Osteochondritis dissecans of the lateral femoral condyle in the adult

Arthroscopy: The Journal of Arthroscopic and Related Surgery 8(4):474-481 Published by Raven Press, Ltd. 0 1992 Arthroscopy Association of North America

Osteochondritis Dissecans of the Lateral Femoral Condyle in the Adult

John C. Garrett, M.D., Kenneth J. Kress, M.D., and Mark Mudano, M.D.

Summary: In osteochondritis dissecans, 15% of the lesions occur in the lateral condyle. In order to understand the significance of these lesions, 27 were studied prospectively from 1983 to 1990 and compared with 20 consecutive cases of lesions of the medial femoral condyle. Lesions of the lateral femoral condyle were larger, and often comprised the entire width of a condyle and resulted in deformation of a significant segment of the femoral condyle. They lay further posteriorly and commonly were associated with mechanical symptoms including buckling or locking. A discernible clunk was unique to these lesions. In addition, lateral lesions were more fragile, often having multiple bony islands that were prone to fragmentation, making replacement difficult if not impossible. Lateral lesions occurred directly within the main force-bearing areas of the condyle, disrupting normal contact areas and possibly leading to more rapid joint deterioration once segments are lost. This has prompted con- cern for reinsertion of articular fragments or reconstruction with osteochondral allografts. Key Words: Osteochondritis dissecans-Osteochondral allograft- Arthritis-Loose body-Abrasion arthroplasty-Avascular necrosis.

Osteochondritis dissecans (OCD) is a rare disorder of unknown etiology. Approximately 15% of the lesions occur in the lateral femoral condyle (l), making these so rare that a general orthopaedic sur- geon may evaluate only one or two within a lifetime and therefore lump them together with those of the medial condyle. However, they differ in subtle but important ways that has caused David Dandy (per- sonal communication) to suggest that they are a distinct entity, if not because of etiology at least by virtue of their clinical behavior. In order to gain a better understanding, 27 cases of OCD of the lateral femoral condyle were studied prospectively.

MATERIALS AND METHODS

Twenty-seven cases of OCD of the lateral femoral condyle diagnosed between 1984 and 1991 were

Address correspondence and reprint requests to Dr. J. C. Gar- rett, 77 Collier Road, Suite 2000, Atlanta, GA 30309, U.S.A.

followed prospectively. All were those of the senior author (J.C.G.) whose interest in OCD and trans- plant surgery may have resulted in an uncommonly high incidence of this disorder due to referral of cases from sources outside his practice. Cases of acute chondral fracture, chondral delamination, and osteonecrosis were eliminated from the study.

The patients included 22 males and five females ages 16-46 years (mean, 25 years) at the time of the study. All exhibited physeal plate closure establish- ing them as adults. All lateral condylar lesions were unilateral although in one case OCD was present in the ipsilateral medial condyle as well as in the medial femoral condyle of the contralateral knee.

Patients were studied clinically as to the type and frequency of symptoms. Each was examined for ev- idence of effusion, atrophy, condylar tenderness, and audible clicks or clunks indicative of mechanical dysfunction.

The diameter of the lesions was measured roentgenographically with standard anteroposterior and lateral views. The product of these two figures was

474

OCD OF LATERAL FEMORAL CONDYLE 475

used as an estimate of the cross-sectional area. An- teroposterior roentgenographs were also used to determine whether lesions occurred in the center or on the margin of the condyle and the diameter of the lesion was divided by the width of the affected femoral condyle to yield an index of involvement. The lateral view was used to determine whether the defect was bisected by or fell anterior or posterior to a line that was drawn as an extension of the posterior cortex of the femur. Apparent widening of the space (calculated by measuring the distance from the base of the crater to the opposing tibia1 plateau and then subtracting the width of the medial joint space) was used as an estimate of the space necessary to fill with fibrous tissue if abrasion arthroplasty was performed or with an osteochondral allograft if transplantation was chosen instead. The number of bony islands within the lesion was noted. At follow-up, a search was made for degenerative changes as evidenced by progressive joint space narrowing, osteophyte formation, and sclerosis.

