Adenoidectomy Does Not Significantly Reduce the Incidence of Otitis Media in Conjunction With the Insertion of Tympanostomy Tubes in Children Who Are Younger Than 4 Years: A Randomized Trial
Objective. To evaluate the efficacy of adenoidectomy in reducing the incidence of otitis media among children who are younger than 4 years and receive tympanostomy tubes.
Methods. A randomized trial was conducted at a tertiary center clinic. A total of 217 children who were aged 12 to 48 months and had recurrent acute otitis media (>3 episodes during the past 6 months) or chronic otitis media with effusion, no obstructive symptoms as a result of adenoid enlargement, and no previous surgical intervention were enrolled in the study. Adenoidectomy in conjunction with the insertion of tympanostomy tubes or insertion of tympanostomy tubes without adenoidectomy was studied. The number of otitis media episodes during the follow-up period of 12 months was measured.
Results. During the follow-up, the mean number of otitis media episodes was 1.7 among children who underwent adenoidectomy with concurrent insertion of tympanostomy tubes and 1.4 among children who received tympanostomy tubes only. The risk for recurrent otitis media (≥3 episodes) could not be reduced by adenoidectomy (odds ratio: 1.66; 95% confidence interval: 0.80–3.46).
Conclusion. Adenoidectomy does not significantly reduce the incidence of acute otitis media in otitis prone children who are younger than 4 years and receive tympanostomy tubes.
Otitis media is one of the most common disorders of childhood.1,2 It is frequently followed by recurrent acute otitis media (RAOM) infections, and it may also lead to chronic otitis media with effusion (COME), characterized by persistent middle ear effusion. Both of these disorders are treated with the insertion of tympanostomy tubes to facilitate drainage of middle ear effusion.3,4
The rationale of adenoidectomy in the treatment of RAOM and COME has been to remove a potentially chronically infected focus in the pharynx.5 Adenoidectomy has been reported to be beneficial in treating children who are older than 4 years and have COME,6–8 but its effectiveness in treating children with RAOM has been controversial9–11 and its effect in COME among children who are younger than 4 years is unknown. Recently, adenoidectomy was shown to be ineffective in reducing the incidence of otitis media in children who were younger than 2 years in a study in which adenoidectomy was performed in conjunction with the insertion of tympanostomy tubes12 and in another survey in which adenoidectomy was performed as the sole procedure.13 The former was not a true randomized study as the children were enrolled after randomization was revealed to the parents, which may have resulted in a bias. The 95% confidence intervals (CIs) in both of these recent studies were relatively wide, suggesting that these studies might not have proved lack of efficacy.
We wanted to assess whether adenoidectomy causes any additional benefit when performed in conjunction with the insertion of tympanostomy tubes among children who are younger than 4 years, the population most frequently affected by otitis media.1,2 To test this, we conducted a randomized trial of the effect of adenoidectomy in conjunction with the insertion of tympanostomy tubes.
Enrollment and Assignment
Children with a history of otitis media were recruited to the study from patients who were referred to the Department of Otorhinolaryngology, Helsinki University Central Hospital, between March 2001 and December 2002 because of recurrent or persistent otitis media. The inclusion criteria were 1 to 4 years of age, RAOM (≥3 episodes of acute otitis media during the preceding 6 months or ≥5 episodes of acute otitis media during the preceding 12 months) or a suspicion of COME as judged by examination with a pneumatic otoscope, and no previous adenotonsillar surgery or placement of tympanostomy tubes. Children with asthma, cleft palate, or diabetes or children who were judged to require prompt removal of adenoids because of obstructive symptoms resulting in continuous mouth breathing or sleep apnea were excluded. Of the 1385 children evaluated, 296 met the eligibility criteria. The parents of 217 eligible children gave a written consent to participate in the study. These children were randomized in blocks of 8. The Ethical Review Committee of the Helsinki University Central Hospital approved the study.
The randomized children were assigned either to undergo adenoidectomy in conjunction with the insertion of tympanostomy tubes or to receive tympanostomy tubes without adenoidectomy. The operations were performed under general anesthesia using standard techniques as described previously.12 All children received Donaldson silicon tympanostomy tubes in both ears.
