 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1999 The Journal of Bone and Joint Surgery, Inc. Excision for the Treatment of Periarticular Ossification of the Knee in Patients Who Have a Traumatic Brain Injury*  , ROME, ITALYInvestigation performed at University of Rome "Tor Vergata" and Istituto di Ricerca e Cura a Carattere Scientifico Santa Lucia, Rome
Background: Patients who are comatose after a traumatic brain injury often have heterotopic periarticular ossification that can be treated with excision to improve the range of motion of the joint. Methods: Areas of periarticular ossification were resected at an average of twenty-three months after recovery from a coma in seven knees of five patients who had a traumatic brain injury. Before the procedure, all of the knees were fixed in a flexed position that ranged from 10 to 40 degrees and they had a painful arc of motion that ranged from 20 to 70 degrees of flexion. None of the patients could walk, and some of them could barely sit in a wheelchair. At the end of the operation, the arc of motion was markedly improved in all of the knees (0 to 130 degrees in three knees, 0 to 120 degrees in three, and 10 to 120 degrees in one). In an attempt to prevent postoperative loss of motion and recurrence of the ossification, continuous passive motion was applied to the involved knee for six weeks before a full rehabilitation program was started. The latest follow-up evaluation was at an average of thirty-four months (range, twenty-five to sixty months). Results: At the time of follow-up, all of the patients could walk and all of the knees were pain-free. One knee had an arc of flexion of 0 to 90 degrees; two, an arc of 10 to 100 degrees; one, an arc of 5 to 110 degrees; two, an arc of 0 to 120 degrees; and one, an arc of 0 to 130 degrees. Ossification did not recur in any of the knees. Conclusions: Patients with good neuromuscular control had the best general functional result. The routine use of a continuous-passive-motion machine was associated with no recurrence of ossification, and there was some late loss of motion after its use was discontinued.
Patients who are comatose after a brain injury are often affected by heterotopic periarticular ossification. Epidemiological studies3,6 have shown that either severe limitation of joint motion or ankylosis may develop in approximately 10 to 20 percent of patients who recover from a coma, with the elbow and the hip being the most frequent locations of heterotopic bone. For example, in a study of 496 adults who had had a severe head injury, Garland et al.3 reported heterotopic ossification that caused pain or limited motion, or both, in 100 joints in fifty-seven patients. Usually, patients who have heterotopic ossification and a good mental and neuromuscular recovery are referred to an orthopaedic surgeon by neurologists or physiatrists of the postcoma unit who deem excision of the heterotopic bone to be necessary to speed up the rehabilitation program. In patients who have severe cognitive and physical deficits, an operation may also be indicated to improve their position in bed and in a wheelchair or to allow personal hygiene4,6. Heterotopic ossification about the knee is very rare in patients who have a head injury. Garland et al.3 identified three knees (among the 100 knees in fifty-seven patients) that had periarticular heterotopic bone that limited motion. Moreover, in a subsequent study on the results of operative resection of heterotopic ossification, Garland et al.4 reported that there were six involved knees in patients who had heterotopic bone removed from the hip, elbow, or shoulder; however, the area of ossification was not removed operatively from any of these knees. To the best of our knowledge, there have been only two previous studies1,2 on the results of excision of heterotopic periarticular bone after a traumatic brain injury; these reports included a total of twelve knees in ten patients. In the present study, we describe the results of operative treatment in seven knees with a severe functional impairment caused by ossification.
From February 1992 to November 1996, 180 patients who had a traumatic brain injury were admitted to our postcoma unit to be evaluated for eligibility for a rehabilitation program. Five of these patients had periarticular ossification that limited motion of the knee, and neurologists and physiatrists sought operative treatment for them after physical therapy and the application of a brace had failed. The radiographic patterns and extents of the heterotopic ossification were very similar in these five patients. The heterotopic bone extended from the anteromedial surface of the distal femoral metaphysis and the medial femoral condyle toward both the femoral trochlea and the patella and inferiorly toward the tibia along the medial collateral ligament, the proximal portion of which was also ossified. We operated on seven knees in five patients. Three patients were male, and two were female. All of the patients had had a traumatic brain injury followed by a coma that lasted from thirty to 210 days (average, seventy-six days). The right knee was involved in two patients; the left knee, in one; and both knees, in two. All of the knees were fixed in a flexed position that ranged from 10 to 40 degrees, and they had a painful limitation of the arc of flexion, which ranged from 20 to 70 degrees. A spastic flexion contracture of the hamstrings was also present in two knees. The patients were operated on between eight and sixty-seven months (average, twenty-three months) after they recovered from the coma, as soon as they could offer sufficient mental and physical cooperation and their general condition had improved enough to have the operation. The ages of the patients at the time of the operation ranged from fifteen to forty-two years (average, twenty-six years) (Table I).
