Impact of Navigation Modality on Radiographic Parameters Following Minimally Invasive Transforaminal Lumbar Interbody Fusion

Abstract

Study Design

Single-center, retrospective cohort review.

Objective

This study aims to compare the impact of robotic and imaging-only navigation modalities on radiographic parameters following minimally invasive transforaminal lumbar interbody fusion (MI-TLIF).

Methods

Patients with degenerative spondylolisthesis who underwent one-level MI-TLIF using navigation or robotic modalities were included and divided into two groups: robotic-assisted (Robo) or imaging-only navigation (Nav). Radiographic parameters were assessed preoperatively and 6 months postoperatively. Outcome measures were screw and cage characteristics, postoperative disc height (DH) improvements, lumbar lordosis (LL), and slip reduction rates. Statistical analyses included paired and independent samples t-tests and multivariable linear regression analysis.

Results

The Robo group often used larger screw diameters, deeper cages, and more expandable cages, while the Nav group used wider cages. The Robo group demonstrated significantly greater DH improvement compared to the Nav group (average improvement: 71.6% vs 22.5%, P = 0.002). However, the Nav group showed significantly greater LL restoration than the Robo group (P = 0.006), and preoperative LL was independently associated with postoperative LL (β = 0.64, P < 0.001). No clear differences in slip reduction or cage subsidence were observed.

Conclusions

In MI-TLIF, robotic guidance was associated with greater DH restoration and the use of larger screws, whereas LL restoration appeared to depend mainly on preoperative sagittal alignment and was greater with image-guided navigation. These findings suggest that each modality can be leveraged to optimize different radiographic goals and underscore the need for studies integrating implant strategies with clinical outcomes.

Keywords

minimally invasive transforaminal lumbar interbody fusion navigation spondylolisthesis robotics disc height lumbar lordosis slip reduction

Introduction

Technological advancements in spine surgery have transformed the field of minimally invasive transforaminal lumbar interbody fusion (MI-TLIF) in recent years. Among these advancements, robotic assistance and advanced navigation are two such technologies that enable high levels of precision in spine surgery, even under conditions of limited direct visualization such as in the minimally invasive technique. These robotic navigation systems enhance the surgeon’s ability of accurate hardware placement. This accuracy helps to maximize the size of the screw based on a patient’s anatomy, leading to an improved biomechanical stability of the construct. Biomechanical stability plays a crucial role in the reduction maneuver particularly when patients require restoration in disc height (DH) and segmental lordosis due to degenerative lumbar spine disease such as spondylolisthesis.^1-5 In contrast, navigation-only techniques remain widely used because of their intraoperative flexibility, allowing real-time adjustment of instrumentation. This flexibility can be advantageous during key steps such as cage positioning, rod contouring, and sequential compression or distraction.

Postoperative restoration of DH is of great interest for surgeons to achieve better alignment and indirect neuroforaminal decompression of nerves. Furthermore, an increased DH is associated with the reduced risk of adjacent segment disease by minimizing stress on adjacent levels.^6-9 Achieving optimal sagittal alignment should also enhance long-term outcomes.^10,11 The precise and larger screw placement achievable with robotic assistance may provide a stronger construct for DH restoration, whereas alignment correction may depend more on the surgeon’s intraoperative strategy and three-dimensional handling, which can be facilitated by navigation-only approaches. However, despite these insights, limited data exist directly comparing the impact of robotic vs navigation-only modalities on postoperative radiographic parameters, including DH restoration, spinal alignment, and spondylolisthesis slip reduction.^1,4,12,13

This study aims to investigate the impact of robot-assisted vs navigation-only modalities on implant details and postoperative radiographic parameters following MI-TLIF for degenerative spondylolisthesis. We hypothesized that the use of robot-assisted modalities would allow surgeons the precision and confidence to use larger screws and interbody cages, thus stabilizing the fusion and leading to the restoration of DH, segmental lordosis, and other spinal alignment parameters. Clarifying these associations will guide surgeons in selecting navigation modalities when conducting MI-TLIFs based on operative goals, preoperative alignment parameters, and patient anatomy.

Methods

Study Design and Patient Population

This study was a single-center, retrospective cohort review, including patients from a multi-surgeon database. Patients who underwent a 1-level MI-TLIF between June 2017 and September 2023 using either robotics or navigation modalities were included. Inclusion criteria were: (1) patients with a primary diagnosis of degenerative spondylolisthesis, (2) patients with a 6-month clinical follow-up, (3) patients over 18 years of age, and (4) patients who underwent MI-TLIFs. We excluded patients with (1) longitudinal midline incision, (2) additional concurrent adjacent level procedures, (3) absence of an interbody cage, and (4) a fusion construct using more or less than four screws. This study included cases from 6 surgeons, with 4 of the 6 performing MI-TLIF using either robotics or navigation modalities, and 2 only using robotic modalities. Of the surgeons who used both modalities, they preferentially selected robotic methods (used in 56.6-91.4% of their cases). Data collection were managed using REDCap (Research Electronic Data Capture) hosted at Weill Cornell Medicine Clinical and Translational Science Center supported by the National Center for Advancing Translational Science of the National Institute of Health under award number: UL1 TR002384.^14,15

Surgical Technique

MI-TLIF were performed as previously described in the literature.^16-18 Patients underwent either 3D intraoperative navigational imaging alone or combined 3D imaging with robotic assistance. Pedicle screws were placed through minimal parasagittal incisions, either manually or with robotic guidance. Facetectomy was performed under microscopic visualization using a high-speed burr and Kerrison rongeurs. After complete discectomy, the disc space was prepared with autograft or allograft bone fragments placed anteriorly, followed by insertion and expansion of an interbody cage under direct visualization.

