INTRODUCTION
Degenerative lumbar spinal canal stenosis with instability requires decompression and fusion of the affected segment. It can be achieved through several approaches including anterior as well posterior approaches. Minimally invasive oblique lumbar interbody fusion (MIS OLIF) is one such approach which was first described by Mayer [1] and Silvestre et al. [2]. It is a retroperitoneal approach which uses interval between great vessels and psoas muscle. Because it is anterior to psoas, it does not have complications related to lumbar plexus injury seen in transpsoas approaches (lateral lumbar interbody fusion/direct lateral interbody fusion) and thus neuromonitoring is not required while performing MIS OLIF [3-5]. Apart from this OLIF is an effective technique even for sever lumbar canal stenosis working on principle of ligamentotaxis with intact anterior longitudinal ligament and posterior longitudinal ligament and opening of disc space, foraminal height and restoration of segmental lordosis [6]. Thus, OLIF is an effective method of indirect decompression [7,8]. It avoids all complications related to direct decompression like more muscle damage, bony resection of lamina and facets, bleeding, nerve root injuries, dural tear, epidural bleeding, hematoma, deep wound infection, perineural fibrosis etc seen in conventional posterior approaches [9,10]. However, it can have its own set of complications including vessel injury, ureteric injury, endplate damage, peritoneal injury etc. [11,12]. Also, poor case selection can lead to failure of effective indirect decompression. Lack of knowledge regarding approach-related complications as well as ideal patient selection can pose challenge especially in early learning curve. In this article author discusses about early learning curve experience of MIS OLIF performed in degenerative lumbar disorders. This includes ideal case selection specifically where there can be apprehension on efficacy of indirect decompression and techniques to avert potential known complications.
MATERIALS AND METHODS
It was a retrospectives analysis of 22 patients and their 24 segments. It was a retrospective analysis of clinicoradiological outcomes of patients who underwent MIS OLIF. It was conducted from 2020 to 2023 over period of 26 months with longest follow-up of 3 years. In all these patients, MIS OLIF was done from left side and percutaneous pedicular fixation was done in the same stage. Inclusion criteria were failed conservative line of management for low back pain with neurogenic claudication in lower limbs which got relieved with rest. All the patients exhibited lumbar canal stenosis with instability on radiological assessment done with dynamic x-rays and magnetic resonance imaging (MRI). Exclusion criteria were rest pain, infection, trauma, spondylolisthesis more than grade 3 and acute lumbar disc prolapse. All patient’s OLIF working operative corridor (space between iliac vessels and psoas major) was measured on axial MRI images and criteria of 10 mL was kept for MIS OLIF as an option of treatment of choice. All included patients were counselled for the requirement of direct decompression in case of persistent symptoms of neurological claudication after OLIF. Clinical and neurological assessment of all the patients was done. Radiological analysis was done using standardized plain anterior-posterior (AP) and lateral dynamic x-rays and MRI of lumbosacral spine. Appropriate size of cage was inserted using retroperitoneal oblique corridor under image intensifier television (IITV) guidance. Artificial bone graft (B tricalcium phosphate) with bone marrow aspirate was used in all of 22 patients. After cage insertion, fixation of the lumbar segment was done in the same stage. Posterior percutaneous fixation was done in 19 patients whereas anterior fixation was done in 3 patients. No direct decompression was done. No neuromonitoring was carried out.
Estimated blood loss (EBL), operative time, and all known perioperative complications were noted. Follow-up was done at 1, 3, 6 months and at 1 year. Postoperative clinical assessment was done using Modified Macnab criteria at 3 months. Postoperative x-rays of lumbosacral spine AP and lateral view were taken in all patients. X-rays were taken at 1-month, 3-month, 6-month, and 1-year follow-up. Immediate postoperative MRI was done in all patients. Radiological fusion was confirmed at 1 year on x-ray. Patients who had lumbar canal stenosis Schizas grade D underwent for MRI postoperatively twice: immediate postoperative period and at 1-year follow-up. Assessment of radiological parameters: Percentage improvement in disc height, foraminal height, segmental lordosis on x-rays, reduction of slippage of vertebrae (Figure 1) and increase in overall area of spinal canal area were noted to assess indirect decompression (Figure 2). All results were statistically analyzed using paired t-test. Surgimap was used for calculation of parameters. Patients were divided into 2 groups. Group one had first 10 patients as initial part of the learning curve and second group had later 12 patients later part of the learning curve. Comparison of MRI picture, clinical outcomes, blood loss, operative time for cage insertion and known perioperative complications were done between the 2 groups. Informed written consent has been obtained from all the patients.
