Preserve or sacrifice the stalk? Endocrinological outcomes, extent of resection, and recurrence rates following endoscopic endonasal resection of craniopharyngiomas.

Objective: Gross-total resection (GTR) of craniopharyngiomas (CPs) is potentially curative and is often the goal of surgery, but endocrinopathy generally results if the stalk is sacrificed. In some cases, GTR can be attempted while still preserving the stalk; however, stalk manipulation or devascularization may cause endocrinopathy and this strategy risks leaving behind small tumor remnants that can recur.

Methods: A retrospective review of a prospective cohort of patients who underwent initial resection of CP using the endoscopic endonasal approach over a period of 12 years at Weill Cornell Medical College, NewYork-Presbyterian Hospital, was performed. Postresection integrity of the stalk was retrospectively assessed using operative notes, videos, and postoperative MRI. Tumors were classified based on location into type I (sellar), type II (sellar-suprasellar), and type III (purely suprasellar). Pre- and postoperative endocrine function, tumor location, body mass index, rate of GTR, radiation therapy, and complications were reviewed.

Results: A total of 54 patients who had undergone endoscopic endonasal procedures for first-time resection of CP were identified. The stalk was preserved in 33 (61%) and sacrificed in 21 (39%) patients. GTR was achieved in 24 patients (73%) with stalk preservation and 21 patients (100%) with stalk sacrifice (p = 0.007). Stalk-preservation surgery achieved GTR and maintained completely normal pituitary function in only 4 (12%) of 33 patients. Permanent postoperative diabetes insipidus was present in 16 patients (49%) with stalk preservation and in 20 patients (95%) following stalk sacrifice (p = 0.002). In the stalk-preservation group, rates of progression and radiation were higher with intentional subtotal resection or near-total resection compared to GTR (67% vs 0%, p < 0.001, and 100% vs 12.5%, p < 0.001, respectively). However, for the subgroup of patients in whom GTR was achieved, stalk preservation did not lead to significantly higher rates of recurrence (12.5%) compared with those in whom it was sacrificed (5%, p = 0.61), and stalk preservation prevented anterior pituitary insufficiency in 33% and diabetes insipidus in 50%.

Conclusions: While the decision to preserve the stalk reduces the rate of postoperative endocrinopathy by roughly 50%, nevertheless significant dysfunction of the anterior and posterior pituitary often ensues. The decision to preserve the stalk does not guarantee preserved endocrine function and comes with a higher risk of progression and need for adjuvant therapy. Nevertheless, to reduce postoperative endocrinopathy attempts should be made to preserve the stalk if GTR can be achieved.