Arthroscopy: The Journal of Arthroscopic and Related Surgery 6(l): 11-17 Published by Raven Press, Ltd. 8 1990 Arthroscopy Association of North America
Osteochondritis
Dissecans of the Trochlea of the Femur James B. Smith, M.D.
Summary: This article defines the characteristic features of osteochondritis dissecans of the trochlea of the femur, and indicates that important differences distinguish it from the more familiar femoral condylar form. The clinical features in 16 knees included: gradual onset of symptoms, pain with running and
jumping, no significant history of injury; inconstant tenderness of the trochlea, and pain with resisted extension at 20 to 45”. Diagnosis was usually difficult, and was often delayed because of subtle radiographic changes. Treatment depended on the stage of presentation. Nonsurgical treatment failed in four of seven knees. Drilling the lesions failed in two of the three cases. Fixation with small screws produced two good results; two others healed, but with short follow-up. Removal of the loose bodies from six knees produced one poor result and five good results. This process differs in presentation from femoral condylar osteochondritis dissecans. Although the results (average follow-up more than 5 years) were generally good, the mild symptoms probably represent incongruity of the patellofemoral joint, and probably foretell osteoarthritis. Key Words: Osteochondritis dissecans-Trochlea-Femur.
The lateral side of the trochlea of the femur is an unusual site for osteochondritis dissecans. A review of the literature revealed 3,329 cases that fit the diagnosis of osteochondritis dissecans; of these, 19 mentioned the trochlea of the femur (l-20). Except for reports by Linden (9) and Lindholm (12), these were single case reports, or brief mentions of the trochlea as an uncommon site. Some authors may have considered this site as part of the lateral femoral condyle; they would then group condylar and trochlear lesions together. This might explain the varied incidence in different series. The site of trochlear lesions is possibly more critical to knee function than the more common medial condylar site; if so, preservation of the detached bone and cartilage is more important in the trochlear osteochondritis dissecans than in the condylar form. The trochlear form has a specific clinical and radiographic presentation that can make the diagnosis difficult, as is indicated by the fact that, in several
of the cases, experienced radiologists and orthopaedists missed the obscure radiographic lesions. The condition is not new. Wells (20) described and illustrated a case in an ancient museum specimen. Mollan (13) described a case located in the “super0-medial aspect of the lateral femoral condyle;” there was a good result after fixation of the fragment with Smillie pins. He said that there was no prior description of the condition. A more recent report of two cases by Kurzweil et al. (7) indicated no knowledge of previous reports. Axhausen (3) published a detailed case description in 1912. His patient was an l&year-old woman. The lesion was in the medial side of the trochlea; his sketch of the lesion (Fig. 1) shows it in the center. Otherwise, this case is identical to those in the present series. Bianchi et al. (4) removed both the bone fragment and the bed in which it lay. Greville’s case (6) apparently followed direct trauma to the patella. The lesion was healing 6 months after treatment with bone pegs. Aichroth (1) attributed his two cases to trauma resulting from patellar dislocation. In a statistical
Address correspondence and reprint requests to Dr. James B. Smith at 2460 S.W. Holden Street, Suite 12, Seattle, WA 98106, U.S.A.
I1
12
J. B. SMITH age duration of symptoms at the time of diagnosis was 15 months, with a range from 3 to 36 months. SYMPTOMS There was pain in 15 knees; it was located behind the patella, usually to the lateral side. Running, and especially running and jumping, were the most common aggravating factors. There was no pain at rest, except for pain after sitting for a time with the knee flexed; moving the knee relieved it. Five patients described effusion. Although four knees presented with a detached loose body and had symptoms of impingement, three of these four had had no effusion. The effusion tended to occur with activity, and to subside with rest. Four patients reported vague sensations of giving way, or incipient “buckling;” most thought that this was due to sudden pain that developed while bearing weight on the slightly flexed knee. PHYSICAL FINDINGS
1.1
rr:3w7
~~~f*~i~~~n
dcs KnnrpP1~ isr 1:s
FIG. 1. Axhausen’s sketch of his case of osteochondritis cans of the trochlea. Reprinted with permission.
disse-
review of osteochondritis dissecans, Linden (9) found six of 156 cases in an anterior location; of the 130 cases described by Lindholm (12), five were in the trochlea, but they did not differentiate the clinical features from those of the more common locations in the femur.
