Near-Infrared Fluorescence Imaging With Indocyanine Green for Intraoperative Nerve Root Visualization in Spinal Surgery: From Preclinical Studies to a Pilot Randomized Controlled Trial

Abstract

Study Design

Pilot randomized controlled trial.

Objectives

Iatrogenic nerve injury is a major complication in endoscopic spinal surgery, potentially causing serious neurological deficits. Near-infrared (NIR) fluorescence imaging with indocyanine green (ICG) has shown promise for intraoperative nerve root identification. This study assessed the feasibility, optimal dosing, safety, and mechanism of ICG fluorescence for nerve root visualization, transitioning from preclinical to clinical phases.

Methods

In the preclinical phase, 36 rabbits were assigned to ICG dose groups (1.4, 2.8, or 5.5 mg/kg, corresponding to 0.5, 1, or 2 mg/kg in humans) and observation times (3, 6, 12, or 24 hours). Fluorescence signals in lumbar nerve roots were quantified by signal-to-background ratio (SBR) and mean fluorescence intensity (MFI). Histological analyses explored ICG retention mechanisms. In the clinical phase, 40 patients undergoing unilateral biportal endoscopic surgery for lumbar disc herniation were randomized into different ICG dose groups (0, 0.5, 1, or 2 mg/kg), administered 1.5 hours preoperatively. Intraoperative fluorescence parameters, nerve root identification time, and perioperative outcomes (VAS and ODI scores) were assessed.

Results

In preclinical studies, the 2.8 and 5.5 mg/kg groups showed peak SBR and MFI at 3 hours post-injection. Histology revealed ICG accumulation in nerve root microvascular regions. In the clinical study, the 2 mg/kg group had the highest SBR and MFI, reducing nerve root identification time without significant adverse events.

Conclusion

ICG fluorescence imaging is a feasible and safe technique for intraoperative nerve root visualization, with ICG accumulation attributed to the enhanced permeability and retention effect.

Keywords

indocyanine green near-infrared fluorescence imaging spinal nerve roots disc herniation endoscopy

Introduction

Iatrogenic nerve injury is a significant surgical complication that can lead to substantial disability across various specialties.¹ In the United States, approximately 600,000 cases of iatrogenic nerve injury are reported annually.² The risk of nerve injury remains a concern in minimally invasive surgeries, where nerve identification often relies on interpreting anatomical landmarks rather than direct visual observation.³

Endoscopic spinal surgery has become a widely adopted minimally invasive technique for treating lumbar disc herniation (LDH), offering advantages such as smaller incisions, reduced muscle and bone tissue damage, less postoperative pain, and faster recovery.⁴ Recent advancements in high-resolution endoscopic imaging have significantly improved intraoperative visualization of neural structures and surrounding tissues, enhancing surgical precision and safety. However, despite these advancements, complications such as nerve injury and dural tears remain significant clinical challenges, especially for less experienced surgeons who are still within the learning curve.^5,6 The incidence of nerve injury in endoscopic spinal surgery ranges from 1.1% to 6.3%, primarily due to the difficulty in distinguishing nerve roots from adjacent structures, such as the ligamentum flavum, dural sac, nucleus pulposus, and adipose tissue.^7-11 This increases the risk of inadvertent nerve damage, potentially leading to postoperative sensory disturbances, motor dysfunction, or long-term neurological deficits. Therefore, optimizing intraoperative neuroprotective strategies and improving the visualization of neural structures are essential for minimizing iatrogenic nerve injuries and enhancing the safety and clinical efficacy of endoscopic spinal surgery.

Intraoperative nerve monitoring can be achieved through various techniques, such as ultrasound, optical coherence tomography, and confocal endoscopic imaging.^12-14 However, these methods have limitations in specificity, spatial resolution, and wide-field imaging capabilities, making real-time nerve visualization challenging. In spinal surgery, intraoperative neurophysiological monitoring (IONM) is widely used for real-time functional assessment of nerves to detect potential nerve injuries.¹⁵ However, IONM is inherently limited as it can only identify functional abnormalities after nerve injury has occurred and does not provide direct anatomical visualization, thus restricting its role in nerve injury prevention.¹⁶ Additionally, some surgeons have attempted methylene blue staining of the intervertebral disc to distinguish nerve roots from the nucleus pulposus.¹⁷ However, this method does not allow direct nerve visualization and may increase operative time and intraoperative fluoroscopic exposure, potentially affecting surgical safety and efficiency.¹⁸

With recent advancements in optical imaging technology, fluorescence-guided surgery has shown significant potential in improving intraoperative anatomical visualization.^19,20 Indocyanine green (ICG), the only fluorescent dye approved by the U.S. Food and Drug Administration (FDA) for intraoperative near-infrared (NIR) imaging, has been widely applied in tumor resection,^21-24 lymphatic mapping,²⁵ and vascular perfusion assessment.²⁶ Recent clinical studies have demonstrated that ICG fluorescence imaging effectively facilitates the intraoperative identification of thoracic sympathetic ganglia,^27,28 facial nerves,²⁹ and pelvic nerves,^30,31 offering a novel approach for nerve protection and precision surgery. However, these applications have focused on peripheral nerve trunks or autonomic ganglia. The feasibility of ICG fluorescence for visualizing the spinal nerve roots within the confined environment of the spinal canal, as well as the systematic optimization of its dose and timing for this specific application, remains unestablished.

This study is a translational exploration from preclinical studies to clinical research, hypothesizing that ICG fluorescence imaging can enable intraoperative visualization of nerve roots and evaluating its potential utility in spinal surgery (Scheme 1). The primary objectives are: (1) to assess the feasibility and safety of ICG fluorescence imaging for visualizing nerve roots during endoscopic spinal surgery; (2) to optimize the dosage and timing of ICG administration to enhance imaging quality and clinical applicability; and (3) to investigate the mechanisms underlying nerve root fluorescence imaging. This study aims to introduce an innovative and intuitive technique for nerve visualization in minimally invasive spinal procedures, thereby improving intraoperative nerve identification and enhancing surgical precision.

