Ultrasonic surgical equipment was initially developed in the 1950s when it was mainly applied in dentistry and maxillofacial surgery [11]. The ultrasonic bone scalpel (UBS) is a relatively novel ultrasonic surgical device that cuts bone and spares soft tissues, and has been used for various spinal disorders [4,5,7,8]. To date, although the advance of instruments and technique, thoracic spinal decompression surgery is still of great challenging. Unfavorable outcomes and high incidence of complications were frequently reported [12,13]. To the best of our knowledge, few studies has described how to use UBS for thoracic spinal decompression [4], and no consistent surgical procedures have yet been established. In view of UBS’S technical advantages in bone cutting, we believe that its standard use in decompression surgery will be conducive to achieved success of surgeons. Therefore, the purpose of this study is to present our experience and technique note of using UBS, and to put forward a standard surgical procedure for thoracic spinal decompression.
Laminectomy is a well-described technique and has been widely used for treating various spinal stenosis diseases [3]. In some extremely cases with severe spinal cord damage because of the existence of OPLL, it can also be performed as an alternative surgery to circumferential decompression due to its lower risk of postoperative paralysis and technical feasibly [14]. However, in some OLF cases, a standard laminectomy may not completely resect the ossifications, since some of the compression may occurred behind the facet joints. In such conditions, an extensive laminectomy including the lamina and the medial half of the bilateral facet joint is required [15]. In this study, we emphasized the decompression width and chose the midline of the articular process as the slotting reference. In addition, choosing this landmark for bone cutting has another technical advantage. Since some authors have suggested that the laminectomy with ultrasonic osteome was performed without direct visualization of dura, and spinal cord injury may occur by insertion of blade beyond the inner cortex of lamina, if penetration of lamina is not felt [4]. In our experience, grooving along the midline of the articular process is safe enough to avoid the possibility of direct contact between the tip and the dura, and even if the blade accidentally penetrates the innermost cortex, the tip will be inserted laterally into the dura without damaging the spinal cord.
En block resection of the lamina and ossified ligamentum flavum has many advantages, as it can reduce the operative time and bleeding, and provide a relatively bloodless surgical field. In the past, laminectomy was commonly performed with a high-speed drill and kerrison rongeurs. The process was technically demanding and time consuming because a large amount of bone was removed piecemeal. Blood loss also increases as the operation time prolonged [3,5]. Use of UBS for en block resection has simplified this surgical procedure and improved the visualization [16]. In this procedure, surgeons only need to cut through the four edges of the lamina and ossified lesions in a relatively safe area, instead of doing operations above the spinal cord like piecemeal resection does, thus significantly reducing the risk of accidental spinal cord injury. By gradually lifting the lamina, the surgeon can quickly assess the adhesion of the ossified lesions to the dura. With the careful operation, iatrogenic durotomy can be avoided in most cases and the ossified dura can be safely removed under direct vision. Otherwise, intraoperative blood loss is expected to be reduced due to reduced operating procedures and the clotting effect of UBS. Lu et al [17] compared the efficacy and safety of UBS with high-speed drill for TSS decompression and found that UBS effectively reduced the operation time and blood loss. In this study, the mean blood loss was 108.3±47.3ml and the operation time was 3.0±1.4min for single-segments laminectomy. These results are comparable with previous reports and highlight the feasibility of UBS in saving time and reducing bleeding.
In spite of the effectiveness of laminectomy in most TSS cases, it may not appropriate in all cases, especially those with ventral lesions of the spinal cord such as OPLL and TDH [18]. Circumferential decompression with good surgical outcomes was considered to be an ideal treatment for these diseases. Two-staged posterior- anterior approach and one-stage posterior circumferential decompression have been described [6,19-21]. However, to the best of our knowledge, there are no consistent surgical procedures have been established due to the rarity of the disease and variety of its complications. The use of UBS in circumferential decompression was less described and is expected to reduce most complications. In this study, UBS was used for one-stage circumferential decompression, bilateral facet joints were excised accurately and nerve roots were preserved. By creating a buffer space beneath the ventral lesion, we can safely cut the OPLL base using our specially designed angled tip. After separating the adhesion between the dura and OPLL, OPLL was completely removed. During this process, we encountered no nerve injury complications except for an unintended dura tear during the separation of dura adhesions in one patient. The incidence of complications was noted to be lower in comparison with studies with conventional instruments [12,22,23].
Dural tear and neurologic deficit are the most common complications of spinal surgery [22,23]. It is reported that the incidence of durotomy varies greatly ranging from 1.1% to 17% [25], and the rate of neurologic deficit ranges from 8.4% to 19.8% [23]. Risk factors analysis showed that the most common instrument leading to durotomies was the Kerrions punch followed by the curate and then the drill [25]. Liu et al [5] compared the safety and efficacy of UBS and high-speed drill in multilevel thoracic OLF, the incidence of durotomies was 5.6% and 13.0% respectively. Chen et al [26] reported 2 cases (12.5%) of dural tear in the high-speed burr group, while no dural injury occurred in the UBS group. In this cohort, six patients have experienced CSF leakage, of which 5 cases were due to the resection of ossified dura mater and the other case is occurred during the separation process of OPLL and dura mater. Although, the dural mater of these 5 cases was resected purposely, the fact that patients with dural ossification has a higher incidence of CSF leakage reminding us the importance of preoperative diagnosis and intraoperative management of such cases. In addition, we encountered two cases instrument-relate nerve root injury, with an incidence of 7.1%. The incidence was comparable to previous reports [12,22,23]. One explanation is that these complications are mainly occurred at the beginning use of UBS, when surgeons had limited experience with the instrument. Another reason is that both patients were diagnosed with severe TSS, one with OLF+OPLL, also with Skeletal Fluorosis, and the other with OPLL. In these two cases, the nerve roots were tightly pinched by ossified lesions and vertebral body. During laminectomy, the blade penetrates the inner cortex of the ossified lesion, and the vibratory force was transmitted directly to the dura mater of the nerve root, resulting in root injury. To overcome this complication, we modified our surgical procedure by retaining the last thin layer of ossification, rather than penetrating it, and use a narrow bone osteotome to pry open and amputate the lamina. Since then, we have used this method to perform more than 300 cases of similar surgery, and no UBS related nerve root injuries occurred. This surgical procedure was proved to be safe enough.
In terms of the neurological recovery, all patients have experienced an uneventful recovery process and the average JOA score was improved from 4.7±0.9 before surgery to 10.1±0.6 at the last follow-up. The average recovery rate was 85.8%, which is much higher than previous reports [3,6,12,27,28]. The results indicated that with this surgical procedure, postoperative long-term neurological recovery could be achieved.
Overall, for the first time, we systematically described the surgical procedures for the treatment of TSS with UBS. The results of this study were comparable with previous reports in which the authors compared safety and efficiency of using UBS and conventional instruments. In addition, two cases of nerve root injury occurred at the user’s beginning use of UBS, reminding us of the learning curve is required to safely use this novel device. Thus, we believed that successful decompression surgery requires not only appropriate tools, but also experience. Before surgeons becoming comfortable with the use of UBS, a system training process is essential.