Deficits in functional performance and gait one year after total knee arthroplasty despite improved self-reported function

The current literature lacks sufficient information about the degree of functional improvement after total knee arthroplasty (TKA) using objectively measured function including gait patterns [21]. Patients with knee osteoarthritis report decreased pain and improved function following TKA surgery [15, 19], and research indicates patient satisfaction correlates well with both self-reported outcomes and a clinician-based score [2]. Mean improvement in self-reported pain and function is reported to range from 50 to 168 % at 6 months after TKA [35]. At 1 year after TKA, improvements in self-reported pain and function range from 52 to 194 %, where the largest improvements were found in a functional subscale, measuring function in sport and recreation [33]. Self-reported measures of function have been found to be largely influenced by pain, and if perceived pain is greatly reduced, the assessment of function may be overestimated or confused with improved function rather than reduced pain [3].

Despite these improvements, 20–30 % of patients report persistent disability, limited function, reduced quality of life, diminished working capacity, and gait deviations after TKA [24, 38]. Gait pattern in patients operated with unicompartmental knee arthroplasty resembles healthy controls more than patients operated with TKA [37]. In patients with knee osteoarthritis, joint replacements are generally considered successful surgeries. Traditionally, these surgeries were performed on older patients with lower functional demands. Thus, functional improvement has not been considered as important as pain relief. Disability associated with TKA may be conceptualized as a surgical failure since the indication for surgery is pain and impaired function. In clinical practice, the array of available functional outcome measures is usually restricted to self-reported measures and performance-based measures as they do not require special equipment. Detailed knowledge of the relationships between objective measures of function derived from three-dimensional (3D) gait analysis and performance-based function would be advantageous when determining which outcome measures to use in clinical practice.

This study aimed to objectively quantify functional improvements in gait patterns and performance-based function 1 year after TKA in patients with knee osteoarthritis. Further, we aimed to explore the relationships between the degree of improvement in gait pattern, with respect to kinematics and kinetics, in performance-based function using the Five Times Sit-to-Stand (5STS) test, and in self-reported function using a disease-specific questionnaire. In this study, it was hypothesized that (1) patients with knee osteoarthritis would significantly reduce the degree of gait pattern deviation and (2) improve performance on the 5STS 1 year after TKA. Further it was hypothesized that patients with knee osteoarthritis, operated with TKA, would nevertheless not reach the level of healthy controls with regard to gait patterns and performance on the 5STS 1 year after surgery.

Forty patients with physician diagnosed primary knee osteoarthritis were included in this prospective cohort study. Patients were recruited from two orthopaedic departments in Stockholm, Sweden (Ortho center, Löwenströmska hospital and Karolinska University Hospital). The inclusion criteria were: being scheduled for TKA within one month; ability to walk 10 m repeatedly without the use of a walking aid; and ability to understand verbal and written information in Swedish. Exclusion criteria were: any previous major orthopaedic surgery in the lower limbs, severe back pain or other lower extremity joint pain, rheumatoid arthritis, diabetes mellitus, neurological disease, BMI >40, and/or other condition affecting walking ability. Twenty-five age- and gender-matched, healthy controls without any known musculoskeletal disease or neurological disorder were recruited through acquaintances between the years 2013 and 2015. The control group was matched to the osteoarthritis group by age strata across five age groups (40–49, 50–59, 60–69, 70–79, and 80–89 years of age). The regional ethical review board in Stockholm, Sweden, approved the study (DNR 2010/1014-31/1). All participants provided informed consent in accordance with the Declaration of Helsinki.

Twenty-eight patients with knee osteoarthritis, with a mean age of 66 (±7) years, 64 % female, completed the 1-year follow-up (Table 1). The most common reasons for not completing the 1-year follow-up were TKA in the contralateral limb within the subsequent year (n=5), not undergoing the planned surgery (n=2), or post-operative infection causing re-operation (n=2) (Fig. 1). Patients with knee osteoarthritis who did not complete the 1-year follow-up (n=12) did not differ statistically from the studied TKA group with respect to the distribution of age, gender, weight, height, BMI, or duration of years with symptomatic knee osteoarthritis (which was evaluated using independent samples t tests or the Mann Whitney U test, depending on the distribution of data, and a Fishers exact test).

The most important finding of the present study was that objectively measured function, represented by gait patterns and the 5STS, improved significantly, but did not reach the level of healthy controls. Based on change in 5STS performance, we identified patients with substantial improvements in gait patterns, while self-reported measures of function could not differentiate between patients improving in functional performance and those who did not.

Pre-operatively, lower GDI-kinetic scores were observed in the contralateral limb compared to the soon-to-be-operated limb. After surgery, the largest gait pattern improvements were found in GDI-kinetic scores of the contralateral (non-operated) limb. These results suggest TKA has a positive impact on gait patterns for both limbs, although deviations are still present 1 year post-operatively as compared to healthy controls. These results corroborate the findings of Metcalfe et al. [23] that gait pattern recovery at the contralateral knee is variable and often incomplete. Using a 1-year follow-up of gait patterns may be a too short period of time with regard to restoring joint loading. It has been suggested that functional recovery continues beyond 1 year and even up to 2 years after surgery [20].

