Low back pain was reported affecting up to 80% of the population during their lifetime [
Open lumbar discectomy was implemented as a standard surgery for LDH therapy, firstly described by Dandy and Peltier [
The overall success rate of conventional microdiscectomy ranged from 75 to 100% [
Recently, the significance of foraminoplasty has been widely emphasized. It was defined as “widening the foramen by undercutting the ventral part of the superior articular process (SAP) with ablation of foraminal ligament with the use of bone trephines, endoscopic drill, and side-firing laser to visualize the anterior epidural space and its contents” [
Upon the development of PELD, a technique of transforaminal PELD without foraminoplasty (TF PELD) was adopted on treating LDH. However, whether the injury during the TF PELF procedure would deteriorate the clinical outcomes compared with TF PELD on LDH treatment is an open question. In the present study, 140 patients with LDH who underwent TF PELF (62 cases, PELF group) or TF PELD (78 cases, PELD group) were recruited. The authors comprehensively compared the postoperative clinical outcomes between the two groups with a 2-year follow-up.
With approval from the Institutional Review Board of the Second Affiliated Hospital and Yuying Children’s Hospital of Wenzhou Medical University, the study population comprised 140 consecutive patients with LDH who underwent TF PELF or TF PELD surgery in our department from July 2014 to August 2016. All of the patient met the inclusion criteria and were followed up to 2 years postoperatively. All the patients provided the informed consents and protocols that described the details of the follow-up.
Inclusion criteria were as follows: (1) preoperative imaging evidence of LDH at L1-2, L2-3, L3-4, L4-5, and L5-S1 (monosegmental or double segmental), with or without canal and/or lateral recess stenosis caused by herniated mass on magnetic resonance images (MRI) and computed tomography (CT) (Figures
The preoperative and postoperative imaging data of patient who received TF PELF. (a and b) A preoperative MRI image shows the sagittal and coronal views of a patient diagnosed with LDH. (c) A preoperative CT image of the same patient. (d and e) The postoperative MRI sagittal and coronal views of the patient after TF PELF. (f) A postoperative CT image. (g) The postoperative 3D CT imaging result. The red round denotes foraminoplasty.
The preoperative and postoperative imaging data of patient who received TF PELD. (a and b) A preoperative MRI image shows the sagittal and coronal views of a patient diagnosed with LDH. (c) A preoperative CT image of the same patient. (d and e) The postoperative MRI sagittal and coronal views of the patient after TF PELD.
All the surgeries were performed by two senior and experienced surgeons (Dr. Zhan and Dr. Xu) in TF PELD and foraminoplasty. All procedures were performed following the standard TF PELF and TF PELD technique with the transforaminal endoscopic spine system (Joimax GmbH, Karlsruhe, Germany). Patients were on the lateral position on an operating table on the contralateral side. The C-arm fluoroscopy technique was used to help surgeons determine the affected discs and pedicle and to draw a line from the midpedicular annulus to the facet lateral margin and the extension to the body surface. The skin entry point from the midline was 10–12 cm. After subcutaneous infiltration of local anesthesia with 1.0–1.5 mL 0.5% lidocaine, the subsequent steps were performed sequentially: (1) An 18-gauge needle was inserted to reach the lower segmental SAP under fluoroscopic guidance with a puncture angle of about 15° until the needle tip reached the posterior rim of the SAP of the distal vertebrate at the lateral view and the medial pedicle line at the anterior-posterior view. (2) After the stylet was retreated, another 20 mL 0.5% lidocaine was injected through the needle for adequate anesthesia. A guide wire was inserted on the same direction of the needle and a 0.8 cm in diameter incision was made, followed by a serial dilation, and a working channel was rotated into the guide wire in succession. (3) Replacing the guide wire and dilation with a guide bar (Figures
Imaging data during surgeries. (a) The tip of the guide bar lay at the posterior rim of the upper endplate of the SAP facet of the distal vertebrate in the lateral view. (b) The tip of the guide bar lay at the medial pedicle line in the anterior-posterior view. (c) Cutting SAP with a trephine under endoscope. (d) Removing the herniated lumbar disc mass under endoscope during procedure. (e) The operation area after the herniated mass was removed. SAP = superior articular process, NRT = nerve root.
The foraminoplasty was individualized for each situation, and the surgeons decided to perform foraminoplasty depending on the operation location and experience. In cases where the working cannula could not be placed near the disc fragment due to the anatomical barrier, especially the SAP, leading to the inability of transforaminal endoscopic access to the dural sac or nerve root in the spinal canal, foraminoplasty also would be carried out to allow the working cannula access near the herniated disc [
Postoperative symptomatic improvement was evaluated by the surgeons on the operation day, and the radiography was further examined by MRI (Figures
Quantitative data were presented as mean ± standard deviation, and qualitative data were presented as frequency (%). The normality of the data was analyzed. Mann–Whitney
Eighty-six males and fifty-four females were included in this study, and the mean age was 54.5 years in the PELF group and 54.6 years in the PELD group (
Comparisons of basic information between PELF and PELD groups.
