Technique and Clinical Evaluation of Arthroscopic

Ankle Arthrodesis

Salvatore J. Corso, M.D., and Timothy J. Zimmer, M.D.

Summary: Arthroscopic ankle arthrodesis has recently been shown to be an effec-

tive procedure with significant advantages when properly indicated. We report on
the results of arthroscopic ankle fusion in 16 patients with idiopathic or posttrau-
matic osteoarthritis and rheumatoid disease. We used standard ankle arthroscopic
technique and simple noninvasive distraction with hanging weights. All 16 patients
had a successful fusion at an average of 9.5 weeks postoperatively. Complications
included 1 lateral cutaneous neuroma, and 1 patient who required removal of
screws because of superficial pain. Postoperative evaluation showed complete
resolution of pain in 14 of 16 patients and significant improvement in gait. Fourteen
of 16 patients were completely satisfied with the result and cosmesis, and only 1
patient required shoe modification. These results substantiate previous reports that
arthroscopic ankle arthrodesis is successful, and where indicated, has significant
advantages over the open technique. Key Words: Ankle-Arthrodesis-Fu-
sion-Arthroscopy-Osteoarthritis.

of ankle arthrodesis in arthroscopy to be applied when indicated. One of the

S ince the first description
1879,’ over 30 different surgical approaches have
been described to achieve fusion of the tibiotalar
technical advantages of the arthroscopic procedure is
the maintenance of normal bony contour of the tibia
joint.‘-15 This is a result of a high complication rate and talus. This can be achieved by debridement and
including infection in up to 25% and pseudoarthrosis abrasion of the articular surfaces made easier by visual-
in 20%.‘6-21 As the trend has moved from external ization of the entire joint. The ankle can then be placed
compression using fixation such as the Charnley clamp in a neutral position, usually without malalignment that
and Calandruccio frame, to internal fixation using large can occur with an erroneous osteotomy. This advan-
cancellous screws, the rate of successful fusions has tage becomes a limitation if the patient has significant
increased and the complication rate decreased.22 varus/valgus or anteroposterior malalignment. In these
The initial indications for operative ankle arthros- instances, contouring the joint is difficult arthroscopi-
copy included removal of loose bodies, osteochondritis tally and fusion might be achieved by an open ap-
dissecans, and debridement of an arthritic, inflamed, proach.
or septic joint. With advancement in these techniques,
the capability to extensively abrade articular surfaces MATERIALS AND METHODS
has been realized, thereby allowing the advantages of
We retrospectively reviewed 16 arthroscopic ankle
From Orthopaedic Research of Virginia, and Tuckahoe Orthopae- arthrodeses performed at our institution from October
die Associates, Ltd., Richmond, Virginia, U.S.A. 1991 to July 1993. Eight men and seven women with
Address correspondence and reprint requests to Timothy .I. Zim- an average age of 53 years were evaluated. Twelve
mer, M.D., Orthopaedic Research of Virginia, 8919 Three Chopt
Rd, Richmond, VA 23229, U.S.A. ankles developed posttraumatic osteoarthritis after a
0 1995 by the Arthroscopy Association of North America fracture and open reduction internal fixation. Two
0749-8063/95/1105-1193$3.00/O ankles had idiopathic osteoarthritis, and two had rheu-

Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 11, No 5 (October), 199.5: pp X5-590 585
586 S. J. CORSO AND T. J. ZIiviV4ER

FIG 1. (A) Simple distraction with sterile cling and hanging weights. (B) After draping the unsterile weights, the procedure is performed
with the surgeon sitting.

