One-Stage Peri-Prosthetic Knee Revision Surgery with Local Antibacterial Hydrogel: Long-Term Results

1. Introduction

Total knee arthroplasty (TKA) is among the most successful procedures in orthopaedic surgery, providing reliable pain relief, restoration of function and substantial improvement in quality of life for patients with end-stage knee disease. Owing to an ageing population, increasing life expectancy and expanding indications, the annual number of primary TKAs continues to rise worldwide. It has been estimated that more than 3.4 million primary TKAs will be performed annually in the United States by 2030, representing an increase of more than six-fold compared with 2005 [1]. Despite these excellent outcomes, periprosthetic joint infection (PJI) remains one of the most devastating complications following arthroplasty, accounting for a large proportion of early revision procedures and imposing a considerable economic burden on healthcare systems [2]. Recent estimates suggest that the annual cost of treating hip and knee PJI in the United States alone will exceed US$1.8 billion [3].

For decades, two-stage exchange arthroplasty has been regarded as the reference treatment for chronic knee PJI, particularly in North America [4]. However, growing evidence accumulated during the last two decades has progressively challenged this paradigm. Several systematic reviews, registry studies and contemporary case series have demonstrated that, in appropriately selected patients, one-stage revision can achieve infection eradication rates comparable to those of two-stage revision while offering important clinical advantages [5,6,7,8]. These include a single surgical procedure, lower perioperative morbidity, shorter hospitalization, earlier rehabilitation, preservation of bone stock, reduced soft-tissue damage, lower healthcare costs and improved patient satisfaction. Consequently, international consensus statements have progressively expanded the indications for one-stage revision. While earlier recommendations considered factors such as culture-negative infection, resistant microorganisms or the presence of a sinus tract as contraindications, the most recent International Consensus Meeting (ICM 2025) recognizes that many of these conditions should instead be considered relative rather than absolute contraindications. Current selection criteria primarily emphasize the possibility of performing radical debridement, obtaining adequate soft-tissue coverage and achieving stable reconstruction in the absence of uncontrolled systemic sepsis [9].

Regardless of the surgical strategy adopted, successful treatment of PJI depends on several fundamental principles: meticulous radical debridement, complete removal of infected tissues and implants, stable reconstruction, targeted systemic antimicrobial therapy and effective local antibiotic delivery. Among these factors, prevention of bacterial adhesion to the newly implanted prosthesis has attracted increasing attention. Experimental studies have demonstrated that bacterial colonization begins within minutes after implantation, initiating biofilm formation that subsequently protects microorganisms from both antibiotics and host immune responses. This concept, often referred to as the “race for the surface”, has stimulated the development of technologies capable of protecting implant surfaces during the critical early postoperative period [10].

DAC^® (Defensive Antibacterial Coating) hydrogel is a fully resorbable implant coating composed of covalently linked hyaluronic acid and poly-D,L-lactic acid [11]. The hydrogel acts as a temporary physical barrier that reduces early bacterial adhesion while simultaneously serving as a carrier for locally delivered antibiotics selected according to microbiological findings. Complete resorption occurs within approximately 72 h, allowing osseointegration of cementless components without leaving permanent foreign material on the implant surface [12]. In vitro studies have consistently demonstrated significant inhibition of bacterial adhesion and biofilm formation, including against methicillin-resistant staphylococci [13]. Experimental animal studies further confirmed the ability of DAC hydrogel to reduce bacterial colonization without impairing bone healing or implant integration [14,15].

Clinical evidence has progressively supported these experimental observations. Several clinical series have reported encouraging infection eradication rates following hip and knee PJI revision surgery when DAC hydrogel was used as an adjunct to meticulous surgical debridement and systemic antibiotic therapy [16,17,18,19,20]. Nevertheless, most available reports include relatively limited patient numbers and, importantly, follow-up durations rarely exceeding three to five years. Consequently, robust evidence regarding the durability of infection eradication and long-term implant survival remains scarce.

Long-term follow-up is particularly relevant in PJI surgery because recurrence may occur several years after revision, and short-term success does not necessarily translate into durable cure. Furthermore, data regarding the use of DAC hydrogel in conjunction with modern hybrid fixation techniques, combining cemented epiphyseal fixation with cementless stems, remain limited despite the increasing popularity of this reconstructive strategy.

