. Higgins JPT , Thompson SG , Deeks JJ , Altman DG . Measuring inconsistency in meta-analyses . BMJ . 2003 ; 327 ( 7414 ): 557 – 560 . Crossref PubMed Google Scholar
18. Higgins JPT , Thompson SG . Quantifying heterogeneity in a meta-analysis . Stat Med . 2002 ; 21 ( 11 ): 1539 – 1558 . Crossref PubMed Google Scholar
19. Borenstein M , Hedges LV , Higgins JPT , Rothstein HR . A basic introduction to fixed-effect and random-effects models for meta-analysis . Res Synth Methods . 2010 ; 1 ( 2 ): 97 – 111 . Crossref PubMed Google Scholar
20. Egger M , Davey Smith G , Schneider M , Minder C . Bias in meta-analysis detected by a simple, graphical test . BMJ . 1997 ; 315 ( 7109 ): 629 – 634 . PubMed Google Scholar
21. Sterne JAC , Sutton AJ , Ioannidis JPA , et al. . Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials . BMJ . 2011 ; 343 : d4002 . Crossref PubMed Google Scholar
22. Romanò CL , Malizos K , Capuano N , et al. . Does an Antibiotic-Loaded Hydrogel Coating Reduce Early Post-Surgical Infection After Joint Arthroplasty? J Bone Jt Infect . 2016 ; 1 : 34 – 41 . Crossref PubMed Google Scholar
23. Whiteside LA . Prophylactic peri-operative local antibiotic irrigation . Bone Joint J . 2016 ; 98-B ( 1 Suppl A ): 23 – 26 . Crossref PubMed Google Scholar
24. Omrani F , Emami M , Sarzaeem M , Zarei R , Yeganeh A . The effect of Intra-wound vancomycin powder application in reducing surgical site infections after total hip arthroplasty . Biosci Biotechnol Res Asia . 2015 ; 12 ( 3 ): 2383 – 2386 . Google Scholar
25. Winkler C , Dennison J , Wooldridge A , et al. . Do local antibiotics reduce periprosthetic joint infections? A retrospective review of 744 cases . J Clin Orthop Trauma . 2018 ; 9 ( Suppl 1 ): S34 – S39 . Crossref PubMed Google Scholar
26. Otte JE , Politi JR , Chambers B , Smith CA . Intrawound vancomycin powder reduces early prosthetic joint infections in revision hip and knee arthroplasty . Surg Technol Int . 2017 ; 30 : 284 – 289 . PubMed Google Scholar
27. Assor M . Noncemented total knee arthroplasty with a local prophylactic anti-infection agent: a prospective series of 135 cases . Can J Surg . 2010 ; 53 ( 1 ): 47 – 50 . PubMed Google Scholar
28. Hanada M , Nishikino S , Hotta K , et al. . Intrawound vancomycin powder increases post-operative wound complications and does not decrease periprosthetic joint infection in primary total and unicompartmental knee arthroplasties . Knee Surg Sports Traumatol Arthrosc . 2019 ; 27 ( 7 ): 2322 – 2327 . Google Scholar
29. Cohen EM , Marcaccio S , Goodman AD , Lemme NJ , Limbird R . Efficacy and cost-effectiveness of topical vancomycin powder in primary Cementless total hip arthroplasty . Orthopedics . 2019 ; 42 ( 5 ): e430 – e436 . Crossref PubMed Google Scholar
30. Dial BL , Lampley AJ , Green CL , Hallows R . Intrawound vancomycin powder in primary total hip arthroplasty increases rate of sterile wound complications . Hip Pelvis . 2018 ; 30 ( 1 ): 37 – 44 . Crossref PubMed Google Scholar
31. Parvizi J , Gehrke T , Chen AF . Proceedings of the International consensus on periprosthetic joint infection . Bone Joint J . 2013 ; 95-B ( 11 ): 1450 – 1452 . Crossref PubMed Google Scholar
32. Parvizi J , Zmistowski B , Berbari EF , et al. . New definition for periprosthetic joint infection: from the Workgroup of the musculoskeletal infection Society . Clin Orthop Relat Res . 2011 ; 469 ( 11 ): 2992 – 2994 . Crossref PubMed Google Scholar
33. Khatri K , Bansal D , Singla R , Sri S . Prophylactic intrawound application of vancomycin in total knee arthroplasty . J Arthrosc Joint Surg . 2017 ; 4 ( 2 ): 61 – 64 . Google Scholar
34. Parvizi J , Gehrke T , Chen AF . Proceedings of the International consensus on periprosthetic joint infection . Bone Joint J . 2013 ; 95-B ( 11 ): 1450 – 1452 . Crossref PubMed Google Scholar
35. Katarincic JA , Fantry A , DePasse JM , Feller R . Local modalities for preventing surgical site infections: an evidence-based review . J Am Acad Orthop Surg . 2018 ; 26 ( 1 ): 14 – 25 . Crossref PubMed Google Scholar
36. Kunakornsawat S , Sirikajohnirun S , Piyaskulkaew C , et al. . Comparison between 1 g and 2 g of Intrawound Vancomycin Powder Application for Prophylaxis in Posterior Instrumented Thoracic or Lumbosacral Spine Surgery: A Preliminary Report . Asian J Neurosurg . 2019 ; 14 ( 3 ): 710 – 714 . Crossref PubMed Google Scholar
37. Heckmann ND , Mayfield CK , Culvern CN , et al. . Systematic review and meta-analysis of Intrawound vancomycin in total hip and total knee arthroplasty: a call for a prospective randomized trial . J Arthroplasty . 2019 ; 34 ( 8 ): 1815 – 1822 . Crossref PubMed Google Scholar
38. Chen AF , Fleischman A , Austin MS . Use of Intrawound antibiotics in orthopaedic surgery . J Am Acad Orthop Surg . 2018 ; 26 ( 17 ): e371 – e378 . Crossref PubMed Google Scholar
