Reliability of a portable device for measuring lower-limb muscle strength in patients who underwent total knee arthroplasty

Abstract

[Purpose] This study aimed to analyze the reliability of using the Locomo Scan to measure quadriceps muscle strength in patients who underwent total knee arthroplasty. [Participants and Methods] Two testers (A and B) used the Locomo Scan to perform three measurements of quadriceps muscle strength on the operated side of 64 patients who underwent total knee arthroplasty and recorded the mean and maximum values. We then assessed intra- and inter-rater reliability using intraclass correlation coefficients and obtained systematic errors via Bland–Altman analysis. We calculated the 95% confidence interval and percentage error for the minimum detectable change. [Results] The intraclass correlation coefficients for case 1 were 0.903 and 0.889 for testers A and B, respectively, and those of case 2 were 0.783 and 0.818 for the mean and maximum values, respectively. Bland–Altman analysis revealed no systematic errors. The 95% confidence interval and percentage error for the minimum detectable change were 0.138 kgf/kg and 39.6% for the mean value and 0.132 kgf/kg and 35.1% for the maximum value, respectively. [Conclusion] The Locomo Scan can reliably measure quadriceps strength in patients who underwent total knee arthroplasty and may be useful in clinical settings.

INTRODUCTION

Total knee arthroplasty (TKA) is a surgical procedure that is performed to treat severe knee osteoarthritis, and the number of TKA patients is expected to rise as the elderly population increases¹⁾. The goals of physical therapy after TKA are to increase the range of motion of the knee joint, enhance lower limb muscle strength, and improve gait function and activities of daily living (ADL). Quadriceps muscle strength is strongly associated with ADL and can be used to assess postoperative functional impairment as well as treatment efficacy^{2, 3)}. Reportedly, quadriceps muscle strength is significantly reduced in the early postoperative period^{4, 5)}, and may require several months to fully recover^{6, 7)}. During physical therapy, it is important to evaluate the postoperative recovery of muscle strength over time; therefore, a reliable quantitative measurement of quadriceps muscle strength is urgently needed for TKA patients. Standard methods for the evaluation of quadriceps muscle strength include the usage of a handheld dynamometer (HHD) or an isokinetic machine to make measurements. However, when using an HHD, the reliability of muscle strength measurements can be affected by device fixation and tester experience^{8, 9)}. Meanwhile, the isokinetic machine is large and nonmobile with a long measurement process, which is not ideal for screening muscle strength over a limited time period.

The device used in the reliability study is a portable device for measuring quadriceps muscle strength (Locomo Scan, ALCARE Co. Ltd., Tokyo, Japan) and can be used in a long sitting position with the knee joint slightly flexed. Unlike conventional knee extension strength measurement systems, which typically require approximately 90 degrees of knee flexion, this device enables measurements to be performed in a more comfortable posture. It also employs the same method used in quadriceps setting exercises, which are commonly performed during post-TKA rehabilitation¹⁰⁾. Therefore, the Locomo Scan would be useful in screening quadriceps muscle strength in post-TKA patients since they would already be familiar with the measurement method and exercise pattern involved. Previous studies have reported reference values for quadriceps muscle strength in healthy Japanese participants as well as high correlations with isokinetic machine measurements^10,11,12), thus the Locomo Scan has been described as a simple and useful measurement device for muscle strength. Although intra-rater reliability of the Locomo Scan has been demonstrated in healthy individuals^{11, 13)}, its intra-rater and inter-rater reliabilities in post-TKA patients remain unclear. It will also be necessary to clarify measurement errors among testers with this device since the same tester may not always be available to perform measurements even at one facility.

Consequently, the purpose of this study was to analyze the reliability of using the Locomo Scan to measure quadriceps strength in post-TKA patients.

PARTICIPANTS AND METHODS

Our study included patients who underwent TKA at one regional Japanese hospital between April 1, 2021, and April 1, 2023. The exclusion criteria were defined as follows: (1) dementia, (2) the inability to understand Japanese, (3) postoperative complications such as nerve palsy, and (4) cases deemed to be at risk for study participation by the attending physician. The study protocol was approved by the Ethics Committee of Suzuka Kaisei Hospital (No. 2021-07), and written informed consent was obtained from all participants. Ultimately, 64 patients (19 males and 45 females) were included in the study (age 71.9 ± 7.3 years, height 155.4 ± 8.3 cm, weight 62.5 ± 10.8 kg). The demographic data of the participants are provided in Table 1, and the algorithm for participant selection is shown in Fig. 1.

Table 1. Demographic data of participants.

Characteristics
Sex	Male	19
	Female	45
Age (years)		71.9 ± 7.3
Height (cm)		155.4 ± 8.3
Weight (kg)		62.5 ± 10.8
BMI (kg/m2)		25.9 ± 3.9
Length of hospital stay (days)		31.5 ± 9.0

Values are mean  ± SD. BMI: Body mass index.

Fig. 1.

Algorithm for participant selection.

