OBJECTIVES: Information is limited regarding the interim and long-term outcomes associated with knee microfracture surgery. The objective of the study was to describe failure rates 1 year, 3 years, and 5 years after knee microfracture surgery in an administrative claims database.  METHODS: Beneficiaries with knee microfracture surgery between 10/1/2007 and 9/30/2008 (index surgery) and continuous enrollment 24 months prior to the index surgery were identified in the Truven Health Marketscan® Commercial, Medicare Supplemental, and Medicaid databases.  Patients with any knee procedures in the 2 years prior to index surgery were excluded.  Patients were followed for a minimum of 3 years after the index surgery and a subgroup of patients were followed for 5 years.  Procedure failure was defined as receiving a total knee replacement (TKR) or revision, microfracture or autologous chondrocyte implantation subsequent to index surgery.  RESULTS: A total of 3,498 patients (2,948 commercial payer, 351 Medicare and 199 Medicaid) met the study inclusion criteria.  Failure rates increased with increasing years of follow-up: 9% within 1 year, 18% within 3 years, and 32% within 5 years.  Among commercial payer patients, 8% were failures after 1 year, compared to 16% after 3 years and 31% after 5 years.  The highest rates of failure were observed among Medicare beneficiaries, in which 18%, 30%, and 41% of patients received knee procedures within 1 year, 3 years, and 5 years, respectively.  Failure rates defined by TKR after microfracture surgery were 5% within 1 year, 12% within 3 years, and 22% within 5 years.  CONCLUSIONS: Although the gold standard, knee microfracture surgeries have high failure rates as defined by repeat procedures within five years.  There is a need for procedures that provide durable improvements in health outcomes and reduce the need for repeat interventions.