Arthroscopic Drilling for the Treatment of
Osteochondral Lesions of the Talus*
BY TSUKASA KUMAI, M.D., PH.D.†, YOSHINORI TAKAKURA, M.D., PH.D.†,
ICHIRO HIGASHIYAMA, M.D.†, AND SUSUMU TAMAI, M.D., PH.D.†, NARA, JAPAN
Investigation performed at Nara Medical University, Nara
Abstract
Background: An osteochondral lesion of the talus
is a relatively rare disorder of the ankle. While a num-
ber of treatment options have been reported, it appears
to be difficult to manage all lesions with a single ap-
proach. We evaluated the indications for and the results
of arthroscopic drilling for the treatment of an os-
teochondral lesion of the talus.
Methods: Eighteen ankles (seventeen patients) with
a symptomatic osteochondral lesion of the talus were
examined. The ages of the patients ranged from ten to
seventy-eight years (mean, 28.0 years) at the time of the
operation, and the patients were followed postopera-
tively for two to 9.5 years (mean, 4.6 years). After the
continuity of the cartilage overlying the lesion and the
stability of the lesion had been confirmed, arthroscopic
drilling was performed with use of a Kirschner wire that
was 1.0 to 1.2 millimeters in diameter. A cast was not
applied postoperatively, and full weight-bearing was al-
lowed six weeks after the procedure.
Results: The clinical result was good for thirteen
ankles and fair for five; all ankles had improvement.
Twelve of the thirteen ankles that were in patients who
were less than thirty years old had a good result. In
contrast, only one of the five ankles in patients who
were fifty years old or more had a good result. Thus,
the clinical results tended to be better for younger
patients. Improvement was seen radiographically in fif-
teen ankles. However, the three ankles in patients who
were more than sixty years old were found to have no
improvement on radiographic examination. Analysis of
the group of patients who had a history of trauma
revealed that the mean interval between the injury and
the operation was 6.3 months for the three ankles that
had a good radiographic result and 11.3 months for
the six that had a fair result. Thus, the radiographic
results tended to be better when the interval between
the injury and the operation was shorter.
Conclusions: Arthroscopic drilling for the treat-
ment of medial osteochondral lesions of the talus does
not require osteotomy of the medial malleolus or post-
*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.
†Department of Orthopaedic Surgery, Nara Medical University,
840 Shijyo-cho, Kashihara, Nara 634-8522, Japan.
Copyright 1999 by The Journal of Bone and Joint Surgery, Incorporated
VOL. 81-A, NO. 9, SEPTEMBER 1999
operative immobilization; thus, the procedure is less
invasive than other types of operative treatment for the
condition and it allows early resumption of daily activ-
ities and sports. On the basis of the results in this study,
we believe that the procedure is effective and useful in
young patients, especially those who have not yet had
closure of the epiphyseal plate. A specific indication
for the procedure is an early lesion with only mild
osteosclerosis of the surrounding talar bone, continuity
of the cartilaginous surface, and stability of the os-
teochondral fragment.
An osteochondral lesion of the talus appears to be
a relatively rare disorder and, although many patients
have a history of trauma, the etiology remains largely
unknown1,4,15. The principles of treatment of this con-
dition have been well established since the description
by Berndt and Harty4 in 1959, and a number of treat-
ment methods have been reported1-4,6,8-10,12-14,16. Berndt and
Harty classified the lesions into four stages. Stage I in-
dicates a small area of compression of subchondral
bone; stage II, a partially detached osteochondral frag-
ment; stage III, a completely detached osteochondral
fragment that remains in the talar crater; and stage IV,
a displaced osteochondral fragment. Treatment is most
commonly based on this radiographic classification
system. Subsequent studies4,6,14 have supported this ap-
proach, which involves a trial of nonoperative treatment
for stage-I and II and medial stage-III lesions and oper-
ative treatment for lateral stage-III lesions and stage-IV
lesions. We have evaluated the treatment of osteochon-
dral lesions of the talus in many patients, and currently
we do not believe that a single approach can be applied
to all lesions; instead, an appropriate treatment must be
selected from the various alternatives for each patient.
Of these alternatives, arthroscopic drilling is simple, is
less invasive than other types of operative treatment,
and necessitates a short duration of hospitalization; thus,
it is considered to be highly advantageous, especially for
students and athletes. In this study, we evaluated the
specific indications for and the results of this procedure.
Materials and Methods
Seventeen (six male and eleven female) patients with
eighteen affected ankles were evaluated (Table I). Their
ages ranged from ten to seventy-eight years (mean, 28.0
years) at the time of the operation, and they were fol-
lowed postoperatively for two to 9.5 years (mean, 4.6
1229
1230 TSUKASA KUMAI, YOSHINORI TAKAKURA, ICHIRO HIGASHIYAMA, AND SUSUMU TAMAI

