Full-endoscopic lumbar decompression versus open posterior interbody fusion for surgical treatment of stable lumbar disc herniation in patients over 65 years: A retrospective cohort study with 2-year follow-up

Background: In the last decade, full-endoscopic techniques to treat lumbar disc herniation (LDH) have gained popularity in clinical practice. However, few studies have described the safety and ecacy of full-endoscopic lumbar decompression (FELD) in treating older patients with LDH. The aim of this study is to evaluate the safety and ecacy of FELD and open posterior lumbar interbody fusion (O-PLIF) for stable LDH in patients over 65 years of age. Methods: Data of 251 consecutive patients over 65 years received surgical treatment for stable LDH in 2 separated groups (minimally invasive spinal surgical group and open spinal surgical group) in single center between January 2014 and April 2018 were retrospectively analyzed. Patients were categorized into 2 groups (FELD group and O-PLIF group) according to the operative procedure. Clinical outcome evaluations were performed preoperatively, at 3, 6, 12, 24 months postoperatively, including Oswestry disability index (ODI) score and Japanese orthopaedic association (JOA) score for function assessment, visual analog scale (VAS) score for low-back pain and leg pain, MacNab criteria for assessment of life quality. In addition, the surgical information, including operative time, blood loss, hospitalization time, and adverse events were assessed. Results: Among 251 patients, 153 underwent FELD and 98 underwent O-PLIF. There was no signicant difference in baseline data between 2 groups (p>0.05). The mean operative time was signicantly shorter, blood loss and hospitalization time were signicantly reduced in patients having FELD than O-PLIF (p<0.001). The mean ODI, JOA, and VAS postoperative scores were signicantly improved over the preoperative scores in both groups (p<0.05). No signicant difference was between 2 groups in postoperative ODI, JOA, VAS scores, and MacNab evaluation (p>0.05). The adverse event rate was 5.2% in FELD group, which was signicantly lower than that in O-PLIF group (16.3%, p<0.05). The recurrence rate was 7.2% in FELD group and the rate of adjacent segment disc herniation was 6.1% in O-PLIF group. Conclusions: FELD could achieve satisfactory safety and ecacy for the treatment of stable LDH in the patients over 65 years of age. With less trauma, faster recovery, and lower adverse event rate, FELD may be an alternative surgical treatment for stable LDH in elderly patients. a signicant proportion of 2 and the distributions were degenerative were severe in these elderly patients and the distributions of disc degeneration, facet joint degeneration, and lateral recess stenosis between the 2 groups were also similar. The mean ODI, JOA, and VAS postoperative scores were signicantly improved over the preoperative scores in both FELD group and O-PLIF group. no signicant difference was found between the 2 groups in postoperative ODI, JOA, VAS scores, and MacNab evaluation. Therefore, the results suggest FELD is as ecacy as conventional O-PLIF in improving function, low-back and leg pain, and quality of life for patients over 65 years.


Introduction
Lumbar disc herniation (LDH) accounts for a large proportion of low-back pain as it affects over 90% of individuals in their lifetime [1]. With an average onset age of 45 years, LDH usually occurs in people between 20 and 50 years of age and is relatively rare in the elderly [2]. LDH is often combined with ligamentum avum hypertrophy and lateral recess stenosis in older population, which lead to lumbar spinal stenosis (LSS) [3]. Although most patients recover spontaneously or by conservative management, a considerable number of patients will eventually undergo surgical procedures [4]. LSS is the most common reason requiring lumbar spine surgery in patients older than 65 years and the decompressive spine surgery for older patients has increased substantially over the last decade [5,6]. However, surgical measures are associated with an increased risk of complications and poor outcomes in this population with multiple risk factors including general condition, comorbidities, and pathophysiological changes [7].

Conventional open posterior lumbar interbody fusion (O-PLIF) for spinal decompression and fusion has
been an effective surgical treatment for lumbar degenerative disease. Nevertheless, denervation of paraspinal muscles caused by open dissection may lead to postoperative low-back pain and muscle atrophy of the surgical segment [8]. As relevant comorbidities for both surgery and anesthesia occur more frequent in elderly than in a younger population, the older patients take greater risks in open procedure [9].
With the development of endoscopes instrument and great demands of patients for minimally invasive procedure, minimally invasive spinal surgery has achieved rapid advancement in the past four decades [10]. Percutaneous endoscopic lumbar discectomy (PELD) is a minimally invasive procedure for treating LDH rst reported in 1973 [11]. It became popular with the invention of Yeung endoscopic spine system (YESS) technique and transforaminal endoscopic spine system (TESSYS) technique [12][13][14].
However, endoscopic treatment for elderly LDH remains controversial due to a single lumbar discectomy without excision of facet joints and ligamentum avum could not achieve adequate decompression for elderly patients.
Full-endoscopic lumbar decompression (FELD) is a newly developed technique, which has been used in the treatment of LDH and LSS. With further development of PELD, FELD could achieve visualized satisfactory spinal decompression from dorsal to ventral and avoid the damage of posterior structures under local anesthesia [15], which made it a potential alternative for older patients. Nevertheless, to the best of our knowledge, few studies have described the safety and e cacy of FELD in treating older patients with LDH. Our study has performed both FELD and O-PLIF as treatment for LDH in patients over 65 years. Therefore, we attempted to evaluated the clinical outcomes, adverse events, and recurrence of FELD versus O-PLIF for LDH in patients over 65 years.

