Epitympanum Volume and Tympanic Isthmus Area in Temporal Bones With Retraction Pockets

Transcription

1 The Laryngoscope VC 2016 The American Laryngological, Rhinological and Otological Society, Inc. Epitympanum Volume and Tympanic Isthmus Area in Temporal Bones With Retraction Pockets Rafael da Costa Monsanto, MD; Henrique Furlan Pauna, MD; Serdar Kaya, MD; Omer Hızlı, MD; Geeyoun Kwon, PhD; Michael M. Paparella, MD; Sebahattin Cureoglu, MD Objectives/Hypothesis: To compare the volume of the epitympanic space, as well as the area of the tympanic isthmus, in human temporal bones with retraction pockets to those with chronic otitis media without retraction pockets and to those with neither condition. Study Design: Comparative human temporal bone study. Methods: We generated a three-dimensional model of the bony epitympanum and measured the epitympanic space. We also compared the area of the tympanic isthmus. Results: The mean total volume of the epitympanum was mm 3 in the retraction pocket group, mm 3 in the chronic otitis media group, and mm 3 in the neither condition group. The mean volume of the anterior, lateral, and medial compartments in temporal bones in the retraction pocket group was significantly smaller than in the two control groups (P < 0.05). Total epitympanic volume was also significantly smaller in the retraction pocket group than in both control groups (P < 0.05). The mean area of the tympanic isthmus was significantly smaller in the retraction pocket group ( mm 2 ) than in the chronic otitis media group ( mm 2 ) or the neither condition group ( mm 2 )(P < 0.05). Conclusion: Our data indicate that temporal bones with retraction pockets have a smaller volume bony epitympanum and a smaller tympanic isthmus area as compared with temporal bones from both control groups. The smaller volume tympanic isthmus in the retraction pocket group may suggest that a blockage in the aeration pathways to the epitympanum could create dysventilation, resulting in negative pressure and ultimately in retraction pockets and cholesteatomas. Key Words: Epitympanum, tympanic isthmus, human temporal bone, retraction pockets, middle ear ventilation, 3D reconstruction. Level of Evidence: NA Laryngoscope, 126:E369 E374, 2016 From the Department of Otolaryngology Head and Neck Surgery, University of Minnesota (R.DC.M., H.F.P., S.K., O.H., G.K., S.C.); the Paparella Ear Head and Neck Institute (M.M.P.), Minneapolis, Minnesota, U.S.A.; the Department of Otolaryngology Head and Neck Surgery, Banco de Olhos de Sorocaba Hospital (R.dC.M.), Sorocaba; the Department of Otolaryngology Head and Neck Surgery, University of Campinas (H.F.P.), Campinas, S~ao Paulo, Brazil; the Department of Otolaryngology Head and Neck Surgery, Gebze Fatih State Hospital (S.K.), Gebze, Kocaeli, Turkey; and the Giresun A. Ilhan Ozdemir State Hospital (O.H.), Giresun, Turkey Editor s Note: This Manuscript was accepted for publication January 29, Source of Funding: This project was funded by NIH NIDCD U24 DC011968, International Hearing Foundation, Starkey Hearing Foundation, Lions 5m International and Scientific and Technological Research Council of Turkey (TUBITAK) Scholarship. The authors have no other funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Sebahattin Cureoglu, MD, Associate Professor, Co-Director, Otopathology Laboratory, Department of Otolaryngology, University of Minnesota, Minneapolis, Minnesota, th St. SE, Lions Research Building, Room 210, Minneapolis, MN cureo003@umn.edu DOI: /lary INTRODUCTION The ventilation of the middle ear has been extensively studied. In 1946, Chatellier and Lemoine 1 introduced the concept of the interatticotympanic diaphragm, which was later renamed the epitympanic diaphragm 2 : it consisted of the malleal ligamental fold, the posterior incudal ligamental fold, the tensor fold, the lateral incudomalleal fold, and the lateral surfaces of the epitympanic portion of the malleus and incus. Since the early studies conducted by Prussak in 1867, 3 much progress has been achieved on this matter. Recently, Palva et al. initiated the investigation of middle ear ventilation patterns and their association with ear anatomy and middle ear disease. 2,4 6 Blockage of the aeration pathways especially the narrow route via the posterior pouch to Prussak s space, which is especially vulnerable to obstruction in patients with chronic otitis media can compromise drainage of mucus and clearance of infection, thereby preventing air from flowing to the upper compartments of the middle ear and causing dysventilation. 2,5,6 Retractions of Shrapnell s membrane represent a likely forerunner of a retraction-pocket cholesteatoma, but studies have shown that even small retractions of the tympanic membrane can hide a cholesteatoma. 6,7 The middle ear is divided into two compartments by the epitympanic diaphragm, which separates the epitympanum from the other middle ear structures; the main pathway for ventilation of the epitympanum through the diaphragm is the tympanic isthmus. The anatomic boundaries of the tympanic isthmus are as follows: the tensor tympani tendon anteriorly, the medial portion of the posterior incudal ligament posteriorly, the attic bone medially, and the body and short process of the incus and the head of the malleus laterally. 8,9 E369