The children had scheduled visits at our hospital 1 month after the surgery and at the end of the follow-up (12 months). During the first follow-up visit, the ears were examined and the patency of the tympanostomy tubes was examined. Patient diaries were used to document otitis media episodes during the 12 months of follow-up. Whenever the child had symptoms of acute respiratory infections or the parents suspected otitis media, they were advised to visit a doctor at a primary care center. All episodes of acute otitis media and other respiratory infections during the visits were recorded. The criteria of acute otitis media consisted of acute symptoms and middle ear inflammation. Each acute episode of otitis media was treated with antibiotics. The parents were contacted by telephone 1 to 2 times during the follow-up to assess compliance. At the end of the 12 months of follow-up, the children were examined at our hospital and the diaries were collected.
The outcome measure was the number of otitis media episodes during the follow-up.
Diagnosis of COME and RAOM at the Time of Enrollment
The 217 enrolled children had tympanostomy tubes inserted on average 93 days (SD: 53 days) after the initial evaluation by the referring colleagues at primary care centers. Of the enrolled children, 57 children were suspected to have COME by the referring colleague. The diagnosis of COME in primary care centers in Finland is based on patient history and repeated pneumatic otoscope examinations generally 6 to 8 weeks apart. The diagnosis of COME could be verified in 55 of these 57 children by the presence of mucoid, serous, or purulent effusion in the middle ear during surgery. All of the remaining 162 children were classified as having RAOM.
The difference in the number of otitis media episodes between the randomization groups was evaluated using Students t test. The risk for frequent episodes of otitis media (≥3 episodes) was evaluated using logistic regression analysis adjusted by gender, age, number of siblings, previous number of otitis media episodes 12 months before surgery, parental smoking, and the clinical diagnosis of either RAOM or COME.
Altogether, 217 children were randomly allocated to the treatment groups (Fig 1). Of the 109 children who were allocated to undergo adenoidectomy with concurrent insertion of tympanostomy tubes, 5 did not receive the allocated treatment because their parents wanted to discontinue the trial before the allocated treatment was given. Two children discontinued the follow-up before the first month follow-up visit. Of the 108 children who were allocated to undergo the insertion of tympanostomy tubes without adenoidectomy, 5 did not receive the allocated treatment because their parents wanted to discontinue the trial before the allocated treatment was given. Seven of the remaining children were withdrawn or discontinued the follow-up. Four of these underwent adenoidectomy because of recurrent otitis media. One child underwent adenotonsillectomy because of obstructive symptoms. One child discontinued because she developed a neck abscess and later on type 1 diabetes. One child was lost during the follow-up.
The 2 randomization groups were similar at the end of the follow-up in gender distribution, mean age at operation, number of siblings, history of acute otitis media episodes, frequency of parental smoking, and frequency of persistent effusion in middle ear (Table 1). There was no significant difference between the randomization groups in the number of otitis media episodes during the follow-up year (Table 2). The mean number of otitis media episodes among children in the tympanostomy with concurrent adenoidectomy group was 1.73 (range: 1–8 episodes) versus 1.44 (range: 1–8 episodes) in the tympanostomy group. The odds ratio of the difference in the mean number of otitis media episodes between the randomization groups was 1.11 (95% CI: 0.94–1.32). Even when the randomization groups were subdivided by the frequency of previous episodes of otitis media (Table 2), by the clinical diagnosis of RAOM or COME (Table 2), or by the age of the patient (data not shown), concurrent adenoidectomy provided no benefit. Adenoidectomy also provided no benefit in decreasing the risk for frequent episodes of otitis media as defined by ≥3 episodes during the follow-up year (odds ratio: 1.66; 95% CI: 0.80–3.46). There was no significant difference in the frequency of reinsertion of tympanostomy tubes between the randomization groups. Ten children in the tympanostomy with concurrent adenoidectomy group and 8 children in the tympanostomy group underwent reinsertion of tympanostomy tubes.