39. Edelstein AI , Weiner JA , Cook RW , et al. . Intra-Articular Vancomycin Powder Eliminates Methicillin-Resistant S. aureus in a Rat Model of a Contaminated Intra-Articular Implant . J Bone Joint Surg Am . 2017 ; 99-A ( 3 ): 232 – 238 . Crossref PubMed Google Scholar
40. Hanssen AD , Osmon DR , Patel R . Local antibiotic delivery systems: where are we and where are we going? Clin Orthop Relat Res . 2005 ; 437 : 111 – 114 . PubMed Google Scholar
41. Fernicola SD , Elsenbeck MJ , Grimm PD , Pisano AJ , Wagner SC . Intrasite antibiotic powder for the prevention of surgical site infection in extremity surgery: a systematic review . J Am Acad Orthop Surg . 2020 ; 28 ( 1 ): 37 – 43 . Crossref PubMed Google Scholar
42. McNamara DR , Steckelberg JM . Vancomycin . J Am Acad Orthop Surg . 2005 ; 13 ( 2 ): 89 – 92 . Crossref PubMed Google Scholar
43. Bokhari R , You E , Zeiler FA , et al. . Effect of Intrawound vancomycin on surgical site infections in Nonspinal neurosurgical procedures: a systematic review and meta-analysis . World Neurosurg . 2019 ; 123 : e7 : 409 – 417 . Crossref PubMed Google Scholar
44. Stravinskas M , Nilsson M , Vitkauskiene A , Tarasevicius S , Lidgren L . Vancomycin elution from a biphasic ceramic bone substitute . Bone Joint Res . 2019 ; 8 ( 2 ): 49 – 54 . Crossref PubMed Google Scholar
45. Hoelen DWM , Tjan DHT , van Vugt R , van der Meer YG , van Zanten ARH . Severe local vancomycin induced skin necrosis . Br J Clin Pharmacol . 2007 ; 64 ( 4 ): 553 – 554 . Crossref PubMed Google Scholar
46. 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 ; 32 ( 3 ): 924 – 928 . Crossref PubMed Google Scholar
47. Liu C , Bayer A , Cosgrove SE , et al. . Clinical practice guidelines by the infectious diseases society of america for the treatment of methicillin-resistant Staphylococcus aureus infections in adults and children: executive summary . Clin Infect Dis . 2011 ; 52 ( 3 ): 285 – 292 . Crossref PubMed Google Scholar
48. Rybak MJ , Lomaestro BM , Rotschafer JC , et al. . Vancomycin therapeutic guidelines: a summary of consensus recommendations from the infectious diseases Society of America, the American Society of Health-System Pharmacists, and the Society of Infectious Diseases Pharmacists . Clin Infect Dis . 2009 ; 49 ( 3 ): 325 – 327 . Crossref PubMed Google Scholar
49. Qadir R , Ochsner JL , Chimento GF , et al. . Establishing a role for vancomycin powder application for prosthetic joint infection prevention-results of a wear simulation study . J Arthroplasty . 2014 ; 29 ( 7 ): 1449 – 1456 . Crossref PubMed Google Scholar
50. Buttaro MA , Gimenez MI , Greco G , Barcan L , Piccaluga F . High active local levels of vancomycin without nephrotoxicity released from impacted bone allografts in 20 revision hip arthroplasties . Acta Orthop . 2005 ; 76 ( 3 ): 336 – 340 . PubMed Google Scholar
51. Morgenstern M , Vallejo A , McNally MA , et al. . The effect of local antibiotic prophylaxis when treating open limb fractures: a systematic review and meta-analysis . Bone Joint Res . 2018 ; 7 ( 7 ): 447 – 456 . Crossref PubMed Google Scholar
52. Edin ML , Miclau T , Lester GE , Lindsey RW , Dahners LE . Effect of cefazolin and vancomycin on osteoblasts in vitro . Clin Orthop Relat Res . 1996 ; 333 : 245 – 251 . PubMed Google Scholar
53. Chu S , Chen N , Dang ABC , Kuo AC , Dang ABC . The effects of topical vancomycin on mesenchymal stem cells: more may not be better . Int J Spine Surg . 2017 ; 11 : 12 . Google Scholar
Author contributions
H. Xu: Screened the literature and extracted relevant data, Drafted and revised the manuscript.
J. Yang: Screened the literature and extracted relevant data, Drafted and revised the manuscript.
J. Xie: Analyzed and interpreted the data.
Z. Huang: Analyzed and interpreted the data.
Q. Huang: Analyzed and interpreted the data.
G. Cao: Analyzed and interpreted the data, Participated in the final approval of the manuscript to be published.
F. Pei: Conceptualized and designed the study, Revised the manuscript.
H. Xu, J. Yang, and J. Xie contributed equally to this work.
Funding statement
The research was supported by Post-Doctor Research Project, West China Hospital, Sichuan University (2018HXBH073); National Clinical Research Center for Geriatrics, West China Hospital, Sichuan University (Z2018B11). No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
ICMJE COI statement
No conflict of interest exits in the submission of this manuscript, and manuscript is approved by all authors for publication.
Acknowledgements
We thank Craig J. Della Valle and Nathanael D. Heckmann from the Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, Illinois, USA for providing literature support. We also thank Chapin Rodríguez, from Creaducate Enterprises Ltd for editing the English text of a draft of this manuscript.
Supplementary material
Table showing the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist.
© 2020 Author(s) et al. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/.
