Background: Despite the growing evidence in the literature there is still a lack of consensus regarding the use of minimally invasive surgical technique (MIS) in total knee arthroplasty (TKA). Methods: A prospective, randomized, international multicentre trial including 69 patients was performed to compare computer-assisted TKA (CAS-TKA) using either mini-midvastus (MIS group) or standard medial parapatellar approach (conventional group). Results: Patients in the MIS group (3.97 ± 2.16) had significant more pain at 2 weeks than patients in the conventional group (2.77 ± 1.43) p = 0.003. There was no significant difference in any of the other primary outcome parameters. Surgery time was significantly longer (p < 0.001) and there were significantly higher blood loss (p = 0.002) in the MIS group as compared to the conventional group. The difference of the mean mechanical leg alignment between the groups was not statistically significant (-0.43° (95 % CI -1.50 - 0.64); p = 0.43). Conclusion: There was no advantage of the MIS approach compared to a conventional approach CAS-TKA in any of the primary outcome measurements assessed, however the MIS approach was associated with longer surgical time and greater blood loss. MIS-TKA in combination with computer navigation is safe in terms of implant positioning. Trial registration number: ClinicalTrials.gov NCT02625311 8 December 2015