The present survey indicated that adenoidectomy during concurrent insertion of tympanostomy tubes did not significantly reduce the incidence of acute otitis media when compared with plain tympanostomy. The results are in line with the study of Paradise et al,11 who showed that adenoidectomy was ineffective in reducing the incidence of acute otitis media among children who had not received tympanostomy tubes previously. Our results are also in line with the previous recommendation of the otitis media guideline panel of the American Academy of Pediatrics that does not recommend adenoidectomy in reducing the frequency of otitis media in children who are younger than 4 years.14
Altogether, 7 children in the tympanostomy-only treatment arm discontinued the study. During the follow-up, 4 of them underwent adenoidectomy as a result of recurrent otitis media. This may bias the results in favor of treatment with tympanostomy only as the 4 children who did not finish the follow-up in the tympanostomy-only treatment arm seemed to have frequent episodes of otitis media. This may explain why the mean number of otitis media episodes was lower in the tympanostomy-only group than in the concurrent adenoidectomy group. If we assume that these 4 children had had 7 to 8 episodes of otitis media during the entire follow-up, then the mean number of otitis media episodes would have been similar but not higher in the plain tympanostomy arm than in the concurrent adenoidectomy arm. Thus, this bias unlikely masked a potential beneficial effect of concurrent adenoidectomy.
Remarkably, although the study children were reported to have on average >5 acute otitis media episodes before surgery, during the follow-up year, the children had on average <2 acute otitis media episodes. As the peak incidence of otitis media is during the first and second years of life, this reduction in the number of otitis media episodes may reflect the nature of recurrent otitis media to resolve spontaneously as the child gets older. The insertion of tympanostomy tubes may have decreased the frequency of subsequent otitis media episodes as well.3,4 However, as both under- and overdiagnosis of otitis media occurs commonly, we cannot rule out the possibility that the children erroneously received a diagnosis of have otitis media more frequently before surgery than what actually occurred.
Coyte et al10 came to the conclusion that adenoidectomy would be favorable in reducing the incidence of childhood otitis media in a retrospective analysis of hospital discharge records of >37000 children. They found that children who had had adenoidectomy during the insertion of tympanostomy tubes had fewer hospitalizations and operations related to otitis media than children who had had the insertion of tympanostomy tubes alone as their first surgery. However, their analysis may be biased, as children who had adenoidectomy for their first surgery were considerably older during surgery than children who had had the insertion of tympanostomy tubes alone as their first surgery. As children who are at a high risk for otitis media are likely to receive surgical intervention at a young age, children who received tympanostomy alone may have been at a higher risk for otitis media than children who received adenoidectomy with concurrent tympanostomy, leading to a result favoring adenoidectomy in the retrospective analysis.
We excluded children with adenoid hyperplasia causing obstructive symptoms as judged by continuous mouth breathing. Although adenoidectomy in our setting provided no apparent benefit when no evident adenoid hyperplasia was present, adenoidectomy is likely beneficial to relieve nasopharyngeal obstruction in manifest adenoid hyperplasia. Adenotonsillar surgery has been shown to result in significant improvement in the quality of life among children with obstructive sleep disorders.15,16 These benefits of adenoidectomy must be weighted against the risks of possible perioperative complications of adenoidectomy.
We also analyzed whether the outcome was influenced by the age of the child, by the diagnosis of COME at the time of surgery, or by the frequency of otitis media episodes before the surgery. All of these subanalyses indicated that adenoidectomy was not efficient in reducing the incidence of acute otitis media. The results suggest that adenoidectomy during concurrent insertion of tympanostomy tubes does not significantly reduce the incidence of acute otitis media among children who are younger than 4 years.
This work received financial support from the Sigrid Juselius Foundation and the Helsinki University Central Hospital Research Funds.
- Accepted January 20, 2005.
- Reprint requests to (P.S.M.) Department of Otorhinolaryngology, Helsinki University Central Hospital, Haartmaninkatu 4 E, PO Box 220, 00290 Helsinki, Finland. E-mail:
No conflict of interest declared.
- ↵Lanphear BP, Byrd RS, Auinger P, Hall CB. Increasing prevalence of recurrent otitis media among children in the United States. Pediatrics.1997;99 (3). Available at: www.pediatrics.org/cgi/content/full/99/3/e1
- ↵Maw R, Bawden R. Spontaneous resolution of severe chronic glue ear in children and the effect of adenoidectomy, tonsillectomy, and insertion of ventilation tubes (grommets). BMJ.1993;306 :756– 760
- ↵Koivunen P, Uhari M, Luotonen J, et al. Adenoidectomy versus chemoprophylaxis and placebo for recurrent acute otitis media in children aged under 2 years: randomised controlled trial. BMJ.2004;328 :487
- ↵Guideline. Managing otitis media with effusion in young children. American Academy of Pediatrics. The Otitis Media Guideline Panel. Pediatrics.1994;94 :766– 773
- Copyright © 2005 by the American Academy of Pediatrics