Endoscopic occipital transtentorial transrecess approach for resection of a challenging midbrain-thalamic cavernous malformation: illustrative case

Abstract

BACKGROUND

Midbrain-thalamic cavernous malformations located adjacent to eloquent neurovascular structures are surgically challenging. The authors evaluated the feasibility and safety of the endoscopic occipital transtentorial transrecess approach (EOTTA) for resection of these lesions.

OBSERVATIONS

A 51-year-old woman presented with headache and acute, nonprogressive left hemiparesis. Imaging showed a 2.1 × 1.8 × 1.7–cm lesion at the right thalamus-midbrain junction consistent with a cavernous malformation; preoperative diffusion tensor imaging (DTI) tractograms fused with neuronavigation delineated corticospinal tracts and informed corridor planning. Through use of a lateral decubitus, gravity-assisted approach, a paramedian occipital craniotomy and a tentorial incision were performed 1 cm below the straight sinus to provide a transrecess route into the third ventricle. High-definition endoscopy and neuronavigation guided staged intertumoral debulking and gross-total resection in 110 minutes with 20 mL of blood loss. Postoperative imaging confirmed gross-total resection; the patient recovered full strength and resumed normal activities by 3 months.

LESSONS

EOTTA, combined with DTI-guided neuronavigation and endoscopic magnification, can permit safe, effective resection of selected midbrain-thalamic cavernous malformations while minimizing cortical retraction and injury to critical tracts.

https://thejns.org/doi/10.3171/CASE25537

Midbrain and thalamic cavernous malformations pose formidable challenges in neurosurgery due to their deep-seated location and proximity to eloquent neural structures.^1,2 Conventional microsurgical approaches can work for selected lesions but are limited by constrained visualization and the need for retraction, which increases the risk of morbidity.³ Advances in endoscopic technology—improved illumination, wide-angle optics, and minimally invasive corridors—have renewed interest in managing deep lesions.^4–6 Nevertheless, clinical data on purely endoscopic resections of combined midbrain-thalamic cavernous malformations remain sparse.

Building on our prior anatomical demonstration of the endoscopic occipital transtentorial transrecess approach (EOTTA) for safely exposing the posterior third ventricle,⁷ we combined high-resolution imaging and frameless neuronavigation to plan a precise, tract-sparing corridor. While emerging studies support endoscopic corridors for deep lesions^8,9 and our group has reported benefits of endoscopic transoccipital and transtentorial techniques,¹⁰ the application of EOTTA to lesions that span the midbrain-thalamic junction has not been previously established. Herein, we present our inaugural clinical experience using EOTTA for gross-total resection of a transitional midbrain-thalamic cavernous malformation, detailing operative nuances and early clinical outcomes.

Illustrative Case

Presentation

A 51-year-old woman with no comorbidities presented with 1 week of persistent headaches and acute-onset, mild, nonprogressive left-sided hemiplegia (motor strength 4/5). On neurological examination, she was fully alert (Glasgow Coma Scale score 15) with preserved right-sided strength. Noncontrast CT demonstrated a hyperdense, round mass measuring 2.1 × 1.8 × 1.7 cm at the right thalamus-midbrain junction. Subsequent contrast-enhanced MRI confirmed a T1- and T2-hyperintense lesion without enhancement, consistent with a cavernous malformation (Fig. 1A–F). Preoperative diffusion tensor imaging (DTI) delineated anterior and lateral displacement of the corticospinal tracts, with attenuation of fibers medial to the lesion (Fig. 1G).

FIG. 1.

Preoperative imaging assessment. A: Noncontrast axial CT scans showing a mildly hyperdense, round mass in the right thalamus extending into the midbrain, measuring approximately 2.1 × 1.8 × 1.7 cm. B: High-resolution axial T1-weighted MR images demonstrating the lesion’s T1 hyperintensity. C: Axial T2-weighted MR images confirming T2 hyperintensity consistent with blood products and cavernous malformation architecture. D–F: Postcontrast axial (D), sagittal (E), and coronal (F) T1-weighted images revealing no abnormal enhancement in any plane, supporting the radiographic diagnosis of a cavernous malformation centered on the right thalamus-midbrain junction. G: DTI tractograms fused to anatomical MR images depicting the corticospinal and adjacent white matter tracts displaced predominantly laterally, anteriorly, and superiorly with relative paucity of fibers medially. The planned EOTTA trajectory is indicated as an arrow on the sagittal MR image (E) and sagittal fused MRI-DTI view (G), and its axial projection is shown as a dashed line on the axial MR image (D) and axial DTI view (G), demonstrating a posteromedial corridor medial to the anterolaterally displaced corticospinal tract.

