Pathogenesis of middle-ear effusion owing to the hydrops ex vacuo theory is now evidence-based from studies in humans and animals.
The chapter includes
• The role of the ET in otitis media and related conditions, on the basis of experimental data in humans and animals, although some of the mechanisms remain speculative.
• The role played by the tube in acute otitis media, persistent middle-ear effusion that follows an acute episode, otitis media with effusion, ET dysfunction, atelectasis of the middle-ear cleft, and barotrauma.
• Contributing factors to the high incidence of otitis media.
It is beyond doubt that sometimes in excessive swelling of the tubal mucous membrane and impermeability of the Eustachian tube there occurs in consequence of the consecutive rarefaction of the air in the tympanum, a transudation of serous fluid.
—Adam Politzer (1835–1920)
Many factors have been identified as being involved in the etiology and pathogenesis of middle-ear disease, other than dysfunction of the ET (see Figure 1–7). The most important of these other factors would be host related, such as genetic predisposition, age, prematurity, gender, race, allergy, immunocompetence, and craniofacial abnormalities. Environmental factors are also important, such as upper respiratory tract infection, season of the year, attendance in child day care, number of siblings, exposure to smoking, absence of breast-feeding, socioeconomic status, and use of a pacifier (see Chapter 2, “Epidemiology”).
Many years ago, I proposed a few concepts relating the role of the dysfunctional ET system in the pathogenesis of otitis media and certain related diseases and disorders,1 but now we have studies in both humans and animals to confirm some of the more common pathogenic sequences leading to otitis media. Chapter 5, “Pathophysiology,” is a review of the pathophysiology of the ET’s system as it relates to pathogenesis.
Many abnormalities of the ET system can be attributed to the pathogenesis of middle-ear disease, some of which do not necessarily have to involve the existence of pathophysiology or pathology of the tube itself to be present. The most obvious example of middle-ear infection that does not involve dysfunction of the tube in its pathogenesis is when there is a traumatic perforation of the tympanic membrane in an individual who has a tube that functions normally; despite normal tubal function, the posterior, middle-ear end of the system is open. These patients frequently develop an attack of acute otitis media, with otorrhea, at the onset of an upper respiratory tract infection. Infected nasopharyngeal secretions can reflux through the tube into the middle ear owing to the loss of the middle ear–mastoid gas cell cushion. Also, these patients can insufflate nasopharyngeal secretions through a normal tube into the middle ear by vigorously blowing the nose.
ET dysfunction is most commonly involved in the pathogenesis of middle-ear disease. In this chapter, the recent studies are described in detail that have provided convincing evidence-based confirmation that the hydrops ex vacuo theory, first advanced by Adam Politzer over 100 years ago,2 is the most logical explanation for the development of middle-ear underpressures, which is the stage that precedes the most frequently encountered types of middle-ear disease: acute otitis media and otitis media with effusion. Clinical studies in adult volunteers and experiments in animal models have shown that ET dysfunction is involved in the pathogenesis of middle-ear underpressures, otitis media with effusion, and acute otitis media.
Overview of Pathogenesis Related to the Pathophysiology of the ET System
The ET system can be either too closed or too open, or there is abnormal pressure at either end. The most common result of a dysfunction of the system is development of middle-ear underpressures, which can progress into middle-ear disease. But, as stated previously, there are circumstances in which negative middle-ear pressure is not involved in the pathogenesis, such as when acute otitis media, with otorrhea, occurs when the tympanic membrane is not intact, owing either to a perforation, a tympanostomy tube in place, or when the patient has had a radical mastoidectomy. The middle ear and mastoid can be infected by two major routes: reflux of infected secretions from the nasopharynx or contamination of the middle-ear cleft from the external auditory canal, most commonly from water in the ear canal. If the acute otitis media does not resolve spontaneously or following treatment, the acute disease can progress to chronic suppurative otitis media, as I discuss in this chapter. But dysfunction of the tubal system is most frequently the cause of underpressures developing in the middle ear, which can progress to middle-ear disease.
