The Journal of Arthroplasty xxx (2015) 1e7
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Volume Increases of the Gluteus Maximus, Gluteus Medius, and Thigh Muscles After Hip Arthroplasty Keisuke Uemura, MD a, Masaki Takao, MD, PhD b, Takashi Sakai, MD, PhD b, Takashi Nishii, MD, PhD a, Nobuhiko Sugano, MD, PhD a, * a b
Department of Orthopaedic Medical Engineering, Osaka University Graduate School of Medicine, Suita City, Osaka, Japan Department of Orthopaedic Surgery, Osaka University Graduate School of Medicine, Suita City, Osaka, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 September 2015 Received in revised form 15 October 2015 Accepted 26 October 2015 Available online XXX
Background: Muscle atrophy in osteoarthritis (OA) patients is expected to recover after total hip arthroplasty (THA) because of the increase in activity levels. Although some reports have shown an increase in the thigh muscles (Th) after THA, no reports of increases in the gluteal muscles and of analyses of the factors related to muscle recovery have been published. In this study, the changes in the gluteal and Th volumes after THA were quantitatively evaluated using computed tomography (CT) images, and the factors related to muscle recovery were analyzed. Methods: The subjects were 40 OA hips with CT images taken 3 weeks after THA (ﬁrst postoperative [postop] CT) and more than 2 years after THA (second postop CT). The cross-sectional areas of gluteus maximus (G-max), gluteus medius (G-med), and Th were measured in both CT images. The factors related to muscle recovery that were measured and evaluated were age, gender, days from operation, surgical approach, KellgreneLawrence grades, and hip functional score (Japanese Orthopaedics Association hip score). Results: All measured cross-sectional areas of the operated side increased signiﬁcantly from the ﬁrst postop CT to the second postop CT (G-max, 40%; G-med, 11%; and Th, 20%). In the analysis of the factors related to muscle recovery, the increase rate of G-max correlated signiﬁcantly with patient age, and the increase rate of G-med correlated signiﬁcantly with the increase in Japanese Orthopaedics Association hip score. Conclusions: Hip OA can lead to atrophy of the gluteal and Th, but the muscle volume can recover more than 2 years after THA. © 2015 Elsevier Inc. All rights reserved.
Keywords: muscle recovery hip arthroplasty cross-sectional area muscle atrophy CT images
It has been reported that atrophy of the muscles around the hip and thigh occurs in patients with hip osteoarthritis (OA) [1e6]. Total hip arthroplasty (THA) reduces pain, with a resulting increase in the quality of life and activity levels. The muscular strengths of the hip and thigh also recover after THA. There have been a few reports [7e9] that quantitatively evaluated the changes in hip and Th volumes after THA. These reports showed a signiﬁcant increase in the volumes of muscles around the hip and thigh (iliopsoas, adductors, hamstrings, and quadriceps femoris) [7e9]. However, no reports have shown a One or more of the authors of this paper have disclosed potential or pertinent conﬂicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical ﬁeld which may be perceived to have potential conﬂict of interest with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2015.10.036. * Reprint requests: Sugano Nobuhiko, Department of Orthopaedic Medical Engineering, Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita City, Osaka 565-0871, Japan. E-mail address: [email protected]
(N. Sugano). http://dx.doi.org/10.1016/j.arth.2015.10.036 0883-5403/© 2015 Elsevier Inc. All rights reserved.
