Handheld, accelerometer-based navigation versus conventional instrumentation in total knee arthroplasty: a meta-analysis

Background Precise and accurate alignments in total knee arthroplasty are important predictors for survivorship and functional outcomes. We aim to compare accelerometer-based navigation (ABN) to conventional instrumentation (CONV), patient-specific instrumentation (PSI) and computer-assisted surgery (CAS) in published literature. Methods A systematic search of publications from databases (MEDLINE, EMBASE and Cochrane) was obtained from inception to 15 August 2018. A random-effects meta-analysis was used to pool odd ratios for outliers greater than 3 degrees for the hip-knee-ankle, coronal and sagittal femoral and tibial angles (CFA, CTA, SFA, STA). Secondary outcomes included procedural characteristics and functional outcomes. Results Thirteen studies, involving 1566 patients, met inclusion that compared ABN (50.2%) to CONV (49.8%) and five comparing ABN to CAS/PSI. The pooled odds ratios for percent outliers of greater than 3 degrees from the mechanical axis for the hip-knee-ankle (relative risk 0.58, P = <0.05) and CFA (relative risk 0.42, P = 0.02) was significantly lower for ABN compared to CONV. The pooled odds ratios for CTA, SFA and STA were not significantly different. No differences were identified in comparison to PSI/CAS. There was no statistically significant difference in procedural characteristics and functional outcomes. Conclusions The use ABN in total knee arthroplasty is a successful method of increased precision and accuracy for the restoration of the mechanical axis. In addition, there is no significant compromise in procedural or functional outcomes.