Year: 2021
Type: peer-review
Abstract: We appreciate the opportunity to respond to the Letter to the Editor. Oatis and colleagues contend that “all three of the trials used in this analysis suffer from a common deficiency, a lack of detail in the content, intensity and dosage of the physical therapy interventions, whether experimental or usual care.” Based on a prior publication,1 Oatis et al claimed that a greater number of exercise progressions, particularly in a weight-bearing position, and a greater total number of physical therapy exercises were associated with better outcomes following knee arthroplasty. We applaud efforts to identify factors associated with less-than-optimal outcomes and subsequently develop interventions for the key ingredient(s) that drive rehabilitation outcomes following knee arthroplasty (KA). Although we tend to take a more pragmatic approach and our trials tested interventions developed specifically for individuals at risk for poor outcome, our Perspective paper did not set out to provide a detailed accounting of the interventions that we tested. Our purpose was to “compare and contrast our three clinical trials of enhanced physical therapy designed specifically for the poor outcome phenotype, to suggest factors that might explain the no-effect findings, and to propose areas for future research directed toward improving outcome for the clinically challenging and relatively common KA poor outcome phenotype.” Detailed descriptions of the interventions and their rationale are provided elsewhere2–5 and were referenced in our paper. Importantly, Oatis and colleagues claim in their letter that their prior work1 demonstrated that exercise type and dosage impacts outcome in individuals following knee arthroplasty. We examined the Oatis et al paper.1 It is a retrospective observational cohort study of 112 patients who underwent knee arthroplasty. The purpose was, in part, to identify associations between specific details of the physical therapist-led exercise and 6-month postoperative self-reported and performance outcomes. Several study characteristics raise substantial credibility concerns. The study was retrospective, data for 38% of the sample were missing, and conclusions were based on 102 linear regression analyses, of which 11 comparisons were statistically significant. With N = 112, the capitalization on chance significance is a serious threat to attempts to gain scientific knowledge from 102 linear regression analyses. In addition to the retrospective nature of the study, the multiple comparisons, and substantial missing data, it also is likely that residual confounding and confounding by indication could explain the significant findings reported by Oatis and colleagues. They contended, for example, that patients who had a higher number of weight-bearing exercise progressions (presumably with greater intensity) had better knee flexion ROM. Observational studies of the effects of interventions are vulnerable to unmeasured confounding and confounding by indication.6 In our view, given the observational design and the limitations noted, it is just as likely that a better knee ROM outcome was attributable to unmeasured confounding and healthier participants receiving more intense exercise over time, and not exercise volume or rate of exercise progression. The study by Oatis and colleagues does not, in our view, demonstrate that more descriptive detail on type and intensity of exercise, or that more rapid progression of exercise, will lead to better outcomes. Only well-designed clinical trials can identify definitive cause-effect relationships among interventions and outcomes. We were surprised that Oatis and colleagues did not reference knee arthroplasty clinical trials that examined effects of high intensity postoperative exercise versus more commonly used lower intensity exercise on performance and self-reported outcomes.7,8 These higher forms of evidence were no-effect trials and suggest that more intense exercise is no better than usual care at improving outcome following knee arthroplasty. We found in our KASTPain trial4 that more exercise also is not better and in some cases is likely to be worse9 and more costly.10
Source: Physical Therapy
DOI: 10.1093/ptj/pzab277
Language: English
Subfield: Surgery
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W4205596291