Definitions of otologic diseases and recommended study designs.

Research is what I'm doing when I don't know what I'm doing.

--Wernher Von Braun (1912-1977)

Introduction

Dr. Joseph E. Dohar: Nonantibiotic treatment guidelines have been at the center of considerable controversy since their recent introduction. But a fundamental concern is that these guidelines were based on studies in which otitis media was poorly defined. Although no such guidelines have been published regarding forms of otitis that are treated ototopically, we believe it is important to examine and define these diseases to preempt similar controversies in the future. By so doing, we hope to ensure that we will know what we're doing when we conduct future research.

Definitions

Dr. Dohar: It may seem curious that topics as basic and fundamental as definitions of otologic diseases and study designs are included in a supplement entitled, "Controversies in ototopical therapy." But as one considers the current state of the art, the inclusion of these topics should not be surprising at all.

Arguments have erupted over definitions. Many otolaryngologists have criticized the recent and widely publicized definition of acute otitis media (AOM) that was developed by the Agency for Healthcare Research and Quality (AHRQ) because they say the AHRQ has actually defined an entirely different disease--namely, otitis media with effusion (OME). (1) Many believe that the acceptance of this erroneous definition by regulatory and registration authorities has resulted in an "evidence-based" recommendation that we should withhold antibiotic treatment in certain situations. Disagreement with the AHRQ's position is deep.

As for study designs, researchers have both consciously and unconsciously manipulated the use of certain important parameters to increase the likelihood of selected study outcomes. In the end, if clinical trials are to have value and significance, specific and accurate disease definitions and optimal methodology must be employed.

Definition of AOM

Prof. Thomas Deitmer: The AHRQ has actually developed two "consensus definitions" of AOM--a short and a long version. (1)

The abbreviated definition of AOM is "the presence of middle-ear effusion in conjunction with the rapid onset of one or more signs or symptoms of inflammation of the middle ear."

In the long version, the AHRQ defines AOM as "(1) the presence of middle-ear effusion as demonstrated by the actual presence of fluid in the middle ear as diagnosed by tympanocentesis or the physical presence of liquid in the external ear canal as a result of tympanic membrane (TM) perforation or indicated by limited or absent mobility of the TM as diagnosed by pneumatic otoscopy, tympanogram, or acoustic reflectometry with or without the following: (a) opacification, not including erythema, (b) a full or bulging TM, or (c) hearing loss, and (2) rapid onset (ie, up to 48 hours from the onset of acute signs or symptoms first noted by the parent or guardian to the time of contact with the health system) of one or more of the following signs or symptoms, with or without anorexia, nausea, or vomiting: (a) otalgia (or pulling of ear in an infant), (b) otorrhea, and (c) irritability in the infant or toddler, or (d) fever."

Is there a diagnostic gold standard for AOM? Let us look at some controversial study data:

* Pichichero studied the accuracy of otoscopic diagnosis of AOM and OME by 383 pediatric residents who watched otoscopic videos. (2) He found that these pediatricians frequently misdiagnosed OME. He also found that the color of the TM is a poor predictor of AOM and that the presence of a bulging TM is more reliable.

* Saeed et al compared the value of otoscopy, tympanometry, and tympanocentesis in detecting AOM, with and without effusion, in 90 children/They found that tympanocentesis did not detect middle ear effusion in 14% of the cases in which it had been detected by the other two methods. This suggests that tympanocentesis is not the gold standard for diagnosing AOM.

Prof. Michael Hawke: It is important to not only strictly define otologic diseases, but also to minimize interobserver variability, as well. Attempts have been made to educate and "certify" otoscopists. This is important in the setting of a clinical trial because, as just noted, Pichichero found that the diagnostic accuracy of pediatric residents was approximately 50%. (2) The diagnostic accuracy of otolaryngologists, on the other hand, was 75%.

Diagnosis may be further enhanced by obtaining otoscopic photographs for use in consultation. Diagnostic accuracy is often improved when more than one pair of experienced eyes evaluate a particular case.

Prof. Deitmer: Let us look at some other studies on the clinical features of AOM:

* Harabuchi et al devised a clinical scoring system to assess changes in signs, symptoms, and TM features in 115 children with AOM who had been treated with a single oral antibiotic for 7 days. (4) These patients were followed for 1 month. The authors concluded that it may be possible to predict the clinical course of AOM on the basis of presenting signs, symptoms, TM features, and identification of nasopharyngeal pathogens. Sign/symptom scores declined at a much faster rate than did TM scores. Also, despite treatment, AOM patients with Streptococcus pneumoniae colonization in the nasopharynx had a significantly longer duration of symptoms than did those with Haemophilus influenzae or Moraxella catarrhalis infections.

