CONTEXT: Children with tympanostomy tubes often develop ear discharge.
OBJECTIVE: Synthesize evidence about the need for water precautions (ear plugs or swimming avoidance) and effectiveness of topical versus oral antibiotic treatment of otorrhea in children with tympanostomy tubes.
DATA SOURCES: Searches in Medline, the Cochrane Central Trials Registry and Cochrane Database of Systematic Reviews, Excerpta Medica Database, and the Cumulative Index to Nursing and Allied Health Literature.
STUDY SELECTION: Abstracts and full-text articles independently screened by 2 investigators.
DATA EXTRACTION: 25 articles were included.
RESULTS: One randomized controlled trial (RCT) in children assigned to use ear plugs versus no precautions reported an odds ratio (OR) of 0.68 (95% confidence interval, 0.37–1.25) for >1 episode of otorrhea. Another RCT reported an OR of 0.71 (95% confidence interval, 0.29–1.76) for nonswimmers versus swimmers. Network meta-analyses suggest that, relative to oral antibiotics, topical antibiotic–glucocorticoid drops were more effective: OR 5.3 (95% credible interval, 1.2–27). The OR for antibiotic-only drops was 3.3 (95% credible interval, 0.74–16). Overall, the topical antibiotic–glucocorticoid and antibiotic-only preparations have the highest probabilities, 0.77 and 0.22 respectively, of being the most effective therapies.
LIMITATIONS: Sparse randomized evidence (2 RCTs) and high risk of bias for nonrandomized comparative studies evaluating water precautions. Otorrhea treatments include non–US Food and Drug Administration approved, off-label, and potentially ototoxic antibiotics.
CONCLUSIONS: No compelling evidence of a need for water precautions exists. Cure rates are higher for topical drops than oral antibiotics.
- CrI —
- Bayesian credible interval
- NNT —
- number needed to treat
- NRCS —
- nonrandomized comparative study
- OR —
- odds ratio
- RCT —
- randomized controlled trial
Children with tympanostomy tubes often develop ear discharge. A recent large cohort study found that 52% of children experienced ≥1 episodes of otorrhea after tympanostomy tube placement.1 Cultures of ear discharge often implicate bacteria commonly associated with acute otitis media, including nontypeable Haemophilus influenza, Streptococcus pneumoniae, and Moraxella catarrhalis. These infections may be otherwise asymptomatic and less troublesome than episodes of acute otitis media in children with intact eardrums.
In some patients, cultures reveal Staphylococcus aureus and Pseudomonas aeruginosa, which are likely to have entered the middle ear via the auditory canal through the tympanostomy tube.2 Thus, there has been concern that water exposure during swimming or bathing might increase the risk for otorrhea. Surveys of otolaryngologists, pediatricians, and family medicine physicians have found wide variation in the advice regarding the need for water precautions. Some physicians suggest avoiding swimming entirely, and others recommend using a barrier device (ie, earplug or swim cap) while swimming or bathing. A survey published in 2008 found that many otolaryngologists (47%) and most primary care physicians (73%) recommended use of a barrier device.3 Recent clinical practice guidelines have discouraged routine prophylactic water precautions (use of earplugs or headbands, avoidance of swimming) for children with tympanostomy tubes.4 However, given the continued discrepancy between guidelines and practice, it is of interest to systematically evaluate the available evidence.5
Otorrhea may be associated with a foul odor, fever, or pain, and it may thus negatively affect quality of life. Treatment of otorrhea is aimed at eradicating bacterial infection and reducing the duration and severity of symptoms. Therapeutic options include oral antibiotics and antibiotic eardrops with or without glucocorticoids. Clinical practice guidelines from the American Academy of Otolaryngology–Head and Neck Surgery recommend that clinicians prescribe topical antibiotic eardrops only, without oral antibiotics, for children with uncomplicated acute tympanostomy tube otorrhea.4 Despite these recommendations, a recent survey found that 54% of pediatric emergency physicians would use oral antibiotics to treat tube otorrhea, compared with 9% of pediatric ear, nose, and throat specialists.6
The objectives for this systematic review were to address 2 questions: Are water precautions necessary for children with tympanostomy tubes? When otorrhea develops, what is the effectiveness of topical drops compared with watchful waiting and oral antibiotics?
