Background: The reported outcomes of patients who underwent total or unicompartmental knee arthroplasty for secondary and spontaneous osteonecrosis of the knee are often from studies that lack the number of subjects necessary to generate meaningful conclusions. We systematically reviewed the available literature in order to define the outcomes of patients after total knee arthroplasty for secondary osteonecrosis and after total or unicompartmental knee arthroplasty for spontaneous osteonecrosis of the knee.
Methods: A literature review yielded twenty cohorts with demographic patient information and outcome scores (global knee scores, radiographic outcomes, and revision rates) for patients who had knee arthroplasty as treatment for either secondary or spontaneous osteonecrosis of the knee. The mean preoperative and postoperative global knee scores, the mean revision rate, and the categorization of the mean "poor" and mean "good" outcomes for the knees with each underlying disease were tabulated and reported. The demographic data and the reported mean global knee scores were weighted by the number of knees in each study.
Results: Total knee arthroplasty was performed in 150 knees with secondary osteonecrosis and 148 knees with spontaneous osteonecrosis, and unicompartmental knee arthroplasty was performed in sixty-four knees with spontaneous osteonecrosis. Total knee arthroplasty for spontaneous osteonecrosis of the knee was associated with the best outcomes (higher "good" and postoperative global knee scores and lower revision [3%] and "poor" outcome [6%] rates compared with the other two groups). The outcomes after total knee arthroplasty in knees with secondary osteonecrosis as well as in knees with spontaneous osteonecrosis were better in the cohorts operated on during or after 1985 than in those operated on before 1985. Similarly, the outcomes after unicompartmental knee arthroplasty in knees with spontaneous osteonecrosis of the knee were also better in the cohorts operated on during or after 1985 than in those operated on before 1985.
Conclusions: Total knee arthroplasty performed as treatment for either secondary osteonecrosis or spontaneous osteonecrosis and unicompartmental knee arthroplasty performed as treatment for spontaneous osteonecrosis were associated with improved outcomes in cohorts with more recent operative dates. The evidence suggests that the use of contemporary cemented implants in total knee arthroplasty and the selective use of stems and augments in patients who have development of secondary osteonecrosis after total knee arthroplasty are producing outcomes that are comparable to those seen after total knee arthroplasty for osteoarthritis. Although the outcomes of patients who have total knee arthroplasty for the treatment of spontaneous osteonecrosis of the knee have historically been favorable, such outcomes have also shown particular improvement in the studies from more recent operative periods. Although poor outcomes were seen after unicompartmental knee arthroplasty in earlier studies of patients with spontaneous osteonecrosis of the knee, it is possible that those results were secondary to inappropriate patient selection, as the authors of the most recent and, to our knowledge, the only study to follow established operative indications regarding the use of unicompartmental knee arthroplasty reported excellent results.
Level of evidence: Therapeutic Level IV. See Instructions to Authors on jbjs.org for a complete description of levels of evidence.