Values | PELF group ( |
PELD group ( |
|
---|---|---|---|
Female | 28 | 26 | 0.153 |
Male | 34 | 52 | |
Mean age (year) | 54.5 ± 15.26 | 54.6 ± 13.63 | 0.973 |
Pain duration (month) | 52.1 ± 89.34 | 22.9 ± 39.67 | 0.051 |
Operation time (minute) | 121.7 ± 46.39 | 108.9 ± 37.70 | 0.094 |
Hospital stay (day) | 11.06 ± 9.18 | 9.08 ± 3.75 | 0.458 |
Follow-up duration (month) | 22.0 ± 6.38 | 23.1 ± 4.55 | 0.181 |
Because of the loss of contact, death resulting from other diseases, or refusal to continue the follow-up, 2 patients were lost to follow-up at postoperative month 1, 2 patients at month 3, 1 patient at month 6, and 1 patient at year 2 in the PELF group. In the PELD group, 2 patients were lost to follow-up at postoperative month 1 and 1 patient at month 3 (Table
Time points of the patients lost to follow-up.
Time | PELF group | PELD group |
---|---|---|
1 month | 2 | 2 |
3 months | 2 | 1 |
6 months | 1 | 0 |
2 years | 1 | 0 |
Total | 6 | 3 |
Of all included patients, 2 cases had herniation at L2-3, 1 was included in the PELF and 1 in the PELD group; 6 cases had herniation at L3-4, 5 were included in the PELF and 1 in the PELD group; 76 cases had herniation at L4-5, 29 were included in the PELF and 47 cases in the PELD group; 43 cases had herniation at L5-S1, 25 were included in the PELF and 18 in the PELD group; 3 cases had herniation at both L3-4 and L4-5 levels, 1 was included in the PELF and 2 in the PELD group; 2 cases had herniation at L3-4 and L5-S1 levels, both were included in the PELD group; and 8 cases had herniation at L4-5 and L5-S1 levels, 1 was included in the PELF group and 7 in the PELD group. The distribution of the surgery sides is also presented in Table
The distribution of surgery levels and sides.
Groups | PELF | PELD | ||
---|---|---|---|---|
Levels | Sides | |||
Left | Right | Left | Right | |
L2-3 | 1 | 0 | 1 | 0 |
L3-4 | 4 | 1 | 1 | 0 |
L4-5 | 13 | 16 | 23 | 24 |
L5-S1 | 17 | 8 | 10 | 8 |
L3-4 and L4-5 | 1 | 0 | 2 | 0 |
L3-4 and L5-S1 | 0 | 0 | 0 | 2 |
L4-5 and L5-S1 | 1 | 0 | 5 | 2 |
Total | 37 | 25 | 42 | 36 |
VAS and ODI were utilized to estimate the surgery clinical outcomes. Compared with those preoperatively, the postoperative low back and leg VAS pain ratings and ODI scores significantly decreased over time in both groups (Figures
The low back pain VAS pain rating before and after TF PELF and PELD. (a) Low back pain was significantly decreased at all time points postoperatively compared with that preoperatively in the PELF group (
Leg VAS pain rating before and after TF PELF and PELD. (a and b) Leg VAS pain rating decreased significantly at postoperative all time points compared with that preoperatively in both PELF and PELD groups (
ODI before and after surgery in PELF and PELD groups. (a and b) ODI in both PELF and PELD groups significantly decreased after surgery compared with that preoperatively (
At the final stage of the follow-up, modified Macnab criteria were used to evaluate the recovery at postoperative year 2 for the remaining 131 patients. In the PELF group, 24 cases reported “excellent” (42.9%), 21 cases reported “good” (37.5%), 6 cases reported “fair” (10.7%), and the other 5 cases reported “poor” (8.9%). In the PELD group, 38 cases reported “excellent” (50.7%), 30 cases reported “good” (40.0%), 5 cases reported “fair” (6.7%), and the remaining 2 cases reported “poor” (2.6%). Hence, the satisfactory rate reached 80.4% in the PELF group and 90.7% in the PELD group (Table
Macnab outcome evaluated at the final visit (postoperative year 2) of the follow-up.
