A magnetic resonance imaging-based morphometric analysis of bilateral L1–L5 oblique lumbar interbody fusion corridor: feasibility of safe surgical approach and influencing factors

Article information

Asian Spine J. 2024;18(6):757-764

Publication date (electronic) : 2024 October 22

doi : https://doi.org/10.31616/asj.2024.0193

Archit Goyal, Mayukh Guha, Rajat Mahajan

Department of Spine Services, Indian Spinal Injuries Centre, New Delhi, India

Corresponding author: Archit Goyal, Department of Spine Services, Indian Spinal Injuries Centre, Vasant Kunj, Sector C, New Delhi, India, Tel: +91-7838193675, 91-9643855224, Fax: +91-11-2689 8810, E-mail: clinical.research@isiconline.org

Received 2024 May 30; Revised 2024 August 24; Accepted 2024 September 5.

Abstract

Introduction

Since Briggs and Milligan introduced lumbar interbody fusion in 1944 [1], a wide array of surgical treatments have evolved to address degenerative lumbar diseases, including lumbar intervertebral disc herniation, lumbar spinal stenosis, adult spinal deformity, and degenerative spondylolisthesis. Initially, posterior lumbar interbody fusion and transforaminal lumbar interbody fusion were the primary approaches for lumbar interbody fusion. With the advent of minimally invasive surgical approaches and advancements in surgical instruments, newer techniques have emerged as safer alternatives to conventional procedures. Anterior lumbar interbody fusion (ALIF) refers to a direct anterior retroperitoneal/transperitoneal approach to the lumbar spine. Lateral lumbar interbody fusion (LLIF) involves accessing the disc space via a lateral retroperitoneal, transpsoas corridor, while oblique lumbar interbody fusion (OLIF) involves minimally invasive access to the disc space via a retroperitoneal corridor anterior to the psoas muscle.

OLIF was first introduced by Silvestre et al. [2] as an innovative solution to mitigate the approach-related complications associated with ALIF and LLIF. The OLIF approach utilizes the natural corridor between the psoas muscle and the large vessels to access the intervertebral disc. The procedure involves an anterolateral skin incision, followed by splitting the abdominal wall muscles along their fibers, and navigating through the oblique retroperitoneal corridor between the psoas laterally and the major blood vessels (inferior vena cava [IVC] or common iliac veins on the right and the aorta or common iliac artery on the left). This approach circumvents the need for neuromonitoring and is associated with a shorter operating time, decreased blood loss, lower incidence of intraoperative and postoperative complications, and shorter hospital stays [2–5].

The OLIF approach has garnered considerable interest in recent times. However, due to its relatively novel nature, there is a paucity of data on the local anatomy and morphology of this approach, with most of the studies focusing on the left-sided corridor. Cadaveric studies have an inherent limitation in terms of clinical pertinence due to key morphometric differences with surgical positioning and poor representation of the living muscles and vessel turgor [6–8]. There is a lack of information on the right-sided corridor, the L1–L2 anatomy, and the impact of age and sex on the morphology of the corridors, especially in the Indian population.

This study aimed to investigate the size and local anatomy of both the right and the left-sided oblique corridors between L1–L5 levels and to identify the potential differences in the sizes of the corridors with increasing age and sex in the Indian population utilizing magnetic resonance imaging (MRI). In-depth characterization of the anatomical basis of the technique can help identify patients who are suitable for OLIF. It can also guide the spine surgeons in preoperative assessment of the corridor, thus avoiding intraoperative complications.