MATERIALS There were 13 cases, all male; three were bilateral, for a total of 16 knees. The average age at onset of symptoms was 13 plus years. All patients were active in athletics. Three patients sought medical attention after trivial injuries; all three patients had had some chronic symptoms before the described injury. Except for the two patients currently receiving treatment, and one who presented with a 16-year history of a loose body (case 12), the aver-
Arthroscopy,
Vol. 6, No. I, 1990
There was tenderness in eight knees. It was located at the lateral margin of the trochlea. Only one patient had limited motion. He had symptoms for 36 months, and already had some radiographic degenerative changes; these changes and a moderate effusion restricted the last 20” of motion. Seven patients had thigh atrophy; the incidence and amount of atrophy did not seem related to the severity or duration of symptoms RADIOGRAPHIC
FINDINGS
The radiographic findings were inconstant, and often subtle. The lesions often were not visible on standard anteroposterior views, but usually were visible on lateral or tangential views. They usually consisted of areas of indistinct relative radiolucence of the proximal articular surface of the trochlea; on the tangential view, they involved the lateral facet (Fig. 2). In some cases, the lesions became visible on standard radiographs only after symptoms had been present for over a year. In several cases, the lesions were visible only with tomography, and in others, tomography helped to define lesions that were indistinct on standard radiographs (Fig. 3). Computerized tomography was helpful in one case (case 7). The patient had had open internal fixation, but symptoms continued. Standard films suggested that the lesion had healed, but computer-
OSTEOCHONDRITIS
DISSECANS
OF THE TROCHLEA
OF THE FEMUR
13
2A,B
FIG. 2. Case 4. A: Lateral radiograph after symptoms had been present for 1 year; a faint radiolucence is present in the trochlea. B: Tangential radiograph. The separated bone fragment is faintly visible in the lateral side of the trochlea. C: Photograph of videotape at arthroscopy, with the arthroscope in a proximal portal. A needle has impaled the fragment, and is lifting it up from its bed in the trochlea, demonstrating instability that was not apparent radiographically.
ized tomography demonstrated nonunion of the fragment and the degree of involvement of the trochlear surfaces (Fig. 4). In a recent case, a magnetic resonance study showed a diffuse area of increased signal on the tangential projection, but did not define the lesion precisely; however, the lesion was clearly visible on an oblique projection (Fig. 5). As with the condylar form, the bone of the fragment is often so thin that it does not appear radiographically, even on tomography, so that the radiographic appearance is deceptively innocuous (Fig. 6). TREATMENT AND RESULTS The small number of patients, the varying stages at presentation, and the variety of treatment methods that were necessary combine to prevent me
from drawing any conclusions about the effect of treatment on the results; however, because the condition is unusual, listing the results seems appropriate. The only rating system found in the literature was that of Hughston et al. (21), but it did not seem appropriate for these cases. We classified the results as “excellent” when there were no symptoms, and examination and radiographs were normal; “poor” when there were disabling symptoms, and obvious deterioration on examination and radiographs; and “good” when there were mild symptoms, minor abnormalities on examination, and slight irregularities of the patellofemoral joint on radiographs , If the patient has pain and a radiographic lesion without evidence of displacement, and without clinical evidence of motion of the fragment, such as effusion, limited motion, or giving way, a cylinder
Arthroscopy, Vol. 6, No. 1. 1990
J. B. SMITH
14 3A,B
RIGHT
LEFT
FIG. 3. Case 5, right knee. A: A faint irregularity of bone density is visible in the trochlea on a lateral radiograph. B: On a tangential radiograph, both trochleae have indistinct radiolucent areas. C: Lateral tomograph defines the lesion precisely.
cast immobilizes the knee in extension for 6-8 weeks, and then the patient gradually returns to normal activities. As with condylar lesions, the presence or absence of pain determines the rate of return to activity; radiographic evidence of healing lags far behind clinical evidence of healing. Seven patients had nonsurgical treatment initially; three healed (cases 3L, SL, and 5R), one with excellent, and two with good results. Four other patients (cases 1, 2, 3R, and 4) failed to heal, and went on to surgical treatment. Treatment consists of arthroscopy if restriction of activity or immobilization fails, or if the patient presents with effusion or limited motion. If the lesion is stable in its bed, drilling through it to the bone beneath the lesion is necessary, followed by immobilization for 6-g weeks. Of the three patients (cases 3R, 4, and 7) treated with this procedure; one had a good result, one healed with a poor result, and the other failed to unite, a poor result. A screw placed through the joint surface fixes the Arthroscopy.
