Clinical and cost-effectiveness of autologous chondrocyte implantation for cartilage defects in knee joints: systematic review and economic evaluation.

OBJECTIVE: To support a review of the guidance issued by the National Institute for Health and Clinical Excellence (NICE) in December 2000 by examining the current clinical and cost-effectiveness evidence on autologous cartilage transplantation. DATA SOURCES: Electronic databases. REVIEW
METHODS: Evidence on clinical effectiveness was obtained from randomised trials, supplemented by data from selected observational studies for longer term results, and for the natural history of chondral lesions. Because of a lack of long-term results on outcomes such as later osteoarthritis and knee replacement, only illustrative modelling was done, using a range of assumptions that seemed reasonable, but were not evidence based.
RESULTS: Four randomised controlled trials were included, as well as observational data from case series. The trials studied a total of 266 patients and the observational studies up to 101 patients. Two studies compared autologous chondrocyte implantation (ACI) with mosaicplasty, the third compared ACI with microfracture, and the fourth compared matrix-guided ACI (MACI) with microfracture. Follow-up was 1 year in one study, and up to 3 years in the remaining three studies. The first trial of ACI versus mosaicplasty found that ACI gave better results than mosaicplasty at 1 year. Overall, 88% had excellent or good results with ACI versus 69% with mosaicplasty. About half of the biopsies after ACI showed hyaline cartilage. The second trial of ACI versus mosaicplasty found little difference in clinical outcomes at 2 years. Disappointingly, biopsies from the ACI group showed fibrocartilage rather than hyaline cartilage. The trial of ACI versus microfracture also found only small differences in outcomes at 2 years. Finally, the trial of MACI versus microfracture contained insufficient long-term results at present, but the study does show the feasibility of doing ACI by the MACI technique. It also suggested that after ACI, it takes 2 years for full-thickness cartilage to be produced. Reliable costs per quality-adjusted life-year (QALY) could not be calculated owing to the absence of necessary data. Simple short-term modelling suggests that the quality of life gain from ACI versus microfracture would have to be between 70 and 100% greater over 2 years for it to be more cost-effective within the 20,000--30,000 pounds sterling per QALY cost-effectiveness thresholds. However, if the quality of life gains could be maintained for a decade, increments relative to microfracture would only have to be 10--20% greater to justify additional treatment costs within the cost-effectiveness band indicated above. Follow-up from the trials so far has only been up to 2 years, with longer term outcomes being uncertain.
CONCLUSIONS: There is insufficient evidence at present to say that ACI is cost-effective compared with microfracture or mosaicplasty. Longer term outcomes are required. Economic modelling using some assumptions about long-term outcomes that seem reasonable suggests that ACI would be cost-effective because it is more likely to produce hyaline cartilage, which is more likely to be durable and to prevent osteoarthritis in the longer term (e.g. 20 years). Further research is needed into earlier methods of predicting long-term results. Basic science research is also needed into factors that influence stem cells to become chondrocytes and to produce high-quality cartilage, as it may be possible to have more patients developing hyaline cartilage after microfracture. Study is also needed into cost-effective methods of rehabilitation and the effect of early mobilisation on cartilage growth.