This review is derived from an Agency for Healthcare Research and Quality–commissioned comparative effectiveness review (Tympanostomy Tubes in Children With Otitis Media) conducted by the Brown Evidence-Based Practice Center. The full review and review protocol (PROSPERO registry number CRD42015029623) are available at http://www.effectivehealthcare.ahrq.gov.
We followed the approach outlined in the Agency for Healthcare Research and Quality’s Methods Guide for Comparative Effectiveness Reviews.7
Search Strategy and Study Selection
For water precautions, we included randomized controlled trials (RCTs) and nonrandomized comparative studies (NRCSs) in children with tympanostomy tubes placed for otitis media with effusion or acute otitis media. For treatment of otorrhea, we included studies of symptomatic or asymptomatic children with acute tympanostomy tube otorrhea beyond the immediate postoperative period. We defined the immediate postoperative period as 30 days after surgery, but we included studies reporting results near that period (eg, 28 days, 4 weeks). We excluded trials enrolling children with early postoperative otorrhea or chronic suppurative otitis media.
We conducted literature searches in Medline, the Cochrane Central Trials Registry and Cochrane Database of Systematic Reviews, Excerpta Medica Database, and the Cumulative Index to Nursing and Allied Health Literature (from inception) to May 19, 2016 to identify primary research studies meeting our criteria. Citations found by literature searches were independently screened by 2 researchers using the open-source online software Abstrackr (http://abstrackr.cebm.brown.edu/).8 Conflicts were resolved by discussion until a group consensus was reached.
Data Extraction and Analysis
Each study was extracted by 1 methodologist. The extractions were reviewed and confirmed by ≥1 other methodologist. We conducted direct pairwise random effects meta-analyses of NRCSs in the R9 package metafor (R Foundation for Statistical Computing, Vienna, Austria).10,11 We conducted network meta-analyses in the Bayesian framework, using the R gemtc package.12 Estimation was done with Markov chain Monte Carlo via the JAGS software,13 based on initial values drawn randomly from the marginal distributions of the priors of respective parameters. We fit 4 Markov chain Monte Carlo chains. After a burn-in of 5000 iterations, we monitored the convergence of random effects means and variances automatically, by checking every 10 000 iterations whether the Gelman–Rubin diagnostic was <1.05 with 95% probability for all monitored parameters. After convergence was reached, another 10 000 iterations were run. All models converged within 10 000 iterations. Model fit was assessed by comparison of the posterior mean of the residual deviance to the number of data points.
Network meta-analysis is an extension of pairwise meta-analyses that simultaneously combines direct (when interventions are compared head to head) and indirect (when interventions are compared through other reference interventions) evidence. Combining the direct and indirect evidence not only improves precision of estimates but also provides estimates for all pairwise comparisons, including those missing from the direct evidence.14 Statistical heterogeneity was explored qualitatively. Because of the small number of studies and the little variability in characteristics, meta-regression analyses were not performed.
We report odds ratios (ORs) for treatment effect estimates. We translated ORs to number of children needed to treat (NNT) to avoid an episode of (recurrent) otorrhea assuming that the baseline frequency of events was the same as seen in 3 specific studies.15–17
Assessment of Study Risk of Bias and Strength of Evidence
We assessed the methodological quality of each study based on predefined criteria. We used the Cochrane Risk of Bias Tool18 for RCTs and the Newcastle Ottawa Scale19 for nonrandomized studies. We graded the strength of evidence as per the Agency for Healthcare Research and Quality methods guide on assessing the strength of evidence.20
Figure 1 shows the results of the literature search and selection process.
