The surgical treatment for thoracic disc herniations remains challenging for the spinal surgeon. The operative results of disc excision via thoracic laminectomy were disappointing. There was a 35% rate of postoperative neurological deterioration [16
]. The manipulation of the spinal cord, during the removal of the disc ventral to the spinal cord, may produce mechanical injury. In addition, this approach can potentially interfere with the spinal cord blood supply [18
Although the transpedicular and transfacet pedicle-sparing approaches are less extensive procedures, the exposure of the ventral view of the spinal cord is very limited. The transpedicular approach, with the aid of a 70°-angled endoscope, might be a solution for the limited visual field. However, it is difficult and disorienting to become familiar with a reversed surgical field [14
Transdural approaches have been successfully adopted for multilevel cervical spondylotic myelopathy with central or paracentral disc herniation or lumbar disc herniation [4
]. For the thoracic disc herniation, the posterior laminectomy has been abandoned because of the limited visualization of the surgical field and the unavoidable manipulation of the spinal cord. Horwitz et al. [11
] reported good outcomes for the ruptured thoracic discs after laminectomy with a dural opening, which facilitates the expression of the disc fragments without trauma to the spinal cord. Since then, the transdural approach has been evaluated in several case reports of thoracic intradural disc herniations during the past decade [1
]. With the sectioning of the dentate ligament and CSF drainage, the transdural approach allows for an adequate exposure without spinal cord retraction. The thoracic spinal cord is approximately 6.5 mm deep and 8.0 mm wide, whereas the thoracic spinal canal is approximately 16.8 mm deep and 17.2 mm wide; the canal widens distally. Therefore, there is a minimal clearance space of 9.2 mm laterally, whereas there is a 10.3 mm of clearance in the AP direction [6
]. With the aid of a microscope and the microsurgical technique, the transdural approach provides direct access to the lesion, permitting reduced trauma while removing the soft disc fragments without excessive cord retraction.
The advantages of the transdural approach are the use of familiar surgical techniques and a short learning curve. This approach can be readily adopted by spine surgeons who are comfortable with the laminectomy approach and requires no special instruments. The posterior transdural approach allows entry into the disc space without significant disruption of the integrity of the facet–pedicle complex. The limited bone removal and soft tissue destruction minimizes the perioperative pain with potential for improvement related to axial pain. The posterior transdural approach might reduce the morbidity associated with entry into the chest; it is optimal especially for the high-risk patient. Especially in the upper thoracic disc, such as illustrative case 1, this approach could be very reasonable given that an anterior approach to the T2–T3 level would likely require splitting the sternum. The potential complications associated with this procedure are CSF leakage and the development of a pseudomeningocele. For case number 3 at the cervicothoracic junction an alternative would have been an anterior approach to the cervical spine and anterior corpectomy and fusion. The advantage of the transdural was to obviate the need of a fusion.
A total six cases of thoracic disc herniation were operated from January 2006 to June 2008. Two cases with posterolateral type of thoracic disc herniation were performed through posterior convenient approach, because the spaces for disc removal were easily exposed via only unilateral laminectomy and medial facetectomy without dural incision. In case with severe calcified central disc, transthoracic approach was used for disc removal due to the possibility of damage caused by too much retraction of spinal cord during the surgery. Three cases in this paper were performed via transdural approach.
In the cases presented, no attempt was made to close the ventral dural defect. The ventral dura densely adheres to the posterior longitudinal ligament and annulus fibrosus, which explains why a small incision less than 1 cm is not problematic. Regarding the neurological complications, no patient demonstrated worsening of spinal cord function postoperatively. Intraoperative neurophysiological monitoring, including somatosensory evoked potential (SSEP), motor evoked potential (MEP), and electromyography (EMG), have been used to monitor the spinal cord during spine and spinal cord surgeries. Combined intraoperative monitoring is helpful for predicting and possibly preventing neurological injury. Therefore, intraoperative monitoring can be adopted in posterior transdural approach for thoracic disc herniation. We did not encounter any neurological complications despite the fact that no intraoperative monitoring was used in these cases. We certainly recommend intraoperative neuromonitoring when this technology is available.
The posterior transdural approach for the thoracic disc allows a more direct approach to the paracentral discs. In addition, ample operating space can be secured without excessive retraction of the spinal cord. Although the follow-up period and the number of patients were limited for a definitive conclusion, the posterior transdural approach could be applied for lateral and paracentral soft-disc herniations. However, calcified central disc herniations might not be appropriate for this procedure. The removal of hard discs risks the occurrence of inadvertent mechanical injury to the spinal cord. Therefore, computed tomography (CT) scan should be performed prior to selecting the surgical approach and to avoid this technique in those with calcified discs. We emphasize that this approach offers an alternative approach in those with soft discs documented by CT and in those that may not be good candidates for an anterior approach.