Full-Endoscopic Transforaminal Ventral Decompression for Symptomatic Thoracic Disc Herniation with or without Calcification: Technical Notes and Case Series

Background Symptomatic thoracic disc herniation is a challenge in spinal surgery, especially for cases with calcification. Traditional open operation has a high complication rate. The authors introduced a modified full-endoscopic transforaminal ventral decompression technique in this study and evaluated its imaging and clinical outcomes. Materials and Methods Eleven patients with symptomatic thoracic disc herniation who underwent full-endoscopic transforaminal ventral decompression in a single medical center were enrolled. The surgical technique was performed as described in detail. Dilator sliding punching, endoscope-monitored foraminoplasty, and base cutting through the “safe triangle zone” are the key points of the technique. Clinical outcomes were assessed by the modified Japanese Orthopedic Association (mJOA) score for neurological improvement and the visual analogy score (VAS) for thoracic and leg pain. The operation time, hospital stay, and complications were also analyzed. Results Postoperative magnetic resonance imaging (MRI) revealed good decompression of the spinal cord. The mJOA improved from 7.4 (range: 5–10) to 10.2 (range: 9–11). Axial thoracic pain improved in 8 of 9 patients. Leg pain and thoracic radicular pain improved in all patients. No complications were observed. The average operation time was 136 minutes (range: 70–180 minutes). The average length of hospital stay was 5.3 days (range: 2–8 days). Conclusion Minimally invasive full-endoscopic transforaminal ventral decompression for the treatment of symptomatic thoracic disc herniation with or without calcification is feasible and may be another option for this challenging spine disease.


Introduction
Symptomatic thoracic disc herniation (TDH) is a relatively uncommon entity, constituting less than 1% of all disc herniations [1]. TDHs often occur in the 4-6 th decade of life and may present with radiculopathy and/or myelopathy. e symptoms range from slight back pain to severe intercostal neuralgia, weakness of the lower extremities, and even bowel and urine abnormalities [2,3]. Surgery is reserved for those patients who are nonresponsive to conservative treatment.
More recently, percutaneous transforaminal endoscopic discectomy (PTED) has been developed and achieved excellent clinical outcomes in the treatment of not only soft lumbar disc herniations (LDH) [9,10] but also of calcified type of lumbar disc herniation [11][12][13]. However, there are few reports about endoscopic transforaminal thoracic discectomy to treat TDH.
In the present study, we introduced a posterolateral fullendoscopic transforaminal ventral decompression technique for thoracic disc herniation with or without calcification.
is surgical technique is a modification of PTED for lumbar disc disease. We used this modified PTED technique to treat thoracic disc herniation and achieved good clinical outcomes.

Materials and Methods
is study was approved by the Ethics Committee of the Hebei General Hospital before data collection and analysis. It was a retrospective study. e diameter of the endoscope system was 4.3 mm, with a view angle of 30°and 181 mm working length through a working channel with an inside diameter of 6.9 mm (SPINENDOS GmbH, Munich, Germany).

Clinical Data.
Between January 2018 and December 2019, eleven patients with thoracic disc herniation underwent full-endoscopic transforaminal ventral decompression in a single medical center. Seven were male and four were female, aged 25-72 years (mean: 53 years). Patients with a history of fracture, infection, or tumors were excluded. Pathological changes were only or mainly located in the ventral side of the spinal cord, which was clinically recognized as the main cause of the symptoms. As confirmed by the MRI and CT scan, the pathological changes included soft disc herniation (2 cases) and calcified disc herniation (9 cases). Two was located at T9-T10, 3 at T10-T11, 5 at T11-T12, and 1 at T12-L1. e presenting symptoms were classified as axial thoracic pain, thoracic radicular pain, leg pain, and myelopathy (6). e mean symptom duration was 8.6 months (range 2-18 months). e clinical characteristics of the patients are given in Table 1.

Preoperative Preparation and
Anesthesia. CT, MRI, and X-ray examinations were performed before the operation. e patients needed to simulate the operation position under the care of the medical staff. ere were two advantages: first, the patients could adapt to the position of the operation in advance to avoid anxiety during the operation. Second, whether there were neurological deficits in a specific position could be identified preoperatively. For the operation, the patient was positioned in a comfortable prone position. Some soft cushions were used to keep the patients in a comfortable position and relieve their neurological symptoms. e anesthesia regime was local anesthesia combined with conscious sedation. During this, an electrocardiograph, blood pressure, respiration, and finger pulse oxygen saturation were monitored. Dexmedetomidine hydrochloride was pumped at a rate of 0.1-0.5 μg/kg/hour. e speed of the pump was adjusted according to the patient's surgical tolerance, so that the patient was maintained in a sober yet sedated state. Anesthesiologists took care of the patients throughout the operation. e local anesthetic was a mixture of 0.5% lidocaine and 0.25% ropivacaine. e anesthesia area included the skin, deep fascia, dorsolateral articular process, and intervertebral foramen. No anesthetic was given directly into the spinal canal. is can prevent the spinal cord from being anesthetized, resulting in serious complications.

