Comparative Analysis of Outcomes in Adult Spinal Deformity Patients with Proximal Junctional Kyphosis or Failure Initially Fused to Upper Versus Lower Thoracic Spine

Abstract

Background: Patients with proximal junctional kyphosis (PJK) or failure (PJF) may demonstrate disparate outcomes and recovery when fused to the upper (UT) versus lower (LT) thoracic spine. Few studies have distinguished the reoperation and recovery abilities of patients with PJK or PJF when fused to the upper (UT) versus lower (LT) thoracic spine. Methods: Adult spine deformity patients ≥ 18 yrs with preoperative and 5-year (5Y) data fused to the sacrum/pelvis were included. The rates of PJK, PJK revision, and radiographic PJF were compared between patients with upper instrumented vertebra (UIV) in the upper thoracic spine (UT; T1-T7) and lower thoracic spine (LT; T8-L1). Mean differences were assessed via analyses of covariance, factoring in any differences between cohorts at baseline and any use of PJF prophylaxis. Backstep logistic regressions assessed predictors of achieving Smith et al.'s Best Clinical Outcomes (BCOs) and complications, controlling for similar covariates. Results: A total of 232 ASD patients were included (64.2 ± 10.2 years, 78% female); 36.3% were UT and 63.7% were LT. Postoperatively, the rates of PJK for UT were lower than LT at 1Y (34.6 vs. 50.4%, p = 0.024), 2Y (29.5 vs. 49.6% (p = 0.003), and 5Y (48.7 vs. 62.8%, p = 0.048), with comparable rates of PJF. In total, 4.0% of UT patients underwent subsequent reoperation, compared to 13.0% of LT patients (p = 0.025). A total of 6.0% of patients had recurrent PJK, and 3.9% had recurrent PJF (both p > 0.05). After reoperation, UT patients reported higher rates of improvement in the minimum clinically important difference for ODI by 2Y (p = 0.007) and last follow-up (p < 0.001). While adjusted regression revealed that, for UT patients, the minimization of construct extension was predictive of achieving BCOs by last follow-up (model p < 0.001), no such relationship was identified in LT patients. Conclusions: Patients initially fused to the lower thoracic spine demonstrate an increased incidence of PJK and lower rates of disability improvement, but are at a lessened risk of neurologic complications if reoperation is required.

Keywords: adult spine deformity; proximal junctional failure; proximal junctional kyphosis; realignment; spine fusion.

Figure 1

Example UT patient. A 63-year-old woman with a body mass index of 20.6 kg/m² and presenting with chronic back and leg pain. She had a past medical history of gastroesophageal reflux disease and diabetes. She underwent posterior spinal fusion from T6 to pelvis and L3-S1 interbody fusion. (A) Baseline sagittal plain film radiograph. (B) Six-week postoperative sagittal radiograph. (C) One-year postoperative sagittal radiograph denoting mild proximal junctional kyphosis (PJK). (D) Five-year postoperative sagittal radiograph illustrating further progression of PJK which had also become symptomatic.

Figure 2

Example LT patient. A 64-year-old woman with a body mass index of 29.6 kg/m² and presenting with chronic back pain. She had a past medical history of hypertension and coronary artery disease. She underwent posterior spinal fusion from T11 to pelvis and L4-S1 interbody fusion. (A) Baseline sagittal plain film radiograph. (B) Six-week postoperative sagittal radiograph. (C) One-year postoperative sagittal radiograph illustrating proximal junctional kyphosis (PJK) which was clinically symptomatic. The patient subsequently underwent reoperation with fusion extension proximally to T2. (D) Five-year postoperative sagittal radiograph.