 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 2000 The Journal of Bone and Joint Surgery, Inc. Electromyographic and Gait Analysis of Forty-three Patients After Rotationplasty* , MÜNSTER, GERMANYInvestigation performed at the Department of Orthopedics and the Motion Analysis Laboratory, Westfälische Wilhelms-Universität Münster, Münster
Background: Rotationplasty is considered to be a treatment option for patients who have had a primary malignant bone tumor of the distal part of the femur or the proximal part of the tibia. The present study was performed to evaluate the muscle activity, the kinetics (range of motion of the hip and knee joints), and the kinematics (joint moments) after rotationplasty and to determine whether there was an association between these parameters and the functional outcome. Methods: Forty-three patients who had been managed with rotationplasty for the treatment of a femoral or tibial bone tumor were evaluated clinically and functionally. The mean age (and standard deviation) at the time of follow-up was 24.4 ± 10.7 years (range, eight to sixty-eight years), the mean age at the time of the procedure was 17.8 ± 10.2 years (range, seven to sixty-three years), and the mean duration of follow-up was 6.7 ± 4.9 years (range, 0.7 to eighteen years). Instrumented gait and electromyographic analyses were performed. The qualitative data were compared with the functional outcome, which was determined with the functional evaluation score of the Musculoskeletal Tumor Society. Results: Gait analysis revealed a fairly normal walking pattern with a slight limp and a lateral lean of the trunk over the ipsilateral limb that led to a reduced joint moment in the hip (moment on involved side, 68 percent [compared with a control group]; moment on uninvolved side, 81 percent). The ranges of motion of the hips (uninvolved side, 42.0 ± 8.2 degrees; involved side, 42.4 ± 8.0 degrees) and the knees (uninvolved side, 59.7 ± 5.0 degrees; involved side [former ankle joint], 58.1 ± 11.6 degrees) were symmetrical even though the knee-motion pattern of the involved limb indicated a slightly reduced extensor mechanism in 51 percent (twenty-two) and a markedly reduced extensor mechanism in 35 percent (fifteen) of the forty-three patients. Electromyography revealed function of the muscles of the involved limb, with comparable amplitudes in the involved and uninvolved limbs. The leg muscles of the involved limb were active in the stance phase (the soleus and the lateral and medial heads of the gastrocnemius) and the swing phase (the peroneus longus and the tibialis anterior) according to their function in relation to the new knee joint. The patients had a good functional result, with a mean score of 23.9 ± 2.7 of 30 points. With the numbers available for study, we could not show the duration of follow-up to be related to the overall outcome, but the age at the time of the operation was related to the total functional score as well as to gait and walking ability (p < 0.05). Conclusions: The results of the electromyographic and gait analyses demonstrated good functional restoration of gait following rotationplasty.
The improved rate of survival, after effective adjuvant or neoadjuvant chemotherapy, of patients who have had a primary malignant bone tumor of the distal part of the femur or the proximal part of the tibia has increased the need for optimum function of the limb after operative treatment. Disarticulation of the hip and above-the-knee amputation are extremely disabling and can be avoided in most cases. Rotationplasty is considered to be a good alternative to endoprosthetic replacement or above-the-knee amputation2,10, and it is indicated when an extensive tumor mass of the distal part of the femur has affected the extensor mechanism of the knee joint. Furthermore, the procedure is advantageous for children younger than ten years old, in whom extendable endoprostheses cannot be used because of the longitudinal extension of the tumor, the higher risk of infection, and the expected number of subsequent operations. When the tumor invades the proximal part of the tibia, the procedure can be performed without modification, as reported by Van Nes20 in 1950. Rotationplasty represents a functional replacement of the knee joint after limb salvage with en bloc resection of a malignant tumor or another major pathological lesion necessitating wide margins of resection that include the knee joint. The tibia and foot segment is rotated 180 degrees and is reattached to the remaining femur while preserving the sciatic nerve. The result is a shortened limb with the ankle joint at the position of, and with the function of, the former knee joint9. This procedure was first described in 1930 by Borggreve1 for the treatment of an extensively