Materials And Methods
This retrospective study was performed after achieving the approval of the Ethic Committee of our hospital, and the patient outcomes were collected independently from informed consent. Case assignment was from the result of the whole section discussion before surgery and eventually determined by experienced surgeons. A total of 251 consecutive patients over 65 years received surgical treatment for stable LDH in 2 separated groups (minimally invasive spinal surgical group and open spinal surgical group) in single center between January 2014 and April 2018 were retrospectively enrolled. Patients were categorized into 2 groups, FELD group (n=153) and O-PLIF group (n=98), according to the operative procedure.
Inclusion criteria were: 1) single-level LDH, 2) veri ed by magnetic resonance imaging (MRI) and computed tomography (CT); 3) clinical symptoms and signs in accordance with imaging changes; 4) and failure of standard conservative treatment for at least 12 weeks. Exclusion criteria were: 1) lumbar spondylolisthesis, or segmental instability suggested by radiographic ndings; 2) vertebral infection, vertebral tumor, or other vertebral lesions; 3) and a surgical history involving the same level.

Radiographic measurements
Radiographic measurements were performed by 2 independent physicians. The chief physician examined and determined the measured results when the 2 observers were in disagreement. The digital images were stored in the picture archiving and communication system (PACS) (GE, USA). Preoperative MRI was used to determine lumbar disc degeneration and nerve root compromise in the lateral recess. Preoperative CT scans were used to measure lumbar facet joint degeneration. Disc degeneration was classi ed into 5 grades according to the grading system proposed by P rrmann et al [16]. Nerve root compromise in the lateral recess was graded as follows: Grade 0, no contact of the disc with the nerve root; Grade 1, contact without deviation; Grade 2, nerve root deviation; Grade 3, nerve root compression [17]. Facet joint degeneration was divided into 4 grades according to the CT scan using Weishaup grading [18].

Surgical technique
Full-endoscopic lumbar decompression (FELD) All procedures were carried out under local anesthesia with mixed anesthetic (2% lidocaine 15ml, 1% ropivacaine 10ml and saline 20ml) and sedated with midazolam and fentanyl, which allowed for monitoring of any change of the patients' symptoms and made it possible for surgeons to get instant feedback from patients so as not to cause damage to any neural structure. The patients were placed in a prone or lateral position on the radiolucent operating table. Preoperative posteroanterior and lateral radiographs were obtained by C-arm uoroscopy to determine an appropriate entry point and approach angle. There were 2 approaches for endoscopic surgeries of LDH, Transforaminal (TF) approach and interlaminar (IL) approach, the indications for each approach were showed in Figure 1. Special instruments were designed and constructed for foraminotomy and laminectomy, including outer working cannula, trepan, and inner working cannula (Endosurgi, China) ( Figure 2).
(1) TF approach The entry point was 10-12 cm lateral to spinal midline above the iliac crest. A long 18-gauge puncture rod was inserted from the entry point toward the midline under intermittent uoroscopic guidance. The superior facet joint was the anatomic marker to avoid puncturing injuries and compressions of nerve root. The working cannula and endoscopic surgical system were inserted toward the intervertebral foramina along the puncture rod and all the subsequent steps were performed under constant endoscopic visualization. With a saw to remove the hyperplastic facet joint, full-endoscopic visualized foraminoplasty was performed to enlarge the intervertebral foramina. Then the hyperplastic ligamentum avum were removed to ensure clearly visible decompression of dura and spinal nerves. Herniated disc, degenerated nucleus, and ventrally hypertrophic osteophytes were removed at last. The endoscopic was removed after freed nerve root could be identi ed and no active bleeding was con rmed ( Figure 3).
(2) IL approach The entry point was 1-2 cm lateral to the spinal midline. Likewise, the working cannula and endoscopic system were inserted from the entry point to the interlaminar ligamentum avum. Laminectomy and subsequent procedures were performed under constant endoscopic visualization. Su cient decompression of bony structure was necessary to expose the cranial and caudal insertion points of interlaminar ligamentum avum. En-bloc resection of the unilateral hyperplastic ligamentum avum was performed to expose the axillary and under-shoulder space of nerve root. Then the herniated disc was removed to achieve adequate decompression. The endoscopic was removed after freed nerve root could be identi ed and no active bleeding was con rmed.