* In a study of 771 children aged 3 to 36 months with AOM, Polachek et al reported that: (1) leukocyte counts were higher in culture-positive middle ear diseases, (2) leukocyte and absolute neutrophil counts were significantly higher in patients with pneumococcal AOM than in patients with H influenzae or culture-negative disease, (3) C-reactive protein levels were higher in bacterial AOM than in viral AOM, and (4) leukocyte counts were significantly reduced after 4 to 6 days of antibacterial therapy, regardless of etiology or bacterial outcome. (5)

* Has the clinical picture of AOM changed? Researchers in Finland say it has, and to a great degree. Joki-Erkkila et al retrospectively studied all cases of AOM that had occurred in children younger than 10 years during a 12-month period in 1978 and 1979. (6) They then compared them with similar cases that had occurred during 12 months in 1994 and 1995. They found that during the intervening 16 years, AOM became less severe overall but that the incidence of recurrences increased significantly, from 6.0 to 12.7%. The authors associated this increase with a general shift in treatment patterns from the widespread use of penicillin V to the use of broad-spectrum antibiotics. In addition, the percentage of patients who had undergone surgical procedures (i.e., adenoidectomy and/or tympanostomy tube placement) doubled from 6.2 to 12.4%.

* Takata et al conducted a metaanalysis of 739 cases of AOM in children aged 4 months to 18 years. (7) They determined that the spontaneous resolution rate at 7 days was 81% and that suppurative complications were uncommon. When antibiotic therapy was administered, the 7-day clinical failure rate fell from 19 to 7%. However, they did not specifically define AOM, and it is doubtful that all of these cases were actually AOM. Scrutiny of these findings illustrates how serious an incorrect definition of AOM can be.

Definition of OME

Prof. Deitmer: Otitis media with effusion is characterized by the presence of fluid from the middle ear without signs and symptoms of acute infection?

Definition of AOM through tympanostomy tubes (AOMT)

Prof. Deitmer: A diagnosis of AOMT requires that the tube be patent. Drainage in a patient with a blocked tube represents simple AOM, which might be made worse by the presence of the foreign body in the TM.

The clinical features of AOMT are an acute onset of symptoms and the presence of purulent secretions in the outer ear canal. The onset of AOMT symptoms often occurs in conjunction with an upper respiratory infection. Systemic symptoms such as fever, pain, and irritability are not common.

In a randomized, controlled, double-blind trial, Ruohola et al gave amoxicillin/clavulanate or placebo for 7 days to 66 children with AOMT. (9,10) Those patients in the active-treatment group experienced a shorter median duration of both otorrhea (3 vs. 8 days) and bacterial growth in middle ear fluid (1 vs. 8 days).

Dr. Ramzi T. Younis: To date, there is no accepted staging system for AOMT. Let us propose definitions of AOMT classified simply as mild, moderate, and severe.

* AOMT is mild when the volume of discharge does not completely obscure the tube.

* AOMT is moderate when the tube cannot be seen without first performing aural toilet.

* AOMT is severe when the otorrhea is associated with granulation tissue.

Prof. Hawke: With regard to that staging system, we must keep in mind that an irritable child with a fever can have a TM that looks normal on examination and yet may become hemorrhagic and rupture within a span of 6 hours. Such a change may confound staging, and therefore it should be anticipated when designing a study.

Another factor that can confound the definition of AOMT is the presence of granulation tissue that serves as the sole source of the otorrhea. Such patients should be excluded from AOMT trials which, by definition, require that otorrhea originate from the middle ear.

Overall, far less interobserver variability should occur with AOMT than with AOM.

Dr. John Rutka: What is the natural history of untreated AOMT, and what is the rate of spontaneous remission? These are very germane issues when we consider interventions.

Prof. Deitmer: In the study by Ruohola et al, the spontaneous remission rate in the placebo group at 7 days was 41%, the median duration of otorrhea was 8 days, and the median duration of bacterial growth in middle ear fluid was 8 days. (9)

Dr. Younis: As is the case with AOM, the natural history of AOMT in very young children (i.e., infants) differs from that in older children. There is also a difference in bacteriology by age and by season. And as is the case with AOM, daycare attendance may affect both the duration and the frequency of AOMT.