1. OLIF Procedure
The procedure has been described in detail in literature however noteworthy points considering learning curve shall be described here. All patients were kept in right lateral decubitus with hip extended. Before beginning the procedure, the desired disc space was localized in dead AP and lateral images under IITV. In patients with high iliac crest where access to L4–5 level can be difficult; placement of a roller-pack was done below L4–5 level which will increase the space between iliac crest and disc space (Figure 3A). However, once disc preparation was complete, roller-pack was removed before final cage placement as it will lead to fusion in iatrogenic scoliotic deformity. Incision was placed 2 fingerbreadths anterior to disc space. Usual size of incision was 3–4 centimeters for single level and it did not vary much in obese/thin patients. Abdominal muscles were split and care was taken after splitting transverses abdominis. Only finger dissection was done after that and fingers were pointed posteriorly otherwise one can breach peritoneum easily. After blunt dissection and entering into retroperitoneal space, anterior-most fibers of psoas were gently retracted posteriorly using peanut shaped gauze piece. Docking was done over disc space using serial dilators. While deciding docking point which is at junction of anterior 1/3rd and posterior 2/3rd of disc space, blunt tipped dilator was used instead of Jamshidi needle (Figure 3B). A distractor pin was placed close to endplate as much as possible to avoid injury to segmental vessels (Figure 3C). It also served to retain the tubular retractor in place over the disc space. After annulotomy, disc removal was done using serial shavers in orthogonal maneuver. It is very important to confirm the position of instruments in AP and lateral views while performing orthogonal maneuver especially while inserting first shaver (Figure 3D). All disc preparation instruments are inserted in orthogonal maneuver only. The annulus fibrosis at the opposite side was released completely with caution using Cobb elevator under fluoroscopic viewing. Cage size was determined by appropriate sizers, which were inserted into disc space and confirmed on IITV images. An appropriate-sized cage was filled with synthetic bone substitute and was inserted in a press-fit fashion into the disc space. Because these maneuvers are performed anterior to psoas muscle and the retraction was gentle and temporary, neuromonitoring was not needed. Upon insertion of cage either anterior fixation was done or the patient was turned to the prone position, and supplemental percutaneous posterior pedicle screws were placed. Compression was done.
RESULTS
Table 1 shows demographic characteristics of this study. Twenty-four segments of 22 consecutive patients (9 males and 13 females) were operated. Average age was 58 years. Minimum follow-up was for 1 year and maximum follow-up was for 3 years with average follow-up of 16 months. Average body mass index (BMI) was 27.7 kg/m2. Seventeen patients had degenerative listhesis, 2 had lytic listhesis, and 3 had adjacent segment disease. Single segment fusion was done in 20 patients, double level fusion done in 2 patients. Out of 24 segments, 7 segments were at L4–5, 13 segments were at L3–4 whereas 4 segments were at L2–3. Three patients had anterior whereas 19 patients had posterior percutaneous pedicular fixation. As this study was on learning curve of MIS OLIF, authors had more patients with L3–4 segment as compared to L4–5 segment. The major reason behind this: in many patients, bifurcation of great vessels is little higher than L5–S1 level. This can narrow down working corridor between psoas and vessels at L4–5 level making it difficult to perform OLIF especially during early learning curve. Sixteen patients had American Society of Anesthesiologists physical status (APA PS) classification II risk for surgery whereas 6 patients had ASA PS classification III risk. Diabetes mellitus was present in 16 patients, cardiac comorbidities was present in 7 patients, and other medical comorbidities were present in 10 patients.
1. Clinical Results
Clinical assessment was done according to modified MacNab criteria at 3-month follow-up. Twenty patients (91%) had excellent result and 2 patients (9%) had good result at 3-month follow-up. Average hospital stay of the patients was 2 days (Table 2).