Open posterior lumbar interbody fusion (O-PLIF)
O-PLIF was performed with the patient under general anesthesia, conventional posterior midline incision, dissection. A standard decompression was carried out at the lumbar disc herniation segment bilaterally, and a single cage packed with an autologous bone graft was inserted. Posterior pedicle screw instrumentation was used in all cases.

E cacy evaluation
Operative time, blood loss, hospitalization time, and adverse events were recorded. Oswestry disability index (ODI) score (range, 0-50) and japanese orthopaedic association (JOA) score (range, -6-29) for function assessment, and visual analog scale (VAS) score for back and leg pain (range, 0-10) were evaluated preoperatively and at 3, 6, 12, and 24 months postoperatively. Patient satisfaction was evaluated according to the modi ed MacNab criteria.

Statistical analysis
The data were analyzed using SPSS (version 26.0, Chicago, Illinois, USA). All data were presented as mean value ± standard deviation. The independent-samples t tests and chi-square tests were used to evaluate group differences. The paired-samples t tests were used to compare preoperative and postoperative data. Statistical signi cance was de ned as p<0.05.

Radiographic results
The preoperative evaluation of lumbar disc degeneration, nerve root compromise in the lateral recess, and facet joint degeneration were summarized in Table 2-4. The grade distribution of these evaluations between 2 groups were also similar (p>0.05).
Clinical outcomes FELD and O-PLIF were technically successful and well tolerated in all patients. Each patient underwent single-level FELD and O-PLIF, and the mean operative time, blood loss, and hospitalization time of both groups were summarized in Table 5. Mean operative time was shorter in the FELD group than in the O-PLIF group (p<0.001). The FELD group was superior to the O-PLIF group with less mean blood loss (p<0.001) and less hospitalization time (p<0.001).
In both FELD group and O-PLIF group, mean ODI, JOA, and VAS postoperative scores were improved signi cantly over the preoperative scores (p<0.05, Table 6). No signi cant difference was found between 2 groups in preoperative and postoperative ODI, JOA, and VAS scores (p>0.05). According to the modi ed MacNab criteria, the outcomes rated as excellent and good were 108(70.6%) and 34(22.2%) in FELD group, 75(76.5%) and 16(21.3%) in O-PLIF group at last follow-up, respectively (Table 6). There was no difference between 2 groups (p>0.05).

Adverse events and recurrence
During the follow-up period, adverse events occurred in 8(5.2%) patients in FELD group and 16(16.3%) patients in O-PLIF group, which summarized in Table 7. The adverse event rate in O-PLIF was signi cantly higher than that in FELD group (p<0.05). Among 11(7.2%) patients who were found recurrence in FELD group, 8 patients underwent FELD once again and 3 underwent O-PLIF for further treatment. New disc herniation was found in 6(6.1%) patients at the adjacent segment of fusion in O-PLIF group, and they all underwent open surgery for treatment.