Vol. 6, No. 1, 1990
fragment to the femur, if there is motion of the fragment in its bed. Arthrotomy is necessary if reduction and fixation are not certain arthroscopically. Immobilization continues for six weeks after the drilling or internal fixation, followed by arthroscopic screw removal. One patient (case 2) had this treatment arthroscopically, and had a good result. Because of difficulty moving the patella out of the way for internal fixation devices, arthrotomy was necessary for adequate fixation in one case (case 6) with a good result. If the fragment is displaced from its bed, and is structurally intact, replacement and internal fixation is advisable to preserve the opposing surface for the patella. Despite this opinion, of the four patients presenting with the fragment detached from its bed in the trochlea, all but one had the fragment removed. The first patient in the series (case 1) had extensor mechanism malalignment; unaware of the nature of the process, we removed the fragment. His result
OSTEOCHONDRITIS
DISSECANS
OF THE TROCHLEA
OF THE FEMUR
I5
DISCUSSION
FIG. 4. Case 7. Healing of the lesion was uncertain on radiographs taken 1 year after drilling; computerized tomographs confirmed the suspected nonunion of the fragment, and demonstrated the extent of the articular surface involved.
was poor, with continued pain and swelling, and osteoarthritis. Two patients declined replacement and fixation of the fragment (cases 8 and 10) to avoid the required period of immobilization. Their results were good. In the fourth patient (case 9), who had bilateral involvement, there were multiple fragments in both knees, and the surgeon did not think it possible technically to replace and fix the fragments. This patient’s surgery was recent, but he is doing well so far. Cases 10, 11, and 12 had their operations recently, and therefore have no result assigned. The other 10 patients (12 knees) have an average followup of 5.7 years and all have reached skeletal maturity. There was one excellent result, eight good, and three poor results.
FIG. 5. Case 13. An oblique magnetic resonance image projection defines the round trochlear lesion.
Osteochondritis dissecans of the trochlea seems to be a distinct clinical form. It involves adolescent boys, and symptoms include a gradual onset independent of trauma, pain during activities involving resisted extension, and tenderness at the margin of the trochlea. A characteristic radiolucence of the trochlea is easy to miss. Its clinical course and response to treatment parallel the condylar form. It seems likely that this condition occurs more frequently than reports in the literature indicate. Judging by the subtlety of the clinical and radiographic changes, and the fact that several cases escaped detection by experienced orthopaedists and radiologists, many cases probably have a mistaken diagnosis. Many of these cases probably heal spontaneously; others, such as case 12, progress to shedding of the osteocartilaginous fragment. When the loose body requires treatment, the site of origin can remain obscure unless there is a conscious effort to find it. Aichroth (1) described two trochlear lesions, and believed that they were related to trauma associated with patellar dislocation. Only one of my patients had evidence of patellofemoral malalignment, and none had a history suggesting subluxation or dislocation. Four of my patients had a history of trivial trauma before the onset of symptoms, and the others had none; it seems that injury in this group of patients did not have a significant role. The general principles of treatment of osteochondritis dissecans in the femoral condylar lesions, as given in the excellent study of Hughston et al. also appear to apply to the trochlear form, with one possible exception. Hughston et al. (21) recommended only exercise and slight modification of activity for patients who were not skeletally mature and whose lesions were stable; 82% of patients achieved satisfactory results. In contrast, all but three of my patients were still growing at the time of presentation, but only two had stable lesions that healed without surgery; all the others had unstable fragments or fragments that had detached before the patients sought treatment. Although the results seem generally satisfactory, the possibility for degeneration seems likely. There was only one excellent result. All the patients are now skeletally mature; those with a good result had one or more significant symptoms or physical or radiographic findings that indicate incongruent patellofemoral joint surfaces. Long-term follow-up Arthroscopy,
Vol. 6, No. 1, 1990
16
J. B. SMITH
FIG. 6. Case 11. A: A definite defect is present on the lateral radiograph, but no separate bone fragment is visible. B: No lesion is visible on the tangential view. C: Surgical photograph shows the lesion in the lateral side of the trochlea; cancellous bone forms the base of the lesion. The proximal margin of the trochlea is on the left; the infrapatellar fat pad is on the right; a retractor displaces the patella to the lateral side.
will probably show deterioration and eventually osteoarthritis, as Linden (11) showed in cases of femoral condylar osteochondritis dissecans. This presentation may increase awareness and stimulate more frequent diagnosis of this condition, so that results of treatment will improve.
squatting or jumping activities, tenderness appears over the lateral surface of the trochlea, and indistinct radiolucence and fragmentation are visible at the site. The methods used to treat osteochondritis dissecans usually produce healing, but the results of treatment have been imperfect.
