Академический Документы
Профессиональный Документы
Культура Документы
A R T I C L E I N F O A B S T R A C T
Article history: Objective: Antibiotic treatment is the standard of care for tympanostomy tube otorrhea. This meta-
Received 16 March 2016 analysis aims to evaluate the efficacy of topical antibiotics with or without corticosteroids versus oral
Received in revised form 21 April 2016 antibiotics in the treatment of tube otorrhea in children.
Accepted 5 May 2016
Data Sources: MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials and ProQuest.
Available online 11 May 2016
Review Methods: The above databases were searched using a search strategy for randomized controlled
trials for optimal treatment of tube otorrhea in the pediatric population. PRISMA (Preferred Reporting
Keywords:
Items for Systematic reviews and Meta-Analyses) guidelines were followed. Primary outcome was cure
Tympanostomy
Tube otorrhea (i.e. clearance of otorrhea) at 2–3 weeks. Secondary outcomes were microbiological eradication and com-
Otitis media plications such as dermatitis and diarrhea. The incidence of these events was defined as dichotomous
variables and expressed as a risk ratio (RR) and number needed to benefit (NNTB) in a random-effects
model.
Results: We identified 1491 articles and selected 4 randomized controlled trials which met our inclu-
sion criteria. Topical treatment had better cure (NNTB = 4.7, pooled RR = 1.35, p < 0.001) and microbiological
eradication (NNTB = 3.5, pooled RR = 1.47, p < 0.001 among 3 of the studies) than oral antibiotics. Oral
antibiotics had higher risk of diarrhea (pooled RR = 21.5, 95% CI 8.00–58.0, p < 0.001, Number needed to
harm (NNTH) = 5.4) and dermatitis (pooled RR = 3.14, 95% CI 1.20–8.20, p = 0.019, NNTH = 32). The use
of topical steroids in addition to topical antibiotics was associated with a higher cure rate (pooled RR = 1.59,
p < 0.001 vs pooled RR = 1.57, p = 0.293).
Conclusion: Topical antibiotics should be the recommended treatment for management of tympanostomy
tube otorrhea in view of its significantly improved clinical and microbiological efficacy with lower risk
of systemic toxicity as compared to oral antibiotics. Further research is necessary to confirm the ben-
efits of topical corticosteroids as an adjunct to topical antibiotics.
© 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction children with otitis media with effusion or recurrent acute otitis
media [3].
Tympanostomy tube insertion is the most common Acute otorrhea is the most common observed complication of
otolaryngologic day-surgery procedure performed in the pediatric tympanostomy tube, with a mean incidence of 26% (range, 4%–
population. In the US, nearly 1 in 15 children would have under- 68%) in observational studies and up to 83% with prospective
gone a tympanostomy tube insertion by 3 years of age [1]. Indications surveillance [4,5]. Tube otorrhea is usually sporadic and painless [6],
for tube insertion include persistence of middle ear effusion, re- but may be accompanied with foul odor, pain and pyrexia [3]. It is
current middle ear infections, or infections recalcitrant to oral postulated to be a manifestation of a recurrent acute otitis media,
antibiotic therapy [2]. It has been shown to significantly restore with bacterial superinfection or infection. Streptococcus pneumoniae,
hearing, reduce effusion prevalence, reduce incidence of recur- Haemophilus influenzae, and Moraxella catarrhalis [7,8] are often im-
rence of otorrhea, and improve disease-specific quality of life for plicated as the predominant bacteria. Treatment is usually with
broad-spectrum antibiotics, which can be delivered either orally or
with topical eardrops.
Trials comparing topical and oral antibiotics in children with tube
* Corresponding author at: Department of Otolaryngology–Head and Neck Surgery,
National University Health System, 1E Kent Ridge Road, NUHS Tower Block Level 7,
otorrhea have had relatively small sample sizes. But indepen-
119228, Singapore. Tel.: +65 6772 5555; fax: +65 6779 5678. dently they suggest that otic drops are as effective as, or more
E-mail address: cheejeremy@gmail.com (J. Chee). effective than, oral treatment. Despite these evidence in the
http://dx.doi.org/10.1016/j.ijporl.2016.05.008
0165-5876/© 2016 Elsevier Ireland Ltd. All rights reserved.
