Osteoporosis (OP) is a degenerative disease characterized by reduced bone mass and bone structural degeneration, which leads to changes in the bone microarchitecture and imbalances in bone homeostasis. OP is often closely related to bone pain and fragility fractures (Akkawi and Zmerly, 2018). In China, the prevalence of OP in women over 50 years old is 29%. In the United States, an estimated 10.2 million people 50 years or older have OP (Black and Rosen, 2016), and it is estimated that by the end of 2025, the economic burden of OP will reach USD 25.3 billion in the United States alone (Johnell and Kanis, 2006).

The vertebral endplate is a thin layer of tissue located on both sides of the intervertebral disc composed of cartilage and bone. The vertebral endplates constitute the mechanical interface between the intervertebral disc and the vertebral body and facilitate the even distribution of compressive stress across the vertebral body. In addition, the endplates contain bone marrow contact channels, which provide nutrients to the intervertebral discs, excrete metabolic wastes by diffusion, and enable the flow of fluid under cyclic loading (Holm et al., 1981; Mattei, 2013). Endplate defects can promote communication and continuous cross-talk between the intervertebral disc and vertebral body, which may lead to a persistent inflammatory state in the vertebral marrow and become an important source of low back pain. The relatively thin and perforated physiological structure of the vertebral endplate cannot bear the long-term compressive load and is prone to fracture and defect. Currently, some of studies have confirmed the connection between osteoporosis and endplate injury (Ferrar et al., 2007; Ferrar et al., 2012; Fujiwara et al., 2019; Homminga et al., 2001; Lentle et al., 2021; Wang et al., 2019; Zhang et al., 2021). With respect to whether OP further aggravates vertebral endplate destruction by affecting the bone microenvironment and increasing the fragility of the endplate, only a few studies have reported the relationships among these diseases. Xiao et al. reported OP-induced abnormal porosity in the cartilage endplates and harmful effects on the endplates caused by OP (Xiao et al., 2018). Wang et al. reported that vertebral endplate injury occurred in OP model rats (Wang et al., 2014). The pathogenic relationship between OP and vertebral endplate injury remains unclear.

As a mature bone mineral density (BMD) evaluation technology, dual-energy X-ray absorptiometry (DXA) is used in clinical diagnosis, but many studies have confirmed that vertebral hyperplasia, facet joint sclerosis, scoliosis, osteophyte and pannus formation can cause DXA measurement errors (Xiao et al., 2018; Bland et al., 2022). Recently, many experts have introduced and recommended an alternative method of assessing BMD in patients via the Hounsfield unit (HU) value through computed tomography (CT). Studies have confirmed that there is a suitable correlation between local BMD and HUs measured via CT (Alawi et al., 2021; Lee et al., 2013; Schreiber et al., 2011; Silva et al., 2012; Zou et al., 2020; Xue et al., 2024), and measurement errors caused by vertebral hyperplasia and facet joint sclerosis are avoided (Leboff et al., 2022).

X-ray, CT, and magnetic resonance imaging (MRI) can all visualize the vertebral endplates. Among them, MRI is considered the most sensitive and effective way to detect subtle changes in the structure of the endplate. Therefore, MRI can better reveal different degrees of endplate injury.

Recent studies on the relationship between OP and vertebral endplate defects involve qualitative grading rather than accurate measurement. Moreover, recent studies have reported that OP is closely related to the total endplate (TEP) score of vertebral endplate defects (Zhuang et al., 2021); however, the size and morphology of the endplate defect often affect other phenotypes, such as the high-density zone (HIZ), disc displacement, and facet joint changes. (Li et al., 2023). In view of these limitations, it is necessary to further investigate the relationship between OP and the phenotype and distribution of vertebral endplate defects and to further understand how OP is involved in the process of vertebral degenerative diseases.

Table 1 summarize and compare the demographic characteristics of the 199 patients who met the inclusion criteria during the study period. Among them, 69 patients were males, and 130 were females. The mean age was 67.1 years (range: 39–87, SD: 6.6 years). The mean BMI was 24.4 kg/m² (range: 16.2–36.4, SD: 3.5 kg/m²).

