Infection after total knee arthroplasty (TKA) remains a difficult complication to treat. The risk of infection ranges from 0.5% to 2% for primary TKAs and 2% to 4% for revision TKAs. Several demographic studies indicate that more infections are occurring after these procedures, and infection is one of the most common reasons for TKA failure. Prevention remains the key to minimizing the risk of infection; however, little evidence-based literature exists to establish the optimal approach. Every patient with a painful TKA should be suspected of having an infection until proven otherwise. An algorithmic approach to these patients should include standard laboratory screening tests to rule out infection. Synovial fluid aspiration remains the best test for diagnosing infection. Synovial fluid white blood cell counts greater than 1,700 cells/µL and a differential greater than 69% polymorphonuclear cells should raise a high index of suspicion for infection. Several options are available to treat deep periprosthetic infection. The timing of the infection as it relates to surgery and the onset of symptoms are critical in determining treatment success. Prosthetic retention is indicated only in patients with an acute onset of infection, but its limited success reported in recent literature brings into question its role in infected TKAs. A two-stage exchange arthroplasty remains the gold standard for treatment of infection following TKA.