Patients with SLDH commonly present with muscle atrophy, especially of the erector spinae and iliocostalis lumborum muscles. The injury and lipogenesis of the paraspinal muscles after open lumbar surgery are the primary causes of long-term low back pain [22–24]. With PELD, the surgeon can perform a discectomy via a channel that measures 0.7–12 mm in diameter, which can effectively reduce muscle injury and the incidence of postoperative low back pain [25, 26]. Comparisons of surgery durations, postoperative VAS scores, ODI scores, and hospitalization periods between PELD and OLM patients have yielded inconsistent results across previous studies. Through a meta-analysis, Manyoung Kim  found that VAS scores of the lower back and leg, ODI scores, surgery durations, and hospitalization periods of patients who underwent PELD were all better than those of patients who underwent OLM, whereas there were no differences in the MacNab grading, complication rates, recurrence rates, or repeat surgery rates between the patients who underwent the two procedures. However, another meta-analysis showed that when the parameters of patients who underwent OLM were compared with those of patients who underwent PELD, those who underwent the latter procedure had shorter surgery durations and hospital stays; however, there were no significant differences in postoperative VAS scores, ODI scores, surgical complications, or reoperation rates between the two groups . In the present study, VAS-BP scores, VAS-LL scores, and ODI scores in both groups were significantly improved after surgery (P < 0.05). The VAS-BP scores at the 1st week and 3rd month after surgery and the ODI scores at the 12th month’s and 24th month’s appointment in the PEID group were lower than those in the OLM group (P < 0.05); however, there were no differences in the VAS-LL score at the postoperative appointment, the VAS-BP score at the 12th month and 24th month, or in the MacNab level at the 24th month after surgery (P > 0.05). We believe that the lower VAS-BP scores at 1st week and 3rd month, and the ODI scores at the 12th month and 24th month appointment in the PEID group may be related to the fact that the patients included in this study were elderly patients. Compared with PELD, OLM has a wider range of paravertebral muscle dissection, laminectomy, and facet arthrotomy. A previous study reports that 23% of patients required conservative treatment for moderate back pain, and 9% of patients underwent fusion surgery at the primary discectomy site for severe back pain . Another study reported that 70% of the patients experienced back pain during the 4–17-year follow-up period . Elderly patients exhibit muscle atrophy and spinal degeneration. Therefore, OLM surgery should aggravate the back muscle injury, which affects the patient’s low back pain and lower back function, and show higher VAS-BP scores and ODI scores in the OLM group than in the PEID group. Our results demonstrate that both PEID and OLM exhibit good clinical efficacy in the treatment of SLDH and that PELD may play a role in reducing postoperative low back pain.
According to previous studies, PELD has advantages in terms of inducing minimal surgical trauma and minimal intraoperative bleeding and being associated with short surgery durations and fast postoperative recovery [30–32]. In the present study, the estimated intraoperative blood loss, postoperative bed stay, and duration of hospitalization of patients in the PEID treatment group were all less than those of patients in the OLM group (P < 0.05), which is similar to the findings of previous studies . Previous studies have shown that PELD is associated with less paraspinal muscle damage and more bony structure preservation than OLM [12, 33]. In the present study, patients could get out of bed with the protection of lumbar support belts according to the clinical situation after surgery in the PEID and OLM groups. All the above reasons could lead to the shortening of postoperative bed stays and hospitalization durations; however, it also confirmed that PEID was characterized by less intraoperative trauma and faster postoperative recovery than OLM. The above factors could also be the reasons why the PEID group had lower VAS-BP scores at the 1st week and 3rd month after surgery and lower ODI scores at the 12th month’s and 24th month’s appointments than the OLM group. In our study, the mean surgery duration in the PEID group was higher than that in the OLM group. The reasons for this finding may be as follows: first, SLDH is often accompanied by the degeneration and hyperplasia of facet joints, both of which make PEID harder in SLDH than in younger patients with lumbar disc herniations. Second, before this study, our team had performed 111 PEIDs, and most of the patients involved were young patients with lumbar disc herniation and without lumbar spinal stenosis. The surgeon did not have much experience when it comes to PEID for senile patients. In contrast, OLM is a classical technique. Prior to this study, OLM had been used to treat more than four hundred patients with disc herniation, including young and elderly patients. The surgeon was already skillful at performing this procedure at the early stage of the present study. The surgeon was able to quickly finish a discectomy even with the presence of severe spinal degeneration. The results all confirm that the surgeon’s proficiency in PEID for SLDH could be further improved.
Nerve injury, dural tearing, and cerebrospinal fluid leakage were severe complications of PELD . Li et al. showed that intraoperative nerve injury may be related to the compression of nerve roots by surgical instruments during the separation of adhesions or the stimulation of nerve roots during hemostasis with a radiofrequency knife head . The elderly patients had lumbar degeneration, which leads to osteophytic hyperplasia, which, in turn, leads to stenosis of the foramina and spinal canal. All three patients had nerve root injury during the implantation of the working channel into a narrow spinal canal. The elderly patients always suffered severe lumbar degeneration and had a thickened ligamentum flavum and adhesions to the dura, all of which resulted in dura injury in two patients during the separation of the ligamentum flavum and dura and also in postoperative cerebrospinal fluid leakage. All the complications had been successfully treated with oral mecobalamin and celecoxib, and there were no neurological sequelae. The incidence of complications in PELD is approximately 4.0–12.5% in the literature [13, 36]. In the present study, 19 patients with SLDH were treated via PEID, and two (10.5%) of them incurred dural injuries and experienced cerebrospinal fluid leakage during surgery while three (15.8%) of them experienced transient nerve-root irritation. The complication rate in the present study was higher than that reported in previous studies, and we believe that this discrepancy is mainly related to spinal degeneration in elderly patients and the paucity of surgical experience at the early stage.
With a combination of our review of the literature and our personal experiences, we can suggest measures to reduce the incidence of complications of PEID. (1) Detailed X-ray, CT, and MRI examinations to understand the location of the disc herniation preoperatively; (2) repeated simulation of the implantation position and the angle of percutaneous endoscopy with the help of images before surgery; (3) careful manipulations during the operation and the flexible use of grinding under the endoscope to remove the degenerative hyperplasia of osteophytes; (4) gentle insertion of the working channel into the lamina space and remembering to avoid excessive force while inserting the channel into the lamina space; and (5) careful removal of the disc after clearly identifying the target nerve and disc, and the avoidance of the blind removal of the disc.
PEID and OLM have the advantages of inducing less trauma, enabling rapid postoperative recovery, and yielding satisfactory clinical results in the treatment of SLDH. However, surgical pointer manipulations should be strictly controlled. Lumbar degeneration and radiographic lumbar spinal canal stenosis are common conditions in elderly patients that render PELD difficult and result in significantly long learning curves for surgeons, especially beginners [27, 37]. OLM enables extensive decompression of the narrow spinal canal visually; however, at the same time, it increases the risk of postoperative lumbar instability and persistent low back pain [8, 9, 38]. Therefore, we recommend that PEID or OLM be used for the treatment of SLDH.
Our present study had some limitations. First, it was a retrospective study and not a prospective study that could provide a higher level of clinical evidence for the comparison of two surgical approaches in SLDH. Second, it had a small sample size.