Chronological Changes in Functional Cup Position at 10 Years after Total Hip Arthroplasty

Introduction

The position of the acetabular cup has a significant effect on the postoperative outcome of total hip arthroplasty (THA). Cup malposition can cause impingement and edge loading, and could consequently lead to dislocation, polyethylene wear, and aseptic loosening (1-2-3-4-5). A computer navigation system has been developed to improve implant positioning in THA (6, 7). Although the computer navigation system utilises the anterior pelvic plane (APP) as a reference, pelvic orientation itself differs significantly during the activities of daily living (8), which can affect the cup position after THA (9).

Several authors have recently proposed the concept of functional cup position in terms of a combination of implant alignment in bone and pelvic orientation relative to the body (10-11-12-13-14-15). Patients who exhibit a significant difference between radiographic and functional cup position ideally require a unique adjustment in cup positioning to maximise stability and range of motion (11, 15, 16). Previous studies have reported contradictory results regarding the predictability of postoperative pelvic orientation (14, 17, 18). Furthermore, when evaluating the long-term results of THA, the postoperative chronological changes in pelvic orientation should also be considered. Although pelvic tilt is considered a major determinant of functional cup position, previous studies have indicated varying results for postoperative change in pelvic tilt (17, 19-20-21). However, these studies had relatively short follow-up periods (1-3 years) and do not actually reflect the progressive degenerative changes that occur with aging. In the present study, we aimed to clarify chronological changes in functional cup position at a minimum follow-up of 10 years after THA, and to identify the risk factors influencing a significant difference between radiographic and functional cup position.

Materials and methods

This study was approved by our Institutional Review Board (ERB-101566). Between March 2002 and August 2005, 81 patients with unilateral osteoarthritis underwent primary THA. 7 patients died within 10 years from unrelated causes, 8 patients were lost to follow-up, and 8 patients were excluded from this study as their radiographs were not suitable for evaluation or because a positive Thomas test result was observed. Thus, a total of 58 patients were included in this study. Only patients with unilateral primary osteoarthritis of the hip were included, whereas those with bilateral osteoarthritis, rheumatoid arthritis, hip contracture, and prior arthroplasty were excluded. The mean follow-up period was 11 years (range 10-13 years).

Anteroposterior pelvic radiographs in the supine and standing positions were obtained before, 3 weeks after, and every 1 year after surgery. Also, lateral pelvic radiographs in the supine and standing positions were obtained before surgery. The anteroposterior and lateral lumbosacral radiographs were obtained to evaluate the possibility of vertebral fractures during the postoperative follow-up period. Clinical evaluations included the determination of the Harris Hip Score (HHS) (22).

Radiographic evaluation

Pelvic tilt in the sagittal plane was expressed as the APP angle, which was measured on lateral pelvic radiographs using the CIS-image (IBM Corporation). This angle was defined as the angle formed by a line drawn from the anterior superior iliac spine to the pubic symphysis (APP) and the vertical axis (Fig. 1) (17). Positive APP angle values represent anterior pelvic tilt, whereas negative APP angle values represent posterior pelvic tilt. Lateral radiographs were obtained in the standing position with the hip in extension and the arm resting at 90° on a support (12), as well as in the supine position at an equal distance from the tube. The criterion for radiographic success included appropriate superimposition of the anterior superior iliac spine.

OPEN IN VIEWER

Fig. 1

Measurements of the anterior pelvic plane (APP) angle. The APP angle was defined as the angle formed by a line drawn from the anterior superior iliac spine to the pubic symphysis and the vertical axis. Positive APP angle values represent anterior pelvic tilt, whereas negative APP angle values represent posterior pelvic tilt.

The radiographic cup position was measured on anteroposterior pelvic radiographs in the supine position, whereas functional cup position was measured in the standing position (14). The radiographic abduction angle of the cup (R-Abd) was defined as the angle formed by the line connecting the tips of the teardrops and the line connecting the upper and lower end of the open plane of the cup in the supine position. The functional abduction angle of the cup (F-Abd) was defined as the angle formed by the line parallel to the ground and the line connecting the upper and lower end of the open plane of the cup in the standing position. The radiographic anteversion angle of the cup (R-Ant) and functional anteversion angle of the cup (F-Ant) were measured using anteroposterior pelvic radiographs in the supine and standing positions, respectively. The anteversion angle was calculated using Lewinnek's formula: cup anteversion angle = arc sin (D1/D2). D1 is the distance of the short axis of an ellipse drawn perpendicular to the long axis of the acetabular component, and D2 is the distance of the long axis (the maximal diameter of the implant) (24). The cup anteversion angles were semi-automatically measured using 2D Template software (KYOCERA Medical) (Fig. 2).

