The highest incidence of invasive pneumococcal disease (IPD) is in children younger than 2 years old and in the elderly.1 Streptococcus pneumoniae is also believed responsible for a large fraction of pneumonia cases, most of which are noninvasive. However, the lack of validated diagnostic techniques makes it difficult to determine the precise contribution of the pneumococcus. Overall the pneumococcus may account for >1 million deaths of young children.2
There is little information regarding the incidence of IPD and of pneumonia in Latin America. In 1 hospital-based study in Santiago, Chile, the annual IPD incidence was 40–50 cases/105 children younger than 24 months old.3 In a follow-up study that also included ambulatory patients with high fever, the reported incidence almost doubled as 1 case of ambulatory pneumococcal bacteremia was detected for every case of hospitalized pneumococcal disease.4
Our study used active, population-based surveillance to determine the incidence of IPD and of community-acquired pneumonia (CAP) without bacterial isolation in young children in the city of Cordoba, Argentina.
MATERIALS AND METHODS
We conducted the study in Cordoba, a city in central Argentina (population, 1,236,000; mean per capita income of 7000 US$). Based on child registrations at the primary health care centers, 57% of the total pediatric population receives health care from the city health system, including 4 pediatric hospitals (Hospital de Niños de la Santisima Trinidad, Hospital Infantil, Hospital Pediatrico del Niño Jesus and Hospital de Nuestra Señora de la Misericordia) and a network of community health centers (CHC). The remaining 43% of the population uses private health care.
Children 2–23 months old residing in Cordoba city and receiving care from the government health system were the subjects. The children came from middle and poor working class families.
Children with an axillary temperature of ≥39°C on presentation to the health care system, or who were reported by their parent or guardian to have had fever ≥39°C during the preceding 24 hours, had a blood culture performed. Approximately 32% of the children presenting with occult bacteremia were reported to have already received antibiotics before blood culture. Children with clinical suspicion of pneumonia (major signs standardized across the hospitals and primary health care centers included tachypnea, abnormal auscultation, chest indrawing, hypoxemia, grunting, nasal flaring) were referred for a chest radiograph (posteroanterior and lateral views); those with radiographically confirmed consolidation as defined below had a blood culture performed.
In patients with clinical syndromes compatible with meningitis, peritonitis, pericarditis, arthritis, osteomyelitis, pleural effusion and/or sepsis, blood cultures were repeated, and an appropriate needle puncture was performed to collect cerebrospinal, pleural or synovial fluid for culture.
Potential cases of IPD were detected at the 80 CHCs and the emergency departments of the 4 hospitals where diagnoses were confirmed. During the study period (December 1, 1999–November 30, 2002), patients who met inclusion criteria were identified by trained pediatricians working closely with personnel at CHCs and the hospitals.
Blood cultures were processed within 8–12 hours by automated culture systems (Bact-Alert). All bacteriologic samples were processed at the laboratories of Hospital Infantil or Hospital de Niños de la Santisima Trinidad according to standard techniques. Isolates were serotyped by the quellung reaction (antisera from Statens Seruminstitut, Copenhagen, Denmark) by Dr Yudowski at the Hospital Infantil of Córdoba with quality control by the Institute of Public Health in Santiago, Chile.
Bacteremia without a focus was defined as isolation of S. pneumoniae from blood cultures in patients with no focus of infection. Pneumococcal pneumonia (PP) represents dense consolidation of >2.5 cm without atelectasis involving a segment, lobe or lung on chest roentgenogram and isolation of S. pneumoniae from blood or pleural fluid. Obvious pneumonia (OP) was shown by dense consolidation >2.5 cm without atelectasis involving a segment, lobe or lung on chest roentgenogram, but without etiologic diagnosis. Radiologic outcomes were defined by agreement between the pediatrician and the pediatric radiologist in each center; in case of disagreement, the radiograph was reread by a radiologist at another hospital, and his decision was final.
Continuous variables were analyzed with mean, median, standard deviations and maximum or minimum quartiles. Nominal variables were analyzed with enumeration and proportions. Confidence intervals were calculated by method of Fleiss. Since the study started in December, the annual incidence was calculated for each 12-month period beginning in December through November of the following year.