This study is the first to describe the morphology and distribution of endplate defects on MR images. Our study revealed that endplate defects were present in 47% of individuals and in 18% of vertebral bodies and that endplate defects were more common in individuals with OP, elderly individuals, and female patients (Table 1). With respect to the distribution of the spine, the endplate defects were mostly distributed in the upper lumbar spine (Figure 3). However, we found that the maximum width and depth of the vertebral endplate defect were not significantly correlated with the vertebral segment. Moreover, the mean age of patients with endplate defects was significantly greater than that of individuals with nonendplate defects, which was consistent with the study by Wang et al. (2012) This finding indicates that the depth of endplate defects increases with age and the progression of OP, which may be related to the fact that vertebral endplates become thinner and porous due to ageing, making it difficult for them to bear the corresponding mechanical load (Zehra et al., 2015). Additionally, the distribution of endplate defects in the upper lumbar spine was more extensive than that in the lower lumbar spine. This may be related to the greater thickness of the cortical bone in the lower lumbar spine than in the upper lumbar spine in order to withstand the increased load. (Edwards et al., 2001). Therefore, the lower lumbar spine had greater resistance to loading-induced endplate defects. In addition, compared with the lower lumbar spine, the upper lumbar spine participates more in rotational movement, the rotational movement puts a great deal of pressure on the endplate, and the repetitive rotational movement creates a tiny crack in the centre of the endplate cartilage, which becomes the weak part of the endplate; then, the crack dilates to form an endplate defect. (Gracovetsky and Farfan, 1986). In addition, some scholars believe that the inferior vertebral body has a larger cross-sectional area; during the period of axial loading, the mechanical stress is inversely proportional to the cross-sectional area of the vertebral body. (Duan et al., 2001). This may be another reason for the low incidence of lower vertebral endplate defects. The incidence of upper lumbar endplate defects was greater than that of lower endplate defects. There is no convincing explanation for this at present. Some scholars have proposed that this may be related to the fact that in the early fourth week of embryonic life in the early vertebral body, the cells of the cranial side are arranged loosely, whereas the cells on the caudal side are arranged densely. (Bury 1995; Moore et al., 2013). Therefore, in the early life cycle, the mechanical strength of the lower half of the vertebral body is weaker than that of the upper half, and the lower endplate is more susceptible to damage. In the lower lumbar spine, the proportion of superior endplate defects was greater, which may be related to the thinner cranial endplates supported by the trabecular bone with lower density (Zhao et al., 2009).

Our study, which was based on MRI and CT examinations, was the first to analyse the relationship between the size of the vertebral endplate defect and vertebral BMD. By measuring the vertebral HU value while avoiding more examinations for the patient, and previous studies have shown that the HU value is more effective than DXA in measuring BMD because, compared with DXA, the HU value measurement method avoids the influence of facet joints of the vertebral body. (Muraki et al., 2004; Pu et al., 2024). In addition, the endplate defect information of the patient was obtained through MRI examination. Analysis revealed that the width and standardized score of the endplate defect were greater in the low-bone-volume group than in the high-bone-volume group. In addition, correlation analysis further confirmed that the bone mass of the vertebral body was negatively correlated with the depth and width of the lamina defect. This finding is similar to that of the study by Zhuang et al. (2021), Ferguson and Steffen (2003). The vertebral endplate is a thin layer of tissue located between the vertebral body and the intervertebral disc. It acts as the mechanical interface between the rigid bone and the elastic disc and plays an important role in the transfer and bearing of mechanical stress in the spine. However, the vertebral endplate also acts as the main channel of nutrient supply, and its thin and porous structure plays an important role in intervertebral disc material transport, nutrient metabolism, and maintenance of nucleus pulposus pressure. However, this structure also makes it more susceptible to damage under the action of mechanical stress. BMD is closely related to the bearing capacity of bone. With decreasing BMD, bone is more susceptible to microfractures under the action of mechanical stress. In OP patients, central endplate deformation is more prominent (Ferguson and Steffen, 2003; DISC, 1988); moreover, the vertebral endplates are thinner, and the trabecular bone under the endplates is sparser. This may be a cause of vertebral endplate defects caused by OP. To further understand other risk factors for endplate defects, we used univariate multivariate logistic analysis and found that there were statistically significant associations between the total score of the endplate defect and age and total spine HUs. Increased age may accelerate bone metabolism disorders. When OP occurs in aging individuals, the vertebral endplates become thinner, and eventually, the vertebral endplate strength decreases. The faster damage accumulation under continuous cyclic loading makes the vertebral endplates thinner. Microcracks are generated to reduce the bearing capacity of the endplate under pressure. In addition, with aging, disc degeneration accelerates, the strength of the fibrous ring decreases, and horizontal cracks and blood vessel penetration of the endplate appear in the centre of the intervertebral disc and cartilage endplates (DISC, 1988; Roberts et al., 2006; Burr and Radin, 2003; Turgut et al., 2003). The weak point on the endplate becomes a factor in the herniation of the nucleus pulposus across the endplate and towards the vertebral body, resulting in endplate injury. Our study also revealed that vertebral endplate defects are closely related to low back pain. This may be related to the fact that endplate defects are often accompanied by vertebral endplate microfractures and the aggregation of inflammatory factors (Roberts et al., 2006; Burr and Radin, 2003; Turgut et al., 2003).

