Search

Background: The association between spontaneous skull base cerebrospinal fluid (CSF) leaks and idiopathic intracranial hypertension (IIH) has been suggested, but its significance remains unclear., Objective: To estimate the prevalence of IIH in spontaneous skull base CSF leak patients., Methods: Systematic collection of demographics, neuro-ophthalmic and magnetic resonance imaging evaluation of spontaneous skull base CSF leak patients seen pre- and post-leak repair in one neuro-ophthalmology service. Patients with preexisting IIH were diagnosed with definite IIH if adequate documentation was provided; otherwise, they were categorized with presumed IIH. Classic radiographic signs of intracranial hypertension and bilateral transverse venous sinus stenosis were recorded., Results: Thirty six patients were included (age [interquartile range]: 50 [45;54] years; 94% women; body mass index: 36.8 [30.5;39.9] kg/m2), among whom six (16.7%, [95% confidence interval, CI]: [6.4;32.8]) had a preexisting diagnosis of definite or presumed IIH. Of the remaining 30 patients, four (13.3%, 95%CI: [3.8;30.7]) had optic nerve head changes suggesting previously undiagnosed IIH, while one was newly diagnosed with definite IIH at initial consultation. One out of 29 patients with normal findings of the optic nerve head at presentation developed new onset papilledema following surgery (3.4%, 95%CI: [0.1;17.8]) and was ultimately diagnosed with definite IIH. Overall, the prevalence of definite IIH was 19.4% (95%CI: [8.2;36.0])., Conclusion: Striking demographic overlap exists between IIH patients and those with spontaneous CSF leak. Definite IIH was present in approximately 20% of our patients. However, its true prevalence is likely higher than identified by using classic criteria. We therefore hypothesize that an active CSF leak serves as an auto-diversion for CSF, thereby "treating" the intracranial hypertension and eliminating characteristic signs and symptoms at initial presentation.

We conducted a retrospective review to identify the characteristics of cerebrospinal fluid (CSF) leak in patients who had undergone septoplasty and in selected patients who had experienced a spontaneous CSF leak. CSF leak is a known but infrequently reported complication of septoplasty; to the best of our knowledge, only 4 cases have been previously published in the literature. A review of our institution's database revealed 3 cases of postseptoplasty CSF leak. We reviewed all the available data to look for any commonalities among these 7 cases. In addition, we reviewed 6 cases of spontaneous CSF leak selected from our database for the same purpose. For all patients, we noted the side of the cribriform plate defect, its size and, for the postseptoplasty cases, the interval between the septoplasty and the leak repair. Overall, we found that leaks were much more common on the right side than on the left. The sizes of the leaks in the 2 postseptoplasty groups were comparable (mean: 14.0 × 6.4 mm). The interval between septoplasty and leak repair ranged from 2.5 to 20 years in our cases and from 3 days to 22 weeks in the previously published cases. All 3 of the postseptoplasty patients in our database presented with clear rhinorrhea. Two of the 3 patients had meningitis; 1 of these 2 also had pneumocephalus. Of the 6 cases of spontaneous CSF leaks, 4 occurred on the right and 2 on the left; the average size of the defect was 5.8 mm in the greatest dimension. The finding that cribriform plate defects after septoplasty were typically right-sided likely reflects the prevalence of left-sided surgical approaches. Also, the fact that the defects were larger in the postseptoplasty cases than in the spontaneous cases is likely attributable to the torque effect toward the thin skull base that occurs when the perpendicular plate is twisted during septoplasty.