Middle fossa floor dehiscence (MFFD) has distinctly different presentations that include a range of syndromes and clinical symptoms. Middle fossa floor compromise can be spontaneous or occur secondary to accompanying disease processes, such as trauma, otitis media, inflammation, malignancy, cholesteatoma, and radiation therapy. The etiopathogenesis of spontaneous MFFD is widely debated, and several theories have been proposed. Spontaneous anterior cranial floor dehiscence and CSF leakage have been associated with empty sella syndrome and benign intracranial hypertension.18 These factors are similarly associated with a female predominance and an increased body mass index (BMI), which have also been associated with MFFD. 3,5,9,12 Other proposed hypotheses include congenital erosion of the middle fossa floor during pneumatization of the middle ear and mastoid air cells, 14,19 as well as atrophy over years of normal CSF pulsation.16 Surgical repair has been described 7,12,13 and includes a range of procedures involving resection of the encephalocele, dural repair, and plugging of the bony dehiscence.A wide range of clinical presentations has been de- obJect Middle fossa floor dehiscence (MFFD) can present as multiple syndromes depending on dehiscence location, tissue herniation, and dural integrity. The authors propose a classification system for MFFD with the potential to guide clinical decision making. methods A retrospective analysis of the electronic medical records (years 1995-2012) of patients who had undergone temporal craniotomy for the surgical repair of an MFFD syndrome at a single institution was undertaken. Reviewed data included demographic, operative, presentation, and outcome details. Middle fossa floor dehiscence was classified as follows: Class A, bony dehiscence without herniation of the brain and/or meninges; Class B, herniation of the brain and/or meninges through the middle fossa floor without CSF leakage; Class C, dehiscence with CSF leakage without meningitis; or Class D, dehiscence with meningitis. results Fifty-one patients, 22 males and 29 females, were included in the analysis. The mean age was 48.7 ± 15.5 years, mean body mass index was 32.65 ± 6.86 kg/m 2 , and mean symptom duration was 33 ± 42 months. Seven patients underwent repeat surgery for symptomatic recurrence; therefore, there were 58 surgical encounters. Repair included bony reconstruction with hydroxyapatite with or without resection of encephaloceles and/or repair of a dural defect. According to the MFFD classification system described, 15, 8, 27, and 8 cases were categorized as Class A, B, C, and D, respectively. The prevalence of hearing loss was 87%, 63%, and 70% in Classes A, B, and C, respectively. Vestibular symptoms were more prevalent in Class A. Seven patients reported persistent symptoms at the last follow-up. Transient complications were similar in each classification (13%-25%), and a single permanent complication related to anesthesia was observed. There were no mortalities or severe neurological morbidities in th...