August 2003

The heptavalent pneumococcal conjugate vaccine (PCV7), introduced in the United States in March 2000, was marketed heavily to pediatricians as a vaccine which effectively prevented invasive disease. With the approval of the AAP, PCV7 quickly became part of the routine immunizations of American children.

Based on results of major pre-approval studies of the effectiveness of PCV7 given to 40,000 infants, combined data from northern California and Finland demonstrated a 7% reduction in cases of acute otitis media (AOM), a 10% to 20% reduction in children who experienced three or more episodes of AOM in the previous 6-months (otitis-prone) and an estimated 20% reduction in children who underwent surgery for insertion of tympanostomy tubes.

In the Finnish study by Eskola and co-investigators, PCV7 was 57% effective overall for homologous serotypes contained in the vaccine. In other words, about 40% of immunized Finnish children became infected with homologous serotype pneumococcal AOM. Moreover, the rate of non-vaccine serotypes of pneumococcus increased by one-third in PCV7 recipients, relative to the placebo group.

While it is too early to know if any of the anticipated benefits of PCV7 against AOM have been realized in the United States, a recent randomized, double-blind study from the Netherlands reported on the efficacy of one or two doses of PCV7.

In that study, children 1 to 7 years who experienced two or more episodes of AOM in the year before study entry were the experimental group. The investigators found no reduction in cases of AOM in children who received PCV7 and a booster dose of pneumococcal polysaccharide vaccine, compared with a similar group of children who received only hepatitis A or hepatitis B vaccines. In addition, there was no reduction in surgery for ventilation tube placement or adenoidectomy in the study group. An excellent editorial by Peltola et al states: “This study adds further to the view that the present vaccine is not an apt weapon against acute otitis media.”

There are three major problems with the Dutch study. The first was the exclusion of children in the first year of life from receiving the primary series of PCV7. The second problem was the study’s imprecise definition of AOM. Signs of AOM did not specify bulging of the tympanic membrane (TM) as a sine qua non of AOM. Entry criteria permitted either bulging of the TM, or redness, or dullness of the TM, plus at least one physical symptom including pain, fever, irritability, or acute otorrhea (only 15 ears). While these criteria are acceptable by the FDA and others, they encourage unnecessary treatment of otitis media with effusion (OME) with antibiotics. Children with OME often have associated symptoms of pain or rubbing the ear(s) along with otoscopic signs of tympanic membrane dullness and limited mobility. The third problem is that only 40% of pneumococcal serotypes recovered from the middle ear cultures of children older than 2 years are homologous to vaccine serotypes.

Changes in bacteriology

Since the widespread use of PCV7, a few U.S. investigators who perform tympanocentesis have observed major changes in the bacteriology of AOM among children 7 to 24 months old who received at least three doses of PCV7. Bardstown, Ky., is located in a rural area of that state, 45 miles south of Louisville. Stan Block, MD, and colleagues, who have been in the forefront of research in AOM conducted in private practice, provide care for most of the children in their area.

Dr. Block compared results of middle ear cultures between two periods, prior to and after the introduction of PCV7. The pre-PCV7 period extended from 1992 to 1998 and the post-PCV7 period began in June 2000 and ended in March 2003. Prior to the widespread use of PCV7, most penicillin non-susceptible pneumococcal isolates were homologous with PCV7 serotypes. The good news is that there was a decline in the overall frequency of Streptococcus pneumoniae from 50% to 35%. Unexpectedly, the frequency of penicillin-non-susceptible (penicillin-resistant) strains of S. pneumoniae failed to follow the same percentage of decrease (from 25% of all pneumococcal strains pre-PCV7 to 21%, post-PCV7). Two pneumococcal serotypes (6A and 19A), not contained in the present PCV7 vaccine increased from <10% pre-PCV7 to 37% post-PCV7. These two non-homologous serotypes have prevented the expected major reduction in penicillin nonsusceptible pneumococcal AOM, at least in Kentucky. Block and coworkers also noted that gram-negative middle ear pathogens (primarily non-typable Haemophilus influenzae) increased in frequency from about 40% to 65% since the introduction of PCV7. ß-lactamase-positive (ampicillin-resistant) strains of H. influenzae and Moraxella catarrhalis) now compromise about 50% of middle ear pathogens in Bardstown, Ky.

Vienna, Va., is in Fairfax County, a suburb of Washington, DC. During 2002 and in the first six months of 2003, 50 children younger than 30 months, with AOM, who enrolled in several investigations of antibiotic efficacy underwent tympanocentesis and middle ear culture. All children had received the full primary series of PCV7. Similar to the findings from Bardstown, H. influenzae was the most frequent bacterial organism recovered (50%), only three of which were ampicillin-nonsusceptible. S. pneumoniae was second in frequency at 33% (only one of 18 was highly penicillin-resistant) and Streptococcus pyogenes was third (10%). Only 8% of the specimens from Vienna had no growth of middle ear pathogens in keeping with our strict criteria for the diagnosis of AOM. If these data are confirmed by other investigators, there has been a notable change in hierarchy of bacterial causes of AOM.

PCV7 is not available in Israel except to those few children enrolled in clinical trials. Nevertheless, a similar reduction in pneumococcal AOM was reported in that country in a tympanocentesis study examining the effectiveness of high dose amoxicillin for AOM.

Although it is widely believed that b-lactamase-mediated resistance to ampicillin (including amoxicillin) will not respond favorably to high dose amoxicillin (80 to 90 mg/kg/day), the Israeli study found that high-dose amoxicillin eradicated 84% of ampicillin susceptible H. influenzae and 62% of ampicillin-resistant H. influenzae. On day 4 to 6, when follow-up tympanocentesis was performed on all study children, 28% of the children who entered the study had a second positive culture. Most of these were gram-negative ampicillin nonsusceptible species.

If these data are confirmed, the major middle ear pathogen recovered in younger children who have been vaccinated with Prevnar has become H. influenzae, followed by S. pneumoniae.

For more information:

• Piglansky L, Leibovitz E, Raiz S, et al. Bacteriologic and clinical efficacy of high dose amoxicillin for therapy of acute otitis media in children. Pediatr Infect Dis J. 2003;22:405-12.
• Eskola J, Kilpi T, Palmu A, et al. Efficacy of a pneumococcal conjugate vaccine against acute otitis media. N Engl J Med. 2001;344:403-09.
• Black S, Shinefield H, Fireman B, et al. Efficacy, safety and immunogenicity of heptavalent pneumococcal conjugate vaccine in children. Pediatr Inf Dis J. 2000;19:187-95.
• Veenhoven R, Bogaert D, Ulterwaal C, et al. Effect of conjugate pneumococcal vaccine followed by polysaccharide pneumococcal vaccine on recurrent acute otitis media: a randomized study. Lancet. 2003;361:2189-95.
• Peltola H, Schmitt J, Booy R. Pneumococcal conjugate vaccine for acute otitis media-yes or no? Lancet. 2003;:361:2170-71.
• Block Stan L, Hedrick J, Harrison CJ, et al. Pneumococcal serotypes from acute otitis media in rural Kentucky. Pediatr Infect Dis J. 2002;21:859-65.