shortened limb following tuberculous ankylosis of the knee joint. In 1950, Van Nes20 used this technique to manage patients who had congenital defects of the femur. In the 1970s, Salzer and Knahr16 started managing patients who had a high-grade malignant tumor in the knee region with this procedure as an alternative to high above-the-knee amputation. Modified techniques allowed the treatment of tumors not only in the distal part of the femur but also in the proximal part of the tibia and femur or in the entire femur, including the acetabulum4,22. Since then, rotationplasty has been an option for the operative treatment of tumors. After rotationplasty, the ankle joint takes over the function of the former knee joint. The remaining leg muscles, the tibialis anterior and the medial and lateral heads of the gastrocnemius, are reattached to the origins of the thigh muscles and are capable of moving the exoprosthesis. The remaining muscles, such as the extensor mechanism of the former knee joint, the adductor group, and the hip and back muscles, are expected to show electromyographic activity appropriate to support the new anatomical situation. The main advantage of rotationplasty is that patients retain an actively controlled knee joint. Although a prosthesis must be worn, these patients achieve a high level of activity in daily life and sports19. Unlike the extendable tumor prosthesis that is sometimes implanted in the immature patient and that must be adjusted operatively to allow for limb growth, rotationplasty requires only a single operation and results in a highly functional limb12. The resulting stump is able to control the prosthesis, as the foot is able to carry the load that is transferred to the plantar and dorsal aspects of the foot9. The problem of phantom pain that is sometimes observed after amputation is minor or not evident after rotationplasty. There have been few investigations that have included electromyographic analyses, and those that have been reported have involved only a few patients14. To our knowledge, it has not yet been demonstrated, in a large population, whether the activity of the muscles in their new anatomical and functional location is comparable with the activity of the muscles of the original knee joint. Therefore, the purpose of the present study was to evaluate the clinical and functional outcome after rotationplasty. Specifically, the investigation included analysis of the gait pattern with respect to the function of the hip, knee, and ankle joints and the muscle coordination in the altered anatomical situation.
Patients Between November 1995 and February 1997, forty-three patients (nineteen female and twenty-four male) who had been treated because of a malignant tumor in the knee region were evaluated. One patient had a proximally located tibial tumor, and the remaining forty-two had a femoral tumor. Because the location was close to the knee, all of the patients received a type-A1 rotationplasty according to the classification described by Winkelmann21,23 (Fig. 1), which is analogous to the Van Nes rotationplasty20. The most common tumor was high-grade malignant osteosarcoma (thirty-nine patients) followed by Ewing sarcoma (two patients) and malignant fibrous histiocytoma (two patients). The mean age (and standard deviation) of the patients at the time of the operation was 17.8 ± 10.2 years (range, seven to sixty-three years). The mean height was 162 ± 18.0 centimeters (range, 123 to 190 centimeters), and the mean weight was 53.4 ± 18.4 kilograms (range, twenty-one to eighty-two kilograms). The mean age of the patients at the time of the most recent follow-up was 24.4 ± 10.7 years (range, eight to sixty-eight years). The mean duration of follow-up was 6.7 ± 4.9 years (range, 0.7 to eighteen years) (Table I). All of the patients had a satisfactory prosthetic fitting and had finished chemotherapy treatment.
To exclude false-negative and false-positive findings for the patients who had a short duration of follow-up, we divided the total group into two groups: one with a follow-up of less than two years (nine patients) and another with a follow-up of more than two years (thirty-four patients) (Tables I and III).
The patients filled out a questionnaire about their current health status and subjective satisfaction with the treatment outcome. This questionnaire was used to perform a functional evaluation and to determine the Musculoskeletal Tumor Society score for function described by Enneking et al.5. This score is based on six categories—pain, emotional acceptance, function, use of orthopaedic supports, walking ability, and appearance of gait—with a maximum of 5 points for each (Table II). The clinical examination involved range-of-motion measurements of the hip, knee, and ankle joints.
