Third-Generation Autologous Chondrocyte Implantation Versus Mosaicplasty for Knee Cartilage Injury: 2-Year Randomized Trial Arnaud Clave, 1,2,3 Jean-FranS cois Potel, 4 Elvire Servien, 5 Philippe Neyret, 5 Frederic Dubrana, 1,2 Eric Stindel 1,2,3 1 Universite de Bretagne Occidentale, Faculte de Medecine, 22 avenue Camille Desmoulins, 29200, Brest, France, 2 Service de Chirurgie Orthopedique et Traumatologique de la Cavale Blanche, CHRU Brest, bd Tanguy Prigent, 29200, Brest, France, 3 LaTIM, INSERM UMR 1101, 2 av Foch, 29609, Brest Cedex, France, 4 Clinique Medip^ ole Garonne, 45 rue Gironis, 31036, Toulouse Cedex 1, France, 5 Centre Albert Trillat, service de chirurgie orthopedique, H^ opital de la Croix-Rousse, Hospices civils de Lyon, 103 grande rue de la Croix-Rousse, 69004, Lyon, France Received 23 September 2015; accepted 29 December 2015 Published online 18 January 2016 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.23152 ABSTRACT: Numerous surgical techniques have been developed to treat osteochondral defects of the knee. A study reported encouraging outcomes of third-generation autologous chondrocyte implantation achieved using the solid agarose-alginate scaffold Cartipatch 1 . Whether this scaffold is better than conventional techniques remains unclear. This multicenter randomized controlled trial compared 2-year functional outcomes (IKDC score) after Cartipatch 1 versus mosaicplasty in patients with isolated symptomatic femoral chondral defects (ICRS III and IV) measuring 2.5–7.5 cm 2 . In addition, a histological evaluation based on the O’Driscoll score was performed after 2 years. We needed 76 patients to demonstrate an at least 10-point subjective IKDC score difference with a ¼ 5% and 90% power. During the enrolment period, we were able to include 55 patients, 30 of them were allocated at random to Cartipatch 1 and 25 to mosaicplasty. After 2 years, eight patients had been lost to follow-up, six in the Cartipatch 1 group, and two in the mosaicplasty group. The baseline characteristics of the two groups were not significantly different. The mean IKDC score and score improvement after 2 years were respectively 73.7 20.1 and 31.8 20.8 with Cartipatch 1 and 81.5 16.4 and 44.4 15.2 with mosaicplasty. The 12.6-point absolute difference in favor of mosaicplasty is statistically significant. Twelve adverse events were recorded in the Cartipatch 1 group against six in the mosaicplasty group. After 2 years, functional outcomes were significantly worse after Cartipatch 1 treatment compared to mosaicplasty for isolated focal osteochondral defects of the femur. ß 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 34:658–665, 2016. Keywords: cartilage; knee; autologous chondrocyte implantation; mosaicplasty; scaffold Osteochondral defects of the knee are mainly caused by posttraumatic lesions and osteochondritis disse- cans. 1 They are common in individuals younger than 50 years and adversely impact quality of life. 1 The numerous surgical procedures developed to restore function and to alleviate pain include marrow stimula- tion techniques (microperforations or Pridie drilling to stimulate the bone marrow), osteochondral transfer (autologous mosaicplasty and allogeneic grafting) and cell-based techniques involving either autologous ma- trix-induced chondrogenesis or autologous chondrocyte implantation (ACI). First- and second-generation ACI procedures con- sisted of growing autologous chondrocytes to mono- layers then injecting them into the defect beneath a periosteal flap or collagen membrane, respectively. 2,3 In controlled randomized studies, first-generation ACI showed comparable results to more established techniques 2–4 even though technical challenges such as leaking, heterogeneous cell distribution, patch overgrowth, and phenotypic dedifferentiation were of clinical concern. 5,6 To overcome these problems, a solid agarose-alginate hydrogel scaffold capable of supporting autologous chondrocytes was developed (Cartipatch 1 , Tissue Bank of France, TBF, Lyon, France). This easy-to-handle scaffold provides the cells with a three-dimensional environment, allowing them to develop into an osteo- chondral-like tissue. 7 In 2008, Selmi et al. reported the 2-year outcomes of 17 patients managed using Cartipatch 1 showing significant improvements of the mean International Knee Documentation Committee (IKDC) score, O’Dris- coll histological score and MRI analysis. 7 Those improvements were greatest in patients with cartilage defects larger than 3 cm 2 . We designed a multicenter randomized controlled trial to compare 2-year functional outcomes after Cartipatch 1 or mosaicplasty 1,8,9 treatment of focal femoral chondral defects (ICRS III and IV) measuring 2.5–7.5 cm 2 . The primary outcome was the subjective IKDC score 2 years after surgery. We hypothesized that this score would be higher by at least 10 points in the Cartipatch 1 group. PATIENTS AND METHODS Patients A multicenter randomized controlled trial was conducted from September 2007 to December 2012 in nine French orthopedic surgery centers specialized in osteochondral knee surgery as 4 years duration was planned for this study (enrolment period 2 years, follow-up period two 2 years). The study protocol was approved by the Regional Ethics Committee and French Healthcare Products Safety Agency and was registered on ClinicalTrials.gov (NCT00560664). The trial complied with good medical practice guidelines. Informed consent was obtained from all patients before study inclusion. Table 1 lists the inclusion and exclusion criteria. Figure 1 is the patient flow chart. After validation of the inclusion, an Conflict of interest: None. Grant sponsor: Ministere des Affaires Sociales et de la Sante; Grant number: ; French Government via a “Programme Hospital- ier de Recherche Clinique (PHRC). Correspondence to: Arnaud Clave (T:þ33-298-347-273; F: þ33 298 347 813; E-mail: arnaud.clave@orange.fr) # 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. 658 JOURNAL OF ORTHOPAEDIC RESEARCH APRIL 2016