TKA: total knee arthroplasty.

Measurements were recorded for the quadriceps muscle on the operated side of the TKA patient, as described in the instruction manual for the Locomo Scan. This device measures the pressure on the knee socket while the knee is extended (Fig. 2). As a preliminary exercise, the participants practiced applying force with the measurement device several times while checking the corresponding monitor. During measurements, patients were instructed not to bend their trunk forward or backward and to contract the quadriceps with maximal voluntary effort. The values measured in this study were standardized by body weight (kgf/kg)¹⁴⁾. Measurements were taken 3 times, with each measurement lasting 10 s and a measurement interval of 30 s. The mean and maximum values of the 3 measurements were recorded, and measurements were taken by 2 male physical therapists (Tester A and Tester B) who were familiar with using the Locomo Scan. At the beginning of the study, Tester A had 4 years of clinical experience and a BMI of 23.7 kg/m², while Tester B had 12 years of experience and a BMI of 22.3 kg/m². Measurements were taken on 2 consecutive days during the 5 days prior to discharge, with each tester taking measurements every other day. The order of the testers was randomized.

Fig. 2.

Measuring quadriceps muscle strength.

The participant performed isometric quadriceps exercises in a long sitting position with their knee socket pressed against the load measuring portion of the device. The direction of the force is indicated by the arrow.

For the statistical analysis, intraclass correlation coefficients (ICCs) were calculated to evaluate intra- and inter-rater reliability. Bland–Altman analysis¹⁵⁾ was also used to determine the presence of systematic errors (additive and proportional errors) between testers. For the presence of additive errors, the mean of the difference between the two testers, the standard deviation of the difference (SDd), and the 95% confidence interval (CI) were calculated. For the presence of proportional errors, a Bland–Altman plot was created and regression equations were used to examine significance. To assess the magnitude of error, the 95% CI of the minimum detectable change (MDC₉₅) was calculated using the following formula:

MDC₉₅=SDd*1.96

To assess the percentage of error to the measured value, the %MDC was calculated using the following formula:

%MDC=100*MDC₉₅/Mean

The mean value (MEAN-V) and the maximum value (MAX-V) of the 3 measurements for each tester were used for the analysis of ICCs, systematic errors, MDC₉₅, and %MDC in order to determine inter-rater reliability.

The software used for statistical analysis was SPSS Statistics version 23 (IBM, Chicago, IL, USA), where the statistical significance level was set to less than 5%.

RESULTS

The ICCs Case 1 and 95% CIs are shown in Table 2. The ICCs Case 1 for Tester A and Tester B were 0.903 and 0.889, respectively. The ICCs Case 2 and the 95% CIs are shown in Table 3. The ICC Case 2 for the MEAN-V of the 3 measurements was 0.783, and the ICC Case 2 for the MAX-V was 0.818.

Table 2. The mean value of each of the three measurements performed by two testers and the intra-rater reliability.

	Mean ± SD (kgf/kg)			ICC	95% CI
	1st	2nd	3rd
Tester A	0.331 ± 0.112	0.353 ± 0.117	0.365 ± 0.117	0.903**	(0.858–0.937)
Tester B	0.323 ± 0.103	0.349 ± 0.105	0.365 ± 0.101	0.889**	(0.838–0.927)

**p<0.01. ICC: intraclass correlation coefficient.

Table 3. Mean values of MEAN-V and MAX-V obtained from two testers and inter-tester reliability.

	Mean ± SD (kgf/kg)		ICC	95% CI
	Tester A	Tester B
MEAN-V	0.350 ± 0.112	0.346 ± 0.101	0.783**	(0.666–0.862)
MAX-V	0.376 ± 0.118	0.372 ± 0.104	0.818**	(0.717–0.885)

**p<0.01. ICC: intraclass correlation coefficient; MEAN-V: Mean value of the 3 measurements; MAX-V: Maximum value of the 3 measurements.

The mean difference between the measurements of the 2 testers, the SDd, and the 95% CI for the difference are shown in Table 4 (for MEAN-V and MAX-V), while the Bland–Altman plot is provided in Fig. 3. The 95% CI for the mean difference between the MEAN-V for the 2 testers was (−0.013 to 0.022), and the 95% CI for the mean difference between the MAX-V for the 2 testers was (−0.013 to 0.021). All the regression equations in the Bland–Altman plot were not statistically significant (p>0.05). The MDC₉₅ and %MDC for the MEAN-V were 0.138 kgf/kg and 39.6%, and those for the MAX-V were 0.132 kgf/kg and 35.1% (Table 5).

Table 4. System error of inter-tester.

	Mean difference (kgf/kg)	SDd	Additive error	Proportional error
			95% CI	Slope of the regression line
MEAN-V	0.0042	0.071	−0.013 to 0.022	0.123	p=0.167
MAX-V	0.0039	0.067	−0.013 to 0.021	0.144	p=0.073

SDd: standard deviation of the difference; MEAN-V: Mean value of the 3 measurements; MAX-V: Maximum value of the 3 measurements.

