After obtaining institutional ethical clearance, a retrospective study was conducted in which data of patients with spine tumors who underwent navigation-assisted resection was collected and reviewed. From 2017 to 2019, a total of seven patients underwent navigation-assisted resection for primary spine tumors in five patients and solitary metastasis to spine in two patients. The baseline demographic characteristics were recorded including age, gender, type of tumor, location, neurological grade as per Frankel grading and any adjuvant treatment. All patients underwent diffusion weighted Magnetic Resonance Imaging (MRI) and Positron Emission Computed Tomography (PET-CT) scanning of the spine preoperatively to determine the margins of resection. The aim was to achieve the widest bone and soft tissue margin possible while maintaining critical anatomic structures for all patients considering functional consequences of additional margin resection. However, no specific minimum margin distance was used for these patients. The specific margin distance was determined by the discretion of a multidisciplinary cancer team of our institution and the judgement of the treating surgeon based on the predicted behavior of the tumor.

Surgical Procedure - Patients underwent surgical treatment with the intent to achieve a negative margin resection. After general anaesthesia, Patient was placed prone on radiolucent Wilson frame and pressure points were padded. Midline incision at planned surgical site was given and dissection was done to expose the posterior elements of desired levels. Stealth Station Spine Surgery Imaging and Surgical Navigation with O-arm (Medtronic^©, Minneapolis, MN, USA) was used. Reference clamp was placed on spinous process not being resected but readily visible to the camera and ensured not to move for the entirety of the procedure for accurate stereotaxy. After registering the reference frame and spine instruments, O-arm takes 360-degree image of the patient and transfers the image to stealth station that helps in providing a real time 3-Dimensional map with stereotactic imaging for guiding the registered spine instruments. The navigation system was used to guide all osteotomies made in whatever manner the surgeon felt was most appropriate for that particular patient. The surgeon checked intraoperative navigation image to confirm whether the surgical resection was sufficient with navigation guide (Fig. 1). A postoperative CT was performed to compare with the resection margin planned preoperatively. Surgical margins for both bone and soft tissue were measured for the resection specimen.

Fig. 1
A representative case of navigation-assisted surgery. (A, B) A 41-year-old male patient had a tumor on the posterior aspect of the vertebral body of T9, which showed hot uptake on a positron emission tomography–computed tomography scan. (C) Intraoperative image showing that the navigation array was attached to the osteotome. A Stealth Station™ S8 surgical navigation system enabled us to precisely track the location of osteotomy. A probe (yellow arrow) connected to the navigation system in real time made it possible to confirm the resection margin that had been determined preoperatively. (D) Postoperative computed tomography after resection and reconstruction with strut bone and instrumentation.

• Click for larger image
• Download as PowerPoint slide

Surgical data was recorded including approach of surgery, instrumentation levels, intra-operative blood loss, operative time, intra-operative (navigation related) or post-operative complication (neurological deterioration/ wound/ implant related). Oncological data included resection margins as per histopathology report, local recurrence, any metastasis, and disease status in follow-up period.

Out of Seven patients who underwent navigation-assisted tumor resection, two patients (28.6%) were males while five patients (71.4%) were females. The mean age was 50.28±17.95 years (range, 16-77 years). Five of seven patients had primary spine tumors while two patients had metastatic tumors to spine secondary to breast carcinoma (n=1) and gastric adenocarcinoma (n=1). Five patients had intact neurology (Frankel grade E) while two patients had incomplete neurological deficit (Frankel grade D and Grade C). Four patients received neo-adjuvant therapy (Chemotherapy, n=1; Radiotherapy, n=1; Chemo +Radiotherapy, n=1; Hormonal therapy, n=1). Table 1 shows the baseline demographic characteristics of patients.

Table 1
Baseline Demographic characteristics of patients

• Click for larger image
• Click for full table
• Download as Excel file

All patients underwent wide excision for the lesion – piecemeal total excision in two patients and en bloc excision in rest five patients. Posterior approach was most commonly used (n=4), Combined Anterior +Posterior approach was used in two patients while Anterior approach was used in one patient to resect anterior mediastinal mass with spine tumor. Four of these seven patients underwent reconstruction in same setting after resection. No intra-operative navigation related problem was encountered during the procedure. In post-operative period, one patient developed superficial Surgical site Infection, one patient had neurological deterioration while one patient had severe radiating pain (Table 2).

Table 2
Surgical procedures and outcomes of each patients

• Click for larger image
• Click for full table
• Download as Excel file

Mean follow-up time was 33.42±17.92 months after surgery (range, 6-65 months). In terms of oncological outcomes (Table 3), six out of seven patients (85.7%) survived with an NED state (n=5) or an AWD state (n=1) while one patient died after 6 months (metastasis D5 secondary to gastric carcinoma). Navigation helped in achieving negative resection margins in 6/7 (85.7%) patients (Fig. 2). Only one patient (metastatic tumor D11 secondary to breast carcinoma) had a close margin which was below 0.1 cm in the soft tissue however, there was no recurrence detected in follow-up. At the last follow-up, there was no evidence of local recurrence in 5/6 surviving patients (83.3%) while only one patient (with sacrum osteosarcoma) was diagnosed with local recurrence at seven months follow-up due to a satellite lesion who had a negative margin at the initial resection). Local recurrence resulted in a revisional wide excision and reconstruction (Fig. 3). In terms of reconstruction, there was no mal-positioned screw or early mechanical failure during the follow-up.

