Pelvic tilt between supine and standing after total hip arthroplasty an RSA up to seven years after the operation

The pose of the prosthetic components after total hip arthroplasty (THA) is commonly evaluated on conventional radiographs. Any change of the pelvic position after the operation in supine and between supine and standing position with time will influence validity of the measurements. We evaluated the changed pelvic tilt angle (PTA) in supine and standing position up to 7 years after operation. The aims of our study were (a) to evaluate if the PTA change over time after THA, (b) to assess any difference in PTA between supine and standing positions, and (c) to investigate whether factors such as gender, the condition of the opposite hip or low‐back pain have any influence on PTA after THA. Repeated radiostereophotogrammetric radiographs of 106 patients were studied. Patients had been examined in the supine position postoperatively, and in both supine and standing positions at 6 months and 7‐year follow‐up. Measurements of supine patients showed an increasing mean posterior pelvic tilt over time. From supine to standing, the pelvis tilted in the opposite direction. At 6 months, the mean anterior tilt was 3.6° ± 3.8° (confidence interval [CI]: 2.8° to 4.3°) which increased to 6.4° ± 3.9° (CI: 5.7° to 7.2°) at 7 years. The mean changes in pelvic rotations around the longitudinal and sagittal axis were less than 1 degree, in both positions. In individual patients, this change reached about 11.0 degrees in supine and 18.0 degrees when standing.

We aimed to use RSA to evaluate if PTA changes over time in patients with total hip prosthesis and if the pelvic tilt differs between supine and standing positions at 6 months and 7 years after the operation. We also evaluated whether factors such as gender, the status of the opposite hip, or presence of any low-back disability influence the PTA after THA. To our knowledge, this is the first report on changes of pelvic tilt using RSA in patients operated with THA.  (Digas et al, 2004) 18 . The hip with most well-scattered pelvic markers included; (c) 52 cemented THAs in an evaluation of two types of bone cement (not published previously); (d) 84 cemented THAs in a study aimed to study different types of stem design (Thien et al, 2010) 19 . THA, total hip arthroplasty [Color figure can be viewed at wileyonlinelibrary.com] examination on at least two occasions (at 6 months and 7 years after surgery), in addition to the supine postoperative RSA examination. The purpose of examining patients in supine and standing positions was to evaluate any potential differences in femoral head penetration between these two positions. 20 Only hips marked in the segment of interest, that is, the acetabular bone with at least three tantalum markers or more and with a scatter corresponding to a condition number less than 125 and a mean error of rigid body fitting corresponding to at most 0.35 mm were included. 21,22 One hundred and six patients (106 hips) fulfilled the inclusion criteria ( Figure 1). All patients had provided informed consent for the insertion of tantalum markers during the operation and the subsequent RSA examinations. Approval was obtained through the Regional Ethics Committee, S 257-00, R312-98, Ö449-00, and Ö450-00. Patient demographic data are presented in Table 1.  (Table 2A). Pelvic rotations about the other two axes (longitudinal, y-axis, and sagittal, z-axis) are also presented (Tables 2B and 2C). We also reviewed the medical records of all patients regarding the incidence of low-back pain. A history of low-back pain was con-

| Radiostereometric analysis
For the purpose of this study, we calculated the motions of tantalum markers inserted into the pelvis (the pelvic bone segment) in relation

| Statistical analysis
Statistical analysis was performed using IBM SPSS, version 20.0.0. All RSA data were normally or close to normally distributed (Shapiro-Wilk; P < .05). The data are presented as the mean ± standard deviation (SD) at a confidence interval (CI) level at 95%. Paired-samples t tests were used to evaluate any differences between two repeated observations from the same patients and an independent sample t test or one-way analysis of variance was used for comparison between different groups. Statistical significance was assumed at P values less than .05.   (Tables 2A and 2B).

| Low-back pain
In the supine position, no significant differences in PTA were observed between the patients with or without low-back pain at any of the two follow-up occasions (P > .05). From supine to standing, a weak and insignificant tendency toward larger anterior tilt was observed in patients with no low-back pain on both occasions.   (Table 4B) were associated with female gender (P = .004) and patient age (P = .02).

| Linear regression analysis
Any influence of the status of the opposite hip was only studied for supine to standing at 7 years, because the exact time frame for the development of OA on the nonoperated side could not be established. Analysis of supine to standing results at 7 years did not reveal any influence of the opposite side (normal, OA, or THA) on the recorded PTA (Table 4B).

| DISCUSSION
A steep acetabular cup inclination increases the risk of dislocation and is also considered as an important reason for increased wear regardless of the material used. [23][24][25] Routinely, the inclination of the cup is determined postoperatively on radio-  up to at least 6 months after the operation and even during longer in some patients. In changing from a supine to standing position, the pelvis tilted in the opposite (ie, anterior) direction. This anterior tilt increased with time and became significant at 7-year follow-up especially in women. The increased pelvic tilt in woman may be associated with age-related changes of the spine due to osteoporosis, which increases with age and especially in females. [29][30][31][32] Female gender and increasing age have previously been found to be consistent with larger variations of PTA, 33 36 Pierrepont et al, 37 Tamura et al, 35 Uemara et al 38 ) matched three-dimensional measurements or reconstructions from computed tomography images to conventional radiographic images. They found a high correlation between these measurements and in one of these studies, a median absolute error of 1.5 degrees was reported, 38 but none of these calculations seem to be based on repetition of the examination procedure. In our study, we only repeated examination supine to study precision.
Further studies with repeated examinations of the standing position in one and the same patient could also be of interest, but at the time period for this study, we had no ethical allowance for expansion of the double examinations.
In the previous studies quoted above, the pelvic positions were determined based on anatomical landmarks and the change of pelvic tilt between these positions were computed. We measured change of pelvic tilt corresponding to rotations about the transverse axis of the RSA calibration cage by determination of position changes of clusters of markers inside the pelvic bone. We were therefore unable to account for any differences between the groups at the starting or reference supine position which is a limitation.
Patients with low-back problems might be expected to have reduced mobility of the spine and the pelvis. The evaluation of low-back pain in our study was retrospective and could, therefore, be regarded as suboptimal. However, the differences between patients with or without low-back pain were small and amounted to only one to two degrees on average. These findings were consistent with a metaanalysis comparing pelvic kinematics in patients with and without | 127 low-back pain, which reported no difference between these patient groups. [39][40][41][42] According to our findings and in patients without spinal fusion, the influence of low-back problems on the stability of a hip prosthesis seems to be small.

| CONCLUSION
Up to 7 years after insertion of a THA, the pelvis tilts slightly posteriorly in the supine position. When rising to standing, we found that