Specimens were successfully collected from 891 (91%) of 981 children with otorrhea. Fifty-one percent (452/891) of specimens were culture positive for at least 1 pathogen (Supplementary Table 1, http://links.lww.com/INF/D154). Overall, Hi was the most frequently isolated organism (21%, 187/891), followed by Spn (18%, 164/891), S. aureus (9%, 83/891) and P. aeruginosa (4%, 38/891). While 79% (356/452) of culture-positive specimens contained a single pathogen, 2 pathogens were detected in 20% (89/452) and 3 in 2% (7/452) of the cases. In cases with mixed pathogens Hi was the predominant organism, being present in 78% (75/96) of the cases. Spn was present in 72% (69/96) of mixed-pathogen cases, and Spn and Hi occurred together in 60% (58/96) of cases.
Although there were differences in prevalence of the pathogens in otorrhea cases, no significant difference between the presence of each of the 4 predominant pathogens and median duration of otorrhea was found (Fig. 3).
Serotype Distribution of S. pneumoniae and H. influenzae and Expected Vaccine Coverage
Ninety-eight percent (56/57) of Hi isolates serotyped were nontypeable (NTHi), and the remaining 1 isolate was Hi type f. On the other hand, 36 different pneumococcal serotypes were detected in this study (Fig. 4). The most common serotype was 19A (16%, 26/164), followed by 19F (14%, 23/164), 3 (8%, 13/164) and 14 (7%, 12/164). Taking the serotype distribution into account, the expected coverage of PCV10 for pneumococcal OM was 46%; expected coverage of PCV10 + 6A was 49%. If protection against 19A was achieved, the coverage of PCV10 + 6A+19A would be 65%. For all OM cases, expected coverage of PCV10 was 8%, PCV10 + 6A 9% and PCV10 + 6A+19A 12%.
Antimicrobial Susceptibility Testing of S. pneumoniae and Nontypeable H. influenzae
Overall, 16% (27/164) of the Spn isolates were susceptible to all antibiotics, while 30% were nonsusceptible to 1, 49% to 2 and 4% to 3 different antibiotics tested (Table 1). Ninety-nine percent (163/164) of the strains were susceptible to penicillin, while 77% (126/164) were nonsusceptible to trimethoprim-sulfamethoxazole and 58% (95/164) nonsusceptible to erythromycin.
Ninety-one percent (69/76) of the PCV10 isolates were nonsusceptible to at least 1 antibiotic, compared with 77% (68/88) of isolates of nonvaccine serotypes (P = 0.02; Table 1). Serotype 19A was found to be the most nonsusceptible serotype; 92% (24/26) of 19A isolates were nonsusceptible to trimethoprim-sulfamethoxazole, and 100% (26/26) were nonsusceptible to erythromycin; all 19A isolates were susceptible to penicillin.
Of all the NTHi isolates, 3% (5/187) were susceptible to all antibiotics and 18% were nonsusceptible to 1, 32% to 2 and 18% to 3 different antibiotics. Although no NTHi isolates were nonsusceptible to ciprofloxacin or ceftriaxone, 92% (172/187) were nonsusceptible to erythromycin, 67% (126/187) to tetracycline, 41% (76/187) to trimethoprim-sulfamethoxazole, 23% (43/187) to ampicillin and 16% (31/187) to chloramphenicol.
History of Antibiotic Consumption by Cases
Of the 891 cases from whom specimens were collected, we were able to obtain information on antibiotic consumption in the preceding 7 days from 890 cases. In total, caregivers reported antibiotic use for 2% (17/890) of children with OM. Antibiotic exposure was reported in 0.7% (3/451) of children with culture-positive specimens, as compared with 3% (14/439) of children with culture-negative specimens (P = 0.006).
Here, we report the first dataset on pathogens of OM in Bangladesh and, to our knowledge, the largest series in South Asia. The data will help institute evidence-based treatment and prevention policies in Bangladesh and in the surrounding region.
