Invasive Pneumococcal Disease in Kanti Children s Hospital, Nepal, as Observed by the South Asian Pneumococcal Alliance Network

Transcription

1 SUPPLEMENT ARTICLE Invasive Pneumococcal Disease in Kanti Children s Hospital, Nepal, as Observed by the South Asian Pneumococcal Alliance Network A. S. Shah, 1 M. Deloria Knoll, 2 P. R. Sharma, 1 J. C. Moisi, 2 P. Kulkarni, 2 M. K. Lalitha, 3 M. Steinhoff, 2 and K. Thomas 3 1 Institute of Medicine and Kanti Children s Hospital, Kathmandu, Nepal; 2 Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland; and 3 Christian Medical College, Vellore, India Background. Pneumonia accounts for 2 million deaths annually among children aged!5 years, with most of these deaths occurring in Africa and southern Asia. The South Asian Pneumococcal Alliance (SAPNA) network in Nepal is generating local epidemiological data to assist in the development of national and regional policies for prevention of pneumococcal and Haemophilus influenzae (Hib) disease. Methods. Children aged 2 months to 5 years with suspected invasive bacterial disease were recruited from Kanti Children Hospital, Kathmandu, Nepal. Specimens of blood, CSF, and normally sterile body fluids were cultured, and analysis of antimicrobial susceptibility patterns and serotyping of Streptococcus pneumoniae isolates were performed. CSF specimens were also tested for S. pneumoniae and Hib antigens by a latex agglutination test and an immunochromatographic test of pneumococcal antigen (NOW S. pneumoniae Antigen Test; Binax). Results. A total of 2528 children with suspected invasive bacterial disease were recruited, of whom 82% had pneumonia, 9.6% had meningitis, 2% had very severe disease, and 0.4% had bacteremia; the remainder received another diagnosis. Before hospitalization, 26.7% had received antibiotic treatment. Fifty children had S. pneumoniae identified as the etiological agent of invasive disease. Of 2461 blood cultures performed, 22 were positive for S. pneumoniae. Of 33 cases of S. pneumoniae meningitis, 11 were detected by CSF culture, and 21 were detected by latex agglutination and pneumococcal antigen tests. The rate of detection of S. pneumoniae in CSF was 3.6% by culture, compared with 7.8% by latex agglutination and 10% by pneumococcal antigen testing. The rate of detection of H. influenzae in CSF was 1.7% by culture and 6.5% by latex agglutination. The most common serotypes found were 1, 5, 2, and 7F, followed by 12A, 19B, and 23F. Of all the invasive isolates, 3.8% were resistant to penicillin, and 68% were resistant to trimethoprim-sulfamethoxazole. Conclusions. The SAPNA network has identified Hib and pneumococci as causes of significant disease in Nepal. The World Health Organization estimates that 1.6 million people die of pneumococcal disease every year, including million deaths among children aged!5 years, most of whom live in developing countries [1]. In populations with high rates of child mortality, pneumonia is the leading infectious cause of mortality and accounts for 20% 25% of all child deaths [2]. Nepal, one of the world s poorest nations, with a per capita income of!$250 and a total population of 26 million, has a population of children aged!1 year of 742,164 Reprints or correspondence: Dr. Aparna Singh Shah, Institute of Medicine, Teaching Hospital, Maharajgunj, Kathmandu, Nepal (rajuladi@yahoo.com). Clinical Infectious Diseases 2009; 48:S by the Infectious Diseases Society of America. All rights reserved /2009/4805S2-0012$15.00 DOI: / and a population of children aged!5 years of 3,633,687. The mortality rate among children aged!5 years is very high, at 91 deaths per 1000 live births [3, 4], with pneumonia as the leading cause of death. Nepal s low rate of treatment of suspected