Because we are performing TKAs on heavier, younger patients, greater stress is being put on the implants and is increasing the importance of implantation accuracy. We performed a prospective randomized study to compare the radiographic results and the 3-month clinical outcomes in 100 patients who had TKAs using an imageless navigation system with 100 patients treated using conventional implantation instruments. We measured component alignment by standard radiographs. Clinical outcomes were based on the Insall score, anterior knee pain, feeling of instability, and the step test. The mechanical axis of the limb was within 3 degrees varus/valgus in 92% of the patients who had navigated procedures versus 76% of patients who had conventional surgery. The tibial slope showed a rate of inaccuracy of 3 degrees or less for 98% of the patients in the navigated TKA group versus 80% of the patients in the conventional group. The surgical time was longer for navigated TKA than for the conventional procedure (88 +/- 16 versus 68 +/- 18 minutes, respectively). Clinical outcomes and postoperative blood loss were similar in both groups. The navigation system increased implantation accuracy but did not prevent outliers and did not solve the problems associated with identifying and obtaining accurate component rotation.