Surgical Procedure

Patient Position, Incision, and Craniotomy

We integrated MRI and DTI into a frameless neuronavigation system for precise intraoperative guidance. Given that the thinnest portion of the lesion was situated medially to the thalamus, with anteroinferior extension and the corticospinal tract positioned anterolaterally, we selected EOTTA to provide a posteromedial corridor while minimizing tract manipulation. The planned endoscopic trajectory defined on fused MRI-DTI datasets is illustrated in Fig. 1E and G (sagittal views), with its axial projection shown as a yellow dashed line in Fig. 1D and G, demonstrating a posteromedial transrecess corridor medial to the displaced corticospinal tracts. The patient was placed in the lateral decubitus position with a 30° upper body elevation to optimize venous drainage; the head was flexed and slightly rotated to align the endoscopic trajectory. A 7-cm paramedian skin incision 2 cm lateral to midline permitted a 4 × 3–cm occipital craniotomy extending from 1 cm below the lambdoid suture to 1 cm above the transverse sinus (Fig. 2).

FIG. 2.

Schematic diagram of the surgical position, incision, and craniotomy. The patient was placed in a 30° elevated lateral decubitus position to optimize venous drainage, with the head flexed forward to align the surgical trajectory. A 7-cm paramedian skin incision (2 cm from midline) was made, allowing creation of a 4 × 3–cm craniotomy. The window’s medial edge spanned the superior sagittal sinus—its superior margin 1 cm below the lambdoid suture and its inferior margin 1 cm above the transverse sinus.

Approach and Tumor Removal

The dura mater was opened as a semilunar flap and reflected toward the superior sagittal sinus, exposing the falx margin (Fig. 3A). Next, a high-definition endoscope was then introduced and neuronavigation localized the straight sinus (Fig. 3B), guiding a tentorial incision 1 cm inferior to its margin (Fig. 3C). Then, sharp arachnoid dissection of the quadrigeminal cistern released CSF and exposed the pineal recess (Fig. 3D). After opening and traversing the pineal recess, the endoscope was introduced into the third ventricle (Fig. 3E and F). Guided by high-definition, wide-angle endoscopic views of the lateral third ventricular wall and adjacent neurovascular structures (Fig. 4A), and with neuronavigation confirming the tumor’s location (Fig. 4B), we performed staged intralesional debulking under endoscopic magnification, meticulously preserving the lesion-brain interface (Fig. 4C). Finally, gross-total resection was achieved (Fig. 4D), and endoscopic inspection verified intact vascular and neural structures without bleeding (Fig. 4E and F).

FIG. 3.

Establishment of the EOTTA corridor. A: A semilunar dural flap was opened and reflected toward the margin of the superior sagittal sinus. B: Frameless neuronavigation localized the straight sinus and guided safe tentorial planning. C: A tentorial incision was made approximately 1 cm inferior to the straight sinus margin. D: Sharp arachnoid dissection of the quadrigeminal cistern released CSF and exposed the pineal region. E: The pineal recess was opened to enlarge the natural corridor. F: The posterior third ventricle was entered through this transrecess route, establishing the working corridor to the medial thalamus and dorsal midbrain.

FIG. 4.

Endoscopic resection under magnified visualization. A:Wide-angle, high-definition endoscopic view showing the posterior aspect of the medial thalamic wall and the surgical corridor. B: Neuronavigation screenshot confirming the lesion’s location relative to the third ventricle. C: Staged intralesional debulking with a clearly defined lesion-brain interface. D:Intraoperative endoscopic view after apparent gross-total resection. E:Inspection of the corridor demonstrates intact vascular and neural structures without active bleeding. F: The occipital lobe is preserved and not retracted aggressively, illustrating minimal cortical manipulation with this posterior approach.

Cranial Closure

The dura was closed in a watertight fashion, the bone flap was secured with titanium plates, and the muscle and skin were closed in standard layered fashion.

