The investigation was performed to determine the effect of CA and offset on dislocation after THA. It was demonstrated that both Widmer’s and Yoshimine’s CA values showed a significant difference in the mean values between the dislocation and non-dislocation groups, supporting the effectiveness of CA. However, many researchers have reported that some degree of difference between preoperative target values and postoperative measured values of implant placement angle is inevitable in THA [18–20]. Considering these previous reports, it is difficult to place the cup precisely at the generally recommended combined anteversion value. Moreover, Widmer et al assume that the radiographic inclination of the cup should be in the range of 40°–45° for CA; however, it is also difficult to place the implant within this range. In this study, only 155 implants were placed within this range. Therefore, based on the aforementioned reports and considering possible CA errors in surgery, we defined good CA as a cup radiographic inclination within the range of 35–55° and within ± 10% of the recommended CA value. The results of the present study demonstrated that dislocation was significantly lower in the good CA group and the cutoff values of CA for postoperative dislocation were 37°±3.3° and 90.8°±10.3° for Widmer and Yoshimine, respectively. This range is almost the same as our definition of good CA, and it covers the pre- and postoperative differences in implant placement angles in previous reports. Significantly fewer dislocations occurred when implants were placed within this range, which is useful as an indicator of dislocation prevention. However, as cases of dislocation occurred even when implants were placed within this range, it would be difficult to prevent dislocation after THA using CA alone as an indicator of implant placement.
CA has been reported to be effective only in the prevention of implant impingement, but not in bony impingement . Previous studies have shown the importance of not only implant impingement but also bony impingement and soft tissue tension in the prevention of dislocation after THA [5, 21, 22]. Studies on bony impingement have shown that increased acetabular offset provides increased range of motion for flexion and internal rotation by decreasing the effect of impingement of the greater trochanter on the anterior acetabulum [23–25]. Increased femoral offset improves hip abductor strength by lengthening the functional lever arm, reducing impingement, and increasing postoperative joint stability [22, 26] and hip range of motion [27, 28]. Thus, dislocation after THA is influenced not only by CA but also by acetabular offset and femoral offset, and all of these indicators should be considered during surgery. In this study, the results revealed no significant difference in the dislocation rate between the groups with and without increased acetabular and femoral offsets but a significant difference in the dislocation rate between the groups with and without increased total offset. In THA of patients with osteoarthritis, medialization by reaming is often necessary to place the acetabular component in an anatomically normal acetabulum . The femoral offset should be increased to compensate for the decrease in acetabular offset due to the medialization. Thus, both offsets need to be evaluated , and avoiding a decrease in the total offset is important , which supports our results. In this study, the distance from the pubic symphysis to the femoral shaft was measured as the total offset; therefore, both the femoral and acetabular offsets were used in the evaluation. In both Widmer’s and Yoshimine’s CA, there were dislocations under the condition of good CA alone but not in cases with good CA and increased total offset. This suggests that dislocation can be further avoided by considering not only CA to prevent implant impingement but also the increase in total offset on the operated side to avoid soft tissue hypotension and bony impingement.
Although a previous report showed no difference in the risk of dislocation between the increased and non-increased global offset groups , all previous reports on offset for dislocation after THA have compared the offset between the healthy and the operative sides. Krishnan et al reported that the mean difference between the right and left sides of the normal hip offset was 2.54 ± 2.31 mm and that hip offsets are not always symmetrical even in normal hips . Additionally, determining the appropriate center of rotation of the hip joint in patients with bilateral morbidity or severe dysplasia is difficult . In our study, we found a significant difference in the pre- and postoperative total offset changes on the operative side between the dislocation and non-dislocation groups. This method of evaluating offset has a great advantage in that it is not affected by the condition of the contralateral hip joint. Our findings also suggest that when postoperative offset on the operative side is reduced from the preoperative offset, soft tissue tension is reduced, thus, resulting in bony impingement and increased susceptibility to dislocation. To our knowledge, no other study has compared the pre- and postoperative offsets on the affected side. In this study, regarding the occurrence of pain on the lateral side of the hip, there was no significant difference between patients with > 5 mm of postoperative femoral offset on the operated side compared with the healthy side and those with < 5 mm. Many studies have reported that pain does not increase with increasing offset, which supports our results [23, 35–38].
The limitations of this study should be noted. First, three types of femoral head sizes were used in this study. Regarding the effect of different head sizes on dislocation, many reports have shown that there was less dislocation in those sized ≥ 32 mm compared to those sized 28 mm . However, in this study, no difference in dislocation was observed after comparing head sizes of 28 and ≥ 32 mm. Similar results were obtained for CA and offset for dislocation only in cases with a head size of ≥ 32 mm. Second, we used a non-flatliner to prevent dislocation in patients with inadequate intraoperative stability. These cases were excluded from the study because postoperative CT could not accurately measure the angle of placement of the non-flatliner, which may overestimate the effect of CA on dislocation. Third, the small sample size of nine patients with dislocations prohibited a multi-factor analysis. More dislocation cases should be evaluated to improve the accuracy of the analysis. Fourth, long-term adverse effects in cases of excessively increased femoral offset were not evaluated. Polyethylene wear reportedly increases when femoral offset increases by more than 5 mm compared with the normal hip joint . In this study, there were 22 cases, in which femoral offset increased by ≥ 5 mm compared with the healthy side in 112 patients with unilateral disease, and the patients need to be carefully monitored in the future. Fifth, the proportion of female individuals in this study was larger than that of male individuals, showing a substantial sex bias. The reason for this is that most of the indications for total hip arthroplasty in Japan are of osteoarthritis associated with acetabular dysplasia, and the prevalence of this disease is approximately three times higher in women . In the present study, the incidence of dislocation was not significantly different between men and women.