Is Congenital Lumbar Spinal Canal Stenosis a Contraindication for Indirect Decompression by Lateral Lumbar Interbody Fusion (LLIF)?

Abstract

Study Design

Retrospective study.

Objectives

To assess whether the Congenital Lumbar Spinal Canal Stenosis (CLSS) affects the effectiveness of indirect decompression achieved through the Lateral Lumbar Interbody Fusion (LLIF).

Methods

Patients who were categorized into 2 groups, those with CLSS (AP canal diameter <14 mm at L4 or L5) and non-CLSS, underwent OLIF at L4-5. Clinical outcomes were assessed by using the ODI and VAS pain scores. The radiographic outcomes, including disc height (DH), foraminal height, foraminal area, AP canal diameter, canal cross-sectional area (CSA), and lateral recess grading, were measured between pre- and postoperatively at the index and adjacent levels.

Results

A total of 98 patients were included (26 in CLSS and 72 non-CLSS group). Both groups showed significantly improved clinical outcomes, including ODI and VAS scores for back and leg pain at 1-month and 1-year follow-ups. Radiographically, both groups demonstrated significant increases in canal expansions at the L4-5 index level; the non-CLSS group had slightly greater absolute increases in CSA (+52.23 mm² vs +35.76 mm²) and AP canal diameter (+2.19 mm vs +1.88 mm), but not statistically significant. At the L3-4 level, statistically significant canal expansions were observed in both groups, without significant differences. At L5-S1 level, CLSS group showed significant increase in disc height (+1.02 mm vs +0.15 mm; P = 0.016) and foraminal area compared to non-CLSS.

Conclusion

With proper patient selection, LLIF provides comparable clinical and radiographic outcomes in patients with and without CLSS. CLSS should not be considered a contraindication for LLIF.

Keywords

LLIF congenital lumbar spinal stenosis (CLSS)indirect decompression OLIF minimally invasive spine surgery

Introduction

Lateral Lumbar Interbody Fusion (LLIF), including Oblique Lateral Interbody Fusion (OLIF) and Extreme Lateral Interbody Fusion (XLIF), is a minimally invasive spinal procedure that has been proven to provide significant clinical and radiographic improvements in treating various degenerative lumbar spinal conditions.^1-3 LLIF applies the indirect decompression technique, which helps relieve neural element compression without requiring direct decompression. Additionally, LLIF achieved several radiographic outcomes, including restoration of disc height and foraminal height (FH), unbuckling of the ligamentum flavum, and expansion of the central canal area.^4-9 However, some studies addressed that the efficacy of indirect decompression is limited, particularly for patients with severe stenosis or structural impediments, locked facets, or congenital lumbar spinal stenosis (CLSS). In such instances, some authors recommended performing additional direct decompression to acquire adequate spinal decompression.^10,11

Congenital (or developmental) lumbar spinal stenosis (CLSS), illustrated by Verbiest et al, refers to a pre-existing narrowing of the bony lumbar spinal canal.¹² There is a lower threshold whereby these patients are more susceptible to neural compression with less severe pathologies, such as mild degrees of ligamentum flavum hypertrophy and disc herniation.¹³ This condition was reported to be a risk factor for indirect decompression failure, and some studies recommended that the indirect decompression technique should not be performed on patients with CLSS.^14,15 However, studies regarding the outcomes of patients with CLSS undergoing LLIF are still lacking. This study aims to evaluate the clinical and radiographic outcomes of CLSS patients who underwent the LLIF procedure and ascertain whether the CLSS serves as a contraindication for LLIF.

Materials and methods

This is a single-center retrospective study of patients who were diagnosed with degenerative lumbar spine diseases and underwent single-level LLIF without direct decompression at the L4-5 level from 2016 to 2023. The study received approval from the Ethics Committee of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand (IRB No.0899/67). Given its retrospective nature, informed consent was waived. Patients who met any of the following criteria were excluded: those who underwent additional direct decompression, multi-level surgery, spinal tumors, infectious pathologies, fractures, prior spinal surgery at the index level, or those who did not have completed medical records.

At our institution, all patients, including those with CLSS, are considered for LLIF only if they meet predefined criteria for successful indirect decompression.¹⁶ The criteria included: (1) dynamic clinical symptoms with pain relief in the supine position, (2) presence of reducible disc height as demonstrated radiographically, (3) absence of profound motor weakness, and (4) no evidence of static stenosis, such as facet cyst or fixed bony lateral recess narrowing. Preoperative CT evaluation of facet morphology or bony lateral recess was not routinely performed; instead, CT was reserved for cases where static compression was clinically suspected, and such patients were excluded from this study and treated with a direct decompression method.

Outcomes Measurements

The primary outcomes were assessed through standardized self-reported Health-related Quality of Life (HRQoL) measures, including the Oswestry Disability Index (ODI) and Visual Analog Scale (VAS) scores for back and leg pain. Data were collected at 3 time points: preoperatively, at 1 month, and 1 year postoperatively.

The secondary outcomes were evaluated radiographically. Every patient included in this study underwent preoperative plain lumbosacral spine radiographs and magnetic resonance imaging (MRI), as well as a 6-month postoperative MRI. Radiographic parameters were measured at the index level (L4-5) and adjacent levels (L3-4 and L5-S1). Parameters included disc height (DH) and foraminal height (FH), assessed on lateral radiographs. DH was defined as the distance between the upper endplate of the lower vertebra and the lower endplate of the upper vertebra, measured at the midpoint of the endplate. FH was measured as the distance between the lower edge of the upper pedicle and the superior edge of the lower pedicle (Figure 1).

Figure 1.

