Microbiology of otorrhea in children with tympanostomy tubes: implications for therapy

https://doi.org/10.1016/j.ijporl.2003.08.043Get rights and content

Abstract

Objective: Based on the prevalence of children with tympanostomy tubes who develop otorrhea, an analysis was performed to identify specific prognostic indicators in this population to enable clinicians to determine the likelihood of specific pathogens and select the best empiric therapy. Methods: Data from two multicenter clinical trials of ofloxacin otic solution 0.3% (OFLX) trials in pediatric patients 1–12 years of age were used to develop a statistical model to predict the likelihood of external auditory canal (EAC) or nasopharyngeal (NPG) pathogens. Data were available for 283 microbiologically evaluable patients. Potential indicators included subject age, season of enrollment, ear aspiration, cleaning, otorrhea, and granulation tissue. The model used a stepwise logistic regression analysis relating the occurrence of NPG or EAC pathogens to the potential prognostic indicators. Results: Among the total study population, 42.8 and 61.5% had NPG and EAC pathogens, respectively; 10.6% had both. The most frequently isolated valid pathogens were Streptococcus pneumoniae and Haemophilus influenzae. Significant prognostic indicators for NPG pathogens were subject age, season of enrollment, and presence of ear odor. Although these indicators were similar for EAC and NPG pathogens, the correlation was reversed, i.e. older subjects had EAC pathogens, and younger ones (<2 years) had NPG pathogens; EAC pathogens were associated with presence of ear odor and NPG pathogens, with absence of ear odor. Conclusions: A statistically and clinically valid model has been developed that has prognostic value for the clinician treating children with otorrhea and tympanostomy tubes and serves as an aid in the appropriate choice of empiric therapy.

Introduction

In the United States alone, it is estimated that one million tympanostomy tube insertions are performed annually [1].

Otorrhea is the most common sequelae requiring treatment after tube insertion that is encountered by the treating physician [2]. Earlier studies had noted the rate of otorrhea after tube placement to range from 5 to 21% [3], [4]. In contrast, in a more recent analysis in which Mandel et al. [5] examined data from three studies in which patients who had tympanostomy tube insertions were followed from 1 to 3 years, 50% of 246 subjects experienced otorrhea. Additional recent data have indicated post-tympanostomy otorrhea rates as high as 84% within an 18-month follow-up period [6]. Therefore, it is clear that management of otorrhea in the pediatric population is an issue of growing concern.

As in all infections, therapy should be directed at the etiologic pathogens. However, many clinicians are unfamiliar with the full spectrum of pathogens that cause otorrhea and direct therapy at the pathogens usually encountered in patients with acute otitis media (AOM) and intact tympanic membranes. Indeed, until recently, systemic antibiotics were the only agents indicated for the treatment of AOM. In light of the recent availability of a broad-spectrum, non-ototoxic ototopical agent that is indicated for AOM in children 1–12 years of age who have tympanostomy tubes, it is important for clinicians to better understand the microbiology of this condition in order to choose appropriate therapy. The major significance of this paper is to assist clinicians in predicting which patients would be inappropriately treated with standard oral antibiotics FDA approved for otitis media with an intact tympanic membrane.

In order to try to identify the types of pathogens, i.e. external auditory canal (EAC) pathogens (Pseudomonas aeruginosa and Staphylococcus aureus) versus nasopharyngeal (NPG) pathogens (Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis), a clinician will encounter in this pediatric patient population, data from two clinical trials of ofloxacin otic solution 0.3% (OFLX), both conducted in children with otorrhea and tympanostomy tubes were analyzed. These data have been used to develop statistical models that are based on various potential prognostic indicators (demographic characteristics, vital signs, clinical examination findings). The ultimate goal of this model is to enable clinicians, using available data for a individual patient, to predict the specific pathogens in children with otorrhea and tympanostomy tubes in order to select the best empiric treatment for each child.

Section snippets

Methods

Data from the two clinical trials of OFLX that are utilized in this statistical analysis have been reported previously [7], [8]. Both studies were multicenter trials that were to enroll pediatric subjects 1–12 years of age with otitis media in the face of a tympanostomy tube. Patients were enrolled if they presented with acute otorrhea through a tympanostomy tube that was untreated. A more detailed account of the inclusion/exclusion criteria and patient data can be found in the previously

Results

The demographic characteristics of the analysis population are summarized in Table 1, and characteristics of the affected ear and associated surgical findings are summarized in Table 2. The majority of subjects were 2–6 years of age and male, had had tubes in place for >6 weeks, did not have ears cleaned, had no ear odor or granulation tissue, and were equally enrolled by season (May to September and October to April). Because the duration of tube placement was predominantly >6 weeks and ear

Discussion

Otorrhea in the presence of a tympanostomy tube (AOT), manifesting principally as otorrhea, is in many ways a distinct entity from acute otitis media in the presence of an intact tympanic membrane (AOM). First and foremost, the microbiology of these two entities is quite different since, as was found in this study, P. aeruginosa and S. aureus are primary pathogens in AOT and are found only rarely in AOM. In addition, the presenting symptoms of AOM are quite different to those of AOT. Unlike

Conclusions

The model presented in this paper is statistically and clinically valid and has prognostic value for the clinician treating this patient population. While this statistical model is helpful in discerning which patients have either EAC or NPG pathogens, physicians need to ultimately prescribe therapy that will be of maximum benefit to the patient. Given the dilemma of the large number of children who have tympanostomy tubes and their tendency to develop otorrhea and AOM, there is a need to

Acknowledgements

This work was supported by Dajichi Pharmaceutical Corporation, Montvale, NJ. The authors gratefully acknowledge Rose M. Yannuzzi for her editorial assistance and preparation of this manuscript.

References (19)

  • E.L. Goldblatt et al.

    Topical ofloxacin versus systemic amoxicillin/clavulanate in purulent otorrhea in children with tympanostomy tubes

    Int. J. Pediatr. Otorhinolaryngol.

    (1998)
  • R.M. Cantor

    Otitis externa and otitis media. A new look at old problems

    Emerg. Med. Clin. North Am.

    (1995)
  • D.E. Theone et al.

    Pharmacotherapy of otitis media

    Pharmacotherapy

    (1991)
  • C.D. Bluestone, J.O. Klein, Otitis media, atelectasis, and eustachian tube dysfunction, in: C.D. Bluestone, S.E. Stool,...
  • F.S. Herzon

    Tympanostomy tubes

    Arch. Otolaryngol. Head Neck Surg.

    (1980)
  • C.A. McLelland

    Incidence of complications from use of tympanostomy tubes

    Arch. Otolaryngol. Head Neck Surg.

    (1980)
  • E.M. Mandel et al.

    Acute otorrhea: bacteriology of a common complication of tympanostomy tubes

    Ann. Otol. Rhinol. Laryngol.

    (1994)
  • C. Ah-Tye et al.

    Otorrhea in young children after tympanostomy tube placement for persistent middle-ear effusion: prevalence, incidence, and duration

    Pediatrics

    (2002)
  • J.E. Dohar et al.

    Topical ofloxacin treatment of otorrhea in children with tympanostomy tubes

    Arch. Otolaryngol. Head Neck Surg.

    (1999)
There are more references available in the full text version of this article.

Cited by (0)

View full text