Need for Water Precautions
We identified 11 publications that reported results from 2 RCTs21,22 and 9 NRCSs,23–31 evaluating water restrictions (eg, no swimming or head immersion while bathing) and physical protection while swimming (using ear plugs or bathing caps). One RCT evaluated the effectiveness of earplugs while swimming or bathing and reported a slightly higher average rate of otorrhea per month in children who did not wear ear plugs (mean 0.10 episodes per month, compared with a mean of 0.07; P = .05) after adjusting for compliance. The OR for having >1 episode of otorrhea for children using ear plugs compared with no precautions was 0.68 (95% confidence interval, 0.37–1.25).21 The second RCT evaluated otorrhea in children told to avoid swimming compared with those who did not avoid swimming. They reported an OR for >1 episode of otorrhea in nonswimmers versus swimmers of 0.71 (95% confidence interval, 0.29–1.76).22
The forest plot in Fig 2 summarizes the random effects meta-analyses of NRCSs, with separate summary estimates for ear plugs and avoidance of swimming. The available evidence does not support the conclusion that either ear plugs or avoidance of swimming reduces the risk of otorrhea.
All NRCSs had high risk of confounding bias, given that patient assignment was based on parent or patient choice.
Treatment of Tympanostomy Tube Otorrhea
We identified 12 articles, representing 11 studies reporting results from 10 RCTs and 1 NRCS, with a total of 1811 patients analyzed (1405 in RCTs and 406 in NRCSs), that assessed the effectiveness of various interventions to treat otorrhea, defined as the proportion clinically cured 7 to 20 days after initiation of treatment.
Eleven studies reported the number of clinically cured patients in each arm. Four studies were excluded. Of these, 1 compared ofloxacin ear drops with historical practice or undefined current practice,32 another compared 2 antibiotic–glucocorticoid preparations (drops versus spray),33 and 2 studies evaluated combination treatments (oral amoxicillin with and without oral prednisolone34 and antibiotic–glucocorticoid drops with or without an oral antibiotic).35 Seven studies were included in the network meta-analysis.15–17,34,36–38 The network of treatment comparisons is shown in Fig 3. As shown in Table 1, treatment strategies that include topical antibiotic drops predominate over both oral antibiotics and watchful waiting or placebo.
Both topical antibiotic–glucocorticoid and antibiotic-only drops are superior to watchful waiting (Fig 4). The odds of clinical cure were 12 (95% credible interval [CrI], 1.9–82) times higher (number needed to treat 2.2, assuming a reference rate of 0.4515) for antibiotic–glucocorticoid drops and 7.3 (95% CrI, 1.2–51) times higher (NNT 2.5, assuming a reference rate of 0.4516) for topical antibiotic drops.
When compared with oral antibiotics (Fig 5), treatment with topical antibiotic–glucocorticoid drops demonstrated higher effectiveness, OR 5.3 (95% CrI, 1.2–27; NNT 3.2, assuming a reference rate 0.5615). The OR for topical antibiotic drops was 3.3 (95% CrI, 0.74–16; NNT 5, assuming a reference rate of 0.6917). Overall, the topical antibiotic–glucocorticoid and antibiotic-only preparations have the highest probabilities, 0.77 and 0.22 respectively, of being the most effective therapies (Table 1).
The RCTs generated and concealed the allocation sequence via appropriate methods. However, 8 of 10 studies had an open label design, which precluded blinding of personnel and care providers. The main conclusions and associated interpretations based on our meta-analysis are summarized in Table 2.
Evidence about water precautions is sparse, and aside from 2 RCTs it comes primarily from NRCSs that have high risk of confounding bias. One RCT with moderate risk of bias reported a slightly higher average rate of otorrhea (after adjusting for compliance) in children who did not wear ear plugs. A second RCT with high risk of bias found no difference in otorrhea rates in swimmers compared to nonswimmers. A meta-analysis of NRCSs which evaluated ear protection and nonswimming tended to favor no precautions and swimming, but these studies are subject to high risk of bias, and the summary estimates did not exclude a null effect.
Our network meta-analysis provides evidence for greater efficacy of topical treatment of acute tympanostomy tube otorrhea compared with watchful waiting or oral antibiotics. This conclusion is congruent with 2 independent meta-analyses of studies that directly compared topical drops with oral antibiotics.39,40
The evidence about the need for water precautions is sparse and consists mainly of nonrandomized comparative studies at high risk of bias.
