Differences at Index Surgery in Operative Complexity and Residual Disease for Earlier and Later Repeat Lumbar Surgery.

Abstract

Retrospective review, single-institution cohort studies. To compare patients with earlier (i.e. <1.5y) and later (i.e. >1.5y) repeat lumbar surgery to patients with no repeat surgery according to clinical characteristics at index surgery. Grouping patients as earlier or later repeat surgery may reveal different associations when compared to patients with no repeat surgery. Patients undergoing index surgery for diverse conditions reported preoperative demographic/clinical variables, including comorbidity and depressive symptoms. Extent (i.e. complexity) of surgery was assigned based on a valid index that included decompression, fusion and instrumentation. Co-existing disease at non-operated levels was ascertained from imaging reports. Postoperative records of all medical visits up to the time of this study (12y) were reviewed for repeat surgery. Patients were grouped as earlier (<1.5y) or later surgery (≥1.5y) and compared to patients with no repeat surgery in separate multivariable analyses. Among 1,334 patients (51% men, mean age 59), 82% did not have repeat surgery, 7% had earlier and 11% had later repeat surgery. Compared to no surgery, earlier surgery was associated with more comorbidity (OR 1.7, CI 1.1-2.6, P=0.02), positive depression screen (OR 1.9, CI 1.2-2.9, P=0.006), opioid use (OR 1.8, CI 1.2-2.8, P=0.008), and greater extent of index surgery (OR 1.1, CI 1.0-1.1, P=0.0009). Compared to no surgery, later surgery was associated with pre-index lumbar surgery (OR 1.9, CI 1.3-2.8, P=0.0005) and disease at non-operated levels at index surgery (OR 1.6, CI 1.0-2.4, P=0.05). Earlier surgeries were more likely to involve only the same vertebra as index surgery (51% vs. 16%) and later surgeries were more likely to involve only other levels (5% vs. 36%, P=0.01). Earlier and later repeat lumbar surgeries differed in complexity and residual disease compared to no repeat surgery. These findings have implications for patient counseling regarding short and long-term postoperative spine health.