Scheme 1.

Schematic illustration of the procedure and imaging mechanism for endoscopic near-infrared fluorescence imaging of the nerve root using ICG

Materials and Methods

Study Drug

Indocyanine green (ICG; 25 mg/vial) was purchased from Ruida Pharmaceutical Co., Ltd. This compound is a near-infrared (NIR) fluorophore with a peak excitation wavelength of 805 nm, a peak emission wavelength of 830 nm, and a molecular weight of 775 Da.

Imaging Device

Both the preclinical and clinical studies utilized the FloNavi® fluorescence imaging system (Guangdong OptoMedic Technologies, Inc). This system offers an endoscopic imaging platform and a handheld open-surgery camera, each capable of high-definition dual-band imaging (white light and NIR). During near-infrared (NIR) fluorescence imaging, the system displayed three imaging modes in separate subwindows: (1) white-light mode for anatomical visualization and surgical field reference; (2) grayscale mode to depict NIR signal intensity and enhance fluorescence contrast; and (3) color-scale fluorescence mode using pseudocolor mapping to illustrate NIR fluorescence intensity and spatial distribution.

Part A: Preclinical Studies

Animals

The study protocol was approved by the institutional animal ethics committee and adhered to the Guide for the Care and Use of Laboratory Animals prepared by the National Academy of Sciences, Institute of Laboratory Animal Resources. A total of 36 male New Zealand white rabbits (mean body weight, 2.5 ± 0.2 kg) were used. All procedures complied with the ARRIVE Guidelines for reporting animal research.³²

ICG Administration

Thirty-six healthy rabbits were randomly divided into 12 groups according to ICG dose (1.4 mg/kg, 2.8 mg/kg, or 5.5 mg/kg) and observation time (3, 6, 12, or 24 h). ICG was administered via slow bolus injection through the ear marginal vein. The doses were determined based on a standard human weight of 60 kg (0.5, 1, or 2 mg/kg) using the Dosage Conversion Formula for Different Types of Subjects.³³

Animal Anatomy and Fluorescence Imaging Procedures

After ICG injection and upon reaching the predetermined observation time, rabbits were anesthetized with pentobarbital and euthanized by air embolism. The lumbar region was shaved, and the skin and fascia were incised. Muscles were bluntly dissected to expose the L5-S1 spinous processes and articular process joints, which were removed to reveal the spinal cord and the L6 and L7 nerve roots for fluorescence imaging. Fluorescence imaging was performed in the following sequence: (1) Endoscopic Imaging: The L6 and L7 nerve roots were imaged using the fluorescence endoscopic system. The endoscopic lens was positioned approximately 1 cm from the target, and room lights were turned off to minimize background interference. (2) Open-Surgery Imaging: A handheld open-surgery camera was used to image the entire surgical field from a distance of about 30 cm, maintaining a vertical orientation relative to the exposed tissues. (3) Ex Vivo Imaging: Excised nerve root and muscle specimens were imaged with the handheld open-surgery camera to further assess fluorescence signals.

Histological and Fluorescence Microscopic Analyses

Nerve root and muscle samples were formalin-fixed, dehydrated in sucrose, embedded in optimal cutting temperature compound, and cryosectioned at 5 µm thickness. Unstained sections were scanned using the Odyssey® Dual-Color Infrared Fluorescence Imaging System (LI-COR, USA) to determine ICG fluorescence intensity and distribution. These sections were subsequently stained with hematoxylin and eosin (H&E).

Near-infrared scan images were compared with the corresponding H&E-stained sections. Adjacent serial sections were processed for immunofluorescence staining of neurofilament protein NF200, with 4′,6-diamidino-2-phenylindole (DAPI) as a nuclear counterstain, to better visualize nerve root structures. Additional nerve root sections were immunohistochemically stained for the endothelial marker CD31 to investigate the mechanism of ICG accumulation in nerve roots.

Part B: Clinical Trial

Study Design

A prospective, randomized, open-label clinical trial was approved by the institutional Ethical Review Authority in 2023 and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants, and the trial was registered at ClinicalTrials.gov. The manuscript follows the Consolidated Standards of Reporting Trials (CONSORT) guidelines.

The trial was designed to include 40 subjects who met the following criteria: age 20-60 years, clinical and imaging evidence of lumbar disc herniation, and a requirement for unilateral biportal endoscopy (UBE) surgery for nerve root decompression. Exclusion criteria were known indocyanine green or iodine allergy, thyroid-related disease, severe chronic liver or kidney disease, ankylosing spondylitis, lumbar instability, or bony spinal stenosis.

After providing informed consent, participants were randomly assigned to one of four groups (control, 0.5, 1, or 2 mg/kg ICG) using a block randomization method with a 1:1:1:1 allocation ratio and a block size of 4. The group assignment was sealed until the time of ICG administration. Based on the preclinical findings, ICG was injected intravenously 1.5 hours before surgery. The CONSORT flow diagram is shown in Figure 1. The sample size (n = 10 per group) was based on established guidelines for pilot studies.³⁴

Figure 1.

CONSORT flowchart of the clinical trial

All procedures were performed under general anesthesia by the same surgical team. In the control group, patients underwent conventional posterior endoscopic nerve root decompression without fluorescence guidance. In the experimental groups, near-infrared fluorescence imaging was utilized to facilitate nerve root identification, while all other surgical steps remained identical. The operative time from ligamentum flavum removal to definitive nerve root visualization was recorded. This endpoint was uniformly defined as the time from ligamentum flavum removal until the primary surgeon verbally confirmed the definitive identification of the nerve root. Additionally, any adverse reactions associated with ICG administration were documented.

