Daniel K Hussey1, Rami Madanat1,2, Gabrielle S Donahue1, Ola Rolfson1,2,3, Charles R Bragdon1,2, Orhun K Muratoglu1,2, Henrik Malchau4,2. 1. Harris Orthopaedic Laboratory, Massachusetts General Hospital, Boston, Massachusetts. 2. Department of Orthopaedic Surgery, Harvard Medical School, Boston, Massachusetts. 3. Department of Orthopaedics, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden. 4. Harris Orthopaedic Laboratory, Massachusetts General Hospital, Boston, Massachusetts hmalchau@mgh.harvard.edu.
Abstract
BACKGROUND: In the follow-up evaluation of patients with metal-on-metal (MoM) hip replacements, current evidence suggests that orthopaedic surgeons should avoid reliance on any single investigative tool. Current risk stratification guidelines can be difficult to interpret because they do not provide guidance when there are several risk factors in different groups (high and low risk). To improve the clinical utility of risk stratification guidelines, we designed a scoring system to assess the risk of revision. METHODS: The study population consisted of 1,709 patients (1,912 hips) enrolled in a multicenter follow-up study of a recalled MoM hip replacement. Eleven scoring criteria were determined on the basis of existing follow-up algorithm recommendations and consisted of patient-related factors, symptoms, clinical status, implant type, metal ion levels, and radiographic imaging results. Forward stepwise logistic regression was conducted to determine the minimum set of predictive variables for the risk of revision and to assign variable weights. The MoM risk score for each hip was then created by averaging the weighted values of each predictive variable. RESULTS: Receiver operating characteristic curve analysis yielded good discrimination between all revised and unrevised hips, with an area under the curve of 0.82 (p < 0.001). The odds of revision for the group with a high MoM risk score were increased by 5.8-fold (95% confidence interval [CI], 3.1 to 11.0) relative to the moderate risk group and by 21.8-fold (95% CI, 9.9 to 48.0) compared with the low risk group. CONCLUSIONS: Although the use of MoM hip arthroplasty has been limited since 2010, we continue to be faced with the follow-up and risk assessment of thousands of patients who have not had a revision. As more knowledge about risk stratification is gained, the complexity of the algorithms is expected to increase. We propose the use of the MoM risk score as a tool to aid in the clinical decision-making process. LEVEL OF EVIDENCE: Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.
BACKGROUND: In the follow-up evaluation of patients with metal-on-metal (MoM) hip replacements, current evidence suggests that orthopaedic surgeons should avoid reliance on any single investigative tool. Current risk stratification guidelines can be difficult to interpret because they do not provide guidance when there are several risk factors in different groups (high and low risk). To improve the clinical utility of risk stratification guidelines, we designed a scoring system to assess the risk of revision. METHODS: The study population consisted of 1,709 patients (1,912 hips) enrolled in a multicenter follow-up study of a recalled MoM hip replacement. Eleven scoring criteria were determined on the basis of existing follow-up algorithm recommendations and consisted of patient-related factors, symptoms, clinical status, implant type, metal ion levels, and radiographic imaging results. Forward stepwise logistic regression was conducted to determine the minimum set of predictive variables for the risk of revision and to assign variable weights. The MoM risk score for each hip was then created by averaging the weighted values of each predictive variable. RESULTS: Receiver operating characteristic curve analysis yielded good discrimination between all revised and unrevised hips, with an area under the curve of 0.82 (p < 0.001). The odds of revision for the group with a high MoM risk score were increased by 5.8-fold (95% confidence interval [CI], 3.1 to 11.0) relative to the moderate risk group and by 21.8-fold (95% CI, 9.9 to 48.0) compared with the low risk group. CONCLUSIONS: Although the use of MoM hip arthroplasty has been limited since 2010, we continue to be faced with the follow-up and risk assessment of thousands of patients who have not had a revision. As more knowledge about risk stratification is gained, the complexity of the algorithms is expected to increase. We propose the use of the MoM risk score as a tool to aid in the clinical decision-making process. LEVEL OF EVIDENCE: Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.
Authors: Gulraj S Matharu; Fiona Berryman; David J Dunlop; Andrew Judge; David W Murray; Hemant G Pandit Journal: Acta Orthop Date: 2019-09-09 Impact factor: 3.717