Definition of chronic suppurative otitis media (CSOM)

Prof. Deitmer: CSOM may be defined as the presence of purulent secretions in the outer ear canal in a patient with a nonintact TM and without a cholesteatoma. Clear definitions of CSOM are scarce, probably because of the difficulty we have in determining when a retraction pocket should be called a cholesteatoma. CSOM can persist for months and even years while patients experience bouts of acute episodes. Again, fever, pain, and irritability are not common.

Dr. Harvey Coates: I would add that CSOM is characterized by recurrent or persistent bacterial infection of the ear with conductive hearing loss. Also, the duration of CSOM is an important aspect of its definition. The World Health Organization defines CSOM as a condition that persists for 2 weeks or longer. But if a child with AOM continues to drain for 1 or 2 weeks, how can we distinguish this condition from CSOM? I prefer to define CSOM as a condition that persists for 4 weeks to 2 months.

Dr. Dohar: The definition of CSOM is more controversial than the definition of these other otologic diseases. CSOM is not the same as serous or secretory otitis media. The term used to describe these latter two entities is chronic otitis media with effusion.

When the federal Food and Drug Administration (FDA) approved ototopical ofloxacin for CSOM, it defined CSOM as a drainage of any duration through a chronic perforation. However, this definition would include patients who do not have what we commonly think of as CSOM. What distinguishes chronic suppurative otitis media from acute suppurative otitis media (i.e., AOMT) are the histopathologic changes. We must consider the point at which the disease changes and whether the change is microbiologic or histologic.

Dr. Patrick J.Antonelli: Otologists usually operate on ears that are nonhealing, and we arbitrarily define CSOM as a persistent drainage of several months' duration or a minimum of 30 days' drainage over a period of 2 or 3 months.

Dr. Ann Edmunds: Some cases of CSOM are difficult to evaluate because patients do not return for regular follow-up visits. Therefore, we do not always know how long their draining episodes last, how often they recur, and whether they are complying with therapy and water-avoidance precautions.

Dr. Scott Manning: From a pediatric viewpoint, CSOM is chronic otorrhea that: (1) persists for 6 to 8 weeks, (2) does not respond to therapy, (3) is accompanied by granulation tissue, and (4) occurs in a patient with a TM perforation and/or a tympanostomy tube.

Dr. Dohar: That is true. Thus far, our discussion has centered only on children who have a chronic perforation. Is there such an entity as CSOM in a child with a tympanostomy tube? If so, what is its operational definition?

Dr. Rutka: The American Academy of Otolaryngology--Head and Neck Surgery (AAO-HNS) defines CSOM as a chronic problem in the historical, the clinical, and the pathologic sense. (11) The AAO-HNS makes no mention of duration, but most of us would agree that, at a minimum, the disease must persist for several weeks before we can establish a diagnosis of CSOM. The AAO-HNS definition also includes the presence of a TM perforation and a chronic persistent discharge that may or may not relapse.

We must also keep in mind that patients with CSOM will experience changes that involve the TM, the mucosa, the middle ear space, and the eustachian tube.

Prof. Hawke: I strongly disagree that the mere presence of a TM perforation represents inactive CSOM. A perforation is nothing more than a hole in the eardrum. It can be caused by an infection or trauma.

Dr. Younis: Another important factor is that we must differentiate between CSOM and cholesteatoma. By and large, CSOM is managed medically and cholesteatoma is managed surgically.

Methodology of AOM studies

Prof. Deitmer: Dagan and McCracken identified several flaws in the way AOM studies are conducted. (12) According to their tutorial:

* Bacterial eradication can be proven only by double tympanocentesis.

* Determinants of the clinical score for otitis media include body temperature, irritability, ear-pulling, and TM bulging and erythema.

* The clinical score correlates well with bacteriologically proven cure.

* Outcomes measures vary among different studies. For example, the timing of the test of cure varies. The FDA requires that test of cure be determined 28 days after the onset of therapy, but this benchmark is problematic because a reinfection might be misinterpreted as a treatment failure. (I believe that a 12-day period is more realistic.)

* The clinical score for otitis media should be used in studies, and this should correlate to bacteriologically proven cure.

* Study results can be influenced by inherent biases ("outcome determinators") if they are not controlled for. Among these influences are disease severity, daycare attendance, age less than 2 years, passive smoke inhalation, the presence of siblings, breast feeding, and pacifier use.

As we know, a relatively ineffective antibiotic can appear to be more efficacious if the patient population is skewed to obtain the desired results. Among the factors that increase the likelihood of a good antibiotic result are the inclusion of older children, children who do not attend a daycare facility, children who have no siblings, and/or children who have no history of recurrent disease or antibiotic use. Another determining factor is the choice of a clinical rather than a bacteriologic outcomes measure.