From April 2014 to March 2015, 15% of the 7342 children seeking care at the ENT-OPD of DSH presented with clinical manifestations of OM. Children of 2 years old or younger accounted for half of all disease episodes, a finding that corroborates with those of other studies.1,2 Overall, 88% of children with OM presented with otorrhea, a proportion that is higher than has been observed in most other studies.23 This high rate of otorrhea in OM children and very few nonotorrhea cases recorded in our surveillance study may be explained by poor care–seeking behavior in Bangladesh24; with no national healthcare system in place, patients pay out of their pockets and sick people often do not come to receive treatment in facilities unless the symptoms are severe like ear discharge. This also suggests that the number of cases seen in the hospital is an underestimation of the total burden of OM among children.
Of the 891 specimens collected, only 51% were culture positive. This is lower than seen in other studies.25,26 As specimens were only collected from cases with otorrhea and a large number of these cases had discharge for several days, our surveillance may be biased toward complex, chronic cases, where growth of biofilms is common.4 Hence, it is plausible that our culture methods were not able to detect pathogens in biofilms. Furthermore, OM can also be caused by viruses, and we did not attempt to detect viral pathogens in this study; it is plausible that some culture-negative OM cases were caused by viruses like Respiratory Syncytial Virus.27
The most commonly isolated organisms were NTHi (21%), Spn (18%), S. aureus (9%) and P. aeruginosa (4%)—pathogens that have been commonly associated with OM.28M. catarrhalis, which has been frequently isolated among samples in other studies, was only found in 0.4% of cases here.4,29 Eleven percent of the cases were polymicrobial, with 2 or 3 pathogens identified. Previous studies have suggested that Spn is more frequently responsible for early episodes of OM and may lead to subsequent polymicrobial infections, which are generally associated with NTHi.29
While 98% of H. influenzae were nontypeable, we identified 36 different serotypes of Spn (N = 164). Serotype 19A was the most common serotype, followed by 19F, 3, 14, 6B and 23F; 19A, 3 and 6A are not included in the PCV10 formulation and together were responsible for 32% of pneumococcal cases. The serotype distribution of pneumococci isolated from OM cases was different from that reported for invasive pneumococcal diseases (IPD) in Bangladesh. A multiyear surveillance study on IPD in Bangladesh identified serotypes 2, 1, 6B, 14 and 5 as the predominant causes, but these serotypes were seen less frequently (33/164) among the OM cases here.5 It is important to note that 19A is rarely seen among invasive cases (4%, ranking 9th),5 although it is the predominant serotype in OM cases.
Based on our findings, the serotype coverage of PCV10 was 46% and PCV10 + 6A was 49% for pneumococcal OM and 8% and 9% for all OM cases in this series. However, several clinical trials and pneumococcal vaccine impact studies conducted in Israel on OM have shown greater impact in practice than in theory.4,30 Previous vaccine impact studies conducted in Bangladesh have also revealed higher than theoretical vaccine impact for other syndrome.31,32 Thus, the coverage rate reported here may be an underestimate. On the other hand, however, there are reports of replacement disease by nonvaccine serotypes after PCV introduction; high rates of replacement OM by non-PCV10 serotypes can jeopardize the impact of the vaccine, which is difficult to predict beforehand. Hence, it is imperative to continue the ongoing surveillance and carefully monitor trends in serotype distribution.
Spn serotype 19A became an increasingly common cause of IPD in some countries after the introduction of PCV7, which lacks the 19A antigen.33–37 Some studies have suggested that the 19F antigen, included in PCV10, can provide cross-protection against 19A disease because of their immunologic similarities, but the magnitude of this protection is yet to be discerned and might depend on several factors in a given setting.12 Hence, it is difficult to accurately predict the impact of PCV10 introduction on 19A in Bangladesh. Furthermore, the H. influenzae protein D is used as a carrier protein in the PCV10 formulation to provide cross-protection against NTHi. However, the protective efficacy of PCV10 against diseases caused by NTHi is yet to be fully comprehended.6 Considering the high burden of NTHi OM in Bangladesh, it is important to continue monitoring impact in the post-vaccine era.
