Highlights
- •
The ratio of vastus medialis (VM) cross-sectional area (CSA) to quadriceps muscle and that of semi-membranosus (SM) CSA to hamstring were low in individuals with medial knee osteoarthritis.
- •
The thigh region to evaluate muscle atrophy differed by muscles.
- •
The thigh region around the mid-thigh should be analyzed for the VM CSA ratio.
- •
The thigh region at the muscle belly should be analyzed for the SM CSA ratio.
- •
The therapeutic effect of VM and SM hypertrophy on knee osteoarthritis is worth testing.
Abstract
Background
The characteristics of thigh-muscle cross-sectional area (CSA) in older adults with knee osteoarthritis (KOA) remain controversial.
Objectives
This study aimed to evaluate atrophy of individual thigh muscles in older adults with KOA and to determine which muscle CSA should be measured to detect KOA-related muscle atrophy of the thigh.
Methods
In older adults, individual thigh-muscle CSA measured by 1.5 Tesla MRI was analyzed at 5% intervals of the femoral length (FL) around the mid-thigh between the proximal 25% of the FL and the distal 25%. Participants with KOA grade ≤ 1 and grade ≥ 2 were compared for ratios of quadriceps muscle (QM) CSA to total thigh, individual QM CSA to QM, and individual hamstring (HAM) CSA to HAM at 5% intervals.
Results
We included 40 older adults [20 males; mean (SD) age 73.3 (4.7) years; 20 with KOA grade ≤ 1 and 20 with KOA grade ≥ 2]. The ratio of vastus medialis (VM) CSA to QM from the proximal 25% to distal 15% and the ratio of semi-membranosus (SM) CSA to HAM at the distal 10% to 25% were significantly lower with KOA grade ≥ 2 than grade ≤ 1; the effect sizes were 0.34 to 0.67 for VM and 0.40 to 0.60 for SM. The effect sizes were greatest for the ratios of VM CSA to QM at the mid-thigh with 5% intervals and the ratio of SM CSA to HAM at the distal 25%.
Conclusions
The ratio of VM CSA to QM and/or that of SM CSA to HAM were low and were the best indicators to detect KOA-related muscle atrophy of the thigh. However, to detect KOA-related muscle atrophy, the VM CSA ratio should be analyzed in the thigh region around the mid-thigh, whereas the SM CSA ratio should be analyzed in the thigh region at the muscle belly.
1
Introduction
Previous studies showed that quadriceps muscle (QM) and hamstring (HAM) strength were low in adults with knee osteoarthritis (KOA) , but recent studies have focused solely on the QM to determine the relation between dysfunctional QM and progressive KOA . To objectively evaluate muscle function, muscle volume measurement has been used to estimate knee extension strength , and simple methods of muscle volume estimation based on muscle cross-sectional area (CSA) are widely used . In fact, longitudinal changes in thigh muscle CSA in adults with KOA are more responsive than changes in muscle strength . However, some authors suggested that QM CSA may not be lower in adults with radiographic KOA . Conversely, some researchers reported a smaller QM CSA related to knee pain but not radiographic KOA . Moreover, Ruhdorfer et al. reported that QM CSA and also HAM CSA were decreased over time in adults with KOA regardless of knee pain. Therefore, the characteristics of thigh muscle CSA in older adults with KOA remain controversial.
We proposed 2 potential causes for inconsistent findings reported in previous studies. First, QM and HAM each consist of 4 individual muscles. Some studies have reported that the change in individual QM CSAs differed from that in QM CSAs in adults with KOA . Moreover, muscle volume and/or CSA variations may be affected by individual variations, including sex, age, body height, body weight, and body mass index (BMI), and generally may be normalized by body height or body weight . Then, when comparing adults with KOA to controls, the use of the ratio of QM and/or HAM CSA to total thigh muscle could remove individual variations. However, studies focusing on the ratios of CSAs of individual thigh muscles are rare, and whether the ratios of CSAs of individual thigh muscles may be related to sex, age, and BMI is unknown. Second, which thigh muscle regions should be measured to detect KOA-related muscle atrophy is unclear. Yamauchi et al. showed that the CSAs around the muscle belly of individual QM and HAM were maximally decreased due to disuse, whereas the CSAs around the muscle ends were not significantly changed. Therefore, when comparing adults with KOA to controls, consistent anatomical localization of studied muscles is necessary; however, the measured muscle region varied across previous studies.
