Antibiotic laden spacer blocks frequently are used to treat an infected total knee arthroplasty. Static spacer blocks make exposure at reimplantation difficult secondary to quadriceps shortening. Unexpected bone loss attributable to migration of the spacer block also has been reported. To avoid these problems, a temporary articulating molded implant made of antibiotic cement was used in a consecutive series. The authors sought to determine whether its use would affect the reinfection rate, improve functional results, or prevent bone loss compared with static spacers. Twenty-five patients were treated with static nonarticulating spacers. Since 1996, 30 patients have been treated with tobramycin-laden articulating spacers. The knee arthroplasties in three patients treated with a static spacer became reinfected (12%). The knee arthroplasty in one patient with an articulating spacer became reinfected (7%). Fifteen of the 25 patients with static spacers had unexpected bone loss between stages. No appreciable bone loss could be measured in the patients who received articulating spacers. The average Hospital for Special Surgery score was 83 points in the patients with static spacers and 84 points for the patients with articulating spacers. Range of motion at final followup averaged 98 degrees in the patients who received static spacers and 105 degrees in the patients who received articulating spacers. Articulating spacers seem to facilitate reimplantation of infected total knee arthroplasty without additional risk of infection. Unexpected bone loss is no longer a concern with this two-stage technique. Articulating spacers offered no functional advantage over static spacers in this study group.