Objective To compare the clinical efficacy of Philos plate and Multiloc intramedullary nail for 3-or 4-part proximal humeral fractures in the middle-aged and elderly patients.Methods A retrospective case control study was conducted to analyze the clinical data of 52 middle-aged and elderly patients with 3-or 4-part proximal humeral fractures admitted to Sichuan Orthopedic Hospital from January 2014 to January 2016.The patients were divided into Philos plate group (27 patients) and Multilloc intramedullary nail group (25 patients) according to different treatment methods.In the Philos plate group,there were 11 males and 16 females,aged (59.3 ± 4.5) years.Based on the Neer classification,there were 16 patients with 3-part fracture,11 patients with 4-part fracture including two patients with 4-part fracture dislocation.There were eight patients with varus fracture and 19 patients with valgus fracture.In the Multiloc intramedullary nail group,there were 10 males and 15 females,aged (62.2 ± 7.4) years.Based on the Neer classification,there were 18 patients with 3-part fracture,seven patients with 4-part fracture including one with 3-part fracture dislocation and one with 4-part fracture dislocation.There were 12 patients with varus fracture and 13 with valgus fracture.The operation time,intraoperative bleeding volume,fracture healing time,neck-shaft angle changes and complications of humeral head ischemic necrosis were compared between the two groups.At the last follow-up,the motion range of affected shoulder joint,American Shoulder and Elbow Surgeons (ASES) scale,Constant-Murley score and visual analogue score (VAS) were compared between the two groups.Results There was no significant difference in the operation time between the two groups (P ＞ 0.05).The intraoperative bleeding volumes were 170-350 ml [(260.1 ± 110.3) ml] in the Philos plate group and 70-250 ml [(172.2 ± 100.3)ml] in the Multiloc intramedullary nail group,with statistically significant difference (P ＜ 0.05).All incisions were healed by first intention,and no infection was found.The patients in the Philos plate group were followed up for 12-36 months [(17.2 ±6.5)months],and patients in the Multiloc intramedullary nail group for 12-36 months [(14.5 ± 4.7) months] (P ＞ 0.05).All fractures were healed,with the healing time for 2.5-4 months [(3.5 ± 0.5)months] in the Philos plate group and for 2-3.3 months [(3.0 ± 0.5) months] in the Multiloc intramedullary nail group.The neck-shaft angle was lost to some degree in both groups at the last follow-up compared with the first day after operation,but the differences were not statistically significant (P ＞ 0.05).The Philos plate group had a higher incidence rate of complication [22％ (6/27)] than the Multiloc intramedullary nail group [12％ (3/25)] (P ＜ 0.05).Three patients in the Philos plate group had partial ischemic necrosis of the humeral head,while none was found in the Multiloc intramedullary nail group.There were no significant differences in active shoulder joint ante-flexion and uplift,lateral intorsion and extorsion,ASES score,Constant-Murley score and VAS between the two groups (P ＞ 0.05).Conclusions For 3-or 4-part proximal humeral fractures in middle-aged and elderly patients,both Philos plate and Multiloc nail can obtain stable and reliable fixation and achieve satisfactory results of shoulder joint function.However,the Philos plate fixation has more intraoperative bleeding,longer fracture healing time,and higher incidence rate of humeral head ischemic necrosis than the intramedullary nail.