pneumonia cases is to blame for the high fatality rates: only 15% 18% of all patients with pneumonia who reside in rural or hilly areas are brought by caretakers to health care facilities, according to Ministry of Health estimates [4]. If diagnosis is early enough and treatment is appropriate, the majority of children with pneumococcal disease can recover. However, most children with these illnesses particularly those with meningitis and pneumonia require hospitalization, which may be difficult in developing countries where there is limited access to hospital or other health care facilities. Even when treated, the increasing numbers of antibiotic-resistant Pneumococcal Disease in Nepalese Children CID 2009:48 (Suppl 2) S123

2 Table 1. Definitions of clinical syndromes and criteria for final diagnoses. Diagnosis, definition and/or criteria Meningitis clinical syndrome Sudden onset of fever (at!48 h after diagnosis) and at least 1 of the following: Stiff neck Altered/reduced level of consciousness Bulging fontanelle (for patient aged!12 months) Lethargy Convulsions For patients aged!6 months: any convulsion For patients aged 16 months: focal, prolonged, or recurrent seizures Petechial or purpuric rash Pneumonia clinical syndrome Tachypnea For patients aged 2 to!12 months: respiration rate 150 breaths/min For patients aged 12 months to!5 years: respiration rate 140 breaths/min CSF Purulent 1 of the following criteria: Turbid CSF WBC count 100 cells/mm 3 WBC count of cells/mm 3 and glucose level!40 mg/dl or protein level 100 mg/dl Abnormal One of the following criteria and nonpurulent CSF: Glucose level normal or not tested Protein level normal or not tested pneumococcal infections lead to higher risks of treatment failure. The remarkable success of pneumococcal conjugate vaccines as evident in efficacy trials in both developed and developing countries and in the surveillance of invasive pneumococcal disease (IPD) after routine infant immunization in North America, the United Kingdom, and Australia challenges public health care practitioners throughout the world to quantify the burden of IPD in their region and to estimate the potential benefits of vaccine use [5]. Pneumococcal conjugate vaccines can prevent most serious pneumococcal diseases, but the distribution of individual pneumococcal serotypes causing serious disease varies from country to country [6]. The aims of the South Asian Pneumococcal Alliance (SAPNA) are to establish the presence of disease and to characterize antibioticsusceptibility patterns and serotype distributions of pneumococcal strains that are prevalent in the region. METHODS Study setting and sample population. Kanti Children s Hospital (KCH), situated in Kathmandu, is the largest children s hospital in Nepal. The Kathmandu Valley has 3 districts: Kathmandu, Lalitpur, and Bhaktapur. In 2001, the Kathmandu Valley had a total population of 1,645,091, of whom 129,784 ( 8%) were aged!5 years [3, 4]. Because KCH is a government hospital and provides affordable health care, 175% of acutely sick children from Kathmandu Valley come to this hospital for treatment. Children aged 2 months to 5 years who were admitted to Table 2. Distribution of the clinical syndromes of pneumonia and meningitis and detection of Streptococcus pneumoniae and Haemophilus influenzae (by culture and/or antigen testing) in patients aged!5 years, according to age group. Age group, months No. (%) of patients ( n p 2528) No. of patients who received the diagnosis No. of patients with specimens positive for S. pneumoniae No. of patients with specimens positive for H. influenzae Pneumonia Meningitis Pneumonia strain Meningitis strain Pneumonia strain Meningitis strain (59) (22) (20) S124 CID 2009:48 (Suppl 2) Shah et al.

