Original Article Journal of Paediatric Respirology and Critical Care Invasive pneumococcal disease in children Iris Mei-Ching CHAN ,1* Daniel Kwok-Keung NG ,1 Ting-Yat MIU ,2 Ping LAM ,3 Cindy Wing- Sze TSE ,4 Chung-Hong CHAN ,1 POK-Yu CHOW ,1 Sharon Wan-Wah CHERK 1 1Department of Paediatrics, Kwong Wah ; 2Department of Paediatrics and Adolescent Medicine, Queen Elizabeth Hospital; 3Department of Paediatrics and Adolescent Medicine, ; 4Department of , , Hong Kong

Abstract Objective: Invasive pneumococcal disease (IPD) carries a significant disease burden worldwide. Since clinical details about IPD in Hong Kong are limited, the aim of our retrospective descriptive case study is to provide additional information on this area about IPD in Hong Kong. Methods: Clinical data were collected from three local district including Caritas Medical Centre, Kwong Wah Hospital and Queen Elizabeth Hospital during the period of 1995 to 2007. Results: Forty cases of IPD were identified and reviewed, including bacteraemia without focus (20), pneumonia (12), meningitis (7) and endocarditis (1). The median age was 3-year. The epidemiological data, clinical features, complications and long term sequelae of the disease were further analysed. Our data showed that 85% of IPD occurred between September and March. Convulsion was the commonest feature during the illness, occurring in 32.5% of the total cases. Convulsion was associated with a 5-fold increase in risk of severe morbidity and/or mortality. Conclusion: High index of suspicion to diagnose IPD is needed, especially in children under the age of 2 years or those with chronic medical illness because the presenting symptoms could mimic upper respiratory tract infection. Furthermore, they were associated with a significantly higher risk of mortality and morbidity.

Keywords: Bacteremia; Child; Meningitis, pneumococcal; Pneumococcal infections; Pneumonia, pneumococcal

Introduction invasive pneumococcal disease admitted to three public non-teaching hospitals in Hong Kong. Streptococcus pneumoniae is a leading cause of bacterial infection amongst children worldwide, being the most common cause of bacterial pneumonia and Methods an important cause of meningitis and bacteraemia. 1,2 Infection caused by S. pneumoniae produces This retrospective descriptive case review study was significant morbidity, mortality and long term sequelae conducted at the Departments of Paediatrics in in children. 2 , 3 The incidences for invasive Caritas Medical Centre, Kwong Wah Hospital and pneumococcal diseases (IPD) in Hong Kong were Queen Elizabeth Hospital. These 3 hospitals are reported to be 18.8 and 15.6 per 100 000 for children located in , Hong Kong. All clinical records aged ≤2 and ≤5 years respectively. 4 However, data of patients with positive culture of Streptococcus on the clinical features of IPD in Hong Kong children pneumoniae in normally sterile body fluid, i.e. blood, is limited. To address the foresaid issue, a pleural fluid, cerebrospinal fluid, were retrieved. The retrospective review was conducted for children with study periods were slightly different between the three hospitals, Kwong Wah Hospital (KWH) between 1995 and 2007, Queen Elizabeth Hospital (QEH) between 1997 and 2007, and Caritas Medical Centre (CMC) *Author to whom correspondence should be addressed. between 2000 and 2006. Their case summaries, Email: [email protected] clinical notes, blood tests, bacteriological culture

