Comparative Study of the Percutaneous Endoscopic Interlaminar Approach and Transforaminal Approach in the Treatment of Calcied Lumbar Disc Herniation

Background: Percutaneous endoscopic lumbar discectomy (PELD) is a relatively safe and effective minimally invasive surgery in the treatment of calcied lumbar disc herniation (CLDH). However, studies on percutaneous endoscopic interlaminar discectomy (PEID) and percutaneous endoscopic transforaminal discectomy (PETD) for CLDH have rarely been reported. This research aimed to compare the clinical ecacy of PEID and PETD for L5-S1 CLDH. Methods: We retrospectively analyzed 42 consecutive patients with L5-S1 CLDH treated with PELD at our institution from August 2016 to December 2019. Patients were divided into PEID group (n = 24) and PETD (n = 18) group according to the surgical methods. The demographic characteristics and surgical results of the two groups were compared. Clinical outcomes were estimated by the visual analog scale (VAS) for leg pain, Oswestry disability index (ODI) and modied MacNab criteria. Results: All patients were successfully operated on by PEID or PETD. No signicant differences in the demographic characteristics, intraoperative blood loss, postoperative hospital stay and complication rate were noted between the PEID and PETD groups (P > 0.05). The excellent and good rates in the PEID group were similar to those in the PETD group (91.67% vs 88.89%, P > 0.05), whereas the PEID group exhibited superior results for operative time (m) (64.58 ± 5.88 vs 85.56 ± 9.06, P < 0.05) and uoroscopy times (n) (2.96 ± 0.96 vs 13.33 ± 2.64, P < 0.05) compared with the PETD group. Conclusions: PEID had achieved good clinical ecacy as PETD for L5-S1 CLDH. Compared with PETD, PEID had the advantages of shorter operative time and a reduced number of uoroscopy times in the treatment of CLDH.


Background
Calci ed lumbar disc herniation (CLDH), with a low incidence rate, is a special type of lumbar disc herniation (LDH). Calci ed disc herniation, with hard structure, usually adheres extensively to surrounding tissues such as nerve roots and the dural sac. Most patients with CLDH have severe low back and leg pain symptoms and even severe neurological symptoms in the acute stage [1]. However, conservative treatments fail to effectively relieve symptoms. with a long incision, extensive stripping of the paravertebral muscles and laminectomy, has some de ciencies, such as signi cant tissue damage, considerable intraoperative blood loss, muscle denervation and atrophy, and even spinal instability [2,3].
Percutaneous endoscopic lumbar discectomy (PELD), including percutaneous endoscopic interlaminar discectomy (PEID) and percutaneous endoscopic transforaminal discectomy (PETD), is a minimally invasive operation for the treatment of LDH. Some studies demonstrated that PELD had similar clinical results as traditional open discectomy [4,5]. PELD has less intraoperative blood loss, less trauma, faster postoperative recovery, and shorter hospital stay than traditional open surgery. However, it is di cult and challenging to treat CLDH with PELD because the calci ed disc tightly adheres to nerve roots and the dural sac [6]. In recent years, with the appearance of surgical instruments such as ultrasonic osteotomes and endoscopic grinding drills, PELD has been gradually applied to the treatment of CLDH. However, there are no studies and reports comparing the e cacy of PEID and PETD in the treatment of L5-S1 CLDH. The research aims to compare the clinical effect of the two surgical approaches and to provide clinical guidance for L5-S1 CLDH.

