transplanted hyaline cartilage. Fibrocartilage „grouting" growing upwards from the prepared cancellous bed will complete the mosaicplasty. The patient is kept non weight bearing for 6-8 weeks, depending upon the size and defect being treated, and encouraged to re-establish full range of motion during this period of time. Determination of the efficacy of the instruments and technique, as well as the crucial importance of non weight bearing in the initial postoperative period is based on animal trials conducted in 1991 (10). In those trials, 18 German Shepherd dogs were treated to mosaicplasty : 18 graftings in weight bearing areas of the knee and 18 graftings in non weight bearing areas. These implantations and a sham group (in which the same operative procedure is done without grafting) were assessed to determine the value of the mosaicplasty at 4, 6, 8, 16, 26 and 52 weeks. Gross, histological and radiographic examinations demonstrated the establishment of hyaline cartilage survival with recipient bonding and the subsidence of grafts when subjected to early weight bearing. Various forms of histological assessment (standard and polarized light, glucoseaminoglycan and collagen typing) showed the viability of the transplanted hyaline cartilage and the fibrocartilage ingrowth between the grafts and in the donor site.

IV. The Patients

Diagnosis of the osteochondral lesions were made by clinical evaluation, X-ray, MRI, ultrasound and CT-arthrography. All patients underwent arthroscopy which was determined to be the definitive method of diagnosing the lesion as well as its size and location. The most consistent preoperative symptom was that of pain at the site of the lesion and locking of the knee joint. The patient's symptoms were present preoperatively for an average of 7 months.

Long term follow up for this report was possible on 27 of the 31 patients who were operated longer than 3 years. The average follow up time has been 46,1 months (32-54 months). There were 17 females and 10 males. The average age of the patients at the time of the operation was 26,4 years (17-36 years).

The patients underwent physical examination including range of motion and strength, and were

asked to complete a subjectiv questionnaire which was part of the modified HSS knee scoring system and modified Cinncinatti scoring system. The possible donor site morbidity was evaluated by the Bandi score system and a part of the patients by MRI and biopsies during controlarthroscopies. Second look arthroscopy was performed on 8 patients at 12 weeks to 4,5 years postop. The donor site and mosaicplasty were evaluated by probing and histologically at these varying states of healing. At 5 patients were indicated a second look to determine the beginning of the professional sport activity and in 3 cases were controlled the patients due to an actual injury (meniscal rupture, rupture of the ACL, etc.).

V. Surgical technique

Often cartilaginous lesions are only defined at arthroscopy. If the preoperative differential diagnosis includes such a lesion, the patient should be advised of the possibility of mosaicplasty. This procedure can lead to a potential overnight stay and altered weight bearing status. Preoperatively, the patient is given i.v. antibiotics and the procedure is done under general or regional anasthesia and tourniquet control. The table set up should allow for knee flexion to 120°. Once the lesion has been identified by arthroscopy, it is measured for size and location. Paramount to the success of the operation is placing the grafts perpendicular to the surface. Lesions greater than 15mm in diameter of the femoral condyles are treated by miniarthrotomy due to the inability to place the grafts perpendicular to the surface arthroscopically throughout the defect. Loose OCD are generally performed by arthrotomy. Lesions under 15mm in diameter on the femoral condyles can be treated arthroscopically. Depending upon which condyle is involved, it is helpful to move this incision medially or laterally a finger breath. Cave, Z and Blount retractors should be available.

Once the exposure has been made, the technique for medial and lateral condyles is identical. With the use of curettes and knife blade, the edges of the defect are brought back to good hyaline cartilage. The base of the lesion is then abraded to viable subchondral bone. The resulting bleeding floor will promote fibrocartilage „grouting" ingrowth

between the grafts. By visual and two hand control, the first recipient hole is made perpendicular to the chondral surface. 4,5 mm and 3,5 mm drillbits are used. The depth of the lesion should be 20 mm for osteochondral lesion. A specially designed drillguide is used to keep the recipient holes divided by 1mm of bone (Fig.1.).

smooth forceps, the graft is delivered into the defect hole. If the implantation is done arthroscopically special guides, drillbits, transfer tubes and dilating rods are used to insure proper placement of the grafts. Once all the grafts have been thus seated, the knee is put through a range of motion with either or valgus stress depending upon which condyle has been treated. a single drain is used and closure is performed in a routine fashion. The patient is treated with antibiotics (Mandokef - Lilly 1 x 1gr praeop. and 4 x 1gr postop.), antiinflammatories and analgesics (narcotics, aspirin, etc.). Instruction is given for a strict non weight bearing for a period of 6-8 weeks depending upon the size of the lesion. Isometric and range of motion exercises are encouraged.

VI. Results

From Februray 06, 1992 until December 31, 1996 47 such operations were performed. 31 patients were operated longer than 3 years. 27 of those patients have been available for long term follow up. Of the 23 patients with condylar cartilage defects, 9 were of the medial femoral condyle and 14 of the lateral femoral condyle. 4 patients with OCD of the patella were treated. The extent of the recipient area defect ranged from 1-8,5 square centimeters. An average of 7 (3-15) grafts were used in creating the mosaicplasty (fig. 2, 3, 4 et 5).