Discussion
The therapeutic management of symptomatic LDH includes conservative treatment, epidural in ltrations, percutaneous therapeutic techniques, and surgical treatment. Most patients are responded well to physiotherapy, pharmacotherapy, and epidural steroid injection, while some patients still require percutaneous therapeutic techniques or surgery. LDH is most common in middle-aged patients, and early results has showed good outcomes for lumbar discectomy in younger populations [19][20][21]. In addition, short-term clinical e cacy of PELD has been showed reliable in treating young patients with LDH [22]. Although early study has reported the outcome of lumbar discectomy in elderly patients is as good as in younger patients [23], the surgical options for patients over 65 years with worse general conditions and more comorbidities is challenging and controversial. Both surgeons and patients should balance the surgical risks and bene ts of functional improvement and quality of life.
With less trauma, shorter operative time and hospitalization, and local anesthesia, the endoscopic procedures are considered to be an alternative choice for LDH in older population to reduce the occurrence of surgical complications. However, the degenerative changes of all lumbar spine structure gradually occur in normal aging process and mainly cause central spinal stenosis which affects the elderly most [24]. These changes, including disc herniation, ligament ava hypertrophy, and ossi cation of longitudinal ligament, increased the complexity of surgery in for elderly patients. Conventional PELD is an effective minimally invasive technique to treat LDH in recent years [13,14], but it is still of several limitations. The working cannula of YESS technique and TESSYS technique were established with the aid of intervention technique, which may lead to the neurological injury. In addition, without adequate foraminoplasty, the nerve root could not be fully exposed and intraspinal adequate decompression is di cult to achieve. Due to the obstructive anatomy, foraminoplasty is challenging at L5-S1 level and the nerve root is not possible to achieve adequate decompression via TF approach [25,26]. Hence, conventional PELD can hardly meet the complex surgical requirement for older patients with severe LDH.
As a developed technique of PELD, FELD is performed with full-endoscopic visualized foraminoplasty, which can ensure adequate decompression of nerve root and spinal canal safely. The uoroscopy is only used to con rm the surgical level and location of intervertebral foramina in the process of FELD, and all the sequent procedures, including foraminoplasty and discectomy, are performed under endoscopic visualization [27]. Thus, compared with conventional PELD, FELD could signi cantly reduce the uoroscopy time and the radiation exposure, and ensure adequate and safe decompression. Besides, IL approach is applied in L5-S1 level to avoid the obstacle of iliac crest. In our current study, we were able to directly compare the clinical outcomes and complications of FELD with O-PLIF for stable LDH in patients over 65 years. Chronic diseases, including hypertension, diabetes, and heart disease, accounted for a signi cant proportion of 2 groups and the distributions were matched. Meanwhile, the degenerative changes were severe in these elderly patients and the distributions of disc degeneration, facet joint degeneration, and lateral recess stenosis between the 2 groups were also similar. The mean ODI, JOA, and VAS postoperative scores were signi cantly improved over the preoperative scores in both FELD group and O-PLIF group. Besides, no signi cant difference was found between the 2 groups in postoperative ODI, JOA, VAS scores, and MacNab evaluation. Therefore, the results suggest FELD is as e cacy as conventional O-PLIF in improving function, low-back and leg pain, and quality of life for patients over 65 years.
Compared with open surgery, endoscopic techniques have signi cant advantages in the following aspects, including local anesthesia, preserving the normal posterior paraspinal structures, less operative time, less blood loss, and short hospital stay. Although annulus brosus fenestration and intraoperative nerve root irritation may be painful under local anesthesia, the surgeon could communicate with patient during the operation to con rm that the symptoms have improved, which indicate adequate decompression of the nerve root [12]. Furthermore, with relatively worse cardiopulmonary function, elderly patients have a lower risk of local anesthesia during surgery. In addition, minimal traumatization of FELD reduces the intraoperative blood loss and further shorten hospital stay. In this study, FELD also showed a signi cantly shorter mean operative time, less mean intraoperative blood loss, and shorter mean hospitalization time than O-PLIF.
The neurological adverse event rate of FELD in current case series was 1.3% (2/153), which was lower than that of O-PLIF (3.1%, 3/98). Therefore, FELD was relatively safe even for patients over 65 years. As for total adverse event rate of our study, 8 Early studies have showed a recurrence rate of 3.7%-6.9% during the follow-up period in all patients with LDH[28, 14,26]. In current study, 11(7.2%) in 153 patients over 65 years underwent FELD had recurrence. The recurrence was relatively higher in older people than in the general population.
Among the 11 patients who were found recurrence in FELD group, 8 patients underwent FELD once again and the other 3 underwent open surgery for further treatment. Adjacent segment degeneration is common after lumbar fusion surgery, with a high rate ranging from 5.2% to 18.5% [29]. In this study, new disc herniation was found in 6(6.1%) elderly patients at the adjacent segment of fusion, which was consistent with previous reports. Therefore, the patients underwent FELD are at risk for recurrence, while the patients underwent O-PLIF are at risk for new disc herniation at the adjacent segment.
Although our study showed some reliable results, there were several limitations. First, since it was a single-centered retrospective study, the generalizability of our ndings was limited and a randomized control trail may be more convincing. Second, the follow-up period was relatively short and a long-term follow-up should be required in the future. In addition, some older patients were illiterate and could not complete the clinical assessments by themselves, therefore, their family members were contacted to help them nish these assessments.

Conclusions
The traditional surgical treatment of lumbar disc herniation has been PLIF but there has been a trend towards minimally invasive procedures. FELD could achieve satisfactory safety and e cacy for the treatment of stable LDH in the patients over 65 years of age. Moreover, with less trauma, faster recovery, and lower adverse event rate, FELD may be an alternative surgical treatment for stable LDH in elderly patients. Further comparative studies and prospective, randomized, controlled studies should be conducted to assess the clinical outcomes. This study was approved by the Ethic Committee of The Peking university Third Hospital, and patient outcomes were collected independently from informed consent.

Consent for publication
Written informed consent for publication of their clinical details and clinical images was obtained from the patient.

Availability of data and materials
We state that data will not be shared because all raw data were used to prepare the gures included in the article.

Competing interests
The authors declare that they have no competing interests.

Funding
This research did not receive any speci c grant from any funding agencies in the public, commercial, or not-for-pro t sectors.        The indications for transforaminal approach and interlaminar approach Instruments designed for foraminotomy and laminectomy a. An outer working cannula was applied as an anchor on the lamina and articular processes. b. A trepan was applied to remove the osteophyte through the working cannula. c. An inner working cannula was applied as a channel of the endoscope. d. The assembled instruments with endoscopic system.