CONCLUSION
Acknowledgment: I thank Drs. John Hendrickson, Daniel Flugstad, Richard Kirby, and James Russo, for permission to include cases 3,9, 11, and 13, respectively.
When osteochondritis dissecans affects the trochlea of the femur, a distinct clinical pattern develops: it occurs in adolescent boys, pain occurs with
Arthroscopy,
Vol. 6, No. I, 1990
I also thank Beverly D. Downing, medical librarian at West Seattle Community Hospital, for her skill and persistence in finding the references.
OSTEOCHONDRITIS
DISSECANS
REFERENCES 1. Aichroth P. Osteochondritis dissecans of the knee. A clinical survey. J Bone Joint Stag 1971;53-B:44&7. 2. Arcq M. Behandlung der Osteochondritis Dissecans Durch Knochenspanbolzung. Arch Orthop Uufall-Chir 1974;79: 297-312. 3. Axhausen G. Die Entstehung der Freien Gelenkkorper und Ihre Beziehunaen. Arch F Klin Chir 1912;104:581-678. 4. Bianchi G, Giidanich IF, Zanasi R. Contributo clinic0 e anatomopatologico alla conscenza dell osteochondrosi dissecante. Chir Organi Mov 1955;41:46&87. 5. Fisher AGT. A study of loose bodies composed of cartilage or of cartilage and bone occurring in joints with special reference to their pathology and etiology. Br J Stag 1920;8:493523. 6. Greville MR. Osteochondritis dissecans: treatment by bone grafting. South Med J 1964;57:886-93. 7. Kurzweil PR, Zambetti GJ, Hamilton WG. Osteochondritis dissecans in the lateral patellofemoral groove. Am J Sports Med 1988;16:308-10. 8. Kunzli HF. Beitrag zur Osteochondritis Dissecans. Schweiz Med Wochenschr
1962;92:292-4.
9. Linden B. The incidence of osteochondritis dissecans in the condyles of the femur. Acta Orthop Stand 1976;47:664-7. 10. Linden B. Osteochondritis dissecans of the femoral condyles. J Bone Joint Stag 1977;59-A:769-76. 11. Linden CB. Osteochondritis dissecans. The natural course
OF THE TROCHLEA
OF THE FEMUR
17
and incidence in femur condyles. [Ph.D. Dissertation] Lund, Sweden: 1976. Malmii General Hospital (University of Lund School of Medicine). 12. Lindholm TS. Osteochondritis dissecans of the knee. A clinical study. Ann Chir Gynaecol Fenn 1974;63:69-76. 13. Mollan RAB. Osteochondritis dissecans of the knee. A case report of an unusual lesion on the lateral femoral condyle. Acta Orthop Stand 1977;48:517-19. 14. Moulonguet P. Foreign bodies in joints. J Bone Joint Suru
1929;l lL353-64. 15. Poulsen K. 3 Tiefalde of Mus Articuli Genus. Hosp Tidende, BD VIII No. 24. 1890. (Cited in Centralblatt fur Chirurnie No. 43, V. 17, 1890;831.j 16. Rehbein F. Die Entstehung der Osteochondritis Dissecans. Arch KZin Chir 1950;265:69-114. 17. Roberts HM. Osteochondritis dissecans. In: Kennedy JC, ed. The injured adolescent knee. Baltimore: Williams &
Wilkins, 1979:1214. 18. Scheller S. Roentgenographic studies of epiphyseal growth and ossification of the knee. Acta Radio1 Sum11 _L 1960:195: l303. 19. Smillie IS. Diseases of the knee joint. Edinburgh: Churchill Livingstone, 1974:360-89. 20. Wells C. Osteochondritis dissecans in ancient British skeletal material. Med Hist 1974;18:365-9. 21. Hughston JC, Hergenroeder PT, Courtenay BG. Osteochondritis dissecans of the femoral condyles. J Bone Joint Surg 1984;66-A:l340-8.
