The Toxicity of Chlorhexidine and Povidone-Iodine to Rotator Cuff Tendon.

Abstract

Background: Antimicrobial wound lavages containing chlorhexidine gluconate (CHG) or povidone-iodine (PI) are commonly used for infection prophylaxis in arthroplasty surgery, including shoulder arthroplasty. However, the potential toxicity of these solutions to the surrounding native tissue, including rotator cuff tendons is not well understood. The purpose of this study was to investigate the ex vivo effects of a 1-minute 0.05% CHG exposure and a 3-minute 0.35% PI exposure on the viability and biochemical content of rotator cuff tendon explants.

Methods: Infraspinatus tendons (n=6 per group) were isolated from the shoulders of Nubian goats. Tendons were submerged in 0.05% CHG in sterile water for 1 minute or 0.35% PI for 3 minutes per manufacturer guidelines, followed by phosphate-buffered saline (PBS) wash and culture in tissue medium. Control tendon specimens were bathed in PBS solution for 1 minute, followed by PBS wash and culture in tissue medium. Seven days after exposure, isolates from both the tendinous and enthesis portions were analyzed for tenocyte viability, biochemical content (collagen and glycosaminoglycan [GAG]), and matrix histology.

Results: Within the tendinous portion of the tendon, CHG and PI exposure led to reduced mean tenocyte viability compared to controls (4.35% ± 4.64% viability in CHG group, P = 0.045; 14.4% ± 20.3% viability for PI group, P = 0.097; versus 35.4% ± 28.4% viability in control group). Within the enthesis, CHG and PI exposure led to smaller and non-significant decreased mean tenocyte viability compared to controls (29.5% ± 22.9%, P = 0.336 for CHG; 37.5% ± 29.4%, P = 0.327 for PI). There were no significant differences in mean collagen content/wet weight or GAG/wet weight among groups within the tendinous and enthesis isolates, nor were there any appreciable differences on histology in collagen and GAG distribution among groups.

Conclusion: Brief 0.05% CHG and 0.35% PI exposures to rotator cuff tendon may lead to reduced tenocyte viability, with the tendinous portion seemingly more susceptible than the enthesis. Additionally, at these concentrations tested, CHG appears to be more toxic than PI. These findings raise concern for the potential cytotoxic effects of low-dose CHG and PI to native tendon tissue, even after brief exposures commonly used for antimicrobial prophylaxis in shoulder arthroplasty.