Examination of Spinal Canal Anatomy with MRI Measurements in Lomber Disc Herniation Patients: An Anesthesiologist Viewpoint.

BACKGROUND AND AIM: The aim of this study is to investigate the magnetic resonance imaging (MRI) of patients with lumbar disc herniation (LDH) to identify the challenges associated with neuraxial anesthesia.
MATERIALS AND METHODS: The MRI images in the supine position of 203 patients admitted to hospital with complaints of lower back pain were studied. Medial sagittal slices of the lumbar spine were imaged from L1 to S1. LDH is classified as either bulging, extrusion, or protrusion.
RESULTS: For this study, 83 males and 120 females with a mean age of 43.18 ± 14.68 years were recruited. The highest herniation level was observed at L4-L5 in 145 (71.4%) patients: 76 instances of disc bulging (37.4%), 56 instances of extrusion (27.6%), and 13 instances of protrusion (6.4%). The longest distance between the skin and spinal cord was 60.06 ± 1.61 mm at L5-S1; the longest distance at width of the epidural space was 6.09 ± 1.95 mm at L3-L4. According to the disc herniation groups, no significant differences were found between the skin-to-dura distance, width of the epidural space, and depth of skin level to spinous process (P > 0.05). Moreover, the anterior dura to cord distances was significantly different from normal patients (P < 0.05). Indeed, there was a statistically weak and negative correlation between both the length and age of the lumbar spinal canal (P < 0.05, r = -0.295).
CONCLUSIONS: Lumbar disc pathologies can cause anatomical derangements in the spinal canal, which may cause neurologic deficits by neuraxial blockade. Copyright:

INTRODUCTION

Neuraxial anesthesia is used for all types of medical procedures (and pain treatments) under the neck level, either independent of, or in conjunction with, general anesthesia.[1] Compared to general anesthesia, neuraxial anesthesia is often preferred due to the fact that, in the early postoperative period, it has lower mortality rates, fewer thromboembolic complications, higher levels of patient satisfaction, better hemodynamic stability, shorter hospital stays, as well as minimal costs and side effects (nausea, vomiting, etc.).[2] Moreover, serious complications are rare, such as spinal hematoma, spinal stenosis, epidural abscess, meningitis, traumatic cord injury, and cauda equina syndrome.[34] To be sure, this is one complication that is concerning; namely, the possibility that the previous neurological deficit worsens, or else a new neurological deficit develops. For this reason, the current neurological deficit is a relative contraindication, due to the fact that it is difficult to identify new neurological deficits that may occur in the perioperative period. Although there are not any controlled studies of regional anesthesia increasing neurological deficits, both case presentation studies and review studies are readily available.[5] Indeed, lower back pain is a troubling health problem that can result in a poor quality of life as well as disability.

One of the most common causes of lower back pain is lumbar disc herniation (LDH), with a reported prevalence range of 20%.[6] Moreover, magnetic resonance imaging (MRI) of asymptomatic patients suggests that disc bulging and protrusions are often responsible.[7] Indeed, a lumbar MRI will yield accurate results with respect to detecting intervertebral disc anomalies; however, the clinical implications of said results are controversial.[8] Unfortunately, a LDH can cause morphological changes in spinal canal structures, such as the narrowing of the spinal canal, lateral recess, or neural foramen due to chronic degenerative changes.[9] In fact, these changes may cause permanent neurological deficits and they may, moreover, be problematic in terms of regional anesthesia.

The primary aim of this study is to evaluate the MRI images of both patients with lower back pain and LDH to determine the challenges associated with neuraxial anesthesia.

MATERIALS AND METHODS

The approval for the research was granted by the ethics committee, and the MRI images of 203 patients admitted with lower back pain or radiculopathy complaints, between January 2017 and December 2018, were reviewed retrospectively.