The average modified HSS score value was 91,3 (67-100) in 27 follow up patients. Patients operated for condylar defects showed higher scores than patellar mosaicplasties.

Possible disturbances of the patellofemoral junction due to graftharvesting were evaluated by the Bandi Score System among the femoral implantations and found no significant alterations to a controlgroup (femoral osteochondral defects treated only by Pridie drilling or abrasion arthroplasty).

22 patients have returned to their normal activity level and have become reinvolved in competitive sports and work.

There was no clinical or radiological evidence of graft loosening, protuberance or subsidence. The ultrasound, MRI with specific sequences and CT-arthrography revealed good congruity and satisfactory cartilage thickness of the replaced area.

Upon completion of tunnels preparations are made for obtaining the osteochondral grafts. The donor site is the periphery of the femur at the patellofemoral joint. By using the superior medial and lateral edges, up to fifteen 4,5mm grafts can be obtained. If the procedure is done by miniarthrotomy, this area of the knee is accessible. If done arthroscopically, then a small 15mm sagitttal incision just medial or lateral to patellar edge is made. Specially designed tubular chisels are then used to obtain the graft. With the use of a

During the second looks the articular surface was evaluated by visual and histological methods. All surfaces were probed for consistency. In 5 cases of the 8 such cases the cartilage was evaluated histologically. The transplanted cartilage remained hyaline in character and bonding occured at the recipient sites by hyaline-fibrocartilage transitions. Direct visualization and probing showed smooth solid surfaces, except in 2 cases where grade II chondromalacia was found.

Of the 27 cases seen in long term follow up, the initial (3 to 6 months) course was benign. Two cases of painful haemarthrosis were treated by postoperative haematoma evacuation. Subsequently all patients have been treated by postoperative drainage.

VII. Discussion

Localized osteochondral lesions on the weight bearing surfaces of the knee present a major problem in treatment because of the inability of full thickness cartilaginous lesions to regenerate. Abrasion arthroplasty and Pridie drilling have been shown only to promote reparative fibrocartilage and is limited by defect size. Osteochondritis dissecans has not responded consitently to loose body removal, internal fixation, or bone grafting particularly in adult. Allograft transplantation runs the risk of disease transmission and immunological alteration of the hyaline surface. According to the basic thesis of the authors the use of autogenous hyaline cartilage transfer should delay develop

ment of osteoarthritis. Recent reports of autogenous chondrocyte transplantation have been encouraging, though the procedure is two stages and laboratory dependent.

Based upon the author's animal research and clinical experience, the mosaicplasty as a one-step procedure can produce hyaline-like quality congruous surfaces at the site of focal osteochondral defects. The transplanted small diameter cylindrical osteochondral grafts have been shown to incorporate at the insertion sites and have not shown evidence of loosening or subsidence. The elasticity of the bone at the graft recipient interface does cause a consistent stable press fit. Important features of the procedure are that it is one-step and the non weight bearing femoral surfaces of the patellofemoral joint serve as suitable donor sites. Animal studies and follow up arthroscopies have demonstrated these donor sites to refill with cancellous bone and fibrocartilage. No morbidity at these donor sites has been experienced.

During the mentioned period mosaicplasty was used as treatment of osteochondritis dissecans of the talar dome in 11 cases. Their follow up was also perfomed in similar manner (Hannover Scoring System, MRI, CT-arthrography, etc.). These results are being reported upon separately.

Based upon the follow up results, mosaicplasty seems to be an alternative in the treatment of

osteochondral defects of the knee, although further follow ups and more data required to determine the final value of the procedure.

VIII. SUMMARY

A technique for the treatment of focal biomechanical cartilage defects, traumatic chondral defects and osteochondritis of the knee is described. This report concerns 57 patients with the above diagnoses who have undergone mosaicplasty of the knee. The size of the focal defects have ranged from 1-8,5 square centimeters. The surgical technique with the use of specially designed instruments for the mosaicplasty is outlined. The treatment technique involves one operation and is relatively easy to perform using surgical techniques familiar to orthopaedic surgeons. If does require attention to detail with the major point being that the recipient drill holes must be perpendicular to the articular surface. Lesions less than 15mm in diameter can be treated arthroscopically. Fixation of the graft is based upon press fit. To date there have been no incidences of subsidence or loosening of the grafts. The mosaicplasty technique is presented as a new option in the treatment of focal chondral lesions and osteochondritis dissecans of the knee.

1. Abernethy, P., J.- Townsend, P., R.- Rose, R., M.- Radin, E., L. (1978)

Is chondro- malacia patellae a separate clinical entity ?

J. Bone Joint Surg. 60-B : 205.

2. Aichroth, P.- Burwell, R., G.- Laurence, M. (1971)

An experimental study of osteoarticular grafts to replace articular surfaces.

