Keywords
cholesteatoma - chronic otitis media - mongolian gerbils - experimental models - otoendoscopy
Introduction
Cholesteatoma is a pathology with high morbidity and high costs to the health system, with significant risk for complications, both intratemporal and extratemporal.[1] Over the past decades, our understanding of the disease has evolved, especially in clinical and pathophysiological aspects. Many gaps persist in our understanding of its pathogenesis; however, several theories have been proposed. Clinical and experimental evidence[2]
[3]
[4]
[5]
[6] has demonstrated that the development of this pathology may follow a sequence of progressive tympanic retraction, pocket formation, loss of self-cleaning properties, keratin accumulation, and cholesteatoma formation. Nevertheless, some previous studies have failed to clearly demonstrate this evolution.[7]
[8]
In the effort to provide adequate answers to lingering knowledge gaps, and considering the difficulty in performing histological and long-term follow-up studies of this disease in humans, the use of animal models for this purpose is highly valuable. Among these, Mongolian gerbils have emerged as one of the most operational species to meet this need, because together with humans, they are the only species to develop cholesteatomas spontaneously. According to Chole et al.,[9] cholesteatomas in Mongolian gerbils exhibit several characteristics similar to those of humans, both macroscopically and microscopically. It has previously been demonstrated that, with experimental obliteration of the external auditory canal or eustachian tube (ET), cholesteatoma can be induced in these animals, with characteristics very similar to those that occur spontaneously.[10]
[11]
[12]
[13]
[14] The obliteration of the external auditory canal has been used more frequently in prior studies, since this technique allows the development of cholesteatoma in almost all the animals that undergo the procedure.[10]
[15] Although this model has provided us added knowledge of cholesteatoma in general, the pathogenesis involved in the formation of cholesteatoma when this technique is used appears to differ from that observed in humans.
The present study aims to evaluate, through serial otoendoscopies, the development and characteristics of pars flaccida retraction pocket and cholesteatoma in Mongolian gerbils after obliteration of the ET and compare it with the control group.
Materials and Methods
The present research was performed at the Animal Experimentation Unit (AEU) of the Hospital de Clínicas de Porto Alegre, between March 2016 and November 2017. All procedures were performed according to the regulations of the Committee on Animal Ethics (CEUA, in the Portuguese acronym) of this institution (16–0010).
Obliteration of the ET in Mongolian gerbils was performed by cauterization of the tubal ostium bilaterally. The animals underwent the procedure while under general anesthesia using inhaled isoflurane, intraperitoneal tramadol (30 mg/kg), and bupivacaine (0.5%, 4 mg/kg). Cauterization was performed using monopolar cautery with a needle tip inserted in a transpalatal approach, without direct visualization of the tubal ostia. Based on an anatomical study by Wolfman et al,[16] the needle was introduced ∼ 5 mm posterior to the transition between the hard palate and soft palate of the animal, maintained for 3 seconds, and angled ∼ 30° to the right initially and, afterwards, using the same angulation, to the left ([Fig. 1]). In the immediate postoperative period, analgesia included intramuscular dipyrone (500 mg/kg) in animals that underwent the intervention. Maintenance of analgesia was performed using tramadol hydrochloride (10 mg/kg) every 12 hours intraperitoneally in animals exhibiting clinical signs of pain.
Fig. 1 Eustachian tube obliteration.
Sixteen weeks was used as a follow-up time ([Fig. 2]), established by Wolfmann et al.,[16] who reported that 75% of the animals developed cholesteatoma by histology after ET obstruction during this follow-up period. To evaluate the natural history of potential pathology, all Mongolian gerbils sequentially underwent bilateral otoendoscopy (Karl-Storz optical fiber, 0°, 2.8mm, and 18cm; Karl Storz SE & Co. KG, Tuttlingen, Germany) under general anesthesia at zero (to exclude previous cholesteatoma), 1, 4, 8, 12, and 16 weeks ([Fig. 3]). Otoendoscopies were evaluated for the presence of normal tympanic membrane ([Fig. 4a]), pars flaccida retraction pocket ([Fig. 4b]) or cholestatoma ([Fig. 4c]).
Fig. 2 Study flow chart.
Fig. 3 Otoendoscopy.
Fig. 4 Evolution of otoendoscopy in “animal 8” - intervention group. a) week zero – normal; b) week 1–pars flaccida retraction pocket; c) week 8–cholesteatoma.