General Information
This research obtained the support of the Ethics Committee of our institution and informed consent of all patients. From August 2016 to December 2019, we collected 42 consecutive patients with L5-S1 CLDH treated with PEID and PETD in our hospital. These patients were divided into PEID (n = 24) and PETD (n = 18) groups according to the surgical methods. The inclusion criteria were as follows: 1) symptoms of pain and numbness of lower limbs; 2) L5-S1 CLDH con rmed by preoperative lumbar X-ray, computed tomography (CT) and magnetic resonance imaging (MRI); 3) failure of conservative treatment or no signi cant improvement in symptoms for greater than 3 months; and 4) the follow-up time was greater than 12 months. The exclusion criteria were as follows: 1) multisegmental lesions; 2) noncalci ed lumbar disc herniation; 3) lumbar spinal stenosis, lumbar instability, lumbar tuberculosis, lumbar infection or spinal tumor. A percutaneous transforaminal endoscopic spine system (Joimax, Karlsruhe, Germany), tipexible bipolar radiofrequency system (Elliquence LLC, USA), ultrasonic osteotome (SMTP, China) and endoscopic instruments (MaxMorespine GmbH, Germany) were used. Surgical Procedure PEID. All patients were placed on the operating table in the prone position after general anesthesia. The surgical segment was identi ed by uoroscopy on an X-ray machine. After marking the puncture site on the body surface, routine disinfection was performed. The puncture point was anesthetized with local in ltration of 0.5% lidocaine. The puncture needle was inserted in 1.0 cm beside the posterior midline.
The needle position was con rmed by uoroscopy of a C-arm X-ray machine. A 0.7-cm longitudinal incision was made at the skin mark. The skin, subcutaneous tissue and lumbodorsal fascia were incised in turn. The dilating tube was placed to the ligamentum avum. The working cannula and endoscope were inserted along the dilating tube. After checking the operating system, the light source was connected, and the operation was performed under continuous saline irrigation. Under the endoscope, the ligamentum avum was cut by scissors. After reaching the epidural space, the lateral dural sac and nerve root were exposed. The locations of the nerve root, dural sac and herniated disc were explored. The degree of calci cation and size of the herniated disc were carefully con rmed. Under the endoscope, the calci ed and herniated disc tissue was removed by an ultrasonic osteotome (SMTP, China), and the herniated free nucleus pulposus was removed using nucleus pulposus forceps. Under direct vision, no obvious compression of the nerve root and dural sac was observed. During the operation, bipolar radiofrequency ablation was used to adequately stop bleeding. After removing the working cannula, the incision was closed intradermally and covered with a sterile dressing. All patients underwent CT and MRI examinations before ( Fig. 1A-D) and after ( Fig. 1E-H) operation.
PETD. The patient was placed in a prone position after local anesthesia, and the responsible segment was identi ed by C-arm X-ray uoroscopy. The level of responsible intervertebral space and the posterior midline line were marked. Routine disinfection and sheet laying were performed. The puncture site was set at 12 ~ 14 cm next to the posterior midline, and the puncture point was anesthetized with 0.5% lidocaine injection. The 18-G puncture needle was inserted in the marked puncture site, and the needle slid along the lateral aspect of the superior articular eminence and entered the spinal canal through the intervertebral foramen. It was con rmed that the puncture needle was located in the midline on the anterior radiograph and the posterior upper edge of the vertebral body on the lateral radiograph by uoroscopy with a C-arm X-ray machine. After inserting the guide wire, a puncture needle was removed.
The skin was cut to approximately 0.7 cm, and the expanding cannula was inserted gradually. Finally, the bevelled cannula was inserted along the dilatation tube. When the working channel was di cult to insert, part of the superior articular eminence could be ground off with a circular saw. The position of the working cannula was con rmed under uoroscopy on a C-arm X-ray machine. The working cannula was ushed with 3000 ml of saline, and the calci ed disc and herniated nucleus pulposus were removed with an ultrasonic osteotome (SMTP, China) and nucleus pulposus forceps under an endoscope. The working cannula was moved appropriately. The nerve roots and dural sac were moved. The nerve roots and dural sac were then explored for relaxation, and no protruding discs compressed the nerve roots and dural sac.
The bipolar radiofrequency ablator was used to perform brous annuloplasty. Bleeding points were carefully explored and given adequate hemostasis. The nerve root and dural sac were con rmed to be free of compression, and no bleeding points were evident. The working cannula was removed. Then, the incision was sutured intradermally and covered with a sterile dressing. The CT and MRI examinations of all patients were done before ( Fig. 2A-D) and after ( Fig. 2E-H) operation.

