INTRODUCTION
Loss of smoothness of the articular cartilage leads to pain, loss of range of motion, crepitus with motion, swelling and eventual deformity. Several different techniques have been employed over the years in an effort to restore this joint surface; however, most of these have only resulted in fibrocartilage repair, rather than true hyaline cartilage restoration. Newer techniques of cartilage transplantation require harvesting of weight-bearing cartilage or expensive in vitro growth of cells followed by open surgery. The purpose of this study was to evaluate the outcome of a novel arthroscopic technique of cartilage transplantation. This IRBapproved prospective study presents the results of a two to twelve year follow-up of 125 patients who received an osteocartilagenous paste graft for traumatic or arthritic lesions.

METHODS
Surgical Technique: The surgical steps included preparation of the arthritic lesion by chondroplasty followed by morselization with an awl to create a complete fracture bed and a robust healing response (Figure 1a). Articular cartilage and bone were harvested from the intercondylar notch and then formed into a paste in a graft impactor (Figures 1b,c). The paste was then impacted into the defect with the aid of a trephine and held into place for a few minutes while clotting occurred to fix the graft (Figure 1d).

Patients: Of these 125 patients, 93 (74.4%) had undergone previous surgical procedures to the affected knee with a mean number of past surgeries of 1.70 (1 to 7). The remaining 32 patients failed to respond to non-operative care. 82 males and 43 females were evaluated with an average age of 46 years, average length of follow-up was 50.4 months. The mean size of the defects was 228 mm2 (range 21-1400 mm2). Lesions grafted included 76 medial femoral condyles, 23 lateral femoral condyles, 24 trochleas and 11 tibial plateaus.

Statistics: The nonparametric Wilcoxon signed ranks test, for repeated measures, was used to compare pre-operative and post-operative effects of surgery. Statistical power with n=125, was over 90%.

RESULTS
Of 145 patients that underwent paste grafting over 12 years, 20 patients were lost to follow-up, (2 deceased, 18 no response to contacts), leaving 125 available for evaluation. 105 were considered successful by at least one grade of improvement in the WOMAC pain scale of 0-3. 2 patients had no improvement and 18 patients pain worsened over time, 10 of which underwent joint arthroplasty, and were considered failures. Therefore, 105 of 125 (84%) patients available for follow-up were successfully treated. WOMAC pain and functioning scores improved significantly, with corresponding p-values of 0.0001 and 0.008. There were no significant differences due to age or gender, nor were there significant differences based on the number of past surgeries a patient underwent or the number of concomitant procedures. Histology: There was spatial variability within the biopsies obtained. None were truly normal. 66 biopsies from 61 patients revealed generally smooth grafts.

Collagen typing revealed a variable concentration of Type I and Type II collagen by qualitative gel electrophoresis and by immunohistochemical labeling. An independent blinded reviewer (DS) could not disprove that 42 of 66 biopsies (63.6%) showed strong and consistent evidence of replacement of their articular surface by histological criteria. 18 of 66 (27.3%) biopsies were judged to have areas of cartilage in the biopsy indistinguishable from normal.

CONCLUSION
Paste grafting for arthritic areas in the knee provides durable tissue and pain relief up to 12 years. There are no complications and no significant risk to the procedure. The percent of successful outcomes compares favorably to other techniques.

DISCUSSION
Structural defects in the smooth articular cartilage of a joint can lead to severe physical ailments such as pain and swelling. Injured articular cartilage loses its ability to withstand mechanical pressures and becomes more susceptible to further damage over time. The osteocartilagenous paste graft composite of articular cartilage and cancellous bone appears to provide a supportive matrix for repair and re-growth of cartilage in traumatic and arthritic knees, regardless of whether or not the patients have undergone previous efforts at treatment. The procedure is inexpensive and pain relief is good with overall results linked to careful surgical technique and a well-defined rehabilitation program. Further prospective studies are planned to compare this technique to other cartilage repair procedures.