TABLE I
PATIENT DEMOGRAPHICS AND RESULTS

Gender, History of Duration of Site of Duration of Clinical Radiographic

Case Age at Op. Trauma Symptoms* Lesion Stage4 Follow-up Result Result
(yrs.) (mos.) (mos.)
1 M, 15 Yes 1 Medial II 114 Good Good
2 M, 27 Yes 4 Medial II 100 Good Good
3 F, 13 No 13 Medial II 84 Good Fair
4 F, 50 No 39 Medial II 66 Fair Fair
5 F, 62 No 62 Medial II 64 Fair Poor
6 F, 66 No 38 Medial II 64 Fair Poor
7 F, 10 No 27 Medial II 60 Good Good
12 No 51 Medial II 36 Good Fair
8 F, 78 Yes NA Medial II 60 Fair Poor
9 F, 13 No 26 Medial II 48 Good Fair
10 M, 13 Yes 12 Medial II 42 Good Fair
11 F, 22 No 15 Medial II 42 Good Fair
12 M, 50 Yes 1.5 Medial II 41 Good Fair
13 M, 16 Yes 14 Medial II 40 Good Good
14 F, 13 Yes 18 Medial II 36 Good Fair
15 F, 15 Yes 6 Medial II 36 Good Fair
16 M, 13 Yes 6 Medial II 30 Good Fair
17 F, 16 Yes 24 Medial II 24 Fair Fair
*NA = not available.

years). Ten patients (ten ankles) had a clear history of
trauma, and seven patients (eight ankles) did not. Of
the patients who had a history of trauma, nine had an
inversion injury; three (Cases 1, 13, and 14) of the nine
had been injured during sports activity. The remain-
ing patient (Case 8) was seen a long time after the injury,
and she did not remember the details.
The preoperative symptom of the condition was
usually pain in the ankle joint during or after exercise.
In addition, mild swelling was observed in many pa-
tients. The site of the lesion was medial in all ankles.
According to the radiographic classification system of
Berndt and Harty4, all of the lesions were stage II. Pre-
operative computed tomography scans were made to
obtain information about the osseous lesion in fourteen
ankles, and magnetic resonance imaging was performed
to evaluate the condition of the bone surrounding the
lesion in thirteen ankles.
The seventeen patients were among a group of
seventy-two patients with an osteochondral lesion of
the talus for whom we considered operative treatment
with one of three operative procedures: excision of the
osteochondral fragment and curettage, reduction and
fixation of the osteochondral fragment with bone pegs,
and arthroscopic drilling. As the stability of the lesion
cannot be determined on the basis of plain radiographs
alone, arthroscopic viewing and probing are essential.
Diagnostic arthroscopy was first performed in all pa-
tients; when intact cartilage over the lesion and a stable
fragment were detected with probing, arthroscopic drill-
ing was carried out. When a small unstable fragment
was detected arthroscopically, excision of the fragment
and curettage was performed. In patients who had a
large unstable fragment, the fragment was reduced and
fixed with bone pegs in an open operation. Seventeen
patients (the study group) were managed with arthro-
scopic drilling; twenty-eight, with excision of the frag-
ment; and twenty-seven, with reduction and fixation.
Continuity of the cartilaginous surface and stability of
the osteochondral fragment, confirmed by arthroscopic
probing, were regarded as essential indications for ar-
throscopic drilling.
Operative Procedure
After induction of spinal or general anesthesia, the
patient is placed in the supine position with the cal-
caneal region slightly protruding from the end of the
operating table to allow free plantar flexion and dorsi-
flexion of the ankle. Although there are reports7,11 of
skeletal traction being used to widen the joint space, we
do not use that technique because the joint space is
relatively wide in patients who have this disease, most
of whom are young, and because the lesions are so small
that skilled manual manipulation is needed for drilling.
A medial portal adjacent to the anterior tibial tendon
is used, and a 2.7-millimeter-diameter arthroscope is in-
serted. As the lesions are frequently located in the me-
dial aspect of the talar dome, they are examined with
the ankle joint in plantar flexion. If the junction between
the lesion and the normal area cannot be determined
with arthroscopic viewing alone, probing of the carti-
laginous surface is helpful because the cartilaginous sur-
face of the lesion is often softened; when this is the case,
the tip of the probe dips into the cartilage in the region
of the lesion. In addition, the cartilaginous surface is
observed with regard to its continuity and signs of de-

