Résumé : Objective. In the recent update of the TRS-398 Code of Practice (CoP), Monte Carlo results were incorporated into the derivation of recommended beam quality correction factors for ionisation chambers (IC) in proton beams. While the underlying Monte Carlo simulations implement detailed models only based on the nominal geometries from manufacturer blueprints, this paper considers the potential geometric deviations in plane-parallel IC arising from manufacturing tolerances that can reach 10%. Approach. A representative model of a plane-parallel IC has been designed using the Monte Carlo code GATE/Geant4, from which beam quality correction factors have been derived. Subsequently, the results of a reference geometry are compared to those of perturbed geometries, in which the parameters are modified according to the tolerances specified in a standard. Main results. The comparison between the reference and perturbed geometries reveals no significant differences, as they show an agreement within one standard deviation for all the cases studied, with relative deviations not exceeding 0.5%. From these results, we estimate the maximum added uncertainty from manufacturing tolerances on Monte Carlo calculated kQ factors to be about 0.7%. Significance. Overall, the current use of nominal dimensions of plane-parallel IC from manufacturer’s blueprints remains consistent for beam quality correction factor calculations via Monte Carlo simulations, which therefore support the latest results recommended by the TRS-398 CoP.