Study design

In this study, patients over the age of 18 years who underwent a lumbar MRI due to lower back pain were recruited. Patients with spinal cord or vertebral column pathology (advanced degenerative changes, spinal stenosis, spondylolisthesis, facet arthritis, spondyloarthritis), tumor, infection, ischemia or spinal canal hemorrhage, and congenital anomalies (transition anomalies, spina bifida, etc.) were excluded. Moreover, patients who had previously undergone lumbar vertebral column surgery were also excluded from the study, due to the associated high error rate.

Magnetic resonance imaging

A 1.5-T unit device (Sigma, General Electric, Milwaukee, WI, USA) with a synergy spine coil was used for the MRI of the lumbar spine. The sagittal series of the lumbar spine MRI scans were obtained with a field of view of 350 mm (matrix 256 × 256, T1 weighted: 4-mm slice thickness, NSA 2., T2 weighted: 4-mm slice thickness, NSA 3.) For the most medial section, the vertebrobasilar vein images of each patient were assessed by an independent radiologist, and the distances of the relevant structures on the image were measured [Figure 1]. The measurements of this study were made electronically using the Philips Sectra PACS system.

Figure 1

Magnetic resonance imaging of lumbar spine and medial sagittal. This figure is an actual image from an examined patient. The arrows indicate the measured distances found in the study: (A) depth of skin level to spinous process, (B) skin to posterior dura, (C) width of the epidural space, (D) posterior dura to spinal cord, (E) cord diameter, and (F) anterior dura to cord

Evaluation of disc herniation

Using the MRI, the interval from the first lumbar vertebral (L1) level to the sacral first vertebral (S1) level was evaluated. The herniation levels of the patients between L1 and S1 were evaluated using the lumbar disc terminology outlined by Fardon and Milette.[10]

Normal disc definition includes normal morphological discs, which are independent of the clinical setting of the patient and degenerative, developmental, or adaptive changes (e.g. normal aging, scoliosis, spondylolisthesis)

Herniation is defined as the herniated disc material extending outside of the intervertebral disc circumference. The disc material may be apoptotic bone that is fragmented, nucleus, annular tissue, or cartilage. The images in the axial plain are defined as “localized” or “focal” in which less than 25% of the disc material extend from the disc circumference

Bulging is defined as the displacement of the disc material greater than 25% of the disc frame

Protrusion occurs when the largest distance of disc material beyond the disc space is less than the disc material beyond the base of the disc space. The weakened and torn annulus is the backward displacement of the disc material

Extrusion occurs when the largest distance of disc material outside the disc space is greater than the disc material beyond the base of the disc space. The extruded disc may migrate up or down behind the adjacent vertebra corpus, though it retains its connection with the main disc. The nucleus has passed the ruptured annulus.

Data collection

After an MRI was performed on the patients by an experienced radiologist, measurements between the L1 and S1 intervertebral levels were grouped under the following headings: Distance between the skin to posterior dura, width of the epidural space measured along the spinal needle placement, anterior dura to spinal cord distance, and depth of skin level to spinous process [Figure 1]. The lumbar spinal cord is continuous at one level only (L1); accordingly, this was where the posterior cord to dura distance along the needle was measured, as well as the posterior dura to cord distance and the spinal cord diameter.

Statistical analyses

The normality of the data was assessed using the Shapiro–Wilk test, with the normally distributed data evaluated using parametric tests and the nonnormally distributed data evaluated using nonparametric tests. Categorical data were evaluated using the Chi-square test, with data given as n (%). Comparisons of normality-matched data were performed with an independent t-test; the data were given as a mean value (± standard deviation). The data, moreover, were presented in graphical format. Correlation tests were performed according to different elements, such as age and gender. A Kruskal–Wallis test was used to compare the four groups in our study. The group pairs were then tested with a Mann–Whitney U-test after the differences were obtained using the aforementioned Kruskal–Wallis test. Moreover, a linear Pearson test was conducted separately for each intervertebral disc level. Nonnormality data were compared using both the Mann–Whitney U-test and the Wilcoxon test; the data were given as median value (minimum–maximum). Differences were considered significant if P < 0.05. Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) version 22.0 for Windows (SPSS Inc., Chicago, IL, USA).