We considered cholesteatoma through otoendoscopy in two possibilities: presence of a plug of keratin filling a pars flaccida retraction pocket; and keratin accumulation adjacent to the tympanic membrane and partially or totally obliterating the external auditory canal.
The sample calculation was performed using an estimated incidence of developing cholesteatoma of ∼ 45% in the control group and of 85% in the intervention group, as reported in the literature.[9]
[11]
[16] The incidence of cholesteatoma formation in each group was reported as n (%). Differences in the incidences between the groups were assessed using the chi-squared test.
Results
Forty Mongolian gerbils > 3 months of age of both genders were included in the study and divided into 2 groups (intervention and control) of 20 animals each ([Fig. 2]). Five losses occurred throughout the study: three in the control group and two in the intervention group. However, one was included in the analysis because it was followed-up for at least 8 weeks, resulting in 72 ears studied. The groups were divided without randomization and did not demonstrate statistically significant differences in their baseline characteristics ([Table 1]).
Table 1
Baseline characteristics
|
Intervention
|
Control
|
|
|
Age
|
5.3 months
|
5.2 months
|
|
|
Weight
|
88.8 g
|
81.8 g
|
p = 0.08
|
|
Gender (M/F)
|
17/3
|
17/3
|
|
In the first two evaluations (week zero, week 1), no cholesteatoma was visualized by otoendoscopic evaluation. In week zero, we verified the presence of pars flaccida retraction pocket in 2 out of 38 (5.2%) ears in the intervention group and in 1 out of 34 (2.9%) ears in the control group (p = 0.654). In week 1, the occurrence of pars flaccida retraction pocket was observed in 6 out of 38 (15.7%) ears in the intervention group and in 3 out of 34 (8.8%) ears in the control group (p = 0.372).
In week 4, the presence of cholesteatoma was observed in 5 out of 38 (13.1%) ears in the intervention group and in 1 out of 34 (2.9%) ears in the control group (p = 0.117). Considering the ears in the intervention group that developed cholesteatoma this week, 3 out of 5 (60%) presented pars flaccida retraction pocket in the previous evaluation. Pars flaccida retraction pocket was seen in 9 out of 38 (23.6% [6 new/3 already present in the previous evaluation]) ears in the intervention group and in 6 out of 34 (17.6% [3 new/3 previous]) ears in the control group (p = 0.528).
In the 8th week, 8 out of 38 (21.05%) ears in the intervention group showed cholesteatoma, and 2 out of 34 (5.8%) ears in the control group (p = 0.063). In this week, we identified 3 new cholesteatomas compared with the previous evaluation, 2 out of 3 (66%) presented pars flaccida retraction pocket at week 4. The presence of pars flaccida retraction was seen in 9 out of 38 (23.6% [2 new/5 previous]) ears in the intervention group and in 6 out of 34 (17.6% [1 new/5 previous]) in the control group (p = 0.528).
In week 12, we observed the presence of cholesteatoma in 10 out of 38 (26.3%) ears in the intervention group and in 4 out of 34 (11.7%) ears in the control group (p = 0.119). In the intervention group, 2 new cholesteatomas emerged, 1 of which (50%) presented pars flaccida retraction pocket in the previous evaluation. Pars flaccida retraction pocket was observed in 12 out of 38 (31.5% [4 new/8 previous]) ears in the intervention group and in 6 out of 34 (17.6% [1 new/5 previous]) ears in the control group (p = 0.172).
At the end of the 16-week follow-up, cholesteatoma was present in 13 out of 38 (34.2%) ears in the intervention group and in 7 out of 34 (20.6%) in the control group, a difference that was not statistically significant (p = 0.197). In this final week, we identified 3 new cholesteatomas compared with the previous evaluation, and 2 out of 3 (66%) presented pars flaccida retraction pocket at week 12. When we consider cholesteatoma and its potential precursor, pars flaccida retraction pocket, combined, these changes were observed in 23 out of 38 (60.8%) in the intervention group and in 11 out of 34 (32.3%) in the control group, a difference with statistical significance (p = 0.016). Using serial otoendoscopic evaluation, the development of cholesteatomas occurred, on average, at 10.5 weeks in animals in the intervention group compared with 14.2 weeks in the control group (p = 0.052).