signiﬁcant increase in the volumes of the gluteus maximus (G-max) and gluteus medius (G-med) after THA. Moreover, there have been no reports of the factors related to the increases in muscles volumes. The maximum follow-up of the previous studies was 2 years, which may not be enough for the G-max and G-med to recover, and we hypothesized that the period after THA is a factor related to muscle volume recovery. Therefore, if we observe the muscle volumes after THA for a longer period, we may ﬁnd signiﬁcant muscle volume recovery in the G-max and G-med. In this study, hip and Th volumes were quantitatively measured using computed tomography (CT) images taken more than 2 years after THA, the recovery of muscle volumes was evaluated, and factors related to muscle volume recovery were examined. Methods Forty patients (6 men and 34 women, mean age 58 years) who had undergone unilateral THA between 2002 and 2012 and then
K. Uemura et al. / The Journal of Arthroplasty xxx (2015) 1e7
underwent contralateral THA with an interval of more than 2 years were retrospectively selected as the subjects of this study (Table 1). During this period, CT images for THA navigation were routinely obtained preoperatively according to a standard protocol (General Electric Spiral scan; GE Medical System, Tokyo, Japan; 120 kV, 150 mA, slice thickness; 2.0 mm, 1-second scan time). Axial images from the iliac brim to the knee were taken. At 3 weeks after THA, postoperative (postop) CT images were obtained for evaluation with the same protocol (ﬁrst postop CT). The second postop CT was taken incidentally at a mean of 4 years (range, 2-8.5 years) after THA for navigation surgery for the contralateral hip. When the initial THA was performed, contralateral hip symptoms were negligible. However, at the time of the contralateral THA, the hip became symptomatic with the progression of OA. Patient data were erased from the CT images, and this research was approved by the Institutional Review Board at Osaka University Hospital. THA was performed through a posterolateral approach (PLA) in 33 cases and a direct anterior approach (DAA) in 7 cases. Hip resurfacing arthroplasty was performed in 4 patients, and conventional THA was performed in 36 patients. The KellgreneLawrence (KL) classiﬁcation  of the operated side was grade 3 or 4, and the grade of the contralateral side was 1-4 at the time of the initial THA (Table 1). Preoperative and postop hip function was evaluated by the Japanese Orthopaedic Association hip score (JOA H-S) . The JOA H-S consists of 4 subcategories: pain, range of motion, ability to walk (gait), and activities of daily living (Appendix 1). It has a maximum of 100 points and has been reported to have a strong correlation with the Harris Hip Score [12, 13]. The postop JOA H-S was measured when the second postop CT was taken. Muscle volume measurements were done on CT images with the use of 3D template software (Kyocera Medical, Osaka, Japan). Because leg length change and offset change may affect the measurements of the muscle volume, the ﬁrst postop CT and second postop CT were used in this study. The cross-sectional areas (CSAs) of the G-max, G-med, and Th of both hips on the axial plane of the CT images were measured as a surrogate for total muscle volume. To measure the 3 CSAs, the coordinate system of the pelvis and the femur was determined on each CT image using anatomic landmarks. The anterior pelvic plane (APP) was deﬁned as the plane touching the most anterior points of the bilateral anterior superior iliac spines and the pubic tubercles (Fig. 1). The CSA of G-max was measured on the plane perpendicular to APP through the bilateral most proximal points of the greater trochanters (Fig. 2). The CSA of G-med was measured on the plane perpendicular to APP through the bilateral anterior superior iliac spines (Fig. 2). The table top plane (TTP) of the femur was deﬁned as the plane touching the most posterior point of the proximal femur and bilateral femoral posterior condyles, with the vertical axis set as a line through the
Table 1 Patients' Characteristics. Parameter
Male/female (cases) Age (y)a Conventional THA/HRA (cases) Intervals between the 2 CT images (y)a PLA/DAA (cases) KL grade of the operated side (cases) KL grade of the contralateral side (cases)
6/34 58 (45-74) 36/4 4 (2-8.5) 33/7 Grade 3: 3, grade 4: 37 Grade 1: 3, grade 2: 8, grade 3: 19, grade 4: 10
CT, computed tomography; DAA, direct anterior approach; HRA, hip resurfacing arthroplasty; KL, KellgreneLawrence; PLA, posterolateral approach; THA, total hip arthroplasty. a Values are expressed as means (range).