Radiographic Measurement: Disc Height (DH, A) and Foraminal Height (FH, B)

T2-weighted MRI was utilized to measure the foraminal area (FA), spinal canal diameter (SCD), and the cross-sectional area (CSA) of the thecal sac. FA was measured bilaterally, and the average of both sides was calculated and recorded (Figure 2). The lateral recess stenosis grading was categorized into 4 grades according to Bartynski et al¹⁷: Grade 0 (normal); Grade 1 (reduced size of the lateral recess but nerve root was visualized and not compressed); Grade 2 (reduced lateral recess size with nerve root compression); and Grade 3 (severe hypertrophy of the facet joints and ligamentum flavum with no epidural space or cerebrospinal fluid identified in the lateral recess and severe nerve root compression, or medially displaced) (Figure 3).

Figure 2.

Foraminal Area (FA, A), Spinal Canal Diameter (SCD, B), and Cross-Sectional Area of the Thecal Sac (CSA, C) on T2-Weighted MRI

Figure 3.

Classification of Lateral Recess Stenosis. (A) Grade 0: Normal. (B) Grade 1: Reduced Size of the Lateral Recess Stenosis, but the Nerve Root is Not Compressed and Visualized. (C) Grade 2: Reduced Size of the Lateral Recess Stenosis and the Nerve Root is Compressed. (D) Grade 3: Severe Hypertrophy of the Facet and Ligamentum Flavum, No Space or CSF is Identified in the Lateral Recess, and the Nerve Root is Compressed Severely or May be Seen as Medially Displaced

Preoperative axial T1-weighted MRI at the L4-5 level was used at the slice with the thickest pedicle, capturing the vertebral body, pedicle, and lamina. For the sagittal view, a midsagittal image was used that bisected the spinous processes to measure the midline AP bony spinal canal diameter. Congenital lumbar spinal canal stenosis was defined as an AP bony spinal canal diameter of less than 14 millimeters (mm) at either L4 or L5 in axial view (Figure 4).

Figure 4.

AP Spinal Canal Diameter on Axial T1-Weighted MRI. A: Non-CLSSS. And B: CLSS

The patients were categorized into CLSS and non-CLSS groups according to CLSS diagnostic criteria reported by Cheung et al.¹⁸ The clinical and radiographic outcomes were compared between the 2 groups.

Surgical Techniques

The procedure was done by spine surgeons who have experienced in minimally invasive spine surgeries (WS or WL). After induction of general anesthesia, the patient was positioned in the right lateral decubitus position. The targeted intervertebral disc levels were identified by a C-arm fluoroscope, and the skin incision was marked accordingly. An incision was made along the anterolateral abdominal area aligned with the targeted disc level. The external oblique, internal oblique, and transversus abdominis muscles were bluntly dissected layer by layer, to enter the retroperitoneal space. The target intervertebral disc was located in the oblique corridor and the self-retaining tubular retractors were subsequently docked within the corridor. Following discectomy and endplate preparation, trial cages were inserted to determine the appropriate implant size. Cage height was selected intraoperatively using sequential trial cages, aiming to optimize disc height restoration without endplate violation. The length and footprint were chosen according to individual anatomic fit, and the final lordotic angle and cage selection were made at the discretion of the operating surgeon. The interbody cage (OLIF25 Clydesdale Spinal System, Medtronic, Minneapolis, MN, USA), filled with demineralized bone matrix (DBM; GRAFTON®, Medtronic, Minneapolis, MN, USA), was then implanted using an orthogonal maneuver, and the lateral incision was closed layer by layer in standard fashion. Subsequently, the patient was repositioned prone for percutaneous pedicle screw and rod fixation.^1,2

Statistical Analyses

Patient characteristics are presented as means and standard deviations (SD) for continuous variables and as frequencies (percentages) for categorical variables. Statistical analyses were conducted using SPSS version 29.0 (IBM Co, Armonk, NY, USA). Within-group comparisons between baseline and the 1-year follow-up were analyzed using the generalized estimating equation (GEE). For comparisons between groups, we applied the independent t-test for continuous variables and the chi-square test or Fisher’s exact test for categorical variables, as appropriate. Pearson correlation analysis was performed to evaluate the relationship between radiographic changes and clinical outcomes. A P-value of less than 0.05 was considered statistically significant.

Radiographic measurements were performed by a spine fellow and repeated after a 2-week interval to evaluate intra-rater reliability. Additionally, a second independent observer assessed the same subset to determine inter-rater reliability. Intraclass correlation coefficients (ICCs) were calculated using a two-way random-effects model for absolute agreement. The ICC demonstrated good reliability for both intra-rater (ICC = 0.81) and inter-rater (ICC = 0.79) assessments.

Results

The total number of 98 patients were included in the study. The patients’ demographic data are summarized in Table 1. The mean age was 66.27 ± 8.53 and 63.74 ± 11.01 years in the CLSS group and non-CLSS group, respectively. Gender distribution was similar between the 2 groups. There were no statistically significant differences in baseline characteristics between groups. The only parameter that showed a significant intergroup difference was the anteroposterior (AP) diameter of the spinal canal at the L4 and L5 levels, which was markedly smaller in the CLSS group than in the non-CLSS group (P < 0.001). The mean cage height was 11.85 ± 0.94 mm in the CLSS group and 12.01 ± 0.92 mm in the non-CLSS group, with no significant difference. The mean cage length was 46.15 ± 4.31 mm in the CLSS group and 45.97 ± 3.52 mm in the non-CLSS group, without a significant difference. Most cages had a 6° lordotic angle, with no variation between groups. Cage subsidence occurred in 8 patients (11.5% in the CLSS group and 6.9% in the non-CLSS group), with no significant intergroup difference, and all cases were asymptomatic and did not require revision surgery.

Table 1.