Operative Procedure The type-A1 rotationplasty was performed according to the method of Winkelmann21,23, which was based on the technique described by Van Nes20. It is applicable for patients who have a malignant tumor in the distal part of the femur when the hip joint is not affected. The procedure is applicable for tumors in the proximal part of the tibia only when the tumor does not extend past the proximal third of the tibia. The osteotomy is performed in the proximal third of the femur, distal to the lesser trochanter, and in the proximal part of the tibia, distal to the tibial tuberosity. The foot is rotated 180 degrees, and the tibia is reattached to the remaining femur. The foot is fitted in a prosthetic shaft and is attached to a conventional foot prosthesis. Proximally, the prosthesis wraps around the thigh and is connected to the foot component with a hinge joint. The operation results in a shortened limb with a single functional joint in which the ankle takes over the function of the former knee.
Kinetic and Kinematic Analysis The patients were asked to walk at a freely chosen, comfortable speed. As soon as a consistent cadence of strides was reached, six trials were recorded and subsequently analyzed. The active range of motion of the hip and knee joints as well as the movements of the trunk and pelvis were determined. The kinematic information from the motion-analysis system, in combination with the ground-reaction forces, was used to estimate the moments acting on the hip joint. The center of rotation of the hip joint was determined with consideration of the individual geometry of the pelvis derived from three-dimensional stereo radiographs according to the method described by Selvik17. Additionally, the torque acting on the knee and ankle joints was determined.
Electromyography
The patients were asked to walk along a twelve-meter-long walkway. The signal was transferred to the electromyographic amplifier with an umbilical cord that enabled the patient to walk freely along the entire walkway. The signals were recorded on a chart-recorder at a paper speed of sixty millimeters per second. For analysis and evaluation, the electromyographic records of each patient were visually inspected. Four cycles of electromyographic activity that appeared to be characteristic with respect to amplitude and duration were selected. Activity of the gluteus medius was used to define the stance phase. Because of pronounced inter-subject differences in the amplitude of the electromyographic signals (due to the thickness of the subcutaneous tissues or skin impedance, or both), a semiquantitative evaluation procedure was chosen. On a scale of 1 to 5 points, 3 points indicated a normal level of activity in the walking cycle of healthy subjects. A score of 0 points represented no activity; 1 point, a decreased level of activity; 2 points, a slightly reduced level of activity; 4 points, a slightly increased level of activity; and 5 points, an increased level of activity. In order to ensure comparability, the scaling was performed by one of us (J. S.) for all patients according to the same criteria.
Statistical Analyses
Kinematics Lateral trunk movement in the frontal plane revealed a slightly increased lateral lean of the trunk toward the stance limb on the involved side (4.9 degrees) as compared with that on the contralateral side (1.5 degrees) (p < 0.0001). Pelvic tilting was significantly more pronounced on the uninvolved side (1.3 degrees) than it was on the involved side (0.2 degree) (p = 0.0023). The range of motion of the hip joint in the sagittal plane was almost equal on both sides (Table V). The maximum range of motion of the knee joint in the sagittal plane on the uninvolved side (59.7 degrees) was similar to that on the involved side (the former ankle joint) (58.1 degrees). However, the overall movement pattern revealed marked differences in some cases. All but two patients had physiological flexion of the knee (about 15 degrees15) on the uninvolved side at mid-stance (Fig. 2, A). However, only six (14 percent) of the forty-three patients had full physiological flexion of the former ankle joint on the involved side at mid-stance. Twenty-two patients (51 percent) had slightly reduced flexion, ranging from 7.5 to 12.5 degrees, at mid-stance (Fig. 2, B), and fifteen patients (35 percent) had a marked reduction of flexion, to 7 degrees or less (Fig. 2, C). Thirty patients had reduced extension of the former ankle joint at the beginning of the stance phase; fifteen had a reduction to between 5 and 10 degrees and fifteen, to between 11 and 15 degrees. All but two patients had full extension of the knee on the uninvolved side.
Kinetics The external hip moment in the frontal plane was normalized to body weight and height and revealed an unloading that was more pronounced on the involved side than on the uninvolved side (68 and 81 percent as compared with a control group investigated in our laboratory). This difference was highly significant (p = 0.0085) even though not all of the patients unloaded the involved limb. In the sagittal plane, we found that the extension moment of the new knee (the former ankle joint) persisted throughout the entire stance phase (with resultant hyperextension of the new knee) in two patients. Fifteen patients had an exclusive flexion moment, revealing a lack of extension on the involved side. On the uninvolved side, all of the patients had a normal knee-joint flexion-extension moment except for two who had a reduced extension moment. In thirteen of thirty-eight patients, the external dorsiflexion torque at the ankle of the prosthetic foot was diminished to 50 percent or less of that at the uninvolved ankle (100 percent) (Fig. 3).
