Meniere’s disease (MD), a syndromal inner ear disease, is commonly associated with a pathological accumulation of endolymphatic fluid in the inner ear, termed “idiopathic” endolymphatic hydrops (iEH). Although numerous precipitating/exacerbating factors have been proposed for MD, its etiology remains elusive. Here, using immunohistochemistry and in situ protein–protein interaction detection assays, we demonstrate mineralocorticoid-controlled sodium transport mechanisms in the epithelium of the extraosseous portion of the endolymphatic sac (eES) in the murine and human inner ears. Histological analysis of the eES in an extensive series of human temporal bones consistently revealed pathological changes in the eES in cases with iEH and a clinical history of MD, but no such changes were found in cases with “secondary” EH due to other otological diseases or in healthy controls. Notably, two etiologically different pathologies—degeneration and developmental hypoplasia—that selectively affect the eES in MD were distinguished. Clinical records from MD cases with degenerative and hypoplastic eES pathology revealed distinct intergroup differences in clinical disease presentation. Overall, we have identified for the first time two inner ear pathologies that are consistently present in MD and can be directly linked to the pathogenesis of EH, and which potentially affect the phenotypical presentation of MD. Electronic supplementary material The online version of this article (10.1007/s00401-018-1927-7) contains supplementary material, which is available to authorized users.

Hypothesis: The vestibular aqueduct (VA) in Menière's disease (MD) exhibits different angular trajectories depending on the presenting endolymphatic sac (ES) pathology, i.e., 1) ES hypoplasia or 2) ES degeneration. Background: Hypoplasia or degeneration of the ES was consistently found in inner ears affected by MD. The two etiologically distinct ES pathologies presumably represent two disease “endotypes,” which may be associated with different clinical traits (“phenotypes”) of MD. Recognizing these endotypes in the clinical setting requires a diagnostic tool. Methods: 1) Defining the angular trajectory of the VA (ATVA) in the axial plane. 2) Measuring age-dependent normative data for the ATVA in postmortem temporal bone histology material from normal adults and fetuses. 3) Validating ATVA measurements from normative CT imaging data. 4) Correlating the ATVA with different ES pathologies in histological materials and CT imaging data from MD patients. Results: 1) The ATVA differed significantly between normal adults and MD cases with ES degeneration, as well as between fetuses and MD cases with ES hypoplasia; 2) a strong correlation between ATVA measurements in histological sections and CT imaging data was found; 3) a correlation between the ATVA, in particular its axial trajectory in the opercular region (angle α exit ), with degenerative ( α exit < 120°) and hypoplastic ES pathology ( α exit > 140°) was demonstrated. Conclusion: We established the ATVA as a radiographic surrogate marker for ES pathologies. CT-imaging-based determination of the ATVA enables endotyping of MD patients according to ES pathology. Future studies will apply this method to investigate whether ES endotypes distinguish clinically meaningful subgroups of MD patients.

Two histopathological subtypes of Meniere's disease (MD) were recently described in a human post-mortem pathology study. The first subtype demonstrated a degenerating distal endolymphatic sac (ES) in the affected inner ear (subtype MD-dg); the second subtype (MD-hp) demonstrated an ES that was developmentally hypoplastic. The two subtypes were associated with different clinical disease features (phenotypes), suggesting that distinct endotype-phenotype patterns exist among MD patients. Therefore, clinical endotyping based on ES pathology may reveal clinically meaningful MD patient subgroups. Here, we retrospectively determined the ES pathologies of clinical MD patients (n = 72) who underwent intravenous delayed gadolinium-enhanced inner ear magnetic resonance imaging using previously established indirect radiographic markers for both ES pathologies. Phenotypic subgroup differences were evidenced; for example, the MD-dg group presented a higher average of vertigo attacks (ratio of vertigo patterns daily/weekly/other vs. monthly, MD-dg: 6.87: 1; MD-hp: 1.43: 1; p = 0.048) and more severely reduced vestibular function upon caloric testing (average caloric asymmetry ratio, MD-dg: 30.2% ± 30.4%; MD-hp: 13.5% ± 15.2%; p = 0.009), while the MD-hp group presented a predominantly male sex ratio (MD-hp: 0.06:1 [f/m]; MD-dg: 1.2:1 [f/m]; p = 0.0004), higher frequencies of bilateral clinical affection (MD-hp: 29.4%; MD-dg: 5.5%; p = 0.015), a positive family history for hearing loss/vertigo/MD (MD-hp: 41.2%; MD-dg: 15.7%; p = 0.028), and radiographic signs of concomitant temporal bone abnormalities, i.e., semicircular canal dehiscence (MD-hp: 29.4%; MD-dg: 3.6%; p = 0.007). In conclusion, this new endotyping approach may potentially improve the diagnosis, prognosis and clinical decision-making for individual MD patients.