Ultrasound Imaging Evaluation of Abdominal Muscles After Breast Reconstruction With a Unilateral Pedicled Transverse Rectus Abdominis Myocutaneous Flap
- L-J. Liaw, PT, MS, Institute of Allied Health Sciences and College of Medicine, National Cheng Kung University, Tainan, Taiwan; Department of Physical Therapy, College of Health Science, Kaohsiung Medical University, Kaohsiung, Taiwan; and Department of Rehabilitation Medicine, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan.
- S-D. Lin, MD, Division of Plastic and Reconstructive Surgery and Division of General Surgery, Department of Surgery, Kaohsiung Medical University Hospital.
- L-Y. Guo, PT, PhD, Department of Sports Medicine, College of Medicine, Kaohsiung Medical University.
- Y-Y. Hou, PhD, Department of Electrical Engineering, Far East University, Tainan, Taiwan.
- M-F. Hou, MD, Department of Surgery, Kaohsiung Medical University Hospital.
- A-T. Hsu, PT, PhD, Institute of Allied Health Sciences and Department of Physical Therapy, College of Medicine, National Cheng Kung University, No. 1 Ta-Hsueh Road, Tainan 701, Taiwan.
- Address all correspondence to Dr Hsu at: arthsu{at}mail.ncku.edu.tw.
Abstract
Background A muscle-sparing (MS) procedure using a full-width pedicled transverse rectus abdominis (RA) myocutaneous (TRAM) flap was developed to reduce abdominal morbidities after breast reconstruction. However, the effects of this procedure on the morphology of the remnant RA muscle and other abdominal muscles remain unclear.
Objective Ultrasound imaging was used to evaluate the morphology of the remnant RA muscle and other abdominal muscles in women with the MS pedicled TRAM flap procedure.
Design A case-control, cross-sectional design was used.
Methods Thirty-four women with an MS unilateral pedicled TRAM flap procedure after mastectomy (TRAM group) and 25 women who were healthy and matched for age (control group) participated. The curl-up test measured trunk flexor muscle strength. Ultrasound imaging measured the thickness of all abdominal muscles in all participants and the cross-sectional area of the RA muscle at rest and in an isometric position with the head raised in women in the TRAM group. Acoustic echogenicity and border visibility assessed the tissue composition of the remnant RA muscle.
Results Trunk flexor muscle strength was weaker in the TRAM group than in the control group. Compared with the remnant RA muscle in the contracted state, the remnant RA muscle in the relaxed state was thinner and had a smaller cross-sectional area. The remnant RA muscle in the relaxed state also was thinner, more echoic, and less visible than its contralateral counterpart. No differences in the thickness of the other abdominal muscles were found between the sides. The abdominal muscles in the TRAM group were smaller than those in the control group.
Limitation Because a prospective, longitudinal design was not used, a definite cause-effect relationship could not be determined.
Conclusions In women with an MS unilateral pedicled TRAM flap procedure, the remnant RA muscle retains its ability to change in size during contraction, albeit at reduced levels. Muscular atrophy occurs in other ipsilateral and contralateral abdominal muscles as well as the remnant RA muscle. Postoperative immobilization is the most likely cause of generalized weakness of the abdominal musculature.
Breast cancer is the most common neoplasm in women in western Europe and the United States.1–3 It also is the current leading cancer among women in Taiwan,4 where it is characterized by onset at a relatively younger median age (45–49 years) at the time of diagnosis.5,6 In the past 3 decades, breast reconstruction after mastectomy has become an integral part of breast cancer management.7 The use of autogenous tissue for patients who have undergone mastectomy has been shown to be psychologically advantageous because it helps patients overcome a sense of deformity after mastectomy.7–9 Ever since it was reported by Hartrampf et al in 1982,10 the pedicled transverse rectus abdominis (RA) myocutaneous (TRAM) flap procedure has gained in popularity and become the standard choice for breast reconstruction. In this procedure, the RA muscle is transferred through a tunnel under the skin to the mastectomy site with its proximal attachments intact, thereby preserving its original blood supply (the superior epigastric vessels) to the overlying skin and fat tissues that are used to form the breast.10 The TRAM flap breast reconstruction results in a better aesthetic appearance than implant reconstruction, with a soft and naturally ptotic breast.7
Various TRAM flap procedures, including the whole-width pedicled TRAM flap, free TRAM flap, muscle-sparing (MS) free TRAM flap, and deep inferior epigastric perforator flap, have been used. The guiding principle has progressed from total muscle harvest to total muscle preservation in an attempt to minimize untoward complications,7,11,12 such as abdominal bulges, abdominal asymmetry, and reduced trunk flexion capacity.13,14 Several outcome studies demonstrated that the pedicled TRAM flap procedure had an impact on abdominal muscle strength and function.14–17 Atisha and Alderman14 reported that up to 53% of women who had undergone pedicled TRAM flap surgery experienced weakness in the rectus and oblique muscles of the abdomen.
