 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
To the Editor.—
Ruohola et al1 demonstrate that placebo-controlled trials can be performed in young children, and time-to-resolution endpoints can be measured in such trials. Recent US Food and Drug Administration meetings2,3,4 have discussed the need for placebo-controlled antimicrobial trials in potentially self-resolving bacterial infections and the endpoints for such trials.
Unfortunately, the clinical relevance of the time to resolution of tube otorrhea is unclear. The authors state that acute otitis media (AOM) is associated with tube otorrhea when accompanied by acute symptoms. The authors show that the majority of children had "respiratory symptoms," but these may not be the same as symptoms of AOM. If acute symptoms differentiate AOM from tube otorrhea alone, then the resolution of acute symptoms should be the primary endpoint. Measuring the time to resolution of acute symptoms rather than measuring outcomes at some fixed time point is most relevant to patients and may allow the greatest chance of showing a difference between antibiotics and placebo in self-resolving diseases. As the authors state, the ultimate goal of administering antibiotics is to help the host recover from disease. Disease is the presence of signs and symptoms, not merely the presence of bacteria and/or serous drainage. Although microbiologic endpoints are important, measurement on therapy may represent suppression, not eradication of organisms. Microbiologic endpoints should be measured after completion of therapy and elimination of the drug to look for bacterial regrowth.
It is uncertain how the results of this study relate to children with routine AOM. Children in the United States with tympanostomy tubes represent a different population from those with routine AOM. The natural history of AOM with patent tubes may differ from a closed-space infection in routine AOM. Antibiotics in routine AOM have little effect on resolution of middle-ear fluid,5 which is the primary endpoint in this trial. This study does not address how the resolution of tube otorrhea and lack of bacterial growth correlates with the resolution of the constellation of signs and symptoms associated with routine AOM. Therefore, the authors’ conclusion that "this study provides evidence for the efficacy of antibiotics in the treatment of acute middle-ear infections"1 should be interpreted with caution.
There is a need for placebo-controlled trials in children with routine AOM to demonstrate the magnitude of benefit of antimicrobials. The most relevant endpoints are time to resolution of acute symptoms specifically referable to AOM in addition to properly performed microbiology.
ACKNOWLEDGMENTS
The opinions expressed in this letter represent those of the authors and not necessarily those of the US Food and Drug Administration.
REFERENCES
In Reply.—
We appreciate that Powers and Johann-Liang opened the discussion about relevant endpoints in the treatment of acute middle-ear infections. We agree with them that the ultimate goal of administering antibiotics is to help the host recover from the disease, and that the disease includes the presence of signs and symptoms, not merely the presence of bacteria. We were also pleased about their comment that time to resolution is more relevant than a fixed time point to measure the outcome in otitis media trials. However, for the following reasons we consider that, in a population of young children suffering from acute otitis media with otorrhea through tympanostomy tube, the time to resolution of otorrhea is the most relevant and reliable endpoint.1
First, although the resolution of subjective symptoms would obviously be a clinically relevant outcome, children with acute middle-ear infections have almost always a concomitant viral upper respiratory infection.2 Therefore, they have the same symptoms as those children with common cold only: rhinitis, cough, fever, irritability, and/or restless sleeping.3–5 The only symptom that can be considered as specific to middle-ear infection is earache, but unfortunately young children who suffer most from otitis media are often unable to express definite earache. Notably, 28% to 40% of all children with acute otitis media with intact tympanic membrane do not have any kind of ear-related symptoms, and those with patent tympanostomy tubes usually lack ear-related symptoms entirely.3,4 The symptoms of our study patients were recorded daily. However, because these symptoms were typical to viral-type upper respiratory infection, we could not include the resolution of such symptoms as outcomes in assessing the efficacy of antibiotics in the treatment of acute otorrhea through a tympanostomy tube.
Second, the time to resolution of otorrhea is the most relevant outcome in this context, because formation of middle-ear fluid is a central clinical manifestation of the pathogenesis of a middle-ear infection. Although Powers and Johann-Liang doubted the clinical relevance of this outcome, we think that the duration of otorrhea has clinical and every-day-life relevance because otorrhea often prevents the child from attending day care or other activities, and otorrhea lasting >3 days has been shown to reduce the quality of life.6 As clinicians, we further think that it is important to restrict the duration of any illness to be as short as possible.
Powers and Johann-Liang stated that microbiologic endpoints should be measured after completion of therapy. This necessitates the use of tympanocenthesis. Several studies have used the so-called double-tap design to assess the efficacy of different antimicrobials in the treatment of acute otitis media. However, it can be questioned whether the use of this study design has caused more suffering than alleviation to children because tympanocenthesis has been shown not to improve the recovery of an individual child and the procedure is traumatic and stressful.7–9 Therefore, we consider it unethical to use a painful and noncurative procedure in young children only for scientific purposes.
The criteria for tympanostomy tube placement are probably more liberal in Finland than in the United States, and, as we explained in our report, our study population was not extremely otitis-prone. Also, the other patient characteristics and bacterial findings of our study population did not differ from those in an average population with acute otitis media with intact tympanic membranes. Although we agree that the resolution of the disease may be different in ears with tympanostomy tubes than in ears with intact tympanic membranes, we believe that the preciseness and objectivity afforded by our study design could provide one alternative model for studying acute middle-ear infections.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2004 American Academy of Pediatrics. All rights reserved. |
||
|
||
