The angular relationship between the foramen ovale and the trigeminal impression: percutaneous cannulation trajectories for trigeminal neuralgia

Abstract

The debilitating pain of trigeminal neuralgia (TN) often necessitates neurosurgical intervention via percutaneous transovale cannulation. While most percutaneous treatments of TN are successful, severe adverse events resulting from failure to properly cannulate the foramen ovale (FO) have been reported. With regard to specific targeting of particular trigeminal divisions (i.e. V₁, V₂, V₃, and combinations thereof), operative techniques have been described; however, these descriptions have not included specific angulation data. This anatomical study analyzed the angular relationship between the centroid and anteromedial- and posterolateral-most aspects of the FO and the boundaries of the trigeminal impression. The study is the first to detail the angular relationship between the FO boundaries and the boundaries of the trigeminal impression in dry human skulls relative to the coronal plane. The information may be used to prevent miscannulation and also target specific branches of the trigeminal nerve for optimal operative results.

Introduction

Trigeminal neuralgia (TN) is the most common form of craniofacial neuralgia and is typically marked by debilitating pain that necessitates treatment.¹ Trigeminal neuralgia has a prevalence of approximately 0.1–0.2 per 1,000 individuals and an incidence ranging from 2 to 25.6 in 100,000 per year.^2–5 With regard to sex, the female-to-male ratio is approximately 3:2.^5–7 With regard to localization, TN tends to affect the right side more often than the left side and V₂ and V₃ more often than V₁ nerve distributions.⁷

Many methods have been utilized for the treatment of TN including microvascular decompression, stereotactic radiosurgery, and percutaneous procedures including radiofrequency rhizotomy, glycerol rhizotomy, and balloon compression.^1,8–10 Percutaneous procedures typically involve cannulation of the foramen ovale (FO) via the transjugal-transovale route of Härtel.¹¹ The FO is relatively small and anatomical variations including foramen stenosis, or ossified pterygospinous or pterygoalar ligaments may complicate the cannulation of the already-small target.^12,13 Likewise, the shape of the FO may contribute to operative difficulties.¹⁴ The diverse shapes of foramina ovalae have been described as “banana-like,” “triangular,” “oval,” “truly oval,” “elongated oval,” “elongated,” “semicircular,” “almond,” “round,” “rounded,” “slit,” “irregular,” “D shape,” and “pear”.^15–24 The diverse morphology of the FO may partly explain why failure to cannulate the FO has been reported to occur in as many as 8% of procedures.^25,26 Even CT paired with navigation technology has proved unsuccessful in cannulating the FO in 5.17% (9:174) of patients due to suspected variation in FO morphology.¹⁴ Moreover, multiple attempts to cannulate the FO may increase the risk of adverse complications.¹⁴

While most percutaneous treatments of TN are successful, a variety of adverse effects have been reported.²⁷ Most of the documented adverse effects are related to the inherent nature of the procedure itself (e.g. meningitis); however, many adverse effects are a direct result of failure to properly cannulate the FO.^27,28 For example, mistakenly cannulating a nearby sphenoidal emissary foramen (SEF) may puncture the cavernous sinus which one report described in eight cases, one of which resulted in a temporal lobe hematoma.²⁹ Another report documented mistaken cannulation of the SEF in seven of 200 (3.5%) procedures.³⁰ Likewise, inadvertent puncture of the foramen lacerum and carotid artery, inferior orbital fissure, superior orbital fissure, and jugular foramen have been reported.^27,31–33 Unsuccessful attempts to penetrate the FO have led to adverse effects including blindness, brainstem hematoma, temporal hematoma, carotid artery hemorrhage, and death.^{27,29–31,33,34}

In addition to the prevention of severe adverse effects as a result of improper cannulation, understanding the angles at which surgical tools are introduced to the FO may optimize operative outcomes. Despite many reports regarding the treatment of TN via percutaneous procedures, only a few have documented angles by which surgical tools are introduced to the FO.^35–39 Most of the studies that have reported the approach angle of the surgical tool have reported angles relative to the axial plane or Reid line, but not the coronal or sagittal planes.^35–39 Yao et al.³⁶ documented the angle between the center of the FO and the trigeminal impression (TI) relative to the coronal / sagittal planes in a population of 120 individuals (presumably Chinese) via CT scan. However, aside from the study by Yao et al.,³⁶ there is little information regarding the angular relationship of the FO and TI in relation to the coronal and sagittal planes. Due to the clinical importance of FO cannulation angles and the paucity of information regarding the angular relationship between the FO and the TI, this study assesses the angular relationship between the boundaries of the FO and the TI relative to the coronal and sagittal planes in dry human skulls.