Materials and Methods

We conducted this study in compliance with the principles of the Declaration of Helsinki. IRB approval was not necessary for this study at Indian Spinal Injuries Centre, New Delhi, India given the retrospective radiological nature of the study. Informed consent was waived due to the nature of the study. We retrospectively reviewed the imaging data of 300 patients (150 males and 150 females; age range, 21–87 years) who underwent a 1.5T MRI of the lumbar spine between January 2023 and January 2024 at Indian Spinal Injuries Centre, New Delhi, India. Patients with MRI findings suggestive of lumbar spondylosis, lumbar degenerative disc disease, and lumbar canal stenosis were preferred for the review as this patient population may require fusion surgery. Patients with adult/degenerative scoliosis, spondylodiscitis, tumor, trauma, congenital vertebral abnormalities, or a previous history of lumbar or retroperitoneal surgery were excluded. The study cohort was evenly distributed across six age groups (21–30, 31–40, 41–50, 51–60, 61–70, and >70 years) and stratified by sex. Each age group and sex combination consisted of 25 patients. Two fellowship-trained spine surgeons (Dr. A.G. with 5 years of experience in spine surgery and Dr. M.G. with 3 years of experience in spine surgery) independently analyzed the T2 weighted axial MRI images from L1–L5 at the mid-disc level (marked using the midsagittal view). The oblique corridor was defined as the shortest distance between the posterolateral aspect of the aorta or IVC or the iliac vessels and the anteromedial aspect of the ipsilateral psoas muscle (Fig. 1). Both surgeons independently recorded the measurements. The average of the two values was calculated for both the right and the left sides at L1–L2, L2–L3, L3–L4, and L4–L5 levels in males and females. Any obstruction of the corridor by the renal cortex, renal vessels, diaphragm, or lower lobe of the liver was identified and recorded. At the lower lumbar levels, the presence of the Mickey Mouse sign deemed the OLIF corridor unfit for the approach [9]. Data analyses were performed using Stata ver. 16.0 software (Stata Corp., College Station, TX, USA). Continuous variables were presented as mean (±standard deviation) and range. Normally distributed continuous variables were compared using the independent t-test while the Wilcoxon rank-sum test/Kruskal-Wallis test followed by multiple comparisons using Dunn’s test was used for non-normally distributed variables. Intra-group differences were assessed using the Sign-rank test or Friedman test followed by multiple comparisons using the Sign-rank test. All p-values <0.05 were considered indicative of statistical significance.

Fig. 1

(A, B) Measurement of the right and left corridor size.

Results

The mean age of male and female subjects was 50.44 years (range, 21–87 years) and 50.68 years (range, 21–85 years), respectively. Tables 1 and 2 summarize the left and right-sided corridor measurements for male and female patients, respectively.

Table 1

The left and right corridor measurements for male patients in different age groups (N=150)

Table 2

The left and right corridor measurements for female patients in different age groups (N=150)

Effect of sex on the corridor size

The effect of sex on corridor size is presented in Table 3.

Table 3

Effect of sex on the corridor size

Difference between the right and left-sided corridors from L1–L2 to L4–L5

The length of the OLIF corridor was significantly (p=0.0001) longer on the left side compared to the right side at L1–L5 levels in both males and females (Tables 4, 5). The left L1–L2 corridor was obstructed by the renal vasculature in 96/150 males (64%) and 90/150 females (60%) and by the renal cortex in 11/150 males (7.3%) and 5/150 females (3.3%). The right L1–L2 corridor was obstructed by the renal vasculature in 100/150 males (66.6%) and 72/150 females (48%) and by the renal cortex in 36/150 males (24%) and 25/150 females (16.6%). The left L2–L3 corridor was obstructed by the renal vasculature in 5/150 males (3.3%) and 5/150 females (3.3%) and by the renal cortex in one male and one female. The right L2–L3 corridor was obstructed by renal vasculature in 3/150 males (2%) and 4/150 females (2.6%) and by the renal cortex in 9/150 males (6%) and two females. The right corridor at L1–L2 was obstructed by the lower lobe of the liver in one patient and the diaphragm in two patients. The left corridor at L1–L2 was obstructed by the aortic bifurcation in one patient. The L4–L5 corridor was inaccessible due to the bulk of psoas muscles (Mickey Mouse sign) in 46/150 male patients (30.6%) and 100/150 female patients (66.6%) (Fig. 2). In two patients, the right L4–L5 was obstructed by the confluence of the iliac veins. At L3–L4, five female patients exhibited the Mickey Mouse sign, making the corridor unsuitable for surgery.

Table 4

The difference in Left corridor size from L1–L2 to L4–L5 in males and females

Table 5

The difference in Right corridor size from L1–L2 to L4–L5 in males and females

Fig. 2

(A, B) Mickey Mouse sign.

Discussion

Lumbar fusion surgery is the most widely practiced spine surgical procedure worldwide. The OLIF approach leverages a natural oblique corridor, bounded by the aorta-IVC (or common iliac vessels) anteromedially and the anteromedial aspect of the psoas muscle posterolaterally. Both OLIF and anterior to the psoas approaches utilize the prepsoas corridor for access. However, a key distinction lies in the technique: anterior to the psoas involves safe vascular retraction for access, while OLIF allows access with mild psoas retraction without any vascular mobilization. However, this approach is also not without complications. Several studies have documented incidents of vascular injury with the OLIF approach [2,7,10,11]. Other complications such as postoperative hip flexion weakness, numbness in the anterior thigh and groin, injury to the ureter, great vessels, segmental artery, iliolumbar vein, peritoneal breach, and postoperative ileus have also been reported [12]. Therefore, meticulous surgical technique and a thorough understanding of the anatomy are essential to minimize complications.

