Microfracture (MF) has been the main intervention in symptomatic articular cartilage knee defects. Autologous chondrocyte implantation (ACI) has looked promising, but was not recommended by the UK National Institute for Health and Care Excellence (NICE) in 2015 due to the short-term follow-up data from trials.

Most long-term data comes from observational studies. We provided new unpublished analyses to NICE based on survival data of these studies, with appropriate caveats. They included: a large ACI study by Nawaz with useful subgroup data by osteoarthritis Kellgren-Lawrence stage and previous repair attempts; a very large MF study by Layton, and a small RCT by Knutsen indicating MF was as ‘good’ as ACI. A Markov model explored the cost-effectiveness of ACI vs. MF. Different scenarios were explored: ACI or MF as a first procedure, followed by ACI or MF in those needing a second repair. A NHS England perspective was adopted. Health outcomes were expressed as quality-adjusted life-years (QALYs).

The revised base-case analysis, used a list price of £16,000 (EUR 17,380 in 2013 prices) for cells, used ACI failure data from Nawaz with no previous procedures for ACI, and pooled MF failure data from two studies-Saris and Knutsen. ACI was more expensive but provided more QALYs. The incremental cost-effectiveness ratio comparing ACI then MF with MF then ACI was £8,000 (EUR 8,690) per QALY. Various sensitivity analyses were conducted assuming a threshold of £20,000 (EUR 21,730) per QALY: previous repair attempts reduced success of ACI (£22,000 (EUR 23,900) per QALY); reducing cell costs, ACI improved its cost-effectiveness; and limiting intervention to patients with higher Kellgren-Lawrence score did not appear cost-effectiveness.

The final NICE guidance published in October 2017 approved the use of ACI for patients who had no previous knee repairs, for people with minimal osteoarthritic damage to the knee, and for people with articular defects of over 2cm2.