2 Fig. 1. A representative horizontal human temporal bone section, showing epitympanic compartments and landmarks (hematoxylin and eosin). A 5 anterior compartment; L 5 lateral compartment; M 5 medial compartment; M/I 5 malleus/incus compartment; P 5 posterior compartment; 1 5 the origin of the anterior tympanic spine of the lateral attic bone; 2 5 the posterior point of the cochleariform process; 3 5 the most anterior point of the head of the malleus; 4 5 the lateral side of the aditus antrum; 5 5 the medial side of the aditus antrum; 6 5 the most posterior point of the incus body. Segmentation performed with Amira software (Amira, Visualization Sciences Group, Bordeaux, France). Shirai et al. 8 and Palva et al. 2,4 6,9 claimed that blockage of the tympanic isthmus can cause inadequate ventilation of the upper middle ear area because the aeration pathway from the Eustachian tube leads directly to the mesotympanic and hypotympanic spaces. In patients with a shortened tympanic isthmus, studies have reported pathologic changes such as cholesteatomas, tympanosclerosis, a swollen mucosa, and mucoid secretions. 10,11 Palva et al. 2,4 6 asserted that, even if Prussak s space is an essential part of the epitympanum for aeration and drainage, it can be blocked or obliterated without affecting the function of the compartments superior to the epitympanic diaphragm or of other middle ear structures, including the mastoid air system. Nonetheless, blockage of the tympanic isthmus can cause extensive pathology and impair middle ear function. Views differ regarding the volumetric impact of retraction pockets on the epitympanum and on its relationship with the area of the tympanic isthmus. To our knowledge, no volumetric studies have been performed. The purpose of this study was to compare the volume of the epitympanic space, and of the area of the tympanic isthmus, in archived human temporal bones from three groups of deceased patients: those with retraction pockets, those with chronic otitis media without retraction pockets, and those with neither condition. MATERIALS AND METHODS Samples From our archived human temporal bone collection at the University of Minnesota, we selected 34 temporal bones from 32 deceased patients with retraction pockets. Exclusion criteria were as follows: a cholesteatoma destroying the bony boundaries, otologic surgery, tumors or metastatic tumors, bony erosion, and processing artifacts. Eighteen temporal bones were then excluded, and thus 16 temporal bones were finally included in our study. We also formed two control groups: 1) a chronic otitis media group (16 temporal bones from 16 age- and sex-matched deceased patients with chronic otitis media without retraction pockets) and 2) a neither condition group (16 temporal bones from 16 age- and sex-matched deceased patients with neither chronic otitis media nor retraction pockets). For the chronic otitis media group, we included temporal bones with any sign of chronic inflammatory changes in the middle ear cleft, such as cholesterol granuloma, fibrosis, or fibrocystic structures. In the retraction pocket group, we observed effusion in nine (56%) of the temporal bones; in the chronic otitis media group, we observed effusion in 12 (75%) of the temporal bones. The human temporal bones had been previously removed at autopsy, fixed in 10% formalin, decalcified, and embedded in celloidin. Each temporal bone was serially sectioned in the horizontal plane at a thickness of 20 lm. Every 10th section was stained with hematoxylin and eosin, and then evaluated using light microscopy. Anatomic Boundaries For this study, the superior limit of the epitympanum was the first temporal bone slide containing the malleus, incus, and incudomalleal joint. The area between the ossicles and the tegmen was not included in our study because the temporal bone sections that included this space were not available for most of the cases. We therefore selected the incudomalleal joint as a landmark because it was the most superior structure present in all of the temporal bones. The inferior limit was outlined as the cut in which the body of the incus disappeared. We examined all slides in between those limits. We divided the epitympanum into five different compartments and defined six bony landmarks (Fig. 1). The five compartments are as follows: 1) the anterior (A), the triangular space with vertices 1, 2, and 3; 2) the lateral (L), the space surrounded by E370