Electromyography The electromyographic recordings of the hip muscles showed no marked differences between the involved and uninvolved sides and could be described as normal (Table VI). Only the activity level of the gluteus maximus was slightly higher than the normal value of 3 points. This effect was significantly more pronounced on the involved side (4.0 points) than on the uninvolved side (3.5 points) (p = 0.0017).
The electromyographic recordings of the thigh muscles on the uninvolved side could be described as normal, with stance-phase activity of the rectus femoris, vastus medialis, and vastus lateralis and swing-phase activity of the adductor magnus. The stance-phase activity of the adductor magnus was more pronounced on the involved side (2.1 points) than on the uninvolved side (0.7 point) (p = 0.0001). The hamstrings were active in the transition from stance to swing. No significant differences in activity between the vastus medialis and the vastus lateralis could be detected with the numbers available for study. On the involved side, the rotated leg muscles appeared to be coordinated with the thigh muscles of the uninvolved side, with the leg muscles functioning as knee extensors and the tibialis anterior, as a knee flexor. The two heads of the gastrocnemius and the soleus generally showed a synchronized activation pattern during the stance phase. The medial head of the gastrocnemius was mainly active during the stance phase, with an activity level of 2.6 points compared with a swing-phase activity level of 0.4 point. The stance-phase activity level of the lateral head of the gastrocnemius was 3.1 points, and the swing-phase activity level was 0.6 point. The soleus revealed a stance-phase activity level of 2.8 points and a swing-phase activity level of 0.6 point. Five patients had a reduced level of swing-phase activity on the involved side. As the antagonist and functional knee flexor, the tibialis anterior had a stance-phase activity level of 1.4 points and a swing-phase activity level of 3.0 points. In six patients we found a considerable level of activity (3 points) and in two patients, a strong level of activity (4 points) during the stance phase. In two patients we found no activity at all, in four we found a markedly reduced level of activity (1 point), and in two we found a slightly reduced level of activity (2 points) in the swing phase. This antagonistic innervation pattern was seen in the same patients. As a stabilizer of the former ankle joint, the peroneus longus had nearly the same activity level throughout the entire gait cycle (swing phase, 2.4 points; stance phase, 2.6 points).
Clinical Outcome
Correlation Analyses The correlation coefficients between the Musculoskeletal Tumor Society score and the patient data indicated a significant negative relationship; the older patients generally had a lower score, a shorter walking distance, and a worse gait pattern (Table VII). On the other hand, the duration of follow-up apparently had no influence on gait, walking ability, emotional acceptance, or the total score (Table VII) even though the range was widespread, with short follow-up periods of less than one year.
Although the range of motion of the knee was comparable on both sides (59.7 degrees compared with 58.1 degrees [Table V]), there was a significant correlation between the range of motion of the rotated former ankle joint and the overall Musculoskeletal Tumor Society score (r = 0.42) as well as the subjective quality of the gait pattern (r = 0.42) (Table VII).