For reducing abdominal wall morbidities, the pedicled TRAM flap procedure has been modified through the use of MS techniques.18,19 The MS pedicled TRAM flap procedure involves preservation of the lateral strip of the RA muscle, both lateral and medial strips of the RA muscle, or the entire RA muscle.19 Nevertheless, the MS pedicled TRAM flap procedure has the potential risk of weakening the abdominal wall because it disrupts the integrity and the continuity of the RA muscle and its sheath and alters the insertion of the oblique muscles and, therefore, the biomechanical relationship between the RA muscle and adjacent structures. Furthermore, scarring in the RA muscle after dissection may further decrease the contractile potential of the muscle.20
Muscle strength is the most appropriate variable for evaluating the performance of a muscle. However, the strength of an individual muscle cannot be measured selectively because a muscle usually does not act alone but acts as part of a functional group. Muscle size measurement is a good alternative as an indirect measurement of strength. Muscle size measurement provides not only information about the status of atrophy or hypertrophy of a muscle but also evidence of treatment effectiveness. Ultrasound imaging is an accurate noninvasive alternative for measuring muscle size, that is, muscle cross-sectional area (CSA) or thickness, in muscle groups as well as individual muscles.21,22 Compared with computed tomography or magnetic resonance imaging, ultrasound imaging is less expensive and is capable of visualizing muscles in static as well as dynamic conditions.23 Ultrasound imaging used in the rehabilitation of neuromusculoskeletal disorders—known as rehabilitative ultrasound imaging (RUSI)—is a tool that physical therapists can use to assess the morphology and behavior of muscle and related soft tissues, such as thickening of the obliquus internus abdominis (OI) muscle and lateral sliding of the transversus abdominis (TrA) muscle, in the relaxed state and during specific activities.22,24,25
Several studies demonstrated the occurrence of muscular atrophy at the donor site of the RA muscle after breast reconstruction with a free TRAM flap or a deep inferior epigastric perforator flap by measuring the thickness of the muscle on the donor side and comparing it with the thickness of the muscle on the contralateral side.17,26,27 However, little is known about the size, at rest and during contraction, of the remaining muscle tissue on the donor side after MS pedicled TRAM flap harvest. Benditte-Klepetko et al26 reported that the remaining RA muscle in a subgroup of patients who underwent the pedicled TRAM flap procedure (n=3) had a smaller diameter at rest and a lack of motion within the contractile tissue during contraction. Hence, a more detailed evaluation of the size and residual function of the remnant RA muscle is necessary.
After the pedicled TRAM flap procedure, patients are usually instructed to immobilize their trunk for a few weeks to protect an abdominal wound from rupture. Immobilization may predispose all abdominal muscles to disuse atrophy, especially when such precautionary behavior becomes habitual. Neither the effects of the pedicled TRAM flap procedure on the RA muscle and the other abdominal muscles on the same side nor the effects of this procedure on the abdominal muscles on the intact side have been investigated.
Therefore, the purposes of this study were: (1) to appraise the morphology of the remnant RA muscle on the donor side and its contralateral counterpart at rest and changes in muscle size during isometric contraction by using RUSI, (2) to assess the thickness of the lateral abdominal muscles bilaterally at rest, and (3) to compare the thickness of the RA, obliquus externus abdominis (OE), OI, and TrA muscles on the intact side in women after the pedicled TRAM flap procedure (TRAM group) with that in women who were healthy and matched for age (control group).
Method
Participants
Between June 2009 and August 2011, 34 patients (mean age=42.6 years, SD=6.1, range=30–56) were recruited to participate in this study (TRAM group). They all had undergone immediate breast reconstruction with an MS unilateral pedicled TRAM flap after modified radical mastectomy for breast cancer at Kaohsiung Medical University Chung-Ho Memorial Hospital, Kaohsiung, Taiwan. Patients were recruited at least 6 months after the MS pedicled TRAM flap procedure and at least 1 month after the conclusion of the last chemotherapy course, if any. A control group comprising 25 female volunteers who were apparently healthy (mean age=40.4 years, SD=5.4, range=30–51) and had comparable socioeconomic backgrounds and physical conditions was recruited by convenience sampling from communities in Kaohsiung, Taiwan. Patients with the following conditions or symptoms that required medical attention were excluded from this study: distant metastasis; a history of low back pain or injury resulting in an inability to perform activities of daily living; scoliosis; spinal surgery; or any neurological, neuromuscular, rheumatological, or systemic diseases.
Group sample sizes were determined from a pilot study (10 women with the TRAM flap and 10 women who were healthy) that we conducted and were based on the power (80%) required to detect a difference of 0.1 cm in the thickness of the RA muscle, representing a large effect size (0.8) between 2 groups. The estimated target sample size for each group was 25 participants. Written informed consent was obtained from each participant before the study.
