Keywords
sciatica - intervertebral disc displacement - lumbosacral region
Introduction
Open microdiscectomy is the gold standard treatment for lumbar disc herniation. Percutaneous
endoscopic lumbar discectomy has a lot of benefits when compared with open surgery,
such as minor surgical trauma and blood loss, shorter hospitalization, faster recovery,
and lower postoperative morbidity due to the preservation of the dorsal musculature
and of the osteoligamentous structures. This minor tissue trauma during endoscopic
surgery results in faster rehabilitation, leading to lower costs to society.[1]
[2]
[3]
[4]
[5]
The treatment of lumbar disc herniation with advanced disc degenerative disease, which
is represented by more advanced stages in the Modic and Pfirrmann classifications,
still has no consensus regarding the best treatment method between discectomy or arthrodesis.
Taking that into consideration, the objective of the present study is to evaluate
the correlation between radiologic abnormalities (Pfirrmann and Modic) and radicular
pain variation (pre- and postoperative) in patients who underwent transforaminal endoscopic
surgery for lumbar disc herniation.
Materials and Method
This is a study of a series of cases with an initial sample of 80 patients who underwent
surgical treatment for lumbar disc herniation by percutaneous transforaminal endoscopic
approach, between January 29, 2018 (1st entrance of this procedure in the data bank) and august 28, 2019 (last patient with
at least 3 months of follow-up) in a spine endoscopic surgery service.
The inclusion criteria were: radicular pain, failure of 12 weeks of conservative treatment,
and diagnosis of lumbar disc herniation with magnetic resonance imaging (MRI). The
exclusion criteria were: other cause of pain than lumbar disc herniation, previous
lumbar arthrodesis, spondylolisthesis, tumor, infection, lumbar fractures, and Pfirrmann
grades I, II and III. In patients who had lumbar pain and sciatica, only those whose
pain source was mainly radicular were included. Due to the small sample and to the
incapability of association analysis, Pfirrmann grade III patients were excluded.
The selected patients were investigated for demographic data, surgical indications,
and operative details registered in medical records, as well as pre- and postoperative
evaluation by the visual analogue scale (VAS) (varying from 0 to 10, with 0 corresponding
to no pain and 10 to the worst pain ever experienced in the leg), where an improvement
of 2 points was considered as good clinical result.[6]
[7]
The radiological parameters (Pfirrmann and Modic classification) were evaluated by
experienced radiologist specialized in spine diseases, without access to the clinical
features of the patients.
The patients were divided into 3 groups according to the Modic classification (Absence
of Modic, Modic 1 and Modic 2), and into 2 groups according to the Pfirrmann classification
(Pfirrmann IV and Pfirrmann V).
All procedures were performed the same surgeon (Carvalho S. T. C.), who had an experience
of ∼ 30 years of spinal endoscopic surgery. The patients had local anesthesia and
were sedated with propofol and remifentanil; they were positioned in the prone position
on a radiograph-permeable table, under orthograde 2-plane fluoroscopic control, over
a hip and thorax roll to relieve the abdominal and thoracic organs and diminish epidural
bleeding. The operating table can be adjusted for kyphosis intraoperatively at the
lumbar level. The surgeon operated from the side of the disc prolapse, and the video
monitor and C-arm were positioned on the opposite side.
After the position of the iliac crest was determined, the skin incision had a distance
from the midline to the puncture point depending on the level approached (6 to 8 cm
for L2–L3, 8 to 10 cm for L3–L4, and 12 to 14 cm for L4–5 and L5–S1). Following disinfection
and sterile draping, the entry point was marked, always superiorly to the iliac crest,
and a line was drawn across the superior articular process (SAP) to the midline of
the lower endplate.
The spinal needle was inserted orthograde to the disc space. Ideally, when seen in
the fluoroscopy, the tip of the needle should be advanced to the posterior vertebral
body line on the lateral view and to the middle of the medial pedicle line on the
anteroposterior view. With the spinal needle lodged into the disc, the nucleus pulposus
was stained blue (using a 2-mL admixture of contrast media and Methylene blue for
discography), and the surgeon proceeded with the following steps: guidewire passage
through spinal needle; removal of spinal needle; limited incision (8-mm) of skin at
entry site; delivery of tapered cannulated obturator along the guidewire; insertion
of obturator into the disc (on reaching the annulus); advancement of beveled, oval-shaped
working cannula (into disc) along the obturator; and obturator removal. Next, the
endoscope was inserted through the cannula, and the pathologic nucleus (stained blue
for easy distinction and attached to the annular fissure) and any fibrous scar tissue
were released and completely removed using endoscopic forceps and a radiofrequency
device. The working cannula was adjusted to find and remove the hyperplastic superior
facet, the herniated disc, the vertebral posterior edge, and the osteophytes that
existed around the traversing nerve root using a high-speed drill and a bone reamer
or a bone cutter (inside-out).
To be considered sufficient, the decompression had to obtain a nerve showing pulsations
similar to the heart rate and an amount of disc material removed matching the amount
seen on the MRI. When complete, the endoscope was withdrawn and the skin was sutured.
