Validation and reliability of six-minute walk test predictive model equation of total knee replacement

Introduction

Total knee replacement (TKR) is a common surgical procedure for treating severe knee injuries or advanced osteoarthritis. Since its inception in the 1970s, TKR has undergone significant changes. The number of TKR surgeries will continue to rise at a faster rate due to the growing number of knee arthritis cases worldwide. By 2030, it is anticipated that these will have increased by 85% to 1.26 million primary TKR treatments. ¹

Knee osteoarthritis is a leading cause of disability worldwide, with a prevalence of 28.7% in India. ² TKR has become a popular treatment option, with increased attention of surgeons to measure outcomes of surgery by utilizing patient self-reported tools. These outcomes include pain, functional status, and health-related quality of life. TKR is considered an excellent treatment option for end-stage knee osteoarthritis, improving functional status and overall quality of life. ³ In any clinical trial of knee OA, an important outcome is an improvement in physical function, this is usually measured through performance-based or patient-reported measures. ⁴ Total knee replacements are increasing, with those under 65 being the fastest-growing group. Concerns have been raised about outcomes in this age bracket. ⁵ The 6-minute walk test (6MWT) is a reliable and easy-to-administer measure, with high intra-class correlation coefficients (0.82–0.99) and strong correlations with physical activity (r = 0.40–0.85). ⁶

The 6-minute walk test (6MWT) is a brief, objective measure of functional walking improvement after total knee replacement (TKR). ⁷ Despite improvements, TKR patients often have residual deficits in quadriceps strength (41% reduced), walking distance (28% lesser), and stair climbing speed (105% lower). ⁸ To address these deficits, outpatient TKR rehabilitation focuses on achieving maximal functional mobility. ⁹ The 6MWT and Timed-Up-and-Go (TUG) tests are commonly used to evaluate functional recovery. Social support and knowing a patient’s functional abilities are crucial in determining the best discharge destination. ¹⁰ A patient’s prognosis after TKR depends on their home environment, physical needs, and current functional abilities.

Reliability and validity are crucial when evaluating patient-reported outcome measures (PROMs); moreover, it depends on the target population and situations. ¹¹ Several studies have shown that total knee replacement (TKR) improves mobility, quality of life, and patient satisfaction. ¹² The validity of a measurement tool refers to its ability to measure what it intends to measure. Technical quality, including reliability and interpretability, is also essential. ¹³

The Knee Injury and Osteoarthritis Outcome Score (KOOS) is a widely used PROM. Twelve KOOS items in total—four for pain, four for function, and four for quality of life—are rated using three different scales and the KOOS-12 total score. ¹⁴

Effective community rehabilitation requires valid measurements of mobility. Current measurements of balance and gait often fail to represent real-life community mobility, which demands adaptability and adjustment to avoid obstacles. Therefore, it’s essential to develop PROMs that accurately capture changes in mobility and community functioning.

Postoperative muscle insufficiency and articular surface replacement can impair knee joint proprioception. Deficiency in proprioception is affected by valgus deformity, OA revision, and recent surgery. Assessing changes after TKR requires thorough and reliable tests. ¹⁵ The six-minute walk test is a valid and reliable instrument for assessing functional recovery after TKR. Antony Leo Aseer et al. developed a prediction equation for 6-minute walk test distance based on a secondary analysis of a randomized controlled trial carried out with 220 TKR patients. ¹⁶ This study aims to explore the validity and reliability of the prediction formula, which can help clinicians estimate functional recovery and plan rehabilitation programs for patients undergoing TKR.

Methods

Procedure

The Institutional Ethics Committee for Student Projects, Sri Ramachandra Institute of Higher Education and Research, approved the cross-sectional study (REF: CSP/23/JUL/133/710).

The study was performed at the outpatient physiotherapy department of a tertiary care hospital. TKR patients meeting the inclusion criteria who had underwent surgery before approximately 3 months were included, and the demographic data were collected. All subjects received an explanation of the study protocol, and informed consent was obtained from all the patients.

The isometric muscle strength of quadriceps muscle was tested with the subject positioned in high sitting with knee joint flexed in 70° with the dynamometer probe placed over the lower end of the shin. Participants were instructed to push the dynamometer in attempting to knee extension. Each contraction of knee extension was held for 5 seconds, and three trials were measured with 30 seconds rest interval between measurements. The mean values of three readings were calculated and considered as isometric muscle strength in kilograms.

The six-minute walk test is a functional ability test measuring the distance covered by a subject in 6 minutes. The procedure of the test was followed in accordance to the proposed guidelines of American Thoracic Society. The walking surface area was kept free of obstacles, and two cones were placed at the two extremes. Participants were instructed to discontinue walking, if any discomfort was felt, and use of walking aid was optional. They were encouraged to walk back and forth the distance with normal speed, and the distance covered in 6 minutes was measured in meters.