During the first week of hospitalization after surgery, the participants were allowed to perform active and active-assistive exercises with the upper extremities. They were given instruction in proper body mechanics to protect the abdominal wound, including positioning the head of the bed at a 45-degree angle or lying in a fetal position and using “log-rolling” techniques for moving safely in and out of bed. Ambulation for short distances began on postoperative day 3. Following discharge from the hospital, the participants were instructed to carry out all activities wearing an abdominal corset for support during the first 4 months of the postoperative period. Participants who had surgery involving the axillary region were also instructed to avoid lifting and carrying heavy loads to minimize the risk of lymphedema. None of the participants received any postoperative physical therapy targeted at improving or maintaining abdominal muscle function. All participants reported that they had returned to their normal daily activities and household cleaning chores. In addition, none of the participants, including those in the control group, had received any abdominal muscle training or engaged in any other regular exercises within 6 months before the study.
Surgical Technique for the Pedicled TRAM Flap Procedure
All participants in the TRAM group received the same surgical procedure, that is, the unilateral pedicled TRAM flap procedure with the contralateral RA muscle but without the use of a mesh for closure of the rectus sheath defect caused by the procedure. In 20 participants, the pedicles were elevated from the right side, and in the remaining 14 participants, they were elevated from the left side. Incisions were made in the lower abdomen (eFig. 1, dashed lines). Skin and subcutaneous tissues within the incisions were left attached to the lower portion of the RA muscle on the donor side. The pedicled TRAM flap was prepared with a muscle-splitting dissection by harvesting only the strip from the middle third and leaving the medial and lateral thirds intact. The middle strip of the RA muscle was transected at the level of the arcuate line (linea semicircularis), elevated to the costal margin, and tunneled beneath the upper abdominal skin to the chest on the opposite side, where it was revised and shaped into a breast (eFig. 1). During the flap elevation, all terminal branches of the lower 6 intercostal nerves were carefully preserved to maintain the integrity of innervation to the lateral strip of the RA muscle. After elevation of the pedicled RA muscle, the lateral portion of the rectus sheath was sutured to its medial counterpart by plication to bring the medial and lateral muscle segments close together. All procedures were performed by the same senior plastic surgeon (S-D.L.), who has performed more than 900 pedicled TRAM flap procedures.
Instrumentation of Ultrasound Imaging
Two ultrasound imaging systems—SSD 550 (7.5 MHz, Aloka Co, Tokyo, Japan) and SonoSite Titan (5-2 MHz, SonoSite Inc, Bothell, Washington)—were used for this study. The SSD 550 unit with a 38-mm transducer was used consistently throughout the study for all ultrasound imaging measurements except for the CSA of the intact RA muscle, for which the SonoSite Titan unit with a 60-mm transducer was used. The selection of each unit was based on the superior capability of the SSD 550 unit for differentiating the fascial borders of the remnant RA muscle and for muscle thickness measurements and the ability of the SonoSite Titan unit to cover the whole intact RA muscle for CSA measurements. The 60-mm curvilinear array transducer for the SonoSite Titan unit was used to measure the CSA of intact RA muscle in 25 women in the TRAM group and 25 women in the control group. For comparing the CSAs of the remnant RA muscle and the intact RA muscle, it was necessary to establish the equivalent reliability of the 2 ultrasound imaging units by obtaining images of intact abdominal muscles at 75 locations in 25 participants. The intraclass correlation coefficient (ICC [3,1]) of the thickness measurements of various abdominal muscles was .998 (95% confidence interval=.996–.998), indicating excellent consistency between the 2 ultrasound imaging systems.
Transducer Locations
For measuring the size of the RA muscle, the transducer was placed over the belly of the RA muscle and oriented transversely perpendicular to the midline of the abdomen at each of 3 locations identified with horizontal skin markings; the transducer was placed on the skin without applying any pressure. The RA muscle was scanned at the level of the costochondral junction of the ninth rib, halfway between the lower rib margin and the iliac crest, and 5 cm below the iliac crest. The largest anteroposterior thickness of the RA muscle was measured while the transducer was moved laterally from the midline until the cross-section of the whole muscle was centered on the image. Images of the RA muscle in women with the pedicled TRAM flap were acquired bilaterally at rest and during an isometric contraction to investigate the size of the remaining RA muscle on the donor side and its contralateral counterpart. Before acquisition of the image of the RA muscle in the contracted state, all women in the TRAM group were given instructions on how to perform a maximum curl-up and were allowed a single practice. The curl-up, an indicator of the strength of trunk flexion, was scored on the basis of the ability of the women to raise the trunk against gravity in the supine position with the knees flexed; scoring followed the criteria shown in eTable 1 with some modifications described elsewhere.28 For women in the age-matched control group, the RA muscle was scanned on 1 side at rest only (not in the contracted state).
