Purpose To determine whether additional stem extension for stability is necessary, we performed mid-term follow-up of patients who had been managed with 5-mm metal block augmentation for a tibial defect, where tibial prosthesis was fixed using bone cement without stem extension. Also, we evaluated clinical and radiologic results including survival rate of patients without stem extension. Methods We retrospectively analyzed patients with tibial bone defect, had undergone primary total knee arthroplasty, and had been treated with 5-mm metal block augmentation without stem extension between March 2003 and September 2013. Among 74 patients (80 cases), 47 patients (52 cases) were followed up for at least 5 years. Results Mean flexion contracture improved from 8.8° (0–40°) preoperatively to 0.4° (−5° to 15°) at final follow-up (P < 0.01), but there was no significant change in the mean angle of great flexion: 124.6° (75–150°) preoperatively and 126.2° (90–145°) at final follow-up (P = 0.488). Mean range of motion improved from 115.8° (35–150°) preoperatively to 125.5° (90–145°) at final follow-up (P < 0.01). Mean knee score improved from 38.7 points (0–66 points) preoperatively to 93.2 points (79–100 points) at final follow-up (P < 0.01), and mean functional score also improved from 50.4 points (10–70 points) preoperatively to 81.8 points (15–100 points) at final follow-up (P < 0.01). The mean postoperative Western Ontario and McMaster University osteoarthritis score was 19.5 points (0–66.0 points). The mean femorotibial angle was corrected from 9.0° varus (23.0° varus–6.3° valgus) preoperatively to 5.5° valgus (2.2° varus–11.1° valgus) at final follow-up (P < 0.01). There was no change in the mean β-angle, which was 90.7° (87.2–94.9°) immediately postoperative and 90.8° (87.2–94.9°) at final follow-up (P = 0.748) and in the mean δ-angle, which was 86.2° (81.3–90.0°) immediately postoperative and 87.2° (83.1–96.5°) at final follow-up (P = 0.272). Radiolucent lines (RLL) were observed in ten cases (26.3%), and the mean RLL scores at final follow-up were 0.34 points (0–3 points) in the anteroposterior view and 0.42 points (0–6 points) in the lateral view. Scores for the RLL were ≤ 4 points in 36 cases, 5–9 points in two cases. Revision surgery due to aseptic loosening (three cases) is rarely required, and the Kaplan–Meier survival rate at 10 postoperative years was 96.4% Conclusion When performing 5-mm metal block augmentation for a proximal tibial defect, no additional tibial stem extension can be a good surgical option for the stability of tibial prosthetic construct and mid-term clinical and radiologic results. Level of evidence IV.