The main purpose of this study was to demonstrate atrophy of individual thigh muscles as measured by MRI in older adults with high-grade KOA and to determine which thigh-muscle regions could be used to detect KOA-related muscle atrophy of the thigh. Our secondary objective was to demonstrate an association between KOA-related muscle mass and sex, age, and BMI.
2
Material and methods
2.1
Participants
This study was a cross-sectional study reported according to STROBE guidelines . Japanese older outpatients (age range, 60–85 years) who reported knee pain (≥ 3 months) in one and/or both knees ( n = 65) were included in the study. Using standing knee radiography, patients were diagnosed with medial KOA, and KOA was graded according to the Kellgren–Lawrence classification .
Of the 65 older adults recruited, informed consent for conducting MRI was obtained from 48, and 17 declined MRI. Among the 48 older adults, 40 (20 males), including 20 consecutive participants with radiographic KOA grade ≤ 1 and 20 with KOA grade ≥ 2, were selected to produce an equal number of males and females and adults with KOA grade ≤ 1 (low-grade KOA) and KOA grade ≥ 2 (high-grade KOA) ( Fig. 1 ). The Japanese Knee Osteoarthritis Measure (JKOM) score (100 total points) was used to evaluate knee pain (32 total points), activity of daily living (60 total points) and health perception (8 total points); higher JKOM scores indicated worse disease . For each participant, the side with the poorer KOA grade was used in the study. For participants with bilateral identical KOA grades, the more painful side was studied.
The Institutional Review Board approved this study (no.: 56824). This study was carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki). Informed consent was obtained from all individual participants included in the study.
2.2
Axial MRI acquisition
Participants were required to rest in the supine position for 15 min to avoid fluid shifts, and their ankles were stabilized in the supine position by using a box to maintain the ideal anterior patellar direction. A 1.5-T closed MRI (Excel ART, Canon Corp., Tokyo) was performed around the mid-thigh. After determining the coronal and sagittal femoral bone axis using a scout scan, T1-weighted spin-echo with 10-mm thick axial images (TR/TE, 625/15 ms; field of view, 250 mm; 512 × 512 matrix) were obtained.
2.3
Femoral length measurement and CSA analysis method
We obtained a standing anteroposterior radiograph of both lower limbs. One of the authors (KY) measured femoral length (FL) as the distance between the top of the greater trochanter and the intercondylar fossa of the femur by using Management Integrate Network Diagnosis Solution (LEOCLAN Co., Osaka, Japan).
To designate consistent anatomical locations, the mid-thigh was defined as the midpoint between the distal end of the lesser trochanter and the proximal end of the rectus femoris (RF) tendon by using continuous axial images . Individual thigh-muscle CSAs [QM: RF, vastus lateralis (VL), vastus intermedius (VI), and vastus medialis (VM); HAM: biceps femoris long head (BFL), biceps femoris short head (BFS), semitendinosus (ST), and semi-membranosus (SM)] were measured by manual tracing just inside the muscle fascial lines by using Medical Image Processing, Analysis, and Visualization v7.3.0 software (US National Institutes of Health, Bethesda, MD, USA) at 5% intervals of the FL around the mid-thigh to the proximal 25% and distal 25% of the FL ( Fig. 2 ). Visible intermuscular fat, vessels and neuronal connective tissue were not included. All measurements were performed by one of the authors (KY), who was blinded to patients’ clinical information.
To calculate muscle volume in cm 3 of the total thigh muscle, QM, HAM, RF, VL, VI, VM, BFL, BFS, ST, and SM, the CSAs were summed at 5% intervals of the FL. To calculate total thigh muscle CSA and cm 2 , the CSAs for RF, VL, VI, VM, BFL, BFS, ST, and SM were summed at each CSA interval. The ratios of QM to total thigh and HAM to total thigh were calculated at each CSA interval. The ratios of individual QM muscles to the QM and individual HAM muscles to the HAM were calculated at each CSA interval.
2.4
Reproducibility of CSA measurements in individual QM
To test the reliability of measurements of ratios of individual QM CSAs to QM in 12 randomly selected thighs, the CSA ratio measurements at the distal 10% were repeated by 2 of the authors (KY and SS) with blinding to the results of the other measurements.
