Second Generation Autologous Chondrocyte Implantation in the Patella: Improved Pain Relief and Functional Outcomes among U.S. Army Servicemembers

Objectives: To evaluate the functional outcomes of autologous chondrocyte implantation (ACI) for high-grade patellofemoral chondral defects in a predominately young, high-demand military demographic. Methods: We conducted a retrospective review of all United States military and Tricare beneficiaries undergoing ACI for Outerbridge grade III or IV patellofemoral chondral defects at a two high-volume military medical centers between 2007 and 2014. Second generation surgical technique was performed using autologous cultured chondrocytes (Vericel, Inc; Boston, MA) secured with a type I/III collagen bilayer membrane. Inclusion criteria required minimum two-year surveillance, although patients with knee-related medical discharge were also included to limit non-responder bias. Exclusion criteria were applied to individuals with predominately condylar involvement, periosteal patch coverage, and/or procedure miscoding Demographic information, surgical characteristics, and clinical outcomes were extracted from the electronic medical record. Averages were calculated with standard deviations (SD). Univariate analysis was used to determine significant independent predictors of surgical failure, which was defined as revision chondral procedure or conversion to arthroplasty. Relative risk was quantified with use of odds ratios (OR) with 95% confidence intervals (95% CI). Results: A total 57 patients with 58 knees were identified at average 2.13 year follow-up within the military health care system. The median age was 33.8 years (range, 20.3 to 49.9). Males comprised 77% of patients (n=44), and 93% (n=53) were active duty military status. Mean patient body mass index (BMI) was 30.3 kg/m2, tobacco use was documented in at least 19.3% (n=11), and only one patient had failed prior marrow stimulation procedure. The average lesion size for ACI was 502 mm2 (SD 318; range, 121-1600 mm2) and most defects were based on patella (91.3%). The average tibial tubercle-trochlear groove (TT-TG) distance was 13.9 mm (SD 3.5), and nearly all knees underwent concomitant anteromedializing or offloading tibial tubercle osteotomy (n=56; 97%). Additionally, only three patients required proximal realignment procedure. Only 1 patient (1.7%) required revision chondral procedure, and there were no cases of graft site hypertrophy or conversion to knee arthroplasty. With a “high” or “severe” average patient-reported pain level (86%), the average postoperative visual analog scale (VAS) was 2.3 (SD 2.2;