THE JOURNAL OF BONE AND JOINT SURGERY

 ARTHROSCOPIC DRILLING FOR THE TREATMENT OF OSTEOCHONDRAL LESIONS OF THE TALUS
FIG. 1
Illustration of percutaneous transmalleolar drilling. A Kirschner
wire, 1.0 to 1.2 millimeters in diameter, is inserted from about three
centimeters proximal to the tip of the medial malleolus and is
directed across the medial malleolus into the lesion through the
intact cartilage. Drilling is continued until healthy bleeding from the
bone marrow or leakage of fat droplets is confirmed.
generation, such as fibrillation. The mobility of the os-
teochondral fragment in relation to the bone surround-
ing the lesion and the extent of the lesion are assessed
with a probe.
FIG. 2
Graph showing the relationship between the age at the time of the operation and the clinical result. Twelve of the thirteen ankles that had
a good result were in patients who were less than thirty years old.
VOL. 81-A, NO. 9, SEPTEMBER 1999
1231
The arthroscope is immobilized, and a Kirschner
wire, 1.0 to 1.2 millimeters in diameter, is inserted from
about three centimeters proximal to the tip of the me-
dial malleolus and is directed across the medial mal-
leolus into the lesion through the intact cartilage. The
size of the Kirschner wire was selected mainly on
the basis of the size of the injured region. The position
of the Kirschner wire is confirmed with arthroscopy,
and the wire is drilled into the lesion (percutaneous
transmalleolar drilling) (Fig. 1). Under arthroscopy, the
Kirschner wire is withdrawn to the articular surface at
the distal end of the tibia, and drilling is performed at
a few other sites after slightly changing the angle of
plantar flexion or dorsiflexion of the ankle joint. Drill-
ing is continued until healthy bleeding from the bone
marrow or leakage of fat droplets is confirmed. Next,
the angle of insertion of the Kirschner wire is changed,
and the same procedure is repeated. The drilling is
performed at several sites, including areas around the
lesion. The operation is completed by irrigating the ar-
ticular space.
After the operation, no cast is applied and the oper-
ative site is gently compressed by an elastic dressing.
Plantar flexion and dorsiflexion of the ankle are allowed
from the day after the operation, gradual partial weight-
bearing with use of a crutch is initiated at five weeks,
and full weight-bearing is permitted at about six weeks
after the operation. Sports activities are not permitted
for at least two months.
Results
The clinical evaluation was performed according to
the criteria of Berndt and Harty4. A patient who had
occasional symptoms but no disabling pain was consid-
1232 TSUKASA KUMAI, YOSHINORI TAKAKURA, ICHIRO HIGASHIYAMA, AND SUSUMU TAMAI

FIG. 3
Graph showing the relationship between the age at the time of the operation and the radiographic result. Three of the four ankles that had
a good result were in patients who were less than twenty years old, and all three ankles with a poor result were in patients who were more than
sixty years old.