Clinical Evaluation

Low back pain and leg pain were evaluated by the 10-point visual analogue scale (VAS) before surgery and at 1 day, 1 week, and 1 month postoperatively. Functional outcomes were assessed using the Oswestry Disability Index (ODI, 0-100%). Any neurological complications were recorded. All postoperative clinical outcomes were collected by an independent assessor who was blinded to the patient’s group assignment.

Statistical Analysis

Fluorescence image data were analyzed post hoc using ImageJ (National Institutes of Health; https://rsb.info.nih.gov/ij/) to quantify fluorescence intensity within regions of interest (ROIs). The signal-to-background ratio (SBR) in each image was calculated by dividing the mean fluorescence intensity (MFI) of the nerve root ROI by the MFI of the background. In the preclinical study, measurements were performed on four nerve roots from three rabbits in each group (n = 12). In the clinical trial, three images per patient were analyzed to obtain average MFI and SBR values, reducing measurement error. All quantitative analyses of fluorescence images were performed by two independent assessors who were blinded to the group assignments, and the average value was used for analysis. An SBR ≥2 was considered fluorescence-positive based on established criteria in NIR fluorescence imaging research.^21,35

GraphPad Prism 10 software was used for statistical analyses. One-way analysis of variance (ANOVA), Student’s t test, or the chi-square test was applied as appropriate. Data are presented as the mean ± standard deviation, and a P value less than 0.05 was considered statistically significant.

Results

Part A: Preclinical Studies

Characteristics of ICG Fluorescence Imaging in Rabbit Nerve Roots

Firstly, this study evaluated the effectiveness of ICG near-infrared fluorescence imaging for visualizing nerve roots in a rabbit model. ICG was administered at doses of 1.4 mg/kg (equivalent to 0.5 mg/kg in humans), 2.8 mg/kg (equivalent to 1 mg/kg in humans), and 5.5 mg/kg (equivalent to 2 mg/kg in humans), with fluorescence endoscopic imaging performed at 3, 6, 12, and 24 hours post-injection, resulting in 12 dose-time subgroups (Figure 2A). Among these, five subgroups exhibited significant fluorescence signals, specifically 3 h 2.8 mg/kg, 3 h 5.5 mg/kg, 6 h 2.8 mg/kg, 6 h 5.5 mg/kg, and 12 h 5.5 mg/kg (Figure 2B). The 3 h 2.8 mg/kg and 3 h 5.5 mg/kg groups demonstrated the highest SBR and MFI (P < 0.001), with no statistically significant difference between them (Figure 2C). These findings indicate that ICG administered at an earlier time point (3 h) and at moderate-to-high doses (2.8-5.5 mg/kg) significantly enhances nerve root fluorescence signals, optimizing intraoperative visualization.

Figure 2.

In vivo fluorescence imaging of rabbit nerve roots. (A) Schematic diagram of the animal experimental workflow (Created with BioRender.com). (B) Endoscopic fluorescence imaging of rabbit nerve roots in different ICG administration time-dose groups; images in each group are presented sequentially as white-light images, NIR images, and color-scale fluorescence images. (C) SBR of nerve roots in different time-dose groups. (D) MFI of nerve roots in different time-dose groups. Data are presented as mean ± standard deviation (n = 12). Ns = no significance; *P < 0.05; **P < 0.01; ***P < 0.001

Fluorescence imaging of the surgical field using a handheld open-field fluorescence imaging system revealed strong fluorescence signals in the L6 and L7 nerve roots, providing clear contrast against the surrounding tissues (Figure 3A). Ex vivo imaging further confirmed that the fluorescence intensity of nerve roots was significantly higher than that of adjacent muscle tissues. In summary, these results suggest that under specific dose-time combinations, ICG provides excellent nerve root visualization, significantly enhancing contrast between nerve and surrounding tissues, with potential applications in intraoperative nerve localization and protection.

Figure 3.

Fluorescence imaging of the surgical field and ex vivo specimens. (A) Fluorescence imaging of the surgical field and ex vivo specimens captured using a handheld open surgery camera. (B) H&E staining and corresponding NIR scanning images of nerve root and muscle tissue sections

NIR fluorescence scanning imaging enables precise histological analysis of the spatial distribution of ICG. The results demonstrated a significant accumulation of ICG within the nerve roots, whereas minimal residual fluorescence was observed in muscle tissue, consistent with both in vivo and ex vivo imaging findings (Figure 3B). Moreover, the distribution of ICG within the nerve roots was heterogeneous, with certain regions exhibiting notably high fluorescence signals. This pattern suggests that ICG enrichment in neural tissue may be influenced by specific regulatory factors. These findings further prompt an in-depth investigation into the underlying mechanisms governing ICG accumulation in nerve roots, aiming to elucidate its biological basis and potential clinical applications.

Mechanistic Investigation of ICG Accumulation in Nerve Roots

To further investigate the histological distribution and underlying mechanisms of ICG within the nerve root, this study combined NIR fluorescence scanning imaging and histopathological analysis to examine ICG accumulation. NIR scanning identified high-fluorescence signal regions corresponding to areas of ICG enrichment, which were mapped onto H&E-stained images (Figure 4A–C). Based on the degree of ICG accumulation and structural differences within the nerve root, three regions were defined: Region a, an area of high ICG accumulation within the nerve root; Region b, an area within the nerve root with relatively low ICG accumulation; and Region c, an area outside the nerve root. High-magnification H&E staining revealed that Region a consisted of wavy fibrous structures, Region b comprised clusters of large, round cell bodies, and Region c exhibited bundled fibrous structures (Figure 4D).

Figure 4.

