SciELO - Scientific Electronic Library Online

vol.40 issue1The burden of disease due to tuberculosis in the state of Santa Catarina, BrazilAssessment of ICU readmission risk with the Stability and Workload Index for Transfer score author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Jornal Brasileiro de Pneumologia

On-line version ISSN 1806-3756

J. bras. pneumol. vol.40 no.1 São Paulo Jan./Feb. 2014 

Brief Comunication

Empyema and bacteremic pneumococcal pneumonia in children under five years of age*, **

Maria Regina Alves Cardoso1 

Cristiana Maria Costa Nascimento-Carvalho2 

Fernando Ferrero3 

Eitan Naaman Berezin4 

Raul Ruvinsky5 

Clemax Couto Sant'Anna6 

Maria Cristina de Cunto Brandileone7 

Maria de Fátima Bazhuni Pombo March8 

Ruben Maggi9 

Jesus Feris-Iglesias10 

Yehuda Benguigui11 

Paulo Augusto Moreira Camargos12 

the CARIBE group

1Associate Professor. Department of Epidemiology, University of São Paulo School of Public Health, São Paulo, Brazil

2Associate Professor II. Federal University of São Paulo Paulista School of Medicine, São Paulo, Brazil

3Researcher. Pedro de Elizalde Children's Hospital, Buenos Aires, Argentina

4Associate Professor. Department of Pediatrics, Santa Casa School of Medical Sciences in São Paulo, São Paulo, Brazil

5Head. Department of Maternal and Child Health, Durand Municipal Hospital, Buenos Aires, Argentina

6Associate Professor. Department of Pediatrics, Federal University of Rio de Janeiro School of Medicine, Rio de Janeiro, Brazil

7Center Director. Adolfo Lutz Institute, São Paulo, Brazil

8Associate Professor. Fluminense Federal University, Niterói, Brazil

9Coordinator. Pediatric Clinic, Professor Fernando Figueira Institute of Integrative Medicine, Recife, Brazil

10Head. Department of Infectious Diseases, Dr. Robert Reid Cabral Children's Hospital, Santo Domingo, Dominican Republic

11Senior Neonatal and Child Health Advisor. Pan American Health Organization, World Health Organization, Washington, DC, USA

12Full Professor of Pediatrics. Federal University of Minas Gerais, Belo Horizonte, Brazil


We compared bacteremic pneumococcal pneumonia (BPP) and pneumococcal empyema (PE), in terms of clinical, radiological, and laboratory findings, in under-fives. A cross-sectional nested cohort study, involving under-fives (102 with PE and 128 with BPP), was conducted at 12 centers in Argentina, Brazil, and the Dominican Republic. Among those with PE, mean age was higher; disease duration was longer; and tachypnea, dyspnea, and high leukocyte counts were more common. Among those with BPP, fever and lethargy were more common. It seems that children with PE can be distinguished from those with BPP on the basis of clinical and laboratory findings. Because both conditions are associated with high rates of morbidity and mortality, prompt diagnosis is crucial.

Key words: Empyema, pleural; Pneumonia, pneumococcal; Pneumococcal infections


Comparamos crianças menores de cinco anos com pneumonia pneumocócica bacterêmica (PPB) àquelas com empiema pneumocócico (EP) quanto aos achados clínicos, radiológicos e laboratoriais. Um estudo de coorte aninhado transversal, com 102 crianças com EP e 128 com PPB, foi realizado em 12 centros na Argentina, no Brasil e na República Dominicana. Nas crianças com EP, a média de idade e a duração da doença foram maiores. Taquipneia, dispneia e contagem de leucócitos alta foram mais comuns nas crianças com EP; febre e letargia foram mais comuns naquelas com PPB. Parece possível distinguir crianças com EP de crianças com PPB a partir de achados clínicos e laboratoriais. Como essas duas doenças estão associadas a altas taxas de morbidade e mortalidade, o diagnóstico rápido é crucial.

Palavras-Chave: Empiema pleural; Pneumonia pneumocócica; Infecções pneumocócicas

Streptococcus pneumoniae is widely recognized as the leading cause of community-acquired pneumonia (CAP)-related mortality during hospitalization, as well as being the most common cause of empyema in children. The incidence of complicated pneumococcal CAP (CP-CAP) has been reported to be increasing worldwide.( 1 ) In Scottish children in the 1-4 year age bracket, there was a tenfold increase in empyema admissions in the period between the 1980s and the year 2005.( 2 ) This increase might be related to host susceptibility, individual characteristics, and pathogen virulence.

