Precise implant positioning, tailored to individual spinopelvic biomechanics and phenotype, is paramount for stability in total hip arthroplasty (THA). Despite a few studies on instability prediction, there is a notable gap in research utilizing artificial intelligence (AI). The objective of our pilot study was to evaluate the feasibility of developing an AI algorithm tailored to individual spinopelvic mechanics and patient phenotype for predicting impingement. This international, multicentre prospective cohort study across two centres encompassed 157 adults undergoing primary robotic arm-assisted THA. Impingement during specific flexion and extension stances was identified using the virtual range of motion (ROM) tool of the robotic software. The primary AI model, the Light Gradient-Boosting Machine (LGBM), used tabular data to predict impingement presence, direction (flexion or extension), and type. A secondary model integrating tabular data with plain anteroposterior pelvis radiographs was evaluated to assess for any potential enhancement in prediction accuracy.Aims
Methods
In-hospital length of stay (LOS) and discharge disposition following arthroplasty could act as surrogate measures for improvement in patient pathways, and have major cost saving implications for healthcare providers. With the ever-growing adoption of robotic technology in arthroplasty, we wished to evaluate its impact on LOS. The objectives of this study were to compare LOS and discharge disposition following robotic-arm assisted (RO THA) versus conventional technique Total Hip Arthroplasty (CO THA). This large-scale, single institution study included patients of any age undergoing primary THA (N = 1,732) for any cause between May 2019 and January 2023. Data extracted included patient demographics, LOS, need for Post Anaesthesia Care Unit (PACU) admission, anaesthesia type, readmission within 30 days and discharge dispositions. Univariate and multivariate logistic regression models were also employed to identify factors and patient characteristics related to delayed discharge. The median LOS in the RO THA group was 54 hours (34, 78) versus 60 (51, 100) in the CO THA group, p<0.001. Discharge disposition was comparable between the two groups. In the multivariate model, age, need for PACU admission, ASA score > 2, female gender, general anaesthesia and utilisation of the conventional technique were significantly associated with LOS > 2 days. Our study showed that robotic-arm assistance was associated with a shorter LOS in patients undergoing primary THA and no difference in discharge destination. Our results suggest that robotic-arm assistance could be advantageous in partly addressing the upsurge of hip arthroplasty procedures and the concomitant health care burden; however, this needs to be corroborated by long-term cost effectiveness analyses and data from randomised controlled studies.