The study was approved by the ethical review board and research committees of the 4 hospitals.
During the 3-year period, the population under surveillance comprised 86,211 children 2–23 months old; 21,903 children (25.4%) met the inclusion criteria, and IPD and OP were diagnosed in 179 and 2112 patients, respectively.
Invasive Pneumococcal Disease.
The 71, 60 and 48 IPD cases in 2000, 2001 and 2002, respectively, represent a mean global incidence of 206.8 cases/105 children per year, (256.6, 215.8 and 165.8 for 2000, 2001 and 2002, respectively) in the population under surveillance. The peak incidence (238.2/105 children/year) was in the group 6–11 months old.
In patients with positive S. pneumoniae cultures, the most frequent clinical presentations were bacteremia without a focus [47.5% of all cases; 95% confidence interval (CI), 40.2–55.1] and PP (45.8%; 95% CI 38.4–53.4). Other invasive infections were meningitis and abscesses (Table 1).
The OP incidences in 2000, 2001 and 2002 were 2622, 2248 and 2398 cases/105 children per year, respectively, for a mean of 2422 (95% CI 2321–2527). The peak incidence was in infants 6–11 months of age (2943 cases/105 children/year).
S. pneumoniae was isolated from blood cultures in 62 (2.8%) of 2174 cases of overall pneumonia (PP + OP), and an additional 20 pneumococcal isolates were cultured only from pleural fluid samples, yielding an incidence of confirmed PP of 95 cases/105 children/year. Pneumococcal isolation rates in pneumonia patients were inversely related to age, at <0.5% in young infants but >7% in children 18–23 months of age.
There were 61.5% of IPD patients managed as outpatients according to guidelines at each hospital; 88% of patients with bacteremia without a focus, but only 37.8% of patients with PP, were treated as outpatients. Younger patients were generally hospitalized, whereas older subjects were generally managed as outpatients (Table 1).
One death occurred in 2000 (meningitis), another in 2001 (PP) and none in 2002, yielding annual case-fatality rates among IPD patients of 1.4, 1.7 and 0%, respectively.
Of 179 isolates, 153 were serotyped. The most frequent serotypes were 14 (45.6%), 6B and 1 (each 10.8%), 18C (9.5%) and 5 (7.6%). Serotype 14 was the most common serotype in each of the 4 age groups; 53 and 39% of serotype 14 isolates were from pneumonia and bacteremia patients, respectively. Serotype 5 was most frequently isolated in young infants and 18C was found only in patients with bacteremia who were older than 6 months old and comprised 15% of occult bacteremia isolates. Finally, although in 2000 and 2002 serotype 1 was a major serotype, it was not isolated during 2001.
Using active ambulatory and hospitalized population-based surveillance, this study determined that the IPD incidence in Cordoba, Argentina is similar to that in the prepneumococcal conjugate vaccine era in the United States5 (206.8 versus 166 cases/105 children younger than 2 years old, respectively) and substantially higher than that recently reported in Chile (79.1).4 Several hospital-based European studies also report much lower rates in children younger than 2 years old, including: the United Kingdom, 42.16; Denmark, 34.97; Germany, 19.78; and Spain, 59.6 cases/105/year.9 The inclusion of outpatients in our study and the U.S. study5 likely contributes to much of this difference. However, given the higher rate of hospitalized IPD in Cordoban 2- to 23-month-olds (78.6/105 per year) compared with the European and Chilean studies, the use of a relatively low clinical threshold for blood culturing in this study is probably also an important factor.10
During the 3 years of the study, we noted a gradual decline in the annual incidences of IPD, but not those of CAP. We suggest that this decline, also seen in older children not within the present study (data not shown), may be caused by secular variations in IPD incidence as noted by others5 and is unlikely to be related to the minimal uptake (estimated at <5% of the eligible population) of the 7-valent pneumococcal conjugate vaccine.
The incidence of OP detected here was 25-fold that of confirmed PP (2422 versus 95 cases/105 per year). It is likely that the PP rates markedly underestimate the true contribution of S. pneumoniae to pneumonia in these children, as pneumococcal conjugate vaccines have been shown to prevent >20% of radiographically confirmed pneumonias with consolidation, most culture-negative, as well as a proportion of nonconsolidated disease (the latter not captured here).2,11,12
An important limitation of this study is that surveillance was restricted to children who used the public health system, and it is possible that children seen at the private health clinics and hospitals have different rates of disease. However, we believe that the high level of surveillance and standardized case detection for the public sector children in this study would support the generalizability of these results to the overall pediatric population in Cordoba.