2.5
Statistical analysis
In our pilot study, the effect size for the Student t test between KOA grade ≤1 and grade ≥2 for the ratio of VM CSA to QM was 0.82. An a priori power analysis by Student t test between the 2 groups showed that with power = 0.8, P = 0.05, effect size d = 0.82 and allocation ratio = 1, we needed a total sample of 40. The power analysis involved using G*Power 3.1. (Heinrich-Heine-Universität, Düsseldorf, Germany). Intergroup comparisons (KOA grade ≤ 1 vs grade ≥ 2) involved parametric testing (Student t test), and data for the 2 groups with normal distribution are expressed as mean (SD). When the data for one or both groups were not distributed normally, nonparametric tests (Mann–Whitney U test) were used, with data for both groups expressed as median (Q1–Q3). The normal distribution was tested by Shapiro–Wilk test. For Student t test, the effect size was calculated as the square root of t 2 divided by t 2 + degree of freedom. For Mann–Whitney U tests, the effect size was calculated as Z divided by the square root of the sample size ( n = 40). Effect sizes > 0.5 were defined as a large effect, 0.3 to 0.5 a medium effect, and ≤ 0.3 a small effect .
To determine sex differences in KOA-related muscle mass, muscle volume with a significant difference between KOA grade ≤ 1 and grade ≥ 2 was compared between males and females. In terms of the individual muscle ratio with a significant difference between KOA grade ≤ 1 and grade ≥ 2, the muscle ratio at the thigh region with the highest effect size within the same muscle was compared between males and females.
Moreover, to determine an association between the selected KOA-related muscle mass and age and BMI, we used correlation analysis with the Pearson correlation coefficient ( r ) because the variables were normally distributed. Correlation coefficients ≥ 0.7 were considered strong, 0.4 to 0.7 moderate, and < 0.4 weak. All statistical tests were performed with SPSS v22 (SPSS, Inc., Chicago, IL, USA). P < 0.05 was considered statistically significant.
The relative intra-observer and interobserver reliability of the CSA ratio measurement was evaluated by the intraclass coefficient between the paired measurements repeated by the first author (KY) and interclass coefficient between the paired measurements performed by the 2 observers (KY and SS). Coefficients ≥ 0.8 were considered almost perfect reliability.
Absolute intra-observer and interobserver reliability of the CSA ratio measurement was evaluated by 95% limits of agreement (LOA) in Bland-Altman plots . The LOA was calculated as the mean difference between each paired measurement ± 1.96 × the standard deviation of the difference. In addition, the standard error of measurement (SEM) was calculated as the standard deviation of the paired measurements × (1–intra- or interclass coefficient) 1/2 , and minimal detectable change at the 95% confidence level (MDC 95 ), calculated as 1.96 × 2 1/2 × SEM .
3
Results
3.1
Comparison of participants’ characteristics and individual thigh muscle volume
Among the 40 older adults [20 males; mean (SD) age 73.3 (4.7) years; 20 with KOA grade ≤ 1 and 20 with grade ≥ 2], BMI was significantly greater and JKOM score overall and scores for pain, activity of daily living, and health perception were significantly higher with KOA grade ≥ 2 than grade ≤ 1, with no significant differences in age, height, and weight ( Table 1 ).
| Characteristics | KOA grade ≤ 1 | KOA grade ≥ 2 | P value |
|---|---|---|---|
| Age, years | 72.3 (3.5) | 74.3 (5.4) | 0.175 |
| Height, cm | 156.0 (9.7) | 155.1 (9.1) | 0.660 |
| Weight, kg | 62.0 (12.3) | 67.1 (6.2) | 0.116 |
| BMI, kg/m 2 | 25.4 (3.7) | 27.7 (2.1) | 0.026 |
| Japanese Knee Osteoarthritis Measure | |||
| Total score | 15.0 (5.0–27.3) | 33.5 (24.0–43.8) | 0.002 |
| Pain score | 8.0 (3.0–10.8) | 12.0 (8.3–15.0) | 0.008 |
| ADL score | 5.5 (2.0–12.5) | 17.0 (10.8–22.5) | 0.002 |
| Health perception score | 2.0 (1.0–3.0) | 4.0 (3.3–5.0) | 0.002 |
VM and SM volumes were significantly lower with KOA grade ≥ 2 than grade ≤ 1, with no significant differences in the other muscle volumes ( Table 2 ). However, in terms of total thigh, ratios of QM, VI, VM, and SM muscle volume to body weight were significantly lower with KOA grade ≥ 2 than grade ≤ 1, and the effect sizes were higher for the muscle-volume to body weight ratios as compared with muscle volumes alone (effect size = 0.35 for total thigh, 0.41 for QM, 0.32 for VI, 0.56 for VM, and 0.50 for SM ratios).