Data Collection

Patient demographic data was obtained from electronic medical records and included age, height, weight, body mass index (BMI), Charlson Comorbidity Index (CCI), gender, race, American Society of Anesthesiologists Physical Status Classification (ASA), navigation modality, operative time, and estimated blood loss. We additionally collected implant details used such as screw length and diameter and cage details (height, width, depth) (Figure 1). These data were collected from specific implant logs for each case, provided by the vendor. For expandable cages, the mean value between the nominal and maximum expanded heights provided by the vendor was used, whereas for non-expandable cages, the nominal height was recorded.

Figure 1.

This graphic of interbody cages used in MI-TLIFs show the dimensions analyzed (depth, width, and height). For expandable cages, the cage height median value was used for data analysis. Globus Medical provided explicit permission for use of images from their website⁴⁴

Radiographic parameters were measured by a board-certified surgeon or clinical fellows specializing in spine surgery based on standing lateral radiographs. Postoperative measurements were taken from the 6-month follow-up, including DH at the surgical segment and global spinopelvic alignment parameters. DH was calculated as the mean of anterior and posterior values (Figure 2). Global spinopelvic alignment parameters included pelvic tilt (PT), sacral slope (SS), pelvic incidence (PI), global lumbar lordosis (LL), proximal LL (pLL, L1-L4) and distal LL (dLL, L4-S1). PT is defined as the angle between the vertical axis and a line through the midpoint of the sacral end plate, SS as the angle between the sacral plate and the horizontal, PI as the angle between a line perpendicular to the sacral plate and a line extending from the femur, and LL as the angle between the superior endplates of L1 and S1. Vertebral slip was quantified as the distance between the posterior walls of the upper and lower vertebral bodies at the index level (Figure 2). Slip reduction rate was quantified as the percent reduction achieved after MI-TLIF.

Figure 2.

Lateral radiographs demonstrating anterior and posterior disc height measurement, as well as slip measurement. (A) shows the direction of the vectors used, and (B) shows the measurements of these vectors

Cage subsidence was additionally measured for patients with 2-year post-operative radiographs available (n = 29). Subsidence was defined as >25% decrease of the average of the anterior and posterior disc heights of this long-term radiograph compared to the average disc height measurement immediately post-operatively (Figure 3).^19,20 Due to the small sample size, statistical analysis was not performed to prevent misleading results.

Figure 3.

Lateral radiographs (A) immediately postoperative and (B) 2 years postoperative demonstrating cage subsidence

Statistical Analysis

Descriptive statistics for demographic characteristics, surgical factors, radiographic parameters, and sagittal alignment factors were reported either as mean ± standard deviation for continuous variables or as a frequency (percentage) for categorical variables.

For normally distributed variables, paired or unpaired t-tests were used. For non-normally distributed variables, Wilcoxon signed-rank test or Mann-Whitney U test was applied. A paired samples t-test was used to compare pre- and postoperative radiographic parameters, including spinopelvic alignment, DH, and vertebral slip. An independent samples t-test was used to compare screw and implant details for Nav vs Robo groups. Multivariable linear regression analysis was conducted to examine the effects of navigation modality (Nav vs Robo) on radiographic outcomes and slip reduction by adjusting spinal alignment and implant parameters. All statistical analyses were conducted on a complete case basis, performed using SPSS (version 29.0, IBM Corp., Armonk, NY, USA). Statistical significance was set at P < 0.05.

Results

Patient Demographic

This study included 176 patients with a mean age of 62.9 ± 11.1 years. Of these, 124 (70.5%) procedures were performed with robotic assistance, while 52 (29.5%) used navigational modalities. The mean operative time was 151.1 ± 68.0 minutes, and the estimated blood loss averaged 64.7 ± 55.7 mL (Table 1).

Table 1.

Patient Demographics and Operative Characteristics

Number of patients	176
Age (years), mean ± SD	62.9 ± 11.1
Height (in), mean ± SD	168.1 ± 9.6
Weight (kgs), mean ± SD	79.3 ± 18.4
Body Mass index (kg/m²), mean ± SD	27.9 ± 5.7
Charlson comorbidity index, mean ± SD	2.3 ± 1.4
Gender (male), n (%)	83 (47.2%)
Race, n (%)
White	144 (81.8%)
Black	13 (7.4%)
Asian	7 (4.0%)
Mixed or other	6 (3.4%)
Declined or unavailable	6 (3.4%)
ASA class, n (%)
I	11 (6.3%)
II	147 (83.5%)
III	18 (10.2%)
Side of approach, n (%)
Right	72 (40.9%)
Left	92 (52.3%)
Bilateral	8 (4.5%)
N/A	4 (2.3%)
Navigation modality, n (%)
Robot	124 (70.5%)
Navigation	52 (29.5%)
Operative time (minutes), mean ± SD	151.1 ± 68.0
Estimated blood loss (mL), mean ± SD	64.7 ± 55.7

ASA class = American Society of Anesthesiologists Physical Status Classification.

