BACKGROUND: The introduction of the Musculoskeletal Infection Society (MSIS) criteria for periprosthetic joint infection (PJI) in 2011 resulted in improvements in diagnostic confidence and research collaboration. The emergence of new diagnostic tests and the lessons we have learned from the past 7 years using the MSIS definition, prompted us to develop an evidence-based and validated updated version of the criteria. METHODS: This multi-institutional study of patients undergoing revision total joint arthroplasty was conducted at 3 academic centers. For the development of the new diagnostic criteria, PJI and aseptic patient cohorts were stringently defined: PJI cases were defined using only major criteria from the MSIS definition (n = 684) and aseptic cases underwent one-stage revision for a noninfective indication and did not fail within 2 years (n = 820). Serum C-reactive protein (CRP), D-dimer, erythrocyte sedimentation rate were investigated, as well as synovial white blood cell count, polymorphonuclear percentage, leukocyte esterase, alpha-defensin, and synovial CRP. Intraoperative findings included frozen section, presence of purulence, and isolation of a pathogen by culture. A stepwise approach using random forest analysis and multivariate regression was used to generate relative weights for each diagnostic marker. Preoperative and intraoperative definitions were created based on beta coefficients. The new definition was then validated on an external cohort of 222 patients with PJI who subsequently failed with reinfection and 200 aseptic patients. The performance of the new criteria was compared to the established MSIS and the prior International Consensus Meeting definitions. RESULTS: Two positive cultures or the presence of a sinus tract were considered as major criteria and diagnostic of PJI. The calculated weights of an elevated serum CRP (>1 mg/dL), D-dimer (>860 ng/mL), and erythrocyte sedimentation rate (>30 mm/h) were 2, 2, and 1 points, respectively. Furthermore, elevated synovial fluid white blood cell count (>3000 cells/µL), alpha-defensin (signal-to-cutoff ratio >1), leukocyte esterase (++), polymorphonuclear percentage (>80%), and synovial CRP (>6.9 mg/L) received 3, 3, 3, 2, and 1 points, respectively. Patients with an aggregate score of greater than or equal to 6 were considered infected, while a score between 2 and 5 required the inclusion of intraoperative findings for confirming or refuting the diagnosis. Intraoperative findings of positive histology, purulence, and single positive culture were assigned 3, 3, and 2 points, respectively. Combined with the preoperative score, a total of greater than or equal to 6 was considered infected, a score between 4 and 5 was inconclusive, and a score of 3 or less was not infected. The new criteria demonstrated a higher sensitivity of 97.7% compared to the MSIS (79.3%) and International Consensus Meeting definition (86.9%), with a similar specificity of 99.5%. CONCLUSION: This study offers an evidence-based definition for diagnosing hip and knee PJI, which has shown excellent performance on formal external validation.