3 Table 3. Results of blood culture, CSF culture, antigen test, and latex agglutination (LA) test. Test performed No. of specimens No. (%) of specimens positive for Streptococcus pneumoniae Blood culture (0.9) CSF Culture (3.7) LA test (7.8) Antigen test a (10.0) a An immunochromatographic test of pneumococcal antigen (NOW S. pneumoniae Antigen Test; Binax). KCH from November 2004 through March 2007 with fever (temperature, 138 C) and/or a possible clinical diagnosis of pneumonia, meningitis, or septicemia were considered for enrollment in the study. Enrollment of patients was performed 24 h per day, 7 days per week. Three full-time research officers (medical graduates) were hired for the project. After written informed consent was obtained, a detailed clinical research form with questions about patient history and findings of physical examination was filled out by the SAPNA research officers. Patient exclusion criteria were hospitalization for any illness within the previous 10 days, fever of duration 15 days, and a history of recurrent wheezing. The study was approved by the ethics review board of the Institute of Medicine, Kathmandu, and the Nepal Health Research Council. Table 4. Diagnosis Identification of organisms, according to disease category, among children aged!5 years. No. of patients No. of specimens No. of specimens without culture growth (no. of deaths) Case definitions. The definitions of clinical syndromes and final diagnoses are listed in table 1. These definitions are based on criteria established by PneumoADIP (see the Appendix in this supplement), with minor alterations for the data collected for our population. Findings on chest radiographs were reported by the local clinical team or radiologist. Laboratory methods and data management. Blood specimens were collected from all study participants for cell count and culture. Blood culture was performed in brain heart infusion broth (BHI), which was prepared, quality checked, and delivered to KCH by the Institute of Medicine Reference Laboratory, Kathmandu, Nepal. BHI bottles were incubated for 12 h after sample inoculation, and specimens were subcultured on chocolate agar, sheep blood agar, and MacConkey agar in accordance with standard microbiology protocols [7, 8]. Lumbar punctures were performed for all suspected cases of meningitis, and CSF specimens were examined for WBC counts and glucose and protein levels. All CSF specimens with WBC counts 110 cells/mm 3 were tested with a latex agglutination test (Combo Kit; Becton Dickinson) for Streptococcus pneumoniae, group B Streptococcus, Escherichia coli, Neisseria meningitidis, and Haemophilus influenzae. An immunochromatographic test of pneumococcal antigen (NOW S. pneumoniae Antigen Test; Binax) was also performed on all available CSF specimens, to detect S. pneumoniae antigens. All isolates grown in the KCH laboratory were confirmed by the Institute of Medicine reference laboratory. All S. pneumoniae isolates (placed on chocolate agar slants) were shipped to the Christian Medical College Labo- Streptococcus pneumoniae a No. of positive results (no. of deaths) Haemophilus influenzae a Neisseria meningitides b Meningitis All (5) 33 (2) 19 (2) 2 23 (1) 0 Definite (2) 17 (1) Probable (1) 3 2 (1) 1 7 (1) 0 Suspected (4) Pneumonia All (24) (6) 0 CXR confirmed, very severe (8) (1) 0 Probable, very severe (1) (1) 0 CXR confirmed, severe (12) (2) 0 Probable, severe (2) CXR confirmed (1) (2) 0 Probable Very severe disease Bacteremia Other (3) Total (32) 51 (2) 19 (2) (7) 1 Other Missing data NOTE. The latex agglutination (LA) and pneumococcal antigen tests were performed only on CSF specimens. CXR, chest radiograph. a By culture or an immunochromatographic test of pneumococcal antigen (NOW S. pneumoniae Antigen Test; Binax). b By culture. Pneumococcal Disease in Nepalese Children CID 2009:48 (Suppl 2) S125