Mortality rates were low (<2%) in Cordoba, similar to those of the United States5 and 1 report13 in Spain, but lower than some other studies (9.9 and 11.8%).4,14 Inclusion of occult bacteremic patients in this study, with a very low case fatality rate even if untreated, is undoubtedly part of the explanation. In addition, we speculate that participation of well-trained pediatricians at primary health centers in Cordoba, along with the low blood culturing threshold, allowed for early diagnosis and rapid hospitalization, thus preventing severe complications.
Regarding serotypes, the isolation rate of serotypes 1 and 5 (18.4%) is similar or slightly higher than that reported for most regions outside North America.15 The 7-valent (serotypes 4, 6B, 9V, 14, 18C, 19F and 23F), 9-valent (plus serotypes 1 and 5) and 11-valent (plus serotypes 3 and 7F) pneumococcal vaccines would cover 73, 91 and 95% of isolated serotypes, respectively.
These data suggest that introduction of a pneumococcal conjugate vaccine containing serotypes 1 and 5 into the Expanded Programme on Immunization could have a major impact on the health of Argentine children.
1. Teele DW. Pneumococcal Infection
. 5th ed. chap 90. Philadelphia, PA: W. B. Saunders Co.; 2003.
2. Cutts FT, Zaman SM, Enwere G, et al. Efficacy of nine-valent pneumococcal conjugate vaccine against pneumonia and invasive pneumococcal disease
in The Gambia: randomised, double-blind, placebo-controlled trial. Lancet
3. Levine M, Lagos R, Levine O, et al. Epidemiology of invasive pneumococcal infections in infants and young children in Metropolitan Santiago, Chile, a newly industrializing country. Pediatr Infect Dis J
4. Lagos R, Muñoz A, Valenzuela MT, et al. Population-based surveillance for hospitalized and ambulatory pediatric invasive pneumococcal disease
in Santiago, Chile. Pediatr Infect Dis J
5. Robinson KA, Baughman W, Rothrock G, et al. Epidemiology of invasive Streptococcus pneumoniae
infections in the U.S., 1995–1998: opportunities for prevention in the conjugate vaccine era. JAMA
6. Ispahani P, Slack RCB, Donald FE, et al. Twenty year surveillance of invasive pneumococcal disease
in Nottingham: serogroups responsible and implications for immunization. Arch Dis Child
7. Kaltoft MS, Zeuthen N, Konradsen HB. Epidemiology of invasive pneumococcal infections in children aged 0–6 years in Denmark: a 19-year nationwide surveillance study. Acta Paediatr
8. von Kries R, Hermann M, Hachmeister A, et al. Prediction of the potential benefit of different pneumococcal conjugate vaccines on invasive pneumococcal disease
in German children. Pediatr Infect Dis J
9. Domínguez A, Salleras L, Cardeñosa N, et al. The epidemiology of invasive Streptococcus pneumoniae
disease in Catalonia (Spain): a hospital-based study. Vaccine
10. Hausdorff WP, Siber G, Paradiso PR. Geographical differences in invasive pneumococcal disease
rates and serotype frequency in young children. Lancet
11. Black S, Shinefield H, Ling S, et al. Effectiveness of heptavalent pneumococcal conjugate vaccine in children younger than five years of age for prevention of pneumonia. Pediatr Infect Dis J
12. 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
13. Perez Mendez C, Solis Sanchez G, Miguel Martinez D, et al. Predictive factors for invasive pneumococcal disease
, a case control study. An Esp Pediatr
14. Espin MI, Sandoval A, Ruiz J, et al. Invasive pneumococcal disease
in children in the region of Murcia (Spain). Gacet Sanit
15. Hausdorff WP, Bryant J, Paradiso PR, Siber GR. Which pneumococcal serogroups cause the most invasive disease: implications for conjugate vaccine formulation and use, part I. Clin Infect Dis
. 2000;30: 100–121.