Demographic Data

	CLSS (n = 26)	No CLSS (n = 72)	P-value
Age, years	66.27 ± 8.53	63.74 ± 11.01	0.291
Gender			0.336
Female	19 (73.1%)	59 (81.9%)
Male	7 (26.9%)	13 (18.1%)
BMI (kg/m2)	26.3 ± 5.58	25.86 ± 4.75	0.722
Smoker	1(3.8%)	3(4.2%)	0.944
Diabetes mellitus	5(19.2%)	15(20.8%)	0.862
Osteoporosis	4(15.4%)	12(16.68%)	0.880
Diagnosis			0.290
Spondylolisthesis	16(61.5%)	51(70.8%)
Spinal canal stenosis	9(34.6%)	13(18.1%)
Herniated nucleus pulposus	1(3.85%)	7(9.7%)
Adjacent segment disease	0	1(1.38%)
AP canal diameter
L4	14.55 ± 2.5	19.39 ± 2.83	<0.001*
L5	10.97 ± 2.26	17.15 ± 2.93	<0.001*
Cage parameters
Cage height (mm)	11.85 ± 0.94	12.01 ± 0.92	0.430
Cage length (mm)	46.15 ± 4.31	45.97 ± 3.52	0.833
Cage lordotic angle (degree)	6.46 ± 1.63	6.17 ± 0.99	0.282
Cage subsidence	3 (11.5%)	5 (6.9%)	0.434

Value present as mean ± SD or number (%).

BMI, Body Mass Index.

The VAS scores for back and leg pain at baseline, 1-month, and 12-month follow-up are presented in Table 2. Both groups demonstrated significant reductions in pain scores at 1-month and 12-month postoperatively compared to preoperative periods, with no statistically significant differences between groups. Similarly, the ODI scores improved significantly at both postoperative time points. Figure 5. Illustrates the trend of VAS back pain, VAS leg pain, and ODI scores in the CLSS and non-CLSS groups at preoperative, 1-month, and 12-month postoperative intervals. The findings indicate that both groups experienced consistent improvement in pain and functional outcomes over time.

Table 2.

Clinical Outcomes

	CLSS (n = 26)	No CLSS (n = 72)	Mean difference (95%CI)	P-value
VAS back
Preop	5.54 ± 2.52	6.43 ± 2.37	−0.89 (−1.99, 0.2)	0.109
Postop 1 month	2.35 ± 2.15	2.47 ± 2.45	−0.13 (−1.21, 0.95)	0.817
Postop 12 month	1.88 ± 1.84	2 ± 2.16	−0.12 (−1.06, 0.83)	0.809
Diff preop/1 month	−3.19 (−4.4, −1.98)^a	−3.96 (−4.79, −3.13)^a	0.77 (−0.59, 2.12)	0.268
Diff preop/12 month	−3.65 (−5.01, −2.3)^a	−4.43 (−5.19, −3.67)^a	0.78 (−0.58, 2.13)	0.261
VAS leg
Preop	6.88 ± 2.32	6.13 ± 2.51	0.76 (−0.36, 1.88)	0.18
Postop 1 month	2.08 ± 2.1	1.49 ± 2.16	0.59 (−0.38, 1.56)	0.231
Postop 12 month	1.58 ± 2.06	1.38 ± 2.09	0.2 (−0.74, 1.15)	0.673
Diff preop/1 month	−4.81 (−6.08, −3.54)^a	−4.64 (−5.32, −3.95)^a	−0.17 (−1.41, 1.07)	0.789
Diff preop/12 month	−5.31 (−6.37, −4.24)^a	−4.75 (−5.49, −4.01)^a	−0.56 (−1.8, 0.68)	0.378
ODI
Preop	48.85 ± 16.13	48.26 ± 14.02	0.6 (−6.03, 7.23)	0.858
Postop 1 month	28.62 ± 15.43	29.02 ± 18.5	−0.4 (−8.48, 7.67)	0.921
Postop 12 month	23.01 ± 13.03	22.66 ± 15.75	0.35 (−6.51, 7.2)	0.92
Diff preop/1 month	−20.23 (−28.65, −11.82)^a	−19.22 (−23.8, −14.65)^a	−1.01 (−8.52, 6.5)	0.792
Diff preop/12 month	−25.85 (−33.99, −17.71)^a	−25.6 (−30.03, −21.16)^a	−0.25 (−7.75, 7.25)	0.948

Value present as mean ± SD or mean difference (95% confidence interval).

VAS, visual analog scale; ODI, Oswestry Disability Index; Diff, difference; Preop, preoperative; Postop, postoperative; Diff Preop 1 month, difference between preoperative and 1 month postoperative periods. Diff Preop/12 month, difference between preoperative and 12 month postoperative periods.

^aStatistically significant difference.

Figure 5.

VAS (A) Back and (B) Leg Pain Scores, and (C) ODI Scores Comparison Between Preoperative, 1 Month, and 12 months Post Operation

Preoperative radiographic measurements revealed significant differences between the 2 groups at the index (L4-5) and adjacent (L3-4 and L5-S1) levels, particularly regarding SCD and CSA of the spinal canal. At the L4-5 level, the CLSS group demonstrated a significantly narrower preoperative SCD (10.92 ± 3.14 mm) compared to the non-CLSS group (12.85 ± 3.34 mm, P = 0.012), as well as significantly smaller preoperative CSA (83.8 ± 27.42 square millimeters [mm²] vs [vs] 105.64 ± 57.22 mm², P = 0.013). Postoperatively, both groups exhibited significant improvements at L4-5, although the non-CLSS group had slightly greater increases in SCD (+2.19 mm vs +1.88 mm, P = 0.439) and CSA (+52.23 mm² vs +35.76 mm², P = 0.072); these differences were not statistically significant. Improvements in DH, FH, and FA at L4-5 were comparable in both groups, showing no significant intergroup differences. Additionally, lateral recess stenosis grading at L4-5 improved similarly postoperatively in both groups, without significant differences.