Influence of Follow-up
The patients in the current study had good restoration of gait after rotationplasty, with a total Musculoskeletal Tumor Society score that was comparable with that of patients managed with prosthetic knee replacement. (In our present study, the mean score was 24 points; in our previous study6, patients managed with rotationplasty had a mean score of 24 points and those managed with endoprosthetic replacement had a mean score of 25 points; and in a study by Kawai et al.7, patients managed with rotationplasty had an average score of 25 points.) Good restoration was also evidenced by the symmetrical ranges of motion of the hip and knee joints and by the similar electromyographic activity patterns of the hip muscles. Furthermore, even the former leg muscles of the involved limb functioned according to their new task of controlling the former ankle joint in the rotated position. In particular, four gait peculiarities were noted. First, we found a slight limp on the involved side, which helped to unload the hip joint and to compensate for the reduced hip abductor strength resulting from the procedure itself. Second, the ranges of motion of the hip and knee joints on the uninvolved side were equal to the ranges of the hip joint and the former ankle (new knee) joint on the involved side and were of normal excursion. However, in most of the patients the motion pattern of the knee joint revealed reduced flexion at mid-stance on the involved side, indicating a limited capability to accept weight. Third, the push-off from the involved limb, as revealed by ankle dorsiflexion torque, was diminished to 50 percent or less in 30 percent (thirteen) of the thirty-eight patients in whom it was measured. Finally, the electromyographic activity of the leg muscles in the involved limb showed very good adaptation to the new function. The tibialis anterior and the peroneus longus were not as well coordinated. The activity pattern and level of the hip muscles appeared normal and were symmetrical. Lateral lean of the trunk toward the involved side indicates a weakness of the hip abductor muscles and might be an unwanted effect because excessive movement of the trunk might cause problems in the spine. Therefore, physiotherapy should include strengthening exercises for the hip abductor and trunk stabilizing muscles. Finally, it should be noted that the extent of the lateral lean of the trunk was not associated with the total Musculoskeletal Tumor Society score, gait, or walking ability. The pelvic tilting that was more pronounced during the stance phase on the uninvolved side may be due to the loss of active dorsiflexion of the prosthetic foot. Therefore, the patient tilts the pelvis toward the stance limb and lifts the hip of the involved limb, thus achieving sufficient toe clearance for the swing phase. The range of motion of the hip joint in the sagittal plane was similar to that in normal subjects13. Furthermore, the range of motion of both the involved knee (former ankle) and the uninvolved knee was similar and symmetrical. Only the range of flexion of the involved knee (former ankle) at mid-stance was slightly reduced in 51 percent (twenty-two) and severely reduced in 35 percent (fifteen) of the forty-three patients. This might indicate a disturbed extensor mechanism of the knee joint that has to be compensated for by the former leg muscles. However, Steenhoff et al.18 found, in a study of patients who had had rotationplasty, a comparable extension torque in the contralateral quadriceps muscle because of the longer moment arm due to the length of the calcaneus even though the strength of the plantar flexors was only 71 percent of that of the normal, contralateral ankle (100 percent). Nevertheless, it seems important to exercise the remaining leg muscles in order to retain or develop their capacity to control and support the function of the new knee (former ankle). Strengthening exercises should be combined with stretching to ensure a sufficient range of motion with maximum plantar flexion (knee extension). The overall electromyographic activity pattern demonstrated very good coordination with the stance-phase activity of the triceps surae muscles and the swing-phase activity of the tibialis anterior, concurring with the new task of the muscles. The influence of individual muscles cannot be evaluated, as the muscle strength cannot be directly determined from the electromyographic amplitude. It appears important, however, that enough muscles be active in order to regain control over the former ankle joint. This observation is also in accordance with the findings of Kawai et al.8, who reported an association between the number of excised quadriceps heads and the functional restoration. Catani et al.3 reported that, even though walking by patients managed with rotationplasty was slightly arrhythmic and slower than walking by a control group of normal patients, the rotationplasty group was well coordinated. Furthermore, those authors found that stride length, velocity, and cadence were markedly reduced in the rotationplasty group. As these parameters were not investigated in the present study, we cannot compare our results with those findings. However, in order to determine the Musculoskeletal Tumor Society score, the gait patterns were visually inspected by experienced clinicians and were judged as inconspicuous for twenty-four patients and as fair for fourteen. McClenaghan et al.11 evaluated eighteen patients managed with different operations (above-the-knee amputation, arthrodesis, and rotationplasty) for the treatment of malignant bone tumor of the knee region. On the basis of measurements of oxygen consumption, they found that the patients who had been managed with rotationplasty walked faster and more efficiently than those managed with above-the-knee amputation or arthrodesis. Correlation analyses revealed that the duration of follow-up of our patients did not influence the functional outcome, even though there was a wide range (nine months to eighteen years). However, the age at the time of the operation showed a distinct influence on gait, walking ability, and the total Musculoskeletal Tumor Society score. The younger patients were better able to adapt to the altered anatomical and functional situation and to develop good function. In conclusion, the findings of electromyographic and gait analyses support the good functional results that have been reported after rotationplasty. Therefore, we believe that this biological reconstruction procedure is a good alternative for the treatment of bone tumors.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |