Abstract：
BACKGROUND: Our study presents a single-center experience of resection of intradural spinal tumors either with or without using intraoperative computed tomography-based registration and microscope-based augmented reality (AR). Microscope-based AR was recently described for improved orientation in the operative field in spine surgery, using superimposed images of segmented structures of interest in a two-dimensional or three-dimensional mode.
METHODS: All patients who underwent surgery for resection of intradural spinal tumors at our department were retrospectively included in the study. Clinical outcomes in terms of postoperative neurologic deficits and complications were evaluated, as well as neuroradiologic outcomes for tumor remnants and recurrence.
RESULTS: 112 patients (57 female, 55 male; median age 55.8 ± 17.8 years) who underwent 120 surgeries for resection of intradural spinal tumors with the use of intraoperative neuromonitoring were included in the study, with a median follow-up of 39 ± 34.4 months. Nine patients died during the follow-up for reasons unrelated to surgery. The most common tumors were meningioma (n = 41), schwannoma (n = 37), myopapillary ependymomas (n = 12), ependymomas (n = 10), and others (20). Tumors were in the thoracic spine (n = 46), lumbar spine (n = 39), cervical spine (n = 32), lumbosacral spine (n = 1), thoracic and lumbar spine (n = 1), and 1 tumor in the cervical, thoracic, and lumbar spine. Four biopsies were performed, 10 partial resections, 13 subtotal resections, and 93 gross total resections. Laminectomy was the common approach. In 79 cases, patients experienced neurologic deficits before surgery, with ataxia and paraparesis as the most common ones. After surgery, 67 patients were unchanged, 49 improved and 4 worsened. Operative time, extent of resection, clinical outcome, and complication rate did not differ between the AR and non-AR groups. However, the use of AR improved orientation in the operative field by identification of important neurovascular structures.
CONCLUSIONS: High rates of gross total resection with favorable neurologic outcomes in most patients as well as low recurrence rates with comparable complication rates were noted in our single-center experience. AR improved intraoperative orientation and increased surgeons\' comfort by enabling early identification of important anatomic structures; however, clinical and radiologic outcomes did not differ, when AR was not used.
METHODS: All patients who underwent surgery for resection of intradural spinal tumors at our department were retrospectively included in the study. Clinical outcomes in terms of postoperative neurologic deficits and complications were evaluated, as well as neuroradiologic outcomes for tumor remnants and recurrence.
RESULTS: 112 patients (57 female, 55 male; median age 55.8 ± 17.8 years) who underwent 120 surgeries for resection of intradural spinal tumors with the use of intraoperative neuromonitoring were included in the study, with a median follow-up of 39 ± 34.4 months. Nine patients died during the follow-up for reasons unrelated to surgery. The most common tumors were meningioma (n = 41), schwannoma (n = 37), myopapillary ependymomas (n = 12), ependymomas (n = 10), and others (20). Tumors were in the thoracic spine (n = 46), lumbar spine (n = 39), cervical spine (n = 32), lumbosacral spine (n = 1), thoracic and lumbar spine (n = 1), and 1 tumor in the cervical, thoracic, and lumbar spine. Four biopsies were performed, 10 partial resections, 13 subtotal resections, and 93 gross total resections. Laminectomy was the common approach. In 79 cases, patients experienced neurologic deficits before surgery, with ataxia and paraparesis as the most common ones. After surgery, 67 patients were unchanged, 49 improved and 4 worsened. Operative time, extent of resection, clinical outcome, and complication rate did not differ between the AR and non-AR groups. However, the use of AR improved orientation in the operative field by identification of important neurovascular structures.
CONCLUSIONS: High rates of gross total resection with favorable neurologic outcomes in most patients as well as low recurrence rates with comparable complication rates were noted in our single-center experience. AR improved intraoperative orientation and increased surgeons\' comfort by enabling early identification of important anatomic structures; however, clinical and radiologic outcomes did not differ, when AR was not used.
摘要：
背景：我们的研究提供了在有或没有使用基于术中CT（iCT）的配准和基于显微镜的增强现实（AR）的情况下切除硬膜内脊柱肿瘤的单中心经验。最近，基于显微镜的AR被描述为在脊柱外科手术中改善手术领域的定向。在二维(2D)或三维(3D)模式中使用感兴趣的分割结构的叠加图像。
方法：本研究将所有在我科接受手术切除硬膜内脊柱肿瘤的患者纳入回顾性研究。评估术后神经功能缺损和并发症的临床结果，以及肿瘤残留和复发的神经放射学结果。
结果：112名患者（57名女性，55男,中位年龄55.8±17.8岁），在使用术中神经监测的情况下，接受了120例硬膜内脊柱肿瘤切除术的患者被纳入研究，中位随访时间为39±34.4个月。9名患者在随访期间死亡，原因与手术无关。最常见的肿瘤是脑膜瘤（n=41），神经鞘瘤（n=37），肌乳头状室管膜瘤（n=12），室管膜瘤（n=10），其他（20）肿瘤位于胸椎（n=46），腰椎（n=39），颈椎（n=32），腰骶椎（n=1），胸腰椎（n=1）和一个颈部肿瘤，胸廓,和腰椎。进行了四次活检，十个部分切除，13次全切除，共切除93次。椎板切除术是常见的方法。在79个案例中，患者在手术前出现神经功能缺损，共济失调和轻瘫是最常见的。手术后,67例没有改变，49例有所改善，4例恶化。手术时间,切除范围,AR组和非AR组的临床结局和并发症发生率无差异.然而,通过识别重要的神经血管结构，在手术领域中使用AR改善了方向。
结论：在我们的单中心经验中指出，在绝大多数患者中，GTR的发生率高，神经系统预后良好，复发率低，并发症发生率相当。AR通过早期识别重要的解剖结构，改善了术中定位并增加了外科医生的舒适度。然而，临床和放射学结果没有差异，没有使用AR。
方法：本研究将所有在我科接受手术切除硬膜内脊柱肿瘤的患者纳入回顾性研究。评估术后神经功能缺损和并发症的临床结果，以及肿瘤残留和复发的神经放射学结果。
结果：112名患者（57名女性，55男,中位年龄55.8±17.8岁），在使用术中神经监测的情况下，接受了120例硬膜内脊柱肿瘤切除术的患者被纳入研究，中位随访时间为39±34.4个月。9名患者在随访期间死亡，原因与手术无关。最常见的肿瘤是脑膜瘤（n=41），神经鞘瘤（n=37），肌乳头状室管膜瘤（n=12），室管膜瘤（n=10），其他（20）肿瘤位于胸椎（n=46），腰椎（n=39），颈椎（n=32），腰骶椎（n=1），胸腰椎（n=1）和一个颈部肿瘤，胸廓,和腰椎。进行了四次活检，十个部分切除，13次全切除，共切除93次。椎板切除术是常见的方法。在79个案例中，患者在手术前出现神经功能缺损，共济失调和轻瘫是最常见的。手术后,67例没有改变，49例有所改善，4例恶化。手术时间,切除范围,AR组和非AR组的临床结局和并发症发生率无差异.然而,通过识别重要的神经血管结构，在手术领域中使用AR改善了方向。
结论：在我们的单中心经验中指出，在绝大多数患者中，GTR的发生率高，神经系统预后良好，复发率低，并发症发生率相当。AR通过早期识别重要的解剖结构，改善了术中定位并增加了外科医生的舒适度。然而，临床和放射学结果没有差异，没有使用AR。