4 Table 5. WBC counts in CSF specimens and results of CSF culture for Streptococcus pneumoniae, Haemophilus influenzae, and Neisseria meningitidis. WBC count, cells/mm 3 No. of culture-positive specimens, by pathogen S. pneumoniae H. influenzae N. meningitidis ratory in Vellore, India, for serotyping and antimicrobial susceptibility testing. Antimicrobial susceptibility testing (by the Kirby-Baur method) was performed on Mueller Hinton agar supplemented with sheep blood, in accordance with World Health Organization guidelines [9]. All clinical and microbiological data were uploaded weekly to a centralized, Web-based data management system in Vellore, India. RESULTS During the 28-month study period, 2528 children aged!5 years who had suspected invasive bacterial disease were admitted to KCH and were enrolled in the study (a total of 14,852 children were admitted to KCH for any cause during this 28-month period). Of 2528 enrolled children, 1485 (58.7%) were aged 0 11 months, 547 (21.6%) were aged months, and 496 (19.6%) were aged months (table 2). Blood specimens were collected from 2461 children, and CSF specimens were collected from 300 children (table 3). According to parental report, 676 (26.7%) of the 2528 children had received antibiotic treatment in the 48 h before hospital admission. Of these 676 children, 4 (0.6%) had culture results positive for S. pneumoniae, and 1 (0.1%) had culture results positive for H. influenzae. Of the 1852 children who did not receive antibiotic treatment before hospitalization, 26 (1.4%) had culture results positive for S. pneumoniae, and 4 (0.2%) had culture results positive for H. influenzae ( P p.73 for comparison of isolation of S. pneumoniae from recipients of antibiotic treatment vs. nonrecipients). Table 6. Outcomes for patients aged 5 years in the South Asian Pneumococcal Alliance, according to diagnostic status. Status Meningitis Pneumonia Clinical syndromes. Among the 2528 enrolled children, pneumonia was diagnosed in 2069 (81.8%), and meningitis was diagnosed in 243 (9.6%). Of 2069 pneumonia cases, 1920 (92.8%) were confirmed by chest radiograph. Of 243 meningitis cases, 92 (37.9%) were suspected, 90 (37.0.1%) were probable, and 61 (25%) were definite meningitis. Diagnosis of pneumonia was 8.5 times more frequent than was diagnosis of meningitis (table 4). Microbiologically confirmed pneumococcal disease. Of a total of 2461 blood cultures, 22 (0.9%) were positive for S. pneumoniae. A total of 300 CSF samples were collected and cultured; of which 11 (3.7%) were positive for S. pneumoniae, and 5 (1.7%) were positive for H. influenzae. A total of 244 CSF specimens were tested by latex agglutination, of which 19 (7.8%) had positive results, and 229 CSF specimens were tested for pneumococcal antigen, of which 23 (10.0%) had positive results (table 3). Of the 1485 children in the age group 0 11 months, 10 (0.7%) had S. pneumoniae pneumonia, and 25 (1.7%) had S. pneumoniae meningitis. Of the 547 children in the age group months, 2 (0.4%) had S. pneumoniae pneumonia, and 4 (0.7%) had S. pneumoniae meningitis. Of the 496 children in the age group months, 4 (0.8%) had S. pneumoniae pneumonia, and 4 (0.8%) had S. pneumoniae meningitis (table 2). WBC counts were determined for 295 CSF specimens obtained from enrolled patients: WBC counts of 0 9 cells/mm 3 were found in 103 (34.9%) of the specimens, WBC counts of cells/mm 3 were found in 85 (28.8%), and WBC counts 100 cells/mm 3 were found in 107 (36.3%). Overall, 9 (8%) of the CSF specimens with a WBC count 100 cells/mm 3 were culture positive for S. pneumoniae (table 5). On the basis of the 2001 census data for the Kathmandu Valley (129,784 children aged!5 years, of whom 97,338 [75%] went to KCH), the estimated crude annual incidence of IPD was 52.4 cases per 100,000 children aged!5 years. A total of 45 (1.8%) of the children enrolled in this study died during their hospital stay. Of these children, 10 had meningitis, 32 had pneumonia, and 3 received another diagnosis Nonmeningitis and nonpneumonia Cured or better Same Sequelae Discharged from the hospital in a morbid state Dead Not recorded or not applicable Discharged from the hospital against medical advice Total S126 CID 2009:48 (Suppl 2) Shah et al.

