Does approach matter? A comparative radiographic analysis of spinopelvic parameters in single-level lumbar fusion

Abstract

Background Context

Lumbar fusion is a popular and effective surgical option to provide stability and restore anatomy. Particular attention has recently been focused on sagittal alignment and radiographic spinopelvic parameters that apply to lumbar fusion as well as spinal deformity cases. Current literature has demonstrated the effectiveness of various techniques of lumbar fusion; however, comparative data of these techniques are limited.

Purpose

This study aimed to directly compare the impact of various lumbar fusion techniques (anterior lumbar interbody fusion [ALIF], lateral lumbar interbody fusion [LLIF], transforaminal lumbar interbody fusion [TLIF], and posterolateral fusion [PLF]) based on radiographic parameters.

Study Design/Setting

A single-center retrospective study examining preoperative and postoperative radiographs was carried out.

Patient Sample

A consecutive list of lumbar fusion surgeries performed by multiple spine surgeons at a single institution from 2013 to 2016 was identified.

Outcome Measures

Radiographic measurements used included segmental lordosis (SL), lumbar lordosis (LL), pelvic incidence (PI), pelvic incidence-lumbar lordosis (PI-LL) mismatch, anterior and posterior disc height (DH-A, DH-P, respectively), and foraminal height (FH).

Methods

Radiographic measurements were performed on preoperative and postoperative lateral lumbar radiographs on all single-level lumbar fusion cases. Demographic data were collected including age, gender, approach, diagnosis, surgical level, and implant lordosis. Paired sample t test, one-way analysis of variance (ANOVA), McNemar test, and independent sample t test were used to establish significant differences in the outcome measures. Multiple linear regression was performed to determine a predictive model for lordosis from implant lordosis, fusion technique, and surgical level.

Results

There were 164 patients (78 men, 86 women) with a mean age of 60.1 years and average radiographic follow-up time of 9.3 months. These included 34 ALIF, 23 LLIF, 63 TLIF, and 44 PLF surgeries. ALIF and LLIF significantly improved SL (7.9° and 4.4°), LL (5.5° and 7.7°), DH-A (8.8 mm and 5.8 mm), DH-P (3.4 mm and 2.3 mm), and FH (2.8 mm and 2.5 mm), respectively (p≤.003). TLIF significantly improved these parameters, albeit to a lesser extent: SL (1.7°), LL (2.7°), DH-A (1.1 mm), DH-P (0.8 mm), and FH (1.1 mm) (p≤.02). PLF did not significantly alter any of these parameters while significantly reducing FH (−1.3 mm, p=.01). One-way ANOVA showed no significant differences between ALIF and LLIF other than ALIF with greater ΔDH-A (3.0 mm, p=.02). Both ALIF and LLIF significantly outperformed PLF in preoperative to postoperative changes in all parameters p≤.001. Additionally, ALIF significantly outperformed TLIF in the change in SL (6.2°, p<.001), and LLIF significantly outperformed TLIF in the change in LL (5.0°, p=.02). Both outperformed TLIF in ΔDH-A (7.7 mm and 4.7 mm) and ΔDH-P (2.6 mm and 1.5 mm), respectively (p≤.02). ALIF was the only fusion technique that significantly improved the proportion of patients with a PI-LL<10° (0.410.66, p=.02). Lordotic cages had superior improvement of all parameters compared with non-lordotic cages (p<.001). Implant lordosis (m=1.1), fusion technique (m=6.8), and surgical level (m=6.9) significantly predicted postoperative SL (p<.001, R²=0.56).

Conclusions

This study demonstrated that these four lumbar fusion techniques yield divergent radiographic results. ALIF and LLIF produced greater improvements in radiographic measurements postoperatively compared with TLIF and PLF. ALIF was the most successful in improving PI-LL mismatch, an important parameter relating to sagittal alignment. Lordotic implants provided better sagittal correction and surgeons should be cognizant of the impact that these differing implants and techniques produce after surgery. Surgical technique is an important determinant of postoperative alignment and has ramifications upon sagittal alignment in lumbar fusion surgery.

Introduction

Degenerative conditions of the lumbar spine are becoming increasingly prevalent with an aging population and result in significant reductions to quality of life caused by immobility, radicular pain, and muscle spasm. Lumbar fusion through various techniques is an effective intervention for restoring the stability of these spinal segments, alleviating compression of neural elements, and reestablishing spinal anatomy.

The critical importance of sagittal alignment and pelvic parameters is well established in spinal deformity and lumbar degenerative surgery alike. High pelvic incidence-lumbar lordosis (PI-LL) mismatch (greater than 10°) is associated with adjacent segment degeneration and a tenfold risk of revision surgery [1], [2], [3], worsened postoperative residual symptoms [4], reduced quality of life [5], severe disability [6], [7], and decreased recovery rate [8]. Similarly, lower postoperative lumbar lordosis (LL) is significantly associated with increased adjacent segment degeneration [3], disability [9], and pain [10].

There are multiple studies in the literature that examine the effects various lumbar fusion techniques have on spinopelvic parameters, albeit with conflicting results. Transforaminal lumbar interbody fusion (TLIF) has been shown to improve segmental lordosis (SL) [11], [12] and LL [13], [14] in previous studies. However, a recent retrospective study and randomized controlled trial demonstrated conflicting results as TLIF did not result in superior SL improvement over posterolateral fusion (PLF) [15], [16]. Lateral lumbar interbody fusion (LLIF) was initially shown to improve only SL [17], [18], [19], [20], [21]; however, later studies revealed the technique can also improve LL [22], [23], [24], [25], [26]. Anterior lumbar interbody fusion (ALIF) outperformed TLIF in the improvement of SL and LL in multiple studies [27], [28], [29], [30], [31].

There are few studies that have directly compared the radiographic outcomes of all the available lumbar fusion techniques together, and none have included PI-LL mismatch as an outcome measure. Watkins et al. compared ALIF, LLIF, and TLIF and determined that only ALIF and LLIF significantly improved SL [32]. Sembrano et al. compared ALIF, LLIF, TLIF, and PLF and found that the former three approaches increased SL and LL, with ALIF to the greatest extent, emphasizing the importance of interbody cages [33]. A recent systematic review found that ALIF had significantly higher postoperative SL than LLIF, TLIF, and PLIF [34].

The geometry of the interbody cage itself may influence sagittal alignment, with wedge-shaped lordotic cages becoming more commonly used over non-lordotic cages as the lordotic cages can significantly increase SL and LL [35]. Lordotic cages in LLIF have been shown to provide a significant increase in SL (p=.01), whereas non-lordotic cages do not (p=.71) [36], [37]. Hyperlordotic cages and anterior longitudinal ligament resection may further impact sagittal alignment as placement of increasingly lordotic cages without the restraint of the anterior longitudinal ligament led to greater gains in SL and LL [38], [39], [40], [41]. In addition to cage geometry, cage position and insertion level are factors that may also influence the change in SL [42].

The literature regarding the various lumbar fusion techniques and interbody cages has demonstrated satisfactory fusion rates and clinical successes; however, their comparative effects on sagittal alignment and pelvic parameters, including SL, LL, and PI-LL mismatch remain unclear. The goal of this study was to compare the capacities of ALIF, LLIF, TLIF, and PLF to change SL, LL, PI-LL mismatch, disc height (DH), and foraminal height (FH), as well as to develop a predictive model for lordosis based on the surgical approach, insertion level, and cage lordosis.