The purpose of the present study was therefore to evaluate the medium- and long-term clinical, radiographic and microbiological outcomes of a consecutive series of patients undergoing one-stage revision for infected total knee arthroplasty using antibiotic-loaded DAC hydrogel. To our knowledge, this represents the longest reported clinical follow-up of one-stage knee revision employing this biodegradable antibacterial coating, providing valuable information regarding its long-term safety, durability and effectiveness in preventing recurrent infection.

2. Results

A total of 52 patients undergoing one-stage revision for peri-prosthetic knee infection (PJI) or for septic knee spacer with the use of DAC antibacterial coating, treated according to the same protocol between year 2014 and 2023 were included in the study.

2.1. Pre-Operative Data

Mean age at surgery was 73.2 ± 8.9 years (range 45–87), and 32 patients (61.5%) were female. Mean body mass index (BMI) was 29.3 ± 4.4 kg/m² (range 20.3–40) (cf. Table 1).

Forty-eight patients (92.3%) underwent revision for PJI, while 4 patients (7.7%) were revised for septic spacer in a previously infected knee prosthesis. Twelve patients (23.1%) presented clinical signs of acute infection, while no patient had active draining fistula or soft tissue defect and implant exposure at the time of our revision surgery. According to the McPherson’s systemic host’s classification [21], 12 patients (23.1%) were classified as type A hosts, 19 (36.5%) as type B and 21 (40.4%) type C. The most frequent comorbidities included cardiopathy (48.1%), diabetes mellitus or other endocrinopathies (19.2%), respiratory tract disease (13.5%), smoking (13.5%), hepatitis (11.5%), rheumatological disorders (9.6%), chronic renal insufficiency (9.6%), oncological conditions (9.6%), peripheral vasculopathy (7.7%), dermatological (5.8%) or neurological (3.8%) disorders. Anesthesiologic evaluation at the time of surgery graded 10 patients ASA 2, 29 as ASA 3 and 13 as ASA 4. Average time from first implant was 4.5 ± 4.0 years (range 0.5–17). Twenty-three patients (44.2%) had received one or more surgeries to treat the infection prior to our first observation.

2.2. Intra-Operative Data

Bone loss has been evaluated intra-operatively, after implant removal, at the end of surgical debridement. According to the AORI classification [22], 7 (13.5%) patients were classified as AORI type 1, 11 (21.2%) as type 2A, 16 (30.8%) as type 2B and 22 (42.3%) as AORI type 3. Based on bone defect and on capsulo-ligament competence, 15 patients received a hinged prosthesis, while the remaining 37 were treated with a semi-constrained implant.

Average surgery duration: 142.5 ± 23.6 min (range 95–184). Average blood loss: 460 ± 110 mL (range 300–650).

2.3. Microbiological Data

Microbiological cultures were positive in the majority of cases (cf. Table 2); in 4 patients (7.7%) no pathogen was found and diagnosis was confirmed by histology and other parameters according to MSIS 2018 and WAIOT 2020 scores [23,24].

Methicillin-sensitive Staph. aureus and epidermidis were identified in 7.7% and in 11.5% of the cases, while methicillin-resistant strains accounted for 9.6% and 11.5%, respectively.

Overall, the most commonly isolated organisms were coagulase-negative staphylococci. Less frequent, but still relevant isolates included Streptococci, Propionibacterium acnes, and Enterococci, while mixed flora infections were detected in 8 patients (15.4%).

2.4. Clinical Outcomes

Four patients were lost to follow-up: two dead for causes unrelated to the surgical intervention one and 3 years after surgery, and two more lost at controls at two and three years from surgery, respectively, leaving a total of 48 patients available for the present analysis with an average follow-up of 7.0 ± 2.7 years (range 3–11).

Table 3 summarizes the main clinical outcomes: an average improvement of 36.7 and 41 points was observed in KSS score and KSS functional score, respectively, while pain at rest and at movement dropped from 6.9 and 8.1 prior to surgery to 0.7 and 1.6 at follow-up.

At final follow-up, 47 of the 48 patients available for clinical evaluation did not show any sign of infection recurrence (97.9%). Kaplan–Meier infection-free survivorship was 98.1% (95% CI, 87.1–99.7%) at 11 years. (cf. Figure 1). The only patient who experienced recurrent infection one year after surgery, due to a relapse of her recurrent leg erysipelas and a concomitant strongyloidiasis, underwent above-knee amputation in another hospital.