Histological analysis of ICG distribution. (A) The regions of high ICG accumulation in the NIR scanning images of nerve root sections are outlined with white dashed boxes and correlated with the corresponding H&E staining images. (B) NIR scanning images of nerve roots. (C) H&E staining images of nerve roots. (D) Magnified images of representative regions a, b, and c marked by black dashed boxes in Figure 4A. (E) NF200 immunofluorescence staining images of the same nerve root tissue sections, with nuclei stained by DAPI. (F) Magnified images of representative regions a, b, and c

NF200 immunofluorescence staining further confirmed that Region a represents the nerve fiber region within the nerve root, Region b corresponds to the aggregation area of pseudounipolar neurons within the nerve root, and Region c consists of nerve bundle fibers outside the nerve root (Figure 4E, and F). The analysis of ICG distribution indicated that its accumulation is independent of pseudounipolar neurons. Furthermore, although both Region a and Region c are composed of nerve fibers, ICG accumulation was significantly reduced in the nerve fiber region outside the nerve root (Region c). This phenomenon suggests that ICG enrichment in neural tissues is influenced not only by structural composition but also by specific microenvironmental factors.

To explore the potential factor contributing to ICG accumulation, CD31 immunohistochemical staining was performed to assess the microvascular distribution inside and outside the nerve root. The results revealed a significantly higher microvascular density in the nerve fiber region within the nerve root (Region a), forming an extensive vascular network, whereas the microvascular density was markedly lower in the nerve fiber region outside the nerve root (Region c) (Figure 5A). Quantitative analysis demonstrated that the blood vessel density (count per high-power field, HPF) in the nerve root (60.0 ± 5.9) was substantially higher than that outside the nerve root (8.1 ± 2.1) (Figure 5B).

Figure 5.

Histological analysis of the mechanism of ICG accumulation. (A) CD31 immunohistochemical staining to evaluate the distribution of blood vessels inside and outside the nerve root. (B) Quantitative analysis of blood vessel counts inside and outside the nerve root. (C) Schematic diagram of the possible mechanism of ICG accumulation in nerve roots. HPF = high-power field. Data are presented as mean ± standard deviation (n = 9). *P < 0.05; **P < 0.01; ***P < 0.001

In summary, ICG accumulation predominantly occurs in the nerve fiber region within the nerve root and is positively correlated with local microvascular density. These findings suggest that ICG enrichment may be mediated by the enhanced permeability and retention (EPR) effect, facilitating its accumulation in the highly vascularized nerve root (Figure 5C). This discovery not only advances the understanding of the specific distribution pattern of ICG in neural tissues but also provides new insights into its potential applications in neural imaging and surgical navigation.

Part B: Clinical Trial

Patient Characteristics

From April 2023 to September 2024, a total of 40 subjects were enrolled in this study, including 17 females, with a mean age of 54.4 ± 15.2 years. All participants successfully completed a 30-day follow-up assessment. Baseline comparisons among the study groups revealed no statistically significant differences in age, sex, body weight, preoperative visual analog scale (VAS) scores for back or leg pain, preoperative Oswestry Disability Index (ODI) scores, or surgical complexity across the different dosage groups (P > 0.05). Detailed baseline clinical characteristics of the study population are presented in Table 1.

Table 1.

Baseline Characteristics of the Included Patients

Characteristic	Control (n = 10)	0.5 mg/kg (n = 10)	1 mg/kg (n = 10)	2 mg/kg (n = 10)	P value
Age (years)	52.0 ± 15.6	56.3 ± 13.0	57.2 ± 13.4	52.0 ± 19.6	0.816
Male/female	4/6	6/4	8/2	5/5	0.475
Weight (kg)	64.0 ± 7.7	64.6 ± 7.2	61.0 ± 7.0	63.5 ± 7.4	0.706
VAS scores for back pain	4.4 ± 1.3	4.1 ± 1.7	4.1 ± 1.5	4.2 ± 1.2	0.959
VAS scores for leg pain	7.5 ± 1.1	7.6 ± 1.2	7.5 ± 1.4	7.6 ± 1.0	0.995
ODI scores	56.6 ± 9.5	58.7 ± 12.3	57.6 ± 11.1	59.0 ± 6.7	0.949
Operation level, n(%)	-	-	-	-	0.964
L3-4	1 (10%)	0 (0%)	1 (10%)	1(10%)	NA
L4-5	7 (70%)	7 (70%)	6 (60%)	6 (60%)	NA
L5-S1	2 (20%)	3 (30%)	3 (30%)	3 (30%)	NA

Intraoperative Fluorescence Imaging

The clinical application of intraoperative fluorescence imaging was preliminarily evaluated in this study. The surgical workflow for fluorescence imaging is illustrated in Figure 6A. NIR fluorescence imaging demonstrated that nerve root fluorescence signals were clearly observed in all cases of the 1 mg/kg and 2 mg/kg groups, whereas no fluorescence was detected in the 0.5 mg/kg group (Figure 6B), suggesting that this dose may be insufficient to reach the effective nerve labeling threshold. Quantitative fluorescence analysis further confirmed a dose-dependent difference, with the nerve root SBR and MFI in the 2 mg/kg group significantly higher than those in the 1 mg/kg group (SBR: 4.9 ± 0.7 vs 2.7 ± 0.2; MFI: 62.4 ± 6.5 AU vs 31.8 ± 3.7 AU) (Figure 6C, and D). These findings indicate that a higher dose of ICG effectively enhances nerve root fluorescence signals, improving imaging contrast and facilitating clearer intraoperative nerve structure identification. Furthermore, in the 2 mg/kg group, no fluorescence signals were observed in the ligamentum flavum, dural sac, nucleus pulposus, or surrounding adipose tissue (Figure 6E), indicating that ICG preferentially accumulates in the nerve root region. This selective localization reduces fluorescence interference from surrounding non-target tissues, thereby enhancing imaging specificity and contrast.

Figure 6.

