Computer assisted surgical techniques in total knee arthroplasty have demonstrated increased accuracy of alignment and decreased risk of outliers. Some studies have also demonstrated improved early functional results and pain scores in comparison to traditional surgical methods. Studies have also shown a slightly increased surgical time for computer assisted surgery. A learning curve for computer assisted surgery is recognized, and there may be different outcomes for cases performed initially during the learning phase. This study reports on a single surgeon’s experience with the initial 261 computer assisted total knee arthroplasties.

A single experienced, fellowship trained surgeon performed computer assisted total knee arthroplasty utilizing either the BrainLab or Ci intraoperative navigation system and either the LCS Complete Mobile Bearing Knee System (DePuy) or Sigma PFC Rotating Platform (DePuy). Preoperative and postoperative data was recorded prospectively (DePuy Captureware) of the initial 261 consecutive cases at minimum of one year follow-up. SAS 9.1 was used to perform univariate and multivariate analyses of four groups of patients: patients 1–77, patients 78–135, patients 136–211 and patients 212–261. Multivariate analyses were performed to control for body mass index, age, sex, implant type, pre-operative range of motion, preoperative function and preoperative pain scores.

Multivariate analysis of these four groups demonstrated that there was no statistically significant difference in the improvement of postoperative function (p=0.29) and pain scores (p=0.28) among the patients in the four groups at minimum one year follow-up. There was a statistically significant difference in improvement of postoperative extension (p=0.0022) and flexion (p=0.0139) scores with subsequent surgeries, however the range of improvement for the groups was not clinically significant (extension ranging from 1.97 to 5.92 degrees gained in the four groups, and flexion loss of 0.67 degrees to gain of 6.18 degrees in the four groups). The number of patients requiring a hospitalization greater than two days decreased with each subsequent group which was clinically significant (p=0.021, p=0.001, p< 0.0001 for the second, third and fourth groups, respectively).

For an experienced reconstructive surgeon incorporating computer assisted surgery into his total knee arthroplasty practice, there is no significant learning curve in regards to intermediate term outcomes. Patients undergoing computer assisted total knee arthroplasty have similar intermediate outcomes whether performed earlier in that surgeon’s experience or later. Patients did initially have shorter hospitalization stays in subsequent groups. However, at an intermediate follow-up period of one year, there is no significant difference in patients’ postoperative improvement in function, pain score, knee flexion and knee extension.