Laboratory tests showed on average pre-operative CRP values of 20.9 ± 35.5 mg/L (range 6–199.4) and post-operative values of 5.1 ± 4.2 mg/L (range 2–11).

Radiographic examination did reveal non progressive lucent lines of <2 mm in 18 patients, while two patients showed radiolucent lines > 2 mm in the tibial component and one patient in both the femoral and tibial component, in the absence of clinical signs of loosening.

Concerning complications, two patients required early stiffness treatment by manipulation under anesthesia and one more patient suffered deep venous thromboembolism. One patient was treated with internal osteosynthesis in another hospital for peri-prosthetic femoral fracture 6 years after revision. No other patient required additional surgery for infection recurrence or for aseptic loosening.

3. Discussion

The main finding of the present study is that one-stage revision combined with DAC antibacterial hydrogel provided excellent long-term infection control and functional outcomes after revision total knee arthroplasty.

At a mean follow-up of seven years, with observations extending to eleven years, infection-free survivorship reached 97.9%, while patients experienced marked improvements in Knee Society Scores, pain relief and maintenance of stable implant fixation. To our knowledge, this represents the longest follow-up currently reported for one-stage knee revision employing DAC hydrogel technology.

Long-term evidence is particularly important in the treatment of periprosthetic joint infection because successful short-term eradication does not necessarily translate into durable infection control. The sustained survivorship observed in our series suggests that the local antibacterial protection afforded by DAC hydrogel during the early postoperative period may contribute to preventing bacterial recolonization of newly implanted components while allowing long-term biological fixation to proceed normally. Importantly, no increase in implant-related complications, delayed osseointegration or mechanical failures attributable to the hydrogel was observed, supporting the long-term safety profile of this biodegradable coating.

Why might DAC hydrogel improve infection control? Although the present study was not designed to isolate the specific contribution of DAC hydrogel from the other components of the one-stage revision protocol, the biological rationale supporting its use is well established. Successful treatment of periprosthetic joint infection depends not only on eradication of established microorganisms through surgical debridement and systemic antimicrobial therapy, but also on prevention of bacterial recolonization of the newly implanted prosthesis during the immediate postoperative period. Experimental studies have demonstrated that bacterial adhesion to implant surfaces occurs within minutes after implantation, initiating the formation of a mature biofilm that subsequently protects microorganisms from host immune defenses and markedly reduces antibiotic susceptibility. This competition between bacterial colonization and host cell integration, commonly referred to as the “race for the surface”, represents one of the critical determinants of implant-related infection [25,26].

DAC^® hydrogel was specifically developed to interfere with this early phase of bacterial adhesion. The biodegradable hyaluronan/poly-D,L-lactide matrix forms a temporary hydrophilic barrier on the implant surface while simultaneously functioning as a carrier for high local concentrations of antibiotics selected according to the microbiological profile. In contrast to permanent antibacterial coatings, DAC hydrogel is completely resorbed within approximately 72 h, after the period of greatest risk for bacterial adhesion has passed, thereby avoiding interference with subsequent osseointegration of cementless implants. Experimental investigations have consistently demonstrated reduced bacterial adhesion and inhibition of biofilm formation against both methicillin-sensitive and methicillin-resistant staphylococci, while in vivo studies have confirmed preservation of normal bone healing and implant integration [20].

The excellent long-term results observed in the present series are consistent with this biological mechanism. Although the hydrogel itself is present only during the first postoperative days, successful prevention of initial bacterial adhesion may have lasting clinical consequences because mature biofilm formation is effectively prevented before it becomes established. This concept may explain why a temporary biodegradable coating can produce durable protection extending many years beyond complete hydrogel resorption. The absence of aseptic loosening or evidence of impaired biological fixation in our cohort further supports the long-term safety of this technology and suggests that transient local antibacterial protection can be achieved without compromising implant integration.

