This study explored differences in the early pattern of recovery for self-report (pain and physical function subscales of the Western Ontario and McMaster Universities Osteoarthritis Index) and physical performance measures (six minute walk and timed up and go test) in patients following arthroplasty. Using hierarchical linear modeling, different patterns of recovery and predictors of change were observed. The physical function subscale did not detect the early deterioration in physical function that was detected by the performance measures. Different important clinical information can be learned from performance measures, supporting the use of both types of measures when monitoring decline and recovery.

To explore differences in the pattern of recovery for self-report and physical performance measures in patients following total hip (THA) and knee (TKA) arthroplasty.

Different patterns of recovery and predictors of change were observed for the pain and physical function subscales of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and the performance measures. The physical function subscale did not detect the early deterioration in physical function demonstrated by the six minute walk (6MWT) and timed up and go (TUG) tests.

WOMAC scores are commonly used to monitor changes in pain and physical function both pre and post-operatively. Failure of the physical function sub-scale to detect decline in physical function experienced post-operatively suggests that the WOMAC may not always accurately reflect physical function.

The postoperative predicted scores for the WOMAC either exceeded or met the preoperative scores within one to two weeks compared to seven to eight weeks with the performance measures. The performance measure models contained a greater number of predictors than the WOMAC subscale models.

One hundred and fifty-two patients (mean age 63.8 ± 10.2 years) who underwent THA and TKA were assessed at several points over the first four postoperative months. Average growth curves for the WOMAC, 6MWT and TUG were characterized using hierarchical linear modeling. Predictors of recovery were sequentially modeled after validation of the basic developmental models.

Differences in models and the time point at which preoperative scores are met suggest that different information is being learned from self-report and physical performance measures.

Funding: This research was not supported by funding from a commercial party. It was supported by a grant from the Orthopaedic & Arthritic Foundation.

Introduction and Aims: A growing number of studies suggest worse pre-operative functional status is associated with worse outcome in terms of both function and pain after total hip (THA) and knee (TKA) arthroplasty. A gap in these studies, however, is the lack of physical performance measures to validate the self-reported findings.

Method: Using a repeated measures design, 152 subjects with endstage osteoarthritis were assessed at multiple points over the first four post-operative months. Outcome measures included the six-minute walk test (6MWT), timed up and go test (TUG), and the pain and physical function subscales of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). Average growth curves and individual variations in the pattern were characterised using hierarchical linear modelling. Following validation of the basic developmental models, predictors of recovery were sequentially modelled. Gender, baseline pre-operative function, site of replacement, age, additional outpatient rehabilitation, and number of co-morbidities were investigated as predictors.

Results: In the models explored, a second-degree polynomial growth term provided a reasonable fit for the data over the study interval. Baseline pre-operative score was a significant predictor (p< .001) in all of the models. A significant interaction (p< .0005) was observed between the growth parameter (weeks post-surgery) and baseline pre-operative score for the 6MWT. Patients with better pre-operative scores are predicted to improve their 6MWT distances faster. Although gender was a significant predictor (p< .0005) of TUG and 6MWT performance at one week post-operatively (women slower), it was not a predictor in the WOMAC models. No significant interaction was observed between the growth parameter and gender in either of the 6MWT or TUG models. Site of replacement (hip or knee) was a significant predictor (p< .001) of one-week scores in all of the models except for the WOMAC physical function subscale. In both performance measure models, the average growth curves demonstrate that patients post-THA are predicted to start with worse function, however, their growth rates are faster over the studied interval. Co-morbidities, age and additional outpatient rehabilitation were not explanatory variables for any of the models.

Conclusions: Performance measurement demonstrated that women started with worse function than men one week after surgery, but thereafter had similar rates of improvement. The importance of pre-operative function as a predictor of post-operative recovery cannot be overlooked. Patient and surgeon’s expectations of outcome need to take pre-operative function into account.

Introduction and Aims: With the current trend to minimally invasive techniques for total hip (THA) and knee arthroplasty (TKA), an understanding of early functional recovery for traditional surgical approaches is required for outcome comparison. Patterns of recovery for self-report and physical performance measures were therefore explored during the early post-operative period.

Method: One hundred and fifty-two patients were assessed pre-operatively and several times over the first four post-operative months. The pain and physical function subscales of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), the six-minute walk (6MWT) and timed up and go test (TUG) were used as outcome measures. Hierarchical linear modelling was used to characterise the average pattern of recovery for each measure. Model development began with a parameter that estimated the patients’ self-reported or actual measured function at one week postoperatively and a parameter that estimated the patients’ rate of change for every week thereafter.

Results: Sixty-nine subjects underwent THA and 83 underwent TKA with a mean age of 63.8 + 10.2 years. Different patterns of recovery and predictors of change were observed for the WOMAC subscales and physical performance measures. The growth models for the performance measures were more complex, including more predictors and interaction terms. The model for the physical function subscale was the simplest. In addition to the growth parameters and the intercept, baseline function was the only other model parameter influencing the estimated score at one week. A striking difference between the WOMAC and performance measure growth curves concerned the point at which pre-operative scores are predicted to be met. The predicted scores for the WOMAC subscales either reach or exceed the pre-operative scores at one to two weeks post-surgery in comparison to the 6MWT and TUG, which do not reach these levels until six to eight weeks post-surgery. It would appear that the physical function subscale of the WOMAC does not reflect the early deterioration that occurs in physical function. The predicted growth curves for the TUG confirm its usefulness as a physical function measure only in the early recovery period, as a ceiling effect occurred around 10 weeks.

Conclusion: The physical function subscale of the WOMAC may not always accurately reflect physical function. Using only self-report measures to compare traditional surgical approaches to minimally invasive techniques might miss significant differences in recovery of physical function. Using both physical performance and self-report measures to monitor early recovery is recommended.