Customize your JAMA Network experience by selecting one or more topics from the list below.
Hartnick CJ, Shott S, Willging JP, Myer CM. Methicillin-Resistant Staphylococcus aureus Otorrhea After Tympanostomy Tube Placement: An Emerging Concern. Arch Otolaryngol Head Neck Surg. 2000;126(12):1440–1443. doi:10.1001/archotol.126.12.1440
To review the treatment of pediatric patients with methicillin-resistant Staphylococcus aureus (MRSA)–positive cultures as a result of otorrhea after tympanostomy tube placement in terms of both medication and isolation strategies and to highlight an emerging problem faced by the clinician with reference to treatment options as well as to the treatment of these patients in an outpatient setting.
Between December 1998 and January 2000, a total of 8 children between the ages of 1 and 11 years had MRSA-positive cultures as a result of otorrhea after tympanostomy tube placement.
Main Outcome Measures
The Department of Infectious Diseases was notified, and a variety of topical antibiotic treatments were administered.
STAPHYLOCOCCAL infections are common findings in any medical practice. Since the advent of antibiotics, staphylococci have shown an impressive ability to develop resistance to the particular antibiotics to which they are exposed. In 1961, the first strains of methicillin-resistant Staphylococcus aureus (MRSA) were noted.1 By the end of the decade, hospital outbreaks of MRSA had been documented,2,3 and recent publications have cited an incidence of up to 61% of hospital-acquired staphylococcal infections.4 Vancomycin has been the antibiotic most commonly used to treat MRSA. However, it necessitates parenteral administration and has potential toxic effects. Alternative treatment strategies have been tried, with variable success.
Methicillin-resistant Staphylococcus aureus infections have been documented in the head and neck region in the otolaryngologic literature,5 but there have been few reports to date of MRSA otorrhea occurring after tympanostomy tube placement. Gottlieb et al5 reported on the association of MRSA with chronic otitis media but did not specify whether tympanostomy tube placement was involved. Suh et al6 reported a 23% incidence of otorrhea and an 8% incidence of MRSA otorrhea after middle ear surgery. In this study, we observed a 0.2% incidence of MRSA otorrhea after tympanostomy tube placement.
Otorrhea after tympanostomy tube placement is a relatively common occurrence, with the incidence cited at 21% to 34% of all children at some point after tympanostomy tube placement.7 In the context of roughly 1 million tympanostomy procedures occurring annually in the United States,8 there is a clear need for practical treatment algorithms for posttympanostomy tube otorrhea. Traditionally, the bacteriology of otorrhea has varied according to age, with children younger than 3 years having cultures that mirror those seen in otitis media (predominantly Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis) and children older than 3 years producing cultures with increased growth of S aureus and Pseudomonas aeruginosa.9 Antibiotic treatment of otorrhea varies according to patient age and route of administration. With isolated otorrhea, the usual therapy consists of either oral or topical antibiotics, or both. Ototopical antimicrobial therapy is complicated by the potential ototoxic effects of the antibiotics themselves or the solvents and antiseptics that are contained in the otic medication.10 With the growing concern of MRSA otorrhea as noted in this study, treatment regimens that are both effective and the least ototoxic are being sought.
Aside from specific treatment of the MRSA infection, the practical question arises as to how to manage these cases in an outpatient setting. The major mode of transmission of MRSA in the hospital setting is via direct contact between patients and health care workers. Colonization occurs at sites that include the anterior nares, axilla, and perirectal area. Protocol regarding isolation precautions for patients with MRSA varies from hospital to hospital and from clinic to clinic. Key questions that must be addressed include: (1) How strict does isolation for MRSA need to be? (2) How does this isolation vary from an in-hospital to an outpatient setting? and (3) How long is a patient considered MRSA positive?
Our purpose was to address these treatment and management issues in 8 cases of MRSA otorrhea after tympanostomy tube placement.
Eight patients underwent tympanostomy tube placement for chronic otitis media with effusion at Children's Hospital Medical Center, Cincinnati (CHMCC), Ohio. In our institution, the ear canals are cleaned of cerumen; iodine-containing solutions or other antiseptics are not routinely administered for tympanostomy tube placement. The patients ranged in age from 1 to 19 years (average age, 6.9 years). After tympanostomy tube placement, the patients developed refractory otorrhea, which, when cultured, demonstrated MRSA. In all cases, MRSA was the only organism yielded. Before MRSA identification, the patients were treated with a range of oral and ototopical medications. The oral antibiotic agents included amoxicillin, amoxicillin clavulanate, ciprofloxacin, and azithromycin. The ototopical agents included 1% hydrocortisone–polymyxin B sulfate–neomycin drops, -ciprofloxacin drops, gentamicin sulfate drops, and chloramphenicol drops. Once the results of the cultures were known, each patient's therapy was tailored according to the sensitivity panel. In all cases, the report of sensitivities was limited and did not include sensitivity to ciprofloxacin (this sensitivity was reported when the full panel was requested). Three of the 8 children were initially treated with ciprofloxacin drops and then placed on alternative regimens when the otorrhea failed to resolve. To date, 6 of the 8 children have had resolution of the otorrhea, 3 after using tobramycin and dexamethasone (TobraDex) drops (response after 4-7 days), and 3 after using chloramphenicol (Chloromycetin) otic solution (response after 5-7 days).
