Postoperative pelvic incidence (PI) change may impact sagittal spinopelvic alignment (SSA) after instrumented surgical correction of adult spine deformity (ASD).

Abstract

To study factors causing postoperative change of PI after surgical correction of ASD and to assess the effect of this variability on postoperative PI-LL mismatch. PI is used as an individual constant to define lumbar lordosis (LL) correction goal (PI-LL < 10). Postoperative changes of PI were shown but with opposite vectors. The impact of the PI variability on the postoperative PI-LL has not been studied. The medical and radiographic data analyzed for patients who underwent long posterior instrumented spinal fusion. Inclusion criteria are age, ≥ 20 years old; ASD due to degenerative disk disease (DDD) or scoliosis (DS); ≥ 3 levels fused; and 2-year follow-up or revision. Studied parameters are LL (L1-S1), PI, sacral slope (SS), pelvic tilt (PT), and PI-LL. Measurement error and postoperative changes were defined. Statistical analysis includes ANOVA, correlation, regression, and risk assessment by odds ratio; P ≤ 0.05 considered statistically significant. Eighty patients were included: mean age, 62.4 years-old (SD, 11.1); female, 63.7%; mean body mass index (BMI), 27.1 (SD, 5.6). Distribution of patients by follow-ups includes preoperative 100%; postoperative (1-3 weeks), 100%; 11-13 months. 90%; 22-26 months, 58%; and revision: 24%. Pre- versus postoperative PI (∆PI) changed both positively and negatively and the absolute value of change|∆PI| exceeded measurement error (P ≤ 0.05) reaching as high as 31°, and progressed with time; R² dropped from 0.73 to 0.45 (P < 0.001); ∆PI depended on disproportional changes of SS and PT, preoperative PI, and change of LL. Obesity, DS, and absence of sacroiliac fixation increased |∆PI|. The risk of LL insufficient correction (PI-LL > 10°) associated with a |∆PI|> 6°, P = 0.05. Sacroiliac fixation diminished PI variability only during the first postoperative year. Preoperative variability and postoperative instability of PI diminish the applicability of the PI-LL < 10° goal to plan correction of LL. An alternative method is offered. IV.