Clinical Outcome of Extended Endoscopic Endonasal Transclival Approach for Skull Base Lesions Authors: Alejandro Vargas-Moreno, Sami Khairy, Mouaz Saymeh, Damanpreet Kaur Lang, Nathan Yang, Jessica Rabski, Shaun Kilty, Fahad Alkherayf Published: 2026 ABSTRACT BACKGROUND AND OBJECTIVES:Surgical access to the clival region remains a significant challenge because of its midline location, deep anatomic position, and complex neurovascular relationships. The extended endoscopic endonasal approach (EEA) offers a direct, minimally invasive corridor to this intricate region of the skull base. This study presents a comprehensive review of the clinical presentation, surgical indications, and outcomes in a series of 46 patients who underwent EEA for lesions involving the clivus.METHODS:This was a retrospective study in which we identified cases of clival pathologies through our institution's skull base database from October 2014 and October 2024 based on our inclusion and exclusion criteria.RESULTS:Surgical indications included lesions at the craniovertebral junction, specifically craniovertebral junction instability requiring endonasal odontoidectomy in 15 patients (32.6%) and space-occupying lesions in the remaining 31 patients (67.4%). Among the latter, chordomas were the most common histopathological diagnosis. Clinical symptoms varied according to tumor location and underlying pathology. The extent of resection and the use of adjuvant radiotherapy were tailored to histopathological subtype. In patients with space-occupying lesions, gross total resection (GTR) was achieved in 46.7% of cases, while adjuvant radiotherapy was administered in 18 patients (38.1%). In cases of chordoma, GTR was more frequently achieved in smaller tumors (mean volume: 9.3 cm3 vs 12.6 cm3; P = .359) and in those confined to the midline. Furthermore, GTR was associated with improved cranial nerve function during follow-up. Residual tumors were most frequently observed in the cavernous sinus and retro-paraclival carotid regions.CONCLUSION:Based on our experience, the EEA represents a safe and effective surgical option for clival lesions, with potential benefits including avoidance of craniotomy and brain retraction, reduced neurovascular manipulation, and lower surgical morbidity. ABBREVIATIONS: craniovertebral junction endoscopic endonasal approach gross total resection. Surgical access to the clival region represents a formidable challenge because of its midline position, deep-seated location, and intricate neurovascular relationships.1 The clivus, situated at the skull base and surrounded by critical structures such as the brainstem, cranial nerves, and major arteries, poses significant obstacles for surgical intervention. Historically, open surgical approaches, which often required extensive bone drilling and neurovascular manipulation, were associated with substantial morbidity.2,3 In recent years, extended endoscopic endonasal approaches (EEA) have emerged as a minimally invasive alternative, providing a direct corridor to this complex anatomic region while eliminating the need for brain retraction. These approaches facilitate safe resection and treatment of various pathologies in the clival region.4 This report reviews the clinical presentation and outcomes of patients who underwent an endoscopic endonasal approach for the treatment of skull base lesions involving the clivus and detailing our surgical experience. METHODS The skull base database of our institution was queried to identify cases of endoscopic endonasal skull base surgery for the treatment of clival pathologies between October 2014 and October 2024. Records of all adult patients with this type of lesion were screened for inclusion. All adult patients with this type of lesion were evaluated for inclusion into our study. Our exclusion criteria consisted of incomplete clinical records of the screened subject, inability to definitively establish clival pathology, and subjects with a history of trauma to the skull base, active infection during the time of surgery, or previous incident of cerebrospinal fluid (CSF) leaks. Two fellowship-trained skull base surgeons, a neurosurgeon and an ear, nose, and throat specialist, performed all the cases. The skull base data set of our institution contains no personal or identifying information about participants and is produced using chart reviews. The database has ethical approval from the institution's ethical board, so the study did not require a separate ethical approval in place. Also, as it is a retrospective study, patient consent was not required. We reviewed every patient's operative note and medical records in greater detail. We collected data on the type of approach, American Society of Anesthesiologists class, surgical time, use of lumbar drain, presence of CSF leak, tumor location and volume (using the modified ellipsoid formula [A*B*C]/2), blood loss, use of dural