RESULTS

The patients who averaged 25 years of age at the time of the study had an average age of onset of symptoms of 18.4 years. Only seven patients were symptomatic for < 1 year prior to diagnosis, the re- mainder being symptomatic for a mean average of 6.6 years. All were studied at the time when they first presented, which was coincident to the time that articular fracture had occurred or loose bodies extruded leaving craters in their wake. All 27 com-

FIG. 1. Osteochondritis dissecans of the lateral femoral condyle. Note the location of the defect in the midportion of the weight-bearing surface.

257r- I

; 20 .I

m b e

* IS-

0

f

L e io- S

i 0

n

S 5-

0 -.25 .25 - .50 .50 - .75 1.00

Width

m Medial Lateral

FIG. 2. The width of the lesion compared to the width of the condyle as noted on anteroposterior roentgenographs.

plained of pain that varied according to the level of activity, and interferred variously with activities of daily living and sports. Swelling occurred in 21 of the 27 patients. Buckling and locking were reported by 19 and a complaint of recurrent, audible popping was noted in three.

Examination revealed atrophy was present in 17 and effusion in 16. Tenderness was noted in 16 and a frank clunk was palpated in 12.

Roentgenographically, anteroposterior films revealed 100% of the lesions to be in the mid portion of the lateral femoral condyle (Fig. 1). Twenty-one extended across the entire width of the condyle, whereas six involved only 50-75% of the transverse diameter (Fig. 2). Of the 27 cases, 21 exhibited apparent joint space widening. As viewed on lateral roentgenograph, seven were bisected by a line extending down the posterior cortex of the femur, whereas 20 lay posterior to the line (Fig. 3); five of these comprised the far posterior aspect of the condyle, had no bony backing posteriorly, and ap- peared to have shifted posteriorly “en mass” (Fig. 4A and B). When measured according to cross- sectional area, none of the lesions was <4 cm2, three were 4-8 cm2, 12 were 8-12 cm’, 10 were 12-16 cm*, and two were 16 cm’. Twenty of the lesions had one or more discernible bony islands, three had two, three had three, and one had four or more (Figs. 5 and 6). Two of the patients, cases 6

Arthroscopy. Vol. 8. No. 4, 1992

476 .I. C. GARRETT ET AL.

FIG. 3. A lateral view demonstrating the typical position of osteochondritis dissecans of the lateral femoral condyle, which is found posterior to a line drawn as an extension of the posterior cortex of the femur.

and 14, exhibited degenerative changes as evidenced by joint space narrowing, sclerosis, and osteophyte formation (Fig. 7).

The cases of OCD of the lateral femoral condyle

4A-C

were compared roentogenographically with 20 consecutive cases of the medial femoral condyle. All of the lesions of the medial condylar were marginal in location with their epicenter in the lateral aspect of the condyle (Fig. 8A). Only one of the 20 involved <25% of the width of the condyle, 12 involved 25- 50% and seven involved 5&75%. None extended the entire width of the condyle (Fig. 2). As viewed on lateral roentgenograms (Fig. 8B), three fell anterior to the line extending down the posterior cortex of the femoral condyle whereas 17 were bisected by this line (Figs. 8B and 9). Three of the medial lesions were <4 cm2, 14 were 4-8 cm2, two were 8-12 cm2 and only one was 12-16 cm2. None was >16 cm2 (Fig. 10). Eighteen of the lesions had only one bony island with the remaining two each having two. None had three or more (Fig. 6). None of the cases exhibited degenerative changes.

TREATMENT

Many of the patients had undergone surgery prior to referral to J.C.G. Indeed, the fact that various forms of treatment had been tried prior to transfer may have created a skewed population representing the worse possible cases rather than a representa- tive group. In situ pinning of loose fragments were performed in five patients. Two of these subse- quently underwent loose body excision as did 23

FIG. 4. A: AP view of gigantic lesion of osteochondritis dissecans of the far posterior aspect of the lateral femoral condyle. Note multiple areas of calcification and apparent flattening of the lateral femoral condyle and widening of the lateral joint space. B: Lateral view. Note extrusion of the osteochondral lesion posteriorly. C: Crater following excision.

