Comparison of Surgical Site Complications With Negative Pressure Wound Therapy vs Silver Impregnated Dressing in High-Risk Total Knee Arthroplasty Patients: A Matched Cohort Study

Abstract

BackgroundClosed incision negative pressure wound therapy (ciNPWT) may reduce surgical site complications following total joint arthroplasty. Although unlikely necessary for all patients, the criteria for utilizing ciNPWT in primary total knee arthroplasty (TKA) remain poorly defined. This study’s purpose was to compare the incidence of incisional wound complications, non-incisional complications (ie, dressing reactions), reoperations, and periprosthetic joint infections (PJIs) among a group of high-risk primary TKA patients treated with ciNPWT vs an occlusive silver impregnated dressing. MethodsOne hundred thirty high-risk primary TKA patients treated with ciNPWT were 1:1 propensity matched and compared to a historical control group treated with an occlusive silver impregnated dressing. High-risk criteria included the following: active tobacco use, diabetes mellitus, body mass index >35 kg/m2, autoimmune disease, chronic kidney disease, Staphylococcus aureus nasal colonization, and non-aspirin anticoagulation. ResultsAge, gender, and risk factor profile were comparable between cohorts. The ciNPWT cohort had significantly fewer incisional wound complications (6.9% vs 16.2%; P = .031) and significantly more non-incisional complications (16.9% vs 1.5%; P < .001). No dressing reactions required clinical intervention. There were no differences in reoperations or periprosthetic joint infections (P = 1.000). In multivariate analysis, occlusive silver impregnated dressings (odds ratio 2.9, 95% confidence interval 1.3-6.8, P = .012) and non-aspirin anticoagulation (odds ratio 2.5, 95% confidence interval 1.1-5.6, P = .028) were associated with the development of incisional wound complications. ConclusionAmong high-risk patients undergoing primary TKA, ciNPWT decreased incisional wound complications when compared to occlusive silver impregnated dressings, particularly among those receiving non-aspirin anticoagulation. Although an increase in dressing reactions was observed, the clinical impact was minimal.