Ultrasound images of the OE, OI, and TrA muscles were obtained in the relaxed state. The transducer was positioned transversely at a site halfway between the lower rib margin and the iliac crest along a vertical line passing through the anterior superior iliac spine.
Imaging Procedures and the Examiner
For standardization of the scanning procedure, the participants were positioned in a hook-lying position with extended upper extremities resting alongside the trunk. The hips were positioned in 60 degrees of flexion by placing pillows under the knees to orient the pelvis in a slightly posteriorly tilted position with the lower back touching the firm surface of the plinth. B-mode images focusing on the abdominal musculature were captured at the end of normal expiration to control the variability due to the influence of respiration. Image acquisition was performed 3 times at each location in the relaxed and contracted states. The order of image capture was randomized for location. All images were obtained by the same examiner, a senior physical therapist with 16 years of clinical experience, including 8 years in assessing abdominal muscles using RUSI.
Images Obtained During Isometric Contraction
Ultrasound images of the RA muscle were obtained first with the participants at rest and then during the holding phase at the end of the curl-up. During the curl-up, the participants were asked to initiate the movement by drawing in the lower abdomen before raising the head to 45 degrees without moving the trunk and then to hold the position isometrically for 3 to 5 seconds. All images were obtained at the end of expiration (eFig. 2). Participants were given instructions to “take a relaxed breath in and out, hold the breath at the end of expiration and gently draw your lower stomach in toward your lower back, then raise your head.” Movement errors, including breath holding during inspiration and excessive posterior pelvic tilt (monitored visually), were corrected. Not until the participants were able to perform this isometric submaximal contraction correctly at least twice was data collection started. A 1-minute rest period was allowed between contractions. Images were stored in a computer for offline processing and analyses.
Measurements of Muscle Size
Still ultrasound images were extracted offline. Measurements of the muscle CSA or linear dimensions (thickness and width) were analyzed offline by use of a public domain Java image processing and analysis program, ImageJ software (ImageJ, version 1.43u, National Institutes of Health, Bethesda, Maryland).29 Thickness for all abdominal muscles was defined as the distance between the superficial and the deep borders of the muscles (Fig. 1 and eFig. 3), and thickness measurements were obtained perpendicularly as the distance between the superficial and the deep fascial borders.21 The thickness of the RA muscle was obtained by measuring the greatest perpendicular distance, and the thickness of the lateral abdominal muscles was obtained at the middle of the images. The width of the RA muscle was defined as the distance between its medial and lateral margins, and the CSA (Fig. 2) was defined as the area enclosed by the borders of the RA muscle.
Ultrasound images of the donor (right) side (left panel) and the intact (left) side (right panel) of the rectus abdominis (RA) muscle at rest and upon contraction in a 45-year-old woman with a muscle-sparing unilateral pedicled transverse rectus abdominis myocutaneous flap. The medial and lateral edges of the RA muscle are represented by asterisks. The interval between the plus signs represents the thickest part of the RA muscle. The thickness of the RA muscle was measured at 3 levels: (A) upper=costochondral junction of the ninth rib, (B) middle=halfway between the lower rib margin and the iliac crest, (C) 5 cm below the iliac crest. The thickest part of the remnant RA muscle upon contraction can be seen in the lateral half of the middle location.
Ultrasound images of the intact (left) side of the rectus abdominis (RA) muscle at rest and upon contraction in a 45-year-old woman with a muscle-sparing unilateral pedicled transverse rectus abdominis myocutaneous flap. The cross-sectional area was measured by tracing around the inner border of the RA muscle at 3 levels: upper=costochondral junction of the ninth rib, middle=halfway between the lower rib margin and the iliac crest, lower=5 cm below the iliac crest. (A) Original image in relaxed or contracted state. (B) Superimposition of tracing on original image in relaxed or contracted state. The area in the tracing was measured.
Tissue Composition Measurements
Ultrasound imaging is capable of detecting tissue changes resulting from the degeneration of the architectural characteristics of a muscle on the basis of changes in its echogenicity.21,30 We used the qualitative evaluation tool developed by Strobel et al to assess the tissue composition of the RA muscle.30 The echogenicity of the muscle tissue and the visibility of its contours were compared with those of the reference muscle on the opposite side (intact RA muscle) and were graded from 0 to 2. For the visibility assessment, grade 0 was given when the muscle contours were clearly visible, grade 1 was given when they were partially visible, and grade 2 was given when they were no longer visible. For the echogenicity assessment, grade 0 was defined as isoechoic or hypoechoic in comparison with the intact RA muscle on the ultrasound image, grade 1 was defined as slightly more echoic, and grade 2 was defined as markedly more echoic.21,30 The echogenicity and visibility of the intact RA muscle were assigned grade 0 for comparison with the RA muscle on the donor side (donor RA muscle). A grade of 2 on at least 1 of the 2 scales is an indication of fatty atrophy.21
Data Analysis
All analyses were conducted with SPSS for Windows, version 14 (SPSS Inc, Chicago, Illinois). Statistical significance was set at P<.05. Descriptive statistics were computed for all demographic data, muscle size and percent change, and tissue composition measurements. The demographic data for the TRAM group and the control group were compared with either the Student t test or the chi-square test, and the Mann-Whitney U test was used for trunk flexion strength.
