As we all know, the center of gravity and visual balance are maintained by regulating the curvature of the spine, the posterior rotation of the pelvis, and the lower extremities [15]. Previous studies have reported that any pattern of the degenerative spine is not a static entity, which resulted from evolution [16]. As reported, the patients with LDH are characterized by a flat spine with lower LL, SS, and TK as well as higher PT, SVA and TPA [8, 17–19], which agree with our results. However, the sagittal profiles of the LDH population are significantly different from the asymptomatic volunteers with matched age and sex distribution. This can be explained by the gradual loss of lumbar lordosis with disc degeneration, which could be considered as the initiating event of the compensatory mechanism [20, 21]. Consequently, patients generate a decrease of SS with an increase of PT (the backward rotation of the pelvis) to keep the balance of the “cone of economy” and to achieve a horizontal gaze. Besides, the loss of lordosis potentially leads to the anterior translation of the gravity axis (increasing SVA and TPA). Nevertheless, no significant difference was found in SVA between the two groups, indicating that the spine was still in a state of balance in the elderly LDH patients. The increase of TPA in patients with LDH indicated that the pelvic compensation might reach the limit, but the value was still within the normal range. In addition, the forward posture to relieve pain could also increase TPA. Our study also found that TK in the elder population was greater than that in the younger population, which is significant in both the LDH and the asymptomatic group [22, 23], confirming the tendency of more frontward postures with aging. It might be that their degenerative muscle system couldn’t maintain the original curve due to muscle atrophy in the elderly population [20]. However, our study found that the TK value of herniated population was also smaller than that of the asymptomatic volunteers, which suggested that TK still had the compensatory ability to maintain the sagittal balance in the elderly population ≥ 60 years old.
The previous study carried by Roussouly et al. [24] suggested that type 3 or type 4 lordosis (almost 70%) were most prevalent in the asymptomatic Caucasian adult. Several studies with Asian populations also showed similar Roussouly type distributions [23, 25]. However, some recent studies involved LDH populations had noted that subjects were mostly classified into type 1 and 2 [9, 10]. In the present study, the number of type 1 and 2 cases (almost 79.1 %) increased significantly in the elderly LDH patients, while the number of type 3 and 4 cases decreased, which is consistent with the findings of the studies above. Those results indicated that a straighter shape of the spine and pelvis (type 1 and 2) might increase the risk of LDH. The possible mechanism might be that the subjects with a straighter shape shifting the vertical load distribution on the intervertebral disc and the limited compensatory capacity (the lumbar hyperextension and pelvic retroversion ) due to the short hyperlordosis and the low PI, which will increase the injury rate to the lumbar disc resulting in disc degeneration and herniation. Another reason might be that the spinal degeneration and its compensation might alter the original Roussouly type [12]. The decrease of LL in the elderly patients leads to the extensive backward tilt of the pelvis, so the subjects who now have a smooth spinal pelvic alignment may have shifted from a greater curved alignment. Emmanuelle et al. [26] also reported that the increased proportion with type 2 is shifted from type 3 in subjects with low back pain. Furthermore, our study also found that compared with the young and middle-aged LDH patients, the proportion of type 1 in the elderly population further increased and became the greatest part. This might be due to further spinopelvic degenerative evolution with aging in the elderly LDH population [12]. The decrease of LL in patients leads to the extensive backward tilt of the pelvis, so the subjects with a smooth spinopelvic alignment might show a steeper pelvis in the original shape. However, these assumptions should be confirmed in future studies. Meanwhile, a high proportion of type 1 and 2 also might be the reason why the PI of the herniated population is smaller than that of volunteers.
In the present study, the severity of disc degeneration is gradually increasing from the proximal to the distal segment in all subjects. Youngbae et al. [2] also concluded with similar results that the degenerative lumbar lordosis loss usually began in the lower lumbar spine. It might be related to the different segment mobility at each lumbar level and the severity of disc degeneration tended to be prevalent in the lower segments, which accounts for a greater lordosis and shows increased mobility compared the upper segments [5, 27]. Besides, the lower segments was bore to heavier loads, which would accelerate disc degeneration [28]. As many studies suggested that the sagittal alignment is an important indicator of the disc load and pressure [5, 17, 19, 20], our study also found that compared with type 3 and 4, type 1 and 2 subjects have significantly lower locations and fewer lumbar disc herniation. Types 1 and type 2 spinal subtypes are characterized by the large thoracic kyphosis, the small lumbar lordosis, and the lumbar curvature is mainly in the distal lumbar spine [1], which might result in greater pressure on the distal lumbar disc. While for type 3 and 4, a greater lumbar lordosis represents the greater mechanical stress on the posterior components instead of on the disc, which was potentially a protection to the intervertebral disc.
Several limitations still exist in this study. Firstly, multiple other factors affecting human posture need to be investigated in future longitudinal studies for more accurate evaluation, like lifestyle and occupation. Secondly, this is a retrospective study at a single center, and lacks experimental or mechanical research support and the generalisability of the results. Therefore, further detailed and longitudinal prospective multicenter studies will be needed.