Arthroscopy, Vol. 6, No. 1, 1990
Osteochondritis
Dissecans of the Trochlea of the Femur James B. Smith, M.D.
Summary: This article defines the characteristic features of osteochondritis dissecans of the trochlea of the femur, and indicates that important differences distinguish it from the more familiar femoral condylar form. The clinical features in 16 knees included: gradual onset of symptoms, pain with running and
jumping, no significant history of injury; inconstant tenderness of the trochlea, and pain with resisted extension at 20 to 45”. Diagnosis was usually difficult, and was often delayed because of subtle radiographic changes. Treatment depended on the stage of presentation. Nonsurgical treatment failed in four of seven knees. Drilling the lesions failed in two of the three cases. Fixation with small screws produced two good results; two others healed, but with short follow-up. Removal of the loose bodies from six knees produced one poor result and five good results. This process differs in presentation from femoral condylar osteochondritis dissecans. Although the results (average follow-up more than 5 years) were generally good, the mild symptoms probably represent incongruity of the patellofemoral joint, and probably foretell osteoarthritis. Key Words: Osteochondritis dissecans-Trochlea-Femur.
The lateral side of the trochlea of the femur is an unusual site for osteochondritis dissecans. A review of the literature revealed 3,329 cases that fit the diagnosis of osteochondritis dissecans; of these, 19 mentioned the trochlea of the femur (l-20). Except for reports by Linden (9) and Lindholm (12), these were single case reports, or brief mentions of the trochlea as an uncommon site. Some authors may have considered this site as part of the lateral femoral condyle; they would then group condylar and trochlear lesions together. This might explain the varied incidence in different series. The site of trochlear lesions is possibly more critical to knee function than the more common medial condylar site; if so, preservation of the detached bone and cartilage is more important in the trochlear osteochondritis dissecans than in the condylar form. The trochlear form has a specific clinical and radiographic presentation that can make the diagnosis difficult, as is indicated by the fact that, in several
of the cases, experienced radiologists and orthopaedists missed the obscure radiographic lesions. The condition is not new. Wells (20) described and illustrated a case in an ancient museum specimen. Mollan (13) described a case located in the “super0-medial aspect of the lateral femoral condyle;” there was a good result after fixation of the fragment with Smillie pins. He said that there was no prior description of the condition. A more recent report of two cases by Kurzweil et al. (7) indicated no knowledge of previous reports. Axhausen (3) published a detailed case description in 1912. His patient was an l&year-old woman. The lesion was in the medial side of the trochlea; his sketch of the lesion (Fig. 1) shows it in the center. Otherwise, this case is identical to those in the present series. Bianchi et al. (4) removed both the bone fragment and the bed in which it lay. Greville’s case (6) apparently followed direct trauma to the patella. The lesion was healing 6 months after treatment with bone pegs. Aichroth (1) attributed his two cases to trauma resulting from patellar dislocation. In a statistical
Address correspondence and reprint requests to Dr. James B. Smith at 2460 S.W. Holden Street, Suite 12, Seattle, WA 98106, U.S.A.
I1
12
J. B. SMITH age duration of symptoms at the time of diagnosis was 15 months, with a range from 3 to 36 months. SYMPTOMS There was pain in 15 knees; it was located behind the patella, usually to the lateral side. Running, and especially running and jumping, were the most common aggravating factors. There was no pain at rest, except for pain after sitting for a time with the knee flexed; moving the knee relieved it. Five patients described effusion. Although four knees presented with a detached loose body and had symptoms of impingement, three of these four had had no effusion. The effusion tended to occur with activity, and to subside with rest. Four patients reported vague sensations of giving way, or incipient “buckling;” most thought that this was due to sudden pain that developed while bearing weight on the slightly flexed knee. PHYSICAL FINDINGS
1.1
rr:3w7
~~~f*~i~~~n
dcs KnnrpP1~ isr 1:s
FIG. 1. Axhausen’s sketch of his case of osteochondritis cans of the trochlea. Reprinted with permission.
disse-
review of osteochondritis dissecans, Linden (9) found six of 156 cases in an anterior location; of the 130 cases described by Lindholm (12), five were in the trochlea, but they did not differentiate the clinical features from those of the more common locations in the femur.