Arthroscopy, Vol. 8. No. 4, 1992


SA,B

FIG. 5. A: Lateral view of the with multiple bony islands. B: puted tomography sagittal view distal femur reveals even greate mentation with multiple minute i of calcification interspersed area brous tissue.

lesion Com- of the :r fraa-

others, making a total of 25. Seventeen of these patients underwent abrasion and drilling of the crater. Lateral meniscectomy was performed in three cases. Arthroscopy was used to delineate the status of the crater in 15 cases. Two patients underwent varus osteotomy of the distal femur. The first (case

6), at age 30 years, 15 years following excision of her fragment, exhibited moderate lateral compartment narrowing, spur formation, and genu valgum. The second (case 14), at 21 years of age, five years following excision of an osteochondral fragment,

7A,B

25, -

I I

f: 20

m b e

L e 10 s i 0

n

I 2 3 4 or more

Number of Bony Islands

m Medial ~ Lateral

FIG. 6. The number of bony islands compared to the number of lesions.

FIG. 7. A: A 31-year-old female. Osteotomy for lateral compartment arthritis and genu valgum 12 years following removal of an osteochondral fragment for osteochondritis disseeans of the lateral condyle. B: A 22-year-old female. Lateral compartment arthritis and genu valgus six years following removal of an osteochondral fragment for osteochondritis dissecans of the lateral condyle.


478 J. C. GARRETT ET AL.

SA,B

exhibited lateral compartment collapse, spur formation, and valgus deformity. She underwent a total of nine operations including femoral osteotomy and total knee replacement, the latter performed at an- other institution. Sixteen of the patients eventually underwent osteochondral allografts with follow-up of 6 months to 7 years.

25

N 20 U

m b e

r 15 0

f

L ; 10

i 0

n

s

5

/ Anterior Bisected Posterior

m Medial ~ Lateral

FIG. 9. The position of the lesions as noted on lateral roentgenographs with reference to a line extending along the posterior cortex of the femur.

FIG. 8. A: Osteochondritis dissecans of the medial femoral condyle. The typical location situated eccentrically toward the lateral aspect of the medial femoral condyle. B: Lateral view demonstrating the anteroposterior position of the lesion which typically straddles a line drawn as an extension of the posterior cortex of the femur.

DISCUSSION

The results of this study indicate that in adults with OCD there are indeed differences between lesions of the lateral when compared to their coun- terparts of the medial condyle. Thus, Dandy seems vindicated in his statement that the two exhibit dis- parate behavior although there was nothing in this study to suggest that this stems from dissimilar etiology. Rather, it appears that size, site, and fragility

16

14 N

1 12 b e r 10

0 f 8 L

t 6 I

:4

S

2

0 L o-4 6-12

Size

12-16 ,16

m Medial ~ Lateral

FIG. 10. The size of the lesions, measured in cm’.

Arthroscopy, Vol. 8, No. 4, 1992

OCD OF LATERAL FEMORAL CONDYLE

are determining factors. Similarly, it might be ex- pected that OCD of the patella (2,3) or trochlea have distinct modes of expression without separate origins.

Care must be taken to distinguish OCD from other focal disorders of articular cartilage, namely osteochondral fracture chondral or delamination, or osteonecrosis (4). Similarly, the juvenile form that carries an excellent prognosis (5) should be differ- entiated from its more troublesome adult counter- part.

As long as the osteochondral fragment remains intact and articular congruity is maintained, then normal loads are sustained and wear is unaltered from the norm. Pain presumably appears when loosening occurs. However, once the articular surface becomes flattened or segments sequestered and lost, the result is dramatic. Probably all craters evoke symptoms at a given level of activity but 2 cm in diameter, 4 cm2 in cross-sectional area, seems the watershed of clinical significance. As previously noted by Hughston et al. (6), lesions of the lateral condyle are strikingly larger than those of the medial condyle and usually more symptomatic when extrusion occurs. None of the lateral lesions was smaller than 4 cm2 compared to three of 20 of those of the medial condyle. At the other end of the spec- trum, 12 of the 27 lesions of the lateral femoral condyle were 12 cm2 or greater in size, winning the appellation “osteochondritis dissecans gigantica” (7) a mark earned by only one of the 20 lesions of the medial condyle. Of 27, 21 of the lateral lesions as compared to none of those of the medial condyle extended the entire width of the condyle leaving no articular lip to guide the opposing tibia1 articular surface as the knee flexed. Because of their location further posterior and lateral on the femoral condyles, lateral lesions tended to disrupt tibiofemoral kinematics in deep flexion, producing a sudden halt or catch in deep flexion in 12 of the 27 cases of craters of the lateral condyle. A discernible clunk indicative of a sudden break in the smooth gliding and rolling of the femur on the tibia that occurs because of the loss of a significant length and full width segment of condyle is unique to lateral lesions. Patellofemoral mechanics may be altered by the more anteriorly situated lesions of the medial condyle [Fig. 11 (8)].