We have used indirect evidence from network meta-analysis to augment the direct evidence relating to the comparisons of interest for treatment of otorrhea. The key assumption of the network meta-analysis is that of consistency of direct and indirect effects. Consistency is likely to hold when the distribution of effect modifiers is similar across trials. If this assumption is violated, there may be inconsistency between the direct evidence and indirect evidence of treatment comparisons.
Our network meta-analysis combines trials of fluoroquinolones with other non–US Food and Drug Administration approved preparations. This combination presumes equivalent effectiveness and does not consider variable potential for adverse effects, such as ototoxicity, which may be associated with some agents. Our synthesis includes trials enrolling nonimmunocompromised patients with acute uncomplicated tympanostomy tube otorrhea. Our conclusions should not be extrapolated to immunocompromised patients or children with “complicated” features (eg, occluded auditory canal, associated cellulitis, and systemic symptoms such as fever). These findings do not apply to patients with chronic otorrhea and interventions for otorrhea occurring in the immediate postoperative period, because our review specifically excluded these groups.
Future Research Needs
With regard to water precautions, we agree with Cochrane Ear Nose and Throat Group’s conclusion (based on their review of the 2 available RCTs) that future high-quality studies could be undertaken but may not be thought necessary given the low baseline rate and morbidity of ventilation tube otorrhea and the associated burdens of any intervention aimed at reducing this rate.41
For otorrhea treatment, given evidence for the superiority of topical treatment, knowledge translation efforts are needed that focus on implementation and dissemination. Health service research is needed to identify important factors (eg, cost or formulary availability) that contribute to continued off-label use of nonsterile or potentially ototoxic agents. Future studies of topical agents are needed to clarify the effectiveness of topical antibiotic–glucocorticoid preparations compared with topical antibiotics alone.
In children in whom tympanostomy tubes have been placed, there is no compelling evidence for the need to avoid swimming or bathing or use ear plugs or bathing caps.
Should otorrhea develop, the available evidence supports topical treatment of acute, uncomplicated tympanostomy tube otorrhea in preference to watchful waiting or oral antibiotics.
Special thanks to Joseph Lau for his advice and guidance; Ian Pan and Nathan Coopersmith for assistance with screening and extraction; and Jenni Quiroz and Jenna Legault and for their assistance throughout the systematic review process.
- Accepted March 15, 2017.
- Address correspondence to Dale W. Steele, MD, MS, Center for Evidence Synthesis in Health, Brown University School of Public Health, Box G-S121-8, Providence, RI 02912. E-mail:
The authors of this report are responsible for its content. Statements in the report should not be construed as endorsement by the Agency for Healthcare Research and Quality or the US Department of Health and Human Services.
FINANCIAL DISCLOSURE: All authors received funding for this project under contract HHSA290-2015-00002-I from the Agency for Healthcare Research and Quality, US Department of Health and Human Services.
FUNDING: Supported by the Agency for Healthcare Research and Quality (contract HHSA290-2015-00002-I).
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- Ridgeon E,
- Lawrence R,
- Daniel M
- Agency for Healthcare Research and Quality
- Wallace BC,
- Small K,
- Brodley CE,
- Lau J,
- Trikalinos TA
- R Development Core Team
- Wallace BC,
- Dahabreh IJ,
- Trikalinos TA,
- Lau J,
- Trow P,
- Schmid CH
- Plummer M
- Higgins JP,
- Altman DG,
- Gøtzsche PC, et al; Cochrane Bias Methods Group; Cochrane Statistical Methods Group
- Wells GA,
- Shea B,
- O’Connell D, et al
- Berkman ND,
- Lohr KN,
- Ansari M, et al
- Granath A,
- Rynnel-Dagöö B,
- Backheden M,
- Lindberg K
- Ruohola A,
- Heikkinen T,
- Meurman O,
- Puhakka T,
- Lindblad N,
- Ruuskanen O
- Copyright © 2017 by the American Academy of Pediatrics