Finally, we attempted to explain the possible mechanisms underlying the relationship between OP and endplate defects. When vertebral body OP occurs, macroscopically, the vertebral endplate becomes thinner, and the remodelling of microfractures causes the trabecular bone under the endplate to be disordered under cyclic loading, the porosity of the endplate surface increases, and a cavity in the endplate occurs. Endplate edge defects or the formation of bone redundancy further reduce the mechanical load-resisting capacity of the vertebral endplate and increase the susceptibility of the vertebral endplate to breakage (Xiao et al., 2018; Luo et al., 2015; Liu et al., 2019; Luo et al., 2013; Sun et al., 2021; Zhong et al., 2016; Li et al., 2017; Xiao et al., 2020). At the microscopic level, OP changes the local bone metabolism microenvironment; bone resorption by osteoclasts is greater than bone formation by osteoblasts, resulting in abnormal bone reconstruction. (Frost, 1990). Pharmacological experiments confirmed that when anti-OP drugs are used, abnormal bone reconstruction of the endplate is inhibited, thus maintaining the structural integrity of the original endplate (Turgut et al., 2003; Liu et al., 2019; Luo et al., 2013; Sun et al., 2021). In addition, OP may accelerate the vascularization of the vertebral endplate. Neovascularization leads to the infiltration of inflammatory factors, macrophages, and mast cells, which also generate large amounts of growth factors, cytokines and enzymes, thus accelerating the destruction of the vertebral endplate and reducing its ability to withstand loads (Turgut et al., 2003).

Our study has a few limitations. First, our research is a cross-sectional study and cannot explain the causal relationship between vertebral endplate defects and osteoporosis. Further prospective experiments are needed for further investigation. In addition our measurement of vertebral endplate defects was based on imaging data. Although the measurements were made by experienced radiologists and orthopaedic surgeons, there are still gaps compared to results of direct measurement on cadavers. Besides, although we used the HU value to reflect the vertebral BMD, which can effectively avoid the influence of sclerotic bone and osteophytes on the vertebral BMD in the DXA measurement results. However, its accuracy is still lower than that of Q-CT. In addition, we used the VAS score scale to collect information on patients’ back pain. However, this cannot rule out other confounding factors causing back pain. Therefore, a more precise back pain scale can be further designed in subsequent studies to further clarify the relationship between back pain and endplate defects.

There is a correlation between OP and the size of the vertebral endplate defect, and the defect size increases with decreasing bone mass. In the population, vertebral endplate defects are more likely to occur in elderly individuals, females, and individuals with OP. With respect to the spinal structure, vertebral endplate defects are more likely to occur in the upper lumbar spine. Additionally, age and bone mass are the main factors associated with vertebral endplate defects.