An additional strength of the present series is the consistent use of hybrid fixation, a solution that provides ease of insertion, possibly improved component alignment, and easier removal if required, compared with fully cemented prostheses [27,28]. In all patients, fixation was achieved with antibiotic-loaded cement limited to the epiphyseal component, whereas the metaphyseal and diaphyseal stems remained uncemented and were coated with antibiotic-loaded DAC hydrogel. This strategy combines the mechanical advantages of biological stem fixation with immediate local antibacterial protection of the uncemented implant surfaces, an aspect that has received little attention in previous reports. The excellent survivorship observed in this cohort suggests that DAC hydrogel may facilitate broader application of hybrid fixation during one-stage septic revision without compromising infection eradication.

The microbiological spectrum reflected the contemporary epidemiology of knee PJI, with coagulase-negative staphylococci predominating and a substantial proportion of methicillin-resistant organisms [29]. Moreover, we did not observe a different outcome in patients with negative cultural examination, a result that is in line with the recent report from Sabater-Martos and co-workers [30]. Despite this challenging microbiological profile, infection control remained excellent. Likewise, almost half of the cohort had undergone previous surgical procedures before referral, and over 40% presented severe bone defects (AORI type 3), highlighting that favorable outcomes were obtained in a complex revision population rather than in highly selected low-risk cases.

The remarkably low loss to follow-up further strengthens the validity of the present findings. Only four patients were unavailable for final analysis, two because of unrelated deaths and two because of loss to follow-up, minimizing attrition bias in a study extending over more than a decade.

3.1. Comparison with Published One-Stage Revision Series

The infection-free survivorship observed in the present study compares favorably with the largest published series of one-stage revision for chronic knee PJI. Historically, one-stage revision has achieved infection eradication rates ranging from approximately 80% to 97%, depending on patient selection, microbiological profile, surgical technique and duration of follow-up [5,6,31,32].

Early European experiences by Freeman [33], Buechel [34] and later by Jenny and colleagues [35] demonstrated that meticulous debridement combined with immediate reimplantation could achieve durable infection control in carefully selected patients. More contemporary series from high-volume revision centers have reported infection eradication rates generally between 85% and 93%, even in complex revision settings. For example, Gehrke and colleagues [36] reported infection control exceeding 90% following standardized one-stage exchange protocols, while Haddad et al. similarly demonstrated that one-stage revision can provide excellent infection eradication together with improved functional recovery compared with staged reconstruction in selected patients [37].

The present study demonstrated an infection-free survival of 97.9%, despite including a challenging cohort characterized by advanced bone loss, multiple previous procedures and a substantial proportion of compromised hosts according to the McPherson classification. Importantly, almost half of the patients had undergone previous surgical treatment for infection before referral, suggesting that our cohort cannot be considered a low-risk population. These findings therefore compare favorably with previously published series and reinforce the concept that excellent outcomes can be achieved when meticulous surgical debridement is combined with optimized local and systemic antimicrobial strategies.

Unlike many previous reports in which fully cemented revision constructs were routinely employed, all patients in the present series underwent hybrid fixation with cemented epiphyseal fixation and cementless stems. This reconstructive strategy has become increasingly popular because it facilitates restoration of alignment and biological fixation while reducing the amount of cement required. However, concerns have occasionally been raised regarding the absence of local antibiotic release around cementless implant surfaces. The routine application of antibiotic-loaded DAC hydrogel to the uncemented stems may represent an effective solution to this potential limitation, providing temporary local antibacterial protection during the critical period before osseointegration. The excellent long-term implant stability observed in our cohort suggests that this approach does not adversely affect biological fixation while maintaining excellent infection control.

3.2. Comparison with Systematic Reviews and Meta-Analysis

The present findings are also consistent with the conclusions of the most recent systematic reviews and meta-analyses evaluating one-stage revision for knee PJI.

Kunutsor et al., [7] in one of the largest meta-analyses comparing one-stage and two-stage revision, concluded that reinfection rates are broadly comparable between the two procedures when patients are appropriately selected. Similarly, the recent meta-analysis by Patel et al. reported no significant difference in infection eradication between one-stage and two-stage revision while confirming several advantages of the one-stage approach, including lower morbidity, fewer surgical procedures, shorter hospitalization and improved functional recovery.

Nagra et al., [38] reviewing exclusively one-stage revision for infected total knee arthroplasty, reported pooled infection eradication rates of approximately 87–89%, although considerable heterogeneity existed among the included studies regarding patient selection, microbiological characteristics and surgical protocols.