2. Complications
There were some approach-related complications. Major per-operative complication was rupture of the ilio-lumbar vein in 1 patient (4.5%). However, it was effectively controlled with help of local hemocoagulants and without any post-op morbidity. Other per-operative complication included fracture of the superior endplate of inferior vertebra in 1 patient (4.5%) during cage preparation who was severely osteoporotic. Further operative steps were carried out usually and press-fit OLIF cage was inserted. There were 2 incidents (9%) of per-operative complication and both were in initial 10 cases/initial part of learning curve. Mean EBL over entire case series was 50 (range, 40–70) mL barring 1 case of ilio-lumbar vein rupture which led to blood loss of approximately 150 mL (Figure 4A). Mean operative time for single level cage insertion was 50 minutes which was in decreasing trend with experience (Figure 4B). It is noteworthy that a gastro-surgeon was kept on standby for initial 10 cases in any case of access related complications. Postoperative complications were: One patient (4.5%) had superficial infection which got resolved with oral antibiotics. Four patients (18%) had postoperative left anterior thigh paraesthesia and 3 (13.5%) were from initial 10 cases and 1 was (4.5%) was from later 12 cases. There was no power deficit and all of them recovered within 3–4 weeks of oral pregabalin 75-mg tablets. None required intravenous/oral steroids. And it did not affect overall clinical recovery.
3. Radiological Results
The following parameters were studied either on x-rays or MRI; foraminal height, disc height, reduction in vertebra slippage (listhesis), improvement in segmental lordosis, improvement in spinal canal area on axial T2-weighted MRI images. Comparison of pre- and post0operative data was done and mean of percentage improvement was calculated (Table 3). Improvement in foraminal height was 20.78% and has been shown in Figure 5. Improvement in disc height was 97.32% and has been shown in Figure 6. Improvement in listhesis reduction was 7.1° and has been shown in Figure 7. Improvement in segmental lordosis was 3.16° and has been shown in Figure 8. It was noted that if the cage position were more anterior, it would achieve more segmental lordosis whereas middle or posterior position would have lesser improvement in segmental lordosis. Improvement in spinal canal area was 46.54% overall.
4. Comparing the Data of 2 Groups
Initial learning curve and later part of learning curve showed 2 important aspects. In initial part of the learning curve, the author did not include any patient of severe lumbar canal stenosis (Schizas grade D) due to apprehension of efficacy of indirect decompression in such cases. However, all the patients with Schizas grade D patient in later part (3 patients) showed excellent clinical outcomes. Also, both the per-operative complications were seen in first group of patient only. Only data of psoas paresis showed statistically significant difference between the 2 groups. However, none of rest of the data compared which included comparison of severity of preoperative lumbar canal stenosis, per-operative complication EBL, mean operative time, hospital stay, etc. showed significant statistical difference (Table 4).
5. Case Example
1) Case 1
A 62-year-old female patient, known case of diabetes mellites type 2 and hypertension. BMI was 29 kg/m2. Chief complaints of significant neurological claudication without rest pain and walking distance of less than 100 m. Neurological assessment showed bilateral L5–S1 hypoesthesia and weak toe extensors. Straight leg raising test was negative. All conservative trials had failed. (Figure 9). Radiology showed grade 1 L4–5 degenerative lysthesis and Schizas grade D stenosis on MRI. She underwent L4–5 OLIF where titanium OLIF cage with bone graft was used. Percutaneous pedicle screw fixation at L4–5 was done in same stage. Postoperatively she had excellent clinical outcome as per MacNab criteria. One-year follow-up MRI shows significant improvement in spinal canal area and thinning of ligamentum flavum.
2) Case 2
A 52-year-old female patient with both lower limb neurological claudication and failed conservative management. Radiology showed presence of grade 1 lytic lysthesis at L4–5 level (Figure 10A). Patient was enrolled for MIS OLIF. However, noteworthy aspect was less working corridor between iliac vessels and psoas muscle (<10 mm) and psoas was also bulky (Figure 10B). After reaching desired disc space, psoas muscle’s anterior fibers could not be reflected posteriorly and satisfactory access corridor could not be created. So, OLIF was abandoned and MIS TLIF was performed (Figure 10C). This patient was in early part of the learning curve and not part of the sample of this study. This shows importance of the working corridor for safe passage of OLIF cage.
DISCUSSION
Lumbar Interbody fusion is choice of treatment in patients with degenerative lumbar spine disorders and radiographic evidence of instability. Literature has shown advantages of oblique lumbar interbody fusion as an effective technique for indirect decompression and fusion [8,13]. The OLIF technique allows an oblique lateral trajectory to access multiple lumbar levels mainly from L1–5 levels. This retrospective study demonstrates that OLIF can be safely performed during the learning curve with favorable clinicoradiological outcomes. This study was conducted by a single surgeon at single center with long-term follow-up and all the surgeries were performed during learning curve without use of neuromonitoring.
