Evidence Profile: Referral

Aggregate evidence quality: C, observational studies.

Benefit: better communication and improved decision-making.

Harm: confidentiality concerns, administrative burden, and/or increased parent or caregiver anxiety.

Benefits-harms assessment: balance of benefit and harm.

Policy level: option.

SURGERY: WHEN A CHILD BECOMES A SURGICAL CANDIDATE, TYMPANOSTOMY TUBE INSERTION IS THE PREFERRED INITIAL PROCEDURE; ADENOIDECTOMY SHOULD NOT BE PERFORMED UNLESS A DISTINCT INDICATION EXISTS (NASAL OBSTRUCTION, CHRONIC ADENOIDITIS). REPEAT SURGERY CONSISTS OF ADENOIDECTOMY PLUS MYRINGOTOMY, WITH OR WITHOUT TUBE INSERTION. TONSILLECTOMY ALONE OR MYRINGOTOMY ALONE SHOULD NOT BE USED TO TREAT OME

This recommendation is based on randomized, controlled trials with a preponderance of benefit over harm.

Surgical candidacy for OME largely depends on hearing status, associated symptoms, the child's developmental risk, and the anticipated chance of timely spontaneous resolution of the effusion. Candidates for surgery include children with OME lasting 4 months or longer with persistent hearing loss or other signs and symptoms, recurrent or persistent OME in children at risk regardless of hearing status, and OME and structural damage to the tympanic membrane or middle ear. Ultimately, the recommendation for surgery must be individualized based on consensus between the primary care physician, otolaryngologist, and parent or caregiver that a particular child would benefit from intervention. Children with OME of any duration who are at risk are candidates for earlier surgery.

Tympanostomy tubes are recommended for initial surgery because randomized trials show a mean 62% relative decrease in effusion prevalence and an absolute decrease of 128 effusion days per child during the next year. HLs improve by a mean of 6 to 12 dB while the tubes remain patent. Adenoidectomy plus myringotomy (without tube insertion) has comparable efficacy in children 4 years old or older but is more invasive, with additional surgical and anesthetic risks. Similarly, the added risk of adenoidectomy outweighs the limited, short-term benefit for children 3 years old or older without prior tubes. Consequently, adenoidectomy is not recommended for initial OME surgery unless a distinct indication exists, such as adenoiditis, postnasal obstruction, or chronic sinusitis.

Approximately 20% to 50% of children who have had tympanostomy tubes have OME relapse after tube extrusion that may require additional surgery. When a child needs repeat surgery for OME, adenoidectomy is recommended (unless the child has an overt or submucous cleft palate), because it confers a 50% reduction in the need for future operations. The benefit of adenoidectomy is apparent at 2 years old, greatest for children 3 years old or older, and independent of adenoid size. Myringotomy is performed concurrent with adenoidectomy. Myringotomy plus adenoidectomy is effective for children 4 years old or older, but tube insertion is advised for younger children, when potential relapse of effusion must be minimized (eg, children at risk) or pronounced inflammation of the tympanic membrane and middle-ear mucosa is present.

Tonsillectomy or myringotomy alone (without adenoidectomy) is not recommended to treat OME. Although tonsillectomy is either ineffective or of limited efficacy, the risks of hemorrhage (~2%) and additional hospitalization outweigh any potential benefits unless a distinct indication for tonsillectomy exists. Myringotomy alone, without tube placement or adenoidectomy, is ineffective for chronic OME, because the incision closes within several days. Laser-assisted myringotomy extends the ventilation period several weeks, but randomized trials with concurrent controls have not been conducted to establish efficacy. In contrast, tympanostomy tubes ventilate the middle ear for an average of 12 to 14 months.

Anesthesia mortality has been reported to be ~1:50 000 for ambulatory surgery, but the current fatality rate may be lower. Laryngospasm and bronchospasm occur more often in children receiving anesthesia than adults. Tympanostomy tube sequelae are common but are generally transient (otorrhea) or do not affect function (tympanosclerosis, focal atrophy, or shallow retraction pocket). Tympanic membrane perforations, which may require repair, are seen in 2% of children after placement of short-term (grommet-type) tubes and 17% after long-term tubes. Adenoidectomy has a 0.2% to 0.5% incidence of hemorrhage and 2% incidence of transient velopharyngeal insufficiency. Other potential risks of adenoidectomy, such as nasopharyngeal stenosis and persistent velopharyngeal insufficiency, can be minimized with appropriate patient selection and surgical technique.

There is a clear preponderance of benefit over harm when considering the impact of surgery for OME on effusion prevalence, HLs, subsequent incidence of AOM, and the need for reoperation after adenoidectomy. Information about adenoidectomy in children less than 4 years old, however, remains limited. Although the cost of surgery and anesthesia is nontrivial, it is offset by reduced OME and AOM after tube placement and by reduced need for reoperation after adenoidectomy. Approximately 8 adenoidectomies are needed to avoid a single instance of tube reinsertion; however, each avoided surgery probably represents a larger reduction in the number of AOM and OME episodes, including those in children who did not require additional surgery.