| Muscle | KOA grade ≤ 1 | KOA grade ≥ 2 | P value | Effect Size |
|---|---|---|---|---|
| Total thigh muscle volume, cm 3 | 647.3 (140.5) | 630.6 (107.9) | 0.676 | 0.07 |
| Ratio to BW, cm 3 /kg | 10.5 (1.6) | 9.4 (1.4) | 0.034 | 0.35 |
| QM volume, cm 3 | 453.8 (98.6) | 424.6 (73.8) | 0.296 | 0.17 |
| Ratio to BW, cm 3 /kg | 7.3 (1.2) | 6.4 (1.0) | 0.011 | 0.41 |
| HAM volume, cm 3 | 193.5 (51.3) | 206.1 (42.9) | 0.407 | 0.14 |
| Ratio to BW, cm 3 /kg | 3.2 (0.54) | 3.1 (0.54) | 0.582 | 0.09 |
| RF volume, cm 3 | 52.1 (12.9) | 55.0 (14.8) | 0.501 | 0.11 |
| Ratio to BW, mm 3 /kg | 84.2 (14.0) | 82.4 (19.9) | 0.752 | 0.05 |
| VL volume, cm 3 | 146.5 (128.1–183.6) | 142.4 (133.1–176.3) | 0.552 | −0.10 |
| Ratio to BW, mm 3 /kg | 238.5 (210.7–284.1) | 232.4 (200.7–261.8) | 0.175 | −0.21 |
| VI volume, cm 3 | 117.2 (108.4–143.3) | 130.5 (107.8–136.1) | 1.0 | 0.00 |
| Ratio to BW, mm 3 /kg | 206.2 (28.6) | 186.2 (31.6) | 0.048 | 0.32 |
| VM volume, cm 3 | 112.9 (30.0) | 93.0 (19.6) | 0.017 | 0.38 |
| Ratio to BW, mm 3 /kg | 182.1 (36.1) | 139.1 (28.4) | < 0.001 | 0.56 |
| BFL volume, cm 3 | 148.0 (36.9) | 143.4 (35.4) | 0.687 | 0.07 |
| Ratio to BW, mm 3 /kg | 235.3 (41.6) | 213.7 (46.7) | 0.141 | 0.24 |
| BFS volume, cm 3 | 31.2 (8.4) | 32.7 (10.5) | 0.618 | 0.08 |
| Ratio to BW, mm 3 /kg | 51.4 (15.8) | 49.0 (15.6) | 0.639 | 0.08 |
| ST volume, cm 3 | 175.6 (49.0) | 166.3 (32.2) | 0.481 | 0.12 |
| Ratio to BW, mm 3 /kg | 285.1 (72.4) | 249.5 (51.2) | 0.089 | 0.28 |
| SM volume, cm 3 | 99.0 (25.2) | 82.3 (22.1) | 0.032 | 0.34 |
| Ratio to BW, mm 3 /kg | 158.7 (33.2) | 123.0 (30.0) | 0.001 | 0.50 |
Related posts:
The Annals of Physical and Rehabilitation Medicine through the 2010s: A generalist journal of rehabilitation with a French touch
Bi-cephalic transcranial direct current stimulation combined with functional electrical stimulation for upper-limb stroke rehabilitation: A double-blind randomized controlled trial
New factors that affect quality of life in patients with aphasia
High sedentary behaviour and low physical activity levels at 12 months after cardiac rehabilitation: A prospective cohort study
Dosage for cost-effective exercise-based falls prevention programs for older people: A systematic review of economic evaluations
Could non-invasive brain stimulation help treat dysarthria? A single-case study
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