SD = standard deviation.

Implant Details

The overall mean pedicle screw diameter and length were 7.2 ± 0.8 mm and 48.9 ± 4.4 mm, respectively. The mean screw diameter for robotic vs navigation modalities was 7.2 ± 0.9 and 6.8 ± 0.6 respectively (P < 0.001), and the mean screw length was similar between groups (Robo, 49.2 ± 4.0 vs Nav, 47.8 ± 4.2 mm; P = 0.183). The mean cage height, width, and depth for robotic vs navigation modalities were 10.2 ± 2.1 and 10.4 ± 1.0 (P = 0.137), 21.9 ± 14.4 and 26.5 ± 9.8 (P < 0.001), and 18.8 ± 10.7 and 15.5 ± 9.4 (P = 0.006) respectively. Mean cage dimensions were 23.5 ± 13.3 mm in width, 17.9 ± 10.4 mm in depth, and 10.5 ± 1.0 mm in height (Table 2).

Table 2.

Implant Details

	Nav	Robo	P-value	Overall
Lumbar pedicle screw diameter (mm), mean ± SD	6.8 ± 0.6	7.2 ± 0.9	<0.001	7.2 ± 0.8
Lumbar pedicle screw length (mm), mean ± SD	47.8 ± 4.2	49.2 ± 4.0	0.183	48.9 ± 4.4
Cage width (mm), mean ± SD	26.5 ± 9.8	21.9 ± 14.4	<0.001	23.5 ± 13.3
Cage depth (mm), mean ± SD	15.5 ± 9.4	18.8 ± 10.7	0.006	17.9 ± 10.4
Cage height (mm), mean ± SD^a	10.4 ± 1.0	10.2 ± 2.1	0.137	10.5 ± 1.0
Patients with expandable cages, n (% of Robo or Nav cases)	34 (65%)	102 (82%)	0.015	136 (77.3%)
Level of cage, n (%)
L2-L3				1 (0.6%)
L3-L4				3 (1.7%)
L4-L5				147 (83.5%)
L5-S1				25 (14.2%)

Screw measurements are reported across all screws (#1-4). All 176 patients had measurements for pedicle screw #1, only 132 had screw #2, 32 had screw #3, and 10 had screw #4. Bolded values indicate statistical significance (P < 0.05).

^a5 patients were excluded who did not have cage height, width or depth values reported. Cage height median value was taken for patients with expandable cages.

Expandable cages were reported in 136 patients (77.3%), with 34 being Nav patients (65%) and 102 Robo patients (82%; P = 0.015). Five patients did not have recorded cage parameters and were excluded from these measurements. The most common level for MI-TLIF was L4/L5 (147 patients, 83.5%), followed by L5/S1 (25 patients, 14.2%; Table 2).

Radiographic Outcomes Comparison Between Pre- and Post-Operative Parameters

The 176 patients in this study had postoperative visits of 6 months, and the radiographic data from this time period was used for the analysis. Postoperative measurements showed significant improvements in anterior DH (11.4 ± 2.9 vs 8.2 ± 3.5; P < 0.001), posterior DH (6.7 ± 2.0 vs 6.2 ± 2.9; P < 0.04), and average DH (9.0 ± 2.0 vs 7.2 ± 2.9; P < 0.001) compared to preoperative values, with the greatest increase observed in anterior DH (3.2 ± 4.0 mm). Robo group achieved a significantly greater mean DH improvement (Nav, 22.5% vs Robo, 71.6%, P = 0.002). In spinal alignment, LL and dLL showed significant improvement from preoperative to postoperative values (LL: 50.5° vs 52.2°, P = 0.008; dLL: 27.8° ± 29.3°, P = 0.008), while the absolute change was small (Table 3). Other spinal alignment parameters did not show significant change. Regarding vertebral slip rate, the Robo and Nav group showed comparable slip reduction rates (Nav, 61.2% vs 65.8%; P = 0.32).

Table 3.

Disc Height Changes and Sagittal Spinal Alignment Changes Following Minimally Invasive Transforaminal Lumbar Interbody Fusion

	Preoperative	Postoperative	P-value^b	Absolute change
Anterior disc height (mm)	8.2 ± 3.5	11.4 ± 2.9	<.001 ^a	3.2 ± 4.0
Posterior disc height (mm)	6.2 ± 2.9	6.7 ± 2.0	0.040 ^a	0.5 ± 2.9
Average disc height (mm)	7.2 ± 2.9	9.0 ± 2.0	<.001 ^a	1.8 ± 3.0
Pelvic tilt (PT)	21.6 ± 8.0	21.2 ± 7.6	0.342	0.4 ± 6.1
Sacral slope (SS)	36.8 ± 9.5	37.7 ± 8.2	0.087	0.9 ± 6.9
Pelvic incidence (PI)	58.3 ± 11.4	58.8 ± 10.8	0.384	0.5 ± 7.4
Lumbar lordosis (LL)	50.5 ± 13.1	52.2 ± 11.6	0.008 ^a	1.7 ± 8.6
L1L4 lumbar lordosis	23.5 ± 9.9	22.9 ± 9.5	0.329	0.5 ± 7.2
L4S1 lumbar lordosis	27.8 ± 9.4	29.3 ± 9.2	0.008 ^a	1.5 ± 7.4
PI-LL	7.9 ± 13.5	6.6 ± 12.4	0.104	1.2 ± 10.1