The ICC (3,1) was used to assess the level of consistency across 3 muscle thickness or CSA measurements at each location. The separate intrasession ICCs for interimage reliability within day were computed for each location of the RA muscle and the OE, OI, and TrA muscles bilaterally for the TRAM group but on only 1 side for the control group. The measurement error was examined by calculating the standard error of measurement and the minimal detectable change.
Muscle thickness, width, or CSA for each location was calculated from the average value obtained from the 3 images and was used for further analyses. Separate paired t tests were used to compare mean thickness or mean CSA measurements for the RA muscle in the contracted state with those in the relaxed state to determine whether differences in muscle thickness or CSA occurred during contraction. In addition, the behavior of the RA muscle during isometric contraction was investigated with ultrasound imaging to calculate the percent change in thickness and CSA in the TRAM group. The percent change in the thickness of the RA muscle was calculated as: [(contracted thickness − relaxed thickness)/relaxed thickness] × 100. Paired t tests also were used to compare the thickness of all abdominal muscles on the donor side with that on the contralateral (control) side at rest to determine the effects of the pedicled TRAM flap procedure on the RA muscle and lateral abdominal muscles on the donor side. In addition, Friedman tests were used to analyze differences in the ordinal data obtained with the 2 grading scales (from 0 to 2) for tissue composition in 3 locations of the donor RA muscle, and post hoc pair-wise comparisons were conducted with Wilcoxon tests. In addition, Wilcoxon tests were used to compare the echogenicity and visibility of the contours of the donor RA muscle with those of its contralateral counterpart.
Two-way analyses of variance with a mixed design (between-group factor: TRAM group or control group; within-subject factor: upper, middle, or lower location) were separately performed for the thickness and CSA of the RA muscle. Muscle thickness and CSA were normalized by body weight before group comparisons were made. When the main effect was significant, pair-wise comparisons were performed with the Bonferroni adjustment. Independent t tests were used to compare differences in the thickness of the OE, OI, and TrA muscles on the intact side in the TRAM group and the corresponding side in the control group.
Role of the Funding Source
This study was supported by grants from the National Science Council (NSC 98-2314-B-037-008-MY2) and Kaohsiung Medical University (KMU-Q098023) in Taiwan.
Results
Descriptive statistics for all demographic data for both groups are shown in Table 1. The groups were similar in age, height, weight, body mass index, and parity, but the trunk was weaker in the TRAM group than in the control group (P<.001) (Tab. 1). The mean follow-up time in the TRAM group was 10.1 months (range=6.1–31.1). Most of the women in the TRAM group (91.2%) could only raise their head or clear the spine of the scapula off the table (score of 2 or 2.5), 2 women (6%) could clear the inferior angle of the scapula off the table (score of 3), and only 1 woman (3%) could lift the inferior angle of the scapula with the hands clasped behind the head (score of 5). Of the 25 women in the control group, 14 (56%), 3 (12%), 4 (16%), and 4 (16%) had curl-up performance scores of 5, 4, 3, and 2.5, respectively.
Characteristics of Study Participantsa
Interimage Reliability
Across the 3 locations in the TRAM group and the control group, the ICC (3,1) for the thickness of the RA muscle ranged from .89 to .96 and the ICC (3,1) for the CSA of the RA muscle ranged from .92 to .98 (eTab. 2). All reliability coefficients indicated that muscle size measurements appeared to have excellent (≥.75) interimage reliability31 as well as acceptable measurement errors (all values for standard error of measurement divided by mean were <10%).32
Thickness, Width, and CSA of Abdominal Muscles in the TRAM Group
Thickness measurements were obtained for all 34 women in the TRAM group, but CSA measurements were obtained for only 25 women. Table 2 shows descriptive statistics for muscle thickness, width, and CSA and results of paired t tests comparing values for the donor side with those for the intact side. The thickness of the RA muscle at the 3 locations ranged from 0.40 to 0.43 cm on the donor side and from 0.72 to 0.79 cm on the intact side (Tab. 2). On average, the thickness and CSA measurements at 3 locations of the RA muscle on the donor side were approximately 58% and 26%, respectively, those on the intact side (Tab. 2). A comparison of RA muscle thickness and CSA measurements at 3 locations on both sides indicated that the values for the RA muscle on the donor side were lower than those on the intact side (P<.001). However, there were no significant differences in the mean thickness of the OE, OI, and TrA muscles between the sides.