Our infection-free survival of 97.9% therefore lies at the upper end of the outcomes reported in these pooled analyses. Although direct comparisons should be interpreted cautiously because of differences in inclusion criteria and follow-up duration, our findings support the growing body of evidence indicating that one-stage revision should no longer be considered an alternative reserved for highly selected patients, but rather a valid standard treatment in experienced multidisciplinary centers.

Perhaps even more importantly, the present study extends the available evidence beyond the follow-up duration reported in most previous investigations. The majority of studies included in current meta-analyses report follow-up periods between two and five years, whereas recurrent infection may occur considerably later. The durability of infection eradication observed after up to eleven years in the present series therefore represents an important contribution to the existing literature and provides additional reassurance regarding the long-term effectiveness of this treatment strategy.

3.3. The Potential Contribution of DAC Antibacterial Hydrogel

Although the retrospective design of the present study does not allow the independent effect of DAC hydrogel to be isolated, several observations deserve consideration.

The antibacterial hydrogel provides temporary local protection during the early postoperative period, when bacterial adhesion and biofilm formation are most likely to occur. Unlike permanent antimicrobial coatings, DAC hydrogel is completely resorbed within approximately 72 h and therefore does not interfere with long-term osseointegration of cementless implants. Previous experimental investigations demonstrated significant inhibition of bacterial adhesion without evidence of cytotoxicity or impairment of bone healing. Furthermore, multicentre clinical studies evaluating the hydrogel in trauma and revision arthroplasty have consistently demonstrated reduced surgical-site infection rates and an excellent safety profile.

The present study substantially expands these previous observations [39] by demonstrating that the favorable early results are maintained over long-term follow-up. In particular, the absence of increased aseptic loosening or mechanical complications supports the long-term biological safety of the hydrogel, while the excellent infection-free survival suggests that temporary surface protection during the critical early postoperative phase may have durable clinical consequences extending many years beyond complete hydrogel resorption.

3.4. Study Limitations

Several limitations should nevertheless be acknowledged. First, this is a retrospective single-centre study including a relatively limited number of patients. Although the cohort is one of the largest homogeneous experiences with DAC hydrogel in one-stage knee revision and follow-up completeness was excellent, larger multicentre investigations would improve external validity. Second, the absence of a contemporary control group prevents direct assessment of the incremental benefit provided by DAC hydrogel. Nevertheless, the observed infection-free survivorship lies at the upper end of the success rates generally reported for one-stage revision in systematic reviews and contemporary series. Finally, because all procedures followed a standardized multidisciplinary protocol combining meticulous debridement, targeted systemic antibiotics and local antibacterial coating, the specific contribution of each individual component cannot be isolated.

Overall, these findings support one-stage revision combined with antibiotic-loaded DAC hydrogel as a safe and durable strategy for treating chronic knee PJI. Future prospective multicenter comparative studies should determine whether the addition of DAC hydrogel provides a measurable advantage over conventional one-stage revision and identify those patient subgroups that may derive the greatest benefit from this technology.

4. Materials and Methods

This retrospective observational study included consecutive patients who underwent one-stage revision total knee arthroplasty for chronic periprosthetic joint infection (PJI) or infected knee spacer at our institution between February 2014 and September 2023. All procedures were performed using a standardized diagnostic, surgical and postoperative protocol.

All patients provided written informed consent for surgery and for the anonymous use of clinical data for research purposes. The study is reported according to the STROBE statement.

4.1. Patient Selection

Patients were eligible if they fulfilled the diagnostic criteria for PJI according to the Musculoskeletal Infection Society (MSIS) 2018 definition [23] and the World Association against Infection in Orthopaedics and Trauma (WAIOT) [24,40] criteria and underwent one-stage revision using DAC^® antibacterial hydrogel.

Patients undergoing revision for aseptic failure or incomplete follow-up of less than three years (unless recurrent infection occurred earlier) were excluded.

Overall, 52 consecutive patients were included.

4.2. Pre-Operative Evaluation

All patients underwent a standardized diagnostic work-up including:

• detailed medical history and clinical examination;
• plain radiographs;
• laboratory investigations including CRP, ESR and complete blood count;
• joint aspiration whenever feasible;
• microbiological cultures;
• assessment according to the MSIS 2018 and WAIOT diagnostic criteria.