MATERIALS There were 13 cases, all male; three were bilateral, for a total of 16 knees. The average age at onset of symptoms was 13 plus years. All patients were active in athletics. Three patients sought medical attention after trivial injuries; all three patients had had some chronic symptoms before the described injury. Except for the two patients currently receiving treatment, and one who presented with a 16-year history of a loose body (case 12), the aver-
Arthroscopy,
Vol. 6, No. I, 1990
There was tenderness in eight knees. It was located at the lateral margin of the trochlea. Only one patient had limited motion. He had symptoms for 36 months, and already had some radiographic degenerative changes; these changes and a moderate effusion restricted the last 20” of motion. Seven patients had thigh atrophy; the incidence and amount of atrophy did not seem related to the severity or duration of symptoms RADIOGRAPHIC
FINDINGS
The radiographic findings were inconstant, and often subtle. The lesions often were not visible on standard anteroposterior views, but usually were visible on lateral or tangential views. They usually consisted of areas of indistinct relative radiolucence of the proximal articular surface of the trochlea; on the tangential view, they involved the lateral facet (Fig. 2). In some cases, the lesions became visible on standard radiographs only after symptoms had been present for over a year. In several cases, the lesions were visible only with tomography, and in others, tomography helped to define lesions that were indistinct on standard radiographs (Fig. 3). Computerized tomography was helpful in one case (case 7). The patient had had open internal fixation, but symptoms continued. Standard films suggested that the lesion had healed, but computer-
OSTEOCHONDRITIS
DISSECANS
OF THE TROCHLEA
OF THE FEMUR
13
2A,B
FIG. 2. Case 4. A: Lateral radiograph after symptoms had been present for 1 year; a faint radiolucence is present in the trochlea. B: Tangential radiograph. The separated bone fragment is faintly visible in the lateral side of the trochlea. C: Photograph of videotape at arthroscopy, with the arthroscope in a proximal portal. A needle has impaled the fragment, and is lifting it up from its bed in the trochlea, demonstrating instability that was not apparent radiographically.
ized tomography demonstrated nonunion of the fragment and the degree of involvement of the trochlear surfaces (Fig. 4). In a recent case, a magnetic resonance study showed a diffuse area of increased signal on the tangential projection, but did not define the lesion precisely; however, the lesion was clearly visible on an oblique projection (Fig. 5). As with the condylar form, the bone of the fragment is often so thin that it does not appear radiographically, even on tomography, so that the radiographic appearance is deceptively innocuous (Fig. 6). TREATMENT AND RESULTS The small number of patients, the varying stages at presentation, and the variety of treatment methods that were necessary combine to prevent me
from drawing any conclusions about the effect of treatment on the results; however, because the condition is unusual, listing the results seems appropriate. The only rating system found in the literature was that of Hughston et al. (21), but it did not seem appropriate for these cases. We classified the results as “excellent” when there were no symptoms, and examination and radiographs were normal; “poor” when there were disabling symptoms, and obvious deterioration on examination and radiographs; and “good” when there were mild symptoms, minor abnormalities on examination, and slight irregularities of the patellofemoral joint on radiographs , If the patient has pain and a radiographic lesion without evidence of displacement, and without clinical evidence of motion of the fragment, such as effusion, limited motion, or giving way, a cylinder
Arthroscopy, Vol. 6, No. 1. 1990
J. B. SMITH
14 3A,B
RIGHT
LEFT
FIG. 3. Case 5, right knee. A: A faint irregularity of bone density is visible in the trochlea on a lateral radiograph. B: On a tangential radiograph, both trochleae have indistinct radiolucent areas. C: Lateral tomograph defines the lesion precisely.
cast immobilizes the knee in extension for 6-8 weeks, and then the patient gradually returns to normal activities. As with condylar lesions, the presence or absence of pain determines the rate of return to activity; radiographic evidence of healing lags far behind clinical evidence of healing. Seven patients had nonsurgical treatment initially; three healed (cases 3L, SL, and 5R), one with excellent, and two with good results. Four other patients (cases 1, 2, 3R, and 4) failed to heal, and went on to surgical treatment. Treatment consists of arthroscopy if restriction of activity or immobilization fails, or if the patient presents with effusion or limited motion. If the lesion is stable in its bed, drilling through it to the bone beneath the lesion is necessary, followed by immobilization for 6-g weeks. Of the three patients (cases 3R, 4, and 7) treated with this procedure; one had a good result, one healed with a poor result, and the other failed to unite, a poor result. A screw placed through the joint surface fixes the Arthroscopy.
