Introduction: The presence and severity of endplate abnormalities and/or Modic changes have been significantly associated with low back pain (LBP). Such endplate changes may manifest or be identified in childhood, occur in isolation or along with disc degeneration (DD) and/or facet joint changes (FJC). However, the risk factors or “drivers” of endplate changes remain debatable. Past studies have noted that spinopelvic alignment may be related to the development of various spine changes, pain and outcome. Pelvic incidence is a spinopelvic parameter that is inherent and a non-modifiable risk factor. However, nothing is known with respect to spinopelvic alignment in relation to structural endplate abnormalities, accounting for DD and FJC. Therefore, this study aimed to determine if spinopelvic parameters are key drivers of endplate abnormalities of the lumbar spine, in the setting of DD and FJC. To address this objective – a pediatric cohort, who is without the multitude of potential confounders related to spine changes that are often noted in adulthood – was analyzed.
Methods: An ambispective study was performed of 190 consecutive pediatric patients who presented with LBP at a single center. Patients with deformity, tumors, infections, and inflammatory conditions of the spine as well as previous surgery were excluded. Lateral plain radiographs of the lumbar spine (L1-S1) and pelvis were assessed to determine the sagittal alignment of the following parameters: pelvic tilt, sacral slope, pelvic incidence (PI), lumbar lordosis (LL), and pelvic incidence-lumbar lordosis (PI−LL) mismatch. MRI phenotypes (L1-S1) were assessed for the presence of structural endplate abnormalities, based on the total endplate damage score (TEPS) by Rajasekaran et al, which included the presence of irregularities, Modic changes, and geometric variations among other criteria. Disc degeneration (DD) was assessed according to the Pfirrmann grading scale, while a cumulative lumbar DD severity score was obtained. Bilateral FJC at each motion segment were evaluated, accounting for various phenotypes (i.e. osteophytes, sclerosis). Severity threshold values of TEPS, DD and FJC were based on their respective mean values in this cohort. Patient demographics, such as age, sex, and body mass index (BMI), were also noted.
Results: The mean age of the included patients was 16.7±2.3 years (mean BMI: 24.8±5.5 kg/m2), 55.2% of them were female. 45.3% of patients had a TEPS severity score >11. Adjusting for patient demographics, multiple regression analysis noted that risk factors for endplate abnormalities were PI-LL>10° (OR:17.5, 95% CI:6.98-43.6, p<0.001), lower PI values (OR:0.97, 95% CI:0.94-1.003, p=0.077), DD severity (OR:2.72, 95% CI:1.09-6.79, p=0.032) and FJC severity (OR:0.55, 95% CI:0.23-1.35, p=0.19).
Discussion: Our study is the “first” to note the presence of “spinopelvic alignment-drivers” of lumbar endplate changes and severity in humans, by approximately an 18-fold increased risk. This association was “independent” of DD and/or FJC of the motion segment, phenotypes that have been traditionally believed to be the main drivers of endplate changes. Our study presents a “paradigm shift” as to the development of endplate changes, which further sheds light into understanding which individuals may be predisposed to vertebrogenic origins of LBP with biomechanical and developmental footprints.