Introduction
Meningiomas are the most frequently occurring intracranial tumors in adults, accounting for one third of cases [
1]. They are classified into three separate World Health Organization (WHO) grades based on histology, mitotic index, and the presence of brain invasion [
2]. Although meningiomas are mostly benign and slow growing, they compress surrounding brain tissue and nerves. Depending on location and time course, symptoms such as seizures, focal neurological deficits (e.g., extremity weakness or visual changes), or (severe) mental status changes can arise. Treatment is most effective when complete surgical radical resection is achieved aiming to remove the complete tumor while preserving neurological function. Incomplete resection is a major risk factor for recurrence [
2]. Currently, differentiation between meningioma tumor invasion in the dura mater or adjacent bone and reactive tissue is limited. Chances of safe gross total resection without compromising neurological outcome may decrease due to location (such as the cerebellopontine angle), invasion of dura and/or bone, and involvement of neurovascular structures. Fluorescence-guided meningioma surgery may be a helpful tool to increase the extent of resection, especially in patients with complex meningiomas.
Introduced in 1948 by Moore et al, fluorescein was applied to visualize brain tumors [
3]. Later, 5-aminolevulinic acid (5-ALA) has been implemented in the field of neurosurgery, mainly in glioblastoma surgery [
4]. Indocyanine green (ICG) has been utilized for video angiography mainly for neurovascular neurosurgery (aneurysms, arteriovenous malformation, and fistulas) [
5]. More recently, these dyes have been applied in meningioma resections.
This review aims to give a descriptive overview of the most frequently applied intra-operative dyes (5-ALA, ICG, and fluorescein) in meningioma surgery and evaluate their role in terms of efficacy, additional extent of resection, and complication rate.
Discussion
The main risk factor for meningioma recurrence is an incomplete resection. In order to improve extent of resection, the ability to adequately identify tumor to normal tissue borders should be enhanced. This may be possible through the application of intraoperative fluorescence-guided surgery. In this review, we evaluated the literature and summarized the intra-operative use of 5-ALA, ICG, and fluorescein aiming to determine the benefit of fluorescent guidance during meningioma surgery in striving for a radical resection.
Current reports have major limitations. In this review, the majority of included studies are case reports or series with low level of evidence. Only a few report large(r) patient cohorts. To date, most studies are only anecdotal without an adequate study design: they cannot provide strong evidence and merely reflect the surgeon’s observations, e.g., extent of resection or additional safe resection. Only few reports are based on (partially) standardized protocols, for example, noting fluorescence status of tumor, bone, dura mater, and adjacent healthy tissue with at least documentation of neurological deficits and post-operative MRIs determining rate of resection. Indeed, few of the included studies followed a structured sampling approach. Thus, these outcome measures are often biased by the investigator(s) and not readily comparable between patients, even within the same study. Moreover, most studies are retrospective and no randomized controlled trials have been reported. Most reports lack reliable patient data, such as basic patient characteristics, histopathological data, and (long term) follow-up. Consequently, currently, the level of evidence for the use of fluorescence is low due to selection bias.
Reported follow-up time was mostly very limited (“post-operative days,” “post-operatively”) without long-term outcome. Short-term outcomes were mostly positive: transient or no neurological deficit and (in only a few reports) complete resection on post-operative MRI. Long-term confirmation of radical resection without neurological deficits could not be determined, e.g., in terms of progression-free survival or recurrence rates. These outcome variables are essential for the evaluation of fluorescent techniques.
Furthermore, most studies used a dichotomous scoring system to assess quality of fluorescence signal. Although this system is easily reproducible, it is not applicable for all clinical and/or experimental issues [
58] and quantification of fluorescence intensity is flawed. A three- or five-level grading score would be more appropriate for the objective recording of fluorescence quality. Due to clinical applicability, the three-level system would be the first choice [
58], e.g., non-fluorescent (1), minimal/moderate fluorescent (2), and highly fluorescent (3). Additionally, quantifying fluorescence using an intraoperative spectrometer enables the comparison of fluorescence intensity in a more standardized fashion by measuring absolute fluorescence levels. Although some studies applied in vivo spectrometry with quantitative analysis using an intraoperative probe, the majority only performed a qualitative analysis.
Numerous drawbacks of presently applied dyes became apparent in this review. Currently, no compelling evidence is available regarding the benefit of fluorescence-guided meningioma surgery. It might aid surgical resection in select cases through detection of meningioma remnants or invasion or by aiding in safe vascular sacrifice. Additionally, the investigated fluorophores severely lack sensitivity and specificity to distinguish meningioma (infiltration) from healthy brain tissue. Multiple studies applying 5-ALA show heterogenic meningioma fluorescent patterns [
12,
33] and emphasize the role of homogenic fluorescence in complete surgical resection. However, fluorescence homogeneity was not discussed in most reports investigating any of the three investigated fluorophores. Tumor margin determination was unreliable, reducing the value of using these fluorophores. Moreover, histological analysis of individual samples was almost never or only partially reported, which diminishes the ability to correlate fluorescence (intensity) with presence of meningioma cells and severely limits the possibility of calculating sensitivity and specificity rates.