3 Statistical Analysis Results are presented as the mean 6 standard deviation. For all three of our groups, we calculated and compared the volumes of every compartment, as well as the total epitympanic volume. In addition, we calculated and compared the narrowest area of the tympanic isthmus. To confirm the normal distribution of data in each group, we used the Kolmogorov-Smirnov test (P 5 0.2). We also used the independent t test in the SPSS 22.0 software for Windows (SPSS Inc., Chicago, IL). Findings were considered statistically significant when P values were less than Fig. 2. A three-dimensional model generated for volume calculation for epitympanic compartments. A 5 anterior compartment; L5 lateral compartment; M 5 medial compartment; M/I 5 malleus/incus compartment; P 5 posterior compartment. vertices 1, 3, 4, and 6; 3) the posterior (P), the triangular space with vertices 4, 5, and 6; 4) the medial (M), the space surrounded by vertices 2, 3, 5, and 6; and 5) the malleus (M)/incus (I), the epitympanic portion of the M and I. The six bony landmarks are as follows: 1) the origin of the anterior tympanic spine of the lateral attic bone; 2) the posterior point of the cochleariform process; 3) the most anterior point of the head of the malleus; 4) the lateral side of the aditus antrum; 5) the medial side of the aditus antrum; and 6) the most posterior point of the incus body. Then, we generated a three-dimensional (3D) model of the five compartments (Fig. 2). We defined the tympanic isthmus boundaries as follows: the tensor tympani tendon, the medial portion of the posterior incudal ligament, the attic bone, the body and short process of the incus, and the head of the malleus (Fig. 3). Our measurements of the tympanic isthmus included only the aerated space (not any fibrous, granulation, or soft tissue). RESULTS We initially selected 34 archived human temporal bones with retraction pockets but excluded 18 of them for the following reasons: four had undergone previous ear surgeries; 11 had cholesteatomas that destroyed the bony boundaries; and three had processing artifacts. Thus, our final retraction pocket group included 16 temporal bones (12 males and 4 females); their mean age was years (range, 8 88 years). Of the 16 temporal bones included in our final retraction pocket group, 13 had other associated pathologies: 10 had chronic otitis media; two had otosclerosis; and one had purulent otitis media without signs of Volumetric Analysis To scan all of the selected slides, we used a high-resolution scanner (PathScan Enabler IV; Meyer Instruments, Houston, TX). Then, we transferred the resulting images to a 3D reconstruction software (Amira, Visualization Sciences Group, Bordeaux, France; and Zuse Institute, Berlin, Germany), in which we generated a 3D model of the five compartments (Fig. 2). To calculate the total tympanic volume, we used the sum of the measurements from each compartment. To measure the area of the tympanic isthmus, we examined all slides from the lower limit to the upper limit of the tensor tympani tendon. To calculate the volume of the epitympanum and the area of the tympanic isthmus area of each temporal bone, we used the material statistic feature of the Amira software (Visualization Sciences Group). In the retraction pocket group, we also classified the temporal bones according to Sade and Berco classification, 12 based on the following graduated categories: 1) simple retraction of the eardrum, not in contact with the incus or stapes; 2) tympanic membrane retraction in contact with the incus or stapes (tympanoincudopexy); 3) tympanic membrane in contact with the promontory wall (not adhered to it); and 4) tympanic membrane adhered to the promontory (adhesive otitis media). Fig. 3. A representative horizontal human temporal bone section, showing the tympanic isthmus (hematoxylin and eosin). (A) Temporal bone from the retraction pocket group. Yellow area 5 tympanic isthmus. (B) Temporal bone from the neither condition group. Blue area 5 tympanic isthmus. E371