4 Volume 9 No. 2, June 2013 Original Article results and radiological imaging reports were retrieved next higher twofold dilution, as recommended in the and reviewed by the authors with entry to a standard instructions. Results were interpreted according to the form. published breakpoints of the Clinical and Laboratory Standards Institute (CLSI). The term non-susceptible Clinical definitions was used to denote both intermediate and IPD was defined as the isolation of S. pneumoniae in resistant isolates. For penicillin, the criteria were blood, cerebrospinal fluid or other normally sterile susceptible, ≤0.06 µg/mL; intermediate, 0.12-1 µg/mL; sites. Bacteraemia without focus (BWF) was defined resistant, ≥2 µg/mL. For cefotaxime, the criteria were as isolation of S. pneumoniae from blood without a susceptible, ≤1 µg/mL; intermediate, 2 µg/mL; focus of infection. A child was considered to have resistant, ≥4 µg/mL. pneumococcal meningitis when the clinical course was consistent with meningitis and S. pneumoniae Data analysis was isolated from cerebrospinal fluid. The diagnosis Statistical analyses were performed using SPSS of pneumococcal pneumonia was defined 11.0 (Mac OS X, SPSS Inc, Chicago, IL) Data were when clinical course and radiographic findings summarized as percentage for categorical variables were consistent with pneumonia with isolation and median for continuous variables. The associations of S. pneumoniae from blood, pleural fluid or between mortality and morbidity related to IPD and bronchoalveolar lavage. The diagnosis of selected risk factors (age <2-year and seizure) were pneumococcal endocarditis was considered definite studied by Fisher Exact test. Odds ratios with 95% when there was a consistent clinical course with confidence intervals were calculated. Two tailed suggestive echocardiogram and S. pneumoniae was p value of less than 0.05 was considered statistical isolated from the blood. Long term sequelae were significant. defined as neurological deficits (e.g. neurosensory hearing loss, blindness, abnormal visual evoked potential, cerebrovascular accident), epilepsy, heart Results failure and death. Forty cases of IPD were identified in the study period, Identification of Streptococcus pneumoniae 21 cases from KWH (52.5%), 14 cases from QEH Colonies of alpha haemolytic streptococcus with (35%) and 5 cases from CMC (12.5%). There was a colony morphology suspected of S. pneumoniae were slight female preponderance with a male: female ratio subcultured onto the blood agar plate with an optochin of 1:1.2. The age ranged from 2-month-old to 13-year- o old. The median age was 3-year. Twenty-six cases disc applied and incubated at 35 C in 5% CO 2. Most S. pneumoniae would have a >14 mm zone of were older than 2-year (65%). Fourteen cases were inhibition. The identification was further confirmed with below or equal to the age of 2-year (35%). Thirty-eight its bile solubility with 2% sodium deoxycholate on the children were Chinese (95%) and 2 of them were agar plate after incubation. South Asian (5%). Thirty-four cases (85%) occurred in cooler weather, i.e. from September to March, For susceptibility testing, E-test strips of penicillin and (Figure 1) which was similar to Ho et al.'s study. 4 cefotaxime (AB Biodisk, Solna, Sweden) were used to determine the minimum inhibitory concentration (MIC) of the isolates to penicillin and cefotaxime. E-test MICs were determined following the manufacturer's recommendations. Test inocula were prepared from pneumococcal colonies grown on horse blood agar that had been incubated for 20 to 24 hours in 5% CO 2. Colonies were suspended in 0.9% saline to obtain a suspension equivalent to a 0.5 McFarland standard of turbidity. From this suspension, E-tests were performed on Mueller-Hinton agar with 5% horse blood (BBL, Becton Dickinson Microbiology Systems, Cockeysville, MD). The plates were incubated at 35 oC Figure 1. Monthly distribution of 40 cases of invasive in 5% CO 2 for 20 to 24 hours. MICs falling between two marks on the E-test strip were rounded up to the pneumococcal disease.

