Comparison of total patellectomy and osteosynthesis with tension band wiring in patients with highly comminuted patella fractures: a 10–20-year follow-up study

BACKGROUND: The purpose of this study was to evaluate and compare the long-term clinical outcomes between the total patellectomy and osteosynthesis with tension band wiring in patients with highly comminuted patella fractures. METHODS: Between January 1987 and December 2003, this retrospective study included a total of 35 patients (mean age, 51.4±16.8 years) with a minimum of 10 years follow-up period, comprising 29 males and 6 females, who were divided into the total patellectomy group (17 patients) or the open reduction and internal fixation (ORIF) group (18 patients) in the Third Affiliated Hospital of Hebei Medical University. We retrospectively collected patient demographics and data on the type of trauma, fracture type, and postoperative complications. Clinical outcomes including knee range of motion (ROM), 36-Item Short-Form Health Survey (SF-36) score [including physical component score (PCS) and mental component score (MCS)], Knee Injury and Osteoarthritis Outcome Score (KOOS), and Kujala score were evaluated and compared between the two groups. Biodex System dynamometer was used to quantitatively evaluate quadriceps femoris muscle power following measurement of peak torque. RESULTS: The mean follow-up periods of the total patellectomy group and the ORIF group were 17.2±5.6 and 16.8±4.9 years, respectively. There were no significant differences between the two groups of patient demographics in terms of the number of patients, age, sex, injury side, time to surgery, type of trauma, and fracture classification (p>0.05). Total patellectomy was comparable to osteosynthesis with tension band wiring in terms of ROM [injured knee: 120.4±3.1° vs 118.6±3.3°; uninjured knee: 126.5±2.8° vs 127.3±1.7°; both p>0.05], peak torque [Injured knee: 96.2±2.3 vs 97.3±2.6, N· m; Uninjured knee: 107.6±2.1 vs 106.3±1.8, N· m; both p>0.05], SF-36 score [PCS: 64.1±18.0 vs 61.5±17.9; MCS: 55.1±13.8 vs 54.3±12.4; both p>0.05], KOOS score [76.3±12.1 vs 73.4±11.7; p>0.05], and Kujala score [67.6±11.8 vs 70.8±11.9; p>0.05] at the final follow-up, while total patellectomy had significantly shorter operation time than ORIF group (47.5±12.1 vs 68.8±22.3, min, p<0.05). In the total patellectomy group, complications occurred in 6 of 17 cases (35.3%), and all occurred with calcification. In the ORIF group, complications occurred in 12 of 18 cases (66.7%), including 2 cases of infection (11.1%), 1 case of non-union (5.6%), 2 cases of implant failure (11.1%), 2 cases of soft tissue irritation (11.1%), and 5 cases of patellofemoral arthritis (27.8%). CONCLUSIONS: Total patellectomy technique was a safe and reliable alternative treatment for treating patients with highly comminuted patella fractures when anatomically reduction and rigid fixation were difficult, although it caused relatively higher rates of calcification.