Failure of lumbopelvic fixation after long construct fusions in patients with adult spinal deformity: Clinical and radiographic risk factors

Woojin Cho, Jonathan R. Mason, Justin S. Smith, Adam L. Shimer, Adam S. Wilson, Christopher I. Shaffrey, Francis H. Shen, Wendy M. Novicoff, Kai Ming G. Fu, Joshua E. Heller, Vincent Arlet

Research output: Contribution to journalArticlepeer-review

81 Scopus citations


Object. Lumbopelvic fixation provides biomechanical support to the base of the long constructs used for adult spinal deformity. However, the failure rate of the lumbopelvic fixation and its risk factors are not well known. The authors' objective was to report the failure rate and risk factors for lumbopelvic fixation in long instrumented spinal fusion constructs performed for adult spinal deformity. Methods. This retrospective review included 190 patients with adult spinal deformity who had long construct instrumentation (< 6 levels) with iliac screws. Patients' clinical and radiographic data were analyzed. The patients were divided into 2 groups: A failure group and a nonfailure group. A minimum 2-year follow-up was required for inclusion in the nonfailure group. In the failure group, all patients were included in the study regardless of whether the failure occurred before or after 2 years. In both groups, the patients who needed a revision for causes other than lumbopelvic fixation (for example, proximal junctional kyphosis) were also excluded. Failures were defined as major and minor. Major failures included rod breakage between L-4 and S-1, failure of S-1 screws (breakage, halo formation, or pullout), and prominent iliac screws requiring removal. Minor failures included rod breakage between S-1 and iliac screws and failure of iliac screws. Minor failures did not require revision surgery. Multiple clinical and radiographic values were compared between major failures and nonfailures. Results. Of 190 patients, 67 patients met inclusion criteria and were enrolled in the study. The overall failure rate was 34.3%; 8 patients had major failure (11.9%) and 15 had minor failure (22.4%). Major failure occurred at a statistically significant greater rate in patients who had undergone previous lumbar surgery, had greater pelvic incidence, and had poor restoration of lumbar lordosis and/or sagittal balance (that is, undercorrection). Patients with a greater number of comorbidities and preoperative coronal imbalance showed trends toward an increase in major failures, although these trends did not reach statistical significance. Age, sex, body mass index, smoking history, number of fusion segments, fusion grade, and several other radiographic values were not shown to be associated with an increased risk of major failure. Seventy percent of patients in the major failure group had anterior column support (anterior lumbar interbody fusion or transforaminal lumbar interbody fusion) while 80% of the nonfailure group had anterior column support. Conclusions. The incidence of overall failure was 34.3%, and the incidence of clinically significant major failure of lumbopelvic fixation after long construct fusion for adult spinal deformity was 11.9%. Risk factors for major failures are a large pelvic incidence, revision surgery, and failure to restore lumbar lordosis and sagittal balance. Surgeons treating adult spinal deformity who use lumbopelvic fixation should pay special attention to restoring optimal sagittal alignment to prevent lumbopelvic fixation failure.

Original languageEnglish (US)
Pages (from-to)445-453
Number of pages9
JournalJournal of Neurosurgery: Spine
Issue number4
StatePublished - Oct 2013
Externally publishedYes


  • Adult
  • Complication
  • Deformity
  • Iliac screws
  • Instrumentation
  • Pseudarthrosis
  • Spine

ASJC Scopus subject areas

  • Surgery
  • Neurology
  • Clinical Neurology


Dive into the research topics of 'Failure of lumbopelvic fixation after long construct fusions in patients with adult spinal deformity: Clinical and radiographic risk factors'. Together they form a unique fingerprint.

Cite this