Spinopelvic alignment and pubic symphysis widening are associated with postpartum low back pain: a retrospective case–control study

Background Postpartum low back pain (PLBP) is a frequent but often under-recognized condition that may be related to pregnancy- and delivery-induced biomechanical alterations of the lumbopelvic complex. However, imaging-based evidence linking postpartum lumbopelvic morphology to PLBP remains limited. Methods This retrospective case–control study analyzed postpartum women (≤1 year after delivery) treated in a rehabilitation department between January 2022 and October 2023. Participants were classified into a PLBP group (n = 84) and a non-PLBP (NPLBP) group (n = 73) based on clinical assessment following guideline-based diagnostic criteria and exclusion of alternative causes. Standard weight-bearing standing anteroposterior and lateral radiographs were used to quantify lumbopelvic parameters in the coronal (Lumbar Cobb Angle, Pelvic Obliquity, Pubic Symphysis Width), transverse (Iliac Width Ratio), and sagittal planes (Pelvic Incidence, Pelvic Tilt, Sacral Slope, Lumbar Lordosis). Group comparisons were performed, and univariable and multivariable logistic regression analyses were used to identify radiographic factors independently associated with PLBP. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC) and the Hosmer–Lemeshow test. Results Baseline characteristics (age, body mass index, postpartum time, delivery mode, gravidity, and pelvic floor dysfunction) did not differ significantly between groups. Compared with the NPLBP group, the PLBP group exhibited significantly greater lumbar Cobb angle (3.82 ± 2.95° vs. 1.85 ± 1.91°), pubic symphysis width (6.85 ± 2.58 mm vs. 5.15 ± 1.64 mm), pelvic incidence (50.58 ± 8.94° vs. 46.38 ± 9.03°), sacral slope (38.01 ± 6.10° vs. 34.34 ± 7.31°), and lumbar lordosis (55.17 ± 9.58° vs. 47.15 ± 9.08°). In multivariable analysis, lumbar Cobb angle (OR 1.31 per 1°, 95% CI 1.10–1.56), lumbar lordosis (OR 1.08 per 1°, 95% CI 1.03–1.14), and pubic symphysis width (OR 1.47 per 1 mm, 95% CI 1.20–1.80) remained independently associated with PLBP. The multivariable model showed acceptable apparent discriminatory ability (AUC 0.829) and goodness-of-fit (Hosmer–Lemeshow P = 0.608). Conclusion Coronal lumbar asymmetry, increased lumbar lordosis, and greater pubic symphysis width were independently associated with PLBP. These radiographic features may reflect altered lumbopelvic load transfer and pelvic ring stability and could inform postpartum musculoskeletal assessment and stabilization-oriented rehabilitation strategies. Prospective longitudinal studies are warranted to clarify temporality and causality.