Comparison of a right versus left sided anterior approach to the thoracic and thoracolumbar spine: A case series.

Aims & Objectives: Traditionally, a left sided anterior approach has been considered the safest way to access the thoracic and thoracolumbar spine due to avoidance of the inferior vena cava (IVC). We challenge this 'go from the left' dogma by presenting a series of patients who underwent a right sided approach to the thoracic and thoracolumbar spine.

Materials & Methods: 71 anterior thoracic fusion procedures treated with either a left or right sided approach were identified and retrospectively studied. Demographic information, indication for surgery, intra-operative time, estimated blood loss (EBL), hospital length of stay (LOS), and complications were collected. Patients were further sub divided into groups according to indication for surgery.

Results: 57 cases with a right sided approach and 14 cases with a left sided approach were investigated. Four mortalities occurred, all within the right sided approach group among patients who were being for osteomyelitis/discitis. There were statistically significant differences between right and left sided approach groups for operative time and EBL when patients were analysed together favoring the left sided approach. However once subgroup analysis was performed, there were no statistically significant differences between right and left sided approach groups aside from operative time in the HNP group and LOS in the fracture group. Neither group had catastrophic intra-operative vascular injury. Perioperative complications occurred in 22.8% of right sided approach patients and 14% of left sided approach patients.

Conclusions: Our results reinforce the fact that anterior thoracic fusion is a major surgical endeavor that is not without risk. Although there were differences in operative time and EBL between right and left sided approach groups when all patients were analysed together, these were no longer significant with subgroup analysis according to indication for surgery with two exceptions. This may be in part due to low sample size or confounding variables related to indication.