Surgical Outcomes and Follow-Up

The procedure lasted approximately 110 minutes with an estimated blood loss of 20 mL (Video 1). Immediate postoperative CT showed no hemorrhage or ischemia, and the patient’s neurological status remained stable. Her headache resolved by postoperative day 7. Left-sided motor strength improved steadily after surgery, and no formal postoperative rehabilitation program was required. At 3 months, MRI confirmed complete lesion removal (Fig. 5), and her left-sided motor strength had recovered to 5/5, enabling unrestricted daily activities. Importantly, no significant intraoperative or postoperative complications were encountered. Per our institutional protocol for EOTTA, if brain relaxation remains insufficient after corridor entry, occipital horn puncture is prepared for CSF drainage to reduce intracranial tension. For venous bleeding, we first gently apply a cottonoid tamponade followed by a short waiting period and reinspection, avoiding direct bipolar coagulation on major venous trunks. Persistent oozing within the lesion cavity is managed in a stepwise manner with continuous warm saline irrigation, repeated gentle cottonoid compression, and bipolar coagulation of the cavity wall or feeding points as needed, supplemented by hemostatic materials (e.g., oxidized cellulose or gelatin-based agents). None of these escalation measures were required in this case.

FIG. 5.

Three-month postoperative MR images demonstrating gross-total resection. A: Axial T1-weighted images showing no evidence of residual lesion. B: Postcontrast axial T1-weighted images confirming the absence of abnormal enhancement or residual cavernoma, consistent with gross-total resection and no postoperative contrast-enhancing collection.

VIDEO 1. EOTTA for resection of a right midbrain-thalamic cavernous malformation. The video opens with the patient’s brief clinical details and preoperative imaging and then demonstrates patient positioning for EOTTA and a paramedian occipital skin incision. After craniotomy, a semilunar dural flap is reflected and a tentorial incision is made approximately 1 cm inferior to the straight sinus. Subsequent segments show sharp arachnoid dissection of the quadrigeminal cistern, opening of the pineal recess, endoscopic entry into the posterior third ventricle, and establishment of a transrecess working corridor. Under high-definition, wide-angle endoscopic visualization with real-time neuronavigation, staged intralesional debulking is performed using microdissection and suction while the lesion-brain interface is meticulously preserved. The final segment documents endoscopic inspection after apparent gross-total resection, demonstrating intact vascular and neural structures and secure hemostasis. The video is edited to emphasize key technical steps and relevant anatomy; all identifiable patient information has been removed, and written informed consent for anonymized intraoperative image/video publication was obtained. Click here to view.

Informed Consent

The necessary informed consent was obtained in this study.

Discussion

Observations

In patients presenting with a first-time hemorrhage of a supratentorial cavernous malformation, conservative management, including observation and corticosteroids, remains a valid option.¹ However, growing evidence suggests that the cumulative risk of rebleeding and persistent neurological deficit often tips the balance in favor of early resection for symptomatic, surgically accessible lesions.^11–14 In our patient, acute-onset left hemiparesis following hemorrhage (without further progression during the preoperative period), together with the lesion’s superficial location along the medial thalamic surface, prompted us to pursue gross-total resection rather than continued observation.

Our principal finding is that EOTTA can enable safe gross-total resection of select transitional midbrain-thalamic cavernous malformations. Three factors supported this outcome: 1) use of a natural cisternal corridor with venous preservation, allowing resection without occipital contusion or injury to corticospinal tracts, the quadrigeminal plate, or galenic venous structures; 2) operative efficiency—the case was completed in 110 minutes with minimal blood loss; and 3) a single posterior corridor, which provided simultaneous access to both midbrain and medial thalamic components, overcoming compartmental limitations of some traditional approaches.

To our knowledge, this is the first reported gross-total resection of a midbrain-thalamic cavernous malformation using EOTTA, extending the approach’s application beyond pineal and third ventricle lesions.