RESULTS

The MRI images of 203 patients were used in this study. The patients consisted of 83 males and 120 females, with a mean age of 43.18 ± 14.68 years, which is a total range of 18–83 years [Table 1]. The MRI showed no herniation between L1 and S1 in 27 patients. Overall, 42 patients had at least one level of herniation, 52 had at least two levels, 44 had at least three levels, 23 had at least four levels, and 15 had at least five levels. Moreover, of these herniation instances, 141 patients had bulging herniations, 103 had protrusions, and 27 had extrusions [Table 1].

Table 1

Demographic data of the patients

Characteristic	n (%)
Age (years)	43.18±14.68 (18-83)
Gender (male/female) (%)	83/120 (40.8/59.2)
MRI disc changes
Disc bulging	141 (69.45)
Disc protrusion	103 (50.73)
Disc extrusion	27 (13.30)
Level of herniation
No herniation	27 (13.30)
1 intervertebral level	42 (20.68)
2 intervertebral level	52 (25.61)
3 intervertebral level	44 (21.67)
4 intervertebral level	23 (11.33)
5 intervertebral level	15 (7.38)
Total	203

Mean±SD. n=Patient number, MRI=Magnetic resonance imaging, SD=Standard deviation

The herniation types were classified as the following for L1–S1: No herniation, bulging, extrusion, and protrusion [Table 2]. The lowest percentage of herniation was found at L2–L3 in 57 patients (28.1%) without any herniation types. The highest percentage of herniation was observed at L4–L5 in 145 patients (71.4%). Moreover, at L4–L5, 76 patients (37.4%) had disc bulging, 56 (27.6%) had extrusion, and 13 (6.4%) had protrusions.

Table 2

The percentage of disc herniation types at each intervertebral level

Questions	L1-L2, n (%)	L2-L3, n (%)	L3-L4, n (%)	L4-L5, n (%)	L5-S1, n (%)
No herniation	91 (44.8)	146 (71.9)	111 (54.7)	58 (28.6)	91 (44.8)
Disc bulging	66 (32.5)	36 (17.7)	60 (29.6)	76 (37.4)	66 (32.5)
Disc extrusion	37 (18.2)	19 (9.4)	29 (14.3)	56 (27.6)	37 (18.2)
Disc protrusion	9 (4.4)	2 (1.0)	3 (1.5)	13 (6.4)	9 (4.4)
Total	203	203	203	203	203

Mean±SD. SD=Standard deviation

The longest distance of skin to posterior dura is 60.06 ± 1.61 (28–106) mm at L5–S1, and the shortest distance is 40.55 ± 1.01 (22–76) mm at L1–L2. The longest width of the epidural space is 6.09 ± 1.95 (1–14) mm at L3–L4, and the shortest distance is 3.39 ± 1.86 (1–11) mm at L5–S1 [Table 3]. The longest anterior dura to spinal cord distance is 9.05 ± 4.09 (0–18) mm at L5–S1, and the shortest distance is 5.90 ± 2.7 (0–12) mm at L3–L4. The distance from skin level to spinous process at L5-S1 level is mean 29.5± 15.39 (1–80) mm, and the shortest distance is at L1-L2 level mean 13.87 ± 9.29 (2–47) mm.

