AOM: Guideline use and antibiotic choice differ among clinicians

It has been six years since the American Academy of Pediatrics and the American Academy of Family Physicians published the Clinical Practice Guideline: Diagnosis and Management of Acute Otitis Media in 2004.

This was the first national evidence-based guideline to endorse use of observation without initial use of an antibiotic for certain infants and children with AOM (based upon age, diagnostic certainty and severity of illness).

In these guidelines, specific antibiotics were recommended for specific roles. Some researchers have since assessed practicing clinicians’ adoption and views on these guidelines. Most recently, researchers analyzed data taken from the National Ambulatory Medical Care Services on antibiotic prescribing habits for infants and children with AOM. This month’s column will review these data and discuss prescribing patterns of antibiotics identified in these surveys.

Coco and colleagues analyzed data from the National Ambulatory Medical Care Survey (NAMCS) — data collected from 2002 to 2006 for the purpose of evaluating changes in antibiotic and analgesic (acetaminophen, ibuprofen and Auralgan) prescribing for children aged 6 months to 12 years diagnosed with AOM.

The NAMCS provides data on U.S. ambulatory medical care services by nonfederally-employed, office-based physicians who are randomly selected. Data were collected for the diagnoses of acute suppurative otitis media, unspecified suppurative otitis media and unspecified otitis media.

Two 30-month periods were analyzed: A pre-guideline period (January 2002 to June 2004) and a post-guideline period (July 2004 to December 2006) for a total of 1,114 patient visits. Antibiotics were not prescribed in 13% of patient visits, and the percent of children diagnosed with AOM who were treated without an antibiotic did not significantly change before and after guideline publication (11% pre-guideline and 16% post-guideline). Prescribing patterns of some antibiotics differed before and after guideline publication: Amoxicillin prescribing increased (40% to 49%); amoxicillin/clavulanate prescribing decreased (23% to 16%); cefdinir (Omnicef, Abbott) prescribing increased (7% to 14%); prescribing of other cephalosporin antibiotics decreased (12% to 6%); and prescribing of macrolide antibiotics did not change. Prescribing of analgesics increased from 14% to 24% (P<.05).

Independent predictors of a patient visit at which an antibiotic was not prescribed included absence of ear pain, absence of fever and analgesic prescribing. These data were limited by two factors: 1) Researchers were not able to identify if prescribers used safety-net antibiotic prescriptions (antibiotic prescription given but recommended to be withheld unless symptoms persisted after 48 to 72 hours), and 2) Prescribing patterns of amoxicillin or amoxicillin/clavulanate dosing were not identified (that is, use of high doses — 90 mg/kg/day as compared with standard doses — 45 mg/kg/day).

Data from the study conducted by Coco and colleagues indicated similar results to a survey published in 2007 (Vernacchio and colleagues). The researchers assessed 299 pediatricians’ views and clinical practice habits for AOM and the 2004 guidelines, using a national practice-based, primary care pediatric research network in 2006. Surveyed pediatricians were asked about antibiotic choice for several common clinical AOM scenarios, including “AOM that failed treatment with amoxicillin at 80 mg/kg/day to 90 mg/kg/day.” The AAP/AAFP guidelines suggested amoxicillin/clavulanate 90 mg/kg/day for this scenario. In this survey, use of high-dose amoxicillin/clavulanate decreased from 51.7% to 42.8%, and use of cefdinir increased from 21.5% to 30.4% from 2004 to 2006. Thus, data from two distinct assessments of clinicians’ prescribing habits for AOM revealed differences in antibiotic choice among nationally endorsed guidelines and community practitioners.

Amoxicillin/clavulanate vs. cefdinir

The AAP/AAFP guidelines recommended amoxicillin or amoxicillin/clavulanate as agents of choice (depending upon severity) for infants and children being treated initially with an antibiotic or failure with observation, and amoxicillin/clavulanate for failed treatment with amoxicillin. Cefdinir is one of three cephalosporin antibiotics recommended for infants or children with a history of nontype I penicillin allergy (not urticaria or anaphylaxis). A further analysis on the use of amoxicillin/clavulanate and cefdinir by physicians described in the studies above is interesting to consider. In the most recently published study (Coco and colleagues), amoxicillin/clavulanate prescribing decreased and cefdinir prescribing increased. How do these antibiotics compare for the treatment of AOM?

Amoxicillin/clavulanate and cefdinir can be compared by FDA-labeled indications; activity toward the major AOM bacterial pathogens; results of clinical trials; and results of direct, comparative clinical trials. Other factors are also important to consider, including dosing, adverse effects, taste and cost.