In tumor imaging, 5-ALA has high sensitivity and specificity for meningioma tissue. However, fluorescence heterogeneity and the absence of correlation between fluorescence and mitotic index remain relevant issues. Sensitivity is low(er) in dural and bone invasion compared to the fluorescence results obtained in the tumor itself. Recently, PpIX formation in a non-fluorescent-invaded dural tail was confirmed using mass spectrometry. Although emphasizing the need for a more accurate technique in fluorescence-guided meningioma surgery, this technique is highly inefficient in clinical practice. Only one study described imaging of the meningioma tumor using ICG [
54]. The imaging protocol should be improved to increase sensitivity rates and tumor-to-background ratio. Other reports apply video angiography for transdural, arterial, and/or venous imaging, claiming the benefit for half of the patients, determined by surgeons personal (subjective) judgment without standardized techniques. The benefit of transdural ICG VA remains highly questionable, as established techniques such as intraoperative neuronavigation already offer accurate transdural visualization of the meningioma. In total, 75% of all investigated cases benefitted from transdural imaging following ICG VA: in the authors opinions, it increased the feeling of safety during dural opening and assessing vessel patency. Data regarding objective long-term follow-up of these “positive” cases is lacking. Authors claim arterial and venous imaging could address the uncertainty of patency, anastomotic flow, and the distinction between tumor draining veins and collateral ones. Its benefit has been described in the minority of patients, without objectifying criteria to determine superiority of this application above surgery without this tool. Fluorescein fluorescence has been performed with filter (filtered surgical microscopy) and without (standard white light microscopy). Despite the convenience and affordability of white light fluorescein surgical microscopy, contrast between tumor and surrounding tissue is markedly less evident compared to filtered surgical microscopy. Additionally, one group [
54] reported that the intraoperative fluorescein-induced meningioma fluorescence did not significantly alter the surgical strategy, emphasizing the questionability of this fluorophore in meningioma surgery.
Differentiation between meningioma invasion and reactive tissue is limited at present. Pre-operatively, changes of dura mater and bone can be observed as a “dural tail sign” or bone hypertrophy. The dural tail sign is often observed on MRI scans as dural thickening adjacent to the meningioma, and is either caused by meningioma invasion or by vascular congestion, both accounting for approximately 50% of cases [
59]. 5-ALA fluorescence was observed in invaded and normal dura mater, reducing sensitivity (50%). Reports describing ICG fluorescence did not determine fluorescence status of adjacent dura mater or histologically investigated dural invasion [
35]. Lastly, the dura mater always fluoresces after fluorescein administration, thus making the ability to differentiate between healthy dura mater and meningioma infiltration highly questionable. This emphasizes the need for highly sensitive and specific fluorescent dyes in meningioma surgery.
Similarly, distinction between (reactive) hyperostosis and tumor invasion cannot always be readily made. Bone invasion has been reported in 20–68% meningioma patients, primarily in the presence of hyperostosis. Regardless, invasion has been described in non-hyperostotic bone as well, in 10–40% of cases [
60,
61]. Indeed, all fluorescent (
n = 21) and none of the non-fluorescent (
n = 18) bone presented meningioma invasion in one study [
28] applying 5-ALA. The only other dye showing bone fluorescence was fluorescein; however, histological results were not provided [
54]. ICG bone fluorescence was not observed. In addition to location, local invasion of bone and soft tissues makes it a technical challenge to determine the tumor margin using available surgical tools. Fluorescence-guided meningioma surgery may be a helpful tool to increase the rate of resection, especially in patients with complex meningiomas.
In summary, we reviewed the current experience of 5-ALA, ICG, and fluorescein in meningioma surgery. In our opinion, fluorescence-guided meningioma surgery should be a reliable, highly specific, and sensitive technique. Despite numerous studies reporting the use of fluorescent dyes, currently there is no evidence that these tools improve the radical resection rate and long-term recurrence-free outcome in meningioma surgery and no evidence exists that safety of surgery increased due to application of these fluorescence techniques.
Future perspectives
Failure of currently available fluorescence techniques clearly show the need for a highly meningioma-specific probe clearly distinguishing the tumor margin from healthy tissue has become apparent after analyzing current data in this review. We propose the development and intraoperative use of targeted tracers, which could be achieved by binding a meningioma-specific peptide to a fluorescent dye, preferably near-infrared due to increased penetration depth. First, this should be confirmed in in vitro cultures and subsequently in in vivo studies with mice and human subjects. As meningiomas are usually of dural origin, passage of the tracer through the blood-brain-barrier should not be an issue. The use of molecular fluorescent-guided surgery (MFGS) has been successful for various tumor types, e.g., in ovarian carcinoma [
62] and peritoneal metastases of colorectal carcinomas [
63] targeting αvβ3-integrin or folate receptor α and vascular endothelial growth factor α, respectively. The use of MFGS seems feasible for the optimization of meningioma surgery, which is currently a high priority in our ongoing research.
Next to this, improving and validating existing fluorescence-guided techniques could be an alternative. Current evidence is mostly based on case reports and case series and therefore an organized trial could be another next step. The organization of such a trial could improve the quality of evidence tremendously and is a necessary step in the further development of this MFGS technique. Outcome measures to be considered are among others extent of resection, fluorescence status and homogeneity, and correlation between fluorescence and WHO grade, along with long-term follow-up.