Materials and Methods

The study was performed on a collection of dry human crania from Bethany College, California University of Pennsylvania, Franciscan University of Steubenville, John Marshall High School, Ohio University-Eastern, Washington & Jefferson College, West Liberty University, and West Virginia University School of Medicine. The sample included 139 dry human skulls without demographic data. Exclusion criteria included the presence of foramina ovalae that were confluent with nearby foramina (e.g. SEF, foramen spinosum, foramen lacerum). Likewise, trigeminal impressions that had unclear boundaries were excluded from the study. From the 139 crania, a total of 223 pairs of FO and TI were able to be assessed (113 left-sided FO-TI pairs and 110 right-sided FO-TI pairs).

The crania were photographed with a digital camera (Canon PowerShot SX50 HS, 12.1 Megapixel). The photographs were examined and measured using ImageJ software.^40,41 A line was drawn through the basilar process of the occiput to identify the midsagittal plane in order to orient the digital image appropriately for measurement with ImageJ software.^42,43

Angles were identified between lines connecting landmarks from the FO to the TI and the coronal plane. These angles were measured on anterolateral aspect of the intersection with the coronal plane. The lines were drawn through the following points at the FO and TI: 1.) the anteromedial-most FO and the anteromedial-most TI; 2.) the anteromedial-most FO and the posterolateral-most TI; 3) the posterolateral-most FO and the anteromedial-most TI; 4.) the posterolateral-most FO and the posterolateral-most TI; 5.) the centroid of the FO and the anteromedial-most TI; and 6.) the centroid of the FO and the posterolateral-most TI (Figure 1).

Figure 1.

Illustrated view of the trigeminal nerve and salient regional anatomy. A.) Norma basalis interna with left foramen and trigeminal impression emphasized with a black rectangle. B.) Magnified view of the left foramen ovale and trigeminal impression from figure 1A. (FO: foramen ovale; TI: trigeminal impression). C.) Trigeminal nerve, ganglion, and divisions superimposed in situ with overlying dura. The trigeminal nerve is shown passing through the porus trigeminus. (PT: porus trigeminus encompassing the trigeminal nerve). D.) Photograph of a left-sided foramen ovale and trigeminal impression with superimposed lines illustrating the angles measured in this study. Lines are drawn between the anteromedial- and posterolateral-most aspects of the foramen and the impression. Likewise, lines are drawn between the centroid of the foramen and the anteromedial- and posterolateral-most impression. The black lines represent the coronal plane. Angles in this study were those formed between these landmarks, measured from the anterolateral side (as opposed to the anteromedial-side angles which can be calculated by subtracting the angles reported in this report from 180 degrees). (AMF: anteromedial foramen ovale; C: centroid of the foramen ovale; PLF: posterolateral foramen ovale; AMI: anteromedial trigeminal impression; PLI: posterolateral trigeminal impression). E.) Sensory components of the trigeminal nerve with a surgical instrument introduced at a 65° angle to the coronal plane toward the interface of V₂ and V₃. (V₁: ophthalmic component of the trigeminal nerve; V₂: maxillary component of the trigeminal nerve; V₃: mandibular component of the trigeminal nerve).

The recorded data were then analyzed using the GraphPad Prism statistical software, version 6.00 (GraphPad Software, La Jolla, CA, USA). Inferential statistical analysis included paired t-tests to assess symmetry between the left- and right-sided angles.