General patient condition was classified according to:

• ASA score [41];
• McPherson host classification; [21]
• Anderson Orthopaedic Research Institute (AORI) classification for bone loss (assessed intraoperatively after implant removal) [22].

4.3. Surgical Technique

All procedures were performed through the previous surgical incision whenever possible.

Following removal of all prosthetic components and cement, extensive radical debridement was performed, including complete excision of infected synovium, scar tissue, necrotic soft tissues and devitalized bone. Multiple tissue specimens (minimum five) were collected from different anatomical sites for microbiological and histological examination before administration of definitive antibiotics.

Bone defects were reconstructed according to the AORI classification using a combination of modular augments, metaphyseal cones, antibiotic-loaded bone cement and long stems whenever required. Structural allografts were not used.

Depending on ligament competence and residual bone stock, patients received either:

• a rotating-hinge revision prosthesis (LINK Endo-Model^®, Waldemar Link, Hamburg, Germany), or
• a semi-constrained TC3 prosthesis (DePuy Synthes^®, Warsaw, IN, USA).

Hybrid fixation was adopted in all cases. Epiphyseal fixation was achieved using commercially available antibiotic-loaded polymethylmethacrylate cement containing gentamicin combined with vancomycin (COPAL^® G+C/V or VancoGenX^®, according to availability), whereas metaphyseal and diaphyseal stems were implanted without cement.

Bone defects were managed using antibiotic-loaded cement, augments, metaphyseal cones and long stems according to defect severity. No allografts were used in any case.

Tourniquet was only applied during cementation. No drains were used.

4.4. DAC Antibacterial Hydrogel

Immediately before implantation, all prosthetic components were coated with a sterile biodegradable hydrogel (DAC^®, Novagenit S.r.l., Mezzolombardo, Italy).

DAC hydrogel was loaded with vancomycin 5% or vancomycin 5% + meropenem 5% in the first ten patients, on the basis of the pathogen isolates. In the remaining 42 patients and in all cases with negative cultural examination the combination vancomycin 5% + meropenem 5% was used. In all the patients a total of 10 to 15 mL of DAC hydrogel was applied both to the uncemented intra-medullary parts of the implant and to the polyethylene insert and to the extra-medullary surface (cf. Figure 2).

4.5. Post-Operative Management

Systemic antibiotic treatment was administered for 31.8 ± 6.3 days on average (range 24–48). All patients received dual antibiotic treatment, based on antibiogram results. During hospital stay and until antibiogram results, a combination of glycopeptide and tazobactam-piperacillin or meropenem were usually administered. At discharge and whenever possible, antibiotic treatment was shifted to oral administration.

Beside antibiotic treatment, all patients received thromboprophylaxis for 5 weeks with low-dose heparin.

Postoperative rehabilitation was initiated on the first postoperative day with immediate mobilization and progressive weight bearing according to implant stability and soft-tissue conditions.

4.6. Follow-Up Examination

Patients were reviewed every 3–6 months during the first postoperative year and annually thereafter.

Clinical assessment included:

• Knee Society Score (KSS) [42];
• Knee Society Functional Score;
• Visual Analogue Scale (VAS) for pain at rest and during movement

Laboratory follow-up included serum CRP.

Standardized anteroposterior and lateral radiographs were obtained at each follow-up visit. Radiographic evaluation assessed component position, radiolucent lines and evidence of aseptic loosening.

The primary endpoint was infection eradication, defined as the absence of clinical, laboratory and radiographic evidence of recurrent infection and no need for additional surgery for infection recurrence.

Secondary endpoints included:

• implant survivorship;
• functional outcome;
• radiographic stability;
• postoperative complications;
• need for further revision surgery

4.7. Statistical Analysis

Continuous variables are reported as mean ± standard deviation (SD) with ranges, whereas categorical variables are expressed as frequencies and percentages.

Changes in clinical outcome scores between preoperative assessment and final follow-up were analyzed using paired Student’s t-tests after confirmation of normal distribution. Statistical significance was set at p < 0.05.

Infection-free implant survivorship was estimated using the Kaplan–Meier method, with recurrent infection considered the endpoint and patients without recurrence censored at their latest follow-up.

Statistical analyses were performed using SPSS version 27 (IBM Corp., Armonk, NY, USA).