matoid disease. One patient had bilateral procedures operating room table is elevated so that the ankle is at a
performed. None of the patients had significant varns comfortable height with the surgeon sitting. Following
or valgus malalignment. The average time to standard prepping and draping, the leg is exsangui-
arthrodesis from onset of symptoms was 24 months. nated and the tourniquet inflated. A sterile cling is
The average follow-up was 14 months with a range of placed around the ankle with a loop anteriorly for
7 to 22 months. All patients had an extended trial placement of 20 lb of hanging weights, which provide
of conservative treatment including anti-inflammatory excellent distraction (Fig 1A). The surgeon is seated
medication and bracing or orthotics. The indication for at the foot of the table and incorporated into the sterile
surgery was intractable pain unresponsive to conserva- draping (Fig 1B).
tive measures. The postoperative evaluation included A standard anteromedial portal is established, fol-
clinical signs of fusion, i.e., loss of tenderness or pain lowed by an anterolateral portal, using a blunt trocar
with weightbearing, and roentgenographic confirma- or Wissinger rod. A posterior portal can be used for
tion, which included continuity of bony trabeculae inflow, although we do not use one. Debridement is
across the fusion site. initiated with an arthroscopic 4.5-mm full-radius resec-
tor to debride synovium and debris along the medial
SURGICAL TECHNIQUE and lateral gutter, as well as the talofibular recess. An
arthroscopic burr is then used to denude the articular
This procedure is performed under general or spinal surface down to bleeding cancellous bone (Fig 2). This
anesthesia with the patient in the supine position. The must be done in a systematic fashion to ensure com-
 TECHNIQUE FOR ARTHROSCOPIC ANKLE ARTHRODESIS 581

FIG 4. External view with the arthroscope placed laterally and the
motorized burr placed medially.

FIG 2. A motorized burr debriding articular cartilage of the talus

down to bleeding, cancellous bone. The abrasion is performed in a
systematic fashion to ensure that all articular cartilage is removed. be repeated for the medial half of the joint, placing the
arthroscope laterally and the instrumentation medially
(Fig 4).
pleteness and smooth contouring of the fusion site. It Once the surfaces are well abraded, the arthroscopic
is unnecessary and undesirable to level the surfaces of instruments are removed. The cling is cut to release
tibia and talus, since maintenance of the normal con- the weights. The ankle is dorsiflexed into a neutral
tour of the tibia and talus is technically easier and position, erring to no more than slight valgus and exter-
allows for easy reduction of the fusion. Curettes can nal rotation. Fixation is achieved using 6.5-mm cannu-
be used to complete the abrasion, especially in the lated cancellous screws. The initial guide pin is drilled
lateral gutter, which may be more difficult to reach obliquely from the medial surface of the tibia into the
with a motorized burr (Fig 3). This process can then talus. This is begun approximately 2 cm above the
ankle joint at an angle of 30” to 40” with the skin.
Pin placement is confirmed using anteroposterior and
lateral fluoroscopy. A small stab incision is made, and
the screw length measured. The screw and washer are
then placed.
A second guide pin is placed laterally through the
fibula and tibia, and into the talus, crossing the initial
screw. We additionally use a third screw placed
obliquely from anterior tibia and angled posteriorly
into the talus. Following placement of the screws, fixa-
tion of the tibiotalar and mobility of the subtalar joint is
confirmed. Anteroposterior and lateral roentgenograms
are obtained (Fig 5). The wounds are closed with nylon
suture, and the patient is placed in a bulky dressing.
On the first postoperative day, the dressing is
changed and the patient is placed in a cast-boot. The
postoperative protocol includes non-weightbearing for
a 4-week period. After 4 weeks, patients begin toe-
touch ambulation progressing to full weightbearing as
tolerated. Fusion is established when, clinically, the
FIG 3. Curettes are used especially in the gutters, where it may
be difficult to reach with a motorized burr. Illustration shows a patients have no pain on ambulation or palpation of
curette used to debride the lateral gutter. the joint. Roentgenographic confirmation is demon-
588 S. J. CORSO AND T. J, ZIMMER

FIG 5. Postoper .ative ant1 x0-

posterior and later al roe :ntgc :no-
grams following arthI USCl 2pic
ankle arthrodesis.