Results

Normative data regarding the angular relationship between the FO and TI can be found in Table 1. The average angles from the anteromedial-most aspect of the FO to the anteromedial- and posterolateral-most aspects of the TI were 65° ± 10.2° and 95° ± 8.5°, respectively (Mean ± SD). From the posterolateral aspect of the FO, the angles at the anteromedial- and posterolateral-most aspects of the TI were 69° ± 10.3° and 32° ± 11.2°, respectively. The average angles from the centroid of the FO to the anteromedial- and posterolateral-most aspects of the TI were 47° ± 11.2° and 83° ± 9.4°, respectively.

Table 1.

Angular relationship data of the lines formed between the anteromedial-most foramen ovale, the centroid of the foramen ovale, and the posterolateral-most foramen ovale with the anteromedial- and posterolateral-most aspects of the trigeminal impression with the coronal plane.^*

	Side(s)	N	Mean	SEM	SD	Min	Max	Range
AMF-AMI Angle (°)	L	113	65	0.8	8.9	41	92	51
	R	110	64	1.1	11.4	34	91	57
	L+R	223	65	0.7	10.2	34	92	58
AMF-PLI Angle (°)	L	113	95	0.8	8.2	68	116	48
	R	110	94	0.8	8.7	70	117	47
	L+R	223	95	0.6	8.5	68	117	49
PLF-PLI Angle (°)	L	113	69	0.9	9.3	44	97	53
	R	110	68	1.1	11.3	43	96	53
	L+R	223	69	0.7	10.3	43	97	54
PLF-AMI Angle (°)	L	113	33	1.0	10.2	9	61	52
	R	110	31	1.2	12.1	4	59	55
	L+R	223	32	0.8	11.2	4	61	57
Centroid-AMI Angle (°)	L	113	47	1.0	10.1	22	78	56
	R	110	46	1.2	12.3	17	76	59
	L+R	223	47	0.8	11.2	17	78	61
Centroid-PLI Angle (°)	L	113	84	0.8	8.7	57	108	51
	R	110	83	1.0	10.1	59	111	52
	L+R	223	83	0.6	9.4	57	111	54

^*The angles were measured at the anterolateral intersection of each line with the coronal plane. (See Figure 1).

Abbreviations:

AMF = anteromedial foramen

AMI = anteromedial trigeminal impression

PLF = posterolateral foramen

PLI = posterolateral trigeminal impression

Centroid = centroid of the foramen

The data from all parameters measured were normally distributed. Also, paired t-tests did not reveal any significant differences in any angles measured between sides (Table 2). Moreover, the mean differences among the angles between sides were all less than 2° (Table 2).

Table 2.

Paired samples t-test between left- and right-sided angular relationships of the foramen ovale and the trigeminal impression.

	Paired Differences
Pair (L vs R)	Mean	SD	SEM	95% C		T	df	P
				Lower	Upper
AMF-AMI angle	1.50	13.1	1.27	−1.0	4.0	1.19	106	0.236
AMF-PLI angle	0.88	11.3	1.09	−1.3	3.0	0.80	106	0.424
PLF-PLI angle	0.43	14.1	1.36	−2.3	3.1	0.31	106	0.754
PLF-AMI angle	1.69	13.1	1.27	−0.8	4.2	1.33	106	0.186
Centroid-AMI angle	1.61	14.1	1.36	−1.1	4.3	1.19	106	0.238
Centroid-PLI angle	0.50	13.1	1.26	−2.0	3.0	0.39	106	0.695

^*: The angles were measured at the anterolateral intersection of each line with the coronal plane. (See Figure 1).

Abbreviations: AMF = anteromedial foramen; AMI = anteromedial trigeminal impression; PLF = posterolateral foramen; PLI = posterolateral trigeminal impression; Centroid = centroid of the foramen

Discussion

The debilitating pain of trigeminal neuralgia often necessitates neurosurgical intervention via percutaneous transovale cannulation.^1,8–10,25 Although most procedures are largely successful and have minimal or transient adverse effects, there is potential for debilitating or fatal results when the angle upon which the surgical tool is introduced to the cranial base deviates from the boundaries of the foramen ovale.^{27,29–31,33,34} Also, with regard to specific targeting of particular trigeminal branches, operative techniques have been described; however, these descriptions have, in large, neglected to included specific angulation data. This study is the first to detail the angular relationship between the foramen ovale boundaries and the boundaries of the trigeminal impression in dry human skulls.

