Tympanocentesis is a procedure that can be used in the office by trained primary caregivers. It can provide benefit to patients by allowing for identification of the pathogens and their antimicrobial susceptibility, distinguishing between viral and bacterial infection, and draining the effusion or abscess. It is indicated in febrile children with a toxic appearance, those with ear pain and a bulging tympanic membrane, and those whose otitis has failed to respond to sequential antimicrobial therapy. [Infect Med. 2001;18:363-366]

Before the discovery of antimicrobial agents, tympanocentesis was a common procedure used not only by otolaryngologists but also by pediatricians.¹ This was often the only effective means of treating an ear infection. The procedure allows for drainage of fluid from the middle ear, and it provides immediate relief of symptoms such as pain,  irritability, and fever. In the preantimicrobial period, many practicing pediatricians and family practitioners were familiar with this procedure and performed it in their clinics. However, when antimicrobial agents became available in the mid-1940s, the procedure became less useful, since the antimicrobial agents were effective in most patients, inducing rapid resolution of the infection. Subsequently, tympanocentesis was used less often, being performed only in select cases, and mostly by otolaryngologists. 

In the past 2 decades, a significant increase in resistance to antimicrobial agents has occurred.² Resistance to many antimicrobials used for the treatment of acute otitis media developed in the predominant organisms causing this infection. Up to 40% of Haemophilus influenzae and more than 95% of Moraxella catarrhalis strains are currently producing the enzyme ß-lactamase³; 30% to 50% of Streptococcus pneumoniae isolates resist penicillin⁴ through alteration of their penicillin protein-binding site and concomitantly also resist other antimicrobials (macrolides, trimethoprim-sulfamethoxazole, and third-generation cephalosporins)⁴; and 5% to 10% of Streptococcus pyogenes strains resist macrolides.⁵

As a result of this growing resistance to antimicrobials, physicians have encountered increased difficulty in treating otitis media, and several reports have described an increased rate of complications such as mastoiditis.⁶ Because of these changes, there has been a resurgence in the use of diagnostic and therapeutic tympanocentesis. Training primary care physicians in this procedure offers them the flexibility of performing it promptly without the need to refer the patient to an otolaryngologist. This may save time, since such a referral can often delay the procedure significantly, thus delaying the diagnostic and therapeutic benefit. However, referral to an otolaryngologist is still appropriate for those physicians who are unfamiliar with the procedure or are unwilling to perform it and in cases where the procedure needs to be done by a specialist because of anatomic or other difficulty.

Recommendations in the literature
Several recent authoritative publications have endorsed the use of tympanocentesis. In 1997, the Committee on Infectious Diseases of the American Academy of Pediatrics recommended its use in patients in whom standard therapy has failed and who are at risk for infection with resistant S pneumoniae.⁷

More recently (in 1999), the Drug-resistant Streptococcus pneumoniae Therapeutic Working Group of the CDC recommended that "in cases of clinically-defined treatment failure, consideration should be given to identification of the etiologic agent by tympanocentesis for susceptibility testing to guide alternative antibiotic therapy."⁸

Stool¹ outlined the details of office-performed tympanocentesis and suggested its use in children with severe ear pain, those with recurrent infection, and those with chronic infection in whom therapy fails. Howie and Ploussard⁹ used a secondary ear tap to study antibiotic cure rates in 280 children. Brook¹⁰ described a simple procedure that reduces the risk of contamination of the specimen by ear canal flora and maintains anaerobiosis to allow for the identification of anaerobic bacteria. Giebink et al¹¹ suggested the use of tympanocentesis to exclude viral infection; Berman¹² recommended it for unresponsive otitis media; Klein¹³ suggested it for an ill-appearing child who may benefit from identification of pathogens and determination of their susceptibility; and Block¹⁴ offered its use as an adjunctive intervention for refractory cases. 

Pichichero and Pichichero¹⁵ were able to define those patients who had no bacteria in their middle ear and therefore required no antimicrobial therapy. Such patients include about a third of those with acute otitis media and half of those with persistent and recurrent infection. Hoberman et al¹⁶ reviewed the proper technique of performing tympanocentesis using a disposable aspirator. Brook and Gober¹⁷ demonstrated the usefulness of tympanocentesis in determining the cause of failure of antimicrobials to clear the ear infection and in determining antimicrobial susceptibility.