5 Figure 1. Antimicrobial-resistance patterns among isolates from patients aged!5 years. (table 6). Of all the deaths due to IPD-associated syndromes, 36 (80%) occurred among children aged!1 year. Of all invasive isolates, 4% were resistant to penicillin (1 CSF isolate was intermediately resistant, with an MIC of 0.5 ug/ml), 68% were resistant to trimethoprim-sulfamethoxazole (cotrimoxazole), 7% were resistant to erythromycin, and 4% were resistant to cefotaxime. None of the isolates showed resistance to chloramphenicol (figure 1). The most common pneumococcal serotypes were 1, 5, 2, and 7F, followed by 12A, 19B, and 23F (figure 2). DISCUSSION The SAPNA network has, for the first time, generated information on disease burden, serotype distribution, and antimicrobial-resistance patterns of invasive pneumococcal isolates from Nepal. Before this time, there was insufficient information available to make an evidence-based decision about whether to introduce pneumococcal vaccine into the national immunization program. Documentation of the presence of disease and characterization of prevalent serotypes are key factors in this decision-making process. The total pneumococcal yield from patients recruited on the basis of study inclusion criteria was 2% (50 of 2528), which is the same as that observed in India during the Invasive Bacterial Infection Surveillance (IBIS) study (2% [128 of 6025]) [10]. In developing countries, prehospitalization antibiotic use grossly affects isolation of S. pneumoniae in culture, which results in underestimation of disease burden. In our study, 676 (26.7%) of the patients had received antibiotics before hospitalization, this is almost double the rate documented in the IBIS study (13%). Antibiotic use reduced the rate of isolation of S. pneumoniae from 1.4% (26 of 1852) to 0.6% (4 of 676) ( P p.1), which is similar to the reduction in isolation rates seen in India (from 5.6% to 4%) [10]. In our study, 91% of pneumonia cases (1974 of 2174) had radiological confirmation, but!1% had blood cultures positive for S. pneumoniae. Although this low rate of isolation could be associated partly with a high rate of prehospitalization use of antibiotics, it is likely also because culture results in true cases of pneumococcal pneumonia are rarely positive for children. This was confirmed by 2 trials of 9-valent conjugate pneumococcal vaccine in Africa; the observed reductions in the rate of disease were 37% for radiologically confirmed pneumonia in The Gambia [11] and 20% in South Africa [12], despite low rates of positive blood culture results. Overall, the rate of penicillin resistance was 4%. The 1 isolate from CSF had intermediate resistance (MIC, 0.5 ug/ml) to penicillin, whereas none of the isolates from blood showed resistance to penicillin. In our study, 19 (68%) of S. pneumoniae isolates were resistant to cotrimoxaxole, of which 3 (16%) showed intermediate resistance. The IBIS study reported 48% resistance to cotrimoxazole, of which 7.5% was intermediate. In addition, 7% of our tested isolates were resistant to erythromycin, and 4% were resistant to cefotaxime, whereas all isolates were susceptible to chloramphenicol. By contrast, in the study by Saha et al. [13], cotrimoxazole resistance was seen in 64.1% of isolates, which is similar to the rate in our study. Very Figure 2. Serotype distribution of Streptococcus pneumoniae isolates found at Kanti Children s Hospital, Kathmandu, Nepal Pneumococcal Disease in Nepalese Children CID 2009:48 (Suppl 2) S127

6 that there is a substantial burden of pneumococcal disease in urban Nepal, which could be reduced by the use of appropriate pneumococcal conjugate vaccines. Figure 3. Projected coverage of Streptococcus pneumoniae serotypes by 7-valent, 10-valent, and 13-valent pneumococcal conjugate vaccines. high resistance to cotrimoxazole could be the result of frequent prescribing, easy availability of the drug, and the practice of prescribing it to treat suspected bacterial pneumonia cases. The data of Mastro et al. [14] showed that 2.8% of strains had resistance to chloramphenicol, and 1.1% had resistance to erythromycin. In our study, isolates showed no resistance to chloramphenicol, which may reflect the rare use of this drug in Nepalese hospitals and in the community. The most common serotypes (for both pneumonia and meningitis) of pneumococcal isolates in this study were 1, 5, 2, and 7F. None of these serotypes are covered by the currently available 7-valent conjugate