4 TABLE I. Mean Bony Volume of Epitympanic Compartments and Total Epitympanic Volume in the Retraction Pocket Group and the Neither Condition Group. Compartment Retraction Pocket Neither Condition P Value Volume(mm 3 ), mean 6 SD Anterior Posterior Lateral Medial Malleus-Incus Total TABLE II. Mean Bony Volume of Epitympanic Compartments and Total Epitympanic Volume in the Retraction Pocket Group and the Chronic Otitis Media Group. Compartment Retraction Pocket Chronic Otitis Media P Value Volume(mm 3 ), mean 6 SD Anterior Posterior Lateral Medial Malleus-Incus Total chronic pathology in the middle ear cleft. To grade all 16 temporal bones in our final retraction pocket group, we used Sade classification 12 : two (12.5%) were grade I; 10 (62.5%) were grade II; three (18.75%) were grade III; and one (6.25%) was grade IV. The mean epitympanic bony volume for each grade was as follows: 44.2 mm 3, grade I; mm 3, grade II; mm 3, grade III; and 32.2 mm 3, grade IV. Our chronic otitis media group included 16 temporal bones (10 males and 4 females) Their mean age was years (range, 8 88 years). Our neither condition group included 16 temporal bones (10 males and 6 females); their mean age was years (range, years). Both of these control groups were age- and sex-matched with the retraction pocket group. Epitympanum The mean volume of the anterior, lateral, and medial compartments in temporal bones in the retraction pocket group was significantly smaller than in each of our two control groups (P < 0.05 for both comparisons); total epitympanic volume was also significantly smaller in the retraction pocket group than in both control groups (P < 0.05). We found no significant difference in the mean volume of the posterior compartment between the retraction pocket group and each of our two control groups (P > 0.5) (Tables I and II). We also found no significant difference in either the mean volume of the epitympanic compartments or the total epitympanic volume between our two control groups (P > 0.05) (Table III). TABLE III. Mean Bony Volume of Epitympanic Compartments and Total Epitympanic Volume in the Chronic Otitis Media Group and the Neither Condition Group. Compartment Neither Condition Chronic Otitis Media P Value Volume(mm 3 ), mean 6 SD Anterior Posterior Lateral Medial Malleus-Incus Total E372

5 TABLE IV. Mean Narrowest Area of the Tympanic Isthmus in Each Group and Comparisons Between Groups. Group Area (mm 2 ), mean 6 SD Retraction pocket group Neither condition group Chronic otitis media group Comparisons P Retraction pocket group vs neither condition group Retraction pocket group vs chronic otitis media group Neither condition group vs chronic otitis media group Tympanic Isthmus The mean narrowest area of the tympanic isthmus was significantly smaller in the retraction pocket group ( mm 2 ) than in the chronic otitis media without retraction pocket group ( mm 2 ; P ) or the neither condition group ( mm 2 ; P ) (Table IV). But we found no significant difference in the mean narrowest area of the tympanic isthmus between our two control groups (P ). In the retraction pocket group, we observed that the area of the tympanic isthmus was reduced mainly because of soft tissue not directly related to the retraction pocket. DISCUSSION According to the epitympanic diaphragm concept, the upper compartment of the middle ear is aerated only through the tympanic isthmus if the tensor fold and the lateral incudo malleal fold are complete 1,4,5,13 If this area has any blockage, the only gas exchange would come from the mucosa of the mastoid cells. 14 Palva et al. 2,4 6 and Marchioni et al. 15 claimed that anatomic factors can compromise aeration of the epitympanum even with a normally functioning Eustachian tube. In our study, the area of the tympanic isthmus in our retraction pocket group was significantly smaller than in each of our two control groups (i.e., our chronic otitis media group and our neither condition group). This finding is in accord with an endoscopic study by Marchioni et al., 15 in which patients with attic disease had a significantly higher rate of isthmus blockage than those without attic disease. The fact that the mean area of the aeration pathway leading to the epitympanum is smaller in patients with retraction pockets (vs. with patients with chronic otitis media without retraction pockets) supports the idea that selective dysventilation of the epitympanum can create negative middle ear pressure inducing atelectasis, retraction pockets, and ultimately cholesteatomas, 16 even though the pathophysiology of this process is yet not fully understood In addition, we found that the bony spaces of the anterior, lateral, and medial epitympanic compartments were significantly smaller in our retraction pocket group than in each of our two control groups (i.e., our chronic otitis media group and our neither condition group). This finding supports the general assumption, and observations, of other previously published studies. 3,4,16 The smaller epitympanic volume that we found in our retraction pocket group is important because the decreased aerated space of the epitympanum and the consequent negative pressure in the middle ear cavity have been described as the genesis of a cholesteatoma. 