ered to have a good result. A patient who had a decrease clinical results tended to be better for younger patients
in symptoms but still had some disabling pain was con- (Fig. 2). Ten of the eleven ankles in patients who were
sidered to have a fair result. The result was considered less than twenty years old had a good result.
to be poor if the symptoms had not decreased. The
clinical result was good for thirteen ankles and fair for
five; all ankles had a decrease in the symptoms (Table
I). All thirteen ankles with a good result had recovered
a full range of motion no later than two weeks after the
operation, and the swelling around the ankle had also
decreased. Pain did not recur after walking with full
weight-bearing. The five ankles that were rated as fair
also regained an almost full range of motion two weeks
after the operation. However, four of these five ankles
gradually became slightly swollen again after walking
with weight-bearing was resumed. The five ankles that
were rated as fair were painful after the patients walked
a long distance. Two (Cases 5 and 6) of the five ankles
were painful when the patients went up and down stairs,
and another ankle (Case 17) was slightly painful after
recreational sports. Seven patients (seven ankles) had
engaged in sports as recreational activities before the
operation, and all of them had resumed sports activity
at the time of the evaluation.
Of the ten ankles that had a history of trauma, eight
had a good result and two had a fair result. In con-
trast, of the eight ankles that did not have a history of
trauma, five had a good result and three had a fair
result. Thus, the clinical results tended to be worse for the
ankles that did not have a history of trauma. Of the thir-
teen ankles that were in patients who were less than
thirty years old at the time of the operation, twelve had
FIG. 4-A
a good result. In contrast, only one of the five ankles that
Figs. 4-A, 4-B, and 4-C: Case 7, a ten-year-old girl.
were in patients who were fifty years old or more had Fig. 4-A: Plain radiograph, made at the initial examination, show-
a good result; the other four had a fair result. Thus, the ing irregularity on the medial side of the talar dome (arrow).

THE JOURNAL OF BONE AND JOINT SURGERY

 ARTHROSCOPIC DRILLING FOR THE TREATMENT OF OSTEOCHONDRAL LESIONS OF THE TALUS 1233

atively satisfactory9,10,12-14,16. We have encountered more

than 100 osteochondral lesions of the talus. The size
and stability of the lesion, the continuity of the cartilag-
inous surface, and the condition of the talar bone sur-
rounding the lesion have varied among patients, and
it appears to be inappropriate to treat all lesions with
a single approach. In fact, we choose among three op-
tions — excision of the osteochondral fragment, reduc-
tion and fixation of the osteochondral fragment, and
arthroscopic drilling — according to the condition of
the individual lesion, and we have obtained relatively
good results with each procedure. For this reason, care-
ful selection of the type of treatment is important, and
various imaging modalities must be used effectively
to evaluate the condition of the lesion accurately13,15-17.
Computed tomography scans and magnetic resonance
imaging are particularly useful for assessing the fea-
tures of the lesion. Magnetic resonance imaging also
provides information about the condition of the talar
bone surrounding the lesion17. As a discrepancy between
FIG. 4-B
the radiographic staging system of Berndt and Harty4
Arthroscopy showed that, although there was roughness of the
cartilaginous surface (arrows), its continuity was preserved. Mild and the arthroscopic appearance has been reported16,
softening of the cartilage was detected by probing. direct arthroscopic viewing and probing are also im-
portant for selection of the best treatment method.
The radiographic result was considered to be good We perform excision of the osteochondral fragment and
when complete osseous union or ossification was ob- curettage in patients in whom the osteochondral frag-
served, fair when osseous union or ossification was ment is small and completely detached from the talar
incomplete but improved compared with the preopera-
tive finding, and poor when no changes were observed.
The ankles with a fair radiographic result showed some
tendency toward ossification compared with the pre-
operative finding; mainly, the radiolucent zone distal to
the osteochondral fragment tended to disappear. The
radiographic result was good for four ankles, fair for
eleven, and poor for three (Table I). Of the ten ankles
that had a history of trauma, three, six, and one had a
good, fair, and poor result, respectively. In contrast, of
the eight ankles that did not have a history of trauma,
one, five, and two were rated as good, fair, and poor,
respectively. Radiographic results tended to be worse
for patients who did not have a history of trauma.
Three of the four ankles in which complete osseous
union or ossification was observed were in patients
who were less than twenty years old (Figs. 3 through
4-C), and all three of the ankles in which there were
no radiographic changes compared with the preopera-
tive status were in patients who were more than sixty
years old. Analysis of the group that had a history of
trauma showed that the mean interval between the
injury and the operation was 6.3 months for the three
ankles that had a good radiographic result and 11.3
months for the six that had a fair radiographic result.

