PURPOSE: The purpose of this study was to assess the impact of post-operative coronal plane alignment of the knee (CPAK) class change on functional outcomes and determine the rate and type of CPAK class change after image-based robotic arm-assisted (RA)-total knee arthroplasty (TKA) performed with functional alignment (FA) at two different centres. METHODS: The present retrospective, observational, multicentre study included 201 patients treated with RA-TKA between October 2020 and April 2022 at two different centres. The radiographic CPAK classification was adapted using CT images to achieve pre- and post-operative knee categorization into CPAK classes. At a minimum of 1 year post-operatively, patients were administered the Forgotten Joint Score-12 (FJS-12) and surveyed about their post-operative satisfaction level using a 5-level Likert scale (5-LLS). RESULTS: The most common preoperative overall CPAK classes were: Types II, I, III, IV and V. Implant positioning after RA-TKA with FA within the alignment boundaries, determined distribution in the CPAK classification, predominantly maintaining classes I, II, IV, and V. No statistically significant FJS-12 differences were detected between subjects who maintained and changed their preoperative CPAK class. The mean 5-LLS for satisfaction in patients where the preoperative CPAK class was maintained intraoperatively was 4.4 ± 1.1 (range = 1-5), while subjects having the CPAK class changed had a mean 5-LLS of 4.4 ± SD 1.0 (range 1-5). CONCLUSION: In the setting of image-based RA-TKA with FA, CPAK can be changed within a 'functional safe-zone', without compromising functional outcomes. Good functional outcomes are the result of a stable and balanced knee with soft-tissue preservation, regardless of the maintenance of the preoperative CPAK class. LEVEL OF EVIDENCE: Level III.