The Knee Injury and Osteoarthritis Outcome Score (KOOS) is a questionnaire that measures how knee injuries and osteoarthritis affect a patient’s symptoms, function, and quality of life. It’s a self-administered tool. Each subscale is scored from 0 (worst knee problems) to 100 (best knee problems).

Clinical variables encompassed isometric strength of the quadriceps was measured using the push-pull dynamometer, knee flexion range measurement was done using the universal goniometer, and pain score of the KOOS tool. The 6MWD was performed following ATS guidelines and checked against the predicted distance using the derived equation.

The six-minute walk test distance 3 months after total knee replacement was predicted by the following model.

( isometric muscle strength of quadriceps × 23.83 + knee flexion range × 2.34 + KOOS − pain score × 1.78 ) – 188.48

Results

The study cohort consisted of 82 TKR patients, with a mean age of 65.4 ± 8.2 years, and a predominantly female population (63.4%, n = 52). Table 1 shows the demographic characteristics of the included subjects. The mean and SD values for the 6MWT, quadriceps strength, KOOS Pain score, and knee flexion range of motion and pain score (NPRS) were 299.04 ± 42.424, 4.73 ± 1.352, 87.72 ± 4.495, 97.82 ± 5.299, and 3.07 ± 1.097, respectively (refer Table 2).

OPEN IN VIEWER

Table 1.

Demographic data (n = 82).

Characteristics	Value
Age (years)	65.40 ± 8.23
Mean ± SD
Height (cm)
Male	167.6 ± 7.3
Female	158.3 ± 7.1
Weight(kg)
Male	71.4 ± 12.3
Female	73.2 ± 13.4
BMI mean ± SD
Male	31. ± 4.9
Female	29.3 ± 5.7
Approach
Midvastus	42 (51%)
Median parapatellar	40 (49%)
Gender
Male	30 (36.5%)
Female	52 (63.4%)
History of diabetes
Yes	49 (60%)
No	33 (40%)
History of hypertension
Yes	69 (84%)
No	13 (16%)
Post op weeks	12.3 ± 0.5
Mean ± SD

OPEN IN VIEWER

Table 2.

Baseline clinical variables.

Outcomes N = 82	Mean ± standard deviation	Median
Actual six-minute walk test (meters)	299.04 ± 42.424	299.50
Knee extension using dynamometer (kilograms)	4.73 ± 1.352	5.00
KOOS pain score (score)	87.72 ± 4.495	86.00
Knee flexion range of motion (degree)	97.82 ± 5.299	98.00
Numerical Pain Rating Scale score	3.07 ± 1.097	3.00
Calculated values of predictive equation 6MWT (meters)	308.29 ± 41.877	307.17

The relationship between the 6MWT and the predictive model equation values was statistically significant with a p < 0.001 using the Pearson correlation coefficient method indicating a very strong correlation between the two measures. The predictor equation showed a strong correlation with the actual 6MWD (r = 0.92, p < 0.001) and also demonstrated very strong concurrent validity with the 6MWD (refer Table 3). Functional improvements in the patients who underwent TKA were also portrayed in the study. The correlation scatter plot is presented in Figure 1; Pearson’s product moment correlation shows significant result (p < 0.05) with coefficient correlation approximately 85%. This suggests a remarkably strong and reliable relationship between the two score, indicating that they are closely linked and can be used to make highly accurate predictions about each other.

OPEN IN VIEWER

Table 3.

Validation of predicted 6MWT equation and actual 6MWT distance.

Calculated value of 6MWT using predicted equation and actual values	Pearson’s r	0.922
	p-value	<.001

OPEN IN VIEWER

Figure 1.

Correlation plot between the actual scores of the 6MWT and the predicted scores.

Test–retest reliability which was examined with the Intraclass Correlation Coefficient (ICC) was equal to 0.9 (refer Table 4). In this study applying the values in the predictive model equation, we should get a number close to the six-minute walk test’s outcomes. The study further commented that the predictive model equation could also be used to identify those patients at an increased risk from poor functional outcomes after TKA.

OPEN IN VIEWER

Table 4.

Test–retest reliability.

Characteristics	Predicted 6MWT (mean± standard deviation)	Actual 6MWT (mean ± standard deviation)
Test	308.29 ± 41.87	299.04 ± 42.42
Retest	306.42 ± 35.21	294.53 ± 63.18
Mean difference	1.87	4.51
ICC (95% CI)	0.948 (0.88, 0.97)	0.902 (0.78, 0.94)

Discussion

The aim of this study was to validate the predictor equation of 6MWT in individuals with TKR after 3 postoperative months and to find the reliability of comparing actual six-minute walk test and predictive model equation in TKR individuals. The current study provides 6MWT values and a prediction equation, in an age group of above 45 years obtained from 82 patients with high correlation.

The study sought to validate the anticipated six-minute walk test (6MWT) in patients after total knee replacement (TKR) by evaluating the accuracy and reliability of prediction equations for 6MWT distances. The study included a cohort of TKR patients assessed at 3^rd postoperative month. The main aim was to ascertain if the anticipated 6MWT distance, based on preoperative and early postoperative factors, properly represented the actual distances traversed by patients throughout the assessment.