Indications in primary care
The indications for tympanocentesis include:

• Acute, severe otalgia in the presence of a bulging tympanic membrane (in children younger than 4 years).
• Febrile, toxic-appearing child.
• Suppurative complications (to establish the diagnosis and treatment).
• Failure of antimicrobial therapy to clear the infection. The higher the number of failures, the more the procedure is indicated.
• Immunocompromised state.
• Identification of unusual pathogens in neonates.

Drainage of the ear can provide immediate pain relief and rapidly reduce fever. We have been able to abort fever and seizures in children who presented with febrile convulsions. The procedure also significantly improves the clinical condition of a toxic-appearing child, in whom accurate identification of the pathogen is of utmost importance. Removal of pus from the middle ear may reduce systemic toxicity. Removal of fluid that may continue to accumulate and cause serious otitis media may be beneficial. Although this is not yet proved, tympanocentesis may prevent complications, such as the development of mastoiditis, bacteremia, and meningitis.

In instances in which antimicrobials have failed to resolve the infection, the identification of bacterial pathogens and their antimicrobial susceptibility makes possible the proper selection of antimicrobial therapy.¹⁵ In instances in which no bacterial pathogens (aerobic or anaerobic) are identified, a viral infection can be assumed to be present. Since the microbial causes of otitis media in immunocompromised patients and newborns may be diverse and may include atypical or low-virulence organisms, and the risk of complication is higher, myringotomy may be indicated in those patients even before the first course of therapy. It should also be done in hospitalized patients in the presence of suppurative complications, including mastoiditis and meningitis.

Contraindications
The procedure should not be done in the office in those with bleeding diathesis, in those who cannot be safely restrained, or when the tympanic membrane cannot be visualized or is thick or scarred. Those cases should be referred to an otolaryngologist for performance of the procedure.

Performing tympanocentesis
We do not routinely use topical anesthesia and only rarely use mild sedation in the performance of tympanocentesis. Analgesia can be used when indicated, with acetaminophen and codeine or with diazepam. (When diazepam is used, the child must be on an oxygen monitor with 2 persons trained in cardiopulmonary resuscitation present, because there is a risk of aspiration.) After the child is restrained appropriately (a papoose should be used for restraint in those younger than 4 years), the external auditory canal is cleaned with a blunt curette.

Cleansing of the canal is the most crucial determinant of success with the procedure. Without clear visualization of the tympanic membrane, tympanocentesis is contraindicated. When complete antisepsis is warranted, the external auditory canal can be cleaned with povidone-iodine solution, followed by copious irrigation with saline. Thereafter, the external auditory canal is dried of excess fluid using sterile cotton swabs and is allowed to dry for 2 to 3 minutes. After antisepsis of the external ear canal, a culture should be obtained to ensure its sterility. 

An 18- or 20-guauge spinal needle is used. After the sylet is removed, it is bent to a 45-degree angle about an inch from the syringe and is connected to a 2-mL syringe. Alternatively, an intravenous cannula set (Medicut, Sherwood Medical Instrument, Inc., St. Louis) can be used; this consists of an 18- or 20-gauge needle attached to a 2-mL syringe covered by a plastic cannula that is slipped forward to cover the bevel.

An operating otoscope should be used. The otoscope is held by the left hand, and the cannula set by the right. The needle is then directed through the otoscope to the tympanic membrane. When the needle is adjacent to the tympanic membrane, the cannula is retracted and the membrane is penetrated in the anterior inferior quadrant; negative pressure is then applied to the syringe. The needle tip does not make contact with the speculum or auditory canal. There is no need to seal the ear canal. However, if drainage persists, cotton can be temporarily placed. Alternatively, a disposable aspirator connected to a suction device can be used. 

The removed fluid is cultured as soon as possible (preferably immediately) for aerobic and facultative bacterial using blood and chocolate agar media. In instances of chronic infection, it is advisable to culture also for anaerobic organisms.¹⁸ Gram stain should be done. If only a small amount of fluid is aspirated, the needle should be rinsed with sterile saline (without inhibitory substances). The content of the syringe should be plated on a solid agar medium, not into blood culture fluid. This will ensure semiquantitave identification of the pathogen(s) and rapidly differentiate multiple pathogens if present. 