Multiple MRI-based, computed tomography-based, and cadaveric morphometric studies have investigated the oblique corridor used in the OLIF approach. To the best of our knowledge, no study in the Indian population has analyzed the differences in the right and left-sided OLIF corridor with subgroup analysis by age, sex, and the lumbar disc level. Molinares et al. [8] and Zhang et al. [13] showed that the largest corridor is present at the higher lumbar levels, and the size of the corridor decreases at the lower lumbar levels. This result is expected primarily because of the conical morphology of the psoas major muscle. Our results are consistent with those of the above studies. In males, the left-sided corridor size significantly decreased from L1–L2 to L4–L5 levels in 21–30, 31–40, 41–50, and 51–60 age groups. Patients in the age group 61–70 years had no significant size difference between the upper lumbar levels but the L4–L5 level had a significantly smaller corridor size. For patients in the age group >70 years, the size difference between the upper and lower lumbar levels was insignificant owing to the larger corridor size at L4–L5 due to psoas atrophy. In female patients, the left corridor size was not significantly different at upper lumbar levels; however, the L4–L5 corridor was significantly smaller in all age groups compared to the upper lumbar levels. The significantly smaller corridor size at L4–L5 in elderly females may be attributed to the less voluminous psoas muscle in females compared to males due to which age-related muscle atrophy does not significantly increase the corridor size.

On the right side, in male patients, the size of the corridor was found to significantly decrease from L1–L2 to L4–L5 in all age groups. In females, the corridor size difference was insignificant at the upper lumbar levels but it was significantly smaller at the L4–L5 level for all groups. The corridor size increased with age, similar to the left side, however, the usual size of the corridor at L2–L3, L3–L4, and L4–L5 was still inadequate for an OLIF approach in the majority of patients, requiring major retraction of the psoas major or manipulation of IVC, both of which carry an inherent risk of neurological (genitofemoral nerve and lumbar plexus) and vascular injury. The corridor size on the left side was significantly different between males and females at the upper lumbar levels but at L4–L5 the difference was statistically significant only in the >70 age group. Similarly, on the right side, there was a significant difference in size based on sex at the upper lumbar levels but not farther down in any age group. A computed tomography angiography-based study by Liu et al. [14] found significantly larger corridor measurements in males at all levels from L1–L5. However, Molinares et al. [8] and Zhang et al. [13] found no significant difference between males and females in this respect. At L1–L2/L2–L3, the corridor on the right was obstructed by the renal vasculature, ipsilateral kidney, the lower lobe of the liver, or the diaphragm, whereas, on the left side, it was obstructed by the renal vasculature, ipsilateral kidney, or rarely by a higher bifurcation of the aorta (Fig. 3). Silvestre et al. [2] reported that the L1–L2 disc can only be approached in selected cases where the floating ribs are relatively horizontal and mobile due to the obstruction by the rib cage. These findings were consistent with those of Julian Li et al. [15], who conducted a similar analysis of the OLIF working corridor on the left and right sides using MRI. Their study also revealed that the OLIF corridor tends to be larger on the left side compared to the right side. Zhang et al. [13] investigated the effect of patient positioning on the size of the surgical corridor using MRI, revealing a decrease in the oblique corridor size on transitioning from the right lateral decubitus position to the supine position. Their findings suggest that relying solely on supine position MRI scans to assess the morphometric features of the oblique corridor may lead to inaccuracies.

Fig. 3

(A, B) Corridor obstruction by renal vasculature.

Some limitations of this study should be considered. The preoperative MRI studies were obtained with the patients in the supine position, whereas OLIF surgery is performed in the lateral decubitus position, which may affect the corridor sizes. Additionally, MRI measurements may not accurately reflect intraoperative sizes, as even slight retraction of the psoas major muscle may significantly affect the access corridor size. The small cohort size in each group may not be representative of the population parameters. Thus, multicenter studies with a larger population may provide more precise measurements. The single-center scope of the study may limit the generalizability of the findings. Lastly, the study did not account for the impact of patient height and weight on the corridor size.

Conclusions

OLIF approach represents an innovative solution for lumbar spinal fusion surgery, mitigating approach-related complications associated with ALIF, LLIF, and posterior spinal fusion techniques. However, the right-sided approach may not be a viable option for most patients. In-depth preoperative evaluation of the left-sided corridor anatomy can help reduce the risk of complications. The left L1–L2 and L2–L3 corridors appear to be morphometrically suitable for most patients, but preoperative planning must consider potential obstructions from renal vasculature, kidney, lower lobe of the liver, and diaphragm. The left L3–L4 and L4–L5 corridors may be more suitable for the elderly population, especially male patients.

Article information Continued