Aims. The use of technology to assess balance and alignment during total knee surgery can provide an overload of numerical data to the surgeon. Meanwhile, this quantification holds the potential to clarify and guide the surgeon through the surgical decision process when selecting the appropriate bone recut or soft tissue adjustment when balancing a total knee. Therefore, this paper evaluates the potential of deploying supervised machine learning (ML) models to select a surgical correction based on patient-specific intra-operative assessments. Methods. Based on a clinical series of 479 primary total knees and 1,305 associated surgical decisions, various ML models were developed. These models identified the indicated surgical decision based on available, intra-operative alignment, and tibiofemoral load data. Results. With an associated area under the receiver-operator curve ranging between 0.75 and 0.98, the optimized ML models resulted in good to excellent predictions. The best performing model used a random forest approach while considering both alignment and intra-articular load readings. Conclusion. The presented model has the potential to make experience available to surgeons adopting new technology, bringing expert opinion in their operating theatre, but also provides insight in the surgical decision process. More specifically, these promising outcomes indicated the relevance of considering the overall limb alignment in the coronal and sagittal plane to identify the appropriate surgical decision

Aims. Currently, there is no single, comprehensive national guideline for analgesic strategies for total joint replacement. We compared inpatient and outpatient opioid requirements following total hip arthroplasty (THA) versus total knee arthroplasty (TKA) in order to determine risk factors for increased inpatient and outpatient opioid requirements following total hip or knee arthroplasty. Methods. Outcomes after 92 primary total knee (n = 49) and hip (n = 43) arthroplasties were analyzed. Patients with repeat surgery within 90 days were excluded. Opioid use was recorded while inpatient and 90 days postoperatively. Outcomes included total opioid use, refills, use beyond 90 days, and unplanned clinical encounters for uncontrolled pain. Multivariate modelling determined the effect of surgery, regional nerve block (RNB) or neuraxial anesthesia (NA), and non-opioid medications after adjusting for demographics, ength of stay, and baseline opioid use. Results. TKAs had higher daily inpatient opioid use than THAs (in 5 mg oxycodone pill equivalents: median 12.0 vs 7.0; p < 0.001), and greater 90 day use (median 224.0 vs 100.5; p < 0.001). Opioid refills were more likely in TKA (84% vs 33%; p < 0.001). Patient who underwent TKA had higher independent risk of opioid use beyond 90 days than THA (adjusted OR 7.64; 95% SE 1.23 to 47.5; p = 0.01). Inpatient opioid use 24 hours before discharge was the strongest independent predictor of 90-day opioid use (p < 0.001). Surgical procedure, demographics, and baseline opioid use have greater influence on in/outpatient opioid demand than RNB, NA, or non-opioid analgesics. Conclusion. Opioid use following TKA and THA is most strongly predicted by surgical and patient factors. TKA was associated with higher postoperative opioid requirements than THA. RNB and NA did not diminish total inpatient or 90-day postoperative opioid consumption. The use of acetaminophen, gabapentin, or NSAIDs did not significantly alter inpatient opioid requirements. Cite this article: Bone Joint Open 2020;1-7:398–404