OPEN IN VIEWER

Fig. 2

Measurements of cup anteversion angle. The cup anteversion angle was calculated using Lewinnek's formula: cup anteversion angle = arc sin (D1/D2). D1 is the distance of the short axis of an ellipse drawn perpendicular to the long axis of the acetabular component and D2 is the distance of the long axis.

All measurements were performed by 2 independent observers (Y.O. and S.T.). The measurements for a random set of 30 radiographs were repeated after an interval of at least 1 week to determine intraobserver reliability.

Results

The study sample comprised 52 women and 6 men. The mean age of patients at surgery was 67 years (range 36-81 years). The mean preoperative APP angle was −7.6° (range −40.2°-17.9°) in the standing position and −2.8° (range −27.6°-11.7°) in the supine position, whereas the mean difference in the APP angle between the standing and supine positions (dAPP) was −4.9° (range −32.4°-7.8°). The positive values of dAPP represent a change towards the forward rotation of the pelvis, whereas negative values of dAPP represent a change towards the backward rotation of the pelvis.

All implants showed radiographic stability and the absence of any radiolucent lines at the final follow-up. At 3 weeks after surgery, the functional cup position was clearly different from the radiographic cup position. Both F-Ant (22.0° ± 2.3°) and F-Abd (45.1° ± 1.9°) were significantly greater than R-Ant (16.7° ± 2.7°) and R-Abd (42.1° ± 1.9°), respectively (p<0.05) (Fig. 3). The F-Ant increased over time, and was found to have significantly increased at the final follow-up compared to that at 3 weeks after surgery (p<0.01) (Fig. 4). The maximum postoperative change in F-Ant was 17.0° anteriorly; 12 cases (21%) showed a postoperative change in F-Ant of >10° anteriorly (Fig. 5).

OPEN IN VIEWER

Fig. 3

Comparison of cup anteversion and abduction angles between radiographic and functional cup positioning. Error bars represent standard error. *p<0.05.

OPEN IN VIEWER

Fig. 4

Chronological changes in angle differences between radiographic and functional cup positioning. Error bars represent standard error. F/U = follow-up. *p<0.05 vs. after 3w.

OPEN IN VIEWER

Fig. 5

The frequency of postoperative change in F-Ant. The maximum postoperative change in F-Ant was 17.0° anteriorly; 12 cases (21%) showed a postoperative change in F-Ant of >10° anteriorly.

1 patient with vertebral fracture required revision surgery due to anterior dislocation of the hip at 6 years after the surgery. Dislocation occurred while the patient was turning around in a standing position. The postoperative change in the F-Ant was 14.4°anteriorly in that patient. Moreover, in all patients, the HHS significantly improved from 54 points (range 15-73 points) preoperatively to 82 points (range 47-100 points) at the final follow-up (p<0.05).

Multiple regression analysis indicated that the preoperative APP angle in the standing position and a history of vertebral fractures during the postoperative follow-up period were significant contributors to the differences in F-Ant during the postoperative follow-up period (Tab. I). The intra- and interobserver ICC were ≥0.90, which suggests good intra- and interobserver agreement of the radiographic measurements.

OPEN IN VIEWER

TABLE I

Multiple regression analysis of factors associated with the chronological changes of functional cup anteversion (F-Ant)

Factors	β	t	Significance (p)
Age	0.138	1.195	0.238
Sex	0.183	1.818	0.750
Vertebral fractures	0.624	5.261	<0.001
Supine APP	-0.293	-1.446	0.154
Standing APP	0.478	2.645	0.011

APP = anterior pelvic plane.

Discussion

The most important findings of the present study were: (i) the change in functional cup anteversion was >10° over the 10-year follow-up period in 21% patients, and the number of outliers in safe zone for combined anteversion was greater after 10 years than immediately after surgery; and (ii) the preoperative posterior pelvic tilt in the standing position and vertebral fractures after THA were significant predictors of increasing functional cup anteversion. Although the radiographic cup position immediately after surgery was acceptable in most cases, the functional cup position at the final follow-up was markedly altered in individual cases. In these cases, it is difficult to evaluate the accuracy of acetabular cup position using Lewinnek's safe zone (25). The findings in the present study strongly suggest the use of the functional cup position concept. Hence, surgeons should be aware of the differences between the radiographic and functional cup positions, as well as the chronological changes in the functional cup position.