Lesions of OCD are typically first diagnosed and evaluated with roentgenograms. Routine anteroposterior and lateral and especially tunnel views remain the mainstay for judging these lesions, although

FIG. 11. Lateral view of the lateral femoral condyle revealing the patellofemoral articulation, tibiofemoral articulation in extension, and tibiofemoral articulation in flexion.

computer assisted tomographic scan and magnetic resonance images are of value as well. Within the roentgenograms are clues to deciphering clinical behavior, specifically fragility when still attached, and the extent of symptoms and arthritic changes once fragments are lost. Not surprisingly, roentgenograms correlate well with operative findings. Other than size and location, there are three roentgeno- graphic features that are of note.

First, lesions with multiple bony islands (seven of 27 lateral, two of 20 medial) often fragment one piece at a time, no single piece being of sufficient size to warrant, or for that matter to permit, reat- tachment (Fig. 5). Second, large lesions of the far posterior aspect of the condyle (five of 27 lateral, zero of 20 medial), lack a proper buttress of articular cartilage to their rear and may shift posteriorly, en mass, when they are subject to the shear stress inherent in normal knee function (Fig. 4B). And third, apparent widening of the joint space seen on tunnel views typically correlated with deformation and actual flattening of the joint surface, with or without a break in the articular cartilage (21 of 27 cases of lateral lesions, zero of 20 of medial lesions; Fig. 4A).

All of these factors are most commonly found in lateral condylar lesions, which as a group seem the most fragile, symptomatic, and difficult to treat.

Biomechanically, the movement of the condyles on the plateau may be studied in a manner similar to a bearing. Wear is related to the coefficient of fric- tion, point contact loads, and use. Surface irregu- larities tend to increase wear because of cutting,


480

12A,B

J. C. GARRETT ET AL.

FIG. 12. A: Plug graft. B: placement of the entire post condyle.

Re- .erior

plowing, and filing. Loss of major portions of the weight-bearing surface, such as occur when osteochondral fragments are lost, increases surface stress in the remaining segments, which is the rim surrounding the crater. Forces may reach extreme levels when craters occur within the center of a condyle, a high stress-bearing area, a common oc- currence with those of the lateral femoral condyle (9). Wear is accelerated but occurs by degrees and is dependent upon the applied load, the presence of a functioning meniscus, the angulation of the limb, and the level of sustained activity. It is evidenced by a gradual thinning of the articular cartilage surrounding the crater rather than catastrophic failure with fragmentation of large portions of the articular surface. Initially the process is subtle and difficult to measure. Later, it is obvious and easy to quan- titate in terms of progressive joint space narrowing.

Which craters carry the worst prognosis when left untreated? This study does not answer this question. Only a few were left unoperated. In addition, the fact that many cases were referred after failing to respond favorably to osteochondral fragment excision may have created a skewed population. However, long-term studies demonstrated that

Arthroscopy, Vol. 8, No. 4, 1992

large lesions can be quite troublesome (10). Loss of the meniscus (three of 27 lateral, zero of 20 medial), often mauled by a roughened articular surface, adds to the woes as does adverse angulation of the limb, excess weight, and activity. Lateral femoral condylar lesions may exhibit narrowing, sclerosis, and osteophyte formation within 10-15 years. The most striking example in this study included cases 6 and 14, who developed lateral joint space narrowing and genu valgum and underwent osteotomy 5 and 12 years, respectively, following excision of an osteochondral fragment (Fig. 7). None of the 20 consecutive cases of lesion is of the medial condyle, which had been followed a maximum of 9 years, exhibited similar findings. Of the 181 cases of OCD of the medial condyle in J.C.G.‘s practice from 1986 to 1991, significant narrowing of the medial joint space has been noted in only three, all following excision of an osteochondral fragment and followed 20, 21, and 25 years, respectively. These figures, albeit small, suggest that loss of a major articular segment is significant and that lesions of the lateral condyle more rapidly develop significant arthritis.