At the adjacent L3-4 level, preoperative measurements showed that the CLSS group also had significantly narrower SCD (14.57 ± 2.98 mm vs 16.28 ± 3.58 mm, P = 0.033) and smaller CSA (137.88 ± 32.09 mm² vs 164.87 ± 50.76 mm², P = 0.003) compared to the non-CLSS group. Postoperative assessments demonstrated statistically significant improvements in SCD (CLSS: 15.11 ± 3.10 mm, P = 0.034; non-CLSS: 16.62 ± 3.38 mm, P = 0.024) and CSA (CLSS: 142.16 ± 33.07 mm²; non-CLSS: 168.86 ± 49.78 mm²), with significant intergroup differences persisting postoperatively (CSA P = 0.013; SCD P = 0.049). However, the magnitude of change (Δ) in these parameters was comparable between groups (ΔSCD: 0.54 mm vs 0.35 mm, P = 0.512; ΔCSA: 4.28 mm² vs 3.99 mm², P = 0.931). Although the DH significantly improved postoperatively only in the non-CLSS group (P = 0.001), no significant intergroup difference was observed. Similarly, FH and FA changes at L3-4 were not significantly different between groups. Lateral recess stenosis grading at L3-4 revealed significant preoperative differences, with a higher prevalence of mild-to-moderate stenosis in the CLSS group (right side P = 0.004, left side P = 0.007); however, postoperative lateral recess stenosis grading remained unchanged within each group.

At the adjacent L5-S1 level, preoperative CSA was significantly smaller in the CLSS group (120.72 ± 34.99 mm² vs 149.48 ± 65.5 mm², P = 0.007). Postoperatively, both groups showed significant CSA increases (CLSS: 127.4 ± 37.97 mm², P = 0.017; non-CLSS: 153.87 ± 70.88 mm², P = 0.020), but the magnitude of CSA change between groups was not significantly different (6.68 mm² vs 4.39 mm², P = 0.55). Notably, the CLSS group exhibited a significantly greater increase in DH postoperatively compared to the non-CLSS group (ΔDH: 1.02 mm vs 0.15 mm, P = 0.016). Additionally, FA significantly improved postoperatively in the CLSS group (93.1 ± 28.56 mm² to 100.31 ± 32.64 mm², P = 0.019), whereas no significant FA improvement was observed in the non-CLSS group. No significant pre- or postoperative differences in SCD, FH, or lateral recess stenosis grading at L5-S1 were observed between the groups. Detailed radiographic outcomes for each level are presented in Table 3.

Table 3.