Table 3

The measurement of distances between various anatomical structures in the lumbar spine

Questions	L1-L2	L2-L3	L3-L4	L4-L5	L5-S1
Skin-posterior dura distance (mm)	40.55±1.01 (22-76)	44.15±1.02 (25-92)	50.32±1.37 (27-105)	59.23±1.57 (28-110)	60.06±1.61 (28-106)
Width of the epidural space (mm)	4.91±1.83 (1-11)	5.65±1.94 (1-11)	6.09±1.95 (1-14)	4.66±1.89 (1-13)	3.39±1.86 (1-11)
Anterior dura to spinal cord distance (mm)	6.09±2.29 (0-11)	6.04±2.60 (0-13)	5.90±2.73 (0-12)	6.53±3.73 (0-16)	9.05±4.09 (0-18)
Depth of skin level to spinous process (mm)	13.87±9.29 (2-47)	14.47±1.01 (1-63)	18.59±12.5 (2-67)	25.42±15.59 (2-70)	9.50±15.39 (1-80)

Mean±SD

The skin to posterior dura distances, width of the epidural space, the anterior dura to spinal cord distances, and depth of skin level to spinous process of the patients were assessed separately according to the herniation classification for each lumbar level [Table 4]. There were no significant differences between the intervertebral levels of the skin to posterior dura distances, width of the epidural space, and depth of skin level to spinous process (P > 0.05). Indeed, the only significant difference was found in the anterior dura to spinal cord distances between the normal and herniated patients (P < 0.05).

Table 4

Analysis of patients according to disc herniation classification at each intervertebral level

Characteristics	L1-L2	L2-L3	L3-L4	L4-L5	L5-S1
Skin-posterior dura distance (mm)
No herniation	40.48±0.77	43.95±0.89	549.86±1.86	60.34±2.16	61.76±1.85
Disc bulging	38.13±1.83	45.05±1.51	53.01±2.05	61.50±1.92	58.57±1.72
Disc protrusion	48.75±4.19	43.84±2.0	46.75±1.52	56.67±1.84	59.51±2.62
Disc extrusion	41.0±2.0	46.0±3.0	46.0±7.21	52.00±3.47	56.0±5.40
P	0.87	0.938	0.185	0.109	0.538
Width of the epidural space (mm)
No herniation	4.95±0.14	5.54±0.15	6.08±0.17	4.63±0.26	2.13±0.12
Disc bulging	4.93±0.41	5.62±0.35	6.05±0.26	4.72±0.20	2.48±0.31
Disc protrusion	3.87±0.54	6.57±0.55	6.17±0.43	4.50±0.25	2.93±0.28
Disc extrusion	5.50±0.50	5.50±0.50	6.66±1.45	5.15±0.43	2.22±0.49
P	0.418	0.191	0.954	0.714	0.162
Anterior dura to spinal cord (mm)
No herniation	6.43±0.16	6.78±0.128	7.26±0.20	4.63±0.26	2.13±0.12
Disc bulging	4.60±0.54	4.52±0.42	4.45±0.31	4.72±0.20	2.48±0.31
Disc protrusion	4.12±0.89	3.78±0.48	3.91±0.50	4.50±0.25	2.93±0.28
Disc extrusion	2.50±0.50	5.0±0.50	4.00±0.57	5.15±0.43	2.22±0.49
P	<0.01*	<0.01*	<0.01 *	<0.01 *	<0.01**
Depth of skin level to spinous process (mm)
No herniation	13.84±0.72	14.52±0.81	18.50±1.12	26.28±2.13	31.63±1.72
Disc bulging	11.17±1.47	15.91±1.51	21.11±1.91	27.28±1.91	26.92±1.66
Disc protrusion	22.62±3.47	13.68±1.78	14.58±1.62	23.44±1.75	29.70±2.65
Disc extrusion	13.0±1.0	15.50±0.50	10.66±4.70	19.19±4.14	26.0±4.38
P	0.93	0.834	0.084	0.238	0.254

*P<0.05, **P<0.01, statistically significant, Kruskal-Wallis test

The mean lumbar spinal canal length was 158.95 ± 9.29 (130–185) mm. The distance between the posterior dura and the spinal cord was measured as 1.32 ± 0.33 (0.7–2.50) mm, while the same distance measured from the slope along the spinal needle pathway was 1.92 ± 0.45 (1.20–4.0) mm [Table 5].