^aRadiographic measurements are reported as mean ± SD. Bolded values indicate statistical significance (P < 0.05).

^bP-value from comparison between preoperative and postoperative parameters with paired t-test.

Multivariable Regression Analyses

Multivariable regression analysis showed significant effects of robotic surgery on both anterior and posterior DH (anterior: β = 0.21, P = 0.007; posterior: β = 0.22, P = 0.005). It indicated that the Robo group is significantly associated with DH improvement after adjusting other covariates such as screw size, cage size, and preoperative spinal alignment. In other covariates, smaller screw diameter and lower preoperative PT also showed a significant association with posterior disc height improvement (P = 0.02 and P = 0.049, respectively). Standardized β suggested the effect of PT was greater than navigation modality (Table 4).

Table 4.

Multivariable Regression Model for Disc Height Improvement

	Anterior Disc Height Improvement					Posterior Disc Height Improvement
	Unstandardized	SE	95% CIs	Standardized	P-value	Unstandardized	SE	95% CIs	Standardized	P-value
Nav vs Robo	61.7	22.8	17.0 to 106.4	0.21	0.007 ^a	48.4	17.0	15.1 to 81.7	0.22	0.005 ^a
Cage height	−4.7	5.3	−15.1 to 5.7	−0.07	0.38	1.5	4.0	−6.3 to 9.3	0.03	0.71
Screw diameter	−23.6	13.0	−49.1 to 1.9	−0.15	0.07	−23.4	9.7	−42.4 to -4.4	−0.19	0.020 ^a
Screw length	−2.4	2.6	−7.5 to 2.7	−0.07	0.37	−1.2	1.9	−4.9 to 2.5	−0.05	0.53
Preop PT	−4.8	9.7	−23.8 to 14.2	−0.29	0.62	−14.4	7.3	−28.7 to −0.1	−1.15	0.049 ^a
Preop SS	−3.9	10.0	−23.5 to 15.7	−0.28	0.70	−13.9	7.5	−28.6 to 0.8	−1.32	0.07
Preop PI	5.5	9.9	−13.9 to 24.9	0.47	0.58	14.0	7.4	−0.5 to 28.5	1.59	0.06
Preop LL	−1.3	1.0	−3.3 to 0.7	−0.12	0.22	0.4	0.8	−1.1 to 1.9	0.05	0.59

Unstandardized, Standardized = coefficient (β); PT = pelvic tilt; SS = sacral slope; PI = pelvic incidence; LL = lumbar lordosis; SE = standard error; CIs = confidence interval.

^aBolded values indicate statistical significance (P < 0.05).

For postoperative LL, navigation modalities demonstrated a significantly larger postoperative restoration of LL than robotic surgery (β = −0.14, P = 0.006). Furthermore, preoperative LL and cage height had significant effects on postoperative LL (β = 0.64, P < 0.001 and β = −0.18, P < 0.001, respectively), meaning patients with larger LL preoperatively and smaller cage heights had more benefits from this surgery. The effect size of preoperative LL was significantly greater than navigation modality (Table 5).

Table 5.

Multivariable Regression Model for Postoperative Lumbar Lordosis

	Unstandardized β	SE	95% CIs	Standardized β	P-value
Nav vs Robo	−3.5	1.2	−5.9 to -1.1	−0.14	0.006 ^a
Cage height	−2.1	0.5	−3.1 to -1.1	−0.18	<0.001 ^a
Cage width	−0.1	0.1	−0.3 to 0.1	−0.08	0.19
Cage depth	−0.1	0.1	−0.3 to 0.1	−0.09	0.17
Screw diameter	0.5	0.7	−0.9 to 1.9	0.03	0.51
Screw length	−0.0	0.1	−0.2 to 0.2	−0.01	0.90
Preop pelvic tilt	−0.8	0.5	−1.8 to 0.2	−0.58	0.11
Preop sacral slope	−0.6	0.5	−1.6 to 0.4	−0.46	0.30
Preop pelvic incidence	0.8	0.5	−0.2 to 1.8	0.77	0.14
Preop LL	0.6	0.1	0.4 – 0.8	0.64	<0.001 ^a

^aBolded values indicate statistical significance (P < 0.05). LL, lumbar lordosis; SE, standard error.

Navigation modality did not have significant effect on slip reduction rate (β = 0.10, P = 0.32), while the higher cage height was significantly associated with a lower slip reduction rate (β = −0.21, P = 0.022), although the clinical effect size was modest. Other covariates also did not have a significant association with slip reduction rate (Table 6).