Measures
Demographic characteristics, such as age, sex, body mass index (BMI) and follow-up time, were compared between the PEID and PETD groups. Surgical outcomes, such as operative time, intraoperative blood loss, intraoperative uoroscopy times, postoperative hospital stay and complications, were compared between the two groups. Clinical outcomes indicators, including the preoperative and postoperative visual analog scale (VAS) [7] for leg pain, Oswestry disability index (ODI)[8], and modi ed MacNab criteria [9]at the last follow-up, were compared between the two groups.

Statistical Assessments
The SPSS 25.0 program (IBM Corporation, USA) was used for statistical analysis of all data.
Comparisons between the two groups were analyzed by independent-sample t tests, whereas pairedsamples t tests were used for intragroup comparison. Sex and complications were compared by chi-square tests. The modi ed MacNab criteria were compared by the Mann-Whitney U test. P < 0.05 indicates a statistically signi cant difference.

Demographic characteristics and surgical outcomes
All patients were successfully operated on by a trained and experienced spine surgeon in our hospital. We followed up all 42 patients (24 in the PEID group and 18 in the PETD group). The demographic characteristics (age, sex, BMI and follow-up time) of the two groups were signi cantly similar as shown in Table 1. In the PEID group, the operative time, intraoperative uoroscopy times, postoperative hospital stay, intraoperative blood loss and complication rate were 64.58 ± 5.88 minutes, 2.96 ± 0.96 times, 2.58 ± 1.21 days, 13.96 ± 4.66 ml and 8.33%, respectively. In the PETD group, the operative time, intraoperative uoroscopy times, postoperative hospital stay, intraoperative blood loss and complication rate were 85.56 ± 9.06 minutes, 13.33 ± 2.64 times, 2.56 ± 1.25 days, 13.61 ± 3.76 ml and 11.11%, respectively. PEID was superior to PETD in terms of operative time and intraoperative uoroscopy times (P < 0.05), but no signi cant difference in intraoperative blood loss, postoperative hospital stay or incidence of complications (P > 0.05) were noted as shown in Table 2. Values are mean ± SD, number, or as otherwise indicated.
Clinical outcomes VAS for leg pain and ODI scores were used to evaluate the clinical results before and 1 day, 1 month, 3 months, 6 months and 12 months after the operation. The postoperative VAS and ODI scores of the two groups were obviously lower than those before surgery (P < 0.05) as shown in Table 3. No signi cant difference was found in VAS and ODI scores between the two groups at any follow-up time point before and after surgery (P > 0.05) as shown in Table 3. On the basis of the modi ed MacNab criteria at the last follow-up, no signi cant difference was found in the excellent and good rates of the two groups (91.67% vs 88.89%, P > 0.05) as shown in Table 2. Values are mean ± SD, number, or as otherwise indicated.

Complications
In the PEID group, 1 patient had a dural sac tear but no cerebrospinal uid leakage; 1 patient had postoperative dysesthesia. In the PETD group, 1 patient had postoperative dysesthesia; 1 patient had residue of herniation. The symptoms of patient with residue were relieved after the second operation while the rest of patients were cured after conservative treatments. During the follow-up period, there were no reports of epidural hematoma, infection or lower extremity deep vein thrombosis.

