Robot-assisted knee arthroplasty improves component positioning and alignment, but results are inconclusive on whether it improves clinical scores or reduces complications and revisions: a systematic overview of meta-analyses

Abstract

The purpose of this systematic overview was to identify, synthesise and critically appraise findings of meta-analyses on robot-assisted versus conventional unicompartmental knee arthroplasty (UKA) and total knee arthroplasty (TKA). The hypothesis was that robotic assistance would reduce complications and revision rates, yield better clinical scores, and improve component positioning and alignment. Two researchers independently conducted a literature search using Embase®, MEDLINE®, Web of Science, Allied and Complementary Medicine™ and Cochrane Database of Systematic Reviews on 2 November 2020 for meta-analyses (Level I–IV) on robotic assistance in UKA and/or TKA. Outcomes were tabulated and reported as weighted mean difference (WMD), risk ratio (RR) or weighted odds ratio (WOR), and were considered statistically significant when p\u2009<\u20090.05. A total of ten meta-analyses were identified; four on robot-assisted UKA (n, 1880 robot-assisted vs. 2352 conventional UKA; follow-up, 0 to 60 months), seven on robot-assisted TKA (n, 4567 robot-assisted vs. 5966 conventional TKA; follow-up, 0 to 132 months). Of the meta-analyses on UKA, one found that robotic assistance reduced complication rates (relative risk (RR), 0.62), one found that it improved clinical scores (weighted mean difference (WMD), 19.67), three found that it extended operation times (WMD, 15.7 to 17.1 min), and three found that it improved component positioning and alignment (WMD,\u2009−\u20091.30 to\u2009−\u20093.02 degrees). Of the meta-analyses on TKA, two found that robotic assistance improved clinical scores (WMD, 1.62–1.71), two found that that it extended surgery times (WMD, 21.5–24.26 min), and five found that it improved component positioning and alignment (WMD,\u2009−\u20090.50 to\u2009−\u200910.07 degrees). None of the meta-analyses reported differences in survivorship between robot-assisted versus conventional knee arthroplasty. Robot-assisted knee arthroplasty enabled more accurate component positioning and placement within target zones, but extended operation time considerably. Although robotic assistance improved component positioning, its benefits regarding clinical scores, patient satisfaction and implant survivorship remains to be confirmed. Finally, this overview revealed that six of the ten meta-analyses were of ‘critically low quality’, calling for caution when interpreting results. IV.