Forst and Hausmann[9] first applied modified arthroscopy to decompression in 1983, greatly improving the efficiency and safety of spinal surgery. More than anything, Kambin 's anatomical description of the neural foramen (Kambin triangle) is one of the cornerstones in developing of a fully endoscopic transforaminal approach[10]. On this basis, significant progress has been made in techniques for minimally invasive treatment of lumbar disc herniation, such as intradiscal injection of chymopapain[11], foraminoplasty[12], automatic nucleotomy aspiration via transforaminal approach with auxiliary catheter[13], laser decompression, and radiofrequency ablation[14]. In 1997, Yeung[15] et al developed a new generation of spinal endoscopic YESS system for intradiscal decompression through the Kambin safe trigone into the intervertebral disc, which has multiple channels and a wider angle, improved the nature of percutaneous endoscopic discectomy[16]. Since then, with the continuous development of spinal endoscopic techniques, tools and the transformation of concepts. Currently, PELD plays an irreplaceable role in the treatment of LDH.
During the process of shearing ligamentum flavum, rotating working channel and removing annulus fibrosus, stimulation of sinovertebral nerve and spinal nerve toot may cause intolerable lumbago and leg pain, and IELD has stronger traction on dura mater and nerve root during operation, which may lead to severe neuropathic pain. Therefore, most scholars recommend general anesthesia (GA) when performing PELD via the interlaminar approach[17][18]. However, some studies have found that the risk of nerve root injury is higher during general anesthesia, and nerve electromyography monitoring should be used during surgery to prevent nerve root injury[19]. Ye[20] et al treated 60 patients with LDH using the IELD technique and showed no significant difference in ODI and VAS scores between LA and GA, whereas 1 patient in GA group had intraoperative nerve root injury. At the same time, it has also been shown that GA is associated with postoperative cognitive dysfunction in elderly patients[21]. Currently, PELD with local anesthesia has been proven to be a safe and effective method. Local anesthesia has the advantages of a faster recovery, shorter hospital stay and fewer complications.
In this study, all patients were given local anesthesia, and the operation was successfully completed. There was no significant difference in postoperative bed rest time and hospitalization time between the two groups. 7 patients in the IELD group developed pain intraoperative, pain was controlled after intravenous sufentanil infusion. So IELD had a higher VAS scores for intraoperative back pain and leg pain than in the TELD group(P < 0.01). This could be a reason for the lower satisfaction rate in the IELD group. However, more intraoperative pain did not affect the clinical outcomes. At each follow-up of 3, 6, and 12 months after surgery, and the VAS and ODI score were significantly lower than preoperative (P < 0.01). At the last follow-up, According to MacNab criteria, the excellent and good rate rates of TELD group and IELD group were 93.3% and 90.0%, separately. Therefore, the efficacy of PELD under LA is similar to that of traditional surgery, but it has the advantages of less trauma, lower medical expense and faster recovery.
However, PELD via the transforaminal route is challenging at the L5-S1 level due to obstruction of the anatomy. The iliac crest and inclination of the L5-S1 levels frequently impedes the transforaminal approach, resulting in steeper trajectory angles away from the extruded disc. The location and angle of the working channel are key factors for successful PELD. Choi DJ[22] established a pathway through iliac drilling for the first time to treat L5-S1DH intervertebral disc herniation with high iliac spine, but this method is easy to cause iliac fracture, superior gluteal nerve and superior gluteal artery and vein injury, so it is difficult to popularize widely. But definition of high iliac crest is vague. Choi KC [23] et al first proposed a grading system for high iliac crest based on the relationship between the highest point of the iliac crest and the adjacent bony markers, They recommended that the iliac height grade ≥ 5 is the threshold of foraminoplasty for transforaminal endoscopic L5-S1 discectomy. We believe that whether or not foraminoplasty is performed is also associated with the severity of foraminal stenosis and the size of facet joints. Therefore, foraminoplasty was performed in this group of patients as appropriate, which not only increased the safe working area of the working catheter, but also reduced the risk of nerve root injury. Similar to previous studies[24][25], we also found that patients in the TELD group had a greater puncture time and radiation time than patients in the IELD group (P < 0.001). Analysis of the reasons: It is well known that the intervertebral foramen becomes progressively smaller in the lumbar spine, especially at the L5-S1 segment. In addition, the iliac crest usually conceals the L5-S1 foramen, which makes the puncture process difficult; therefore, a sufficient foraminoplasty is necessary to reveal axillar herniation. Conversely, the wider L5-S1 intervertebral disc space lowered the difficulty of PEID and thereby reduced the puncture time and radiation time. In the process of localization and puncture, the surgeon fully understand the spinal anatomy, master the puncture technique, can reduce the fluoroscopy time to a certain extent and shorten the puncture time.
L5-S1 has the largest interlaminar space distance and provides adequate operating space for the IELD. The majority of S1 nerve roots origin of the L5-S1disc level, and some may even originated superior L5-S1disc level[26]. There is a certain space between the S1 nerve root and the dural sac, and axillary type of herniation will increase the space. And, it is easy to remove the herniation by placing the working catheter under the nerve root axilla. In contrast to axillary type of herniation, the shoulder space is relatively small as well as the restriction of the articular process, and the working catheter is difficult to reach to the shoulder of the nerve root. it is risky to rotate the working catheter directly in towards the disc. Therefore, a laminectomy is therefore needed to create operating space to reduce the risk of nerve root injury. In this study, a laminectomy was required in 23.5% of patients. We believe that the difficulty of the IELD approach with a huge central LDH. Which leads to greater tension of the dura and nerve roots, and the rotating working catheter is easy to damage the dura mater and nerve roots. Therefore, in this study, the position of the tongue of the working catheter was first placed in the disc under the axillary herniated disc, part of the herniation was removed for intraspinal decompression, and then the catheter was rotated to the shoulder of the nerve root under the protection of a nerve dissector for subsequent procedures.
At the same time, Nie[27] et al found that PETD total operation time was higher than PEID in their study, which was contrary to our present study. In the study, we found that operative time unde the endoscope PEID group was longer than that of the PETD group. This is because PEID was done under general anesthesia in their study, whereas this study was done under local anesthesia. And some patients required more time to add intravenous sufentanil because of severe intraoperative pain caused by cutting the ligamentum flavum and laminectomy[28].
Although this technique is widely used in the treatment of LDH, some unavoidable complications that may affect the final treatment outcome have also been reported[29]. In this study, the most common complication was postoperative dysesthesia, which we speculated was associated with excessive intraoperative traction of the nerve root, The patients’ symptoms were improved by mecobalamin, gabapentin. In addition, complications such as recurrence of LDH and fragment Omissions were also observed in this study, but they were finally solved by various methods. There was no significant difference in complications between the two groups, and no serious complications such as nerve root injury and dural tear occurred. Surgeons should carefully consider appropriate technical factors and have a thorough understanding of patient anatomy to avoid complications.
The treatment results showed that both PEID and PETD could obtain satisfactory results for L5-S1 DH. PEID has shorter puncture time and less radiation exposure, but longer operative time unde the endoscope and greater intraoperative pain. Therefore, in clinical practice, operators should strictly grasp the indications, combined with patient anatomical variation and operator experience to choose a personalized surgical approach.
The present study had several limitations. Because TELD approach is the preferred approach for far-LDH, such patients were not included in this study. This was a preliminary analysis of the treatment results, without a detailed discussion of the relevant influencing factors or a comparative analysis with other surgical methods. Further investigations, especially muti-centered trails with a larger sample size should be conducted to overcome the limitations of our study.