Randomized controlled clinical trial of ICG fluorescence imaging. (A) Schematic diagram of clinical trial grouping and key time points (Created with BioRender.com). (B) Intraoperative fluorescence imaging of nerve roots in patients receiving different ICG doses (0.5, 1, or 2 mg/kg). NR = nerve root. (C) MFI of nerve roots in different dose groups. (D) SBR of nerve roots in different dose groups. (E) Endoscopic fluorescence imaging of key anatomical structures in the 2 mg/kg group (LF = ligamentum flavum, DC = dura mater, VB = vertebral body, EF = epidural fat). (F) Statistical analysis of the time required from opening the ligamentum flavum to identifying the nerve root during surgery. Data are presented as mean ± standard deviation (n = 10). *P < 0.05; **P < 0.01; ***P < 0.001

Notably, intraoperative NIR imaging exhibited a dose-dependent effect on nerve root identification time across different dosage groups (Figure 6F). Specifically, the time from ligamentum flavum removal to clear nerve root visualization was similar between the control group and the 0.5 mg/kg group (approximately 15 minutes in both), suggesting that a low-dose ICG (0.5 mg/kg) did not significantly improve nerve root visualization. However, in the 1 mg/kg and 2 mg/kg groups, this time was reduced to approximately 12.5 minutes and 9.4 minutes, respectively, reflecting the potential of higher-dose ICG to enhance intraoperative nerve root contrast and improve surgical efficiency. Statistical analysis revealed no significant difference in nerve root identification time between the 1 mg/kg group and the control or 0.5 mg/kg groups, indicating that at this dose, fluorescence signal enhancement may not yet reach the threshold for optimized nerve root visualization. In contrast, the 2 mg/kg group showed a significantly shorter identification time compared to the control group (P < 0.05), suggesting that a higher dose of ICG markedly enhances intraoperative nerve root visibility and accelerates the surgical process.

Additionally, no ICG-related adverse drug reactions were observed in any dosage group, indicating a favorable safety profile of this fluorescent contrast agent within the studied dose range. Overall, the findings suggest that intraoperative NIR fluorescence imaging effectively enhances nerve root visualization, improves intraoperative identification efficiency, and maintains a good safety profile at appropriate doses, providing robust support for precise navigation in spinal surgery (Figure 7A).

Figure 7.

Clinical evaluation outcomes. (A) Schematic diagram of fluorescence-guided spinal endoscopic surgery. (B) Time-dependent changes in back pain VAS scores. (C) Time-dependent changes in leg pain VAS scores. (D) Time-dependent changes in ODI scores. Data are presented as mean ± standard deviation (n = 10)

Clinical Outcomes and Complications

Postoperative follow-up assessments indicated a significant reduction in VAS scores across all dosage groups on postoperative day 1, week 1, and month 1, along with varying degrees of improvement in ODI scores. Although all groups exhibited a trend toward pain relief and functional recovery, no statistically significant differences in VAS and ODI scores were observed between the groups (Figure 7B–D).

Regarding postoperative complications, one patient in the control group experienced lateral femoral sensory abnormalities, which resolved spontaneously within 12 days without resulting in long-term neurological impairment (Table S1). No nerve injury-related complications were observed in any of the experimental groups.

Discussion

Intraoperative nerve fluorescence imaging has demonstrated significant potential in reducing iatrogenic nerve injuries.²³ However, clinical evidence remains relatively limited, restricting its widespread application.^36-38 This study systematically evaluated the feasibility, safety, and preliminary efficacy of ICG fluorescence imaging for nerve root visualization, progressing from animal studies to a randomized clinical trial. The results indicate that, within an appropriate dosage and administration window, ICG significantly enhances fluorescence contrast between the nerve root and surrounding tissues, thereby improving identification accuracy, reducing identification time, and enhancing surgical efficiency. No severe adverse reactions were observed, providing important evidence for the application of ICG fluorescence imaging in intraoperative nerve identification and laying the groundwork for further optimization and expanded clinical use.

Preclinical Findings: Dose and Time-dependent Fluorescence

Animal model experiments established the time-dependent and dose-dependent characteristics of ICG fluorescence in nerve roots. In the 2.8 mg/kg and 5.5 mg/kg groups, maximum fluorescence intensity (MFI and SBR) was observed approximately 3 hours post-administration, followed by a gradual decline until complete disappearance, confirming the 3-hour mark as the optimal imaging window. Considering the intraoperative timing requirements for nerve root identification in UBE procedures, the clinical trial selected a preoperative administration time of 1.5 hours to ensure optimal fluorescence contrast during surgery. Additionally, the 1.4 mg/kg group exhibited no detectable fluorescence signals between 3 and 24 hours, while fluorescence completely disappeared after 24 hours in all groups receiving doses between 1.4 and 5.5 mg/kg, indicating that insufficient dosage or an excessively prolonged administration interval failed to achieve optimal imaging outcomes. Furthermore, fluorescence duration was positively correlated with dose: the 2.8 mg/kg group exhibited fluorescence for approximately 6 hours, while the 5.5 mg/kg group extended up to 12 hours. This suggests that increasing the dose can effectively prolong the imaging window, providing stable nerve visualization support in prolonged or complex surgical procedures. These findings clarify not only the optimal ICG administration timing but also the relationship between dose and fluorescence duration, offering critical insights for intraoperative nerve visualization strategies.

It is acknowledged that a fixed 1.5-hour interval may conflict with variable operative start times in typical clinical workflows. However, the preclinical data from this study, showing fluorescence duration up to 12 hours in the high-dose group, suggests a flexible “imaging window” rather than a precise “time-point”. While the 1.5-hour timing was standardized in this pilot study to enable a fair comparison across dosage groups, this flexibility is a key practical consideration for future application.