Fig. 3.

Bland–Altman plot.

(a) Mean value, (b) Maximum value.

Table 5. MDC₉₅ and %MDC of inter-tester.

	MDC95 (kgf/kg)	%MDC (%)
MEAN-V	0.138	39.6
MAX-V	0.132	35.1

MDC₉₅: minimal detectable changes at the 95% confidence interval; MEAN-V: Mean value of the 3 measurements; MAX-V: Maximum value of the 3 measurements.

DISCUSSION

We analyzed whether the Locomo Scan can provide reliable measurements in post-TKA patients. The device can measure with the knee joint flexed approximately 30 degrees, without requiring the knee joint to be flexed more than 90 degrees or fully extended. Therefore, even postoperative patients who complain of pain during knee flexion and extension can exert muscle strength without difficulty. Our findings suggest that this device is useful for postoperative assessment of quadriceps strength in TKA patients.

With regard to ICC criteria, Landis and Koch¹⁶⁾ defined scores from 0.81 to 1.00 as “almost perfect”, 0.61 to 0.80 as “substantial”, and 0.41 to 0.60 as “moderate”. The ICC Case 1 was validated with 2 testers in this study, and both testers had a score of 0.81 or higher (Tester A: 0.903, Tester B: 0.889), indicating that the Locomo Scan has good intra-rater reliability. On the other hand, the ICC Case 2 was validated using the MAX-V and MEAN-V for 3 measurements by 2 testers. The MAX-V was 0.818, confirming high inter-rater reliability, but the MEAN-V was 0.783, which was slightly lower than the MAX-V. Bland–Altman analysis is a method that can be used to distinguish between additive errors and proportional errors, both of which are systematic errors. Statistically, if the 95% CI of the mean of the difference between 2 measurements does not include zero, then an additive error is present because the measurements are distributed in a certain direction. Additionally, if the regression test of the Bland–Altman plot is significant, then a proportional error is present. In this study, the 95% CI for the difference between testers for the MEAN-V or MAX-V included zero, and all regression equations in the Bland–Altman plot were not statistically significant, so there was no systematic error for either the MEAN-V or MAX-V. Furthermore, the MDC₉₅ and %MDC were lower when the MAX-V was used compared to when the MEAN-V was used, and random errors were smaller as well. These results suggest that it would be ideal to adopt the maximum value as the measurement value in clinical scenarios where more than 1 person is expected to be involved in the measurement process. Previous studies examined the reliability of the HHD, a commonly used muscle strength device, in TKA patients and reported intra-rater and inter-rater reliabilities that were greater than 0.9^{17, 18)}. Meanwhile, the %MDC values for the Locomo Scan in our study were 39.6% (MEAN-V) and 35.1% (MAX-V), and had not been previously investigated in TKA patients. In comparison, Buckinx et al.¹⁹⁾ analyzed the %MDC of quadriceps muscle strength measurements using the HHD in elderly individuals and reported a value of 50.75%. Based on these findings, we believe that the Locomo Scan can make measurements with the same level of accuracy as the HHD, even though its ICC is slightly inferior. Based on our literature review, no studies have been found that have investigated the intra- and inter-rater reliability of isokinetic machines in patients after TKA. In this population, the use of isokinetic machines may be limited due to difficulties in achieving the required test position caused by restricted knee range of motion, pain during maximal voluntary contractions, and fear of movement. However, during our study period, all TKA patients included in the study were able to undergo muscle strength measurement using the Locomo Scan, suggesting that this device is applicable to a wide range of TKA patients.

This study has several limitations. First, we did not examine test-retest reliability by the same tester, given the burden on participants. Therefore, the systematic error, MDC₉₅, and %MDC for the same tester are unknown. Previous studies have reported lower intra-rater MDC₉₅ and %MDC values compared to inter-rater values for knee extension muscle strength measurements using the HHD²⁰⁾. In general, inter-rater reliability is lower than intra-rater reliability, so intra-rater MDC₉₅ values may be smaller than inter-rater values. Second, although the reliability of the Locomo Scan was verified in this study with 2 male physical therapists who were familiar with its usage, factors that can impact device reliability, such as measurement proficiency and tester characteristics, were unknown and warrant further analysis in future studies. Finally, the measurements in this study were taken approximately 1 month postoperatively. Therefore, the reliability of this device in assessing early postoperative and long-term outcomes cannot be verified.

In conclusion, the Locomo Scan provided highly reliable measurements of quadriceps muscle strength in post-TKA patients. There were also no systematic errors in both inter-rater scenarios, suggesting that this device is a useful method for testing postoperative muscle strength. In addition, our findings indicated that the maximum value should be adopted as the measurement result rather than the mean value.

Conference presentation

A summary of this paper was presented at the 10th Annual Meeting of the Japanese Society of Musculoskeletal Manipulative Physical Therapy.

Funding

No funding was received for this study.

Conflict of interest

The authors have no conflicts of interest to declare.