184 J. Chee et al. / International Journal of Pediatric Otorhinolaryngology 86 (2016) 183–188
2.3. Statistical analysis Antibiotics reduce otorrhea, fever and pain from otitis media. A
meta-analysis performed by Rovers et al concluded that oral anti-
All statistical analyses were performed with STATA Version 13.0. biotics were more effective than placebo in the treatment of children
Meta-analysis of binomial data using the random-effects model was with acute otitis media and otorrhea (NNTB = 3) [15] with regard
performed to derive a summary estimate of relative risks. The to the above outcomes. In adults with chronic suppurative otitis
random-effects model was used because it takes into account both media, topical antibiotics are more efficacious in reducing otor-
variation caused by sampling error and also random variation of the rhea than either oral [16] or intramuscular [17]antibiotics, possibly
underlying effect sizes between studies. The effect of topical versus because the tympanic membrane perforation permits a higher local
oral antibiotics in terms of cure rates and microbiological eradica- drug concentration to be achieved with topical antibiotic therapy.
tion were calculated using relative risks with its p-value. Significant For pediatric patients with tube otorrhea, studies have shown failure
difference was set at p < 0.05 for all analyses. A fixed-effects rates of between 10% and 23% with ciprofloxacin-containing ear
(weighted with inverse variance) or a random-effects model was drops versus 20%–70% with oral antibiotics alone [12,13,18–20].
used where appropriate, after computing the chi-squared and I2 sta- The most recent clinical practice guidelines by the American
tistics. When p < 0.05 or I2 > 50%, the assumption of homogeneity Academy of Otolaryngology-Head and Neck Surgery suggest a strong
was rejected and a random effects model was adopted. Random- recommendation for topical antibiotics over oral antibiotics based
effect estimates were used when there was significant between- on Grade B aggregate evidence [21], based on 3 RCTs performed
study heterogeneity. Peter’s test was used to test for evidence of between 1998 and 2010. We included a fourth well-designed study
publication bias when the heterogeneity variance tau-squared was which was recently published, and conducted a meta-analysis to
less than 0.1. statistically analyze the utility of topical versus oral antibiotics in
J. Chee et al. / International Journal of Pediatric Otorhinolaryngology 86 (2016) 183–188 185
(n = 27 ) (n = 23 )
Studies included in
qualitative synthesis
(n = 4 )
Included
Studies included in
quantitative synthesis
(meta-analysis)
(n = 4 )
From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-
Analyses: The PRISMA Statement. PLoS Med 6(7): e1000097. doi:10.1371/journal.pmed1000097
order to achieve better evidence for topical antibiotics. There was a lack of well-reported data, it was not possible to statistically analyze
some heterogeneity in terms of study enrollment, inclusion and ex- the difference in time to resolution of otorrhea. However, the studies
clusion criteria. Subgroup analysis was difficult given the small which included a patient diary as part of its methodology (3 of 4)
number of trials. In two of the included trials, children who were did report a significant improvement in median time to cessation
pre-treated were allowed to participate, and group A strep and Pseu- of otorrhea in the topical arm. Although patients with negative
domonas infections were excluded [11,12]. There was also some cultures at the outset were excluded from the analysis for micro-
variation in terms of time of measurement of outcomes, as well as biological eradication, the calculated RR was similar for both cure
ototopic solutions used (i.e. different classes of antibiotics as well and microbiological eradication. This excluded group is unlikely to
as usage of steroids). In light of this high degree of heterogeneity have a large effect on the effect size calculated.
(I2 = 80.8%), a random effects model was utilized in our analysis. It has been postulated that there are three main reasons for the
However, the general outcomes were similar and we believe that efficacy of topical ear drops in the treatment of tube otorrhea. First,
the results of the analysis remain clinically significant. the mechanical removal or dilution of bacterial load may augment
Our meta-analysis showed a convincing benefit of topical over the efficacy of local antimicrobial activity for topical ear drops. This
oral antibiotics in achieving cessation of otorrhea, microbiological was seen in the study by Heslop et al [13] where mechanical saline
eradication and in the side-effect profile. Unfortunately, in view of rinsing had a lower failure rate as compared to oral amoxicillin (58%
186 J. Chee et al. / International Journal of Pediatric Otorhinolaryngology 86 (2016) 183–188
Table 2
ITT (microbiology)
ITT (microbiology)
ITT (microbiology)
Intention-to-treat
Quality of the 4 studies as assessed by Jadad score.
