The Challenge of Recalcitrant Acute Otitis Media

Pathogens, Resistance, and Treatment Strategy

Abstract

Recalcitrant acute otitis media (AOM), including 2 well-defined and sometimes related entities (persistent AOM and recurrent AOM), is a common pediatric problem. Episodes of AOM in children with either persistent or recurrent AOM were shown to be associated with recovery of middle ear fluid bacterial pathogens (particularly Streptococcus pneumoniae) with higher resistance to antibiotics compared with children with AOM not recently treated with antibiotics. The relationships between nasopharyngeal carriage of Streptococcus pneumoniae and persistence of middle ear fluid pathogens despite clinical improvement or cure and subsequent recurrence are discussed in depth. The findings emphasize the importance of bacteriologic eradication in the prevention of recurrent AOM episodes.

The purpose of the present review is to analyze the experience accumulated during 1996–2003 with a large group of children with acute otitis media (AOM) enrolled at the Pediatric Infectious Disease Unit at the Soroka University Medical Center, Beer-Sheva, Israel, in various double-tympanocentesis studies evaluating the efficacy of different antibiotics in the treatment of AOM.

RECALCITRANT AOM

Recalcitrant (difficult to treat) AOM is a common pediatric problem, affecting up to 20% of children in their first years of life.^1,2 Recalcitrant AOM includes 2 main entities: (1) persistent AOM, defined as lack of improvement in signs and symptoms for ≥48–72 hours after initiation of antibiotic therapy; and (2) recurrent AOM (R-AOM), classified as ≥3 AOM episodes occurring in the previous 6 months or ≥4 episodes in 1 year preceding the actual AOM.

The microbiology of recalcitrant AOM has evolved as a reflection of the major changes that occurred in resistance patterns among the 3 major otopathogens.³ In the early and mid-1980s, β-lactamase producing organisms (mainly nontypeable Haemophilus influenzae and Moraxella catarrhalis) were frequently responsible for therapeutic failures in patients treated with amoxicillin. In the early 1990s, the emergence of penicillin-nonsusceptible Streptococcus pneumoniae became the major therapeutic challenge in the management of persistent or recurrent AOM.

Although the overall pathogen distribution is considered to be similar in recalcitrant AOM compared with noncomplicated AOM, S. pneumoniae isolates with higher minimum inhibition concentrations (MICs) were more involved in cases of persistent/R-AOM compared with AOM not recently treated with antibiotics (Fig. 1). ^4,5 No information is available from the last 2 decades on the timing of recurrence after completion of therapy, distribution of true recurrences versus new infections, and the possible relationship between bacterial eradication and recurrence of infection.

FIGURE 1.: MICs to penicillin of S. pneumoniae isolated from MEF in patients recently treated with antibiotics. The number of isolates with MIC between 0.38 and 1.0 μg/mL (intermediately resistant) was significantly higher in the group of patients recently treated with antibiotics compared with those not treated. Adapted from Leibovitz E, et al.³

BACTERIOLOGY OF CLINICALLY RECURRENT-AOM: RELATIONSHIP WITH ORIGINAL PATHOGEN

Although the microbiology of AOM is well defined, data on the bacteriologic correlates of R-AOM are limited, and when R-AOM reflects the persistence of the original infection or a new infection remains debatable. In limited published studies on this topic, patients were followed for various periods of time from the original AOM episode, tympanocentesis for microbiologic diagnosis at time of recurrence was rarely performed, and some children received antibiotic prophylaxis after the initial episode. Barenkamp et al⁵ and Carlin et al⁶ performed tympanocentesis in 13 and 36 children, respectively, with R-AOM occurring within a 1-month period from the initial AOM episode. They reported that in 62% and 75%, of the patients, respectively, the recurrences were due to a new pathogen.