At the evaluation performed just before the operation, we categorized our patients according to the five classes of Garland et al.4. No patient was in class I (minimum cognitive and physical deficits), two patients were in class II (minimum cognitive deficits with moderate physical disability), one patient was in class III (minimum cognitive deficits with severe physical disability), two patients were in class IV (moderate-to-severe cognitive deficits with minimum-to-moderate physical disability), and no patient was in class V (moderate-to-severe cognitive deficits with severe physical disability). All five patients were quadriparetic. Periarticular heterotopic ossification severely affected both elbows in one patient, one hip and one elbow in another patient, and one elbow in a third patient. One patient had a severe spastic contracture of the ipsilateral triceps surae with an equinus deformity of the foot. All of the patients were unable to walk at the time of the operation, but they were judged by our neurologists to be good candidates for the operation because mental and neuromuscular recovery was proceeding satisfactorily. The heterotopic bone was resected for the first time in all of the knees. Neither measurement of the serum level of alkaline phosphatase nor a bone scan was included in the preoperative study of our patients. The operation, which lasted for seventy to 100 minutes, was performed under tourniquet-induced ischemia. No blood units were transfused. The area of ossification was removed through an anteromedial approach. Bone was located under the vastus medialis muscle, and it extended from the anteromedial surface of the distal femoral metaphysis and the medial femoral condyle toward both the femoral trochlea and the patella and inferiorly toward the tibia along the medial collateral ligament, the proximal part of which was also ossified (Figs. 1-A, 1-B, 1-C and 1-D). In two patients (two knees), an unyielding spastic contracture of the hamstrings was found after removal of the area of ossification. Both the semitendinosus and the biceps tendon were lengthened in an attempt to achieve full extension of the knee, but it was actually achieved in only one of the two patients. In one of these two patients, we also lengthened the Achilles tendon to correct a severe ipsilateral equinus deformity of the foot.
Immediately after the operation, the range of motion of the involved knee was recorded and a continuous-passive-motion machine was applied. The machine was set up to work along the maximum painless range of motion, which was gradually increased every day until the joint had regained the whole arc of motion that had been achieved at the time of the operation. The stitches were removed at the end of the second week, and the patient's family and his or her physical therapist were instructed to keep the machine on for one more month at night and during the day between the periods of active exercise. All of the patients were examined at regular intervals after the operation until the latest follow-up evaluation. Both clinical and radiographic findings concerning the involved joint were recorded as were the results of a complete neurological examination. The latest follow-up evaluation was performed at an average of thirty-four months (range, twenty-five to sixty months) after the operation.
The arc of flexion that was attained at the end of the operation was 0 to 130 degrees in three knees, 0 to 120 degrees in three, and 10 to 120 degrees in one. Two patients, one who had had lengthening of the hamstrings alone and one who also had had lengthening of the Achilles tendon, were managed with an above-the-knee plaster splint to immobilize the knee in maximum extension for ten days, and thus to obtain initial healing of the tendon, before the application of continuous passive motion. In one knee, a subcutaneous hematoma was drained four days after the operation and there was wound dehiscence. After the wound was resutured, the continuous passive motion was stopped for four days and then was started again. The stitches were removed one week later, without additional complications. Radiographs made at the end of the operation showed remnants of ossification that did not interfere with motion of the joint. Those remnants were essentially unchanged in size on the radiographs made at the time of follow-up (Figs. 2-A, 2-B, 2-C, 2-D, 2-E and 2-F). Ossification did not recur in any of the knees.
No noticeable changes in the range of motion were observed during the first months after the operation, but at the latest follow-up evaluation one knee had an arc of flexion of 0 to 130 degrees; two, an arc of 0 to 120 degrees; one, an arc of 5 to 110 degrees; two, an arc of 10 to 100 degrees; and one, an arc of 0 to 90 degrees (Table I and Figs. 3-A, 3-B, and 3-C). Stability of the knee was also tested. The abduction stress test at 30 degrees of knee flexion was slightly positive for two knees and was negative for the other five, whereas both the Lachman and the drawer tests were negative for all seven knees.
Functional recovery was slower and more painful for the two patients who wore a plaster splint for ten days before starting continuous passive motion than it was for those who started continuous passive motion immediately after the operation. However, additional soft-tissue releases for contractures had been needed in these two patients. The patient who had wound dehiscence did not have any problems when he resumed passive motion four days later. At the time of follow-up, all of the patients could walk and only one of them needed to use a cane. They could eat, maintain personal hygiene, stand up, and sit down by themselves, but they had to rely more or less on their family for other activities of daily living because of their low cognitive level. No patient was able to run, but one could walk rapidly. None of them were able to return to their previous job or to the school that they had attended before the accident. Nonetheless, after the operation, the cognitive and neurological status of two patients (Cases 1 and 5) improved, from class IV to class III and from class II to class I, according to the system of Garland et al.4.
