Clinical decision support tools are software that match the input characteristics of an individual patient to an established knowledge base to create patient-specific assessments that support and better inform individualized healthcare decisions. Clinical decision support tools can facilitate better evidence-based care and offer the potential for improved treatment quality and selection, shared decision making, while also standardizing patient expectations. Predict+ is a novel, clinical decision support tool that leverages clinical data from the Exactech Equinoxe shoulder clinical outcomes database, which is composed of >11,000 shoulder arthroplasty patients using one specific implant type from more than 30 different clinical sites using standardized forms. Predict+ utilizes multiple coordinated and locked supervised machine learning algorithms to make patient-specific predictions of 7 outcome measures at multiple postoperative timepoints (from 3 months to 7 years after surgery) using as few as 19 preoperative inputs. Predict+ algorithms predictive accuracy for the 7 clinical outcome measures for each of aTSA and rTSA were quantified using the mean absolute error and the area under the receiver operating curve (AUROC).Introduction
Methods
Essex-Lopresti injuries are often unrecognized acutely with resulting debilitating adverse effects. Persistent axial forearm instability may affect load transmission at both the elbow and wrist, resulting in significant pain. In the setting of both acute and chronic injuries metallic radial head arthroplasty has been advocated, however there is little information regarding their outcome. The purpose of this study was to assess the efficacy of a radial head arthroplasty to address both acute and chronic Essex-Lopresti type injuries. A retrospective review from 2006 to 2016 identified 11 Essex-Lopresti type injuries at a mean follow-up of 18 months. Five were diagnosed and treated acutely at a mean of 11 days (range, 8 to 19 days) from injury, while 6 were treated in a delayed fashion at a mean of 1.9 years (range, 2.7 months to 6.2 years) from injury with a mean 1.5 (range, 0 to 4) prior procedures. The cohort included 10 males with a mean age was 44.5 years (range, 28 to 71 years). A smooth stem, modular radial head arthroplasty was used in all cases. Outcomes included range of motion and radiographic findings such as ulnar variance, capitellar erosion, implant positioning and implant lucency using a modification of the method described by Gruen. Reoperations, including the need for ulnar shortening osteotomy, were also recorded. Three patients in each group (55%) reported persistent wrist pain. The mean ulnar variance improved from +5 mm (range, 1.8 to 7 mm) to +3.7 mm (range, 1 to 6.3 mm) at the time of final follow-up or prior to reoperation. Three (50%) patients in the chronic group underwent a staged ulnar shortening osteotomy (USO) to correct residual ulnar positive variance and to manage residual wrist pain. There were no reoperations in the acute group. Following USO, the ulnar variance in those three cases improved further to +3.5, +2.1, and −1.1 mm. No radial head prostheses required removal. Capitellar erosion was noted in five (45%) elbows, and was rated severe in one, moderate in two, and mild in two. Lucency about the radial head prosthesis stem was noted in eight (73%) cases, and rated as severe in 2 (18%), based on Gruen zones. Treatment of acute and chronic Essex-Lopresti lesions with radial head arthroplasty often results in persistent wrist pain. In the chronic setting, a planned USO was often necessary to restore axial forearm stability after radial head arthroplasty. Essex-Lopresti lesions represent a rare clinical entity that are difficult treat, particularly in the chronic setting. Early recognition and management with a smooth stem modular radial head arthroplasty may provide improved outcomes compared to chronic reconstruction.