Clinical Validation and Dose Optimization

In spinal endoscopic surgery, precise differentiation of the nerve root from the dura mater, intervertebral disc, and surrounding adipose tissue after ligamentum flavum removal is crucial for minimizing iatrogenic injuries. This study confirmed that preoperative administration of 1 or 2 mg/kg ICG, combined with fluorescence imaging, effectively enhanced nerve root visualization. Clinical trial results demonstrated that the MFI and SBR values in the 2 mg/kg group were significantly higher than those in the 1 mg/kg group. In contrast, preclinical experiments showed no significant differences in fluorescence intensity between the 2.8 mg/kg and 5.5 mg/kg groups (corresponding to 1 mg/kg and 2 mg/kg in humans). This discrepancy may be attributed to specific interspecies differences in physiology and pharmacokinetics. Specifically, this likely involves variations in nerve root microvascular density and permeability, as well as distinct drug metabolism rates and hemodynamic profiles, which collectively influence ICG accumulation and signal intensity. While preclinical experiments provided an initial reference for dosage selection, clinical trials further optimized the dosage parameters, demonstrating that 2 mg/kg offers superior fluorescence contrast for nerve root imaging in spinal surgery, thereby improving anatomical localization and surgical safety.

Fluorescence Imaging Enhances Surgical Efficiency and Safety

By measuring the time required from ligamentum flavum removal to clear nerve root identification, this study systematically assessed the practical value of ICG fluorescence imaging in intraoperative nerve visualization. The results showed that the 2 mg/kg group exhibited a significantly shorter identification time compared to the control and 0.5 mg/kg groups, with a statistically significant difference. These findings indicate that ICG fluorescence imaging enhances the surgeon’s ability to recognize key anatomical structures and reduces intraoperative localization time. In terms of surgical outcomes, all experimental groups demonstrated significant postoperative relief of nerve root compression symptoms, as evidenced by improvements in VAS and ODI scores for back and leg pain. These outcomes were comparable to those of conventional surgery, suggesting that fluorescence imaging does not compromise surgical efficacy. Additionally, no severe ICG-related adverse reactions were observed, further confirming the safety of this technique. Overall, ICG fluorescence imaging enhances nerve root visualization and surgical efficiency while maintaining a favorable safety profile, making it particularly suitable for cases with challenging nerve root exposure or complex anatomical structures.

Mechanisms Underlying Nerve-specific ICG Retention

The precise mechanism underlying ICG fluorescence imaging in nerve visualization remains incompletely understood, and multiple hypotheses have been proposed. One study suggests that sympathetic nerve visualization is associated with ICG retention in connective tissue membranes,²⁸ while others hypothesize a relationship with axonal transport.^30,39 Histological analysis in this study revealed that ICG primarily accumulates in nerve fibers within the nerve root, with minimal fluorescence signal detected in pseudounipolar neurons or surrounding tissues. Immunohistochemical staining for CD31 indicated a relatively rich microvascular network within the nerve root, suggesting that vascular characteristics may play a crucial role in ICG retention. Similar findings have been reported in tumor fluorescence imaging, where ICG signal intensity correlates with vascular density, possibly influenced by the EPR effect.⁴⁰ Furthermore, while ICG accumulation in nerve roots appears to be a baseline physiological phenomenon (as observed in healthy preclinical models), the pathological state of LDH may amplify this process. It is plausible that confounding factors specific to the LDH microenvironment, such as local inflammation-induced increases in microvascular permeability (correlating with the ‘E' in EPR) and hemodynamic stasis resulting from mechanical compression (correlating with the ‘R' in EPR), act as modulators that enhance ICG accumulation in the clinical setting. Understanding these mechanisms not only deepens insights into ICG imaging principles but also provides a foundation for optimizing intraoperative nerve visualization strategies.

Clinical Implications and Future Perspectives

ICG near-infrared fluorescence imaging provides a real-time, non-invasive approach for precise intraoperative nerve root identification, enhancing both surgical precision and efficiency. For surgeons in the early stages of their learning curve, this technology reduces the need for repeated anatomical probing and increases intraoperative confidence. In complex procedures such as recurrent disc herniation, scoliosis correction, and spinal tumor resection, ICG imaging can facilitate precise nerve root localization, which may be particularly valuable in these challenging anatomical situations.

While ICG fluorescence imaging has been widely adopted in other surgical fields, its application in orthopedics, particularly spine surgery, remains in the exploratory phase. The findings of this study validate its clinical utility in minimally invasive spinal surgery and provide a foundation for broader clinical adoption. As further research and clinical experience accumulate, this technique may be incorporated into standard surgical protocols, ultimately contributing to expert consensus and clinical guidelines aimed at improving precision and safety in spinal procedures.

Study Limitations and Future Research Directions

Despite its promising findings, this study has several limitations. First, the clinical trial sample size was relatively small, as this was a pilot study, and the follow-up period was limited to 30 days. These factors make it insufficient to comprehensively assess the impact of this technique on reducing nerve injury incidence, medium-to long-term clinical outcomes, surgical recurrence rates and the detection of rare adverse events. Larger, multicenter studies with extended follow-up are needed to comprehensively evaluate the long-term safety and efficacy of ICG fluorescence imaging. Second, this study focused primarily on patients undergoing UBE surgery for LDH. Its effectiveness in other conditions such as spinal stenosis, spondylolisthesis, or more complex pathologies remains uncertain and requires further investigation. Additionally, the widespread clinical adoption of near-infrared fluorescence imaging is challenged by the capital cost of hardware requirements and the need for specialized training, which may limit its accessibility across different healthcare institutions. It should be noted, however, that the per-procedure cost is low, given that the ICG dye itself is inexpensive and widely available. Implementation may also be streamlined in centers already familiar with NIR imaging technology from other surgical specialties.

Future research should explore the integration of near-infrared fluorescence imaging with intraoperative navigation systems and artificial intelligence-based image recognition to enhance precision and automation in nerve identification. Moreover, the development of novel fluorescent probes with smaller molecular weights or neural-specific targeting capabilities could further reduce background interference and improve imaging sensitivity and specificity, expanding the potential applications of this technique in neurosurgical visualization.