Radiographic Outcomes

L4-5 level	CLSS (n = 26)	No CLSS (n = 72)	Mean difference (95%CI)	P-value
SCD (mm)
Preop	10.92 ± 3.14	12.85 ± 3.34	−1.93 (−3.43, −0.44)	0.012^a
Postop	12.8 ± 2.98	15.04 ± 3.09	−2.24 (−3.63, −0.85)	0.002^a
Diff preop/postop	1.88 (1.2, 2.57)	2.19 (1.78, 2.59)	−0.31 (−1.09, 0.47)	0.439
P-value	<0.001^a	<0.001^a
CSA (mm ² )
Preop	83.8 ± 27.42	105.64 ± 57.22	−21.84 (−38.98, −4.7)	0.013^a
Postop	119.56 ± 39.85	157.87 ± 64.44	−38.31 (−60.03, −16.59)	0.001^a
Diff preop/postop	35.76 (24.76, 46.76)	52.23 (42.09, 62.36)	−16.47 (−34.46, 1.52)	0.072
%Diff preop/postop	47.21 (31.02, 63.4)	66.15 (52.74, 79.56)	−18.95 (−43.09, 5.2)	0.123
P-value	<0.001^a	<0.001^a
DH (mm)
Preop	8.74 ± 2.86	8.96 ± 2.52	−0.22 (−1.41, 0.97)	0.711
Postop	13.51 ± 1.96	13.05 ± 1.96	0.46 (−0.43, 1.35)	0.311
Diff preop/postop	4.77 (3.97, 5.57)	4.09 (3.5, 4.68)	0.68 (−0.4, 1.76)	0.216
P-value	<0.001^a	<0.001^a
FH (mm)
Preop	16.96 ± 3.62	16.49 ± 3.64	0.47 (−1.18, 2.12)	0.572
Postop	19.34 ± 3.34	19.53 ± 2.98	−0.19 (−1.59, 1.2)	0.784
Diff preop/postop	2.38 (1.55, 3.2)	3.04 (2.4, 3.68)	−0.67 (−1.83, 0.5)	0.26
P-value	<0.001^a	<0.001^a
FA (mm ² )
Preop	80.56 ± 31.56	83.02 ± 34.92	−2.45 (−17.93, 13.03)	0.754
Postop	112.66 ± 37.36	117.03 ± 38.62	−4.37 (−21.76, 13.02)	0.619
Diff preop/postop	32.1 (21.99, 42.21)	34.02 (26.37, 41.66)	−1.92 (−15.89, 12.05)	0.786
%Diff preop/postop	47.07 (29.29, 64.84)	59.54 (41.1, 77.98)	−12.47 (−44.78, 19.84)	0.445
P-value	<0.001^a	<0.001^a
Lateral recess rt
Preop				0.052
0	2 (7.7%)	4 (5.6%)
1	1 (3.8%)	22 (30.6%)
2	12 (46.2%)	26 (36.1%)
3	11 (42.3%)	20 (27.8%)
Postop				0.166
0	6 (23.1%)	15 (20.8%)
1	8 (30.8%)	38 (52.8%)
2	8 (30.8%)	15 (20.8%)
3	4 (15.4%)	4 (5.6%)
Lateral recess Lt
Preop				0.121
0	2 (7.7%)	6 (8.3%)
1	2 (7.7%)	21 (29.2%)
2	11 (42.3%)	27 (37.5%)
3	11 (42.3%)	18 (25%)
Postop				0.164
0	6 (23.1%)	18 (25%)
1	8 (30.8%)	35 (48.6%)
2	10 (38.5%)	18 (25%)
3	2 (7.7%)	1 (1.4%)
L3-4 level
SCD (mm)
Preop	14.57 ± 2.98	16.28 ± 3.58	−1.7 (−3.26, −0.14)	0.033^a
Postop	15.11 ± 3.1	16.62 ± 3.38	−1.51 (−3.02, −0.01)	0.049^a
Diff preop/postop	0.54 (0.04, 1.03)	0.35 (0.05, 0.65)	0.19 (−0.38, 0.76)	0.512
P-value	0.034^a	0.024^a
CSA (mm ² )
Preop	137.88 ± 32.09	164.87 ± 50.76	−26.99 (−44.31, −9.68)	0.003^a
Postop	142.16 ± 33.07	168.86 ± 49.78	−26.71 (−47.6, −5.81)	0.013^a
Diff preop/postop	4.28 (−0.18, 8.73)	3.99 (−0.9, 8.88)	0.28 (−6.22, 6.79)	0.931
%Diff preop/postop	3.48 (0.05, 6.9)	3.49 (0.25, 6.73)	−0.01 (−5.75, 5.72)	0.996
P-value	0.059	0.108
DH (mm)
Preop	11.15 ± 1.61	10.78 ± 2.12	0.37 (−0.54, 1.28)	0.418
Postop	11.36 ± 1.62	11.15 ± 1.9	0.21 (−0.63, 1.04)	0.623
Diff preop/postop	0.2 (−0.28, 0.69)	0.37 (0.15, 0.59)	−0.17 (−0.62, 0.29)	0.474
P-value	0.392	0.001^a
FH (mm)
Preop	18.85 ± 2.6	18.47 ± 3.23	0.38 (−1.02, 1.78)	0.59
Postop	18.49 ± 2.52	18.48 ± 3.26	0.01 (−1.39, 1.41)	0.99
Diff preop/postop	−0.36 (−1.11, 0.38)	0.01 (−0.57, 0.59)	−0.37 (−1.42, 0.68)	0.486
P-value	0.328	0.971
FA (mm ² )
Preop	95.91 ± 26.09	102.23 ± 32.38	−6.32 (−20.33, 7.7)	0.373
Postop	101.45 ± 30.55	110.89 ± 37.12	−9.44 (−25.58, 6.69)	0.248
Diff preop/postop	5.53 (−1.38, 12.45)	8.66 (3.49, 13.83)	−3.12 (−12.6, 6.35)	0.514
%Diff preop/postop	6.18 (−0.83, 13.2)	10.88 (5.32, 16.44)	−4.69 (−14.77, 5.38)	0.357
P-value	0.112	0.001^a
Lateral recess rt
Preop				0.004^a
0	9 (34.6%)	51 (70.8%)
1	14 (53.8%)	19 (26.4%)
2	3 (11.5%)	2 (2.8%)
3	0	0
Postop				0.004^a
0	9 (34.6%)	51 (70.8%)
1	14 (53.8%)	19 (26.4%)
2	3 (11.5%)	2 (2.8%)
3	0	0
Lateral recess Lt
Preop				0.007^a
0	10 (38.5%)	49 (68.1%)
1	13 (50%)	21 (29.2%)
2	3 (11.5%)	1 (1.4%)
3	0	0
Postop				0.007^a
0	10 (38.5%)	49 (68.1%)
1	13 (50%)	21 (29.2%)
2	3 (11.5%)	1 (1.4%)
3	0	0
L-S1 level
SCD (mm)
Preop	11.98 ± 2.01	13.1 ± 3.4	−1.12 (−2.53, 0.28)	0.116
Postop	12.27 ± 2.22	13.32 ± 3.37	−1.04 (−2.46, 0.37)	0.146
Diff preop/postop	0.3 (0.02, 0.57)	0.22 (−0.03, 0.47)	0.08 (−0.37, 0.53)	0.724
P-value	0.034^a	0.088
CSA (mm ² )
Preop	120.72 ± 34.99	149.48 ± 65.5	−28.76 (−49.31, −8.22)	0.007^a
Postop	127.4 ± 37.97	153.87 ± 70.88	−26.48 (−48.74, −4.21)	0.02^a
Diff preop/postop	6.68 (1.29, 12.07)	4.39 (−1.14, 9.92)	2.29 (−5.31, 9.88)	0.55
%Diff preop/postop	5.84 (1.29, 10.39)	2.61 (−1.36, 6.58)	3.23 (−3.86, 10.32)	0.369
P-value	0.017^a	0.118
DH (mm)
Preop	9.52 ± 2.64	10.35 ± 2.53	−0.83 (−2, 0.33)	0.158
Postop	10.54 ± 2.84	10.5 ± 2.43	0.04 (−1.12, 1.19)	0.952
Diff preop/postop	1.02 (0.47, 1.56)	0.15 (−0.23, 0.53)	0.87 (0.17, 1.57)	0.016^a
P-value	0.001^a	0.439
FH (mm)
Preop	15.06 ± 3.41	16.13 ± 2.52	−1.07 (−2.33, 0.19)	0.096
Postop	15.27 ± 3.54	15.79 ± 2.35	−0.52 (−1.75, 0.71)	0.405
Diff preop/postop	0.21 (−0.31, 0.74)	−0.34 (−0.76, 0.08)	0.55 (−0.21, 1.31)	0.155
P-value	0.414	0.113
FA (mm ² )
Preop	93.1 ± 28.56	107.55 ± 36.55	−14.45 (−30.18, 1.29)	0.071
Postop	100.31 ± 32.64	107.6 ± 34.99	−7.29 (−22.91, 8.33)	0.357
Diff preop/postop	7.21 (1.31, 13.11)	0.05 (−5.53, 5.63)	7.16 (−2.72, 17.04)	0.153
%Diff preop/postop	8 (1.53, 14.48)	3.64 (−3.03, 10.3)	4.37 (−7.31, 16.05)	0.46
P-value	0.019^a	0.986
Lateral recess rt
Preop				0.170
0	20 (76.9%)	62 (86.1%)
1	6 (23.1%)	6 (8.3%)
2	0	3 (4.2%)
3	0	1 (1.4%)
Postop				0.170
0	20 (76.9%)	62 (86.1%)
1	6 (23.1%)	6 (8.3%)
2	0	3 (4.2%)
3	0	1 (1.4%)
Lateral recess Lt
Preop				0.125
0	20 (76.9%)	61 (84.7%)
1	6 (23.1%)	7 (9.7%)
2	0	4 (5.5%)
3	0	0
Postop				0.208
0	20 (76.9%)	61 (84.7%)
1	6(23.1%)	8 (11.1%)
2	0	3 (4.2%)
3	0	0

