< Back to previous page

Publication

Preoperative Planning of Shoulder Arthroplasty Based on Patient-Specific Computer Models

Book - Dissertation

Shoulder arthroplasty is the third most common joint replacement and is indicated for patients with severe glenohumeral osteoarthritis or cuff tear arthropathy. Anatomic or total shoulder arthroplasty (TSA) and reverse shoulder arthroplasty (RSA) have a complication rate of respectively 10 and 16%. In TSA, the joint anatomy is restored, which is challenging due to the unknown shape of the native unaffected glenoid. In RSA, the joint anatomy is reversed, without clear guidelines for implant design and position. For both types, the design and position of the glenoid and humeral implant components have a significant impact on the postoperative outcome. To limit complications and improve the functional patient outcome, it is important to make a suitable preoperative plan for every patient. The goal of this thesis is to develop patient-specific models to support surgeons in selecting a suitable implant design and position for each patient. More particularly, this research work focuses on two aspects of preoperative planning: glenoid bone analysis and preoperative plan evaluation. Glenoid bone analysis covers the assessment of the shape and severity of a patient's glenoid bone defect to select the preferred treatment option. Since an accurate assessment is difficult due to the unknown shape of the native glenoid, statistical shape modeling technology is applied and evaluated to virtually reconstruct the native glenoid and to quantify the glenoid bone defect by comparing the diseased and native glenoid. The defect quantification can assist surgeons in selecting a suitable glenoid implant design for each patient and allows for objective selection and comparison of patients with similar defects. The virtual reconstruction of the native glenoid shape supports surgeons to restore the joint anatomy during TSA planning. Compared to alternative methods using the healthy contralateral bone as a template, the presented statistical shape model-based methods show accurate results for reconstructing the native glenoid surface, predicting relevant anatomic parameters and quantifying the glenoid bone defect, without the need for a healthy contralateral scapula. Furthermore, the method is fully automated to facilitate integration into clinical practice. Evaluation of a preoperative plan may result in fewer complications and an improved functional patient outcome. This is particularly the case for RSA, where many uncertainties about adequate implant selection and positioning remain. Therefore, patient-specific evaluation methods are developed and evaluated for use in clinical practice. First, an impingement-free range-of-motion (ROM) evaluation method is established to reduce the risk for impingement and notching, which are common complications for RSA patients. Second, a method to measure muscle elongations is developed and evaluated, to support surgeons in assessing muscle tensions and joint stability. Sensitivity analyses show the effect of implant positioning on impingement-free ROM and muscle elongations, which are in line with other clinical results found in literature. As a conclusion, the patient-specific evaluation methods can support surgeons to compare different implant designs and positions during preoperative planning of shoulder arthroplasty. The main contribution of this work is the application of state-of-the-art modeling techniques in clinical practice. The presented patient-specific models enable surgeons to make a more thoughtful and objective decision on implant design and position. Therefore, we believe that this work is an important step towards the ultimate dream of having an optimal preoperative plan for each patient, that limits complications and maximizes the postoperative functional patient outcome.
Publication year:2020
Accessibility:Closed