Conclusion

ICG fluorescence imaging is a promising technique for nerve root identification, improving surgical efficiency. The optimal dosing regimen is 2 mg/kg administered 1.5 hours before surgery. The imaging mechanism is likely related to the EPR effect due to the dense vasculature in nerve roots. This method provides a foundation for the development of real-time intraoperative imaging technologies.

Supplemental Material

Supplemental Material - Near-Infrared Fluorescence Imaging With Indocyanine Green for Intraoperative Nerve Root Visualization in Spinal Surgery: From Preclinical Studies to a Pilot Randomized Controlled Trial

Supplemental Material for Near-Infrared Fluorescence Imaging With Indocyanine Green for Intraoperative Nerve Root Visualization in Spinal Surgery: From Preclinical Studies to a Pilot Randomized Controlled Trial by Huayi Huang, Zhenyi Chen, Shenjia Wu, Linlong Wang, Xiaobin Zhu, Lingfei Xiao, Zhouming Deng, Renxiong Wei, Meijia Gu, Lin Cai, Jun Lei, Yuanlong Xie in Global Spine Journal.

Footnotes

Acknowledgments

The authors acknowledge Guangdong OptoMedic Technologies, Inc. for providing NIR imaging equipment.

ORCID iDs

Huayi Huang

Yuanlong Xie

Ethical Considerations

This prospective, randomized controlled, open-label clinical trial was approved by the Medical Ethics Committee of Zhongnan Hospital of Wuhan University (2023008). The Experimental Animal Welfare Ethics Committee of Zhongnan Hospital of Wuhan University approved the animal research in this study (ZN2023173).

Consent to Participate

Written informed consent was obtained from all participants.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Research Fund from Medical Sci-Tech Innovation Platform of Zhongnan Hospital, Wuhan University (PTXM2023037), and Key R&D plan of Hubei Province (2022BCA032).

Declaration of Conflicting Interests

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

Data Availability Statement

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.*

Clinical Trial Registration Information

Registered at ClinicalTrials.gov (NCT05808140).

Supplemental Material

Supplemental material for this article is available online.

References

Antoniadis

Kretschmer

Pedro

König

Heinen

CPG

Richter

. Iatrogenic nerve injuries prevalence, diagnosis and treatment. Dtsch Arztebl Int. 2014;111(16):273-279. doi:10.3238/arztebl.2014.0273

Burke

Shorten

. When pain after surgery doesn’t go away. Biochem Soc Trans. 2009;37(1):318-322. doi:10.1042/Bst0370318

Attia

Mahmoud

d’Hooghe

Bariteau

Labib

Myerson

. Outcomes and complications of open versus minimally invasive repair of acute achilles tendon ruptures: a systematic review and meta-analysis of randomized controlled trials. Am J Sports Med. 2023;51(3):825-836. doi:10.1177/03635465211053619

Pan

, et al. Percutaneous endoscopic lumbar discectomy: Indications and complications. Pain Physician. 2020;23(1):49-56.

Lee

. Complications and management of endoscopic spinal surgery. Neurospine. 2023;20(1):56-77. doi:10.14245/ns.2346226.113

Lee

Rhee

Chang

Kim

. Deep learning in spinal endoscopy: U-net models for neural tissue detection. Bioengineering-Basel. 2024;11(11):1082. doi:10.3390/bioengineering11111082

Ahn

Kim

Lee

. Comparison of outcomes of percutaneous endoscopic lumbar discectomy and open lumbar microdiscectomy for young adults: a retrospective matched cohort study. World Neurosurg. 2016;86:250-258. doi:10.1016/J.Wneu.2015.09.047

Hsu

H-T

Chang

S-J

Yang

Chai

. Learning curve of full-endoscopic lumbar discectomy. Eur Spine J. 2012;22(4):727-733. doi:10.1007/s00586-012-2540-4

Ruetten

Komp

Merk

Godolias

. Surgical treatment for lumbar lateral recess stenosis with the full-endoscopic interlaminar approach versus conventional microsurgical technique: a prospective, randomized, controlled study. J Neurosurg Spine. 2009;10(5):476-485. doi:10.3171/2008.7.17634

10.

Hoogland

van den Brekel-Dijkstra

Schubert

Miklitz

. Endoscopic transforaminal discectomy for recurrent lumbar disc herniation - a prospective, cohort evaluation of 262 consecutive cases. Spine. 2008;33(9):973-978. doi:10.1097/BRS.0b013e31816c8ade

11.

Kim

Lee

Jung

, et al. Targeted percutaneous transforaminal endoscopic diskectomy in 295 patients: comparison with results of microscopic diskectomy. Surg Neurol. 2007;68(6):623-631. doi:10.1016/j.surneu.2006.12.051

12.

Lopez

Zlatev

Mach

, et al. Intraoperative optical biopsy during robotic assisted radical prostatectomy using confocal endomicroscopy. J Urol. 2016;195(4):1110-1117. doi:10.1016/j.juro.2015.10.182

13.

Rais-Bahrami

Levinson

Fried

, et al. Optical coherence tomography of cavernous nerves: a step toward real-time intraoperative imaging during nerve-sparing radical prostatectomy. Urology. 2008;72(1):198-204. doi:10.1016/j.urology.2007.11.084

14.

Ukimura

Magi-Galluzzi

Gill

. Real-time transrectal ultrasound guidance during laparoscopic radical prostatectomy: impact on surgical margins. J Urol. 2006;175(4):1304-1310. doi:10.1016/S0022-5347(05)00688-9

15.

Charalampidis

Jiang

Wilson

JRF

Badhiwala

Brodke

Fehlings

. The Use of intraoperative neurophysiological monitoring in spine surgery. Glob Spine J. 2020;10(1_suppl):104s-114s. doi:10.1177/2192568219859314

16.

Daniel

Botelho

Milano

, et al. Intraoperative neurophysiological monitoring in spine surgery. Spine. 2018;43(16):1154-1160. doi:10.1097/brs.0000000000002575

17.