Data was processed in IBM SPSS Statistics for Windows, Version 22.0 (IBM Corp., Armonk,
NY, USA) license #10101131007, with the calculation of the means and standard deviation
(SD). The comparison of the variables of the Modic group were done by the likelihood
ratio and Kruskal-Wallis tests, and the likelihood ratio, the t Student for paired
data and the Mann-Whitney tests were used for the Pfirrmann group. The significance
level considered was p < 0,005.
The present study was accepted by the Ethics and Research Committee under the acceptance
number 4.191.443.
Results
From and initial sample of 80 patients, 39 were selected, and 50 intervertebral discs
were evaluated. There was no difference between the gender of the patients; age: 50,36 ± 15,05
years old; surgery level: L2–L3 1 (2%), L3–L4 2 (4%), L4–L5 9 (18%), L5–S1 8 (16%),
L3–L4 + L4–L5 4 (8%), and L4–L5 + L5–S1 26 (52%); herniation location: right foraminal
7 (14%), left foraminal 15 (30%), central 9 (18%), and diffuse 19 (38%); leg pain:
left 25 (50%), right 11 (22%), and both sides 14 (28%); VAS: preoperative 9,5 ± 0,91,
postoperative 2,5 ± 1,79; surgery duration 100 ± 31,36 minutes; and follow-up: 8,4 ± 6,7
months.
When comparing the groups with the Modic alterations, it was noted that there was
a lower intensity of radicular pain in the postoperative of the Modic 2 group when
compared with the Modic 1 group ([Table 1]).
Table 1
|
ABSENCE OF MODIC
|
MODIC 1
|
MODIC 2
|
p-value
|
|
Gender
|
|
Male
|
11/46%
|
6/25%
|
7/29%
|
0.751[1]
|
|
Female
|
10/38%
|
9/35%
|
7/27%
|
|
Age (years old)
|
50.14 ± 13.58
|
49.67 ± 20.61
|
46.64 ± 13.90
|
0.800[2]
|
|
Surgery Level
|
|
L5–S1
|
5/62%
|
1/13%
|
2/25%
|
0.219[1]
|
|
L4–L5
|
3/33%
|
1/11%
|
5/56%
|
|
L3–L4
|
1/50%
|
1/50%
|
0
|
|
L2–L3
|
1/100%
|
0
|
0
|
|
L4–L5 L5–S1
|
10/38%
|
9/35%
|
7/27%
|
|
L3–L4 L4–L5
|
1/25%
|
3/75%
|
0
|
|
Herniation location
|
|
Right foraminal
|
3/43%
|
3/43%
|
1/14%
|
0.813[1]
|
|
Left Foraminal
|
6/40%
|
3/20%
|
6/40%
|
|
Central
|
3/33.3%
|
3/33.3%
|
3/33.3%
|
|
Diffuse
|
9/47%
|
6/32%
|
4/21%
|
|
Leg pain
|
|
Left
|
10/40%
|
6/24%
|
9/36%
|
0.466[1]
|
|
Right
|
6/55%
|
4/36%
|
1/9%
|
|
Two sides
|
5/36%
|
5/36%
|
4/28%
|
|
Leg Pain Visual Analogue Scale
|
|
Preoperation
|
9.71 ± 0.78
|
9.20 ± 0.86
|
9.64 ± 0.92
|
0.872[2]
|
|
Postoperation
|
2.33 ± 1.68
|
2.93 ± 2.25
|
1.93 ± 1.14
|
p
< 0.0001
[2]
|
|
∆ (Post-Pre)
|
- 7.38 ± 1.77
|
- 6.26 ± 2.31
|
- 7.71 ± 1.54
|
0.183[2]
|
|
Follow-up (months)
|
11.25 ± 7.33
|
7.65 ± 7.89
|
7.36 ± 5.81
|
0.132[2]
|
When analyzing the Pfirrmann groups (Pfirrmann IV versus Pfirrmann V), there was no
difference when comparing the clinical and radiologic characteristics, neither in
the intensity of sciatica ([Table 2]).