J. Bone Joint Surg. 53-B : 554.

3. Bobic, V. (1996)

Arthroscopic osteochondral autograft transplantation in anterior cruciate ligament reconstruction : a preliminary clinical study.

Knee Surg. Sports Traumatol. Arthroscopy 3 : 262.

4. Brittberg, M.- Lindahl, A.- Nilsson, A. et al. (1994)

Treatment of deep cartilage defects in the knee with autologous chondrocyte transplantation.

New England Journal of Medicine 331 : 889.

5. Campbell, C., J. (1969)

The healing of cartilage defects.

Clin. Orthop. 64 : 65.

6. Coutts, R., D.- Woo, S., L.- Amiel, D. et al. (1992)

Rib perichondrial autografts in full-thickness articular cartilage defects in rabbits.

Clin. Orthop. 275 : 263.

7. Friedlaender, G., E.- Horowitz, M., C. (1992)

Immune responses to osteochondral allografts : nature and significance.

Orthopedics 15 : 1171.

8. Garrett, J., C. (1986)

Treatment of osteochondritis dissecans of the distal femur with fresh osteochondral allografts.

Arthroscopy 2 : 222.

9. Hangody, L.- Kárpáti, Z. (1994)

A new surgical treatment of localised cartilaginous defects of the knee.

Hung. J. of Orthop. Trauma. 37 : 237.

10. Hangody, L.- Kish, G.- Kárpáti, Z et al. (1997)

Autogenous osteochondral graft technique for replacing knee cartialge defects in dogs.

Orthopaedics International Edition 5 : 3 : 1.

11. Hangody, L. Kish, G.- Kárpáti, Z. et al. (1996)

Treatment of osteochondritis dissecans of the talus : The use of the mosaicplasty technique - preliminary report.

Foot and Ankle International accepted for publication.

12. Hangody, L.- Sükösd, L.- Szigeti, I.- Kárpáti, Z. (1996)

Arthroscopic autogenous osteochondral mosaicplasty.

Hung. J. of Orthop. Traum. 39 : 49.

13. Hangody, L.- Szigeti, I.- Kárpáti, Z.- Sükösd, L. (1996)

Neue Methode in der Behandlung der schweren, lokalen Knorpelschäden im Kniegelenk.

Osteosynthese International 3 : 106.

14. Hungerford, D., S.- Barry, M. (1979)

Biomechanics of the patellofemoral joint.

Clin. Orthop. 144 : 9.

15. Insall, J. (1974)

The Pridie debridement operation for osteoarthritis of the knee.

Clin. Orthop. 101 : 61.

16. Johnson, L., L. (1986)

Surgical arthroscopy : principles and practice.

St.Louis The C. V. Mosby Co.

17. Lexer, E. (1908)

Substitution of whole or half joints from freshly amputated extremities by freeplastic operation.

Surg. Gynecol. Obstet. 6 : 601.

18. Mankin, H., J.- Doppelt, S., H.- Tomford, W., W. (1983)

Clinical experience with allograft implantation : the first 10 years.

Clin. Orthop. 174 : 69.

19. Matsusue, Y.- Yamamuro, T.- Hama, M. (1993)

Arthroscopic multiple osteochondral transplantation to the chondral defect in the knee associated with anterior cruciate ligament disruption : case report.

Arthroscopy 9 : 318.

20. O'Driscoll, S., W.- Salter, R., B. (1986)

The repair of major osteochondral defects in joint surfaces by neochondrogenesis with autogenous osteoperiosteal grafts stimulated by continous passive motion : an experimental investigation in the rabbit.

Clin. Orthop. 208 : 131.

21. O'Driscoll, S., W. (1997)

Cartilage regeneration through periosteal transplantation : basic scientific and clinic studies.

64^th Annual Meeting of the AAOS San Francisco February 13-17. 1997.

22. Oateshott, R., D.- Farine, J.- Pritzker, K., P., H., et al. (1988)

A clinical and histological analysis of failed fresh osteochondral allografts.

Clin. Orthop. 233 : 283.

23. Minas, T. (1997)

Basic scientific studies of chondrocyte transplantation.

64th Annual Meeting of the AAOS San Francisco February 13-17. 1997.

24. Pridie, K., H. (1959)

A method of resurfacing osteoarthritic knee joint.

J. Bone Joint Surg. 41-B : 618.

25. Steadman, J., R.,- Stereet, W., I. (1995)

The surgical treatment of knee injuries in skiers.

Med. Sci. Sports Exerc. 27(3) : 328.

26. Tomford, W., W.- Springfield, D., S.- Mankin, H., J. (1992)

Fresh and frozen articular cartilage allografts.

Orthopedics 15 : 1183.

27. Yamashita, F.- Sakakida, K.- Suzu, F.- Takai, S. (1985)

The transplantation of an autogenic osteochondral fragment for osteochondritis dissecans of the knee.

Clin Orthop. 210 : 43.