vaccine, which would cover only 15% of serotypes isolated from our population. The 10-valent vaccine would cover 56% of serotypes isolated in our study, and the 13-valent vaccine would cover 63% (figure 3). In our study, the most common serotype was serotype 1, which accounted for 29% of isolates. This is similar to the findings in India, for which the IBIS study reported serotype 1 to be the most common, accounting for 25% of all isolates [10]. The common serotypes in our study also included serotypes 5 and 7F, which have been reported in Bangladesh [15]. Suboptimal access to facilities with adequate laboratory capacity combined with easy access to antibiotics may lead to underestimates of the incidence of IPD in Nepal. The potential efficacy of vaccine against IPD may result in a substantial impact on the high disease burden in developing countries; however, for a pneumococcal conjugate vaccine to be optimally effective in Nepal, it is preferable for more serotypes to be included. There is no nationwide surveillance program for pneumococcal disease in Nepal. Our study data are only from hospitalbased surveillance in the Kathmandu Valley, so the pneumococcal serotypes identified may not be representative of the entire country. However, our data may be used to inform policy decisions about the appropriateness of various pneumococcal vaccine formulations for Nepali children. This study suggests Acknowledgments We are grateful to all South Asian Pneumococcal Alliance team members in Nepal and Vellore and also to the staff and patients at Kanti Children Hospital, Kathmandu, Nepal, for their assistance in undertaking this project. Financial support. PneumoADIP and the Hib Initiative at Johns Hopkins University (the PneumoADIP and the Hib Initiative are funded in full by the GAVI Alliance and The Vaccine Fund). Supplement sponsorship. This article was published as part of a supplement entitled Coordinated Surveillance and Detection of Pneumococcal and Hib Disease in Developing Countries, sponsored by the GAVI Alliance s PneumoADIP of Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland. Potential conflicts of interest. All authors: no conflicts. References 1. World Health Organization. Pneumococcal conjugate vaccine for childhood immunization WHO position paper. Wkly Epidemiol Rec 2007; 82: Williams BG, Gouws E, Boschi-Pinto C, Bryce J, Dye C. Estimates of world-wide distribution of child deaths from acute respiratory infections. Lancet Infect Dis 2002; 2: Nepal Ministry of Health. Expanded program on immunization Fact sheet. 4. Nepal Ministry of Health and Population. Fact sheets. Kathmandu: Nepal Ministry of Health, Scott JAG. The preventable burden of pneumococcal disease in the developing world. Vaccine 2007; 25: Levine O, O Brien K, Knoll M. Pneumococcal vaccination in developing countries. Lancet 2006; 367: Lund E, Henrichsen J. Laboratory diagnosis, serology and epidemiology of Streptococcus pneumoniae. In: Bergen T, Norris J, eds. Methods in microbiology. Vol. 11. London: Academic Press, 1978: Lennette EH, Balows A, Hausler WJ Jr, Shadomy HJ. Manual of clinical microbiology. 4th ed. Washington, DC: American Society for Microbiology, 1985: World Health Organization. Guidelines for antimicrobial susceptibility testing. WHO document LAB/79.3. Geneva: World Health Organization, 1979: Invasive Bacterial Infection Surveillance Group, International Clinical Epidemiology Network. Prospective multi-centre hospital surveillance of Streptococcus pneumoniae disease in India. Lancet 1999; 353: Cutts FT, Zaman SMA, Enwere G, et al. Efficacy of nine-valent pneumococcal conjugate vaccine against pneumonia and invasive pneumococcal disease in The Gambia: randomized, double-blind, placebocontrolled trial. Lancet 2005; 365: Saha SK, Rikitomi N, Ruhulamin M, et al. Antimicrobial resistance and serotype distribution of Streptococcus pneumoniae strains causing childhood infections in Bangladesh, 1993 to J Clin Microbiol 1999; 37: Saha SK, Rikitomi N, Biswas D, et al. Serotypes of S. pneumoniae causing invasive childhood infections in Bangladesh, 1992 to J Clin Microbiol 1997; 35: Mastro TD, Gafoor A, Khalid N, et al. Antimicrobial resistance of pneumococci in children with acute lower respiratory tract infection in Pakistan. Lancet 1991; 337: Klugman K, Madhi S, Huebner R, et al. A trial of a 9-valent pneumococcal conjugate vaccine in children with and those without HIV infection. N Engl J Med 2003; 349: S128 CID 2009:48 (Suppl 2) Shah et al.