How should lesions be treated? The ideal is to maintain articular congruity (7,11,12). Pinning and


bone grafting of in situ lesions (13,14), and replacement of lost fragments when possible and bone grafting when necessary remains the norm. For craters, abration arthroplasty may suffice especially with those <2 cm, especially those in younger patients (15). The more closely the resultant tissue mimics hyaline articular cartilage, the better the result in terms of symptoms and wear. Thus, medial lesions, which often are small, should fare better. The larger the lesion and the more ingrowth resem- bles fibrous tissue and fails to recreate a convex surface, the less it will resist compressive loads thus accepting the normal share of stress applied to the condyle. Therefore, lateral lesions, which are typically larger, fare worse.

This experience and logic has led us to use transplantation for large articular deficits, a solution that was used in 16 of the 27 cases (Fig. 12). Similar grafts have been utilized successfully in cases of trauma (16-18). Indeed, the strongest argument for the use of these grafts is for the large lesions of the lateral condyle, especially those that have not re- sponded adequately to abrasion arthroplasty. Fif- teen of the grafts followed up to 7 years have suc- ceeded in eliminating pain and swelling, the 16th having crumbled leaving a crater in its wake. The results of the study, however, should not be con- strued to provide proof of the validity of this ap- proach, but rather, should be seen as a basis for the logic of the treatment regimen.

I.

2.

REFERENCES

Aichroth P. Osteochondritis dissecans of the bone. J Bone Joint Surg [Br] 1971;53:44&7. Desai SS, Pate1 MR, Michelli LJ, Silver JW, Lidge RT. Os- teochondritis dissecans of the patella. J Bone Joint Surg [Br] 1987;69:320-5.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

Stougaard JJ. Osteochondritis dissecans of the patella. Acta Orthop Stand 1974;45: 11 l-18. Bradley J, Dandy DJ. Osteochondritis dissecans and other lesions of the femoral condyles. J Bone Joint Surg 1989;71B:538-22. Green W, Banks H. Osteochondritis dissecans in children. J Bone Joint Surg [Am] 1958;14:2w7. Hughston JC, Hergenroeder PT. Courternay BG. Osteo- chondritis dissecans of the femoral condyles. J Bone Joint Surg [Am] 1984;66:13w8. Outerbridge RE. Osteochondritis dissecans of the posterior femoral condyle. Clin Orthop 1983;175:121-9. Freeman MAR. The Anatomy of the Distal Femur with Spe- cial Reference to the Kinematics of the Knee: Scientific Meeting, Program/Abstracts. The Knee Society, February 11. 1990; p. 12-. Baratz MR, Fu FH, Mengato R. Meniscal tears: the effect of meniscectomy and of repair on intra-articular contact areas and stress in the human knee: a preliminary report. Am J Sports Med 1986;14:270-75. Linden B. Osteochondritis dissecans of the femoral condyles: a long-term follow-up study. J Bone Joint Surg [Am] 1977;59:769-76. Smillie IS. Treatment of osteochondritis dissecans. J Bone Joint Surg [Br] 1957;39:248-4iO. Lipscomb PR Jr., Lipscomb PR, Bryan RS. Osteochondritis dissecans of the knee with loose fragments. J Bone Joint Surg [Am] 1978;60:23540. Lee CK, Mercurio C. Operative treatment of osteochondritis dissecans in situ by retrograde drilling and cancellous bone graft. Clin Orthop 1961;158:129-36. Thomson NL. Osteochondritis dissecans and osteochondral fragments managed by Herbert compression screw fixation. Clin Orthop 1987;224:71-8. Ewing JW, Voto SJ. Arthroscopic surgical management of osteochondritis dissecans of the knee. Arthroscopy 1988;4: 3740. Meyers M, Akeson W, Convery R. Resurfacing of the knee with fresh osteochondral allograft. J Bone Joint Surg 1989; 71A:70&33.

Gross AE, McKee NH, Pritzker KPH, Langer F. Recon- struction of the skeletal deficits at the knee. Clin Orthop 1983;174:96-111. Locht R, Gross A, Langer F. Late osteochondral allograft resurfacing for tibia1 plateau fractures. J Bone Joint Surg Br 1984;3:328-35.

Arthroscopy, Vol. 8. No. 4, 1992

Documents

Osteochondritis dissecans of the lateral femoral condyle in the adult