Value present as mean ± SD, mean difference (95% confidence interval).

SCD, Spinal canal diameter; CSA, Cross-sectional area (of the thecal sac); DH, Disc height; FH, Foraminal height; FA, Foraminal Area.

Preop, preoperative; Postop, postoperative; Diff preop/postop, difference between preoperative and 6-month postoperative periods; %Diff preop/postop, percent difference between preoperative and 6-month postoperative periods.

^aStatistically significant difference.

Correlation analysis demonstrated no significant associations between changes in radiographic parameters (ΔSCD and ΔCSA) and improvements in ODI or VAS back scores in either group. In the non-CLSS group, ΔSCD showed a significant negative correlation with VAS leg pain improvement (r = −0.236, P = 0.046). When analyzing all patients combined, ΔSCD also demonstrated a significant negative correlation with VAS leg pain improvement (r = −0.211, P = 0.037). No other significant correlations were observed. Details of these results are demonstrated in Table 4.

Table 4.

Correlation Analysis of Radiographic Parameters With Clinical Outcomes in CLSS and Non-CLSS Groups

	ODI		VAS back		VAS leg
	r	P	r	P	r	P
CLSS
Diff SCD	−0.154	0.452	−0.311	0.122	−0.159	0.438
Diff CSA	−0.078	0.704	−0.340	0.089	−0.248	0.221
Non-CLSS
Diff SCD	−0.105	0.381	0.212	0.074	−0.236	0.046^a
Diff CSA	0.166	0.163	0.130	0.275	0.108	0.365
Total
Diff SCD	−0.117	0.250	0.065	0.524	−0.211	0.037^a
Diff CSA	0.116	0.256	0.022	0.833	0.066	0.517

VAS, visual analog scale; ODI, Oswestry Disability Index.

Diff SCD, difference between SCD (mm) preoperative and 6-month postoperative periods.

Diff CSA, difference between CSA (mm²) preoperative and 6-month postoperative periods.

^aStatisticaly significant difference.

Discussion

Lateral lumbar interbody fusion (LLIF) achieves neural decompression indirectly through the ligamentotaxis effect.^2,3 By inserting a large interbody cage at the affected disc space, LLIF restores disc height, expands the neural foramina, tensions the posterior longitudinal ligament, reduces bulging of the annulus, and unbuckles the ligamentum flavum.¹⁰ Consequently, this indirect decompression enlarges the central canal and foraminal areas without necessitating subsequent direct decompression.¹⁸

Congenital (or developmental) lumbar spinal stenosis (CLSS) is defined by developmental narrowing of the spinal canal from early adulthood, independent of significant degenerative changes.¹² Typically manifesting as uniform and multilevel narrowing, CLSS patients exhibit consistently smaller spinal canal dimensions, with cross-sectional areas under 200 mm², compared to over 250 mm² in normal individuals. Mid-sagittal canal diameters are also typically 2-3 mm shorter per lumbar level.¹⁹ Shortened pedicles, averaging approximately 6 mm in CLSS compared to roughly 9 mm in normal population, produce a flattened or trefoil-shaped canal, significantly limiting the anterior-posterior (AP) canal diameter.^19,20 These anatomical characteristics are primarily osseous, distinguishing congenital stenosis from stenosis caused by degenerative changes that usually has abnormal soft tissue involvement such as ligamentum flavum hypertrophy.²¹

These anatomical constraints can restrict the ligamentous stretching achievable by ligamentotaxis, potentially increasing the risk of inadequate indirect decompression in CLSS patients undergoing LLIF.¹⁰

Oliveira et al¹⁰ demonstrated that XLIF achieved significant increases in DH (41.9%), FH (13.5%), FA (24.7%), and SCD (33.1%). However, they noted that indirect decompression may be less effective in patients with congenital stenosis or locked facets, as these conditions are limited by the limited space of the natural thecal sac and foraminal dimensions. The study highlighted that 9.5% of patients required additional direct decompression due to persistent symptoms and suggested that preoperative anatomical factors, such as congenital narrowing and bony overgrowth, may reduce the effectiveness of LLIF in achieving sufficient indirect decompression. Similarly, Li et al²² also emphasized that indirect decompression may be limited in congenital spinal canal stenosis due to the restricted ligamentotaxis effect within a rigid bony canal. In contrast to those studies, our clinical results showed significant clinical improvement following single-level L4-5 LLIF in both groups. We also found that OLIF can achieve excellent clinical results in patients with congenital stenosis, comparable to those without congenital narrowing.

Radiographically, the preoperative measurements confirmed that the CLSS group had significantly narrower spinal canals compared to non-CLSS patients at the L4-5 index level, with a mean mid-sagittal diameter of approximately 10.9 mm vs 12.9 mm, and mean CSA of roughly 84 mm² vs 106 mm², respectively. Postoperatively, both groups showed meaningful canal enlargement at L4-5, while CSA increased by approximately 36 mm² (47%) in CLSS and 52 mm² (66%) in non-CLSS patients. Although the absolute increases were slightly greater in non-CLSS patients, this difference was not statistically significant, indicating that indirect decompression effectively expanded the canal even in patients with developmental spinal canal narrowing.

