3.1. Basic conditions of surgery.
There was no significant difference between the two groups in operation time, hospitalization time, blood loss and blood transfusion. (P > 0.05). The comparison of intraoperative data between the two groups was shown in Table 2.
3.2. Clinical Results
120 patients were followed up with an average of 74.8 months(12-114months). At the last follow-up, the Harris hip score in the cylindrical group increased from 41.1±6.1 to 84.3±4.4 (P < 0.05), the tapered group increased from 40.1±6.6 to 85.5±3.8 (P < 0.05), the VAS score in the cylindrical group decreased from 7.6±1.6 to 2.0±0.4 (P < 0.05), and the tapered group decreased from7.5±1.2 to 1.8±0.2 (P < 0.05). At the last follow-up, there was no significant difference in Harris score (P > 0.05), but VAS score of cylindrical group was higher than that of tapered group. (P=0.047). (Figure. 1) There was no significant difference between cylindrical group (87.3%) and tapered group (90.1%) in the overall satisfaction of the recent follow-up results. (Table 3)
3.3. Radiographic Results
At the last follow-up, prosthesis subsidence in cylindrical group was 0-15 mm, with an average of (4.17 ± 4.20) mm, and in tapered group was 0-8 mm, with an average of (2.17 ± 1.49) mm (Figure. 2). There was significantly difference between cylindrical group and tapered group in prosthesis subsidence. All prosthesis subsidence stopped within 1 year after operation.
In the cylindrical group, 45 (84.1%) were fixed by bone growth, 7 (13.0%) were fixed by fibers and 1 (1.8%) was unstable. In the tapered group, 64 (98.4%) were fixed by bone growth, 1 (1.5%) was fixed by fibers and 1 (1.6%) was unstable. Compared with the tapered group, the failure rate of osseointegration (fibrous or unstable) in the cylindrical group was significantly higher (P < 0.05).
In the imaging changes of proximal femoral host bone, the proportion of bone repair type in the tapered group (39.4%) was significantly higher than that in the cylindrical group (7.4%) and the proportion of bone loss type in the tapered group (13.6%) was significantly lower than that in the cylindrical group (43.6%).（Table 4）
In the cylindrical group, 21 hips (38.9%) had stress-shielded bone resorption of degree I and degree II in the greater trochanter, which was manifested by femoral moment atrophy or cortical bone mineral density decreased from sharply blunt margin to trochanter level. However, in the tapered group, 9 hips (13.6%) had stress-shielded bone resorption of degree I and degree II in the greater trochanter.
The 8-year cumulative survival rate was defined as the end point of any reoperation due to septic or aseptic complications. The 8-year cumulative survivorship of the cylindrical stem was 94.43%, (95% confidence interval [CI], 86.13%-97.82%) and the 8-year cumulative survivorship of the tapered stem was 96.69% (95% confidence interval, 91.39%-98.75%). There was no significant difference between the two groups (Figure. 3).
3.5. Postoperative complications
In the cylindrical group, 8 cases (14.8%, Figure.4) had intraoperative fractures, while 3 cases (4.5%, Figure.5) in the tapered group. There was significantly difference between the two groups (P＜0.05). In the cylindrical group, 5 cases were femoral trochanteric fractures (with steel wire binding), 3 cases occurred in the femoral shaft (with steel wire binding), all fractures were healed. In the tapered group, 3 cases had femoral shaft fracture (with steel wire binding), and the fracture healed after operation.
There were 2 cases (3.7%) of periprosthetic fracture in cylindrical group and 1 case (1.5%) of periprosthetic fracture in tapered group (P > 0.05). Periprosthetic fractures were treated with open reduction and internal fixation.
There were 2 cases (3.7%) of hip dislocation in cylindrical group after operation. One patient underwent cup exchange with recurrent instability, while the other 1 was treated with closed reduction, regaining hip stability. Two cases (3.0%) of hip dislocation occurred in the tapered group (P > 0.05). One case of frequent dislocation after closed reduction was treated with replacement of lining and femoral head size.
One hip in cylindrical group was revised due to osteolysis around the cup at 9 years after surgery (Figure.6). The last follow-up radiography showed osteolysis around the cup and Cup revision arthroplasty carried out for an osteolytic lesion.
Postoperative thigh pain: 3 of 66 hips in tapered group had mild thigh pain (VAS score l-3), but most of the symptoms disappeared one year after operation; 7 of 54 hips in cylindrical group had mild thigh pain (VAS score 1-4), of which 2 developed persistent thigh pain.