Primary and Secondary Causes of Middle-Ear Underpressure Owing to Dysfunctions of the ET System
Dysfunction of the system may be a primary or a secondary cause of middle-ear negative pressure. Abnormalities of the system that are a primary cause are due to either anatomic or functional obstruction of the tube or both. Anatomic obstruction is most frequently caused by inflammation, which is most commonly viral in etiology. Evidence for viruses as the cause of tubal dysfunction and middle-ear negative pressure has been demonstrated in studies that involved adult volunteers; however, as described subsequently, subjects who had preexisting tubal dysfunction were at highest risk of developing middle-ear underpressures after being challenged with a respiratory virus inoculated into the nasal cavity. Development of underpressures in the middle ear could be from any of the intrinsic or extrinsic anatomic conditions that are related to pathophysiology of the tube itself (such as allergic inflammation); at either end of the tubal system, such as obstruction at the nasopharyngeal end, as might be caused by adenoids or tumor; or at the middle-ear end of the system, such as cholesteatoma or polypoid granulation tissue.
Functional obstruction of the tube within its system can also cause primary development of middle-ear negative pressure. The possible underlying causes of functional obstruction of the tube are presented in Chapter 5 but include a floppy cartilage, dysfunction of the tensor veli palatini muscle, and constriction of the tube during swallowing.3 Functional obstruction of the tube has been identified by our team4–7 and confirmed by other investigators.8–10 A more dramatic and rapid example of a primary cause of middle-ear negative pressure by functional (as opposed to anatomic) obstruction is otitic barotrauma owing to “locking” of the tube during descent in an airplane, scuba diving, and hyperbaric oxygen therapy; the sudden negative pressure at the middle-ear end of the tubal system functionally obstructs the tube. Patients who have preexisting dysfunction of the ET are at highest risk.11 Functional tubal dysfunction can also be induced at the proximal end of the tubal system and cause middle-ear negative pressure by either functionally obstructing the tube at the nasopharynx (the Toynbee phenomenon or in infants using a pacifier, thumb sucking, or sucking on a nonventilated milk bottle) or by aspirating gas out of the middle ear by sniffing.
A secondary cause of middle-ear underpressures would not be primarily related to anatomic or functional obstruction but secondarily to inflammation that occurs in the middle ear. Because middle-ear inflammation can obstruct the osseous portion of the tube, preventing adequate pressure regulation of the middle ear (anatomic obstruction at the middle-ear end of the system, which can impair the pressure regulatory and clearance functions of the tube), secretions are trapped in the middle ear. From my description of the inverted flask in Chapter 5, as liquid flows down the narrow neck, negative pressure develops in the bulbous portion of the flask. Negative middle-ear pressure has been identified in children who have otitis media with effusion.12 The most likely cause of inflammation occurring within the middle ear in this scenario is either from reflux (i.e., the tube is too open, too short, or both) of nasopharyngeal secretions into the middle ear or by insufflation of secretions from the nasopharynx into the middle-ear cavity by nose blowing or crying in infants or from the Toynbee phenomenon (if positive pressure is transmitted to the middle ear). From knowledge of fluid dynamics through a collapsible tube, liquid flows more readily than gas. Thus, if positive pressure has been identified in the middle ear during these events, nasopharyngeal secretions are even more likely to be forced into the middle ear.
Middle Ear Negative Pressure Can Cause Otitis Media with Effusion and Acute Otitis Media
The next step in the sequence of events leading to otitis media is the development of middle-ear effusion owing to the presence of underpressures in the middle ear; various mechanisms causing these pressures are discussed earlier. As postulated by Politzer, middle-ear negative pressure can cause transudation of fluid from the middle-ear mucosa into the cavity of the middle ear. Since his time, subsequent observations and now studies in humans and animals have provided evidence to support his hydrops ex vacuo theory.2 Obviously not available to Politzer 100 years ago, development of barotitis is a now a well-known complication of flying in an airplane (especially when the cabin is not pressurized), scuba diving, and during hyperbaric oxygen treatment in a pressure chamber, none of which is physiologic.11,13–15 Another dramatic example of how middle-ear negative pressure can cause transudation of an effusion was shown by Ingelstedt and colleagues in which they aspirated the middle ear and mastoid with a syringe in aviators and produced middle-ear effusion.16
Even though these rapid alterations in pressures are impressive examples of effusion developing in the middle-ear cleft owing to the extreme and rapid onset of negative middle-ear pressure, they are not related to the pathogenesis of middle-ear disease in children and adults who have not been exposed to these conditions. We must, therefore, turn to the studies in humans and animals described later. Also, the originally proposed classic ex vacuo theory does not explain if middle-ear negative pressure is a stage prior to an attack of viral or bacterial acute otitis media. But as shown subsequently, following intranasal inoculation of a virus in an adult volunteer, tubal obstruction occurred, followed by middle-ear negative underpressure and the subject developing an attack of acute viral and bacterial otitis media.17 It is probable that the nasopharyngeal pathogens were aspirated into the middle ear; thus, development of negative pressure in the middle ear can be an antecedent event for the pathogenesis of both otitis media with effusion and acute otitis media. As stated previously, acute otitis media can be caused by other dysfunctions of the tube and pathogenetic mechanisms, such as reflux or insufflation of nasopharyngeal secretions into the middle ear, which do not have to have underpressures in the middle ear to occur; however, if negative middle-ear pressure is present, it will enhance reflux and insufflation into the middle ear.