Most reports regarding percutaneous procedures for the treatment of TN focus principally on outcomes rather than specific details of the cannulation. A few studies have addressed cannulation angles; however, most reported the angulation relative to the axial plane.^35–39 Yao et al.³⁶ measured angular relationships between the FO and TI via CT scan in a population of 120 individuals, presumably Chinese, by using the sagittal plane as a reference. Their study measured the angle between the center of the FO and boundaries of the TI.³⁶ However, the term “center” is ambiguous and therefore may lead to confusion with regard to the specificity of cannulation angles. This study measured from the centroid of the FO – a quantitative, reproducible location, in addition to the anteromedial- and posterolateral-most boundaries of the FO.

The study by Yao et al.³⁶ identifies a 50.47°±11.93° (Mean±SD) angle formed by a line connecting the center of the FO to the AMI and the sagittal plane. Therefore, by subtracting 50.47° from 90°, a 39.53° angle would exist between the aforementioned line and the coronal plane. This study documented an average angle formed by the centroid-AMI line and the coronal plane of 47° ± 11.2° (Mean±SD). Also, Yao et al.³⁶ reported a 6.62° ± 14.52° angle formed by a line connecting the center of the FO to the PLI and the sagittal plane. Again, by subtracting 6.62° from 90°, an 83.38° angle was found to exist between the center-PLI line and the coronal plane. This study documented an average angle formed by the centroid-PLI line and the coronal plane of 83° ± 9.4° (Mean±SD), essentially the same angle as that reported by Yao et al.³⁶ from the ‘center’ of the FO to the posterolateral TI. The information presented by Yao et al.³⁶ is limited with regard to the absence of angles formed by projections from the anteromedial and posterolateral boundaries of the FO to the TI, whereas the angles presented in this report account for these parameters – a distinction that is important with regard to cannulation methods.

Van Kleef et al.⁴⁴ note that the stylet should be aimed at the medial FO for the treatment of V₁, the middle of the FO for the treatment of V₂, and the lateral FO for the treatment of V₃. Likewise, it has been advised to direct the stylet to the medial porus trigeminus for V₁ pain, the center of the porus for V₂ or multidivisional pain, and the lateral porus for V₃ pain.^45,46 In lieu of these technical notes, the data of this report may be particularly useful. For example, for the treatment of V₃, a stylet passed from the lateral FO (posterolateral) to the lateral porus would meet the coronal plane at an average angle of 69° ± 10.3° (reported here as the PLF-PLI angle). For the treatment of V₂ TN, a stylet passed from the “middle” of the FO (i.e. centroid) toward the center of the porus would meet the coronal plane at an angle of approximately 65° (=[46°+83°]/2) (Figure 1E). Finally, for the treatment of V₁ TN, a stylet passed from the medial FO (anteromedial) to the medial porus would meet the coronal plane at an average angle of 65° ± 10.2° (Reported here as the AMF-AMI angle). Because these approach angles are nearly the same, it appears that the translation of the surgical tool, in addition to the rotation, is of particular importance to target specific manifestations of TN.

For practical purposes, the angular information presented in this report can be interpreted alongside data which has documented angles relative to the horizontal plane. For example, the angle between the surgical tool and the axial plane has been reported to be 49.37 degrees for men and 52.26 degrees for women.³⁵ Huo et al.³⁷ noted that the approach angle ranged from 38.47°–51.89° with an average of 46.09° relative to the Reid line, which deviates from the axial plane (i.e. Frankfort horizontal line or orbitomeatal line) by approximately 7° to 10°.⁴⁷ Additionally, Peris-Celda et al.⁴⁸ noted that the approach angle should be approximately 45° angle with the hard palate in the lateral radiographic view.

Surgeons should be aware of the angular relationship between the FO and TI when performing percutaneous procedures for the treatment of TN. Understanding these relationships will help to prevent potentially severe adverse events and also aid in targeting specific regions of the trigeminal ganglion for localized treatment of V₁,V₂,V₃, or any combination of these distributions. The data presented in this report can be used to optimize surgical approaches for the percutaneous treatment of TN.