Cefdinir is FDA–labeled for use in infants and children aged 6 months or older for AOM caused by Streptococcus pneumoniae (penicillin-susceptible strains only); Haemophilus influenzae (including beta-lactamase–producing strains); and Moraxella catarrhalis (including beta-lactamase–producing strains). Cefdinir is labeled for dosing durations of five to 10 days (7 mg/kg every 12 hours) and for 10 days (14 mg/kg every 24 hours).

Amoxicillin/clavulanate (Augmentin ES-600 suspension, GlaxoSmithKline) is FDA–labeled for use at 90 mg/kg/day in infants and children aged 3 months or older for recurrent or persistent AOM due to S. pneumoniae (penicillin minimum inhibitory concentration ≤2 mcg/mL); H. influenzae (including beta-lactamase–producing strains); and M. catarrhalis (including beta-lactamase–producing strains). The S. pneumoniae MIC value, ≤2 mcg/mL, for penicillin susceptibility stated in this labeling includes susceptible, nonsusceptible- intermediate and some nonsusceptible-resistant S. pneumoniae strains.

In vitro studies

Researchers from several studies assessed in vitro antibiotic activity toward common pediatric respiratory pathogens.

In a 2001 study conducted by Black and colleagues, 153 S. pneumoniae isolates were recovered from children, and amoxicillin/clavulanate was the most active beta-lactam antibiotic when compared with four cephalosporin antibiotics including cefdinir. Of the NSSP-I (penicillin intermediate nonsusceptible S. pneumoniae) strains tested, 100% were susceptible to amoxicillin/clavulanate and 33% were susceptible to cefdinir. Of the NSSP-R (penicillin-resistant nonsusceptible S. pneumoniae) strains tested, 52% remained susceptible to amoxicillin/clavulanate as compared with 0% susceptible to cefdinir. In a study by Doern and colleagues, 1,531 S. pneumoniae isolates were recovered from children and adults (34.2% were NSSP); 93.7% of these isolates were susceptible to amoxicillin/clavulanate; and 72.8% were susceptible to cefdinir. Susceptibilities for NSSP-R were 70.8% to amoxicillin/clavulanate and 0.3% to cefdinir. Nonsusceptible S. pneumoniae isolates were most commonly recovered from children aged 5 years or younger and from middle ear fluid. In a multicenter study by Sader and colleagues, 3,200 S. pneumoniae isolates were recovered from children and adults with respiratory tract infections in North America from 2000 to 2002. Among these people, 78.5% were susceptible to cefdinir and 94.1% were susceptible to amoxicillin/clavulanate. These studies’ results suggested that amoxicillin/clavulanate is more likely to display activity toward S. pneumoniae than cefdinir, including nonsusceptible strains.

Clinical studies

Cefdinir and amoxicillin/clavulanate were evaluated in clinical studies of infants and children with AOM, including three comparative trials. Both antibiotics demonstrated high clinical success rates. In an open-label (n=125), single-tap (tympanocentesis with pathogen isolation) trial of cefdinir (7 mg/kg twice daily for five days), children aged younger than 2 years had lower clinical cure rates than older children, and the study researchers recommended using a 10-day treatment duration for children aged younger than 2 years (Block and colleagues).

In another study conducted to evaluate differing cefdinir dosages (7 mg/kg twice daily, 14 mg/kg every day), cefdinir was compared clinically to amoxicillin (40 mg/kg/day divided three times per day) for a 10-day treatment duration in 595 children (aged 6 months to 12 years) in a randomized, single blind manner (Adler and colleagues). Cure rates were similar in all three groups. Cefdinir (7 mg/kg twice daily or 14 mg/kg every day) was compared to amoxicillin/clavulanate (40 mg/kg/day divided three times per day) in another randomized, single blind, single-tap study, conducted by Block and colleagues, in 384 children (aged 6 months to 12 years). Clinical cure rates were similar among the groups. However, presumptive bacterial eradication rates of S. pneumoniae were significantly lower for cefdinir given twice daily (55.2%) compared with amoxicillin (89.5%), and they were marginally lower for cefdinir given daily (80%) as compared with amoxicillin (P=.054).

In another clinical trial, Block and colleagues compared high-dose amoxicillin/clavulanate (90 mg/kg/day divided twice daily) with cefdinir (14 mg/kg/day divided twice daily) for 10 days in a randomized, single blind manner in 318 children (aged 6 months to 6 years). Intent-to-treat clinical cure rates were similar for both groups (85% vs. 82%). However, per-protocol cure rates were significantly higher for amoxicillin/clavulanate (90%) compared with cefdinir (82%) because of reduced response to cefdinir in children with recurrent AOM and children aged younger than 2 years.