The procedure and its benefits and risks should be openly discussed with parents and when possible also with the child, and informed consent should be obtained. Patients generally tolerate the procedure well. In instances in which the drum is bulging and the tympanic membrane is thin, there seems to be no perception of pain. However, minimal discomfort may be experienced in other instances. The perforation in the membrane generally seals itself within 24 to 48 hours. The procedure can be taught by an experienced physician or an otolaryngologist. A short hands-on course that teaches the tympanocentesis is offered now across the country by a group of specialists (OMEW). Once the procedure is mastered, it can be performed in the clinic.

Risks and complications
Complications associated with tympanocentesis occur only rarely. They include dislocation of ossicles; entering an aberrant jugular bulb and causing excessive bleeding; persistent perforation; leakage of blood and pus for several hours or, rarely, for several days. When present, bleeding is generally self-limited and ceases within 1 to 2 hours. A follow-up visit to examine the ear is recommended 7 to 10 days later, or sooner if indicated. 

In more than 1000 tympanocentesis procedures performed over the past 25 years, I have encountered no major complications; minor, self-limited bleeding occurred in fewer than 5% of the procedures.

References

1. Stool SE. Myringotomy: an office procedure. Clin Pediatr. 1968; 7:470-473
2. Jacoby GA. Prevalence and resistance mechanisms of common bacterial respiratory pathogens. Clin Infect Dis. 1994; 18:951-957.
3. Richter SS, Brueggmann AB, Huynh HK, et al. A 1997-1998 national surveillance study: Moraxella catarrhalis and Haemophilus influenzae antimicrobial resistance in 34 US institutions. Int J Antimicrob Agents. 1999;13:99-107.
4. Jacobs MR, Bajaksouzian S, Zilles A, et al. Susceptibilities of Streptococcus pneumoniae and Haemophilus influenzae to 10 oral antimicrobial agents based on pharmocodynamic parameters: 1997 US Surveillance study. Antimicrob Agents Chemother. 1999;43:1901-1908.
5. Cizman M, Pokorn M, Seme I, et al. Influence of increased macrolide consumption on macrolide resistance of common respiratory pathogens. Eur J Clin Microbial Infect Dis. 1999;18:522-524.
6. Hoppe JE, Koster S, Bootz F, Niethammer D. Acute mastoidistis - relevant once again. Infection. 1994;22:178-182.
7. American Academy of Pediatrics. Report of the Committee on Infectious Diseases. In: Peter G, ed. 1997 Red Book. 24th ed. Elk Grove Village, Ill: American Academy of Pediatrics; 1997.
8. Dowell SF, Butler JC, Giebink GS, et al. Acute otitis media: management and surveillance in an era of pneumococcal resistance - a report from the Drug-resistant Streptococcus pneumoniae Therapeutic Working Group. Pediatr Infect Dis J. 1999;18:1-9.
9. Howie VM, Ploussard JH. Efficacy of fixed combination antibiotics versus separate components in otitis media. Effectiveness of erythromycin estolate, triple sulfonamide, ampicillin, erythromycin estolate-triple sulfonamide, and placebo in 280 patients with acute otitis media under two and one-half years of age. Clin Pediatr (Phila). 1972;11:205-214.
10. Brook I. Apractical technique for tympanocentesis for culturing aerobic and anaerobic bacteria. Pediatrics. 1980;65:626-627.
11. Giebink GS, Canafax DM, Kempthorne J. Antimicrobial treatment of acute otitis media. J Pediatr. 1991;119:495-500.
12. Berman S. Otitis media in children. N Engl J Med. 1995;332:1560-1565.
13. Klein JO. Otitis media. Clin Infect Dis. 1994;19:823-833.
14. Block SL. Strategies for dealing with amoxicillin failure in acute otitis media. Arch Fam Med. 1999;8:68-78.
15. Pichichero ME, Pichichero CL. Persistent acute otitis media, I: causative pathogens. Pediatr Infect Dis J. 1995;14:178-183.
16. Hoberman A, Paradise JL, Wald ER. Tympanocentesis technique revisited. Pediatr Infect Dis J. 1997;16(suppl 2):S25-S26.
17. Brook I, Gober AE. Microbiologic characteristics of persisitent otitis media. Arch Otolaryngol Head Neck Surg. 1998;24:1350-1352.
18. Brook I, Finegold SM. Bacteriology of chronic otitis media. JAMA. 1979;241:487-488.