Vol. 6, No. 1, 1990
fragment to the femur, if there is motion of the fragment in its bed. Arthrotomy is necessary if reduction and fixation are not certain arthroscopically. Immobilization continues for six weeks after the drilling or internal fixation, followed by arthroscopic screw removal. One patient (case 2) had this treatment arthroscopically, and had a good result. Because of difficulty moving the patella out of the way for internal fixation devices, arthrotomy was necessary for adequate fixation in one case (case 6) with a good result. If the fragment is displaced from its bed, and is structurally intact, replacement and internal fixation is advisable to preserve the opposing surface for the patella. Despite this opinion, of the four patients presenting with the fragment detached from its bed in the trochlea, all but one had the fragment removed. The first patient in the series (case 1) had extensor mechanism malalignment; unaware of the nature of the process, we removed the fragment. His result
OSTEOCHONDRITIS
DISSECANS
OF THE TROCHLEA
OF THE FEMUR
I5
DISCUSSION
FIG. 4. Case 7. Healing of the lesion was uncertain on radiographs taken 1 year after drilling; computerized tomographs confirmed the suspected nonunion of the fragment, and demonstrated the extent of the articular surface involved.
was poor, with continued pain and swelling, and osteoarthritis. Two patients declined replacement and fixation of the fragment (cases 8 and 10) to avoid the required period of immobilization. Their results were good. In the fourth patient (case 9), who had bilateral involvement, there were multiple fragments in both knees, and the surgeon did not think it possible technically to replace and fix the fragments. This patient’s surgery was recent, but he is doing well so far. Cases 10, 11, and 12 had their operations recently, and therefore have no result assigned. The other 10 patients (12 knees) have an average followup of 5.7 years and all have reached skeletal maturity. There was one excellent result, eight good, and three poor results.
FIG. 5. Case 13. An oblique magnetic resonance image projection defines the round trochlear lesion.
Osteochondritis dissecans of the trochlea seems to be a distinct clinical form. It involves adolescent boys, and symptoms include a gradual onset independent of trauma, pain during activities involving resisted extension, and tenderness at the margin of the trochlea. A characteristic radiolucence of the trochlea is easy to miss. Its clinical course and response to treatment parallel the condylar form. It seems likely that this condition occurs more frequently than reports in the literature indicate. Judging by the subtlety of the clinical and radiographic changes, and the fact that several cases escaped detection by experienced orthopaedists and radiologists, many cases probably have a mistaken diagnosis. Many of these cases probably heal spontaneously; others, such as case 12, progress to shedding of the osteocartilaginous fragment. When the loose body requires treatment, the site of origin can remain obscure unless there is a conscious effort to find it. Aichroth (1) described two trochlear lesions, and believed that they were related to trauma associated with patellar dislocation. Only one of my patients had evidence of patellofemoral malalignment, and none had a history suggesting subluxation or dislocation. Four of my patients had a history of trivial trauma before the onset of symptoms, and the others had none; it seems that injury in this group of patients did not have a significant role. The general principles of treatment of osteochondritis dissecans in the femoral condylar lesions, as given in the excellent study of Hughston et al. also appear to apply to the trochlear form, with one possible exception. Hughston et al. (21) recommended only exercise and slight modification of activity for patients who were not skeletally mature and whose lesions were stable; 82% of patients achieved satisfactory results. In contrast, all but three of my patients were still growing at the time of presentation, but only two had stable lesions that healed without surgery; all the others had unstable fragments or fragments that had detached before the patients sought treatment. Although the results seem generally satisfactory, the possibility for degeneration seems likely. There was only one excellent result. All the patients are now skeletally mature; those with a good result had one or more significant symptoms or physical or radiographic findings that indicate incongruent patellofemoral joint surfaces. Long-term follow-up Arthroscopy,
Vol. 6, No. 1, 1990
16
J. B. SMITH
FIG. 6. Case 11. A: A definite defect is present on the lateral radiograph, but no separate bone fragment is visible. B: No lesion is visible on the tangential view. C: Surgical photograph shows the lesion in the lateral side of the trochlea; cancellous bone forms the base of the lesion. The proximal margin of the trochlea is on the left; the infrapatellar fat pad is on the right; a retractor displaces the patella to the lateral side.
will probably show deterioration and eventually osteoarthritis, as Linden (11) showed in cases of femoral condylar osteochondritis dissecans. This presentation may increase awareness and stimulate more frequent diagnosis of this condition, so that results of treatment will improve.
squatting or jumping activities, tenderness appears over the lateral surface of the trochlea, and indistinct radiolucence and fragmentation are visible at the site. The methods used to treat osteochondritis dissecans usually produce healing, but the results of treatment have been imperfect.