Meticulous attention to precise and stringent definitions of disease and study design will optimize the quality of "the evidence" that will be used in the future to guide good clinical practice.

Dr. Dohar: Keep in mind that Prof. Deitmer's remarks on clinical scores pertain to the registration process in the United States; systems may be different in other countries. For new drug approvals for respiratory tract infections, both clinical outcomes measures and clinical scores are required. Proof of efficacy should be demonstrated in either placebo-controlled trials or in trials that are powered to demonstrate superiority. "Noninferiority" studies with a single clinical parameter, such as cure or no cure, are discouraged. Microbiologic eradication should rarely, if ever, stand as the sole endpoint for new antibiotic approval.

Dr. David Stroman: In any trial of an antiinfective, the FDA requires that both clinical and microbiologic cure be the primary endpoints. In the United States, we now need two primary endpoints rather than one primary and one secondary endpoint.

Dr. Dohar: It is easier to identify both the microbiologic and inflammatory components in AOMT than in AOM. Ruohola et al suggested that perhaps a surrogate model for AOM studies should be tympanostomy tube otorrhea because the microbiology is readily available; more than 80% of patients have actual pathogens. (9) Moreover, the problem of interobserver variability in diagnosing AOMT is not as large a factor as it is with AOM, as Prof. Hawke noted.

With regard to sign/symptom scores, Pichichero's most recent data show that one way to markedly increase bacteriologic yield in tympanocentesis, which is traditionally only about 50%, is to study TMs that are bulging. (2)

The study from Finland by Joki-Erkkila et al that Prof. Deitmer mentioned showed that the severity of AOM has decreased since the 1970s. (6) This, too, has an impact on disease definition and determinations of disease outcome. This has also occurred in the United States. In part, this may be a result of universal immunization with pneumococcal conjugated vaccine in the United States.

References

(1.) Management of acute otitis media. Summary. Evidence Report/ Technology Assessment Number 15. Rockville, Md.: Agency for Healthcare Research and Quality; www.ahrq.gov/clinic/epcsums/otitisum.htm (accessed Aug. 23, 2005).

(2.) Pichichero ME. Diagnostic accuracy, tympanocentesis training performance, and antibiotic selection by pediatric residents in management of otitis media. Pediatrics 2002;110:1064-70.

(3.) Saeed K, Coglianese CL, McCormick DE Chonmaitree T. Otoscopic and tympanometric findings in acute otitis media yielding dry tap at tympanocentesis. Pediatr Infect Dis J 2004;23:1030-4.

(4.) Harabuchi Y, Kodama H, Faden H. Outcome of acute otitis media and its relation to clinical features and nasopharyngeal colonization at the time of diagnosis. Acta Otolaryngol 2001;121:908-14.

(5.) Polachek A, Greenberg D, Lavi-Givon N, et al. Relationship among peripheral leukocyte counts, etiologic agents and clinical manifestations in acute otitis media. Pediatr Infect Dis J 2004;23:406-13.

(6.) Joki-Erkkila VP, Pukander J, Laippala P. Alteration of clinical picture and treatment of pediatric acute otitis media over the past two decades. Int J Pediatr Otorhinolaryngol 2000;55:197-201.

(7.) Takata GS, Chan LS, Shekelle P, et al. Evidence assessment of management of acute otitis media: I. The role of antibiotics in treatment of uncomplicated acute otitis media. Pediatrics 2001; 108: 23947.

(8.) Rosenfeld RM, Culpepper L, Doyle KJ, et al. Clinical practice guideline: Otitis media with effusion. Otolaryngol Head Neck Surg 2004;130(5 suppl):S95-118.

(9.) Ruohola A, Heikkinen T, Meurman O, et al. Antibiotic treatment of acute otorrhea through tympanostomy tube: Randomized doubleblind placebo-controlled study with daily follow-up. Pediatrics 2003;111(5 Pt 1):1061-7.

(10.) Powers JH, Johann-Liang R. Tubes and ear infections [letter]. Pediatrics 2004;113(1 Pt 1):178-9; author reply 178-9.

(11.) Hannley MT, Denneny JC III, Holzer SS. Use of ototopical antibiotics in treating 3 common ear diseases. Otolaryngol Head Neck Surg 2000; 122:934-40.

(12.) Dagan R, McCracken GH, Jr. Flaws in design and conduct of clinical trials in acute otitis media. Pediatr Infect Dis J 2002;21:894-902.