strated by osseous bridging traversing the fusion site. One patient, who continued to have pain in her foot,
The patients can then begin activity as tolerated. had rheumatoid arthritis. Her postoperative pain was
not localized to the tibiotalar joint, but rather was dif-
RESULTS fuse pain throughout the foot, and was felt to be sec-
ondary to her disease. The other patient with pain had
Patients were evaluated at 2- to 3-week intervals undergone open reduction internal fixation of a previ-
postoperatively until fusion was achieved. At follow- ous ankle fracture; she had some improvement after
up examination, patients respond to questions about removal of her hardware. Her pain seemed to have
pain, activity, shoe modification, cosmesis, foot numb- originated from the subtalar joint, which roentgeno-
ness, and satisfaction with surgery. Physical examina- graphically revealed degenerative changes.
tion and gait evaluation are performed. A lateral roent- One patient had a lateral cutaneous neuroma. How-
genogram is used to measure the angle of the fusion. ever, even with this minor setback, he was back to
Roentgenographic fusion occurred in all patients; how- work and was completely satisfied with his surgery.
ever, two patients continued to have severe pain post-
operatively and were dissatisfied with the result. These DISCUSSION
will be discussed subsequently. Fourteen of 16 patients
reported complete relief of their severe preoperative Advances in arthroscopy and distraction techniques
pain. All of these patients returned to the daily activi- of the ankle allow the orthopaedic surgeon to perform
ties they were engaged in before they began experienc- ankle arthrodesis arthroscopically where open
ing the significant preoperative pain. Some of the arthrodesis was previously indicated. The advantages
younger, more active patients ,returned to sports, in- of improved cosmesis and decreased risk of wound
cluding basketball, tennis, and soccer. Four of the 14 complications in compromised patients, i.e., rheuma-
patients had an antalgic gait, but only 1 patient required toid arthritis patients on steroid therapy, is obvious.23
a shoe modification. Swelling was residual in 6 of 14 Specific to the ankle, there are the added advantages
patients, although it was significant only in the 2 pa- of the ability to visualize the entire tibiotalar joint and
tients with continued pain. The average time to fusion to maintain the normal bony contour of the convex
was 9.5 weeks. talus and concave tibia1 plafond, which allow for a
 TECHNIQUE FOR ARTHROSCOPIC ANKLE ARTHRODESIS 589