The research identified a strong correlation between the predicted and actual 6MWT distances, suggesting that the predictive equations were reliable. Knee flexion ROM< isometric quadriceps strength and pain scores were significant predictors of the 6-minute walk test (6MWT) distance. The validation of these predictive equations is essential, enabling physicians to establish realistic rehabilitation objectives and efficiently monitor progress.

Predicted distances for the 6MWT have been determined for healthy persons, with variables like age, gender, height, and weight. Equations established by Enright and associates in 1998 yield sex-specific forecasts for walking distance in healthy adults. These predictions facilitate the comparison of an individual’s performance against normative values and evaluate the extent of impairment or functional restriction. The 6-minute walk test (6MWT) is also utilized across several clinical groups, including those with arthritis, multiple sclerosis, Parkinson’s disease, spinal cord injuries, stroke, pulmonary disorders, and knee arthroplasty. ¹⁷

Agarwal et al. study aimed to provide reference values and a prediction equation for the six-minute walk test in a large cohort of healthy Indians aged 6–75 years. They stratified the population into five age groups over 65, 41–65, 18–40, 13–17, and 6–12—after participant recruitment. ³ They had aimed their study to determine the cardio respiratory endurance by predicting six-minute walk distance in post-COVID patients.

In accordance with the earlier study findings, in our study also patients were comfortable while walking at the third month’s follow-up and there was no adverse event reported. All the patients were comfortable during the six-minute walk test without any trouble.

The reliability and validation of prediction equations of the 6MWT assume a key position in their application in the clinical setting. In a study by Duncan et al., six-minute walk distance using predicted equations was validated using previously established equations in healthy European population and found to be statistically significant with less variations between the actual and predicted values. In line with their recommendation, the predicted six-minute walk test can be used to predict the actual six-minute walk distance if in case the patient is not able to perform the lengthy testing protocols and increase safety in occasions where mobility impairment or safety issue is a concern. Moreover, prediction equations can be used to monitor the patient’s progress over time, which enables adjustment by the clinicians in treatment plans. ¹⁸

According to Harada et al., this test is an important tool, especially for rehabilitation purposes, for assessing functional capacity and mobility in older adults. In these situations, the equation should be used in predicting the functional capacity and endurance as an alternative measurement method. ¹⁹

In addition to its clinical application, the 6MWT has been used as an investigative tool in many studies assessing exercise interventions and their relationship to the outcomes used.

The accurate estimation of 6MWT distance is critical for predicting postoperative outcomes and guiding treatment decisions after TKR. ²⁰ As such, continued research into the validation and reliability of predictive model equations is essential for improving patient care and optimizing outcomes after TKR. By refining these models and incorporating additional variables, researchers can develop more accurate prediction tools that better capture the complexities of patient outcomes after TKR.

This formula hence can be used in the establishment of the distance covered without administering the 6MWT, and this would be an alternative. One such major benefit of this predictive formula is that it replaces the 6MWT and also this would be useful for the therapist to design the tailored rehabilitation program post-TKR. This could be easily used to monitor patients over time to allow for adjustments that clinicians can make to rehabilitation programs without using the 6MWT; predictor equations help in selecting patients who may require additional support or interventions to enhance functional capacity and endurance; predictor equations can be used in identifying patients at risk of poor functional outcome after TKA; and predictor equations help development of a rehabilitation program that is particularized to the patient’s unique requirements and capacity.

This research puts forward a better understanding of the walking distance covered by people at 3 months post-TKA surgery and the results acquired so far have shown that this model can be conveniently used, applying the 6MWT test methods in simplifying the work of health professionals when assessing patients’ functional capacity and endurance at a faster, more precise, and accurate level, without the necessity of a big and open area.

The advantage of predictive equations is instead of performing a six-minute walk test, we can use the predictive equation for calculating the six-minute walk test distance, as the current study also shows a high correlation. This can also be used to track the prognosis of the patients who undergo total knee replacement.

Future research may, therefore, be directed toward broadening the applicability of the model to varied groups of patients and, most importantly, the validity of long-term predictions to make the model clinically useful and reliable for rehabilitation purposes.

Limitation

One of the study’s limitations was the potential for bias due to the order in which the six-minute walk test was conducted. The assessor is not blinded to the intervention of the study. The current study was done in 3^rd month follow-up, 6 months and a baseline follow-up could have also been done. Patients’ willingness to participate may have been impacted by their initial experience with pain, which also made it challenging to assess quadriceps muscle strength using a dynamometer as they were reluctant to perform the test while in a state of discomfort.

Conclusion

In conclusion, the results showed that this predictive formula can be used in approximating the distance covered without administering the 6MWT in post total knee replacement patients. This essentially means that the formula is an accurate predictor of a person’s walking distance, which uses substituted parameters within the model for convenience and not necessarily conducting the 6MWT.