We reported on 1077 AOM children enrolled from 1995 through 2000 in double-tympanocentesis studies of whom 834 (77%) completed antibiotic treatment successfully [no pathogen in middle ear fluid (MEF) on days 4–6 of therapy as well as documented clinical improvement or cure at the end of therapy].⁷ None of these patients received the pneumococcal conjugate vaccine. The patients enrolled were followed for 3 weeks to 1 month after the end of therapy, and MEF cultures were obtained when AOM recurred. Serotype and pulsed field gel electrophoresis (PFGE) analyses were performed to establish the similarity or dissimilarity between isolates from the initial and subsequent episode. MEF cultures were performed in 108 consecutive patients with early R-AOM. No significant differences could be found in the relative proportions between the pathogens recovered from the MEF in the initial AOM episodes compared with those recovered at AOM recurrence. Most R-AOM episodes developed during the first 2 weeks of follow-up: 39 (36%), 38 (35%), 21 (19%), and 10 (9%) R-AOM episodes occurred between days 1–7, 8–14, 15–21, and 22–28 after the end of therapy, respectively. True bacteriologic R-AOM was found in 30/108 (28%) patients; 16/39 (41%), 10/38 (26%), 3/21 (14%), and 1/10 (10%) of all episodes occurring on days 1–7, 8–14, 15–21, and 22–28 after the end of therapy, respectively (P = 0.01) (Fig. 2). ⁷ All true bacteriologic R-AOM occurred during the first 14 days of the follow-up period. This study was unique by analyzing only those cases of R-AOM occurring after effective antibiotic treatment of the initial AOM episode (resulting in bacterial eradication), therefore excluding the possibility that the recurrence of AOM represented a persistent infection. The study clearly demonstrated that (1) most (72%) clinical R-AOM episodes occurring within 3 weeks to 1 month from completion of antibiotic therapy for the initial AOM episode are, in fact, new infections; and (2) most of the true bacteriologic R-AOM episodes were concentrated within the 2-week period after completion of therapy for the initial AOM episode.

FIGURE 2.: Distribution of the 35 pathogens causing true bacteriologic AOM recurrences according to timing of clinical recurrence of AOM. NS = not significant.

RELATIONSHIP BETWEEN NASOPHARYNGEAL CARRIAGE OF S. PNEUMONIAE AT COMPLETION OF ANTIBIOTIC THERAPY FOR AOM AND SUBSEQUENT AOM RECURRENCE

Children with R-AOM are colonized in the nasopharynx with the same pathogen that results in AOM both during and also between episodes of upper respiratory infections.⁸ In addition, significant changes are known to occur in nasopharyngeal colonization with S. pneumoniae as a function of the antibiotic treatment administered in children with AOM.⁹ However, limited information is available on the relationship between nasopharyngeal colonization with AOM pathogens during an initial AOM episode and the etiology of subsequent episodes of R-AOM. Libson et al reported on 530 AOM children (417 [79%] with culture-positive MEF) enrolled in double-tympanocentesis studies who completed treatment successfully during 1996–2002.¹⁰ None of them were immunized with pneumococcal conjugate vaccine. Nasopharyngeal cultures were obtained from 494/530 (92%) patients at the end of therapy; S. pneumoniae was isolated in 208/494 (42%) patients. AOM recurred in 130/494 (26%) patients and tympanocentesis was performed in 103 of these (79%); 83/103 (81%) had positive MEF cultures. Recurrent pneumococcal OM was more common in patients with nasopharyngeal cultures positive for S. pneumoniae at the end of therapy compared with patients with nasopharyngeal cultures negative for S. pneumoniae [34/198 (17%) versus 17/269 (6%), P < 0.01]. In addition, nasopharyngeal cultures positive for S. pneumoniae were more frequent at the end of therapy for the initial episode in patients with subsequent S. pneumoniae R-AOM [34/51 (67%)] than in patients without AOM recurrence [142/364 (39%)] or with R-AOM caused by other pathogens [12/32 (38%); P < 0.01]. The same serotype was identified in 24/30 (80%) evaluable S. pneumoniae pairs (obtained from nasopharynx at the end of therapy and from MEF at AOM recurrence), and PFGE showed complete identity between isolates in 22/23 (96%).¹⁰