Table 5

The measurement of the distances between various anatomical structures in the lumbar spine and their statistically significance values between lumber disc herniation classifications

Characteristics	L1-L2	P
Posterior dura to spinal cord length (mm)	1.32±0.33 (0.7-2.50)	0.832
Spinal cord diameter	7.42±0.92 (5-10)	0.322
Posterior dura to cord length along the needle pathway	1.92±0.45 (1.20-4.0)	0.595
Length of lumbar spinal canal	158.95±9.29 (130-185)	0.273

A statistically significant difference was found between the normal pathway and the spinal needle pathway lengths of the posterior spinal cord and the spinal cord, which was calculated using a paired t-test (P < 0.05). Indeed, there were no significant differences between the groups with respect to the normal, bulging, extrusion, and protrusion classifications (P > 0.05).

There was a significantly negative correlation between the lumbar spinal canal length and age (P < 0.05, r = −0.295) [Figure 2]. Figure 3 depicts the graphical plots (box and whisker) of the anterior dura to spinal cord distances and the epidural space distances of each lumbar intervertebral level.

Figure 2

Correlation of the lumbar spinal canal length with age (years)

Figure 3

The box and whisker plots of the anterior dura to spinal cord distance and the epidural space of each lumbar intervertebral level

DISCUSSION

In this study, we investigated the lumbar MRI of patients admitted with lower back pain. The following results were obtained:

The highest ratio of hernias was found at L4–L5

The L5–S1 intervertebral level is the longest distance between the skin to spinal cord and the depth of skin level to spinous process

The anterior dura to spinal cord distances was lower in patients with some form of disc herniation compared to patients with no herniation (P < 0.05)

A significantly negative correlation was observed between age and lumbar spinal canal length

The disc degeneration of the lumbar spine is an important factor in causing recurrent lower back pain.[11] Although there are different morphologic classifications of lumbar disc degeneration, we used the classification of bulging, extrusion, and protrusion in the posterior aspect of the MRI.[12] Lumbosacral intrathecal anatomy has a complex structure due to the dense placement of nerve roots as well as frequent damage to this area. Indeed, disc hernias were common at L4–5, which is consistent with the majority of related studies.[13] Historically, anesthesiologists have medicolegally avoided using a neuraxial blockade for patients with LDH. This is because of the concern that the neuraxial block may increase a patient's preexisting neurological deficits or else cause new neurological damage. The majority of patients with LDH are skeptical of spinal or epidural anesthesia, due to the fact that both methods can worsen lower back pain. After all, case presentation studies indicate an increased risk of neurological complications in patients with spinal instability or LDH. Here, possible damage mechanisms are assumed to be ischemic, mechanical trauma, local anesthetic toxicity, or multifactorial etiology.[3141516] The possible pathophysiologic mechanism was first defined by Upton and McComas in 1973; a damaged neural nerve is particularly susceptible to further damage in another region along the nerve line. This was named the double crush phenomenon.[17] Hebl et al. concluded that spinal or epidural anesthesia with respect to patients with spinal canal pathology and LDH is associated with increased neurological complications (1.1%; 95% confidence interval: 0.5% −2.0%).[16]

Changes in the anatomy of the spinal canal may present some difficulties with respect to a neuraxial blockade. The lower lumbar levels have a crescentic pattern, while the nerve roots form a group when moving toward the cephalad. Lower sacral levels have a dorsal appearance and discrete stalks of thecal sheet.[18] We think that these morphological changes may cause difficulties in neuraxial anesthesia with the contribution of the degenerative changes caused by disc hernias in the spinal canal. Arslan et al. hypothesized that as the L4 and L5 nerve roots pass through more mobile spine sections until they exit the spine, they may suffer more traumatic damage at the degenerative disc levels during flexion.[19] Since these intrathecal nerve roots do not contain structures such as endoneurium and perineurium that protect neural tissues from mechanical damage, they can be exposed to stretch-induced injury.[19] Therefore, the movement of the lumbar flexion in the neuraxial block processes becomes more vulnerable to degenerative intervertebral disc injuries, such as advanced disc protrusion. In our study, it was observed that the anterior dura to spinal cord distance decreased significantly in patients with disc herniation. Accordingly, intrathecal nerve tissues are compressed, which increases intrathecal pressure, as well as the risk of both thecal sac damage and nerve damage. Because of this, we do not recommend the application of neuraxial blockage at the level of disc herniation.