Table 2
|
Pfirrmann iv
|
Pfirrmann v
|
p
|
|
Sex
|
|
Male
|
16/73%
|
6/27%
|
0.723[2]
|
|
Female
|
17/68%
|
8/32%
|
|
Age (years)
|
47.45 ± 16.17
|
53.29 ± 16.95
|
0.271[2]
|
|
Surgery Level
|
|
L5-S1
|
7/78%
|
2/22%
|
0.562[1]
|
|
L4-L5
|
7/78%
|
2/22%
|
|
L3-L4
|
2/100%
|
0
|
|
L2-L3
|
1/100%
|
0
|
|
L4-L5 L5-S1
|
13/59%
|
9/41%
|
|
L3-L4 L4-L5
|
3/75%
|
1/25%
|
|
Herniation Location
|
|
Right Foraminal
|
7/100%
|
0
|
0.096[1]
|
|
Left Foraminal
|
10/62%
|
6/38%
|
|
Central
|
7/78%
|
2/22%
|
|
Diffuse
|
9/60%
|
6/40%
|
|
Leg Pain
|
|
Left
|
16/62%
|
10/38%
|
0.201[1]
|
|
Right
|
9/90%
|
1/10%
|
|
Two Sides
|
8/73%
|
3/27%
|
|
Preoperation
|
9.42 ± 0.97
|
9.29 ± 0.91
|
0.539[3]
|
|
Postoperation
|
2.42 ± 1.56
|
2.43 ± 2.20
|
0.737[3]
|
|
∆ (Post-Pre)
|
- 7.00 ± 1.83
|
- 6.85 ± 2.17
|
0.902[3]
|
|
Follow-up (months)
|
8.57 ± 7.30
|
7.77 ± 6.90
|
0.822[3]
|
Discussion
A meta-analysis study with nine randomized clinical trials comparing endoscopic surgery
with the open approach for symptomatic lumbar disc herniation found that the satisfaction
of the patients and the hospitalization time were, respectively, higher and lower
in the group that underwent endoscopic surgery.[8]
When considering lumbar spine anatomy, where the intervertebral foramen dimensions
diminish as the interlaminar spaces increases from L1 to L5; the transforaminal approach
is recommended for high lumbar levels to L3/L4 and for those cases with foraminal
or lateral recess stenosis; the interlaminar approach is suggested for L4/L5 and L5/S1
levels and cases with central and lateral recess stenosis.[9]
The most significant advantage of the transforaminal approach in detriment of the
interlaminar would be the fact that the first can be done under local anesthesia in
an outpatient environment, avoiding the risk of general anesthesia (used for the interlaminar
approach), especially in elderly patients with comorbidities, resulting in lower costs
for the health care of these type of patients.[10]
Xu et al.[11] studied the difference in the clinical results of patients who underwent transforaminal
endoscopic surgery for lumbar disc herniation, considering the Modic alterations.
In the three analyzed groups (control, Modic type I, and Modic type II), there was
a significant improvement in the radicular pain measured by the VAS in 3 months, 1
year and in the last year of postoperative follow-up when compared with the preoperative
values. It is important to mention that there was no difference between the three
groups.
These authors found the presence of nerve root lesion in 5 patients: 3 in the control
group, 1 in the Modic type I, and 1 in the Modic type II, with complication rates
of 6,6, 6,8 and 8% in these groups, respectively. Recurrence happened in 16 patients:
8 in the control group, 4 in the Modic type 1, and 4 in the Modic type II, with rates
of 4,4, 9.1, and 8 in these three groups, respectively.[11]
In our study, a significant improvement in post-operative sciatica was noted in the
three groups analyzed (Absence of Modic, Modic 1 and Modic 2) after transforaminal
endoscopic discectomy. However, there was a statistically significant difference in
the post-operative VAS value between Modic 1 and Modic 2 groups, being of less intensity
in the last group. The difference was not considered clinically significant, because
it wasn't greater than 2 points in the VAS scale.
The present research showed the presence of paresis (strength grade IV on the Medical
Research Council Scale) in six patients preoperatively: two in the Absence of Modic
group, one in the Modic 2, and three not classified. One patient in the Modic 1 group
presented paresis after surgery. These patients were treated conservatively with normalization
of muscle weakness approximately after 3 months of treatment. The recurrence was found
in five patients: three in the Absence of Modic group, one in the Modic 1 group, and
one not classified.
Considering the complications, a multicentered study with > 26,000 cases found a prevalence < 1%
for percutaneous endoscopic discectomy, such as: dysesthesia 0.45%, dural tears 0.17%,
discitis 0.25%, motor or sensitive impairment 0.32%, and recurrence 0.79%.[12]
The general risk factors for recurrence in percutaneous endoscopic surgery are: male
gender, obesity, age > 50 years old, trauma history, and central disc herniation.
However, there are factors related to the surgical technique, such as: less experienced
surgeons (< 200 cases) and the usage of inadequate material.[9]
[13]
In our study, all patients were submitted to the transforaminal technique for lumbar
disc herniation in discs with advanced degeneration (Pfirrmann IV and V). Significant
improvement was noted not only statistically but also clinically in the sciatica pain
level according to the VAS scale in the last postoperative follow-up; however, there
was no difference between the two groups.
In the present research, the endoscopic transforaminal surgery for lumbar disc herniation
with advanced disc degenerative disease, represented by more advanced stages in the
Modic and Pfirrmann classification, showed to be clinically efficient in diminishing
pain (strong preoperatively to mild postoperatively). There was no clinical difference
for those patients who had advanced degenerative discal disease.
The limitations of the present research were the sample size, the relatively short
patient follow-up, and the fact that it was performed in only one specialized center;
however, it still brings important data about advanced degenerative disc disease treated
with a minimally invasive method.
Conclusion
The present study showed that, even in advanced disc degeneration, the transforaminal
percutaneous endoscopic discectomy appeared to be an efficient method in reducing
radicular pain in patients with lumbar disc herniation.
The improvement of the postoperative pain was noticed in all groups analyzed by the
Modic classification (Absence of Modic, Modic 1, and Modic 2) or by the Pfirrmann
classification (Pfirrmann IV and Pfirrmann V), with no clinical difference in the
sciatica reduction between the groups.