Descriptive Statistics for Averaged Muscle Measurements in Women With the Muscle-Sparing Unilateral Pedicled TRAM Flap Procedurea
Muscle Thickness and CSA During Contraction in the Donor RA Muscle
On the basis of the results obtained from 3 locations of the RA muscle, the mean percent changes in the thickness and CSA of the RA muscle on the donor side were determined to be 17.22% and 19.79%, respectively; those on the intact side were determined to be 27.74% and 21.51%, respectively. The thickness and CSA of the RA muscle on the donor side in the relaxed state were significantly different from those in the contracted state (P≤.001) (eTab. 3).
Size of Abdominal Muscles on the Intact Side in the TRAM Group
The thickness and CSA of the RA muscle on the intact side revealed significant main effects of groups and locations (P<.05) but no interaction of group and location (P>.05) (Tab. 3). Similar results were found for the thickness of the lateral abdominal musculature (Tab. 3). In the TRAM group, the means for the thickness and CSA of the intact RA muscle at the 3 locations measured in a superior-inferior sequence ranged from 0.72 cm (SD=0.19) to 0.79 cm (SD=0.16) and from 2.89 cm2 (SD=0.84) to 3.72 cm2 (SD=0.94), respectively. The corresponding values for the thickness and CSA of the RA muscle in the control group ranged from 0.82 cm (SD=0.16) to 0.89 cm (SD=0.12) and from 3.44 cm2 (SD=0.86) to 4.55 cm2 (SD=0.92), respectively (Tab. 3).
Comparison of Muscle Sizes in Women Who Had Breast Reconstructiona and Women Who Were Healthyb
The intact RA muscle in the TRAM group was thinner and its CSA was smaller than in the control group both before (P<.05) and after (P<.05) normalization for body weight, especially for measurements at the middle and lower locations (P≤.002) (Tab. 3). In addition, there were main effects of location (P<.001); the largest thickness and CSA measurements were obtained at the lower location of the RA muscle, and the smallest were obtained at the upper location of the RA muscle. Likewise, the mean thickness of the OE, OI, and TrA muscles on the intact side also was significantly smaller in the TRAM group than in the control group (P<.01).
Tissue Composition
Data for the visibility of the outer contour and echogenicity and the sum scores for visibility and echogenicity at the 3 locations of the RA muscle in the TRAM group are shown in Table 4. Friedman tests were conducted to evaluate differences in the median values for visibility and echogenicity and the sum scores for visibility and echogenicity at the upper, middle, and lower locations, and no differences were found (P>.05). However, the remnant RA muscle was less visible and more echoic than its contralateral counterpart (all P<.001). Of the 34 women in the TRAM group, 15 had at least 1 location with a grade of 2 on at least 1 of the 2 scales for tissue composition. All women in the TRAM group had at least 1 location that was not clearly visible, but of the 34 women in the TRAM group, 2 (6%) had the same echogenicity score for each location of the remnant RA muscle and its contralateral counterpart.
Visibility of Muscle Contours and Echogenicity of the Remnant Rectus Abdominis Muscle at 3 Locations (n=34)a
Discussion
The present study is the first to simultaneously appraise the tissue composition and the size of the remnant RA muscle at rest and during isometric contraction with RUSI in women who underwent breast reconstruction with the MS pedicled TRAM flap procedure. One of the major goals of the present study was to ascertain the fate of the remaining RA muscle on the donor side after the removal of its middle third for breast reconstruction. We also examined the thickness of the other abdominal muscles. We found that after the MS pedicled TRAM flap procedure, the remaining RA muscle became thinner and more echoic than its contralateral (intact) counterpart in most women (94%) at rest and larger upon contraction from a relaxed state. We also found that women in the TRAM flap group had smaller RA, OE, OI, and TrA muscles on the intact side than women in the control group.
We found a decrease in the thickness of the remnant RA muscle relative to its contralateral (intact) counterpart (Tab. 2). Furthermore, ultrasound imaging revealed a more hyperechoic appearance of the remnant RA muscle than of its contralateral counterpart as a consequence of an increase in the intramuscular fat content or scarring. Atrophy and higher fat content are well-known signs of denervated muscle.33 These findings could be consequential to the damage of segmental nerves to the remaining RA muscle during dissection or direct mechanical damage of the muscle during intramuscular dissection. These results are consistent with those of a previous study of the MS pedicled TRAM flap procedure.26 Benditte-Klepetko et al26 also found that the middle and lower segments of the remnant RA muscle were thinner than the muscle on the intact side in 3 patients who had undergone a unilateral pedicled TRAM flap procedure. In the present study, no differences in muscle thickness were found in the upper, middle, and lower portions of the donor RA muscle. Overall, the findings of the present study are consistent with those of Benditte-Klepetko et al, except that we included a larger sample size and added muscle CSA measurements for describing changes between the contracted state and the relaxed state.
