Peri-Operative Complications in Revision Anterior Lumbar Spine Surgery: Incidence and Risk Factors

Methods: 129 consecutive revision surgeries (1998 - 2003) in 108 patients (70% female; average age 50.6 years) were retrospectively reviewed. Peri-operative complications were recorded. Revisions were either at the same vertebral levels (pseudarthrosis; 70 surgeries) or at adjacent levels (59 surgeries). Original implants were most commonly tricortical allograft (80%), cages in 8% and femoral ring allograft in 6%. Number of levels treated was similar between groups (1-level 69%; 2-level 19%; three or more 12%). Revision cases were more commonly L4-L5 and/or L5-S1; extensions were more likely upper lumbar spine.

Results: Revision cases (same operative level) had a higher overall complication rate (42%) compared to extensions (20%; p=0.007). Complications included: vein lacerations, ileus, peritoneal lacerations, infections. This difference was primarily due to vein lacerations (23.7% vs 3.6%, p=0.002). There were no arterial tears; few dural tears (3.5%) and one ureter injury (0.9%). The rate of peritoneal lacerations was similar between groups (Revisions: 22.0%; Extension: 17.9%). There were no intraoperative deaths or permanent neurologic complications.

Revisions at same level were associated with longer operative time (p=0.004) and higher blood loss (p=0.003). Blood loss greater than 1000cc occurred in 12% of extensions and 17% of revision cases (p>0.05).

Discussion: Complication rates for revision lumbar surgery in this series are three to five times higher than reported non-revision lumbar exposures. Complication rates are significantly higher for revision anterior lumbar cases compared to anterior extensions. All revision surgeries should be undertaken with proper planning and availability of specialized co-surgeons.

Complication Rate by Revision Type

-Comparison between Extension and Revision
-NS: Not statistically significant, p>0.
-05.

Blood Loss & OR Time by Revision Type