A simple surgical technique for sellar closure after transsphenoidal resection of pituitary adenomas in the context of risk factors for cerebrospinal fluid leaks and meningitis

Abstract

OBJECTIVE The transsphenoidal approach is the standard for most pituitary tumors. Despite low morbidity, postoperative CSF fistulas and meningitis are specific complications. Various surgical closure techniques for intraoperative CSF (iCSF) leak and sellar reconstruction have been described. For many years the authors have applied synthetic materials for iCSF leak repair and sellar closure in a standardized fashion in their department. Here they analyze the surgical outcome as well as risk factors for iCSF leak and meningitis. METHODS All patients with transsphenoidal resection of a pituitary adenoma performed by the same surgeon between January 2013 and December 2019 were screened retrospectively. A small amount of iCSF flow without a diaphragmatic defect was classified as a minor leak, and obvious CSF flow with or without a diaphragmatic defect was classified as a major leak. In case of iCSF leak, a fibrin- and thrombin-coated sponge was used to cover the diaphragmatic defect and another one was used for the sellar opening. A gelatin sponge was placed in the sphenoid sinus as an abutment. The primary and secondary outcomes were the number of postoperative CSF (pCSF) leaks and meningitis, respectively. Clinical, histological, and perioperative data from medical records were collected to identify risk factors for CSF leak and meningitis. RESULTS Of 417 transsphenoidal surgeries, 359 procedures in 348 patients with a median age of 54 years were included. There were 96 iCSF leaks (26.7%; 37.5% major, 62.5% minor). In 3 of 359 cases (0.8%) a pCSF fistula occurred, requiring revision surgery in 2 patients and a lumbar drain in 1 patient. Meningitis occurred in 3 of 359 cases (0.8%). All 3 patients recovered without sequelae after antibiotic therapy. According to univariate analysis, risk factors for iCSF leak were macroadenoma (p = 0.006) and recurrent adenoma (p = 0.032). An iCSF leak was found less often in functioning adenomas (p = 0.025). In multivariate analysis recurrent tumors remained as a risk factor (p = 0.021) for iCSF leak. Patients with iCSF leak were at increased risk for a pCSF leak (p = 0.005). A pCSF leak in turn represented the key risk factor for meningitis (p = 0.033). CONCLUSIONS Patients with macroadenomas and recurrent adenomas are especially at risk for iCSF leak. An iCSF leak in turn increases the risk for a pCSF leak, which carries the risk for meningitis. The authors’ surgical technique leads to a very low rate of pCSF leaks and meningitis without using autologous graft materials. Hence, this technique is safe and improves patient comfort by avoiding the disadvantages of autologous graft harvesting.