Abstract
A matched comparison was made between femoral neck-preserving short, tapered stems (n = 50) and conventional length femoral stems (n = 50) in cementless total hip arthroplasty between January 2008 and January 2012. Patients were matched for age, sex, body mass index, height, surgical approach, and surgeon. In group A, mean preoperative HHS and WOMAC scores of 55.0 and 53.0, respectively, improved to mean postoperative scores of 98.6 and 3.3, respectively, at an average follow–up of 37.2 months. In group B, mean preoperative HHS and WOMAC scores of 53.0 and 49.5, respectively, improved to mean postoperative scores 97.8 and 4.4, respectively, at an average follow–up of 35.3 months. In addition, no significant differences in mean postoperative HHS (P = 0.168) or WOMAC scores (P = 0.158) were observed between the two groups according to the independent sample t-test. Table 1 shows two stems (4%) located in valgus (greater than 5° from neutral). The mean preoperative and postoperative HHS and WOMAC scores were similar to those of stems neutrally located in group A. All group A stems displayed bone bridging and endosteal spot welds distributed in Gruen zones 2 and 6 as evidence of bony ingrowth with no radiolucencies (Fig. 1). Two patients in group B had the slight decrease in bone density, mostly in Gruen zone 1 and 7. No radiographic evidence of osteolytic lesions, cortical hypertrophy, or acetabular fractures was detected in either group. Furthermore, no patient required revision surgery for aseptic loosening. The chi–square (Fisher's exact) test showed no significant difference between the two groups with respect to patient complications (Table 2). One patient in group A with a CCD angle of 135° had subsidence (greater than 2 mm, P =0.313) that displayed bony ingrowth with no further progression of subsidence at final follow-up. An intraoperative minor femur neck fracture (P =1.00) occurred in two patients (one in group A and one in group B). These patients were treated by cerclage wires without further incident. Three patients (two in group A: valgus and a CCD angle 135°, and one in group B: varus) had malalignment (P = 0.554) that was not associated with loosening. One patient in group A with a CCD angle of 135° had greater than 1 cm leg length discrepancy (shortening, P = 0.313). Two patients in group B had thigh pain (P = 0.151) that disappeared after a few months. Three patients (one in group A and two in group B) had heterotopic ossifications (P = 0.554) that were categorized as Brooker class I in two patients and class II in one patient without limiting their activities. In conclusion, no significant differences in the clinical and radiographic outcomes as well as component-specific complications were observed between the two groups, showing satisfactory performance at the 5-year follow-up. Group A had a higher incidence of malalignment and subsidence and a lower incidence of thigh pain and proximal bone resorption than group B.