trochanteric fossa and the knee center (Fig. 3). The CSA of Th was measured on the plane perpendicular to the TTP at the center of the femur, which was deﬁned as the center of the femoral head and the knee center (Fig. 4). After each slice to measure CSA was reconstructed, the border line of the muscle and the other tissue was traced manually, and the muscle volume was measured using imaging analysis software (ImageJ version 1.48; National Institutes of Health, Bethesda, MD). After identifying the border line of the muscles, the 3 CSAs were measured on the ﬁrst postop CT and the second postop CT and compared. The factors related to muscle recovery that were measured and evaluated were age, gender, days from operation, surgical approach (DAA or PLA), KL grade of the operated side (grade 3 or 4), KL grade of the contralateral side (grade 1, 2, 3 or 4), and the JOA H-S. To assess the positional change of the pelvis and the femur between the ﬁrst postop CT and the second postop CT, the ﬂexion angle and the abduction angle of both hips and pelvic sagittal tilt were also measured on both CT images. The hip ﬂexion angle was measured as the angle between the CT table and the TTP, and the hip abduction angle was measured as the angle between the craniocaudal axis of the APP and the vertical axis of the TTP. Pelvic sagittal tilt was measured as the angle between the APP and the CT table and was deﬁned as positive for posterior inclination. All statistical analyses were performed with SPSS statistical software, version 23 (IBM Japan, Tokyo, Japan). Bivariate analysis was carried out with the Wilcoxon signed-rank test and ManneWhitney U test, and correlations between the factors and CSA increase rates were assessed by Spearman’s rank correlation coefﬁcients. The correlation coefﬁcient (R) was also calculated. Intraobserver and interobserver reliabilities were tested using intraclass correlation coefﬁcients (ICCs) and their 95% CIs to assess the reliability of measuring the CSA of each muscle. Since a previous analysis of a similar study by Grimaldi et al  studying the CSAs of the muscle around the hip by magnetic resonance imaging (MRI) found an ICC of greater than 0.9, it was determined that a sample size of 20 subjects (half of the subjects) with 2 replicates and 2 evaluators per subject would achieve more than 80% power to detect an ICC value of 0.9 with the minimum value of the 1-sided 95% CI of 0.70. Values of P < .05 were considered to represent signiﬁcant differences and signiﬁcant correlations. Results All the measured CSAs of the operated side increased signiﬁcantly after THA. G-max increased from 25.1 cm2 (interquartile range [IQR], 21.5-29.5 cm2) to 35.1 cm2 (IQR, 30.6-40.2 cm2); G-med increased from 20.4 cm2 (IQR, 18.3-23.4 cm2) to 22.6 cm2 (IQR, 20.127.0 cm2); and Th increased from 76.5 cm2 (IQR, 69.2-85.4 cm2) to 92.0 cm2 (IQR, 84.6-102.7 cm2; all P < .01; Table 2). JOA H-S of the operated side improved signiﬁcantly from 48 (range, 25-72) to 79 (range, 63-91; P < .01). The average postop JOA H-S was relatively low because of OA of the contralateral hip. In the analysis of the factors related to the increase rate of the 3 muscles, gender, surgical approach, KL grade of the operated side, and KL grade of the contralateral side were not signiﬁcant factors (Appendix 2). Furthermore, no signiﬁcant correlations were seen between the increase rates of the 3 muscles and days from the operation. There were also no correlations between patient age and the CSA increase rates of G-med and Th. However, a negative correlation was found between the CSA increase rate of G-max and patient age (R ¼ 0.359, P ¼ .02). In the analysis of hip function and muscle recovery, there was no signiﬁcant correlation between the preoperative JOA H-S and the increase rate of each CSA. There was also no signiﬁcant correlation between the postop JOA H-S and the increase rate of each CSA. However, there was a signiﬁcant
K. Uemura et al. / The Journal of Arthroplasty xxx (2015) 1e7
Fig. 1. (A) Front view of the pelvis. (B) Lateral view of the pelvis. APP (yellow plane) is deﬁned as the plane including the most anterior points of the bilateral ASISs (red circles) and the pubic tubercles (blue circles). APP, anterior pelvic plane; ASIS, anterior superior iliac spine.
Fig. 2. (A) Front view of the pelvis and the femur. (B) Lateral view of the pelvis and the femur. The plane to measure the CSA of G-max (red plane) is created perpendicular to the APP (yellow plane) at a level including the bilateral ASISs (red circles). The plane to measure the CSA of G-med (blue plane) is created perpendicular to the APP (yellow plane) at a level including bilateral tips of the greater trochanters (blue circles). CSA, cross-sectional area; G-max, gluteus maximus; G-med, gluteus medius.
K. Uemura et al. / The Journal of Arthroplasty xxx (2015) 1e7
the second postop CT was 0.1 for the operated side and 1.0 for the contralateral side. For the hip abduction angle, the median change between the ﬁrst postop CT and the second postop CT was 1.0 for the operated side and 0.8 for the contralateral side (Table 3). The intraobserver and interobserver reliabilities, which were evaluated by ICC, were more than 0.9 for each parameter, indicating an acceptable level of reliability  (Table 4). Discussion
Fig. 3. (A) Lateral view of the femur. The yellow line indicates the table top plane (TTP), which includes the most posterior point of the proximal femur (yellow circle) and the bilateral posterior condyles (green circles). (B) Front view of the femur. The blue line indicates the vertical-axis of the TTP, which is set as a line through the trochanteric fossa (blue circle) and the knee center (red circle).
