Computerized TKR Beats Conventional TKR in New RCT Study

Norwegian researchers recently pitted “man versus machine” in the surgical arena.

Their work, “Computer-Assisted Compared with Conventional Total Knee Replacement: A Multicenter Parallel-Group Randomized Controlled Trial,” appears in the August 1, 2018 edition of The Journal of Bone and Joint Surgery.

Co-author Gunnar Petursson, M.D., with The Norwegian Arthroplasty Register, Department of Orthopaedic Surgery, Haukeland University Hospital in Bergen, Norway, told OTW, “This paper is part of a project investigating the need and value of computer navigation in TKR [total knee replacement] surgery.”

“In this study, 190 patients were randomly allocated to undergo either computer-navigated or conventional total knee replacement surgery. One hundred seventy-two patients were available for clinical evaluation at two years and 167 (88%) of those answered all patient reported outcome measures (PROMs), including the Knee Injury and Osteoarthritis Outcome Score (KOOS), Western Ontario & McMaster Universities Osteoarthritis Index (WOMAC), Knee Society Score (KSS), and EuroQol (EQ-5D).”

“Patients and clinical evaluators were blinded to the method of surgery. Surgical outcome was assessed using the OMERACT-OARSI [Outcome Measures in Rheumatology-Osteoarthritis Research Society International] criteria to calculate responder rates divided into high, moderate and non-responders.”

“The computer-navigated group had significantly more high responders at two years, with the number needed to treat (NNT)=8. Comparison of the computer-navigated and conventional surgery groups also indicated that the computer-navigated group had significantly more improvement than the conventional surgery group in mean scores for two KOOS subscales, ‘symptoms’ and ‘sport/rec’, and in the WOMAC subscale for ‘stiffness.’”

“We found more high responders with computer assisted surgery, and the number of patients which needed additional treatment was 8 at two years follow up. We found that 3 out of 11 patient reported outcome measure variables were better with computer assisted surgery compared to conventional surgery after two years and that these differences were both statistically significant and of clinical importance.”

“This randomly controlled trial was designed to compare computer assisted surgery to conventionally operated total knee replacement surgery. A high rate of reporting completeness is essential for the internal validity of the data and one of our strengths is the high follow-up rate of 88% at two years.”

“We have also followed the CONSORT Statement by accounting for all patients lost to follow-up. It is important to acknowledge that randomly controlled trial results do not always represent the average outcome in the average patient. Both patient and observer can be influenced by the study and rare events can be missed.”

“We have tried to minimize observation bias, by blinding both the patient and observer by using sham incisions. We used block randomisation for each participating surgeon with randomly varying block sizes of two and four, to achieve approximately equal numbers of surgeries in both treatments groups for each surgeon and to minimize selection bias. A central randomisation office performed the computer-generated allocation to the trial groups, with concealment by identical sequentially-numbered opaque sealed envelopes. Our patient groups are comparable to patients operated with total knee replacement surgery in Norway and there were no preoperative differences between the groups in terms of patient characteristics.”

“We found that computer assisted surgery total knee arthroplasty patients had significantly better clinical results after two years, using KOOS. WOMAC also showed significantly better clinical results for computer assisted surgery.”

“More patients were pain-free and had better function in the computer assisted surgery group at two years compared to the conventional group. In contrast, the KSS and EQ-5D scores did not show any significant differences.”

“Similar results have been previously described and are thought to be attributable to ceiling effects. The higher proportion of pain-free knees in the computer assisted surgical group is important, both at an individual patient level and from a health system perspective, considering a previous report on the economic expenses involved in investigating and treating painful total knee replacement surgery and the burden to the individual patients.”