Groups | Excellent | Good | Fair | Poor | Total |
|
---|---|---|---|---|---|---|
PELF | 24 (42.9%) | 21 (37.5%) | 6 (10.7%) | 5 (8.9%) | 56 (100%) | 0.329 |
PELD | 38 (50.7%) | 30 (40.0%) | 5 (6.7%) | 2 (2.6%) | 75 (100%) |
Excellent: free of pain and deficit, without restriction of mobility; good: residual symptoms or deficits not impeding a normal life; fair: some improvement in functionality but remained handicapped; poor: no improvement at all.
To further determine whether the injury from the TF PELF procedure would deteriorate the clinical outcomes compared with TF PELD for LDH patients, low back and leg VAS pain ratings, ODI and Macnab outcomes were compared between the 2 groups (Figures
This is a retrospective study to explore a clinical question of whether the damage during the TF PELF procedure would deteriorate the clinical outcomes compared with the TF PELD. Sixty-two LDH-diagnosed patients who received TF PELF and 78 patients who received TF PELD were included for the 2-year follow-up. We found that low back and leg pain VAS pain ratings and ODI scores significantly decreased in both the PELF and PELD groups after surgery, although a fluctuation was observed during the follow-up period. The satisfactory rate was evaluated with modified Macnab criteria at postoperative year 2, which reached 80.4% in the PELF group and 90.7% in the PELD group. However, no significant difference between the two groups of low back and leg VAS, ODI, or satisfactory rate was recorded.
The transforaminal PELD procedure is being developed these years, and the indications of transforaminal PELD are being expanded with the invention and development of the instruments, such as ultrathin high-speed surgical drill, bone remears, cutting forceps, and firing laser. In addition to LDH, this technique can also be utilized to treat lumbar disc stenosis [
Recurrent herniation was defined as (1) patients with a successful PELD confirmed by a pain-free interval of at least 1 month; (2) reappearance of the initial symptoms and MRI evidence of recurrent herniation on the same level [
Despite the evolution, transforaminal PELD cannot be adopted in all patients due to narrow foraminal area and high iliac crest hindered by the L5 transverse process. It was reported that transforaminal PELD could be performed at the L4-5 level in 94.4% (right) and 90.4% (left) patients and only 24.1% and 19.2% at the L5-S1 level [
The disadvantages of PELF were reported as more bleeding and pain, longer operation time, prolonged postoperative recovery time, needing more expensive equipment, and higher risk of heat-damage to the surrounding spinal nerves, including neural injury [
VAS and ODI were evaluated for all patients at each visit during the follow-up. We found that low back and leg VAS pain ratings and ODI scores decreased at postoperative day 1 compared with those preoperatively in both PELD and PELF groups, but both VAS and ODI changed significantly compared with postoperative day 1, suggesting that the symptoms of the patients would fluctuate during postoperative recovery. Low back VAS pain rating increased within 6 months postoperatively compared the first day after operation was observed in the PELD group but not in the PELF group. We postulated that it might be related with a greater range of working channel motion during the procedure in the PELD group, thus causing more damage of the peripheral spinal muscle and even local edema of nerve root, which may extend the recovery period. Nerve root injury occurred in 1 patient in the PELD group, and dural tears occurred in 2 patients in the PELF group, indicating that foraminoplasty did not result in more nerve root or ganglion, but caused complications such as dural tears, but the difference was not significant. No difference was found in Macnab outcomes between the two groups. Therefore, we considered that both TF PELD and PELF were effective and comparable for LDH treatment.
This study has some limitations. (1) Dynamic flexion-extension radiographs were not used to assess stability and hidden dynamic instability after surgeries, especially in the foraminoplasty group. (2) This is a retrospective study without controls from open discectomy, and no valid evidence from randomized controlled trials on the effectiveness of TF PELD and PELF was provided. (3) Randomized controlled trials with longer-term follow-ups compared with other surgical techniques are needed in the future. (4) Despite the LDH-diagnosed patients who reached the inclusion criteria were included in this study, the indication for accepting PELF and PELD is different, the surgeons decided to perform foraminoplasty mainly depending on operation location and experience. If the working cannula could not access the disc fragment due to the anatomical barrier, the foraminoplasty would also be performed. (5) The dimension of foramens of PELF and PELD groups were not recorded in this study, and most of the patients did not receive postoperative CT examine besides MRI, so the authors did not compare pre and postoperative foramens.
Both procedures are demonstrated as safe and effective for the treatment of LDH, and the clinical outcomes of TF PELF and PELD are comparable for LDH treatment. TF PELF would not deteriorate prognosis compared with PELD. However, because of the limitations of the present study, further randomized controlled trials are needed to explore the prognosis of the two procedures in future.
The data used to support the findings of this study are available from the corresponding author upon request.
The authors declare no conflicts of interest.
This work was supported by the National Science Funding of China (81801320) and Wenzhou Science and Technology Project (Y20160392 and Y20130292).