correlation between the improvement in JOA H-S from preoperative condition to postop condition and the CSA increase rate of G-med (R ¼ 0.402, P ¼ .01). There were no obvious correlations between the JOA H-S and the increase rates of the CSAs of G-max and Th. On the contralateral side, the CSAs of G-med and Th decreased signiﬁcantly from the ﬁrst postop CT to the second postop CT (both P < .01), whereas there was no signiﬁcant change in the CSA of G-max (Table 2). When CSAs were compared between the operated side and the contralateral side, all the CSAs on the operated side were signiﬁcantly smaller than those of the contralateral side on the ﬁrst postop CT (all P < .01). However, the CSAs of G-med and Th on the operated side became signiﬁcantly larger than those of the contralateral side on the second postop CT (both P < .01; Fig. 5). In the assessment of the position of the pelvis and the femur on the CT table, the median change of pelvic sagittal tilt between the ﬁrst postop CT and the second postop CT was 0.2 , and the median change of the hip ﬂexion angle between the ﬁrst postop CT and
This study was the ﬁrst to evaluate the recovery of muscle volume more than 2 years after THA and the ﬁrst study showing the increases in the CSAs of the G-max and G-med after THA. Moreover, this study was the ﬁrst to evaluate the factors related to the increases in muscle volumes. However, there are some limitations to this study. First, there were some changes in the position of the pelvis and the femur between the ﬁrst postop CT and second postop CT. With the positional change of the pelvis and the femur against the CT table, muscle tension may change and lead to changes in the CSAs of the measured muscles. However, the median change of pelvic tilt between the ﬁrst postop CT and second postop CT was only 0.2 , and the changes in the hip ﬂexion angle and the hip abduction angle of the operated side and contralateral side were no more than 1.0 . These differences may affect CSAs to some extent but can be considered small. However, to determine the exact extent, another simulation study would be needed. The second limitation is the small number of cases for analysis of related factors. Although the difference in muscle increase was examined by gender, KL grade of the operated side, KL grade of the contralateral side, and surgical approach, these failed to show a signiﬁcant correlation. However, because each of these groups only consisted of a small number, there may be some correlation between these factors and the increased rates in the CSAs if more cases were analyzed. The third limitation is that the subjects of this study were a heterogeneous cohort. Both conventional THA and hip resurfacing arthroplasty were included and both posterior and DAAs were included. Although we did not ﬁnd differences in muscle volume increase by these variables, a larger number of subjects are needed to generalize the results for any types of implants and approaches. It is assumed that the increase in the activity level and quality of life after THA results in the recovery of the muscles of the hip and thigh in OA patients. In the previous report that investigated unilateral OA patients, the CSAs of the adductors, vastus, and hamstrings increased signiﬁcantly from the preoperative CT to the CT image taken 2 years after THA . However, there were no signiﬁcant increases in the CSAs of the G-max and G-med, leading us to hypothesize that 2 years may not be enough for the gluteal muscles to recover. As expected, there were signiﬁcant increases of CSAs not only of Th, but also G-max and G-med in this study. In the analysis of factors related to the increased CSAs, no signiﬁcant correlation between the CSA increase rates and the period from the operation was seen. Because all patients included in this study had 2 CT images with an interval of more than 2 years, it can be assumed that a signiﬁcant increase in the muscle volume may have occurred within 2 years after THA and reached an equilibrium state. Various reasons can explain why the previous reports showed no signiﬁcant increases of the gluteal muscles. One of the reasons may be the differences in age and number of subjects. The number of subjects in the present study was larger, and the subjects were younger than in the previous reports. Rasch et al  evaluated muscle changes in 22 patients with a mean age of 67
K. Uemura et al. / The Journal of Arthroplasty xxx (2015) 1e7
Fig. 4. (A) Front view of the femur. Green circle indicates the center of the femur, which is deﬁned as the center of the femoral head center (orange circle) and the knee center (red circle). The blue line indicates the vertical axis of the femur. (B) Lateral view of the femur. The yellow line indicates the TTP. The plane to measure the CSA of Th (red plane) is created perpendicular to the TTP at a level including the center of the femur. Th, thigh muscles.
Table 2 Results. CSA (cm2)
First Postop CT
Second Postop CT
Increase Rate (%)
G-max OS G-med OS Th OS G-max CS G-med CS Th CS
25.1 20.4 76.5 32.6 24.0 87.5
35.1 22.6 92.0 32.2 22.2 82.9
40 11 20 1 8 5