Discussions
Although both PEID and PETD could achieve good clinical e cacy in the treatment of L5-S1 CLDH, our study showed that PEID was superior to PETD based on operative time and intraoperative uoroscopy times.
Compared with children, CLDH is more common in adults. Children usually are treated by conservative treatments, while adults fail to. Calci ed disc herniation adheres extensively to nerve roots and the dural sac, which not only increases the di culty of removal by PELD, but also enlarges the risk of nerve root injury and dural sac tear. Some previous studies showed that PELD was applied to treat CLDH [10][11][12][13][14]. Dabo et al. [15] showed that 30 patients of CLDH were treated by PEID with a trephine instrument, but 16 patients had postoperative dysesthesia 3 months after operation. Yu et al. [16] reported that the symptoms of 25 CLDH patients treated by PETD were relieved, but 7 patients had postoperative dysesthesia and 1 patient had recurrence of herniation.
In our study, postoperative outcomes demonstrated that the symptoms of all patients were signi cantly relieved by PEID or PETD. For the treatment of CLDH, we used an ultrasonic osteotome to remove the calci ed disc. Ultrasonic osteotome had selective osteotomy properties and retains the adjacent soft tissue [17,18]. Compared with grinding drills, ultrasonic osteotome was considered to be safe, accurate and effective for the removal of bone tissue [19]. Calci ed intervertebral disc, nerve roots and dural sac should be carefully separated and exposed during the procedure. Care should be taken when pulling the nerve root and dural sac. Part of the soft herniated intervertebral disc should be taken out to create a large enough safe space. Small pieces of calci ed intervertebral disc could be removed directly. Large calci ed disc could be rst divided into small pieces with an ultrasonic osteotome and removed in turn. The above measures can reduce the risk of the injury of nerve root and dural sac. Both PEID and PETD may be safer and more effective surgical methods for olderly, obviously frail or economically di cult CLDH patients.
Compared with PEID, PETD has some de ciencies in the treatment of L5-S1 CLDH. PETD is a relatively complex procedure with a long learning curve. Multiple punctures and adequate foraminoplasty are needed, especially in the L5-S1 level with foraminal stenosis and a high iliac crest, which not only increases the operative time and uoroscopy times, but also enlarges the risk of the exiting nerve root injury. Moreover, surgeons and patients are also exposed to much radiation [20]. The working channel is not exible enough during the procedure of PETD, which makes it di cult to su cient decompression of the herniated calci ed disc. Nie et al. [21] found that the operative time and uoroscopy times of PEID were signi cantly shorter than those of PETD. The results were similar to those of the present study.
In our research, no signi cant difference was observed in the complication rate between the PEID and PETD groups, which may be due to the small number of cases or short follow-up time. In the PEID group, 1 patient had a dural sac tear but no cerebrospinal uid leak, and 1 patient had postoperative dysesthesia. In addition, 1 patient had postoperative dysesthesia, and 1 patient had residue of herniation in the PETD group. According to our experience, repeated punctures and foraminoplasty may cause the nerve root injury, which may lead to postoperative dysesthesia. The tight adhesion between calci ed intervertebral disc and surrounding nerve tissues may result in nerve root injury, dural sac tear and even cerebrospinal uid leakage. Residue of herniation may be caused by the poor position of the working channel due to the obstruction of the high iliac crest, resulting in insu cient decompression.
There were some limitations in our research. First, this was a retrospective study. Moreover, the sample size is small and the follow-up period is short. Further studies with multicenter, large sample and longterm follow-up will be conducted in the future.

Conclusions
Both PEID and PETD had achieved good clinical outcomes for L5-S1 CLDH. The clinical e cacy of PEID and PETD in the treatment of CLDH at L5-S1 level were comparable. Compared with PETD, PEID had the advantages of shorter operative time and a reduced number of uoroscopy times in the treatment of L5-S1CLDH.
Abbreviations CLDH: calci ed lumbar disc herniation; LDH: lumbar disc herniation; MRI: magnetic resonance imaging; CT: computed tomography; PELD: percutaneous endoscopic lumbar discectomy; PEID: percutaneous endoscopic interlaminar discectomy; PETD: percutaneous endoscopic transforaminal discectomy Declarations Ethics approval and consent to participate This study was approved by the ethics committee of China-Japan Union Hospital of Jilin University and was therefore performed in accordance with ethical standards. All patients provided their informed consent prior to inclusion in the study.

Consent for publication
Written informed consent was obtained from all participants.

Availability of data and materials
The corresponding author could provide the date of the study if necessary.
Authors YP C and H W designed the study. Authors YP C, QR Z, XK C, YB L and XP C gathered the clinical data and conducted the statistical analysis. Author YP C wrote the manuscript. Authors YPC and H W revised the manuscript. The nal manuscript was approved by all authors.

Figure 1
Pre-and post-operative MRI and CT (A-H). A-D Preoperative CT and MRI revealed lumbar disc herniation combined with calci cation. E-H Postoperative CT and MRI showed that the calci ed intervertebral disc was removed and the compressed nerve root had been relieved by PEID.