Introduction This study aimed to describe an anatomic medial knee reconstruction technique for combined anterior cruciate ligament (ACL) and grade III medial collateral ligament (MCL) injuries and to assess knee function and stability restoration in patients who underwent primary MCL reconstruction compared with primary repair. Methods A total of 105 patients who had undergone anatomic ACL reconstruction between 2008 and 2017 were enrolled in this retrospective study and divided into two groups according to concomitant MCL ruptures. Group A included patients with isolated ACL ruptures without MCL injuries. Group B included patients with both ACL and MCL injuries, and it was subdivided into three groups according to the severity of the MCL injury and treatment modality: B-1, grade I or II MCL injury treated conservatively; B-2: grade III MCL injury treated by primary MCL repair; and B-3: grade III MCL injury treated by primary reconstruction. Knee stability was measured via Telos valgus radiography at 6-month and 2-year postoperative. The Lysholm score, Tegner activity level, Likert scales (satisfaction), and return to previous sports were evaluated at 2-year postoperative. Results At 6-month postoperative, there was no significant difference in medial laxity between the B-2 and B-3 groups. However, at 2-year postoperative, medial laxity were significantly higher both at 30° of flexion (5.2° versus 2.2°, p = 0.020) and at full extension (3.4° versus 1.1°, p < 0.001) in patients in B-2 group compared to those in B-3 group. There were no statistically significant differences between the two groups with respect to Lysholm scores, Tegner activity levels, Likert scales (satisfaction), and returning to previous sports at the 2-year follow-up. Conclusion Primary medial reconstruction combined with severely injured MCL in ACL reconstruction may decrease residual medial laxity more than primary repair. Level of evidence Retrospective observational study, IV.