1778 - Postoperative Pain and Opioid Usage in Patients Undergoing Robotic-Assisted Total Knee Arthroplasty (TKA) versus Conventional TKA

INTRODUCTION: Degenerative joint diseases, such as osteoarthritis (OA), are characterized by progressive deterioration of cartilage; the thin, connective tissue covering bones in articular joints that is uniquely adapted for load bearing. In OA, cartilage is compromised by acute or chronic insults, including joint injury, obesity, and age [1]. OA can be painful and debilitating and may eventually require surgery for symptom relief. Techniques for partially replacing articular surfaces, including osteochondral allografts [2], tissue engineered cartilage [3], and contoured metal inlays [4], are aimed at treating large focal cartilage lesions in younger patients for whom non-surgical approaches failed. Favorable outcomes are associated with placement of these materials in ways that maintain joint congruity [5]. In our laboratory, we observed that metal implants placed in the weight bearing regions of ovine femoral condyles that protrude above adjacent cartilage resulted in significant damage to the opposing tibial plateau and meniscus in as little as 4 weeks (unpublished data). In this study, we aimed to 1) develop an in vitro model to approximate tibial plateau cartilage lesions observed in vivo, and 2) to quantify changes in pressure distribution on the tibial plateau due to proud placement of implants over time. METHODS: Both knee (stifle) joints were obtained from a 4-year-old, 75 kg sheep and stored at -20u00baC until testing. Surrounding musculature was removed but cruciate and collateral ligaments left intact to maintain joint stability. 6.2 mm diameter titanium implants with curved surfaces (radius of curvature of 35 mm) were press-fit into the weight bearing regions of both femoral condyles. Each knee received implants aligned 1 mm proud (medial condyle) and flush (lateral condyle) relative to adjacent articular surfaces. Joint explants were subjected to cyclic loading (Instron ElectroPuls) in increments of 1000 cycles at 1 Hz with an amplitude producing flexion angles between 50u00ba and 70u00ba [5]. Thin, flexible, pressure sensors (Tekscan K-scan, Sensor 4000) were inserted beneath the menisci after extensive preparation to optimize accuracy [6]. Pressure measurements were obtained at 50u00ba flexion during a sequence of ten cycles where displacement was applied at 2 mm/s, held for 5 seconds, and unloaded at 2 mm/s. Sequential measurements were made of native cartilage, immediately after implant placement, and following 1000, 2000, and 3000 flexion-extension cycles. Average and peak pressures were calculated for the last three cycles over a 40 mm2 area covering the implant (30.2 mm2) and adjacent cartilage. Comparisons were made with a one-way ANOVA and Tukeyu2019s post hoc (Statistica v.8).RESULTS: Pressure distributions on the tibial plateau, in areas covered and uncovered by meniscus, became progressively elevated as the number of flexion-extension cycles increased in the presence of a proud implant on the opposing femoral condyle (Fig. 1). Average pressures observed in the joint compartment containing the proud implant increased significantly (p<0.0002) to approximately double those measured in native cartilage (Fig. 2A). Peak pressures exceeded 6 MPa after cyclic flexion-extension compared to 1.79 u00b1 0.03 (n=3) measured in native cartilage (p<0.0002), while flush implants remained relatively stable at approximately 2 MPa (Fig. 2B). India ink staining revealed localized cartilage fibrillation and softened cartilage on the tibial plateau directly opposing proud implants. The tibial plateau opposing flush implants exhibited pristine cartilage with no India ink uptake. There was no evidence of meniscal fraying.DISCUSSION: Proud placement of metal implants, 1 mm above the surrounding articular surface, produced progressive cartilage damage on the opposing tibial plateau following cyclic flexion-extension (Fig.1). Peak pressures due to proud positioning were comparable to those measured at a similar flexion angle in human cadavers [8], although the change relative to native cartilage was greater in the present study. Implants flush with the articular surface approximately recapitulated pressures observed in native cartilage (Fig. 2) with no visible cartilage damage observed after 3000 cycles of flexion-extension. These findings are consistent with macroscopic and microscopic assessments of tibial plateau cartilage 12 weeks after focal resurfacing of femoral condyles with a metallic implant [9]. Loads experienced during flexion-extension cycles were sub-physiological [5], suggesting observed cartilage damage is an underestimate of that experienced in in vivo sheep models as they weight bear in the immediate post-operative period. We successfully modelled cartilage lesions observed on tibial plateau in as little as 4 weeks post-operatively (unpublished) and demonstrated that these lesions develop even in a low-load environment. These data support recommendations that materials used to repair focal cartilage lesions must never protrude above the level of native articular cartilage. SIGNIFICANCE/CLINICAL RELEVANCE: This study revealed that significant cartilage damage could be produced by proud implants in relatively few cycles of flexion-extension at sub-physiological loads and thereby emphasizes that precise placement of materials by surgeons is critical for successful repair of large chondral lesions. REFERENCES: [1] Scott & Athanasiou (2006) Crit Rev Biomed Eng 34(5):347. [2] De Caro (2015) Arthrosc 31:757. [3]Lee (2017) Acta Biomater 62:352. [4] Brennan (2013) Bone Joint J 95-B(3):301. [5] El-Rashidy (2011) J Bone Joint Surg 93(17):1634-40. [6]Taylor (2006) J Biomech 39:791. [7] Herregodts (2015) Sustain. Constr. Des. 6(2):8. [8] Becher (2008) Knee Surg Sports Traumatol Arthrosc 16:56. [9] Martinez-Carranza (2013) Osteo Cartil 21:739-45. ACKNOWLEDGEMENTS: Funding provided by DePuy Synthes Canada Ltd.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.