Viral Nasal Challenge Studies in Adult Volunteers
Clinical studies of adult volunteers at our center have assessed nasal and ET functions and the status of the middle ear following intranasal challenge with viruses. These studies demonstrated the role that the tube plays in the pathogenesis of middle-ear underpressures, otitis media with effusion, and acute otitis media and are summarized in Table 6–1.17–23
In an early study, Doyle and colleagues determined the effect of an upper respiratory tract infection (a cold) on ET function and the status of the middle ear after intranasal challenge of rhinovirus in a group of 40 adult volunteers.18 After rhinovirus was inoculated into the nose, all subjects were found to be infected, but only 80% developed the signs and symptoms of a clinical illness. Before and periodically after this nasal challenge, assessments were made of tubal function (using sonotubometry and the nine-step test), middle-ear pressure (using tympanometry), and nasal patency (using active posterior rhinometry). All subjects who had a cold had decreased nasal patency, 50% had ET obstruction, and 30% had abnormal negative middle-ear pressure for approximately 1 week after the inoculation. All outcomes completely resolved within 16 days, but none of the volunteers developed a middle-ear effusion.
In a subsequent study by McBride and colleagues that employed similar methods and design as described in the previously cited study, 32 adult volunteers were recruited.19 After the challenge with rhinovirus, abnormal findings were limited to the 24 (75%) subjects who developed clinical signs and symptoms of infection. After 2 days, 80% had ET obstruction, 50% had high negative middle-ear pressure, and 46% had decreased nasal patency. Again, none of the subjects developed a middle-ear effusion. These abnormal findings resolved 6 to 10 days after the challenge.
A similar study by Buchman and colleagues evaluated 60 adult volunteers using a design and methods similar to the previous two studies.20 Figure 6–1 shows that after nasal inoculation with rhinovirus, 95% became infected and 60% had a clinical cold. Before the nasal challenge, three volunteers (5%) had abnormal middle-ear pressure, and in two of these subjects, a middle-ear effusion developed. Of the 60 subjects, 22 (39%) had high negative middle-ear pressure. None of the subjects who had normal middle-ear pressure before the challenge developed an effusion, indicating that a rhinovirus infection may result in a middle-ear effusion if the patient has a preexisting dysfunction of the tube.
In still another study, but with a different respiratory virus, Doyle and colleagues reported that intranasal challenge with influenza A virus in 33 healthy adult volunteers resulted in 80% demonstrating tubal obstruction and 80% having negative middle-ear pressure21; however, with this virus, 5 (23%) of the 21 infected subjects also developed a middle-ear effusion. Most likely, the influenza A virus is more virulent than rhinovirus in the pathogenesis of ET and middle-ear abnormalities.
Adapted from Bluestone CD.78
AOM = acute otitis media; ET OBS = Eustachian tube obstruction; HNP = high negative pressure; MEE = middle-ear effusion; NT = not tested.
*Eustachian tube dysfunction and abnormal middle-ear pressures in individuals are more prevalent when the tube is found to have poor function prior to challenge (see text).
In a highly enlightening study, Buchman and colleagues demonstrated the events leading to the development of not only otitis media with effusion but, more importantly, an acute otitis media that developed in one subject.17 Using a design similar to the previous studies, they recruited 27 adult volunteers, in whom influenza A was inoculated into the nose. Figure 6–2 shows that all subjects developed a nasal infection and 16 (59%) subsequently developed high negative middle-ear pressure. In one subject, acute otitis media was present. Using polymerase chain reaction, a middle-ear aspirate revealed the virus and Streptococcus pneumoniae; traditional viral and bacterial culture methods failed to grow these organisms from the middle-ear effusion. It is possible that these microorganisms were aspirated from the nasopharynx into the middle-ear cavity owing to the high negative middle-ear pressure.