Table 6.

Multivariable Regression Model for Slip Reduction Rate

	Unstandardized β	SE	95% CIs	Standardized β	P-value
Nav vs Robo	5.6	5.6	−5.4 to 16.6	0.10	0.32
Cage height	−3.0	1.3	−5.5 to −0.5	−0.21	0.022 ^a
Screw diameter	2.3	3.3	−4.2 to 8.8	0.07	0.49
Screw length	0.1	0.6	−1.1 to 1.3	0.01	0.90
Preop pelvic tilt	−0.4	2.0	−4.3 to 3.5	−0.11	0.86
Preop sacral slope	−0.4	2.1	−4.5 to 3.7	−0.15	0.84
Preop pelvic incidence	−0.1	2.1	−4.2 to 4.0	−0.04	0.97
Preop LL	0.5	0.3	−0.1 to 1.1	0.22	0.09

Bolded values indicate statistical significance (P < 0.05). LL, lumbar lordosis; SE, standard error.

Cage Subsidence

Cage subsidence assessment was conducted on patients who had follow-up radiographs of 24 months or greater (n = 29). Of these 29 patients, 16 used intraoperative navigation and 13 used robotic modalities. Four patients experienced subsidence (18.2%), with two (12.5%) from the Nav group and two from the Robo group (15.4%).

Discussion

This study evaluated the influence of robotic arm on postoperative radiographic outcomes, including DH, spinal alignment, slip reduction, and cage subsidence, in patients undergoing 1-level MI-TLIF. Robo group was more likely to use expandable and deeper cages and used screws with larger diameters, while the navigation group used slightly longer screws and wider cages on average, although screw length differences were not statistically significant. In terms of radiographic outcomes, Robo group achieved superior disc height restoration compared to the Nav group. In contrast, greater LL restoration was achieved through imaging-only navigation modalities. Preoperative PT and LL were also significantly associated with posterior DH improvement and LL restoration, respectively. Slip reduction and subsidence were similar between Robo and Nav groups. Overall, preoperative alignment factors and implant characteristics seem to have a larger effect size than navigation modality. These results highlight the potential benefits of robotic navigation in enhancing disc height restoration and imaging-only navigation in LL restoration while underscoring the nuanced role of implant-related factors in surgical outcomes.

Numerous studies regarding spondylolisthesis have examined the influence of fusion techniques on disc height improvement.^12,21 However, no studies to date have directly compared different navigation modalities. Traditionally, robotic navigation systems are designed to improve the accuracy of implant placement.^22,23 Robotic navigation has been shown to facilitate the placement of larger screws, which may enhance construct strength.²⁴ Among patients with similar spinal rigidity and cage sizes, achievable radiographic reduction goals may vary depending on construct strength.^25-27 Furthermore, surgeons may hesitate to perform radical reductions in cases with insufficient screw parameters due to the risk of posterior instrumentation failure. While the reduced mental workload associated with robotic navigation could theoretically allow surgeons to focus more on achieving optimal radiographic outcomes,^28,29 it is also possible that the observed differences reflect the preferences and surgical philosophies of those who preferentially use robotic systems rather than the technology itself. Therefore, the apparent radiographic advantages of robotic-assisted surgery should be interpreted with caution, acknowledging the potential influence of surgeon selection and practice patterns.

Restoring LL plays a crucial role in the long-term stability of the construct, potentially reducing the risk of subsequent complications such as adjacent segment disease.³⁰ The effectiveness of TLIF in LL restoration has been extensively studied, and Aoki et al. demonstrated that LL restoration is closely associated with preoperative LL values and can lead to improved mid-term clinical outcomes.³¹ Consistent with these findings, another study reported that both MI and open TLIF approaches achieve comparable improvements in LL.³² In our multivariable regression model, lower cage height, greater preoperative LL, and the use of navigation rather than robotic guidance were independently associated with postoperative LL. After adjusting for cage height and other implant-related variables, the residual association between navigation and greater postoperative LL likely reflects differences in preoperative sagittal alignment patterns and intraoperative positioning rather than screw size or cage width alone. Recent evidence suggests that, although overall LL correction may be similar between approaches, an open approach may provide an advantage in selected patients with high PI, indicating that postoperative regional alignment is influenced by the interaction between patient-specific spinopelvic anatomy and surgical technique rather than by the approach alone.³³ Moreover, because a single-level TLIF has an inherently limited capacity to substantially increase global LL due to compensatory mechanisms in other lumbar spine levels,³⁴ surgeons should consider both preoperative sagittal profiles and construct selection when planning realistic segmental alignment correction.³¹ Future research should clarify which specific aspects of surgical modality, including navigation platforms, robotic systems, and exposure techniques, meaningfully influence alignment-related outcomes beyond patient-specific anatomical factors.

