BACKGROUND: Cochlear implantation may be complicated by concurrent injury to the vestibular apparatus, potentially resulting in disabling vertigo and balance problems. Information on vestibular function before implantation as measured by the video head impulse test (VHIT) and cervical vestibular evoked myogenic potentials (cVEMPs) is scarce and literature on long-term effects is non-existing. OBJECTIVE: We aimed to evaluate how vestibular function was affected by cochlear implantation (CI), as measured by VHIT and cVEMPs in the late phase after implantation. METHODS: Retrospective repeated measurement study. Patients: Among the 436 patients elected for CI surgery during 2013 to 2018, 45 patients met the inclusion criteria (CI recipients with a vestibular assessment prior to the first CI and a repeated vestibular assessment after the CI operation). Intervention: VHIT and cVEMPs before and after cochlear implantation. Main outcome measures were vestibular function as evaluated by VHIT gain, saccades and cVEMPs. RESULTS: The mean time between first and second vestibular screening was 19 months. The mean VHIT gain on implanted ears was 0.79 before the operation and showed no change at follow-up (p=0.65). Likewise, the number of abnormal VHIT gain values was equal before and after the operation (p=0.31). Preoperatively, saccades were present on 12 ears (14%) compared with 25 ears (29%) postoperatively (p=0.013) and were associated with significantly lower VHIT gain values. Preoperatively, positive cVEMPs were found in 20 CI ears (49%) and 24 contralateral ears (62%). 10 CI ears lost cVEMP postoperatively compared with 2 ears on the contralateral side (p=0.0047). CONCLUSIONS: Even though VHIT gain has been reported to be affected in the immediate post-implantation period, the findings in this study show that VHIT gain is normalized in the long-term. However, cochlear implantation is associated with the occurrence of VHIT saccades in the long-term and these are associated with lower VHIT gain values. In addition, cVEMP responses are significantly reduced long-term on implanted ears, which agrees with other studies with shorter follow-up.