Fig. 2
Microscopically, a representative tumor lesion (asterisk) and surrounding fibrous tissues were identified. The tumor was composed of hyaline matrix and lobules of mature chondrocytes without significant atypia (hematoxylin and eosin, ×200 magnification).

• Click for larger image
• Download as PowerPoint slide

Fig. 3
The patient had a negative surgical margin at surgery with O-arm navigation (A, B), but recurrence due to activation of the satellite lesion (white arrow) occurred after surgery (C). Total sacrectomy and partial resection of the ilium were performed, and a 3D-printed sacrum-replacing prosthesis was inserted (D).

• Click for larger image
• Download as PowerPoint slide

Table 3
Evaluation of surgical margins and oncological outcomes of each patients

• Click for larger image
• Click for full table
• Download as Excel file

In the recent decades, there have been immense advances in Imaging and technology in the field of spine surgery to improve the precision, accuracy and safety of complex spine surgeries. Computer assisted navigation is one of the technology that has proved be effective ad useful in spine surgery. It has proved to improve the precision and accuracy of pedicle screw insertion, reduce radiation exposure to the surgeons and patient, and has allowed to expand the field of minimally invasive surgeries.10, 11) Spine tumor represents some of the most challenging issue in orthopedic surgery. Major neurovascular structures and major organs existing around the spine and complex anatomical structure of the spine itself make it difficult to excise the tumor in the spine. Because of this complexity, local recurrence rate is high due to incomplete resection.3) Recently, many studies have been conducted on navigation-assisted surgery for assessing resection margins in musculoskeletal tumors.4, 5) We conducted this study to find out the pathological results of patients who underwent navigation-assisted surgeries for spine tumor resection in our institute and the proportion of local recurrence.

Achieving adequate surgical margins is emphasized by the fact that local recurrence rate is highly decided by the surgical margins. By using standard surgical techniques, Jeys et al12) in a study of primary tumors of pelvis reported the rate of intralesional resection to be 29%, with a local recurrence rate of 27% While in other study, Jeys et al13) reported Intralesional resection rate of 8.7% and local recurrence rate of 13% in patients with pelvis and sacral tumors who underwent navigation assisted resection. These serial studies have demonstrated that navigation-assisted surgery is safe and that it can decrease intralesional resection rate and local recurrence for primary tumors of the pelvis and sacrum. Bosma et al8) in a retrospective study on pelvic and sacral bone tumors reported that adequate bone margins can be achieved for more patients in the navigated group than in the non-navigation group (81% versus 50%; odds ratio, 4.14 [95% CI, 1.43-12.01]; p=0.007). Zhang et al14) analyzed the risk factors for local recurrence in 125 spine tumor patients and found recurrence rate of 45.6% in their study. The authors observed that all patients who received incomplete resection had local recurrence in follow-up and incomplete resection was a significant risk factor in logistic regression analysis.

Bandlera et al6) in a case series of navigation guided spine tumor resection in 7 patients showed no local recurrence at 18 months mean follow-up. Ando et al15) in a prospective study on 18 patients who underwent navigation-assisted surgery for a primary spine tumor reported no local recurrence in any case after mean follow-up of 25.1 months duration. Our study has strength of long follow-up duration (mean 33.42 months) which is important to report any local recurrence. In the present study, despite a small size of cohort, we found that adequate resection margin was achieved in 85.7% (six out of seven) patients. One patient who had a positive resection margin did not show local recurrence during follow-up. Local recurrence was detected only in one out of seven patients (14.3%). One patient who was diagnosed with a local recurrence at seven months follow-up had a primary osteosarcoma located in S2 and S3 left side pedicle. Patient underwent wide excision by S2 and S3 total sacrectomy and histopathology showed negative tumor margins. Local recurrence arised from right ala region of S1 was not from inadequate resection margin, but from a satellite lesion (Fig. 3).

Surgical navigation system is a promising technique that addresses in another point of view. Real time navigation techniques in spinal instrumentation can improve screw placement accuracy. Especially, in tumor resection surgery, anatomy is distorted. Thus, the landmark for inserting pedicle screw can be obscure. Under such situation, navigation will be helpful for placing the instrumentation. Ando et al15) reported no malposition of pedicle screws in 18 patients who underwent O-arm assisted spine tumor resection surgeries. Shin et al16) in a meta-analysis of 20 studies have found an increased accuracy of navigated pedicle screw placement, with pedicle breach in only 6% of navigated screws compared to 15% with a free hand. Moreover, navigation can help surgeon decrease potential injury to critical neurovascular structures step by step and in real time with a 3D point of view. Our study also showed precision of navigation in pedicle screw placement with zero malposition rate and no mechanical failure in follow-up.

This study has few limitations. The size of cohort was small. With such as small number, there is a limitation to conduct and interpret meaningful analyses. We cannot conclude that navigation assisted surgeries are superior to conventional tumor resection surgeries either as this study is a case series study, not a case control study. Secondly, although the surgical margin determined by the multidisciplinary cancer team and the resection margin performed in the surgery were compared through CT image, it was impossible to quantify the matching rate. Moreover, because postoperative chemotherapy and radiotherapy regimen were different from each patient, it was hard to conclude that local recurrence was only due to inadequate resection margin.

Navigation-assisted surgery for spinal tumors could be useful not only for accurate reconstruction, but also for determination of surgical margin. Although the number of cases was limited in this study, both thorough preoperative radiological review and intraoperative navigation could provide us benefits for obtaining safe resection margin.