Cage size plays a critical role in restoring spinal morphology appropriately.³⁵ Most of the cages used in this patient series were expandable cages with a 15-degree lordotic angle. This system effectively improved disc height in minimally invasive approaches, optimizing fit while maintaining alignment. However, the surface design of these cages tends to grip the endplate, limiting segment mobility during rod and screw stabilization.³⁶ This may explain why higher cages could lock the segment in place, reducing its ability to slide or move and preventing an effective reduction maneuver. Additionally, greater cage height in non-expandable cage can also fix the segment, diminishing the reduction effect of the posterior construct, as reflected by the regression analysis. Finally, there was a statistically significant difference in use of expandable cages among patients in the Robo and Nav groups; the Robo group had a greater use of expandable cages, which may account for the greater disc height achieved with robotic modalities.

The necessity of slip reduction in spondylolisthesis remains debated. A systematic review suggested comparable intraoperative blood loss, operative time, and clinical outcomes between reduction and in-situ fusion in patients with low-grade spondylolisthesis.³⁷ On the other hand, a randomized controlled trial showed better radiological outcomes including fusion rates in the intentional reduction cohort although clinical outcome scores were aligned with other studies.^38,39 In our cohort, the degree of slip correction did not differ between robotic and navigation modalities, suggesting that the extent of reduction may depend more on the surgeon’s intraoperative decision-making than on the surgical technology itself. Considering the inherent precision and stability that both systems provide for screw trajectory, the absence of a difference in postoperative slip reduction is reasonable. However, this study did not capture the surgeon’s intended amount of reduction or intraoperative strategy. Future research should prospectively evaluate how each modality influences the surgeon’s comfort, confidence, and ability to achieve the desired reduction under real operative conditions, including the relationship between planned and achieved slip correction.

Finally, greater interbody height restoration may increase the risk of subsidence due to the additional forces experienced by the endplates, so it was necessary to analyze long-term radiographic outcomes to determine if this was a durable finding.⁴⁰ Marchi et al. demonstrated that larger cage-endplate interfaces offer superior construct stability and decreases the chance of subsidence.¹⁹ Depending on patient anatomy, surgeons may use different technologies to optimize this interface; for example, intraoperative navigation can facilitate the use of wider interbody cages, whereas robotic guidance may help position cages more anteriorly along the anteroposterior axis. In the subgroup of patients with at least 24 months of follow-up radiographs, no clear difference in the rate of cage subsidence was observed between MI-TLIFs performed with robotic vs navigation modalities. However, this analysis was limited by a small sample size (n = 29) and the low number of subsidence events (n = 4), and was likely underpowered to detect modest between-group differences. Therefore, the absence of an observed difference should not be interpreted as evidence of equivalence between modalities. The overall subsidence rate in this cohort was 18.2%, which is comparable to previous reports showing rates between approximately 10% and 35% after MI-TLIF, including studies using expandable cages.^20,41-43 Therefore, our findings suggest that the subsidence rate observed in our series falls within the expected range and does not indicate an increased risk attributable to either navigation or robotic guidance.

There are several limitations to this study. First, this is a retrospective analysis conducted at a single academic institution, which may introduce selection bias and limit the generalizability of our findings. Surgeons’ preferences for navigation modality and instrument selection may also have influenced which patients underwent each technique and thus represent an important source of residual confounding. Second, neither intraoperative navigation nor the robotic arm was directly employed to guide sagittal alignment correction, so the observed radiographic differences are indirect and may be confounded by unmeasured factors such as mental workload, construct strategy, and case complexity. We attempted to account for these factors using multivariable regression analysis, but unmeasured confounding cannot be excluded. Third, the actual cage height post-expansion was unclear. While a consistent strategy aimed to match the adjacent or healthy disc height, adjustments were made intraoperatively based on bone quality and endplate conditions, potentially introducing bias. Fourth, the relatively small overall sample size, particularly in the subsidence subgroup, represents a major limitation and reduces the statistical power to detect modest differences between modalities. Fifth, due to the limited number of patients, we did not perform propensity score matching, as matching would have further reduced the analyzable cohort and potentially produced unstable estimates. Instead, we used multivariable regression analysis to adjust for baseline imbalances between groups. Therefore, causal inferences should be interpreted with caution. Finally, this study did not include clinical outcome measures, limiting the ability to determine whether the observed radiographic improvements translate into patient benefits. Therefore, the clinical utility of the robotic arm should be interpreted in the context of other factors such as cost-effectiveness, learning curve, and accessibility. Nevertheless, our findings contribute to the growing discussion on the role of robotic arms not only in enhancing screw placement accuracy but also in improving surgical efficiency to achieve experienced surgeons’ goals.

Conclusions

In conclusion, among patients undergoing MI-TLIF, the use of a robotic arm was associated with greater disc height restoration compared with intraoperative navigation, whereas LL restoration appeared to be influenced primarily by patients’ preoperative sagittal alignment rather than with the choice of navigation modality itself. Surgeons should select the operative modality based on patient anatomy, baseline alignment, and specific surgical goals, rather than expecting any single technology to overcome unfavorable preoperative conditions. Future studies should integrate clinical outcomes with radiographic parameters to more comprehensively evaluate the impact of robotic and navigation guidance in minimally invasive spine surgery.