A few studies4,7,8 have addressed the results of excision of heterotopic bone about the hip and elbow in postcomatose patients after a brain injury, but as far as we know only two previous reports1,2 have described the results of such an excision in the knee. Gacon et al.2 operated on six knees. The ossification recurred in one patient who had bilateral involvement, but the authors did not mention the level of neuromuscular involvement of their patients or the range of motion of the involved knees before and after the operation, although they concluded that the final result was positive. Their patients did not receive any adjuvant postoperative treatment. Charnley et al.1 reported on six knees in five postcomatose patients who were given indomethacin after excision of heterotopic bone about the knee. The authors reported results that were very similar to those of the present study. However, the duration of postoperative follow-up of some of their patients was very short and, for that reason, some good results could have deteriorated. We operated on seven knees in five patients who were unable to walk before the operation because of a fixed flexion position of one or both knees, which was initially caused by spastic contracture of the hamstrings after a brain injury and, later, by ossification, which gradually became a mechanical obstacle to motion. All of our patients had pain with motion of the knee. Immediately after removal of the heterotopic bone, we obtained a nearly full range of passive motion of all but two of the knees, in which we had to lengthen the hamstring tendons. At the end of this procedure in one of these knees (which was treated sixty-seven months after the patient recovered from the coma), a residual flexion contracture of approximately 10 degrees was still present, probably because of retraction of the posterior part of the joint capsule. However, the other knee that had lengthening of the hamstrings regained a full range of extension. After the operation, use of a continuous-passive-motion machine, a device that is employed postoperatively for different pathological conditions of the knee9,10, seemed to be successful in preventing stiffness of the joint and recurrence of the ossification. Moreover, continuous passive motion applied at a low speed was painless for the patients, who were usually restless and refused to practice early active exercises with the aid of a physical therapist. However, during the following months, when a full rehabilitation program was started and continuous passive motion was stopped, the patients who had incomplete neuromuscular control of the involved knee began to lose some motion. Ossification did not recur in any of the knees, although radiographs showed that small remnants of the ossification (which did not enlarge during the follow-up period) were present in all of the involved joints. According to Charnley et al.1, indomethacin exerts a good protective effect against the recurrence of ossification, and their results were similar to those obtained after the application of a continuous-passive-motion machine in our study. We did not have a comparable control series of patients in whom ossification about the knee was resected without any postoperative adjuvant treatment. However, in the series of Gacon et al.2, in which none of the patients received any postoperative prophylaxis, one patient had a recurrence of the ossification bilaterally and although the others had a so-called positive result the authors did not report the preoperative, postoperative, or most recent range of motion of the knee. Recurrence of ossification and loss of motion have also been reported by other authors2,4 who did not use any postoperative prophylaxis after resection of heterotopic bone from other joints of postcomatose patients. While we recognize that the number of patients who have been managed with continuous passive motion or indomethacin, or both, is small, we believe that these modalities are useful in preventing the recurrence of ossification in the early postoperative phase, although the long-term functional result of the operation is strictly related to the degree of the patient's neurological recovery. At the most recent follow-up evaluation, all of the knees in the present study were painless and even those that had lost some motion did not show radiographic signs of recurrence. Moreover, in spite of the resection of the ossified medial collateral ligament, only two knees showed a slightly positive result on the abduction stress test and neither of the two patients noted instability of the joint during walking. The scar in the medial portion of the joint capsule, which had been extensively incised to remove the area of ossification, might have acquired sufficient strength to fully replace the medial collateral ligament in our patients. Also, the functional demand on the knees was limited because the neuromuscular system in these patients was damaged. All of our patients could walk well at the time of the most recent follow-up; none of them wore a brace, and only one needed to use a cane and limped because of a residual 10-degree knee flexion contracture and quadriceps weakness. Garland et al.4 showed that the chances of recurrence of ossification after resection of heterotopic bone were not associated with the serum level of alkaline phosphatase or the results of a bone scan. In fact, many of their patients who had a low serum level of alkaline phosphatase and a negative bone scan had a recurrence after resection. On the basis of these findings, we did not include those tests in the preoperative clinical study of our patients. Garland et al. also found that mature heterotopic periarticular bone in postcomatose patients is less likely to recur after resection than is less mature heterotopic bone. They considered the bone to be mature when at least eighteen months had elapsed from the time of the head injury. However, McAuliffe and Wolfson5 recently reported very good results in patients with a central neurological lesion who had had resection of an area of ossification of the elbow soon after the injury. All of the patients in their series were managed prophylactically after the operation with radiation therapy. We found similar results in our patients. In fact, we did not find any difference, in terms of the quality of the results and the rate of recurrence, between the knees that had been operated on before eighteen months had elapsed from the time of the head injury and those that had been operated on after eighteen months had elapsed. The only differences between our series and that of McAuliffe and Wolfson were the site of heterotopic ossification and the prophylactic postoperative treatment.
*No benefits in any form have been received or will be received from a commercial partly related directly or indirectly to the subject of this article. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was a grant from the Istituto di Ricerca e Cura a Carattere Scientifico Santa Lucia.
  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. |