Discussion
Various operative techniques for the treatment of FIG. 4-C
osteochondral lesions of the talus have been reported, Plain radiograph made one year after arthroscopic drilling. Heal-
and the results of each type of treatment have been rel- ing is complete.

VOL. 81-A, NO. 9, SEPTEMBER 1999

1234 TSUKASA KUMAI, YOSHINORI TAKAKURA, ICHIRO HIGASHIYAMA, AND SUSUMU TAMAI

bone. We also carry out open reduction and fixation of
large unstable osteochondral fragments, which allows
direct observation of the condition of the lesion and
the talar bone as well as the most accurate morpholog-
ical repair of the articular surface. However, when the
lesion is located on the posteromedial side, osteotomy
of the medial malleolus may be required to expose
the lesion. Moreover, immobilization in a cast and non-
weight-bearing are needed for approximately five and
ten weeks, respectively, and some bone atrophy is in-
evitable10,12,13. Removal of the internal fixation is also
necessary after bone union. In contrast, arthroscopic
drilling does not require osteotomy of the medial mal-
leolus or postoperative immobilization in a cast, so the
procedure is less invasive and allows early resumption
of daily activities and sports2,8,10,13-16.
The consistency of the results of this procedure must
still be evaluated, but fairly good results can be obtained
if the patients are selected carefully. We performed this
procedure in patients in whom the continuity of the
cartilaginous surface and the stability of the lesion had
been confirmed arthroscopically. Angermann and Jen-
sen2 believed that, if the overlying articular cartilage
appears intact, then drilling into the lesion through the
cartilage is adequate treatment. However, on the basis
of the results of the present study, this procedure ap-
pears to be ineffective in middle-aged and older individ-
uals who have a reduced ability to regenerate bone.
Particularly, when a patient has a subchondral bone
cyst, filling in of bone is unlikely to be induced by, and
no radiographic improvement is expected from, drilling
from the articular surface alone, even if the continuity
of the cartilaginous surface is maintained. Although
no patient who had a subchondral cyst was included in
this series, we previously performed drilling in an el-
derly woman who had intact cartilage overlying a sub-
chondral cyst. Filling in of bone was not observed after
the operation, and the patient did not have a good result.
In contrast, this procedure is markedly effective and
useful in young patients, especially those who have not
yet had closure of the epiphyseal plate5. With modern
techniques (arthroscopy and fluoroscopy), we can drill
into the bone without crossing normal cartilage, either
of the medial malleolus or of the dome of the talus.
There have been recent reports on the use of retro-
grade transtalar drilling that does not injure normal car-
tilage8. Recently, we also have been using retrograde
transtalar drilling, but we have managed only a small
number of patients and the duration of postoperative
follow-up is short. This procedure may become more
popular in the near future.
In the present study, the clinical and radiographic
results tended to be better for patients who had a history
of trauma. This may be due to the fact that, when there
is no history of trauma, there is a long duration of symp-
toms with a resultant poor situation for remodeling be-
cause of the sclerotic changes surrounding the lesion.
Patients who did have a history of trauma but a longer
period between the injury and the operation also had
less radiographic improvement. Many patients who
were managed long after the injury had sclerotic talar
bone surrounding the lesion; this sclerosis is represented
as an osteosclerotic image on radiographs and as a rel-
atively wide low-intensity area on T1-weighted mag-
netic resonance images. Therefore, such patients may
have extensive avascularity in the region of the lesion.
Destruction of the sclerotic bone must be ensured at the
time of the operation to induce reestablishment of the
blood flow. The effects of drilling from above the osteo-
chondral fragment alone may be limited in such patients.
Our results suggest that arthroscopic drilling, which
is less invasive than other operative treatments and
can be performed with short hospitalization, should be
considered first for young patients who have an osteo-
chondral lesion of the talus. A specific indication for this
procedure is an early osteochondral lesion with only
mild osteosclerosis of the surrounding talar bone, conti-
nuity of the cartilaginous surface, and stability of the
osteochondral fragment.
In conclusion, arthroscopic drilling for patients who
are less than thirty years old has a major advantage
compared with the traditional osteotomy of the medial
malleolus, as it allows early mobilization and return to
the preinjury status. An early stable osteochondral le-
sion in a young patient should be considered an absolute
indication for the procedure.