sealants, length of stay, readmission at 30 days, extent of resection, postoperative radiotherapy treatment, recurrence at last follow-up, intraoperative and postoperative complications, use of nasoseptal flap, World Health Organization pathological tumor grade, and mean follow-up. The techniques for the endoscopic endonasal transclival approach to the various regions of the clivus, as well as the craniovertebral junction (CVJ), have been extensively described elsewhere.4-7 At the skull base, clivus forms a smooth, sloping midline surface that anteriorly extends from foramen magnum to the dorsum sellae's posterior edge. This structure is a result of the anatomic fusion between sphenoid body with the basiocciput at the spheno-occipital synchondrosis. In this region, articulation happens between the occipital bone articulates and the petrous segment of the temporal bone where it meets the petroclival fissure. The clivus is lined inferiorly by the rounded projections of the jugular tubercles, which marks the transition between the basiocciput and the lateral jugular regions of the occipital bone.8 In our cohort, we identified 2 surgical groups: patients with space-occupying lesions and those with CVJ instability requiring odontoidectomy. Cases in the first group selected for endoscopic endonasal approaches were reviewed by a multidisciplinary treatment board, with consideration given to lesion extent, presumed diagnosis, and the patient's clinical status. For patients with CVJ pathology, all cases required a staged surgical procedure: an initial endoscopic approach, followed by an open posterior cervical fusion. The extent of the posterior fusion was determined on a patient-specific basis. The extent of tumor resection was classified as total and subtotal. The goal of surgery was gross-total resection (GTR) of the tumor in all cases except in those with multicompartmental lesions in which decompression of the brainstem was the main goal. The degree of tumor resection was evaluated on postoperative imaging studies by a trained neuroradiologist and included MRI in all cases. During the follow-up, we conducted 3T brain MRI imaging on the patients to assess any residual or recurrent tumoral tissue. Descriptive statistics for continuous variables are described as mean, range, and SD, whereas frequency and percentage are reported for categorical variables. We applied the Shapiro-Wilk test to ensure normal distribution of our data. With a 95% significance level, we conducted χ2 and Fisher exact test to compare relationship between categorical variables. We used IBM SPSS Statistics version 26.0 software (International Business Machines Corporation) to conduct our statistical analysis. RESULTS A total of 46 patients underwent an extended endoscopic endonasal approach to the clivus. Preoperative American Society of Anesthesiologists classification was type III in 31 patients (67.4%), type IV in 13 patients (28.6%), and type II in the remaining 2 patients (4.3%) (Table 1). TABLE 1.Patient Demographics and General CharacteristicsVariableSymptomsHeadaches, n (%)38 (82.6)CN palsy, n (%)15 (32.6) Abducens nerve 9/15 Oculomotor nerve 4/15 Glossopharyngeal nerve 1/15 Hypoglossal nerve 1/15Visual dysfunction, n (%)5 (10.9)Myelopathy, n (%)13 (28.3)ASA Class, n (%)II2 (4.3)III31 (67.4)IV13 (28.3)Pathology, n (%)Odontoid pannus15 (32.6)Chordoma13 (28.3)Meningioma6 (13)Pituitary adenoma3 (6.5)Chondrosarcoma2 (4.3)Schwannoma1 (2.1)Meningogele1 (2.1)Lymphangioma1 (2.1)Fibrous dysplasia1 (2.1)Craniopharyngioma1 (2.1)Skull base carcinoma1 (2.1)Metastasis1 (2.1)Tumor location in clivus, n (%)Upper third12 (26)Upper-middle third19(41.3)Lower third—CCJ15(32.6)Mean surgical time (min) (min-max)308 (59-721)Mean blood loss (mL) (min-max)424 (30-2500)Mean LOS (d)12.2Readmission at 30 d, no. (%)7 (15.2)Postop. CSF leak, no. (%)3 (6.5)Postop. DI, no. (%)2 (4.3)ASA, American Society of Anesthesiologists; CCJ, craniocervical junction; CN, cranial nerve; CSF, cerebrospinal fluid; DI, diabetes insipidus; LOS, length of stay; postop, postperative. Patient Demographics and General Characteristics ASA, American Society of Anesthesiologists; CCJ, craniocervical junction; CN, cranial nerve; CSF, cerebrospinal fluid; DI, diabetes insipidus; LOS, length of stay; postop, postperative. Surgical indications included CVJ lesions, specifically CVJ instability requiring endonasal odontoidectomy in 15 cases (32.6%) and space-occupying lesions in the remaining 31 cases (67.4%). Among the latter, chordomas represented the most frequent histopathological diagnosis. Regarding lesion location, 12 cases (26%) involved the upper third of the clivus, 19 cases (41.3%) were located in the upper-middle third, and the remaining 15 cases involved the lower third or craniocervical junctionregion. Clinical presentation varied depending on tumor location and underlying pathology. Headache was the most common symptom, present