Peng

Chen

C-M

, et al. Patent blue versus methylene blue and indigo carmine as a better dye for chromodiscography: in vitro staining efficacy and cytotoxicity study using bovine coccygeal intervertebral discs. Spine J. 2023;23(7):1079-1087. doi:10.1016/j.spinee.2023.02.008

18.

Zhang

Liu

Huang

, et al. Toxicity effects of methylene blue on rat intervertebral disc annulus fibrosus cells. Pain Physician. 2019;22(2):155-164.

19.

Zeng

Wang

, et al. Intraoperative resection guidance and rapid pathological diagnosis of osteosarcoma using B7H3 targeted probe under NIR-II fluorescence imaging. Adv Sci. 2024;11(33):e2310167. doi:10.1002/advs.202310167

20.

Regmi

Wang

, et al. Intraoperative fluorescence redefining neurosurgical precision. Int J Surg. 2025;111(1):998-1013. doi:10.1097/Js9.0000000000001847

21.

Huang

Wei

, et al. Near‐infrared (NIR) imaging with indocyanine green (ICG) may assist in intraoperative decision making and improving surgical margin in bone and soft tissue tumor surgery. J Surg Oncol. 2023;128(4):612-627. doi:10.1002/jso.27306

22.

Wang

, et al. Indocyanine green fluorescence imaging may detect tumour residuals during surgery for bone and soft-tissue tumours. Bone & Joint Journal. 2023;105b(5):551-558. doi:10.1302/0301-620x.105b5

23.

Walsh

Cole

Tipirneni

, et al. Fluorescence imaging of nerves during surgery. Ann Surg. 2019;270(1):69-76. doi:10.1097/Sla.0000000000003130

24.

Liu

, et al. Tumor lesion detection in patients with cervical cancer by indocyanine green near-infrared imaging. Eur J Nucl Med Mol Imag. 2022;50(4):1252-1261. doi:10.1007/s00259-022-06030-1

25.

Chen

Xie

Zhong

, et al. Safety and efficacy of indocyanine green tracer-guided lymph node dissection during laparoscopic radical gastrectomy in patients with gastric cancer A randomized clinical trial. JAMA Surg. 2020;155(4):300-311. doi:10.1001/jamasurg.2019.6033

26.

Gerken

ALH

Nowak

Meyer

, et al. Quantitative assessment of intraoperative laser fluorescence angiography with indocyanine green predicts early graft function after kidney transplantation. Ann Surg. 2022;276(2):391-397. doi:10.1097/Sla.0000000000004529

27.

Pei

Liu

, et al. The safety and feasibility of intraoperative near-infrared fluorescence imaging with indocyanine green in thoracoscopic sympathectomy for primary palmar hyperhidrosis. Thorac Cancer. 2020;11(4):943-949. doi:10.1111/1759-7714.13345

28.

Zhou

Yang

, et al. Near-infrared intraoperative imaging of thoracic sympathetic nerves: from preclinical study to clinical trial. Theranostics. 2018;8(2):304-313. doi:10.7150/thno.22369

29.

Chen

Wang

, et al. Fluorescence-assisted visualization of facial nerve during mastoidectomy: a novel technique for preventing iatrogenic facial paralysis. Auris Nasus Larynx. 2015;42(2):113-118. doi:10.1016/j.anl.2014.08.008

30.

Zhang

, et al. Intraoperative near-infrared fluorescence imaging can identify pelvic nerves in patients with cervical cancer in real time during radical hysterectomy. Eur J Nucl Med Mol Imag. 2022;49(8):2929-2937. doi:10.1007/s00259-022-05686-z

31.

Kanno

Aiko

Yanai

Sawada

Sakate

Andou

. Clinical use of indocyanine green during nerve-sparing surgery for deep endometriosis. Fertil Steril. 2021;116(1):269-271. doi:10.1016/j.fertnstert.2021.03.014

32.

Kilkenny

Browne

Cuthill

Emerson

Altman

. Improving Bioscience research reporting: the ARRIVE guidelines for reporting animal research. PLoS Biol. 2010;8(6):e1000412. doi:10.1371/journal.pbio.1000412

33.

Wei

. Experimental Methodology of Pharmacology. 4th ed. People's Medical Publishing House; 2010:70-72.

34.

Hertzog

. Considerations in determining sample size for pilot studies. Res Nurs Health. 2008;31(2):180-191. doi:10.1002/nur.20247

35.

Predina

Keating

Newton

, et al. A clinical trial of intraoperative near‐infrared imaging to assess tumor extent and identify residual disease during anterior mediastinal tumor resection. Cancer. 2018;125(5):807-817. doi:10.1002/cncr.31851

36.

Montaño

Masillati

Szafran

, et al. Matrix-designed bright near-infrared fluorophores for precision peripheral nerve imaging. Biomaterials. 2025;319:123190. doi:10.1016/j.biomaterials.2025.123190

37.

Hernández-Gil

Chow

Chatras

, et al. Development and validation of nerve-targeted bacteriochlorin sensors. J Am Chem Soc. 2023;145(26):14276-14287. doi:10.1021/jacs.3c02520

38.

Barth

Rizvi

SZH

Masillati

, et al. Nerve-Sparing gynecologic surgery enabled by A near-infrared nerve-specific fluorophore using existing clinical fluorescence imaging systems. Small. 2024;20(41):e2300011. doi:10.1002/smll.202300011

39.

Yang

Wang

, et al. Real time monitoring peripheral nerve function with ICG and BDA-ICG by NIR-II fluorescence imaging. Mater Today Bio. 2024;26:101084. doi:10.1016/j.mtbio.2024.101084

40.

Madajewski

Judy

Mouchli

, et al. Intraoperative near-infrared imaging of surgical wounds after tumor resections can detect residual disease. Clin Cancer Res. 2012;18(20):5741-5751. doi:10.1158/1078-0432.Ccr-12-1188

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.20 MB