Working Port Establishment.
First, we located the puncture trajectory under X-ray fluoroscopy. e puncture trajectory was consistent with the direction of the intervertebral space. e entry point was located on the medial side of the highest point of the posterior rib. e distance to the spinous process was approximately 6-8 cm. A 16G spinal needle was used for the anesthesia of the skin and the deep fascia. After making an incision, a hollow dilator with a diameter of 6.5 mm was used for the following puncture process. A dilator with a blunt tip could reduce the risk of pleura and spinal cord injury. e punching target was the lateral side of the articular process. e pinhole in the dilator can be used to inject anesthetics along the puncture path. After AP and lateral view were checked by X-ray, the dilator was slid into the lower part of the intervertebral foramen ( Figures 1 and 2). e working sheath could be introduced along the dilator. Foraminoplasty and spinal cord decompression were performed subsequently under the endoscopic view.

Endoscopic Operation.
After the soft tissue in the foramen was removed, the bony anatomical structure of the intervertebral foramen could be clearly exposed to view. Foraminoplasty was performed with a high-speed diamond burr or a Kerrison rongeur ( Figure 3). In the foraminoplasty, the excised bony structure included not only the ventral side of the superior articular process but also part of the posterior edge of the vertebral body, even the upper part of the pedicle.
e key point of the foramen foraminoplasty is to increase the range of motion of the working channel. is modified foraminoplasty could not only preserve the stable structure of the spine as much as possible but also expose the ventral pathology clearly. When the space of the intervertebral foramen area was ample enough, the working sheath could be further deepened. ere was no extra pressure on the spinal canal during this process. Forced entry of the working sheath was forbidden.
After entering the lateral spinal canal, for a soft disc herniation, discectomy was performed by the "out-in-out" technique. e fibrous ring was exposed and cut open. e soft nucleus pulposus in the intervertebral space could be cleaned. When the space was large enough, the endoscope and surgical instruments would be inserted. e disc herniated into the spinal canal could be seen at 12 o'clock in the field of vision. Discectomy can be achieved by the 30degree visual field angle of the endoscope and variable angle forceps. Central spinal canal decompression can be achieved by a broad foraminoplasty and working channel pressed down. However, this technique was only suitable for soft lesions. In some cases, the pathogenic factor was a calcified disc, osteophytes, or both. First, the soft disc was removed as described, and a tough shell remained. Bony decompression began at the cranial and caudal edge of the vertebrae, which was the base of the hypertrophic fibrous ring and osteophyte. We can see the spinal cord was jacked up at the distal side of the calcification. Epidural fat is filled in this area. After exposure and hemostasis, we could say a space between the spinal cord and calcification. e space was safe for manipulation, and it was named the "safe triangle zone" by our team (Figure 4). e three sides of the triangle were calcification, spinal cord, and posterior wall of the vertebral body. e tools could operate in this area without touching the spinal cord. After the base was disrupted by the high-speed diamond burr, the hump could be cut off piece by piece. Whether the posterior longitudinal ligament needed to be removed was dependent on the rate of dural sac relaxation. Adequate decompression should be the first aim of the surgery ( Figure 5). e operation is over after careful hemostasis. No drainage system was applied. Patient feedback is essential throughout the entire operation.   Figure 6). e neurological outcome was assessed using the modified Japanese Orthopedic Association (mJOA) score (11 points) [14]. e degree of axial/radicular thoracic and leg pain was assessed using the visual analogy score (VAS). e operation time and the length of hospital stay were also recorded. e operation time was defined as the time from puncture to suturing. It did not include the time of the operative positioning or the operation target locating. e hospital stay for rehabilitation and physical therapy for patients with myelopathy was not reckoned in the length of hospital stay.  e articular process, the ligamentum flavum, and the thoracic disc.

Results
All patients in this study successfully underwent surgery as described. Follow-up was available for all of the patients, and it ranged from 13 months to 24 months (average: 15 months). Postoperative MRI imaging revealed good decompression of the spinal cord in all 11 patients.
Intraoperative blood loss was not measured due to continuous fluid irrigation. However, no patient required any intra or postoperative blood transfusion. ere were no serious complications such as nerve injury, infection, or hematoma. e average operation time was 136 minutes (range: 70-180 minutes). e average length of hospital stay was 5.3 days (range: 2-8 days). e mJOA improved from 7.4 (range 5-10) preoperatively to 10.2 (range 9-11) at the last follow-up. Axial thoracic pain improved in 8 of 9 patients with this symptom. Leg pain and thoracic radicular pain improved in all of the patients with this symptom ( Table 2).

