As oophorectomy, menopause, and pregnancy can effect estrogen level[25, 26], women who had undergone hysterectomy after menopause or oophorectomy were excluded in this series. In addition, there was no significant difference in the number of previous pregnancies between the two groups.
In this study, the degeneration of lumbar paravertebral, facet joints, intervertebral discs, and endplates was more severe in the HR group than in the NHR group. This phenomenon might be explained by hormonal changes that occur after hysterectomy. It was reported that estrogen plays an important role in degeneration of lumbar discs, facet joints, vertebral muscles, and vertebral endplates[3–7]. Hysterectomy might cause decreased ovarian function and a drop in estrogen level[12, 27]. These results support the hypothesis that lumbar spine degeneration may be accelerated by hysterectomy.
Another explanation for worsening lumbar spine degeneration after hysterectomy is that hysterectomy might cause pelvic floor and abdominal muscle deficits, leading to degenerative changes[28]. Panjabi et al. proposed that the diaphragm, pelvic floor muscles, transversus abdominis, and deep lumbar extensor muscles are responsible for spinal stability[29]. Moreover, previous studies have shown that impaired function of the muscular stability system is associated with low back pain[30, 31]. Therefore, the weakened pelvic and abdominal muscles may be also related to lumbar spine instability and, in the long run, lumbar spine degeneration.
Paravertebral Muscles
Several studies have reported that CT images can be used to measure the muscle FCSA and radiological denstiy[15–17]. The paravertebral muscles maintain the stability of the spine, lumbar lordosis and extension. In this research, the FCSA, SFCSA, and radiological density results are similar to those of previous studies[17, 32]. However, no significant differences in FCSA and SFCSA of the multifidus were found between the two study groups. Several studies have reported that multifidus atrophy is more susceptible to aging than other paravertebral muscles[33, 34]. Therefore, aging, rather than hysterectomy, may have played a dominant role in multifidus degeneration. Another possible explanation may be the lack of adequate sample size.
FJOA
The prevalence of FJOA was found higher in the HR group. Meanwhile, in another unpublished research by our team, more severe sacroiliac joint degeneration was also found in hysterectomized patients, suggesting that hysterectomy may be a risk factor for articular cartilage degeneration. Notably, the incidence of FJOA in this study was is consistent with a number of studies in East Asian countries[35], but lower than those in the US[36]. The difference may be attributed to ethnicity.
Intervertebral Disc Index And Pfirrmann Grade
DHI, an advanced sign of lumbar disc degeneration, was significantly lower in the HR group than the NHR group. Meanwhile, the Pfirrmann grades were significantly higher in the HR group at L1-L2 to L3-L4. Notably, there was no significant difference at L4-L5 to L5-S1 between the two groups. A reasonable explanation for this phenomenon is the L4-L5 and L5-S1 segments are responsible for most of the body’s weight, and the biomechanical characteristic may play the key role in the degeneration of the discs at the two segments. Disc degeneration is closely associated with degenerative lumbar spine disease, such as spinal stenosis[37]. Therefore, hysterectomy could be an associated risk factor.
Endplate Damage
In this study, EPD was more severe in the HR group from L3-L4 to L5-S1. Endplates are mainly composed of water, collagen II, and aggrecan. The endplates provide nutrition to the disc through diffusion and contribute to disc stability. Rajasekaran et al. reported that EPD can lead to disc degeneration via degradation of the matrix, decrease in nutrient diffusion, and loss of integrity[38]. In general, the endplate is fragile and susceptible to mechanical damage[39]. This study suggests that EPD may be also associated with estrogen, but further research is needed.
Modic Changes
Modic changes are changes in MRI signal intensity in the vertebral body marrow adjacent to the endplate. Type I change indicates acute inflammatory changes in the subchondral bone marrow, type II indicates lipid replacement, and type III indicates osteosclerosis[22]. Incidence of Modic changes is higher in patients with low back pain (median, 43%) and types I and II are the most common[40]. In our study, the incidence of Modic changes was significantly higher in the HR group. The incidence of type II Modic change was highest in both groups. Shahmohammadi et al. reported that Modic changes are related to age, disc degeneration, and sagittal lumbopelvic parameters[41]. Hormonal factors are also associated with incidence of Modic changes, which are more common in women over 45 years[42], and a variety of underlying mechanisms are thought to be implicated, including accelerated degeneration of discs, paravertebral muscles, and facet joints. The higher incidence of Modic change in the HR group indicated that Modic changes may associated with hormonal factors.
This study has some limitations. First, the estrogen level in patients was not examined before and after hysterectomy. Second, the authors only examined radiological measures of degeneration; clinical symptoms were not considered.