Of note, at the time of MRSA identification, surveillance cultures were obtained from the anterior nares, axilla, and perirectal area. The patients were then declared "MRSA positive" and remained so until 3 sets of surveillance cultures obtained 7 days or more apart were reported as negative for MRSA.
Our 8 cases represent an emerging and challenging health care concern that merits investigation. First, with regard to possible prevention, options include screening health care workers as potential carriers, as well as the routine implementation of proper aseptic techniques. Suh et al6 used ribotyping to study the infection rates and route of infection in patients with MRSA who had undergone middle ear surgery. They found that there were identifiable health care workers who were carriers of MRSA. Once carriers were treated and widespread information regarding MRSA infection had been disseminated, with routine precautions implemented, the MRSA infection rate decreased from 11.9% to 7.7%.6 Given the extremely low incidence of MRSA otorrhea after tympanostomy tube placement seen at CHMCC, screening for carriers and subsequent treatment would appear unwarranted. Routine hand washing and the use of gloves, however, are time-honored and sensible practices.
With regard to isolation precautions, at CHMCC both the Department of Infectious Diseases and the responsible physician are notified of a positive MRSA culture result. There is a standard protocol at CHMCC for children with positive MRSA cultures (as was followed and described earlier). Surveillance cultures are obtained from the anterior nares, axilla, and perirectal area for documentation of carrier status. Children who are inpatients are placed in isolation units, where gowns, gloves, and masks are worn by health care workers. Noncritical medical equipment, such as stethoscopes, blood pressure cuffs, and thermometers, is left in the room. The children themselves are asked not to leave the isolation unit. Family and friends enter and leave as they need to; they are instructed about strict hand washing, and gowns, gloves, and masks are made available, although they are not required. Once a child leaves the hospital, environmental services sanitizes the room thoroughly.
The issue arises as to what happens to these children when they leave the hospital. As far as infectious diseases protocol at CHMCC is concerned, the patients are considered MRSA positive until 3 sets of either surveillance cultures (obtained from the nares, axilla, and perirectal area) or cultures from the original site of infection (eg, blood or cerebrospinal fluid) are reported as negative for MRSA. Cultures are obtained a minimum of 7 days apart. Until that point,the patients who are MRSA positive are placed back in isolation if they return as inpatients. However, when they leave the hospital, they go back to school or day care and resume their normal activities. The rationale for this rests in the theory that an otherwise healthy child who is exposed to MRSA will shed the organisms quickly, whereas a sick child in a hospital who is continually exposed to nosocomial MRSA will not shed the organisms and is more likely to become infected.
The question then arises as to how to care for patients with MRSA who are seen in an outpatient setting. Children with MRSA otorrhea after tympanostomy tube placement who are otherwise healthy can be followed up in a clinic or an office setting. Initially, each of our patients was seen at CHMCC as the last patient of the day. The families or caregivers would call the office intermittently to find out when office hours were concluding so that the child could be seen rapidly and not come in contact with other children unnecessarily. After the child was seen, environmental services would shut the room down for thorough cleaning. It is easy to see how this process might encumber a busy clinic or office practice if the number of children with MRSA otorrhea increases. The policy at CHMCC regarding outpatient visits for MRSA-positive children now requires strict hand-washing procedures, and it makes certain that all sites that the child comes into contact with are cleaned and wiped down after the visit. There is no longer a mandate to see these children at the end of the day, nor is it necessary to shut down the room for a prolonged period. These changes in policy are institution specific, and each institution establishes isolation protocol individually. The national trend has mirrored that seen at CHMCC, where strict isolation procedures have been modified to be both safe and cost-effective.11,12
With regard to treatment of MRSA otorrhea, options must take into account both efficacy and possible toxic effects. Admitting each child with MRSA otorrhea for parenteral vancomycin therapy is not necessarily first-line treatment. Traditional treatment of otorrhea has included ototopical agents such as chloramphenicol, gentamicin, tobramycin, and 1% hydrocortisone–polymyxin-neomycin drops. Each of these agents has been documented to cause ototoxic effects.10,13-15 Faced with this information, it can be argued that the majority of ototoxic data have been compiled from animal research and that there are clear differences between animal and human round window membranes, the site presumed to be most important with reference to potential ototoxins. Moreover, these ototopical agents have been widely used in patients despite the animal ototoxic evidence, and the incidence of ototoxic effects has been dramatically low. Gyde16 reported using topical gentamicin on 300 patients with otorrhea, with no signs of ototoxic effects. House17 commented in a response to the article on ototoxicity by Pickett et al10 that Cortisporin suspension was routinely administered into the middle ear space in patients at the House Ear Clinic with no ototoxic events. In a large survey of practicing otolaryngologists, 2235 of whom responded, 93.7% of responders used ototopical agents in patients with patent tympanostomy tubes, and only 3.4% responded that they had witnessed ototoxic effects of these agents.18 Still, when physicians are recommending treatment, the possibility of ototoxic drug reactions must be taken into account. Topical fluoroquinolones, such as ciprofloxacin, have recently entered the market, with much promise. Ofloxacin has also been shown to be effictive against MRSA in vitro, so this class of agents holds promise of effective therapy.19 However, reports of audiologic testing for potential adverse effects of ciprofloxacin are sparse; the 1 article that has reported clinical audiologic testing after topical ciprofloxacin administration does not delineate the results of the testing, nor does it specify whether high-frequency tones were examined.20 Further work is needed in this area.