Analysis type References Year Randomization Blinding Description Quality
(ITT) (cure)
of dropouts score
ITT (cure)
ITT (cure)
ITT (cure)
Goldblatt et al 1998 2 0 1 3
Dohar et al 2006 2 0 1 3
Heslop et al 2010 2 0 0 2
van Dongen et al 2014 2 0 1 3
of cure (days)
assessment
Timing of
18–21
17–20
14
14
ear mucosa will be higher with topical antibiotics as compared to
oral antibiotics. The middle ear mucosa is traditionally considered
Duration of
treatment
10
7
7
10 mg/kg TID
Amoxicillin/
Amoxicillin/
Type of oral
Amoxicillin
clavulanate
clavulanate
clavulanate
4 drops BID
0.25 ml BID
5 drops TID
part of the therapy [26]. Another study showed that the addition
Ciprofloxacin 0.3%
Ciprofloxacin 0.3%
0.8%–colistin 2%
Hydrocortisone
Ofloxacin 0.3%
1%–bacitracin
in 1 or both ears
no comorbidity.
2014
2006
2010
Year
140/146
controls
treated
76/77
39/40
22/20
of systemic antibiotics.
There are concerns of ototoxicity with topical therapy in the
Table 1
Fig. 2. Forest plot of otorrhea cure rates in ototopical versus oral antibiotics.
Fig. 3. Forest plot of microbiological eradication rates in ototopical versus oral antibiotics.
fluoroquinolones have been shown to have very limited to no oto- topical treatment should be the first choice in treatment of tube otor-
toxic effects [31,32]. Furthermore, steroids have been shown to have rhea. In terms of healthcare economics, treatment of tube otorrhea
a protective effect on potential ototoxic effects of otic drops [33]. with topical antibiotics has been shown to have up to 12% savings
There were no reports of patients who complained of hearing loss as compared to oral antibiotics [34].
after starting ear drops in the 4 studies we analyzed.
This meta-analysis incorporates a small number of studies with 5. Conclusion
a small pooled sample size. Heterogeneity in study design and
conduct and variations in choice of topical antibiotic and steroids Ototopical therapy should be the first choice in the treatment
has resulted in relatively large study heterogeneity and broad con- of acute tube otorrhea in view of its excellent cure rates and mi-
fidence intervals. We have attempted to adjust for this via the crobiological eradication coupled with a safe side-effect profile. The
utilization of random effects models as well as via subgroup anal- addition of steroids appears to confer an advantage for both cure
ysis. While the studies were not double-blinded, we believe that rates as well as microbiological eradication. Further research is
the difference in delivery is one of the reasons why topical antibi- needed to identify which subsets of children are most likely to benefit
otics is more efficacious than oral antibiotics (as seen in the study from topical steroids in addition to topical antibiotics.
by Heslop et al [13] where patients randomized to receive topical
saline had a lower failure rate as compared to oral amoxicillin [58%
Conflict of interest
vs 70%]). This effect would not surface if a true double blind study
had been performed. Last but not least, the study done by Dohar
No sponsorships or competing interests have been disclosed for
et al [12] and Goldblatt et al [11] are indeed prone to inherent bias
this article.
due to industry sponsorship and potential conflict of interest.
However, we note that the effect sizes of these two studies are in
fact smaller than those by Heslop et al [13] and van Dongen et al References
[14] as seen in Figs. 2 and 3 – suggesting that the advantage of topical
treatment is indeed real. [1] M.D. Kogan, M.D. Overpeck, H.J. Hoffman, M.L. Casselbrant, Am. J. Public Health
90 (2000) 245–250.
We feel that the improvement in side-effect profile, coupled with [2] G.G. Browning, M.M. Rovers, I. Williamson, J. Lous, M.J. Burton, Cochrane
the increase in efficacy, provides reasonably strong evidence that Database Syst. Rev. (10) (2010) CD001801.
188 J. Chee et al. / International Journal of Pediatric Otorhinolaryngology 86 (2016) 183–188
[3] R.M. Rosenfeld, M.H. Bhaya, C.M. Bower, P.E. Brookhouser, M.L. Casselbrant, K.H. [19] P.S. Roland, L.S. Kreisler, B. Reese, J.B. Anon, B. Lanier, P.J. Conroy, et al., Pediatrics
Chan, et al., Arch. Otolaryngol. Head Neck Surg. 126 (2000) 585–592. 113 (2004) e40–e46.
[4] C. Ah-Tye, J.L. Paradise, D.K. Colborn, Pediatrics 107 (2001) 1251–1258. [20] L. Vaile, T. Williamson, A. Waddell, G. Taylor, Cochrane Database Syst. Rev. (2)
[5] M.M. Rovers, N. Black, G.G. Browning, R. Maw, G.A. Zielhuis, M.P. Haggard, Arch. (2006) CD001933.