Several studies from the United States and countries in Asia and Europe have reported high rates of antibiotic nonsusceptibility, specifically to penicillin, among pneumococcal isolates. This has significant implications for treatment and in replacement/unmasking diseases after PCV introduction.38–41 However, this trend of resistance is not common in Bangladesh and India for either IPD or OM isolates.5,42 Here, we found that most Spn isolates were resistant to trimethoprim-sulfamethoxazole (77%) and erythromycin (58%); while resistance to trimethoprim-sulfamethoxazole was previously reported in 2003, a rise in erythromycin resistance is a more recent phenomenon.20,43 If trimethoprim-sulfamethoxazole and erythromycin were to be excluded from this analysis, 91% (150/164) of the strains would be susceptible. Remarkably, 99% of the strains were susceptible to penicillin and can be used for successful treatment. We also found that while most strains of NTHi were nonsusceptible to erythromycin and tetracycline, all strains were susceptible to ciprofloxacin and ceftriaxone.
The results of this study should be considered within the context of some limitations. First, tympanocentesis was not performed on OM children without otorrhea, and hence, the pathogens involved in such cases could not be detected. However, as 88% of the suspected OM cases presented with otorrhea, this limitation had a relatively small if any effect on our data. Furthermore, the possibility of missing bacterial pathogens in this study was low as we used selective and enriched (nonselective) media specifically for Spn and NTHi. However, as we did not assess presence of viruses in the specimens as discussed above, our study missed viruses, both as pathogens and copathogens. Another limitation of this study is that the surveillance was performed for only a length of 1 year, and hence year-to-year variations remain unknown. Finally, the surveillance reporting is not population based; it does not have a denominator and thus does not allow for incidence calculation. However, we are currently working to overcome this by combining the existing laboratory data with data from community-based surveillance of utilization of our study hospital to generate incidence estimates. Furthermore, previous work on impact of PCV has shown that for studies with large numbers of cases such as ours, it is possible to monitor vaccine impact in hospital-based surveillance systems in the absence of population-based studies.44
This is the first report on elucidation of pathogens of OM in Bangladesh and will contribute to the design of evidence-based treatment and prevention guidelines. We demonstrate that Spn is indeed a dominant pathogen found in OM cases in Bangladesh, and our data suggest that introduction of PCV10 will have substantial impact on the burden of OM in Bangladesh. We have established ongoing surveillance and are monitoring impact of PCV10 on OM. Based on the number of OM cases identified in our hospital and the high prevalence of Spn serotype 19A and NTHi, this ongoing surveillance will help us monitor the impact of PCV10 on overall pneumococcal and nonpneumococcal OM and also to estimate PCV10’s ability to protect against OM caused by vaccine-related serotypes 6A, 19A, as well as NTHi.
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
We are grateful to Rajib Chandra Das and Afiour Rahman for assistance with statistical analysis and to Dr. Asif Sattar, Zillur Rahman, Palash Mozumder and Shahina Tarafder for technical help.
1. Daly KA, Giebink GSClinical epidemiology of otitis media
. Pediatr Infect Dis J. 2000;19(5 suppl):S31–S36.
2. Alho OP, Koivu M, Sorri M, et alThe occurrence of acute otitis media
in infants. A life-table analysis. Int J Pediatr Otorhinolaryngol. 1991;21:7–14.
3. Vaz LE, Kleinman KP, Raebel MA, et alRecent trends in outpatient antibiotic use in children. Pediatrics. 2014;133:375–385.
4. Dagan R, Pelton S, Bakaletz L, et alPrevention of early episodes of otitis media
by pneumococcal vaccines might reduce progression to complex disease. Lancet Infect Dis. 2016;16:480–492.
5. Saha SK, Hossain B, Islam M, et alEpidemiology of invasive pneumococcal disease in Bangladeshi children before introduction of pneumococcal conjugate vaccine. Pediatr Infect Dis J. 2016;35:655–661.
6. Cerquetti M, Giufrè M, Cerquetti M, et alWhy we need a vaccine for non-typeable Haemophilus influenzae. Hum Vaccin Immunother. 2016;12:2357–2361.
7. Mittal R, Lisi CV, Gerring R, et alCurrent concepts in the pathogenesis and treatment of chronic suppurative otitis media
. J Med Microbiol. 2015;64(10):1103–1116.