We compared our perioperative complication rates with previous studies. Our data showed favorable outcome in terms of complication rate in comparison to systemic review of OLIF complication [12]. These authors reported a mean blood loss of 110 mL, 6-day hospital stay, and overall complication rate of 11%. In comparison, the results from the current series were 50-mL blood loss, 2-day hospital stay, and 15% complication rate which all of them recovered. We also compared our data with studies showing perioperative complication rate during learning curve by Oh et al. [14]. This study had psoas paresis as most common perioperative complication (36.4%) against 15% in our study. Also, study by Oh et al. [14] demonstrated other complications like ureteric injury, cage malposition, peritoneal injury and ilieus etc., which were not present in our study. However, EBL and average hospital stay did not show any significant difference in both the studies. It has been shown that operator experience does not influence the rate of complications with OLIF [15]. We observed the same finding although low complication rates and small sample size limited our ability to detect such trends.
Psoas paresis following OLIF has been described in literature extensively as one of the major complications following OLIF [4,15]. And this stands true in initial learning curve. Difficulty while reflecting anterior-most fibers of psoas and prolonged retraction with self-retaining retractors without intermittent relaxation can be potential cause for that. We adopted technique of reflecting anterior fibers with blunt dissection starting at edge using gauze piece and using handheld long-blade retractors to hold back bulky psoas instead of self-retaining one to avoid constant pressure. In majority of the patients psoas paresis presented as sensory symptoms only and recovered within 4–6 weeks of time.
Vertebral body fracture is one of the potentials but relatively rare complication in OLIF [16]. In our study, patient had significant osteoporosis and concave upper endplate of lower vertebral body. While disc space preparation; use of shavers must be done in orthogonal maneuver and position must be confirmed in AP as well lateral fluoroscopic images. Also, larger size of cage placement can lead to fracture. We recommend taking cranial/normal foramina height as reference to decide cage size. Shavers used to prepare disc space must be 2-mm sizes smaller than the intended cage size to be placed.
Vascular injury is a potential and dreaded complication of OLIF. Preoperative analysis of working corridor is must. Also, one must be aware of the variants of the ilio-lumbar vein and segmental arteries crossing at disc space more common at L4–5 level [17]. We recommend using distractor pins under direct vision and close to endplates as much as possible.
OLIF is an effective technique to achieve satisfactory clinicoradiological outcomes. We could achieve clinically significant improvement. Also, radiological parameters showed effective indirect decompression, lysthesis reduction and segmental lordosis improvement. This was comparable to other Indian studies done by Parikh et al. as well [8]. However, to achieve satisfactory clinicoradiological outcomes; one must be prudent regarding ideal case selection. Resting pain, positive nerve root tension sign on clinical assessment and disc herniation, severe bony canal stenosis, grade 3 or more lysthesis are relative contraindications to achieve indirect decompression using OLIF. However, in case of dynamic instability even in presence of severe lumbar canal stenosis (Schizas grade C/D); OLIF has been proven technique to achieve satisfactory indirect decompression [18]. So, even in presence of severe canal stenosis, one can adopt this technique in early learning curve.
OLIF can be challenging sometimes at L4–5 levels due to 2 reasons. High iliac crest and higher bifurcation of iliac vessels. To cope up with high iliac crest, incision can be put 1–2 cm anterior to desired sight and oblique trajectory to target disc space can be taken. Placement of roller-pack below opposite iliac crest can also widen the space between entry point and crest upto some extent. To tackle second issue of bifurcation of vessels, careful assessment of the axial cut at L4–5 disc space is mandatory. If psoas is not bulky and bifurcation is just below the disc space, one can easily access the L4–5 disc space.
Overall, this study has shown that OLIF remains safe and effective procedure through 2-year follow-up despite all the cases performed during a surgeon’s learning curve. This study addressed the learning curve challenges pertaining case selection and potential perioperative complications. There were also several limitations of this study that warrant mention. This was a retrospective study of only 20 patients treated by a single surgeon. Therefore, the potential for bias, concerns regarding the generalizability of findings, and inability to detect uncommon complications is there. Also, comparative data of learning curve from other lumbar interbody fusion and decompression techniques is also required to prove OLIF as a superior technique to adopt in initial part of learning curve.
CONCLUSION
MIS OLIF is an effective technique to achieve satisfactory cllinicoraiological outcomes in degenerative lumbar spine disorders. This study shows its efficacy in learning curve and feasibility to adopt in absence of major perioperative complications. However, prudent case selection is mandatory. Larger sample size and multicentric studies are required to prove this hypothesis.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Print 
