A study based on data derived from 602, 733 osteoarthritis (OA) patients enrolled in the UK’s Health Improvement Network during 2000-2012 suggested that patients who had a knee replacement (KR) for osteoarthritis had a 28% lower risk of mortality compared with non-KR recipients.
The participants ranged in age from 50 to 89 and 57% were female. Their mean body mass index was 29.2, and mean follow-up was 4.42 years for KR patients and 4.31 years for their non-KR counterparts.
Because knee replacement can relieve pain in patients’ knees, and allow for greater physical activity the assumption has been that enhanced mobility may translate into a decreased risk of long-term mortality.
According to writer Diana Swift, who wrote about the study for MedPage Today, with stratification by age the protective effect on all-cause mortality emerged only in those older than 63, and may have been overestimated.
Devyani Misra, M.D., of Boston University School of Medicine, suggested that the mortality benefit effect seen in older patients may be subject to confounding by selection in which good candidates have the procedure and sicker patients do not: “If you have two 80-year-old patients, one frail and one robust, you would select the robust patient to undergo surgery, ” she pointed out.
There were 1, 159 deaths in the KR group and 1, 418 in the non-KR group.
While the mortality rate among KR recipients decreased as candidacy for KR increased, no similar consistent trend emerged for non-KR recipients.
“While a protective effect of KR on mortality cannot be ruled out, findings of lower mortality among older KR subjects and those with higher propensity scores suggest that prognosis-based selection for KR may lead to intractable confounding by indication, ” Misra and her associates wrote in their evaluation of the study published in Annals of the Rheumatic Diseases.
Other similar studies report conflicting results. A large study using data from the Swedish Knee Arthroplasty Registry found a lower standardized mortality rate with KR and a German study identified no KR-related difference in mortality risk compared with the general population.