in 38 patients (82.6%), followed by cranial nerve palsy in 15 patients (32.6%) and myelopathy in 13 patients (28.3%). The mean operative time for endoscopic procedures was 308 minutes (range: 59-721 minutes; SD: 168 minutes), with a mean estimated blood loss of 472 mL (range: 30-2500 mL; SD: 407 mL). Three major vascular injuries occurred during resection of chordomas involving encased arteries; however, none resulted in ischemic complications. Red blood cell transfusions were required in 4 cases (8.7%). No procedures were aborted, and no perioperative mortality occurred. The mean length of stay was 12.2 days. Postoperative CSF leaks occurred in 3 patients (6.5%)—one with a meningioma and 2 with chordomas—all managed successfully with lumbar drainage; none required reoperation, and no postoperative infections were reported. There was no CSF leak in the odontoidectomy group. Diabetes insipidus developed in 2 patients (4.3%)—one with a pituitary adenoma and another with a craniopharyngioma. One patient with a skull base carcinoma was readmitted within 30 days because of a pulmonary embolism. Among patients with space-occupying lesions, GTR was achieved in 46.7% of cases, and adjuvant radiotherapy was administered in 18 patients (38.1%). Resection extent and radiotherapy use varied by histopathological diagnosis (Table 2). Although statistical significance was not reached because of limited sample size, trends suggested that GTR in chordoma cases was more frequently achieved in tumors with smaller volumes (mean: 9.3 cm3 vs 12.6 cm3; P = .359) and in those located along the midline. GTR was also associated with improved cranial nerve function during follow-up (Table 3). TABLE 2.Resection Outcomes in Non Odontoidectomy CasesVariablesAmount of resection according to pathology, n (%)GTRSTRChordoma6 (46.2)7 (53.8)Meningioma1 (16.6)5 (83.7)Adenoma3 (100)0Chondrosarcoma1 (50)1 (50)Skull base carcinoma01 (100)Metastases1 (100)0Craniopharyngioma1 (100)0Fibrous dysplasia01 (100)Lymphangioma1 (100)0Meningocele1 (100)0Residual locations, no. of casesCavernous sinus8Meckel cave5Occipital condyle2Retro paraclival carotid space6Petroux apex2Ambient cistern1Ventral jugular foramen2Adjuvant treatmentsRadiotherapy, n (%)18 (38.1%)Chordoma13Chondrosarcoma2Meningioma1Skull base carcinoma1Craniopharyngioma1Mean FU (y)5.93Progression, n (%)3 (9.6)FU, follow-up; GTR, gross total resection; STR, subtotal resection. Resection Outcomes in Non Odontoidectomy Cases FU, follow-up; GTR, gross total resection; STR, subtotal resection. TABLE 3.Chordoma Extent of Resection AnalysisVariablesGTR6STR7P valueEffect size (95% CI)Mean preop vol9.312.6.369Mean difference: −3.3 (−8.65 to 2.05)Midline location (%)5 (38.5)2 (15.4).048OR 12.5 (1.01-154.3)Mean blood loss (mL)425893Postop CSF leak11.99OR 1.2 (0.06-22.9)Persistent ophthalmoparesis at last FU (%)1 (7.7)3 (23.1).033OR 0.27 (0.02-3.45)Recurrence at last FU (%)03 (23.1).067OR 0.10 (0.004-2.37)CSF, cerebrospinal fluid; FU, follow-up; GTR, gross total resection; OR, odds ratio; postop, postperative; preop, preoperative; STR, subtotal resection. Chordoma Extent of Resection Analysis CSF, cerebrospinal fluid; FU, follow-up; GTR, gross total resection; OR, odds ratio; postop, postperative; preop, preoperative; STR, subtotal resection. Residual tumors were most commonly located in the cavernous sinus and behind the para clival carotid artery (Figure). Our cohort's mean follow-up duration was 5.93 years (range in years: 2-10; SD in years: 2.62). We observed tumor recurrence/progression in 3 patients, all with chordomas, respectively, at 2, 3, and 5 years postoperatively. Top-down view of the clival region. Heat map representation of residual tumor distribution across key anatomic regions. The cavernous sinus exhibits the highest incidence of residual tumor, as indicated by the red color. This is followed by the Meckel cave and the retro paraclival carotid space, which demonstrate moderate to high residuals. The occipital condyle, petrous apex, and ventral jugular foramen are highlighted in light yellow, reflecting a low to moderate frequency of residual disease. The ambient cistern, marked in green, appeared to be the least affected region in our analysis. DISCUSSION Over the recent decades, various open approaches have been developed for accessing the clival region.4,9,10 Although the transcranial procedures provide sufficient exposure with favorable outcomes, there are important limitations.11 The choice of surgical approach depends on various factors, including the tumor's location, its primary axis of extension, and the surgeon's experience and preference. Endoscope-assisted surgery offers visualization that is comparable with—or even superior to—traditional microsurgical techniques.4,8,12 High-resolution endoscopic imaging enhances the identification of anatomic and pathological structures during resection, allowing access to otherwise