With respect to cage parameters, the mean cage height and length did not differ significantly between CLSS and non-CLSS patients, and most implants had a 6° lordotic angle. Similarly, the rate of cage subsidence was low (11.5% in CLSS and 6.9% in non-CLSS) and did not adversely affect clinical outcomes. These findings suggest that implant selection and construct stability were consistent across groups, minimizing confounding effects on radiographic or clinical outcomes.

We found that changes in CSA and SCD in both CLSS and non-CLSS groups demonstrated negative but nonsignificant correlations with the clinical outcomes, including ODI and VAS scores. However, in the total cohort analysis, ΔCSA was significantly correlated with a reduction in VAS leg pain at 12 months. This finding suggests that canal expansion, particularly CSA enlargement, is a key driver of radicular pain relief after LLIF, irrespective of underlying canal morphology. The lack of significance within each subgroup might be due to sample size limitations rather than the absence of effect.

None of the patients in this study needed to undergo additional direct decompression following the OLIF. This is because we strictly adhered to the patient selection criteria for identifying patients who will most benefit from LLIF and have less risk of indirect decompression failure, which is already established in the literature.^16,23 These criteria included dynamic clinical symptoms, no profound motor weakness, radiographically reducible disc height, and no static stenosis such as facet cysts or bony lateral recess.

Regarding the adjacent segments, the results revealed distinct behaviors at L3-4 and L5-S1 following the OLIF at L4-5. At the L3-4 level, neither group demonstrated significant adverse radiographic changes in terms of DH or CSA, indicating that OLIF did not negatively impact this segment. Conversely, the L5-S1 level exhibited notable radiographic improvements, particularly in the CLSS group, which had significantly greater DH increment (+1.02 mm) compared to minimal changes (+0.15 mm) in non-CLSS patients.

This improvement may reflect compensatory segmental adjustments described by reciprocal alignment theory, wherein spinal realignment at 1 level induces compensatory changes at adjacent segments to maintain sagittal balance.²⁴ Moreover, prior studies support the importance of evaluating adjacent-level behavior: Kim et al²⁵ demonstrated that OLIF can induce measurable changes in foraminal dimensions and canal area at adjacent levels. Additionally, Verst et al²⁶ reported that adjacent segmental angles adapt dynamically within 6 months after single-level LLIF. Our findings are consistent with these observations, suggesting that OLIF at L4-5 not only improves the index level but also promotes favorable remodeling at the lower adjacent level, particularly in CLSS patients.

This study has several limitations. First, as a retrospective single-center study, inherent selection biases and the lack of randomization may have influenced the results. While we strictly adhered to predefined selection criteria, patient variability and surgeon discretion could have played a role in determining the surgical approach, potentially affecting the generalizability of our findings. Second, our follow-up duration was limited to 12 months, which may not fully capture longer-term effects. Prior longitudinal MRI studies demonstrate continued enlargement of the thecal sac area and progressive ligamentum flavum thinning beyond the early postoperative period, extending to 10-14 months follow-up following LLIF.^23,27 Future research with long-term follow-up is necessary to assess whether indirect decompression through OLIF maintains its benefits for patients with CLSS. Third, the lack of standardization in evaluating CLSS severity remains a challenge. Although we used AP canal diameter <14 mm at L4 or L5 as the diagnostic threshold, factors such as facet hypertrophy, ligamentum flavum thickening, and lateral recess morphology could also have influenced the degree of neural compression and outcomes following LLIF. In our practice, preoperative CT scans were not routinely obtained to grade facet arthropathy or quantify bony lateral recess stenosis unless static compression was suspected. Such patients were selected for direct decompression and excluded from this analysis. This selection pathway mirrors published LLIF frameworks that prioritize dynamic symptoms with reducible disc height and exclude static stenosis, such as facet cyst and bony lateral recess stenosis.^16,23 Prior research shows that indirect decompression can increase the spinal canal and foraminal area and also decrease ligamentum flavum thickness over time,^4,9,23 but that congenital canal stenosis and bony lateral recess stenosis may attenuate this effect and increase the likelihood of requiring direct decompression.^10,15,22 In our series, lateral recess stenosis was graded on MRI, and postoperative improvement at the index level was comparable between groups, supporting the adequacy of indirect decompression under these selection criteria. Additionally, our relatively small sample size, especially within the CLSS group, restricted detailed subgroup analyses that could identify specific predictors of indirect decompression failure. Larger, multicenter studies with more extensive subgroup analyses based on anatomical severity are warranted to develop appropriate patient selection guidance. Furthermore, although we reported cage parameters and found no significant differences between groups, implant selection may still introduce subtle biomechanical effects that were not captured in this study. The correlation analysis revealed that CSA enlargement was significantly associated with leg pain improvement at 12 months in the overall cohort, but subgroup analyses (CLSS vs non-CLSS) did not reach significance, likely due to limited power. These aspects underscore the need for larger prospective studies to validate the role of cage geometry and canal remodeling in driving clinical outcomes.

Conclusions

OLIF yielded promising and comparable clinical and radiographic outcomes in both CLSS and non-CLSS patients. While radiographic improvements were more pronounced in the non-CLSS group, clinical benefits were similar, suggesting that, with appropriate patient selection, CLSS should not be considered a contraindication for LLIF.

Footnotes

ORCID iDs

Weerasak Singhatanadgige

Thada Nashinoros

Teerachat Tanasansomboon

Vit Kotheeranurak

Worawat Limthongkul

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Singhatanadgige

Chatareeyagul

Tanasansomboon

, et al.

Should a narrow corridor be a contraindication for performing oblique lateral interbody fusion procedure or not?