5 Journal of Paediatric Respirology and Critical Care

Most of the cases presented as pneumonia (n=12, Pneumococcal meningitis was associated with poor 30%), fever without focus (n=11, 27.5%), febrile outcome. 5 In the current study, there were seven convulsion (n=7, 17.5%), upper respiratory tract cases of pneumococcal meningitis. Six of them had infection (n=6, 15%) and meningitis (n=4, 10%). Final convulsion during the illnesses and the remaining diagnoses included: 20 (50%) bacteraemia, 12 (30%) one had obvious clinical features of meningitis pneumonia, 7 (17%) meningitis, and 1 (2.5%) i.e. neck stiffness, headache and fever. Three out of endocarditis (Table 1). 7 meningitis patients recovered completely whilst 3 had long term neurological deficits (deafness, The most common presentation was convulsion blindness, cerebral palsy, epilepsy, abnormal visual (13/40, 32.5%). Only 6 out of these 13 convulsion evoked potential) and 1 died. cases had meningitis. The 7 convulsion patients without meningitis had no long term sequelae. Pneumococcal endocarditis was a rare and unusual Convulsions occurred significantly more frequent in complication in IPD that usually occurred in patient patients under 2-year (11/13), the odds ratio was with pre-existing heart structural lesion, which carried 5.6 (95% CI=1.33 to 23.6). Amongst those 13 cases significant morbidities and mortality. 6 However, our under 2 year, 6 were diagnosed to have meningitis case, a 39-month-old, had normal heart structure and (6/13, 46%), 7 cases had bacteraemia (7/13, 54%). good past health. The endocarditis was complicated Patients under 2-year and having convulsion (n=11) by perforation of the anterior mitral valve leaflet had a 15-fold increase in risk for long term sequelae resulting in severe mitral regurgitation and heart or mortality (95% CI=2.1-109.99). failure. The patient also suffered a stroke due to the septic embolism to the left middle cerebral artery Most patients who suffered from pneumonia recovered resulting in right hemiparesis. fully without any sequelae. Pleural effusion/empyema were identified in 2 out of 12 cases (17%). One of Four patients (10%) had complications due to IPD these 2 cases presented with shock, pericardial including pleural effusion/empyema, subdural effusion and disseminated intravascular effusion, syndrome of inappropriate ADH secretion coagulopathy. Both of them recovered without any (SIADH), central diabetic insipidus, shock, long term sequelae. disseminated intravascular coagulopathy, neurological deficits, perforated mitral valve and cerebrovascular event (Table 2).

Table 1. Distribution of sites of invasive pneumococcal Twenty-seven children (68.5%) were previously healthy. diseases and the respective age groups One or more underlying diseases were identified in 13 Diagnosis N (%) Age group, years (32.5%) children. Seven (17.5%) had an Advisory <= 2-year >2-yea r Committee on Immunization Practices (ACIP) defined Pneumonia 12 (30) 3 9 high risk medical condition 7 including 1 congenital Meningitis 7 (17) 5 2 immunodeficiency (i.e. X-linked agammaglobulinaemia); 3 acquired immunocompromised conditions (i.e. Bacteraemia 20 (50) 10 10 malignancy on chemotherapy, systemic lupus on Endocarditis 1 (2.5) 1 0 systemic steroid and nephrotic syndrome on systemic Total 40 (100) 19 (47.5%) 21 (52.5%) steroid); 3 congenital heart diseases (2 of them were

Table 2. Mortality and morbidity of invasive pneumococcal diseases Diagnosis N (%) Significant morbidity (%) Mortality (%) Full recovery (%)

Pneumonia 12 (30) 0 0 12 (100) Meningitis 7 (17) 3 (42.9) 1 (14.3) 3 (42.9) Bacteraemia 20 (50) 0 1 (5) 19 (95) Endocarditis 1 (2.5) 1 (100) 0 0 Total 40 (100) 4 (10) 2 (5) 34 (85)