Footnotes

ORCID iDs

Zora C. Hahn

Sereen Halayqeh

Adrian T. H. Lui

Francis C. Lovecchio

Sheeraz A. Qureshi

Sravisht Iyer

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: no direct funding was received for this study. However, this study used REDCap (Research Electronic Data Capture) hosted at Weill Cornell Medicine Clinical and Translational Science Center supported by the National Center for Advancing Translational Science of the National Institute of Health under award number: UL1 TR002384.

Declaration of Conflicting Interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Globus Medical and Stryker (Paid presenter or speaker); Vertebral Columns/International Society for the Advancement of Spine Surgery (ISASS) (Editorial or Governing Board).

Previous Presentations

No portion of this paper has been previously presented.

References

Cui

Han

Wei

, et al. Robot-assisted minimally invasive transforaminal lumbar interbody fusion in the treatment of lumbar spondylolisthesis. Orthop Surg. 2021;13(7):1960-1968.

Garcia

Akinduro

De Biase

, et al. Robotic-assisted vs nonrobotic-assisted minimally invasive transforaminal lumbar interbody fusion: a cost-utility analysis: a cost-utility analysis. Neurosurgery. 2022;90(2):192-198.

Han

Tang

Han

, et al. Comparison of outcomes between robot-assisted minimally invasive transforaminal lumbar interbody fusion and oblique lumbar interbody fusion in single-level lumbar spondylolisthesis. Orthop Surg. 2021;13(7):2093-2101.

Wang

Liu

Zhu

. Effectiveness and safety of robot-assisted minimally invasive transforaminal lumbar interbody fusion for degenerative lumbar spinal diseases: a systematic review and meta-analysis. J Robot Surg. 2024;18(1):37.

Shahi

Subramanian

Araghi

, et al. Comparison of robotics and navigation for clinical outcomes after minimally invasive lumbar fusion. Spine. 2023;48(19):1342-1347.

Siepe

Hitzl

Meschede

Sharma

Khattab

Mayer

. Interdependence between disc space height, range of motion and clinical outcome in total lumbar disc replacement. Spine. 2009;34(9):904-916.

. Comparison of biomechanical performance among posterolateral fusion and transforaminal, extreme, and oblique lumbar interbody fusion: a finite element analysis. World Neurosurg. 2019;129:e890-e899.

Kepler

Rihn

Radcliff

, et al. Restoration of lordosis and disk height after single-level transforaminal lumbar interbody fusion: lordosis and disk height after TLIF. Orthop Surg. 2012;4(1):15-20.

Tian

Guo

, et al. Adequate restoration of disc height and segmental lordosis by lumbar interbody fusion decreases adjacent segment degeneration. World Neurosurg. 2018;118:e856-e864.

10.

Debnath

Chatterjee

McConnell

Jha

Chakraburtty

. Interbody fusion in low grade lumbar spondylolsithesis: clinical outcome does not correalte with slip reduction and neural foraminal dimension. Asian Spine J. 2016;10(2):314-320.

11.

Floman

Millgram

Ashkenazi

Smorgick

Rand

. Instrumented slip reduction and fusion for painful unstable isthmic spondylolisthesis in adults. J Spinal Disord Tech. 2008;21(7):477-483.

12.

Lin

Liu

Lin

. Robot-guided versus freehand fluoroscopy-guided minimally invasive transforaminal lumbar interbody fusion: a single-institution, observational, case-control study. Neurosurg Focus. 2022;52(1):E9.

13.

Wang

, et al. Comparison of robot-assisted versus fluoroscopy-assisted minimally invasive transforaminal lumbar interbody fusion for degenerative lumbar spinal diseases: 2-year follow-up. J Robot Surg. 2023;17(2):473-485.

14.

Harris

Taylor

Minor

, et al. The REDCap consortium: building an international community of software platform partners. J Biomed Inf. 2019;95(103208):103208.

15.

Harris

Taylor

Thielke

Payne

Gonzalez

Conde

. Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inf. 2009;42(2):377-381.

16.

Qureshi

. Pearls: improving upon minimally invasive transforaminal lumbar interbody fusion. Clin Orthop Relat Res. 2019;477(3):501-505.

17.

Liu

Zuo

Zhang

Zhou

. Does MIS-TLIF or TLIF result in better pedicle screw placement accuracy and clinical outcomes with navigation guidance? BMC Musculoskelet Disord. 2022;23(1):153.

18.

Virk

Vaishnav

Sheha

, et al. Combining expandable interbody cage technology with a minimally invasive technique to harvest iliac crest autograft bone to optimize fusion outcomes in minimally invasive transforaminal lumbar interbody fusion surgery. Clin Spine Surg. 2021;34(9):E522-E530.

19.

Marchi

Abdala

Oliveira

Amaral

Coutinho

Pimenta

. Radiographic and clinical evaluation of cage subsidence after stand-alone lateral interbody fusion. J Neurosurg Spine. 2013;19(1):110-118.

20.

Subramanian

Merrill

Shahi

, et al. Predictors of subsidence and its clinical impact after expandable cage insertion in minimally invasive transforaminal interbody fusion. Spine. 2023;48(23):1670-1678.

21.

Chang

Kim

Choo

Lee

. Transforaminal lumbar interbody fusion (TLIF) versus oblique lumbar interbody fusion (OLIF) in interbody fusion technique for degenerative spondylolisthesis: a systematic review and meta-analysis. Life. 2021;11(7):696.