Caregivers of children enrolled in this study were additionally surveyed about each antibiotic, with significantly more caregivers reporting increased ease of use and taste with cefdinir compared with amoxicillin/clavulanate. The majority of caregivers reported similar satisfaction with each antibiotic and would use each antibiotic again for future therapy in their children. Clinical and microbiologic (double-tap, before and after four to six days of antibiotic therapy) studies of children with AOM are most informative of antibiotic efficacy, as a strong correlation exists between bacterial pathogen eradication and clinical success.

Studies of antibiotic treatment in children with AOM where outcome is measured only clinically can be misleading, as clinical endpoints can overestimate efficacy of antibiotics with reduced activity and can underestimate efficacy of antibiotics with increased activity (the Pollyanna phenomenon). High-dose amoxicillin/clavulanate has been evaluated in a double-tap, open-label trial (Dagan and colleagues) of infants and children (n=355; aged 3 to 48 months). The majority (72%) of infants in this study were younger than 2 years old. Of 154 S. pneumoniae isolates identified, 55% were penicillin-susceptible, 18% were NSSP-I and 28% were NSSP-R. Bacterial eradication rates were high: 98% for S. pneumoniae (100% for NSSP-I; 91% for NSSP-R) and 94% for H. influenzae (no difference if beta-lactamase positive or negative). A strong correlation (90%) was demonstrated between pathogen eradication and clinical success. Diarrhea was reported in 3.6% of children, and 12.5% had watery stools for at least two consecutive days.

Conclusions

Six years after publication of national guidelines endorsed by major medical professional academies, some data have suggested that antibiotic-prescribing habits of clinicians providing care to infants and children may differ from guideline recommendations. A complex mixture of reasons for these differences likely exists.

Two antibiotics described in these guidelines, cefdinir and amoxicillin/clavulanate, may be prescribed not as the guideline authors intended. Both of these antibiotics have demonstrated high clinical cure rates in trials of infants and children with AOM, but important differences between them exist. Cefdinir possesses some advantages compared with amoxicillin/clavulanate, as taste tests have shown that children rate cefdinir’s taste better. One should also consider that many pharmacies are able to add more than 20 flavorings (including chocolate banana) to amoxicillin/clavulanate to improve taste. Risk for diarrhea may be lower for cefdinir, as shown in several clinical trials. Results of other trials have demonstrated no difference in rates of diarrhea between cefdinir and amoxicillin/clavulanate, and package labeling describes a risk for diarrhea of 2.9% for amoxicillin/clavulanate and 8% for cefdinir. Amoxicillin/clavulanate administration with a meal or snack can reduce risk for gastrointestinal upset. Both cefdinir and amoxicillin/clavulanate suspensions are available as generic products and are priced similarly. It seems prudent to use amoxicillin/clavulanate as intended in the guidelines, as it has excellent activity toward the major AOM bacterial pathogens.

Importantly, amoxicillin/clavulanate, especially when given in high doses (90 mg/kg/day), is active toward many penicillin-nonsusceptible S. pneumoniae strains, including some penicillin-resistant strains. These pathogens are more likely to be found in younger (aged younger than 2 years) children and children with recurrent or persistent AOM. In vitro and clinical trial data suggest that high-dose amoxicillin/clavulanate is more likely to be effective in these children than cefdinir. There are differences in amoxicillin/clavulanate suspension products, and only the 600-mg/5-mL product (Augmentin ES-600 suspension) should be used when high doses are prescribed — as other suspension products (200 mg/5 mL, 400 mg/5 mL) contain differing amounts of clavulanate, and their use for high doses increases risk for diarrhea and GI upset. Cefdinir may be better used as described in the guidelines (nontype I penicillin allergy) or for children who do not tolerate amoxicillin/clavulanate because of GI upset or taste.

Edward A. Bell, PharmD, is a Professor of Clinical Sciences at Drake University College of Pharmacy, Blank Children’s Hospital and Clinics in Des Moines, Iowa.

For more information:

• Adler M. Pediatr Infect Dis J. 2000;19:S166-S170.
• Black J. Diagn Microbiol Infect Dis. 2001;39:195-197.
• Block SL. Curr Med Res Opin. 2006;22:1839-1847.
• Block SL. Pediatr Infect Dis J. 2000;19:S153-S158.
• Block SL. Pediatr Infect Dis J. 2000;19:S159-S165.
• Coco A. Pediatrics. 2010;125:214-220.
• Dagan R. Pediatr Infect Dis J. 2001;20:829-837.
• Doern GV. Antimicrob Agents Chemother. 2001;45:1721-1729.
• Sader HS. Diagn Microbiol Infect Dis. 2003;47:515-525.
• Vernacchio L. Pediatrics. 2007;120:281-287.