Postoperative changes in measured pelvic incidence after thoracolumbar fusion: a retrospective cohort study

Retrospective cohort. Pelvic incidence (PI) has been regarded as a fixed anatomical parameter; however, accumulating evidence challenges this assumption. While postoperative PI variability has been reported in adult spinal deformity populations, it remains unclear whether similar variation occurs in adolescents and young adults after limited thoracolumbar fusion. This study investigated the frequency and associated factors of postoperative PI change in this demographic. Furthermore, we conducted an exploratory analysis to assess whether such alterations are associated with the development of proximal junctional kyphosis (PJK). We retrospectively reviewed patients with adolescent idiopathic scoliosis or Scheuermann kyphosis who underwent posterior thoracolumbar instrumentation and fusion, with the upper instrumented vertebra proximal to T5 and the lower instrumented vertebra between L1 and L4, and with at least 2 years of follow-up. PI change was defined as a postoperative difference of at least 7° between the last preoperative standing lateral radiograph and the first early postoperative standing lateral radiograph. Among 85 patients, 9 (10.6%) demonstrated a postoperative PI change of at least 7°. Patients with PI changed had lower baseline PI and lower preoperative sacral slope. Extension of the construct to L4 was also associated with PI variability. In the secondary analysis, the PI-change group showed a lower incidence of PJK, but this difference was not statistically significant. Our findings suggest that measured standing PI may vary following thoracolumbar fusion, particularly in patients with low baseline PI and longer constructs. Because the study is retrospective and the sample size is limited, these findings should be interpreted as exploratory and hypothesis-generating. Further prospective research is required to clarify whether the observed changes reflect true anatomical alteration, postoperative positional adaptation, or measurement variability. Not applicable.