Comparing Three and Four-Rod Constructs in Adult Spinal Deformity Patients With Long Fusions to the Pelvis: Is There a Difference in Rod Fractures and Revision Rates at Minimum 2-Year Follow-up?

Abstract

Single-center retrospective cohort study. To determine if the 4-rod construct (4RC) is protective against the occurrence of rod fractures when compared with the 3-rod construct (3RC) in adult spinal deformity (ASD) patients with long fusions to the sacrum. Past studies have explored the different outcomes in subjects with dual-rod versus multirod constructs. There is a lack of literature distinguishing the ramifications of 3RC versus 4RC, particularly in the prevalence of rod fractures and rod fractures requiring subsequent revision surgery as a result of pseudarthrosis. ASD patients undergoing long instrumented fusions to the sacrum were dichotomized between the 3RC and 4RC cohorts. Outcomes of interest include the occurrence of rod fractures (RFs) and RFs requiring revision (RFR). Two-tailed independent-sample t test with Welch's correction and χ2/Fisher exact test were used for continuous and categorical variables, respectively. Multivariable logistic regression analysis was performed to assess whether the 4RC is protective against rod fractures when compared with the 3RC. One hundred forty-five patients with a minimum 2-year follow-up were included (3RC=57, 4RC=88). Four RC had a greater body mass index (BMI) (P=0.002), longer operating room (OR) time (P=0.002), greater estimated blood loss (EBL) (P=0.002), total instrumented levels (TIL) (P=0.028), and more 3-column osteotomies (3COs) performed (P=0.028). Four RC had greater baseline coronal vertical axis (CVA) (28.2±24.9 vs. 18.5±16.9 mm, P=0.006) and sagittal vertical axis (SVA) (55.1±64.8 vs. 30.4±41.8 mm, P=0.006). Three RC had a greater rate of RF (28.1% vs. 14.8%, P=0.0506) and RFR (14.9% vs. 10.2%, P=0.486) when compared with 4RC, but these findings were not statistically significant. After controlling for BMI, the use of 3-column osteotomies, total instrumented levels, change in coronal alignment (∆CVA), change in sagittal alignment (∆SVA), use of bone morphogenetic protein, and number of interbody fusions, 3RC was associated with 4.93× greater odds of experiencing RFs (P=0.0078). However, significance fell short when adjusting for the occurrence of RFR [OR=2.58 (0.60-11.19), P=0.2067]. ASD patients having long fusions to the sacrum with 4RC across the lumbosacral junction are shown to be at lesser risk of developing rod fractures but not revision surgery as compared with 3RC at 2-year follow-up.