Changes in muscle thickness are often interpreted as changes in muscle contraction. Positive (curvilinear or linear) correlations between changes in muscle thickness and the level of electromyographic activity were observed during isometric contraction of the TrA and OI muscles.34,35 However, no extant studies have explored such a relationship for the RA muscle. In the present study, we demonstrated a significant increase in the thickness of the remnant RA muscle at the end of a curl-up. However, the assumption of a direct influence on the retention or recovery of the contractility of the remnant RA muscle in women who have undergone the TRAM flap procedure should be made with caution because numerous factors unrelated to RA muscle contraction may affect its thickness.21,25,36 Additional investigations with simultaneous electromyography (preferably with intramuscular wire electrodes) and RUSI are needed to ascertain the relationship between the thickness and the contractility of the remnant RA muscle. Brown and McGill36 recently reported that the high laterally directed forces generated by the adjacent muscles (OE, OI, and TrA) appeared to create lateral movements and thinning of the RA muscle during abdominal contraction. Additional studies are needed to investigate changes in RA muscle morphology in relation to the activity levels of the RA muscle and adjacent muscles and the associated movements of the medial and lateral fascial margins of the RA muscle during contraction.
Currently, little is known about differences in the thickness of lateral abdominal muscles between the donor side and the intact side in patients undergoing TRAM flap breast reconstruction. Blondeel et al37 used computed tomography or magnetic resonance imaging to investigate the abdominal wall after the free TRAM flap procedure. They demonstrated a 25% reduction in the thickness of both oblique muscles relative to their contralateral counterparts after the unilateral free TRAM flap procedure.37 However, we found no differences in the thickness of the OE, OI, and TrA muscles between the donor side and the intact side, in contradiction with the findings of Blondeel et al. The interpretation of such results may be difficult because of the complicated nature of the anatomic and biomechanical interconnections among structures located at the anterior and lateral abdominal wall. Increased tension resulting from the decreased width of the ipsilateral rectus fascia may simultaneously cause overstretching of all contralateral as well as ipsilateral lateral abdominal muscles.
Using ultrasonography for patients in a standing position, Suominen et al17 demonstrated that the midline was retracted to the donor side by 1 cm, on average, in patients who had the pedicled TRAM flap procedure and were evaluated 6 months to 2 years after surgery. In addition, Blondeel et al37 used computed tomography or magnetic resonance imaging for patients who had the pedicled TRAM flap procedure and found that the linea alba was deviated to the donor side and that the insertion lines with the medial border of the oblique muscles were shifted medially. These results were consistent with those of previous studies38,39 and indicated that the medial margin of the contralateral RA muscle was pulled over the midline. Although we did not document the position of the linea alba, the examiner in the present study also found a tendency for the midline to be shifted to the donor side in most participants (19/25) during ultrasound image capture.
Postsurgical immobilization has long been suspected as the culprit for abdominal muscle atrophy, especially for muscles not damaged by the surgery. According to Momeni et al, the percentage of abdominoplasty patients who subjectively perceived a decrease in abdominal muscle strength postoperatively was the same as that of patients who underwent breast reconstruction with an MS-2 TRAM flap (the use of a small central slip of muscle while preserving the medial and lateral muscle segments).40 This result implied that the decreased abdominal strength was due not only to partial loss of the RA muscle but also to disuse atrophy resulting from postoperative inactivity.40 However, to clarify the effect of abdominoplasty on the strength of the abdominal wall, objective measures are necessary. For participants in the present study, great care was taken to protect the wound in the abdominal wall after the MS unilateral pedicled TRAM flap procedure. These precautions included lying with the trunk, hip, and knee in flexion simultaneously for 7 days postsurgery and wearing an abdominal corset for 4 to 6 months. These instructions greatly limited trunk movements and the activity levels of the participants and resulted in generalized disuse atrophy of all abdominal muscles on the donor side as well as the intact side. Additionally, holding a muscle in a shortened position exacerbates disuse atrophy.41
The results of the present study support the claim that the thickness or CSA of all abdominal muscles on the intact side in women in the TRAM group was smaller than that of women in the control group. These results should be treated circumspectly because, with exercise, this type of atrophy can be prevented and can be reversed after it has happened. In the present study, the size of the RA muscle was measured, normalized by body weight, and compared with the size of the contralateral RA muscle and the size of the RA muscle in a matched control group in a cross-sectional design; a prospective, longitudinal study is required to confirm the results.
