Day: September 18, 2026

Operative Neurosurgery. 2026;31(3):555-569

Persistent hormonal secretion after pituitary surgery may reflect disease extending beyond the apparent sellar tumor, including focal invasion of the medial cavernous sinus wall and its ligamentous attachments. Carpenter and colleagues examine this anatomical problem through a selected series of wall resections, combining endocrine outcomes with descriptions of intraoperative invasion patterns. The paper’s practical contribution is to clarify where tumor may remain concealed and how exposure can be adapted, while emphasizing the additional vascular and cranial nerve risks of extending the operation.

Objective

To assess endocrine outcomes after selective medial cavernous sinus wall resection and compare intraoperative inspection with preoperative Knosp grading for detection of wall invasion.

Methods

The authors retrospectively reviewed 223 endoscopic endonasal pituitary operations performed between January 2021 and December 2023. Twenty-five patients with functioning tumors underwent medial wall resection on the basis of intraoperative assessment, all by one surgeon. Eighteen were primary operations and seven reoperations.

The selected group included 12 somatotroph, nine corticotroph, two lactotroph and two plurihormonal tumors. Resected wall tissue was submitted separately for histopathological examination. Biochemical outcomes were assessed using subtype-specific criteria, and postoperative MRI was obtained at six to eight weeks. Mean follow-up was 26.2 months, with a range of 13-41 months.

Main results

Histological invasion was demonstrated in 20 cases. An important denominator distinction is that one specimen was lost: Table 2 therefore reports 20 positive and four negative results among 24 evaluable cases, or approximately 83%. The 80% quoted in the abstract expresses the 20 confirmed cases against the entire 25-patient surgical cohort.

The reported postoperative hormonal remission rate was 80% overall, approximately 89% after primary surgery and 57% after reoperation. These rates should not be equated with permanent cure: one patient with Cushing disease relapsed one year after an initial remission. Patients with persistent secretion required additional medical treatment or radiosurgery.

The authors reported 95% sensitivity and 75% specificity for direct endoscopic assessment of invasion. Histological involvement also occurred in selected patients with low Knosp grades, showing that limited radiographic extension does not exclude microscopic wall disease. These observations concern a highly selected resection cohort, not a screening population of all pituitary tumors.

Complications included one cerebrospinal fluid leak requiring repair, two transient abducens palsies that resolved within 10 days, one deep-vein thrombosis and three cases of postoperative hyponatremia requiring readmission. No internal carotid artery injury was reported.

Interpretation – operative relevance

The anatomical contribution is the recognition of four invasion patterns. Seven cases had focal wall involvement without adjacent ligament invasion. Six had a small wall perforation associated with inferior parasellar ligament involvement. Eight showed caroticoclinoid ligament involvement with extension into the dorsal clinoidal space. Four had diffuse wall thickening and more extensive ligamentous involvement, a pattern encountered particularly in recurrent disease.

These patterns explain why a largely intact-looking wall may conceal tumor beyond a small defect, and why the caroticoclinoid ligament and dorsal clinoidal space deserve attention when interpreting lateral tumor extension. They are descriptive operative findings, not a validated classification that independently determines how much tissue should be removed.

The authors describe wider lateral sellar exposure on the selected side to identify the carotid artery and medial opticocarotid recess. Doppler localization of the artery preceded opening of the anterior cavernous wall in every case. Medial wall mobilization then proceeded from anterior to posterior, with identification and management of its parasellar ligamentous and inferior hypophyseal arterial attachments. The relevant technical principle is controlled exposure of anatomical attachments rather than traction on a poorly defined wall.

Wall appearance and tactile findings informed selection: focal infiltration, irregularity, fibrosis and loss of the normal glossy appearance raised suspicion. Nevertheless, intraoperative inspection complements MRI; it does not justify disregarding radiological anatomy or routinely removing the wall in every functioning tumor.

The series also illustrates the importance of recognizing the limits of safe extension. Carotid injury and ocular motor dysfunction remain relevant risks despite their absence or transient nature in this small cohort. The reported technique requires specific cavernous sinus anatomical expertise, and persistent or inaccessible disease may still require complementary endocrine or radiation treatment.

Limitations

There was no comparable group in which a similarly suspicious wall was deliberately preserved. The study therefore cannot isolate the benefit of wall removal from tumor resection, patient selection or surgical expertise. Histological verification was limited to selected resected tissue, constraining the generalizability of the diagnostic accuracy estimates.

The small, heterogeneous endocrine subgroups and variable follow-up limit estimates of durable remission. The reported Knosp AUC of 0.300 is difficult to reconcile with the stated 65% sensitivity and 75% specificity at the selected threshold without further clarification of the analysis; it should not be used to conclude that MRI grading is generally unreliable. The clinically persuasive observation is narrower: focal wall invasion can be present despite a low Knosp grade.

Clinical takeaway

Treat the medial cavernous wall as a potential site of persistent disease, but make the decision to resect it selective and anatomically explicit. This paper helps identify concealed extension through wall defects and ligamentous attachments; it supports careful assessment and appropriate exposure, not routine wall excision or an assumption that early biochemical remission guarantees lasting cure.