The main message from this study was that early R-AOM was mostly caused by S. pneumoniae if the organism was still present in the nasopharynx at the end of therapy for the initial AOM episodes. Furthermore, we demonstrated that most pneumococcal R-AOM episode were caused by the same persistent S. pneumoniae isolates present in the nasopharynx and not eradicated at the end of the antibiotic therapy. These findings provide evidence that the persistence of pneumococcal nasopharyngeal carriage is a risk for subsequent recurrence and that antimicrobial selection should incorporate impact on nasopharyngeal colonization.

RELATIONSHIP BETWEEN PERSISTENCE OF MIDDLE EAR FLUID PATHOGENS DESPITE CLINICAL IMPROVEMENT OR CURE IN ANTIBIOTIC-TREATED AOM AND EARLY AOM RECURRENCE

Over time, most of children with AOM improve clinically and by days 10–14 have resolved symptomatically. This has led to a reevaluation of the role of antimicrobial therapy in treatment of AOM. Even when bacteria persist in the middle ear (with or without antimicrobial therapy), 80% of the children improve clinically. Practically, the complexity of this process may lead the treating physician to incorrectly believe that antibiotics with poor bacteriologic activity are as effective as those with a good or excellent bacteriologic profile against common otopathogens.

In a recent study, 676 culture-positive AOM patients seen between 1996 and 2003 had initial tympanocentesis and MEF cultures that were repeated on days 4–6; the patients then completed their antibiotic therapy.¹¹ None of these patients received pneumococcal conjugate vaccine. On days 4–6 of treatment, 192/676 (28%) patients remained culture positive. More patients with culture-negative MEF on days 4–6 were considered clinically improved or cured than patients with culture-positive MEF [476/484 (98%) versus 153/192 (80%); P < 0.001]. Patients with clinical improvement or cure on days 4–6 and culture-positive MEF had an increased rate of recurrent AOM compared with those with culture-negative MEF and clinical improvement or cure [54/133 (35%) versus 126/476 (27%); P = 0.035] (Fig. 3). On further analysis, 41 (76%) of the 54 culture-positive patients with clinical improvement or cure on days 4–6 underwent tympanocentesis at the time of recurrence and 29 (71%) were culture-positive. PFGE identity between pathogens isolated at the recurrent episode and those persisting on days 4–6 was found in 66% patients (without initial sterilization of middle ear) whereas only 36% evaluable patients with culture-negative MEF on days 4–6 went on to a true bacteriologic relapse (P = 0.005). This study demonstrated, on a large number of patients, that those patients with persistence of MEF pathogens after 3–5 days of antibiotic therapy and an outcome considered to be clinical improvement or cure had a significantly higher rate of R-AOM compared with those who achieved culture-negative MEFs on days 4–6 of treatment.¹² These results emphasize the importance of early bacteriologic eradication in the outcome of AOM.

FIGURE 3.: Flow diagram of the study patients: 676 culture-positive AOM patients who completed a 3-week follow-up after the end of therapy visit. Cx = culture.

CONCLUDING COMMENTS

The studies presented here demonstrate that most R-AOM episodes occurring within 1 month after successful completion of antibiotic therapy are due to new otopathogens. Most true bacteriologic AOM recurrences occur within 14 days after completion of therapy for the initial AOM episode, but even during this time interval most of the recurrences are caused by new pathogens. In addition, lack of sterilization of the MEF or eradication of nasopharyngeal colonization during therapy is associated with higher rates of clinical R-AOM, even if the patients demonstrated clinical improvement or cure at the end of therapy for the initial episode. Finally, we demonstrated that these recurrences are mostly caused by MEF and/or nasopharyngeal pathogens not eradicated initially by the antibiotic treatment. Therefore, the optimum strategy for prevention of AOM recurrences is sterilization of the middle ear and eradication of nasopharyngeal carriage of otopathogens during antimicrobial therapy.