Department of General, Abdominal, and Transplant Surgery, Heidelberg University, Im Neuenheimer Feld 110, 69120, Heidelberg, Germany,
Surg Endosc. 2013 Oct;27(10):3663-70. doi: 10.1007/s00464-013-2941-4. Epub 2013 Apr 3.
Navigation systems potentially facilitate minimally invasive esophagectomy and improve patient outcome by improving intraoperative orientation, position estimation of instruments, and identification of lymph nodes and resection margins. The authors' self-developed navigation system is highly accurate in static environments. This study aimed to test the overall accuracy of the navigation system in a realistic operating room scenario and to identify the different sources of error altering accuracy.
To simulate a realistic environment, a porcine model (n = 5) was used with endoscopic clips in the esophagus as navigation targets. Computed tomography imaging was followed by image segmentation and target definition with the medical imaging interaction toolkit software. Optical tracking was used for registration and localization of animals and navigation instruments. Intraoperatively, the instrument was displayed relative to segmented organs in real time. The target registration error (TRE) of the navigation system was defined as the distance between the target and the navigation instrument tip. The TRE was measured on skin targets with the animal in the 0° supine and 25° anti-Trendelenburg position and on the esophagus during laparoscopic transhiatal preparation.
On skin targets, the TRE was significantly higher in the 25° position, at 14.6 ± 2.7 mm, compared with the 0° position, at 3.2 ± 1.3 mm. The TRE on the esophagus was 11.2 ± 2.4 mm. The main source of error was soft tissue deformation caused by intraoperative positioning, pneumoperitoneum, surgical manipulation, and tissue dissection.
The navigation system obtained acceptable accuracy with a minimally invasive transhiatal approach to the esophagus in a realistic experimental model. Thus the system has the potential to improve intraoperative orientation, identification of lymph nodes and adequate resection margins, and visualization of risk structures. Compensation methods for soft tissue deformation may lead to an even more accurate navigation system in the future.
导航系统通过改善术中定位、器械位置估计以及淋巴结和切缘的识别，有可能促进微创食管切除术并改善患者预后。作者自主开发的导航系统在静态环境下具有高度准确性。本研究旨在测试导航系统在真实手术室环境中的整体准确性，并确定改变准确性的不同误差源。
为了模拟真实环境，使用猪模型（n=5），在食管中使用内镜夹作为导航目标。进行计算机断层扫描成像后，使用医学成像交互工具软件进行图像分割和目标定义。光学跟踪用于注册和定位动物和导航器械。术中，实时显示器械相对于分割器官的位置。导航系统的目标注册误差（TRE）定义为目标与导航器械尖端之间的距离。在动物处于 0°仰卧位和 25°反 Trendelenburg 位时，在皮肤上测量 TRE，在腹腔镜经食管裂孔入路准备时，在食管上测量 TRE。
在皮肤上，25°位置的 TRE 明显高于 0°位置，分别为 14.6±2.7mm 和 3.2±1.3mm。在食管上的 TRE 为 11.2±2.4mm。误差的主要来源是术中定位、气腹、手术操作和组织解剖引起的软组织变形。
在真实实验模型中，微创经食管裂孔入路的导航系统具有可接受的准确性。因此，该系统有可能改善术中定位、识别淋巴结和适当的切缘以及可视化风险结构。软组织变形的补偿方法可能会使未来的导航系统更加精确。