Over the 16 weeks, 13 cholesteatomas developed in the intervention group and 7 in the control group ([Fig. 5]). In the intervention group, 8 out of 13 (61.5%) cholesteatomas developed in ears that presented pars flaccida retraction pockets in the immediately previous evaluation, compared with 4 out of 7 (57.1%) in the control group (p = 0.744).
Fig. 5 Findings of each ear studied during the 16-week follow-up.
Discussion
An animal model of cholesteatoma has value in the study of pathogenesis. The Mongolian gerbil has been considered a suitable model of study, since it is the only species, other than humans, prone to spontaneous cholesteatoma. The majority of previous studies is based on observations made after closure of the external auditory canal to increase the incidence of cholesteatoma formation, followed by serial sacrifice of the animals and histological analysis of the bulla. Although these studies have shed light into many facets of this intriguing disease, it is often difficult to fully translate their findings into the human clinical scenario. Particularly, one of the most accepted theory for the development of cholesteatoma in humans is related to malfunction of the ET, with the consequent generation of negative pressure in the middle ear, inflammation, progressive retraction of the tympanic membrane and, finally, cholesteatoma formation.[17] To follow this progression, the clinician can best asses this with serial inspection of the affected ear with the aid of otoscopes, microscopes or endoscopes. In other words, experimentally-induced cholesteatomas obtained by suturing the ear canal are critically dissimilar to natural occurring cholesteatomas. Similarly, relying on histopathology to make the clinical diagnosis of cholesteatoma is not consistent with real-world clinical care.
Taking this into account, we chose ET obliteration as a technique to induce cholesteatoma in the present study. We create an experimental model that could simulate, as close as possible, the development and the diagnosis of cholesteatomas in humans. To do so, we have decided to induce the development of the disease through ET obliteration and follow-up the animals with serial otoendoscopies. This method allows us to analyze evolutionary characteristics, formation pathways and cholesteatoma progression over time in the same animal, a scenario not possible using histology. To our knowledge, this is the first time that endoscopes have been used to monitor cholesteatoma in Mongolian gerbils. Using this methodology, we found that ET obliteration showed a tendency to increase the incidence of cholesteatoma in Mongolian gerbils, although not achieving statistical significance. However, when we included pars flaccida retraction pockets in the analysis (a condition also closely related to ET obliteration and theoretically a potential precursor of cholesteatoma), we obtained a high combined incidence of these changes in the intervention group, with statistical significance when compared with the control group. The inclusion of these findings furthers the concept of a spectral progression to cholesteatoma.
The present study has some limitations. We faced difficulties in the complete visualization of the tympanic membrane through otoendoscopy in cases in which the external auditory canal was completely occluded by the accumulation of keratin. The impossibility of directly visualizing the tubal ostia during the cauterization procedure was another difficulty of the present study. We cauterized in an extended way (∼ 3 seconds) and bilaterally in an attempt to increase the chance of success. However, at no point in the study was there confirmation that the obliteration was really effective, which is an important limitation. Furthermore, we must be very careful when projecting results obtained in experimental studies on humans. In addition to the evident anatomical differences, Mongolian gerbils have a substantially higher incidence of cholesteatoma than humans, demonstrating a greater propensity for the development of this pathology.
Considering our findings in serial otoendoscopies, we hypothesize that ET obliteration and the consequent desestabilization of the middle ear (negative pressure and inflamatory process) seems to increase the incidence of cholesteatoma in Mongolian gerbils. This would occur either by pars flaccida retraction and gradual accumulation of keratin, followed by the probable loss of the self-cleaning ability of the retraction pocket, or by progressive accumulation of keratin in the lateral wall of the tympanic membrane with later advancement toward the middle ear and bulla. In addition, we observed that even in the control group there is a high incidence of cholesteatomas that presented pars flaccida retraction pocket in their immediately previous evaluation. Therefore, we must consider this formation pathway (retraction pocket – cholesteatoma) as responsible for at least part of the spontaneous cholesteatomas in these animals. Taking this into account, the present study corroborates the possibility of using Mongolian gerbils as an animal model to study also the formation pathways of cholesteatoma, and otoendoscopy proved to be an excellent tool for this purpose.
Conclusion
Over the 16 weeks follow-up, serial otoendoscopies enabled us to evaluate the formation and development of pars flaccida retraction pockets and cholesteatomas in Mongolian gerbils and proved to be an excellent diagnostic tool.