Discussion
Because of the underlying anatomy, ventral decompression of the thoracic spinal canal is a technical challenge. Open decompression inflicts great trauma, has a high rate of complications, and possibly requires additional internal fixation [4,7,8]. With the improvement of minimally invasive surgery, video-assisted thoracoscopic surgery, microendoscopic surgery, and full-endoscopic surgery have achieved satisfactory clinical outcomes and fewer complications in the treatment of TDH [15][16][17]. In this retrospective clinical study, we introduced a modified fullendoscopic ventral decompression technique for TDH. e imaging and clinical results were satisfactory. Postoperative MRI imaging in all cases showed sufficient decompression of the spinal cord. MRI also showed the presence of cerebrospinal fluid between the spinal cord and the dura. mJOA and VAS were improved in all cases except 1 case of axial thoracic pain. Similar surgical techniques and findings have been presented in some previous reports [18][19][20][21][22].
We have made some modifications to the previously reported transforaminal endoscopic ventral decompression techniques. In the punching procedure, we used a dilator downward sliding technology instead of the needle targeting puncture technique to avoid neurovascular injury and pleural injury. Pulmonary complications are one of the troublesome complications of thoracic spine surgeries. Entering the thoracic cavity is the immediate cause, which is more likely to occur in the anterior approach [23,24]. Although no such complication has been reported for transforaminal endoscopic surgery, we believe that this potential risk exists.
In the foraminoplasty procedure, we used a high-speed diamond burr and a Kerrison rongeur under an endoscopic view instead of a circular saw or bone drill under fluoroscopy to avoid spinal cord injury and to preserve the stability of the spinal posterior column as much as possible. Hua et al. applied this similar technique in lumbar surgery. e neurological complication rate was 1.4%, which is much lower than for traditional PTED surgery (12.4%) [25]. In the establishment of the working channel procedure, we used the "step-by-step" technique instead of the "one-step" technique [19,21] to avoid any iatrogenic pressure on the spinal canal during the working channel insertion. Wagner et al. applied this working channel establishing technique to a 31-year-old female with T8-9 disc herniation. e imaging and clinical outcomes were excellent [20]. We also reported the sequence applied for endoscopic decompression. Rutten et al. reported a similar technique called "box-shaped" [26]. However, there are some differences. First, for central calcified discs, it is difficult to resect the base directly with the burr because of the obstruction of the spinal cord. A strong pair of forceps needs to reach the safe triangle zone to complete the bone resection. Second, clear exposure of the safe triangle zone before bone resection can help us accurately judge the boundary of resection and the distance from the spinal cord. As far as we know, this is the first time anyone has described the concept of the "safe triangle zone." e clinical outcomes in this trial are similar to those in previous reports. Guo et al. reported 6 cases of symptomatic TDH. e mJOA improved from 4.4 preoperatively to 6.6 one year after surgery [19]. Choi et al. presented a mean VAS improvement from 6.5 to 3.0 for back pain and 5.8 to 2.5 for leg pain. However, only patients with soft disc herniations were included [27]. Rutten et al. reported a 20% (5/25) rate of complications including 1 dural tear, 1 epidural hematoma, 2 transient intercostal neuralgias, and 1 deterioration of myelopathy [28]. In our previous research on the endoscopic surgery for thoracic OLF, we found complications such as neck pain and dural tear [29]. However, no complication was observed in this study or in Choi and Guo's. is may be due to the small sample size. Because of the low incidence rate of TDH, this study is limited by its small sample size. It is not sufficient to show that this method is a safe and effective surgical technique. Larger controlled studies are warranted. Long-term followup and analysis will be required in the future.  Pre-op  Post-op  Pre-op  Post-op  Pre-op  Post-op  1  70  7  10  11  4  1  6  0  2  160  5  6  10  3  1  0  0  3  120  7  9  11  3  3  6  2  4  95  7  6  9  0  2  3  0  5  180  8  8  10  0  0  5  0  6  180  8  5  9  2  2  3  1  7  80  4

Conclusion
Full-endoscopic transforaminal ventral decompression for the treatment of symptomatic thoracic disc herniation with or without calcification is feasible and may be another option for this challenging spine disease. Data Availability e data generated or analyzed during this study are included in the article.

Disclosure
is study was performed at Hebei General Hospital.

Conflicts of Interest
e authors declared no potential conflicts of interest for the research, authorship, and publication of this article.

Authors' Contributions
Shangju Gao and Jingchao Wei contributed equally to this work. All authors reviewed the original study data, reviewed the data analysis, and approved the final paper, and all authors are responsible for archiving the study files.