Close microbiological analysis of the culture results is perhaps the best initial guide for treating the patient with MRSA otorrhea. Routine cultures and sensitivity profiles that are reported at CHMCC include a standard panel and do not include agents such as ciprofloxacin. However, when asked, the microbiology department can provide a complete sensitivity list, which includes ciprofloxacin, chloramphenicol, gentamicin, and a host of the more common ototopical agents. The otolaryngologist faced with a child with MRSA otorrhea should request a full sensitivity panel. In cases involving MRSA, it is important to remember that penicillins, cephalosporins, and β-lactams, such as amoxicillin–clavulanic acid, ampicillin-sulbactam, and ticarcillin–clavulanic acid, may appear active in vitro but have not been shown to be clinically effective.
A close review of the culture sensitivity profiles of our 8 patients with MRSA otorrhea reveals that all 8 cultures showed sensitivity to vancomycin, sulfamethoxazole-trimethoprim, and gentamicin, and that 7 of the 8 were sensitive to chloramphenicol. However, only 1 of the 8 cultures showed sensitivity to ciprofloxacin. Sensitivity to 1% hydrocortisone–polymyxin B–neomycin drops was not tested. Sulfamethoxazole-trimethoprim has been shown to be a viable alternative to vancomycin for the treatment of MRSA infections.21 It has numerous advantages, including oral administration and fewer adverse effects.
With regard to possible ototopical agents, fluoroquinolone-containing solutions would seem of little use, given the resistance to ciprofloxacin. For children who are sensitive to ciprofloxacin or similar agents, such ototopical medication would be a reasonable first-line choice. For those who are resistant, chloramphenicol, gentamicin (or tobramycin), or 1% hydrocortisone–polymyxin B–neomycin drops are an option. The latter combination has been shown to be active in vitro against MRSA.19 On the horizon, fosfomycin may be used in conjunction with 1% hydrocortisone–polymyxin B–neomycin drops, as it has been shown to reduce the ototoxic effects that are attributed to polymyxin B.22 In all cases, patients and their parents must be told of the potential ototoxic effects of the various topical antibiotics.
Methicillin-resistant Staphylococcus aureus otorrhea after tympanostomy tube placement represents a growing concern. Knowledge of infectious disease precautions and isolation protocols at one's own hospital is critical. Culturing and assessing these cultures for a full antibiotic sensitivity panel may help guide management. Treatment decisions are made with a sound knowledge of the necessity for therapy as well as of the potential adverse effects of the treatment that is chosen. Possible treatment strategies, with attendant risks and benefits, are as follows:
Vancomycin remains a treatment option for MRSA. However, weighing the risks of vancomycin therapy and the need for parenteral administration against the benefits drawn from using other less toxic and more readily dispensable antibiotics, vancomycin should be used as a last resort.
Sulfamethoxazole-trimethoprim is a viable option for an oral antibiotic. In light of the potential ototoxic effects of ototopical agents, an effective oral antibiotic would seem to be a reasonable alternative.
Topical antibiotic medications, such as aminoglycoside- and chloramphenicol-containing solutions, are effective against MRSA and have been effective in our practice. The use of these antibiotic solutions must be tempered by the knowledge of their potential ototoxic effects, which have been demonstrated in animal models.
In our series, cultures revealed MRSA to be resistant to fluoroquinolones. A careful review of the individual culture and sensitivity panel is warranted for every child with MRSA otorrhea. If the culture shows sensitivity to fluoroquinolones, these agents may represent an excellent first-line therapy.
Given the extremely low incidence of MRSA after tympanostomy tube placement in our institution, changing practices with reference to the possibility of adding the routine use of antiseptic-type irrigations before each myringotomy would seem unwarranted. However, routine preventive measures, such as thorough hand-washing and the use of gloves, would seem appropriate.
Surveillance cultures obtained at the time of MRSA identification and follow-up cultures are important and should be performed according to the infectious disease protocols implemented at each institution. Close dialogue between the infectious disease and otolaryngology departments is necessary to define how the children in such cases should best be cared for in the operating room as well as in an outpatient setting.
Accepted for publication June 16, 2000.
Presented in part at the Fall Meeting of the Society for Ear, Nose, and Throat Disorders in Children, Chicago, Ill, October 26-29, 2000.
Corresponding author: Christopher J. Hartnick, MD, Department of Pediatric Otolaryngology, Children's Hospital Medical Center, 3333 Burnet Ave, Cincinnati, OH 45229 (e-mail: firstname.lastname@example.org).