Lessons

Our experience with EOTTA yields three principal lessons and invites a balanced comparison with established surgical strategies for midbrain-thalamic cavernous malformations. First, EOTTA’s ability to bridge anatomical compartments offers distinct advantages but is not universally superior. Its trajectory through the quadrigeminal cistern and pineal recess provides precise, simultaneous access to the medial thalamus and midbrain—regions that cannot be reached concurrently by other conventional approaches. This posterior corridor allows gross-total resection while preserving critical neural and vascular structures. However, this benefit is highly anatomy-dependent and becomes less applicable when lesions extend purely anteriorly or laterally. In comparison, the transsylvian-cerebral peduncle approach provides anterior and anterolateral access, often preferred for lesions in the lateral midbrain or cerebral peduncle. Yet it necessitates extensive sylvian fissure dissection, carries an increased risk of corticospinal tract injury, and often provides suboptimal exposure of medial thalamic structures.^14,15 The orbitozygomatic craniotomy, although versatile, is constrained by anterolateral working angles and may be further limited by anatomical variability or the potential for damage to corticospinal fibers or cranial nerve nuclei.^16,17 Similarly, the contralateral transcallosal-choroidal fissure approach grants access to medial thalamic lesions but poses significant risks of callosal or cortical injury, potentially resulting in memory impairment or seizures.^18,19 The supracerebellar infratentorial approach, widely used for pineal and dorsal midbrain lesions, remains a familiar route for many neurosurgeons;^5,6,20 however, its restricted anterior and inferior reach may preclude adequate visualization of lesions extending into the midbrain. This approach may also necessitate sacrifice of bridging veins, increasing the risk of postoperative complications. By tailoring the surgical corridor to the lesion’s specific topography rather than adhering to predefined anatomical compartments, EOTTA facilitates gross-total resection with minimal retraction and maximal preservation of surrounding neurovascular structures. That said, EOTTA is best suited for lesions situated at or near the posterior third ventricular wall and may not be optimal for lesions with purely lateral or anterior thalamic extension. Furthermore, the narrow working corridor can limit bimanual manipulation and depth perception, potentially affecting resection efficiency in complex lesions.

Second, venous preservation is a strength but requires anatomical precision. The transrecess access helps preserve venous integrity by creating a natural corridor into the third ventricle, thereby safeguarding critical pineal veins and the galenic drainage system²¹— in contrast to suprapineal techniques that risk venous infarction from stalk transection. Preservation of these veins is paramount to avoid traction injury to the superior internal cerebral veins and vein of Galen.⁷ Nevertheless, the tentorial window in EOTTA is narrow and demands meticulous preoperative planning, particularly in cases with variant venous anatomy. Anatomical variability in venous angulation or low-lying straight sinus may render tentorial incisions hazardous without precise neuronavigation.

Third, technological dependence enhances precision but limits generalizability. Combining high-definition endoscopy with gravity-assisted retraction and DTI-based neuronavigation enables precise avoidance of the corticospinal tracts and eliminates microscopic blind spots. This synergy provides direct visualization of structures such as the quadrigeminal plate, cerebral aqueduct, and posterior commissure—preventing ocular motility disturbances, diplopia,^22–24 and visual field loss from occipital lobe contusion, a complication once common with microscopic approaches.^25–27 Collectively, these advances minimize injury to deep venous complexes and eloquent pathways. However, this technique relies on advanced imaging and equipment that are not universally available, and the learning curve for precise endoscopic manipulation is steep. Thus, selection of EOTTA should be balanced against surgeon experience, available equipment, and patient-specific anatomy.

This report is limited by its single-case design and cannot define comparative effectiveness. EOTTA should be viewed as a complementary technique rather than a replacement for established approaches. Prospective, multicenter series comparing outcomes across lesion topographies and anatomical variants are needed. Future refinements, for example, intraoperative MRI, 3D tractography overlays, and augmented-reality guidance, may further improve the safety and individualization of corridor planning.

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Zhang, Liu, Tian. Acquisition of data: Sun, Zhu. Analysis and interpretation of data: Yin, Yang, Wang. Drafting the article: Liu, Tian, Sun, Yin, Yang. Critically revising the article: Zhang, Liu, Zhu. Reviewed submitted version of manuscript: Sun, Yin, Wang. Approved the final version of the manuscript on behalf of all authors: Zhang. Statistical analysis: Tian. Administrative/technical/material support: Zhang, Yin, Yang. Study supervision: Tian.

Supplemental Information

Videos

  Video 1. https://vimeo.com/1148671814.

Correspondence

Xiaobiao Zhang: Zhongshan Hospital, Fudan University, Shanghai, China. xiaobiao_zhang@163.com.