6 Volume 9 No. 2, June 2013 Original Article associated with syndromal disorder and other co- who presented with convulsion had a 15-fold increase morbidities). Two cases had other chronic medical in risk for long term sequelae or mortality (p=0.01, conditions not defined as conferring a high risk for 95% CI=2.1-109.99). The other novel finding was that pneumococcal infection including systemic lupus 15% of the IPD cases in the current series presented erythematosus not on systemic steroid, congenital with symptoms of upper respiratory tract infection. CMV infection with cerebral palsy and epilepsy. Four This highlights the need for a high index of suspicion cases were born preterm. For these 13 children with for diagnosis of IPD especially in children below the pre-existing diseases, 7 had bacteraemia, 5 had age of 2 years or those with chronic medical illness pneumonia and 1 had meningitis. as they are more likely to suffer from mortality and morbidities as a result of IPD. There were 2 deaths reported in the current case series. One died of meningitis and the other died of Meningitis and endocarditis were associated with long septicaemia. The mortality rate was 5% in the current term sequelae in the current study. The long term IPD series. Both of them were under the age of sequelae included neurological deficits, heart failure, 2 years, giving a mortality rate of 14.3% in patient cerebral palsy and epilepsy. Pneumonia usually under 2 years, compared to no mortality in older recovered without any sequelae. The mortality in this patients. One child did not have any risk factor study was 5%. All the deaths occurred in those below whereas the other had velo-cardio-facial syndrome the age of 2-year. and repaired congenital heart disease. Both patients succumbed within 24 hours after admission. In recent years, Pneumococcal sp . resistance to penicillin and other antibiotics were reported in several Blood culture was positive for S. pneumoniae in countries including North America, 10-12 Hong Kong 13 33 cases (82.5%). Amongst 7 cases of meningitis, and Taiwan. 14 The penicillin resistance rate was 6 cases had positive culture in both blood and CSF increasing as it rose from 0.2% (1988) to 7.6% (1992) culture; whereas 1 case was positive only in CSF to 28.9% (1994). 15 In the current study, the penicillin culture. Around 58% of S. pneumoniae isolates were resistance rate was 42%, comparable with previous susceptible to penicillin; 42% were penicillin non- study that yielded 41%. 13 susceptible strains (PNSS). For erythromycin, 16% (4 out of 25) were susceptible and 84% were non- In order to decrease the morbidity and mortality in susceptible. Forty-one percent were trimethoprim- these young children, 2 types of Pneumococcal sulfamethoxazole susceptible. The high incidence of vaccines: pneumococcal conjugated and macrolide and trimethoprim-sulfamethoxazole polysaccharide vaccine were developed and resistance among clinical isolates of S. pneumoniae advocated recently. 7,16 Decline in incidence, disease was similar to that reported from Taiwan 8 and Hong burden of IPD was noted in many places after Kong. 4 All strains isolated in the current study were introduction of the new conjugate vaccine. 17-19 The sensitive to cefotaxime and vancomycin. disease burden in Hong Kong children aged below 2 years old can also be potentially reduced by vaccination. 20 The use of new conjugate vaccine has Discussion also showed reduced antimicrobial resistance in some places already e.g. Israel, South Africa. 21-23 Therefore, IPD is predominantly a disease of young children vaccination should be considered in those high risk worldwide, affecting mainly children younger than 5 groups as recommended by ACIP. 7 years of age. 1-3 Similar to other reports, 1-4,9 our report confirmed that children in Hong Kong younger than In conclusion, the current study demonstrated the 2 years were more likely to suffer from any morbidity increased susceptibility to IPD in children under the (e.g. heart failure, neurological deficits) or mortality age of 2 years or those with chronic medical illness with the odds ratio being 14 (95%CI=1.42-135.5). and IPD may present only with symptoms of upper respiratory tract infection. A high index of suspicion The current study showed that convulsion is the is required to allow early diagnosis and treatment that commonest presenting feature of IPD, accounting for prevent mortalities and morbidities associated with 32.5% of the total cases. Patients under 2-year-old IPD.