Dis. Child. 90 (2005) 480–485. [21] R.M. Rosenfeld, S.R. Schwartz, M.A. Pynnonen, D.E. Tunkel, H.M. Hussey, J.S.
[6] D.J. Kay, M. Nelson, R.M. Rosenfeld, Otolaryngol. Head Neck Surg. 124 (2001) Fichera, et al., Otolaryngol. Head Neck Surg. 149 (2013) 8–16.
374–380. [22] Y. Huang, Z. Yang, L. Cartier, B. Cheung, R.J. Sawchuk, Antimicrob. Agents
[7] J. Dohar, Int. J. Pediatr. Otorhinolaryngol. 67 (2003) 1317–1323. Chemother. 51 (2007) 4336–4341.
[8] P.S. Roland, D.A. Parry, D.W. Stroman, Otolaryngol. Head Neck Surg. 133 (2005) [23] B. Barry, M. Muffat-Joly, P. Gehanno, Clin. Microbiol. Infect. 3 (Suppl. 3) (1997)
585–595. S37–S42.
[9] V. Badalyan, R.H. Schwartz, S.L. Scwhartz, P.S. Roland, Pediatr. Emerg. Care 29 [24] D.M. Campoli-Richards, J.P. Monk, A. Price, P. Benfield, P.A. Todd, A. Ward, Drugs
(2013) 203–208. 35 (1988) 373–447.
[10] D. Moher, A. Liberati, J. Tetzlaff, D.G. Altman, P. Group, BMJ 339 (2009) b2535. [25] R.G. Thomas, C. Ojano-Dirain, P.J. Antonelli, Laryngoscope 121 (2011) 1067–
[11] E.L. Goldblatt, J. Dohar, R.J. Nozza, R.W. Nielsen, T. Goldberg, J.D. Sidman, et al., 1071.
Int. J. Pediatr. Otorhinolaryngol. 46 (1998) 91–101. [26] C.M. Alper, J.E. Dohar, M. Gulhan, A. Ozunlu, D. Bagger-Sjobak, P.A. Hebda, et al.,
[12] J. Dohar, W. Giles, P. Roland, N. Bikhazi, S. Carroll, R. Moe, et al., Pediatrics 118 Arch. Otolaryngol. Head Neck Surg. 126 (2000) 165–173.
(2006) e561–e569. [27] S.N. Park, S.W. Yeo, Acta Otolaryngol 121 (2001) 808–812.
[13] A. Heslop, T. Lildholdt, N. Gammelgaard, T. Ovesen, Laryngoscope 120 (2010) [28] J.W. Kutz Jr., P.S. Roland, K.H. Lee, Expert Opin. Pharmacother. 14 (2013)
2516–2520. 2399–2405.
[14] T.M. van Dongen, G.J. van der Heijden, R.P. Venekamp, M.M. Rovers, A.G. Schilder, [29] P.A. Hebda, S. Yuksel, J.E. Dohar, Laryngoscope 117 (2007) 522–528.
N. Engl. J. Med. 370 (2014) 723–733. [30] G. Matz, L. Rybak, P.S. Roland, M. Hannley, R. Friedman, S. Manolidis, et al.,
[15] M.M. Rovers, P. Glasziou, C.L. Appelman, P. Burke, D.P. McCormick, R.A. Otolaryngol. Head Neck Surg. 130 (2004) S79–S82.
Damoiseaux, et al., Lancet 368 (2006) 1429–1435. [31] K.R. Kavanagh, K. Parham, S.R. Schoem, Arch. Otolaryngol. Head Neck Surg. 135
[16] S. Esposito, G. D’Errico, C. Montanaro, Arch. Otolaryngol. Head Neck Surg. 116 (2009) 238–241.
(1990) 557–559. [32] R. Samarei, Glob. J. Health Sci. 6 (2014) 144–149.
[17] S. Esposito, S. Noviello, G. D’Errico, C. Montanaro, Arch. Otolaryngol. Head Neck [33] S.K. Park, D. Choi, P. Russell, E.O. John, T.T. Jung, Laryngoscope 114 (2004)
Surg. 118 (1992) 842–844. 768–771.
[18] P.S. Roland, J.B. Anon, R.D. Moe, P.J. Conroy, G.M. Wall, S.J. Dupre, et al., [34] T.M. van Dongen, A.G. Schilder, R.P. Venekamp, G.A. de Wit, G.J. van der Heijden,
Laryngoscope 113 (2003) 2116–2122. Pediatrics 135 (2015) e1182–e1189.