In the present study, we investigated conventional and novel measurement patterns of spinopelvic sagittal alignment in healthy adult volunteers. The standing sagittal alignment was similar to that in previous studies. The correlations among the radiographic parameters of sagittal alignment were also comparable with those of previous reports [24, 25]. The correlation between SIP and SS, PIA, and LL is remarkably similar to the correlation between PI and SS, PT, and LL. Just as PI plays a key role in the regulation of positional pelvic and spinal parameters, SIP has the potential to be a key parameter in spinopelvic alignment [25].
As mentioned previously, the evaluation of spinal sagittal alignment, as well as pelvic version according to pelvic morphology, especially PI, are crucial in hip or spine surgeries. Although use of HA, which is the centre of pelvic version, is optimal to assess pelvic morphology, an accurate assessment of PI is impossible on lateral plane radiographs in cases with deformed or dislocated femoral heads due to severe osteoarthritis or necrosis. Recently, alternative morphologic parameters have been proposed, without use of HA. Imai et al. reported a strong correlation between PI and anatomical-SS as a morphological parameter, which is the value of the SS measured from the pelvis adjusted in the anterior pelvic plane [26]. Wang et al. suggested sacrum pubic incidence as another alternative parameters with strong correlation with PI for determining the morphology of the pelvis [24]. Instead of the HA, the upper edge of the pubic symphysis has been proposed as a key anatomic landmark of the anterior pelvic plane for assessing pelvic orientation, and we have followed a similar concept with our novel parameter. Although the superior plate of the S1 at its midpoint is identified as a bending point in the traditional parameters, identification of the posterior edge of S1, which is needed to find the midpoint, is difficult to identify in some cases with strong L5/S1 degeneration. Therefore, we used the superior plate of the S1 at its anterior edge as a bending point for our novel parameter, SIP, which does not require the posterior edge of S1. Furthermore, the PIA can be used to evaluate pelvic orientation instead of the PT [22]. In the present study, both SIP and PIA showed very high intra-observer and inter-observer reliability (all ICCs > 0.95), and strong correlation with PI and PT, respectively (both R > 0.9). Therefore, these two parameters could be alternative morphologic or positional parameters of the pelvis to assess whole body sagittal alignment.
Previous studies have reported that deterioration of sagittal alignment, especially loss of LL due to common spinal disorders, decreases HRQOL [7-11]. Therefore, the purpose of reconstructive surgery is to restore physiological LL and PT, which is equivalent to approximating LL to PI [12, 18]. Although PI is a crucial target during reconstructive surgery, accurate identification of PI is not possible in patients suffering from hip-spine syndrome due to deformed or dislocated femoral heads. Therefore, the alternative morphological pelvic parameter that we propose and our novel concepts for prediction of ideal LL are useful for evaluation of lumbar degeneration as well as alignment correction planning in such complex cases during reconstructive surgery. There have been several concepts related to the prediction of the ideal LL, and the lordosis predictive equation can be based for not only age and pelvic morphologic parameters, but also for thoracic kyphosis and L1 or T9 tilt because of the reciprocal chain of correlations between the positional pelvic and spinal parameters and the morphological PI [25, 27]. However, since positional parameters, especially spinal parameters, can be frequently affected by common aging conditions, such as spinal fractures and degenerative disc diseases, our simple equation using a definitive morphologic parameter to predict ideal LL would be practical in the clinical setting.
Two limitations of our study should be acknowledged. Firstly, the proportion of female volunteers we recruited in this study was as high as 75%. However, a previous report showed that sex has no effect on spinopelvic parameters [25]. Secondly, the study was conducted on individuals with a wide age range, in which age-related degeneration may have affected spinopelvic alignment in some cases. Although our previous study with 136 healthy patients showed that there was no statistically significant correlation between age and LL [20], LL decreases slightly with age [28]. Therefore, age was selected as a contributing factor for ideal LL in our study. Even if participants with a wide age range are analysed, the strong correlation between SIP and PI can be said to be applicable to cases of any age.
On the other hand, the present study has some strong points compared to the previous studies. Firstly, we investigated sagittal whole-body skeletal alignment and used a scanning X-ray imager with a biplanar upright scanning imaging modality to achieve reduced X-ray particle scatter, improved image quality and significantly reduced radiation to the patient [23]. Secondly, we strictly excluded all patients with anomalous vertebrae, such as transitional vertebrae and scoliosis with a Cobb angle > 20°, which can affect measurement precision.
From the perspective of future possibilities, we need to investigate the accuracy or reliability of SIP and PIA measurements on conventional X-ray images compared to PI and PT. Moreover, further studies are needed on whether our novel concepts can predict the HRQOL status in patients with progressive hip arthritis with hip dislocation or femoral head deformity. We believe that use of our novel concepts can support future research in the elucidation of the various pathologic conditions on hip-spine syndrome.