7 Journal of Paediatric Respirology and Critical Care

References Prevalence of antimicrobial resistance among respiratory tract isolates of Streptococcus pneumoniae in North America: 1997 1. Bjornson GL, Scheifele DW, Halperin SA; Canadian Paediatric results from the SENTRY antimicrobial surveillance program. Society/Health Canada Immunization Monitoring Program, Clin Infect Dis 1998;27(4):764-70. Active. Population-based epidemiology of invasive 13. Ip M, Lyon DJ, Cheng AF. Patterns of antibiotic resistance, pneumococcal infection in children in nine urban centers in serotype distribution, and patient demographics of Canada, 1994 through 1998. Pediatr Infect Dis J 2002;21 Streptococcus pneumoniae in Hong Kong. Chemotherapy (10):947-50. 2001;47(2):110-6. 2. Ulloa-Gutierrez R, Avila-Aguero ML, Herrera ML, Herrera JF, 14. Chen YY, Yao SM, Chou CY, Chang YC, Shen PW, Huang Arguedas A. Invasive pneumococcal disease in Costa Rican CT, et al. Surveillance of invasive Streptococcus pneumoniae children: a seven year survey. Pediatr Infect Dis J 2003;22 in Taiwan, 2002-2003. J Med Microbiol 2006;55(Pt 8):1109- (12):1069-74. 14. 3. Lim LH, Lee WS, Parasakthi N. Childhood invasive 15. Lai WM. Antimicrobial susceptibility and serotype distribution pneumococcal disease: a hospital-based study from Malaysia. among clinical isolates of Streptococcus pneumoniae in Hong J Paediatr Child Health 2007;43(5):366-9. Kong. Bulletin of the Hong Kong Society for Infectious Diseases 4. Ho PL, Chiu SS, Cheung CH, Lee R, Tsai TF, Lau YL. Invasive 1998;2:1. pneumococcal disease burden in Hong Kong children. Pediatr 16. Kyaw MH, Jones IG and Campbell H. Prevention of Infect Dis J 2006;25(5):454-5. pneumococcal disease in children. Pneumococcal conjugate 5. Casado-Flores J, Aristegui J, de Liria CR, Martinon JM, vaccines: their use globally could have a major impact on Fernandez C; Spanish Pneumococcal Meningitis Study Group. public health. Acta Paediatr 2001;90(5):473-6. Clinical data and factors associated with poor outcome in 17. Bjornson G, Scheifele DW, Bettinger J, Patrick DM, Gustafson pneumococcal meningitis. Eur J Pediatr 2006;165(5):285-9. L, Daly P, et al. Effectiveness of pneumococcal conjugate 6. Givner LB, Mason EO Jr, Tan TQ, Barson WJ, Schutze GE, vaccine in Greater Vancouver, Canada: 2004-2005. Pediatr Wald ER, et al. Pneumococcal endocarditis in children. Clin Infect Dis J 2007;26(6):540-2. Infect Dis 2004;38(9):1273-8. 18. Center KJ. Prevenar vaccination: review of the global data, 7. Centers for Disease Control and Prevention (CDC). Preventing 2006. Vaccine 2007;25(16):3085-9. pneumococcal disease among infants and young children. 19. Poehling KA, Talbot TR, Griffin MR, Craig AS, Whitney CG, Recommendations of the Advisory Committee on Immunization Zell E, et al. Invasive pneumococcal disease among infants Practices (ACIP). MMWR Recomm Rep 2000;49(RR-9):1-35. before and after introduction of pneumococcal conjugate 8. Hsueh PR, Teng LJ, Lee LN, Yang PC, Ho SW, Luh KT. vaccine. JAMA 2006;295(14):1668-74. Extremely high incidence of macrolide and trimethoprim- 20. Ho PL, Chiu SS, Chow FK, Mak GC, Lau YL. Pediatric sulfamethoxazole resistance among clinical isolates of hospitalization for pneumococcal diseases preventable by Streptococcus pneumoniae in Taiwan. J Clin Microbiol 1999; 7-valent pneumococcal conjugate vaccine in Hong Kong. 37(4):897-901. Vaccine 2007;25(39-40):6837-41. 9. McIntosh ED, Booy R. Invasive pneumococcal disease in 21. Dagan RG, Porat N, Sikuler-Cohen M, et al. Immunization of England and Wales: what is the true burden and what is the toddlers attending day care centers (DCC) with transmission of potential for prevention using 7 valent pneumococcal Streptococcus pneumoniae (Pnc) and antibiotics resistant conjugate vaccine? Arch Dis Child 2002;86(6):403-6. S. pneumoniae (R-Pnc) to their young siblings. 40th Interscience 10. Butler JC, Hofmann J, Cetron MS, Elliott JA, Facklam RR, Conference on Antimicrobial Agents and Chemotherapy 2000; Breiman RF. The continued emergence of drug-resistant 687. Streptococcus pneumoniae in the United States: an update 22. Dagan R, Melamed R, Muallem M, Piglansky L, Greenberg D, from the Centers for Disease Control and Prevention's Abramson O, et al. Reduction of nasopharyngeal carriage of Pneumococcal Sentinel Surveillance System. J Infect Dis pneumococci during the second year of life by a heptavalent 1996;174(5):986-993. conjugate pneumococcal vaccine. J Infect Dis 1996;174(6): 11. Centers for Disease Control and Prevention (CDC). Geographic 1271-8. variation in penicillin resistance in Streptococcus pneumoniae 23. Mbelle N, Huebner RE, Wasas AD, Kimura A, Chang I, --selected sites, United States, 1997. MMWR Morb Mortal Wkly Klugman KP. Immunogenicity and impact on nasopharyngeal Rep 1999;48(30):656-61. carriage of a nonavalent pneumococcal conjugate vaccine. 12. Doern GV, Pfaller MA, Kugler K, Freeman J, Jones RN. J Infect Dis 1999;180(4):1171-6.

8