RESUMEN
BACKGROUND: Three-dimensional (3-D) endoscopic optics use 2 cameras to simulate the different perspectives of the right and left eye, creating the illusion of spatial depth. Optimised orientation as well as improved hand-eye coordination compared to 2-D-optics could be proven in standardised test setups (black box) and in laparoscopic use. This retrospective study examines whether these results can also be applied to thoracoscopic vertebral body replacement at the thoracolumbar junction. HYPOTHESES: 1. Ventral vertebral body replacement using 3-D-thoracoscopy results in a shorter operation time than with 2-D-thoracoscopy. 2. Perioperative blood loss is less, due to better spatial orientation (faster haemostasis) and reduced tissue laceration. MATERIAL AND METHODS: 29 patients met the inclusion criteria of this retrospective study. Between 08â-â2012 and 08â-â2017, all of these received ventral thoracoscopic vertebral replacement at the thoracolumbar junction (Th11 to L2). Patients with additional anterior procedures (e.g. anterolateral plate) were excluded. Perioperative data such as blood loss, duration of surgery and length of hospital stay were analysed. Conventional 2-D-optics were used in n = 14 patients and 3-D-optics in 15 patients. Aesculap EinsteinVision® 2.0 was used as the 3-D-optics. Statistical significance was calculated using Student's t-test. RESULTS: The most common diagnosis was a L1 fracture (n = 18, 62%). Mean OR time was 24 minutes shorter in the 3-D group (149 ± 29, 107â-â198 min) than in the 2-D group (173 ± 39, 125â-â260 min), but this difference was not significant. Total perioperative blood loss in the 3-D group was significantly lower than in the 2-D group (**p = 0.043). Proportional intraoperative blood loss in the 3-D group was also lower (mean around 115 ml), but not significantly so. Significantly lower values were found for the delivery rate of the thoracic drainage in the 3-D group (248 vs. 560 ml, *p = 0.195). Inpatient stay with the 3-D group was on average 1.5 days (d) shorter (8.7 d for the 3-D group, 10.2 d for the 2-D group) but this difference was not significant (p = 0.27). CONCLUSION: Thoracoscopic-assisted vertebral body replacement at the thoracolumbar junction is a safe and reliable surgical procedure using conventional 2-D-optics or the new 3-D-optics. Both methods allow thoracoscopic vertebral body replacement in comparable operation times but in our study the 3-D group presented with significantly lower postoperative blood loss. Due to the small number of cases and because of the retrospective design, the present study is considered to be a pilot study only.