Glob Spine J. 2025;15(2):891-897.

Limthongkul

Tanasansomboon

Yingsakmongkol

Tanaviriyachai

Radcliff

Singhatanadgige

. Indirect decompression effect to central canal and ligamentum flavum after extreme lateral lumbar interbody fusion and oblique lumbar interbody fusion. Spine. 2020;45(17):E1077-E1084.

Castellvi

Nienke

Marulanda

Murtagh

Santoni

. Indirect decompression of lumbar stenosis with transpsoas interbody cages and percutaneous posterior instrumentation. Clin Orthop Relat Res. 2014;472(6):1784-1791.

Fujibayashi

Hynes

Otsuki

Kimura

Takemoto

Matsuda

. Effect of indirect neural decompression through oblique lateral interbody fusion for degenerative lumbar disease. Spine. 2015;40(3):E175-E182.

Ahmadian

Bach

Bolinger

, et al. Stand-alone minimally invasive lateral lumbar interbody fusion: multicenter clinical outcomes. J Clin Neurosci. 2015;22(4):740-746.

Alimi

Hofstetter

Tsiouris

Elowitz

Härtl

. Extreme lateral interbody fusion for unilateral symptomatic vertical foraminal stenosis. Eur Spine J. 2015;24(Suppl 3):346-352.

Campbell

Nunley

Cavanaugh

, et al. Short-term outcomes of lateral lumbar interbody fusion without decompression for the treatment of symptomatic degenerative spondylolisthesis at L4-5. Neurosurg Focus. 2018;44(1):E6.

Formica

Quarto

Zanirato

, et al. Lateral lumbar interbody fusion: what is the evidence of indirect neural decompression? A systematic review of the literature. HSS J. 2020;16(2):143-154.

Sato

Ohtori

Orita

, et al. Radiographic evaluation of indirect decompression of mini-open anterior retroperitoneal lumbar interbody fusion: oblique lateral interbody fusion for degenerated lumbar spondylolisthesis. Eur Spine J. 2017;26(3):671-678.

10.

Oliveira

Marchi

Coutinho

Pimenta

. A radiographic assessment of the ability of the extreme lateral interbody fusion procedure to indirectly decompress the neural elements. Spine. 2010;35(26 Suppl):S331-S337.

11.

Gabel

Hoshide

Taylor

. An algorithm to predict success of indirect decompression using the extreme lateral lumbar interbody fusion procedure. Cureus. 2015;7(9):e317.

12.

Verbiest

. A radicular syndrome from developmental narrowing of the lumbar vertebral canal. J Bone Joint Surg Br. 1954;36(2):230-237.

13.

Lai

MKL

Cheung

PWH

Samartzis

Cheung

JPY

. Prevalence and definition of multilevel lumbar developmental spinal stenosis. Glob Spine J. 2022;12(6):1084-1090.

14.

Zou

Cao

, et al. Lumbar spinal stenosis: an update on the epidemiology, diagnosis and treatment. AME Med J. 2017;2:63.

15.

Malham

Parker

Goss

Blecher

. Clinical results and limitations of indirect decompression in spinal stenosis with laterally implanted interbody cages: results from a prospective cohort study. Eur Spine J. 2015;24(Suppl 3):339-345.

16.

Yingsakmongkol

Jitpakdee

Kerr

Limthongkul

Kotheeranurak

Singhatanadgige

. Successful criteria for indirect decompression with lateral lumbar interbody fusion. Neurospine. 2022;19(3):805-815.

17.

Bartynski

Lin

. Lumbar root compression in the lateral recess: MR imaging, conventional myelography, and CT myelography comparison with surgical confirmation. AJNR Am J Neuroradiol. 2003;24(3):348-360.

18.

Cheung

JPY

Samartzis

Shigematsu

Cheung

KMC

. Defining clinically relevant values for developmental spinal stenosis: a large-scale magnetic resonance imaging study. Spine. 2014;39(13):1067-1076.

19.

Singh

Samartzis

Vaccaro

, et al. Congenital lumbar spinal stenosis: a prospective, control-matched, cohort radiographic analysis. Spine J. 2005;5(6):615-622.

20.

Kitab

Alsulaiman

Benzel

. Anatomic radiological variations in developmental lumbar spinal stenosis: a prospective, control-matched comparative analysis. Spine J. 2014;14(5):808-815.

21.

Akar

Somay

. Comparative morphometric analysis of congenital and acquired lumbar spinal stenosis. J Clin Neurosci. 2019;68:256-261.

22.

Zhou

Jiang

. Development and application of oblique lumbar interbody fusion. Orthop Surg. 2020;12(2):355-365.

23.

Limthongkul

Thanapura

Jitpakdee

, et al. Is direct decompression necessary for lateral lumbar interbody fusion (LLIF)? A randomized controlled trial comparing direct and indirect decompression with LLIF in selected patients. Neurospine. 2024;21(1):342-351.

24.

Tung

K-K

Hsu

F-W

H-C

, et al. The morphological changes in adjacent segments amongst patients receiving anterior and oblique lumbar interbody fusion: a retrospective study. J Clin Med. 2021;10(23):5533. doi:10.3390/jcm10235533

25.

Kim

Kumar

Drolet

, et al.

What is the early fate of adjacent segmental lordosis compensation at L3-4 and L5-S1 following a lateral versus transforaminal lumbar interbody fusion at L4-5?

Eur Spine J. 2024;33(9):3503-3508.

26.

Verst

Drolet

Shen

, et al.

What is the fate of the adjacent segmental angles 6 months after single-level L3-4 or L4-5 lateral lumbar interbody fusion?

Spine J. 2023;23(7):982-989.

27.

Lee

Cho

, et al. The change of spinal canal according to oblique lumbar interbody fusion in degenerative spondylolisthesis: a prospective observational study. Neurospine. 2022;19(3):492-500.