Our results showed that only functional cup anteversion, not abduction, was significantly increased over the 10-year follow-up period. Aging had a greater effect on the sagittal spinopelvic alignment than on the coronal alignment (14, 26, 27). Aging of the spine is characterised by facet joint arthritis, degenerative discs, degenerative lumbar diseases, and atrophy of extensor muscles, which can lead to a loss of lumbar lordosis and an increase in pelvic tilt (26). In addition, vertebral fractures frequently occur in the aging spine. Fechtenbaum et al (28) showed that lumbar Cobb index and pelvic tilt differed in patients with and without vertebral fractures. In fact, vertebral fractures during the postoperative follow-up period were significant contributors of increasing functional cup anteversion in the present study. Hence, the postoperative changes in spinopelvic alignment should be carefully considered, particularly those in the sagittal plane. As the anteroposterior pelvic radiographs in the standing position were influenced to a greater extent by the postoperative changes of spinopelvic alignment, as compared to those in the supine position, we believe that the postoperative follow-up examinations should include anteroposterior pelvic radiographs in the standing position.

Although pelvic tilt is considered as a major determinant of functional cup position, previous studies have indicated varying results regarding the postoperative change in pelvic tilt (17, 19-20-21). Blondel et al (17) reported that pelvic tilt did not significantly differ between the preoperative stage and that at 3 years after THA. Moreover, Murphy et al (20). reported that the mean standing pelvic tilt was altered by <1° after THA. In contrast, other studies from Japan showed that pelvic tilt in the standing position changed by >10° in a significant number of patients (19, 21). The differences between studies could be attributed to age, gender, and racial differences. Furthermore, these studies had a relatively short follow-up period (range 1-4 years) and did not actually exhibit any progressive degenerative changes with aging.

Although chronological changes in functional cup position do occur after THA, their magnitude is relatively low. Thus, in our cohort, only 1 case required revision surgery because of anterior dislocation of the hip. 1 of the reasons is that, in patients with a retroverted pelvis, posture during walking leads to an increase in the trunk inclination angle compared to that in the standing position (29). However, in cases with forced hyperextension, posterior impingement is likely to occur, which may cause edge loading, wear or breakage of the polyethylene liner, and anterior dislocation of the hip (30). Although rare, such cases have been previously reported (31, 32), necessitating careful monitoring during the long-term follow-up after THA.

In the present study, we found that the preoperative posterior pelvic tilt in the standing position was a significant predictor of increasing functional cup anteversion, consistent with a previous study (21). Based on our results, the mean difference between F-Ant and R-Ant was approximately 6° immediately after surgery, and 21% of patients had a change in functional cup anteversion of >10° over the 10-year follow-up period. Hence, we recommend that surgeons should aim to reduce cup anteversion to within impingement-free limits in patients with preoperative posterior pelvic tilt in the standing position. We believe that the safe zone for the combined anteversion technique should probably be between 5 and 10° narrower than that in standard cases.

The present study has certain limitations. 1^st, this was a retrospective study with a small number of patients. All patients were Japanese, and most of the patients were women. 2^nd, preoperative spinal alignment and knee contracture were not evaluated in this study; however, these preoperative factors could affect the functional cup position (33, 34). 3^rd, the radiographic cup position was measured on anteroposterior pelvic radiographs using the 2-D Template software. For the more detailed measurement for the interpretation of the magnitude of chronological changes, towards the Radio Stereometric Analysis method (35) is desirable. Despite these limitations, we believe that the study provides clinically important information regarding functional cup position for long-term success in THA.

In conclusion, functional cup anteversion significantly increased over the 10-year follow-up period; this was associated with a greater number of outliers at the 10-year period than immediately after surgery. Moreover, preoperative posterior pelvic tilt in the standing position and vertebral fractures after THA were significant predictors of increasing functional cup anteversion. Although chronological changes in functional cup position do occur after THA, their magnitude is relatively low. However, posterior impingement is likely to occur, which may cause edge loading, wear of the polyethylene liner, and anterior dislocation of the hip. We believe that, for the combined anteversion technique, the safe zone should probably be 5°-10° narrower in patients predicted to show considerable changes in functional cup position compared with standard cases. Indeed, appropriate treatment is essential to reduce the risk of osteoporotic fractures during follow-up after THA.