3,4,13,14,16 Marchioni et al. 20 observed that cholesteatomas limited to the attic can be associated with a reduced volume of the anterior epitympanum and suggested that the small anterior epitympanic cavities might be proof of a selective attic dysventilation. In addition, Marchioni et al. 20 as well as Miura et al. 16 reported that negative middle ear pressure can induce an atelectasis of the tympanic membrane, which in turn results in retraction pockets of the pars tensa, mainly in the posterior superior quadrant or at the level of the pars flaccida in the epitympanic region. Subsequently, such an atelectasis can lead to adhesive otitis media that adheres to the substrate of the cholesteatoma. 16 Marchioni et al. and Miura et al. described this process as selective epitympanic dysventilation, which can decrease aeration of the epitympanum even with a functional Eustachian tube. 16,20 Moreover, in our study, we had a high exclusion rate of temporal bones with retraction pockets because of an associated cholesteatoma destroying the bone boundaries. The decreased bony volume in our retraction pocket group might parallel the silent sinus syndrome, in which negative pressure in the maxillary sinus can decrease the sinus volume; however, the design of our study does not allow us to prove or disprove any possible parallel. Our results provide insights regarding the anatomy and the causes of epitympanic hypoxia that lead to the genesis of retraction pockets and cholesteatomas. If an alternative aeration pathway can be created, or if the tympanic isthmus can be surgically reopened, the pathologic continuum that ultimately culminates in cholesteatomas might be stopped. CONCLUSION Our data indicate that temporal bones with retraction pockets have a smaller volume bony epitympanum and a smaller tympanic isthmus area as compared with temporal bones from patients with chronic otitis media without retraction pockets and from patients with neither condition. The smaller volume tympanic isthmus in the retraction pocket group suggests that a blockage in the aeration pathways to the epitympanum could create dysventilation, resulting in negative pressure and ultimately in retraction pockets and cholesteatomas. BIBLIOGRAPHY 1. Chatellier HP, Lemoine J. Le diaphragme inter-attico-tympanique du nouveaune. Ann Otolaryngol Chir Cervico-Fac (Paris) 1946;13: Palva T, Johnsson LG. Epitympanic compartment surgical considerations: reevaluation. Am J Otol 1995;16: Prussak A. Zur Anatomie des menschlichen Trommelfells. Arch Ohrenheilkd 1867;3: Palva T, Ramsay H. Incudal folds and epitympanic aeration. Am J Otol 1996;17: Palva T, Ramsay H, Bohling T. Tensor fold and anterior epitympanum. Am J Otol 1997;18: Palva T, Johnsson LG, Ramsay H. Attic aeration in temporal bones from children with recurring otitis media: tympanostomy tubes did not cure disease in Prussak s space. Am J Otol 2000;21: E373

6 7. Lee JH, Hong SM, Kim CW, Park YH, Baek SH. Attic cholesteatoma with tiny retraction of pars flaccida. Auris Nasus Larynx 2015;42: Shirai K, Schachern PA, Schachern MG, Paparella MM, Cureoglu S. Volume of the epitympanum and blockage of the tympanic isthmus in chronic otitis media: a human temporal bone study. Otol Neurotol 2015; 36: Palva T, Ramsay H. Chronic inflammatory ear disease and cholesteatoma: creation of auxiliary attic aeration pathways by microdissection. Am J Otol 1999;20: Miyajima I, Honda Y. The clinical significance of the tympanic isthmus related to the development of cholesteatoma. Auris Nasus Larynx 1985;12: Proctor B. The development of the middle ear spaces and their surgical significance. J Laryngol Otol 1964;78: Sade J, Berco E. Atelectasis and secretory otitis media. Ann Otol Rhinol Laryngol 1976;85: Marchioni D, Alicandri-Ciufelli M, Molteni G, Artioli FL, Genovese E, Presutti L. Selective epitympanic dysventilation syndrome. Laryngoscope 2010;120: Hergils L, Magnuson B. Morning pressure in the middle ear. Arch Otolaryngol 1985;111: Marchioni D, Mattioli F, Alicandri-Ciufelli M, Presutti L. Prevalence of ventilation blockages in patients affected by attic pathology: a casecontrol study. Laryngoscope 2013;123: Miura M, Takahashi H, Honjo I, Hasebe S, Tanabe M. Influence of the gas exchange function through the middle ear mucosa on the development of sniff-induced middle ear diseases. Laryngoscope 1998;108: Wells MD, Michaels L. Role of retraction pockets in cholesteatoma formation. Clin Otolaryngol Allied Sci 1983;8: Knutsson J, Bagger-Sjoback D, von Unge M. Structural tympanic membrane changes in secretory otitis media and cholesteatoma. Otol Neurotol 2011;32: Ars BM. Tympanic membrane retraction pockets. Etiology, pathogeny, treatment. Acta Otorhinolaryngol Belg 1991;45: Marchioni D, Grammatica A, Alicandri-Ciufelli M, Aggazzotti-Cavazza E, Genovese E, Presutti L. The contribution of selective dysventilation to attical middle ear pathology. Med Hypotheses 2011;77: E374