The present study had several limitations. Ultrasound images were arranged in a random order for analyses. Information regarding a participant's name, group allocation, and the side of surgery was not included on the images or in their file names. However, the participant's identification number was not masked. Because the same therapist was responsible for both collecting and analyzing ultrasound images, subjective bias from the examiner could not be eliminated completely. For the CSA measurement of the RA muscle, eliminating bias in analyses for the TRAM group was difficult for 2 reasons: different ultrasound imaging units were used for the remnant RA muscle (SSD 550 with a 38-mm transducer; 7.5 MHz) and the intact RA muscle (SonoSite Titan with a 60-mm transducer; 5-2 MHz), and the ultrasound imaging appearance of the remnant RA muscle was readily distinguishable from that of the intact RA muscle. Therefore, a more elaborate process should be arranged in future research to eliminate any possibility of subjective bias. Additionally, although the intrarater reliability of muscle thickness or CSA measurements with ultrasound imaging in the control group was good, the fascial borders of the RA muscle appeared to become less clear in the TRAM group. Data on intrarater and interrater reliability across different days in women who undergo the TRAM flap procedure are needed.
In addition, we found that the RA muscle and the lateral abdominal muscles on the intact side in the TRAM group were smaller than those in the control group; however, it is possible that muscle size was influenced by occupation, exercise levels, or inactivity while on adjuvant chemotherapy. We collected no preoperative baseline data; therefore, the assessment at the time of follow-up was based on data obtained from women who were healthy rather than data obtained longitudinally from women who were scheduled to undergo the MS pedicled TRAM flap procedure, preferably before receiving the mastectomy. However, considering the psychological and physical stresses of breast cancer and impending surgery, the request for an abdominal strength measurement appears to be unreasonable.
Another limitation is that a decrease in the thickness of the lateral abdominal muscles may be related to overstretching of these muscles resulting from the fascial suture technique and the tightness with which the anterior fascia is closed. Additional research is needed to clarify this issue by documenting distances between the midline and the insertion line of the oblique muscles on both sides and examining the relationships between these distances and the thickness and strength of the lateral abdominal muscles.
Finally, most women in the present study were examined at a 10-month follow-up; 10 months may not be long enough to represent the possible long-term postoperative results or the full recovery potential of the abdominal muscles because the peripheral nerves are known to continue regenerating or sprouting from other intact nerves more than 1 year after injury.42 Additional research with a longer follow-up period may be necessary for ultrasound imaging to document preoperative and postoperative muscle thickness in women undergoing the MS pedicled TRAM flap procedure.
Clinical Implications
The ultrasound evaluation used in the present study showed that all abdominal muscles in women who have undergone the pedicled TRAM flap procedure may have disuse atrophy. Our results indicated that the prime movers of trunk flexion and rotation are both affected after the TRAM flap procedure. Our findings may help physical therapists find ways to effectively solve or mitigate problems in patients undergoing this procedure. Because the strength of the RA muscle cannot be evaluated individually or separately from the other abdominal muscles, ultrasound imaging (or RUSI) is an appropriate tool for evaluating muscle morphology and changes in muscle size during the contraction of individual muscles. Early and appropriately planned interventions by physical therapists are necessary to prevent or lessen the potential for weakness and wasting of the intact abdominal muscles while alleviating potential abdominal morbidity in patients undergoing the pedicled TRAM flap procedure.
Conclusion
Our results demonstrated that, in women 10 months after the MS pedicled TRAM flap procedure, the remnant RA muscle retained or recovered its capability to increase thickness and CSA upon isometric contraction from a relaxed state, but a decrease in relaxed muscle thickness and fatty atrophy were evident. In addition, a reduction in muscle size was found in the other intact abdominal muscles bilaterally; this finding may have been due to postoperative inactivation. Therefore, additional research is essential to explore exercise prescriptions and the efficacy of physical therapy for preventing or reducing weakness of the abdominal muscles in women undergoing the MS TRAM flap procedure.
Footnotes
-
Ms Liaw, Dr M-F. Hou, and Dr Hsu provided concept/idea/research design. Ms Liaw and Dr Hsu provided writing and fund procurement. Ms Liaw and Dr Guo provided data collection. Ms Liaw, Dr Y-Y. Hou, and Dr Hsu provided data analysis. Ms Liaw provided project management. Dr Lin provided study participants. Dr Guo provided facilities/equipment. Dr Lin, Dr Guo, and Dr Y-Y. Hou provided consultation (including review of manuscript before submission). The authors thank all of the patients and volunteers who were healthy for participating in this study.
-
This study was approved by the Institutional Review Board of Kaohsiung Medical University Chung-Ho Memorial Hospital, Kaohsiung, Taiwan.
-
This study was supported by grants from the National Science Council (NSC 98-2314-B-037-008-MY2) and Kaohsiung Medical University (KMU-Q098023) in Taiwan.
- Received February 15, 2012.
- Accepted October 9, 2012.
- © 2013 American Physical Therapy Association