We hypothesized that performance on the 5STS would improve after surgery, and in this sample the performance improved significantly. Nevertheless, the improvement only exceeded the MDC in 10 (36 %) of the 28 TKA patients, while the rest remained unchanged. Patients classified as having a “Good 5STS outcome” displayed significant improvements in GDI and GDI-kinetic scores, whereas patients classified as having a “Bad 5STS outcome” did not. Performance on this test is associated with quadriceps strength, and this test is considered a surrogate measure for lower limb strength [22, 27, 31]. Studies also report factors such as balance, age, weight, and sensorimotor competence play an important role in 5STS performance [22, 31]. Christiansen et al. [6] evaluated weight-bearing asymmetry during the 5STS in patients with knee osteoarthritis and concluded that greater amounts of weight-bearing asymmetry correlated with poorer functional performance up to 6 months after TKA. Alnahdi et al. [1] found patients with TKA displayed unloading of the operated limb, shifting the load to the contralateral limb when performing a sit-to-stand test 1 year after surgery.

Comparing self-reported function and pain between the “Good 5STS outcome” and “Bad 5STS outcome” groups, we found no differences, suggesting that self-reported measures are not able to detect change in performance on this test. Results of our study are consistent with prior studies demonstrating that improvements in self-reported measures are greater than improvement in performance-based function [17, 25, 35]. Boonstra et al. [3] suggest self-reported measures of function are highly influenced by pain. After a successful TKA, pain is substantially reduced and this pain relief may lead patients to overestimate self-reported function or confuse improvements in function with improvements in pain [3].

Limitations of the present study include that our sample consisted of individuals without comorbidities who were able to ambulate without the use of a walking aid. Consequently, we cannot generalize the results to all patients with knee osteoarthritis. The sample size is small, yet consistent with, and even larger than other studies using 3D gait analysis [18, 23, 36]. However, these patients presented with significantly higher weight and BMI compared to the control group, and this difference in BMI may have influenced gait patterns and 5STS performance [31]. We recognize that the use of seven different senior consultant orthopaedic surgeons to perform the surgeries increases the generalizability of the results, as this is representative of surgical practice. However, individual capability of an orthopaedic surgeon could be argued to influence the outcome of the surgery. Examining the inter-surgeon effect was beyond the scope of the present study. Additionally, there was no standardization of the post-operative rehabilitation following TKA to control the impact of rehabilitation on gait pattern or 5STS performance. Anecdotally, the standard protocol for rehabilitation following TKA did not differ between the two orthopaedic departments in this study. The rehabilitation programme included inpatient physiotherapy (>1 week), followed by physiotherapy in a primary care setting of the patient’s choice for varying lengths of time. Median time for post-operative rehabilitation in this sample was 3 months. The MDC of the 5STS is based on a sample of 29 females, mean aged 73.6 years who were able to walk at least 10 m and stand at least 10 min without an assistive device and who had no neurological disease, diabetes, visual deficits, or amputated extremities [14]. Caution should be taken not to overestimate the importance of this value, since the two studied groups (ours and Goldberg’s) might differ substantially. Additionally, we used the MDC reported for GDI [10], to evaluate change in GDI-kinetic scores. Since the GDI-kinetic is an analogue of the GDI and the magnitude of change of the two measures was similar, the MDC of 5.4 units was considered to be useful in the evaluation of GDI-kinetic.

Based on our findings, we advocate for use of the 5STS in clinical practice, as improvement beyond the MDC in 5STS performance appears to be accompanied by significant reductions in kinematic and kinetic gait pattern deviations. Additionally, our data indicate self-reported measures of function have limited use in detecting change in performance-based function. This has implications for pre-operative patient education. Specifically, we believe it is important to inform patients undergoing surgery that TKA is an effective treatment for alleviating pain and symptoms of the affected joint, but activities of daily living and more strenuous activities requiring power, strength, and balance may not become fully restored after surgery. Future research should evaluate the impact of specific exercise programmes on restoring normal joint loading patterns and functional performance. Additionally, we recommend studies to examine whether more extensive post-operative rehabilitation could lead to greater gains in performance-based function and in reductions in kinetic gait deviations. It would also be of interest to evaluate whether joint loading deviations persist at two or more years after surgery.

Measures of overall gait patterns and the 5STS revealed improvements in function 1 year after TKA, but were not restored to the level of healthy controls. Based on change in 5STS performance, we identified patients with substantial improvements in gait patterns. Self-reported measures of function could not detect differences between patients improving in 5STS performance and those who did not. These findings highlight the use of the 5STS in clinical practice since improvement on this test seems to follow the reduction in gait pattern deviations.