8. Roy E, Hasan KZ, Richards JL, et alBurden and outcome of acute otitis media
in children aged under 2 in rural Bangladesh. S Afr J CH. 2012;6:1–5.
9. Kamal N, Joarder AH, Chowdhury AA, et alPrevalence of chronic suppurative otitis media
among the children living in two selected slums of Dhaka City. Bangladesh Med Res Counc Bull. 2004;30:95–104.
10. World Health Organization Publication. Pneumococcal vaccines WHO position paper—2012—recommendations. Vaccine. 2012;30(32):4717–4718.
11. Poolman J, Frasch C, Nurkka A, et alImpact of the conjugation method on the immunogenicity of Streptococcus pneumoniae serotype 19F polysaccharide in conjugate vaccines. Clin Vaccine Immunol. 2011;18:327–336.
12. Hausdorff WP, Hoet B, Schuerman LDo pneumococcal conjugate vaccines provide any cross-protection against serotype 19A? BMC Pediatr. 2010;10:4.
13. Brandileone MC, Zanella RC, Almeida SCG, et alEffect of 10-valent pneumococcal conjugate vaccine on nasopharyngeal carriage of Streptococcus pneumoniae and Haemophilus influenzae among children in São Paulo, Brazil. Vaccine. 2016;34:5604–5611.
14. Vesikari T, Forsten A, Seppä I, et alEffectiveness of the 10-valent pneumococcal nontypeable Haemophilus influenzae
protein D-conjugated vaccine (PHiD-CV) against carriage and acute otitis media
-A double-blind randomized clinical trial in Finland. J Pediatric Infect Dis Soc. 2016;5:237–248.
15. Bosch AATM, van Houten MA, Bruin JP, et alNasopharyngeal carriage of Streptococcus pneumoniae and other bacteria in the 7th year after implementation of the pneumococcal conjugate vaccine in the Netherlands. Vaccine. 2016;34:531–539.
16. Hau I, Levy C, Caeymaex L, et alImpact of pneumococcal conjugate vaccines on microbial epidemiology and clinical outcomes of acute otitis media
. Paediatr Drugs. 2014;16:1–12.
17. Hammitt LL, Akech DO, Morpeth SC, et alPopulation effect of 10-valent pneumococcal conjugate vaccine on nasopharyngeal carriage of Streptococcus pneumoniae and non-typeable Haemophilus influenzae in Kilifi, Kenya: findings from cross-sectional carriage studies. Lancet Glob Health. 2014;2:e397–e405.
18. Grevers G, Wiedemann S, Bohn JC, et alIdentification and characterization of the bacterial etiology of clinically problematic acute otitis media
after tympanocentesis or spontaneous otorrhea in German children. BMC Infect Dis. 2012;12:312.
19. World Health Organization Publication. Chronic suppurative otitis media
. August 2004:1–84.
20. Saha SK, Baqui AH, Darmstadt GL, et alComparison of antibiotic resistance and serotype composition of carriage and invasive pneumococci among Bangladeshi children: implications for treatment policy and vaccine formulation. J Clin Microbiol. 2003;41:5582–5587.
21. Falla TJ, Crook DW, Brophy LN, et alPCR for capsular typing of Haemophilus influenzae. J Clin Microbiol. 1994;32:2382–2386.
22. Jorgensen JH, Howell AW, Maher LAQuantitative antimicrobial susceptibility testing of Haemophilus influenzae and Streptococcus pneumoniae by using the E-test. J Clin Microbiol. 1991;29:109–114.
23. Berman SOtitis media in developing countries. Pediatrics. 1995;96(1 pt 1):126–131.
24. Najnin N, Bennett CM, Luby SPInequalities in care-seeking for febrile illness of under-five children in urban Dhaka, Bangladesh. J Health Popul Nutr. 2011;29:523–531.
25. Sierra A, Lopez P, Zapata MA, et alNon-typeable Haemophilus influenzae and Streptococcus pneumoniae as primary causes of acute otitis media
in colombian children: a prospective study. BMC Infect Dis. 2011;11:4.