CONCLUSION
Acknowledgment: I thank Drs. John Hendrickson, Daniel Flugstad, Richard Kirby, and James Russo, for permission to include cases 3,9, 11, and 13, respectively.
When osteochondritis dissecans affects the trochlea of the femur, a distinct clinical pattern develops: it occurs in adolescent boys, pain occurs with
Arthroscopy,
Vol. 6, No. I, 1990
I also thank Beverly D. Downing, medical librarian at West Seattle Community Hospital, for her skill and persistence in finding the references.
OSTEOCHONDRITIS
DISSECANS
REFERENCES 1. Aichroth P. Osteochondritis dissecans of the knee. A clinical survey. J Bone Joint Stag 1971;53-B:44&7. 2. Arcq M. Behandlung der Osteochondritis Dissecans Durch Knochenspanbolzung. Arch Orthop Uufall-Chir 1974;79: 297-312. 3. Axhausen G. Die Entstehung der Freien Gelenkkorper und Ihre Beziehunaen. Arch F Klin Chir 1912;104:581-678. 4. Bianchi G, Giidanich IF, Zanasi R. Contributo clinic0 e anatomopatologico alla conscenza dell osteochondrosi dissecante. Chir Organi Mov 1955;41:46&87. 5. Fisher AGT. A study of loose bodies composed of cartilage or of cartilage and bone occurring in joints with special reference to their pathology and etiology. Br J Stag 1920;8:493523. 6. Greville MR. Osteochondritis dissecans: treatment by bone grafting. South Med J 1964;57:886-93. 7. Kurzweil PR, Zambetti GJ, Hamilton WG. Osteochondritis dissecans in the lateral patellofemoral groove. Am J Sports Med 1988;16:308-10. 8. Kunzli HF. Beitrag zur Osteochondritis Dissecans. Schweiz Med Wochenschr
1962;92:292-4.
9. Linden B. The incidence of osteochondritis dissecans in the condyles of the femur. Acta Orthop Stand 1976;47:664-7. 10. Linden B. Osteochondritis dissecans of the femoral condyles. J Bone Joint Stag 1977;59-A:769-76. 11. Linden CB. Osteochondritis dissecans. The natural course
OF THE TROCHLEA
OF THE FEMUR
17
and incidence in femur condyles. [Ph.D. Dissertation] Lund, Sweden: 1976. Malmii General Hospital (University of Lund School of Medicine). 12. Lindholm TS. Osteochondritis dissecans of the knee. A clinical study. Ann Chir Gynaecol Fenn 1974;63:69-76. 13. Mollan RAB. Osteochondritis dissecans of the knee. A case report of an unusual lesion on the lateral femoral condyle. Acta Orthop Stand 1977;48:517-19. 14. Moulonguet P. Foreign bodies in joints. J Bone Joint Suru
1929;l lL353-64. 15. Poulsen K. 3 Tiefalde of Mus Articuli Genus. Hosp Tidende, BD VIII No. 24. 1890. (Cited in Centralblatt fur Chirurnie No. 43, V. 17, 1890;831.j 16. Rehbein F. Die Entstehung der Osteochondritis Dissecans. Arch KZin Chir 1950;265:69-114. 17. Roberts HM. Osteochondritis dissecans. In: Kennedy JC, ed. The injured adolescent knee. Baltimore: Williams &
Wilkins, 1979:1214. 18. Scheller S. Roentgenographic studies of epiphyseal growth and ossification of the knee. Acta Radio1 Sum11 _L 1960:195: l303. 19. Smillie IS. Diseases of the knee joint. Edinburgh: Churchill Livingstone, 1974:360-89. 20. Wells C. Osteochondritis dissecans in ancient British skeletal material. Med Hist 1974;18:365-9. 21. Hughston JC, Hergenroeder PT, Courtenay BG. Osteochondritis dissecans of the femoral condyles. J Bone Joint Surg 1984;66-A:l340-8.
Arthroscopy, Vol. 6, No. 1, 1990