more “anatomic” reduction with less chance of mal- geon, more importantly, must recognize the limitations
alignment. This latter advantage is predicated on the of this procedure, especially in those patients with sig-
important contraindications of this procedure. Patients nificant preoperative deformity or bone loss. When
with significant varus/valgus or anteroposterior defor- properly indicated, arthroscopic ankle fusion compares
mity that prevents correction of the ankle joint to a favorably with the more conventional open technique
neutral position arthroscopically should undergo the with significant advantages and predictability.
open procedure to allow for proper correction by oste-
otomy. Other reports in the literature have shown a
faster fusion rate following the arthroscopic method,24 REFERENCES
compared with the 3- to 12-month fusion rate reported
Albert E. Zur resektion des Kniegelenkes. Wien Med Press
with the open technique.” This more rapid union rate 1879;20:705-708.
was evidenced in this study, probably because of the Charnley J. Compression arthrodesis of the ankle and the shoul-
avoidance of extensive dissection and periosteal strip- der. J Bone Joint Surg Br 1959;33:180-191.
Heinig CF, Dupuy DN. Anterior dowel fusion of the ankle. In:
ping. This is especially applicable to patients with vas- Bateman JE, ed. Foot science. Philadelphia: Saunders,
cular compromise or previous surgery. One such pa- 1976:150-155.
tient in this study had a previous open fracture with a 4. Morgan CD, Hankey JA, Bailey RW, Kaufer H. Long-term
results of tibiotalar arthrodesis. J Bone Joint Surg Am 1985;
vascularized muscle flap that would have been compro- 67:546-549.
mised by an open arthrodesis. Arthroscopically, we 5. Staples OS. Posterior arthrodesis of the ankle and subtalar joints.
were able to avoid the area, and the patient responded J Bone Joint Surg Am 1956;38:50-56.
6. Verhelst MP, Mulier JC, Hoogmartens MJ, Spaas F. Arthrodesis
with an early fusion and excellent result. of the ankle joint with complete removal of the distal part of
Successful debridement and abrasion of the ankle the fibula. Experience with the transfibular approach and three
different types of fixators. Clin Orthop 1976; 118:93-99.
arthroscopically is dependent on adequate distraction
I. White AA III. A precision posterior ankle fusion. Clin Orthop
of the joint. An external distractor, applied medially 1974;98:239-250.
as described by Morganz3 or laterally as described by 8. Bingold AC. Ankle and subtalar fusion by a transarticular graft.
J Bone Joint Surg Br 1956;38:862-868.
GuhIz6 is commonly used. Another less invasive
9. Canale ST, Kelly FB, Jr. Fractures of the neck of the talus:
method involves use of the arthrobot distractor, which Long-term evaluation of 71 cases. J Bone Joint Surg Am
is an expensive device with less adequate results. The 1978;60:143-156.
10. Chuinard EG, Peterson RE. Distraction-compression bone graft
distraction method we used was similar to one de- arthrodesis of the ankle: A method especially applicable in chil-
scribed by Grana.” It is a safe, simple, noninvasive dren. J Bone Joint Surg Am 1963;45:481-490.
technique that provides excellent distraction at mini- 11. Graham CE. A new method of arthrodesis of an ankle joint.
mal cost. It avoids the added puncture wounds and the Clin Orthop 1970;68:75-79.
12. Stewart MJ, Beeler TC, McConnell JC. Compression arthrodesis
risks of placing pins into bone, and is technically easier of the ankle and evaluation of a cosmetic modification. J Bone
to apply. Twenty pounds of weight was all that was Joint Surg Am 1983;65:219-225.
necessary for excellent visibility of the entire joint, and 13. Morris HD, Hand WL, Dunn AW. The modified blair fusion
for fractures of the talus. J Bone Joint Surg Am 1971;53:1289-
it was not necessary to add weight or convert to an 1297.
open procedure because of lack of visualization. 14. Thomas RB. Arthrodesis of the ankle. J Bone Joint Surg Br
One of our clinical failures occurred in a patient 1969;51:53-60.
15. Johnson KA. Arthrodesis of the foot and ankle. In: Surgery of
with rheumatoid arthritis. Roentgenographic fusion the foot and ankle. New York: Raven, 1989:151-168.
seems to have been accomplished and her pain is more 16. Ahlberg A, Henricson AS. Late results of ankle fusion. Actu
diffuse, although she denies any improvement of her Urthop Stand 1981;52:103-105.
17. Boobbyer GN. Long-term results of ankle arthrodesis. Acta Or-
preoperative pain. Patients with more extensive thop Stand 1981;52:107-110.
involvement of their foot, as in rheumatoid arthritis, 18. Johnson EW, Jr, Bosecker EH. Arthrodesis of the ankle. Arch
may be more amenable to an open procedure with Surg 1968;97:766-773.
19. Mot-rev _ BF. Wiedeman GP, Jr. Comolications and lonn-term
fusion of multiple joints, i.e., triple arthrodesis. results of ankle arthrodesis following trauma. J Bone Joint Surg
Our results are consistent with previous studies that Am 1980;62:774-784.
reported excellent fusion rates and satisfactory clinical 20. Said E, Hunka L, Siller TM. Where ankle fusion stands today.
J Bone Joint Surg Br 1978;60:211-214.
function.23,24 As with other arthroscopic procedures, 21 Groth HE, Fitch HF. Salvage procedures for complica-
there is a learning curve with performing arthroscopic tions of total ankle arthroplasty. Clin Orthop 1987;224:244-
ankle arthrodesis. Some of the more simple measures 250.
22. Morgan CD. Tibiotalar arthrodesis. Instructional Course pre-
described here, such as noninvasive distraction, may sented at the Annual Meeting of the Arthroscopy Association
aid the less experienced ankle arthroscopist. The sur- of North America, Palm Desert, CA, April 1993.
590 S. J. CORSO AND T. J. ZIMMER

23. Morgan CD. Arthroscopic tibiotalar arthrodesis. In: Guhl JF, ankle: Deep autogenous inlay grafts with maximum cancellous-
ed. Ankle arthroscopy-Pathology and surgical techniques. bone opposition. .I Bone Joint Surg Am 1974;56:63-70.
Thorofare, NJ: SLACK, Inc, 1988:119-122. 26. Guhl JF. History and development of ankle arthroscopy. In:
24. Meyerson MS, Quill G. Ankle arthrodesis: A comparison of an Guhl JF, ed. Ankle arthroscopy-Pathology and surgical tech-
arthroscopic and an open method of treatment. Clin Orthop niques. Thorofare, NJ: SLACK, Inc, 1988:1-6.
1991;268:84-95. 27. Yates CK, Grana WA. A simple distraction technique for ankle
25. Campbell CJ, Rinehart WT, Kalenak A. Arthrodesis of the arthroscopy. Arthroscopy 1988;4:103-105.