challenging areas. This is particularly valuable in clival lesions, where conventional open approaches may be limited by the need for brain retraction, neurovascular manipulation, extensive bone removal, and increased morbidity.5,9,13 Clival lesions are rare and encompass a wide spectrum of pathologies, continuing to pose substantial challenges for neurosurgeons.5,8 Our series demonstrates that endoscopic approaches to the clival region are both safe and effective, achieving high rates of GTR across diverse lesion types and sizes, with no major complications. A broad spectrum of adverse events may occur during extended endonasal skull base exposures for tumor resection. In our cohort, the overall rate of CSF leak was consistent with that reported in large contemporary series14,15 and was successfully managed with lumbar drain placement. Nevertheless, there should be a low threshold for reintervention in cases of refractory leaks or neurological deterioration due to pneumocephalus. Vascular injury may also occur, particularly in larger lesions and multicompartmental tumors. In our series, 3 cases were identified, all involving posterior communicating artery injury encased by chordomas. These were managed with microsurgical bipolar coagulation without parent vessel sacrifice or the need for endovascular therapy. Careful preoperative vascular assessment with noninvasive imaging studies should be routinely performed to minimize these risks. Ultimately, meticulous patient selection remains essential to reducing adverse outcomes and limiting residual tumor burden by recognizing the anatomic constraints of the endonasal corridor.16 The location and extent of a lesion within the clivus not only influence its clinical presentation but also dictate technical aspects of the surgical approach. Cranial nerve palsy because of direct compression and myelopathy were the most common symptoms in our series, consistent with other reports.17 We successfully treated lesions across all anatomic regions of the clivus. Notably, in chordoma cases, a pure midline extension was identified as a prognostic factor for achieving GTR, emphasizing the importance of preoperative planning and detailed imaging analysis—a finding supported by previous studies.18-20 Additionally, the endonasal approach to the lower third of the clivus and the CVJ has established itself as a valuable tool in skull base surgery. Its minimally invasive nature makes it especially attractive for managing degenerative pathology in this region, offering the potential to reduce morbidity associated with more extensive approaches while improving patient comfort.6 The wide range of histological diagnoses encountered in this area increases the complexity of effective lesion management.21 Benign pathologies demonstrated the highest GTR rates in our cohort, likely due to their well-defined boundaries and non-invasive behavior. In contrast, malignant tumors—where extent of resection is a critical prognostic factor—present greater challenges. Although GTR improves outcomes in these cases, recurrence remains common because of microscopic residual disease at the margins.22-24 These limitations highlight the need to consider the boundaries of the endoscopic transclival approach during surgical planning, particularly in aggressive pathologies that may require combined approaches for optimal results.22 In our cohort, heatmap analysis revealed that the majority of residual tumor was located in lateral regions, such as the cavernous sinus and behind the paraclival carotid artery. This underscores the importance of meticulous exposure, dissection, and tumor removal, while maintaining a balance between maximal safe resection and avoidance of excessive morbidity.4,5,7 Recently, the concept for supra marginal resection applied to clival pathology has been proposed in order to overcome this limitations by expanding the anatomical boundaries of the current endoscopic approaches.25 In selected cases of subtotal resection, adjuvant radiotherapy was used to improve overall outcomes and was frequently employed in our series.26 Limitations of our study include its retrospective design and the inherent biases associated with this methodology. Furthermore, the relatively small sample size and the single-center nature of the study may limit the generalizability and prognostic strength of our findings. CONCLUSION The endoscopic endonasal approach is a safe and effective alternative for managing clival pathologies. The clear cut advantages are mitigating craniotomy and brain retraction, reducing the need for neurovascular manipulation as well as attenuating morbidity. Although major complications are rare with this technique, it does possess some limitations. This approach should be incorporated into a cranial base surgeon's armamentarium and be regarded as a viable option for selected patients presenting with these complex pathologies.