26. Madhi SA, Govender N, Dayal K, et alBacterial and respiratory viral interactions in the etiology of acute otitis media
in HIV-infected and HIV-uninfected South African children. Pediatr Infect Dis J. 2015;34:753–760.
27. Danishyar A, Ashurst JVOtitis, Media, Acute. StatPearls. 2017Treasure Island (FL): StatPearls Publishing;
28. DeAntonio R, Yarzabal JP, Cruz JP, et alEpidemiology of otitis media
in children from developing countries: a systematic review. Int J Pediatr Otorhinolaryngol. 2016;85:65–74.
29. Kilpi T, Herva E, Kaijalainen T, et alBacteriology of acute otitis media
in a cohort of Finnish children followed for the first two years of life. Pediatr Infect Dis J. 2001;20:654–662.
30. Ben-Shimol S, Givon-Lavi N, Leibovitz E, et alImpact of widespread introduction of pneumococcal conjugate vaccines on pneumococcal and nonpneumococcal otitis media
. Clin Infect Dis. 2016;63:611–618.
31. Sultana NK, Saha SK, Al-Emran HM, et alImpact of introduction of the Haemophilus influenzae type b conjugate vaccine into childhood immunization on meningitis in Bangladeshi infants. J Pediatr. 2013;163(1 suppl):S73–S78.
32. Baqui AH, El Arifeen S, Saha SK, et alEffectiveness of Haemophilus influenzae type B conjugate vaccine on prevention of pneumonia and meningitis in Bangladeshi children: a case-control study. Pediatr Infect Dis J. 2007;26:565571.
33. Hanquet G, Lernout T, Vergison A, et alImpact of conjugate 7-valent vaccination in Belgium: addressing methodological challenges. vaccine. 2011;29(16):28562864.
34. Pilishvili T, Lexau C, Farley MM, et alActive Bacterial Core Surveillance/Emerging Infections Program Network. Sustained reductions in invasive pneumococcal disease in the era of conjugate vaccine. J Infect Dis. 2010;201:32–41.
35. Williams SR, Mernagh PJ, Lee MH, et alChanging epidemiology of invasive pneumococcal disease in Australian children after introduction of a 7-valent pneumococcal conjugate vaccine. Med J Aust. 2011;194:116–120.
36. Miller E, Andrews NJ, Waight PA, et alHerd immunity and serotype replacement 4 years after seven-valent pneumococcal conjugate vaccination in England and Wales: an observational cohort study. Lancet Infect Dis. 2011;11:760–768.
37. Tan TQPediatric invasive pneumococcal disease in the United States in the era of pneumococcal conjugate vaccines. Clin Microbiol Rev. 2012;25:409–419.
38. Dagan R, Klugman KPImpact of conjugate pneumococcal vaccines on antibiotic resistance. Lancet Infect Dis. 2008;8:785–795.
39. Garau JTreatment of drug-resistant pneumococcal pneumonia. Lancet Infect Dis. 2002;2:404–415.
40. O’Brien KL, Millar EV, Zell ER, et alEffect of pneumococcal conjugate vaccine on nasopharyngeal colonization among immunized and unimmunized children in a community-randomized trial. J Infect Dis. 2007;196:1211–1220.
41. Weinberger DM, Malley R, Lipsitch MSerotype replacement in disease after pneumococcal vaccination. Lancet. 2011;378:1962–1973.
42. Mamishi S, Moradkhani S, Mahmoudi S, et alPenicillin-resistant trend of Streptococcus pneumoniae in Asia: a systematic review. Iran J Microbiol. 2014;6:198–210.
43. Hasanuzzaman M, Islam M, malakar R, et alRecent emergence of highly macrolide-resistance pneumococcus
in Bangladesh- relation to antibiotic use and genetic determinants. 9th World Congress of the World Society for Pediatric Infectious Diseases. November 2015:1.
44. Hampton LM, Zell ER, Schrag S, et alSentinel versus population-based surveillance of pneumococcal conjugate vaccine effectiveness. Bull World Health Organ. 2012;90:568–577.
otitis media; pneumococcus; nontypeable Haemophilus influenzae; pneumococcal serotype
Supplemental Digital Content
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.