22.

Asada

Simon

, et al. Robot-navigated pedicle screw insertion can reduce intraoperative blood loss and length of hospital stay: analysis of 1,633 patients utilizing propensity score matching. Spine J. 2024;24(1):118-124.

23.

Asada

Subramanian

Simon

, et al. Level-specific comparison of 3D navigated and robotic arm-guided screw placement: an accuracy assessment of 1210 pedicle screws in lumbar surgery. Spine J. 2024;24(10):1872-1880.

24.

Shafi

Pompeu

Vaishnav

, et al.

Does robot-assisted navigation influence pedicle screw selection and accuracy in minimally invasive spine surgery?

Neurosurg Focus. 2022;52(1):E4.

25.

Delman

Cage

Lausé

Roberto

Gupta

Klineberg

. Anterior and posterior fusion for large, rigid idiopathic scoliosis: does implant density matter? World Neurosurg. 2020;134:e37-e45.

26.

Tohmeh

Khorsand

Watson

Zielinski

. Radiographical and clinical evaluation of extreme lateral interbody fusion: effects of cage size and instrumentation type with a minimum of 1-year follow-up. Spine. 2014;39(26):E1582-E1591.

27.

Khodaee

Sager

Götschi

, et al. Quantifying in vitro load-sharing in spinal fusion surgical constructs using strain sensor-equipped fixation rods and force sensor-equipped intervertebral cages. Spine J. 2025;25(9):2108-2116.

28.

Guru

Shafiei

Khan

Hussein

Sharif

Esfahani

. Understanding cognitive performance during robot-assisted surgery. Urology. 2015;86(4):751-757.

29.

Moore

Wilson

McGrath

Waine

Masters

RSW

Vine

. Surgeons’ display reduced mental effort and workload while performing robotically assisted surgical tasks, when compared to conventional laparoscopy. Surg Endosc. 2015;29(9):2553-2560.

30.

Segar

Robertson

Boody

Kepler

. Does restoration of anatomic segmental lordosis reduce the rate of adjacent segment disease in 1 level fusions for a degenerative spondylolisthesis? Clin Spine Surg. 2020;33(3):89-91.

31.

Aoki

Inoue

Takahashi

, et al. Postoperative change in lumbopelvic alignment after short-segment transforaminal lumbar interbody fusion is related to preoperative postural difference in lumbar lordosis. J Orthop Sci. 2023;28(2):321-327.

32.

Ali

EMS

Abdeen

Saleh

. Minimally invasive versus mini-open transforaminal lumbar interbody fusion in managing low-grade degenerative spondylolisthesis. Acta Neurochir. 2024;166(1):365.

33.

Hirase

Vaishnav

Zhang

, et al. Impact of single-level minimally invasive versus open L4-L5 and L5-S1 transforaminal lumbar interbody fusion on postoperative distal and lumbar lordosis. Spine. 2025. Published online March 7, 2025. doi:10.1097/BRS.0000000000005325

34.

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.

35.

Gambhir

Wang

Pelletier

Walsh

Ball

. How does cage lordosis influence postoperative segmental lordosis in lumbar interbody fusion. World Neurosurg. 2019;126:e606-e611.

36.

Buttermann

Beaubien

Freeman

Stoll

Chappuis

. Interbody device endplate engagement effects on motion segment biomechanics. Spine J. 2009;9(7):564-573.

37.

Bai

Chen

Liu

, et al. Is reduction better than arthrodesis in situ in surgical management of low-grade spondylolisthesis? A system review and meta analysis. Eur Spine J. 2017;26(3):606-618.

38.

Lian

Hou

, et al.

Posterior lumbar interbody fusion for aged patients with degenerative spondylolisthesis: is intentional surgical reduction essential?

Spine J. 2013;13(10):1183-1189.

39.

Scheer

Auffinger

Wong

, et al. Minimally invasive transforaminal lumbar interbody fusion (TLIF) for spondylolisthesis in 282 patients: in situ arthrodesis versus reduction. World Neurosurg. 2015;84(1):108-113.

40.

Pisano

Fredericks

Steelman

Riccio

Helgeson

Wagner

. Lumbar disc height and vertebral hounsfield units: association with interbody cage subsidence. Neurosurg Focus. 2020;49(2):E9.

41.

Levy

Pinter

Reed

, et al. Transforaminal lumbar interbody fusion subsidence: computed tomography analysis of incidence, associated risk factors, and impact on outcomes. J Neurosurg Spine. 2024;41(4):463-472.

42.

Lin

Kim

Kotheeranurak

Chen

Rui

. Does the application of expandable cages in TLIF provide improved clinical and radiological results compared to static cages? A meta-analysis. Front Surg. 2022;9:949938.

43.

Yao

Chou

Lin

Wang

Liu

Chang

. Risk factors of cage subsidence in patients received minimally invasive transforaminal lumbar interbody fusion. Spine. 2020;45(19):E1279-E1285.

44.

“AlteraTM product page.” Expandable Technology, 2024. https://www.globusmedical.com/expandabletechnology/altera/