The differences in the termination of spinal cord increase the risk of nerve damage in neuraxial block procedures. Conus medullaris ends at the level of vertebral body L1 in thecal sheet in adults. However, this placement can vary between T12 and L3 vertebral levels.[20] Our study suggests that the L1 level is the termination level in the majority of patients. The conus medullaris then forms the structures of the cauda equina and filum terminale. Frequently, the termination of the dural sac was at the S2 level.[21] Indeed, we determined that the spinal canal length varies between S1 and S4 in different patients, and the length increases with increasing age. Female patients with lumbar disc bulging were shown to have a lower conus medullaris terminal in our study. This may be a random incident, so there is a need to use a large number of patients to justify this result.

The epidural space extends from the base of the skull (as a cylindrical structure) to the S2 level, where the dural sac ends. Many procedures are performed in the epidural space, such as analgesia, anesthesia, and pain therapies. It is located between the dura and ligamentum flavum, and it contains fat, spinal veins and arteries, loose areolar connective tissue, and lymphatics.[22] The dimensions of the epidural space vary according to the position and level of the intervertebral. In adults, its length is roughly 0.4 mm at C7–T1, 7.5 mm at the upper thoracic levels, 4.1 mm at T11–12, 4–7 mm at the lumbar region, and up to 2 mm at S1. The lumbar epidural space begins at the lower level of L1 and continues to the upper level of the S1 vertebra. The epidural space in the lumbar region varies in size, from 4 to 7 mm.

In our study, the width of the epidural space between L1 and S1 was 4.91 ± 1.83 mm (L1–L2), 5.65 ± 1.94 mm (L2–L3), 6.09 ± 1.95 mm (L3–L4), 4.66 ± 1.89 mm (L4–L5), and 3.39 ± 1.86 mm (L5–S1), respectively. No statistically significant differences were found in the distances between the groups of patients with intervertebral disc herniation, possibly suggesting that the epidural blockade is safe to conduct. In animal studies, inflammatory reaction was detected in the epidural space in 82% of the pathologies of animals with disc extrusion.[23] However, results of the disc material examination of humans are controversial. After all, it is thought that the lysosomal enzymes secreted from inflammatory mediators and macrophages in the epidural space with respect to LDH are the main cause of chronic back pain.[24] Accordingly, it is unclear whether inflammatory reactions affect the epidural block dynamics. Moreover, although there is no significant difference between the sizes of epidural space between the disc herniation types, it is currently unknown how the disc herniated mass and the inflammatory reaction affect the epidural pressure (normally negative pressure).

This study has limitation that the study design was retrospective, there was no data about the heights and weights of patients, which might affect the measurement results. After all, the patients' measurements were taken in several positions (the supine, upright, lateral decubitus, flexion, extension, etc.) which can change according to a person's weight and height.[25]

CONCLUSIONS

According to the data, we obtained in our study, neuraxial anesthesia applications in patients with LDH pose difficulties due to the anatomical changes of the disc herniation in the spinal canal. Therefore, we do not recommend applying neuraxial anesthesia from the vertebral column segment containing LDH. Indeed, intervertebral disc pathologies constitute a large challenge for anesthesiologists. Because of this the high incidences in the general population, the morphological changes associated with disc degeneration in the spinal canal, and the risk of worsening existing neurological deficits. For these reasons, in the preoperative examination, anesthesiologists should evaluate LDH patients in detail with respect to the risk of postoperative neurological deficits. Moreover, guidelines should be prepared for this topic, with a large series of studies in terms of the possible medicolegal risks for anesthesiologists.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.