In a more recent study, Doyle and colleagues infected 18 adult subjects with influenza A, showing that those individuals who had preexisting good tubal function reduced the otologic complications of the viral upper respiratory tract infection.22 Also, Buchman and colleagues inoculated respiratory syncytial virus into the nasal cavities of 32 adult volunteers.23 Only 56% had detectable infection, but all of the subjects who did have infection had rather substantial signs and symptoms, and 54% of these subjects developed abnormal middle-ear pressures.
These studies in adult volunteers were unique because they demonstrated the relationship of viral upper respiratory tract infection, middle-ear negative pressure, effusion, and acute otitis media.
Studies of Upper Respiratory Tract Infections in Children
An informative clinical investigation by Moody and colleagues also demonstrated a similar sequence of events in children, as was documented in adult volunteers.24 In this study, the parents of 20 children between the ages of 2 and 6 years monitored the middle-ear status of their children every day using a tympanometer. They reported that when an upper respiratory tract infection developed in the children, many soon also developed middle-ear underpressures, and some then developed a middle-ear effusion.
In a follow-up study, Antonio and colleagues prospectively followed 40 children in their homes, using daily tympanometry, symptom diaries, and weekly otoscopic examinations, during the respiratory seasons (fall, winter, and early spring) and during periods of common cold, which occurred in 22%; 63% of all otitis media episodes occurred during an upper respiratory tract infection.25 Interestingly, 30% of the episodes of otitis media were evident 1 to 7 days prior to the onset of the cold, whereas in 26%, the middle-ear disease occurred between 8 and 14 days after the onset of the cold; the remaining 44% had otitis media on the same day or up to the seventh day after the onset of the cold. These prospective studies in children who developed a “wild cold”—not induced by intranasal inoculation of viruses, as in the adult studies—showed that almost one-third of episodes of middle-ear effusion occurred prior to the signs and symptoms being apparent. Thus, the virus had already caused ET dysfunction and otitis media.
There is now equally convincing evidence from studies in adult volunteers that the ET is involved in the pathogenesis of otitis media. These six experiments show that the ET has an important role in the development of otitis media in animal models. This is described subsequently.
Experiments in Animals
In our laboratory over the past 40 years, we successfully produced both underpressures and middle-ear effusion in animal models using several methods. Some of our more important studies are summarized in Table 6–2.26–31
We demonstrated that when the tensor veli palatini muscle is experimentally impaired (altered or inactivated) in animal models, active opening of the ET is impaired. This results in negative middle-ear pressure followed by middle-ear effusion. In one experiment, excision of a portion of the tensor veli palatini muscle at the pterygoid hamulus in the palate resulted in negative pressure in the middle ear followed by an effusion.26 A similar experiment in which the muscle was completely excised, the superficial muscle bundle was transected, or the tendon medial to the hamular process was transposed had comparable outcomes.27 Figure 6–3 shows the surgical procedures used and the number of animals in each group. Complete excision of the tensor tendon resulted in middle-ear underpressures followed by persistent middle-ear effusion. Transection of the muscle resulted in negative middle-ear pressure, or effusion, or both (and in some animals, the middle ear returned to normal after the muscle healed). When the tendon was transposed, outcomes were similar to surgical alteration, but the middle ear rapidly returned to normal in a few weeks. Using a noninvasive method, Casselbrant and colleagues injected botulinum toxin into the tensor muscle, which resulted in negative pressure and then effusion.28 When the effect of the botulinum toxin resolved, the middle-ear status returned to normal.
In these earlier studies, middle-ear status was diagnosed with otomicroscopy and tympanometry. We have also used magnetic resonance imaging (MRI) to more accurately diagnose the presence of effusion in the middle ear and mastoid. Figure 6–4 shows the flow diagrams for two such studies in the monkey. Alper and colleagues used MRI and tympanometry to identify middle-ear and mastoid effusion.31 These investigators also injected botulinum toxin into the tensor veli palatini muscle of monkeys, resulting in an ET that failed to open, middle-ear underpressure, and effusion. In ears that developed underpressure within the middle ear, increased vascular permeability was observed on the MRI. These created a functional obstruction of the ET (i.e., an impairment of the active opening of the tube) impeding pressure regulation of the middle ear and resulting in an effusion. As in the earlier study, when the effect of the botulinum toxin resolved, the middle-ear status returned to normal.
Adapted from Bluestone CD.78