References
1. Alexander, A. H., and Lichtman, D. M.: Operative treatment of transchondral talar-dome fractures (osteochondritis dissecans). Long-
term follow-up. J. Bone and Joint Surg., 62-A: 646-652, June 1980.
2. Angermann, P., and Jensen, P.: Osteochondritis dissecans of the talus: long-term results of operative treatment. Foot and Ankle, 10:
161-163, 1989.
3. Bauer, M.; Jonsson, K.; and Lindén, B.: Osteochondritis dissecans of the ankle. A 20-year follow-up study. J. Bone and Joint Surg.,
69-B(1): 93-96, 1987.
4. Berndt, A. L., and Harty, M.: Transchondral fractures (osteochondritis dissecans) of the talus. J. Bone and Joint Surg., 41-A: 988-1020,
Sept. 1959.
5. Bradley, J., and Dandy, D. J.: Results of drilling osteochondritis dissecans before skeletal maturity. J. Bone and Joint Surg., 71-B(4):
642-644, 1989.
6. Canale, S. T., and Belding, R. H.: Osteochondral lesions of the talus. J. Bone and Joint Surg., 62-A: 97-102, Jan. 1980.
7. Ferkel, R. D., and Fischer, S. P.: Progress in ankle arthroscopy. Clin. Orthop., 240: 210-220, 1989.
8. Ferkel, R. D.: Osteochondral lesions of the talus. In Arthroscopic Surgery: the Foot and Ankle, pp. 145-170. Edited by T. L. Whipple. Phil-
adelphia, Lippincott-Raven, 1996.

THE JOURNAL OF BONE AND JOINT SURGERY

 ARTHROSCOPIC DRILLING FOR THE TREATMENT OF OSTEOCHONDRAL LESIONS OF THE TALUS 1235
9. Flick, A. B., and Gould, N.: Osteochondritis dissecans of the talus (transchondral fractures of the talus): review of the literature and new
operative approach for medial dome lesions. Foot and Ankle, 5: 165-185, 1985.
10. Frank, A.; Cohen, P.; Beaufils, P.; and Lamare, J.: Arthroscopic treatment of osteochondral lesions of the talar dome. Arthroscopy, 5:
57-61, 1989.
11. Guhl, J. F.: New concepts (distraction) in ankle arthroscopy. Arthroscopy, 4: 160-167, 1988.
12. O’Farrell, T. A., and Costello, B. G.: Osteochondritis dissecans of the talus. The late results of surgical treatment. J. Bone and Joint Surg.,
64-B(4): 494-497, 1982.
13. Parisien, J. S.: Arthroscopic treatment of osteochondral lesions of the talus. Am. J. Sports Med., 14: 211-217, 1986.
14. Pritsch, M.; Horoshovski, H.; and Farine, I.: Arthroscopic treatment of osteochondral lesions of the talus. J. Bone and Joint Surg., 68-A:
862-865, July 1986.
15. Shea, M. P., and Manoli, A., II: Osteochondral lesions of the talar dome. Foot and Ankle, 14: 48-55, 1993.
16. Van Buecken, K.; Barrack, R. L.; Alexander, A. H.; and Ertl, J. P.: Arthroscopic treatment of transchondral talar dome fractures. Am. J.
Sports Med., 17: 350-356, 1989.
17. Yulish, B. S.; Mulopulos, G. P.; Goodfellow, D. B.; Bryan, P. J.; Modic, M. T.; and Dollinger, B. M.: MR imaging of osteochondral lesions
of the talus. J. Comput. Assist. Tomog., 11: 296-301, 1987.

VOL. 81-A, NO. 9, SEPTEMBER 1999