Few studies have investigated the characteristics of CP-CAP, especially those of CP-CAP associated with pleural empyema or bacteremia-although few cases of pneumococcal pneumonia are bacteremic-in children under 5 years of age, who are at the highest risk of death from CAP. Bacteremic pneumococcal pneumonia (BPP) is considered an advanced stage of severe pneumococcal pneumonia, in which early recognition and appropriate management can have a favorable impact on the outcome.( 3 ) Given that pleural empyema and BPP are potentially severe, it is important to recognize their clinical peculiarities in order to provide early and appropriate management.

In a recent study of hospitalized CAP patients, the prevalence rates of pleural effusion and empyema were reported to be 27% and 17%, respectively.( 3 ) In another study, empyema was found in 83 (11%) of 767 children hospitalized with CAP; in comparison with the children with CAP without empyema, those with empyema were older, had longer symptomatic periods, and were more likely to receive nonsteroidal anti-inflammatory drugs.( 4 ) The objective of the present study was to compare children under 5 years of age with BPP and those with pneumococcal empyema (PE) in terms of their clinical, radiological, and laboratory features.

A cross-sectional nested cohort study was conducted at 12 centers in Argentina, Brazil, and the Dominican Republic and included 2,536 children 3-59 months of age hospitalized with severe CAP, the results having been published elsewhere.( 5 ) On admission, blood cultures and, when appropriate, pleural fluid cultures were performed. S. pneumoniae was isolated from 283 children by standard procedures used in local referral laboratories. Most (90%) of the children lived in urban settings, and none had received the conjugate pneumococcal vaccine or steroidal/nonsteroidal anti-inflammatory drugs. Cases of BPP were defined as those in which S. pneumoniae was isolated from blood culture. Cases of PE were defined as those in which S. pneumoniae was isolated from pleural fluid culture.

The exclusion criteria were as follows: showing signs of very severe pneumonia (including severe malnutrition, stridor in a calm child, unconsciousness, convulsions, nasal flaring, and central cyanosis); and presenting with concurrent infections. Patients with concomitant PE and BPP (n = 3) and those with pleural effusion of unknown cause (n = 50) were also excluded.

A favorable response to the initial treatment (i.e., conventional doses of intravenous ampicillin or penicillin G for BPP and PE patients alike) was defined as unequivocal clinical improvement within the first 48 h after hospital admission; conversely, treatment failure was defined as no improvement (persistent fever-body temperature being measured at least every 6 h-tachypnea, difficulty breathing, or hypoxemia) after at least 48 h of antibiotic therapy or deterioration during antimicrobial therapy.( 5 ) Demographic, clinical, radiological, and laboratory data were obtained on admission. In addition, the length of hospital stay and the treatment outcome were recorded. Descriptive statistics were calculated, and two multiple logistic regression models were used, both of which were controlled for age. One of the models included variables obtained on admission and the other included variables obtained during hospitalization.

The present study included 230 children under 5 years of age (102 children with PE and 128 children with BPP). There was no significant difference between the two groups in terms of antibiotic use prior to hospital admission (p = 0.23). Table 1 displays the frequency distribution and the comparison (multivariate analysis) of the two groups of children on admission and during hospitalization.

Table 1  Comparison (multivariate analysis) of children with pneumococcal empyema and those with bacteremic pneumococcal pneumonia on admission and during hospitalization. 

Characteristic PE BPP Adjusted OR (95% CI) p
(n = 102) (n = 128)
On admission
Age (months)a 19 (12-33) 13 (9-22) 1.03 (1.00-1.05) 0.01
Duration of diseaseb
= 3 days 17.7 34.4 1 0.013
4-6 days 39.2 35.1 2.84 (1.23-6.58)
= 7 days 43.1 30.5 3.12 (1.35-7.20)
Tachypneab,c 91.2 75.6 4.16 (1.54-11.22) 0.005
Difficulty breathingb 87.9 70.9 3.61 (1.51-8.63) 0.004
Body temperature > 37.5°C on admissionb 42.4 68.2 0.32 (0.17-0.63) 0.001
Lethargyb 53.5 67.7 0.35 (0.17-0.71) 0.004
Leukocyte count > 15,000 cells/mm3b 48.5 29.8 2.13 (1.10-4.13) 0.02
During hospitalization
Difficulty breathing on the 3rd dayb 57.6 27.0 2.9 (1.52-5.56) 0.001
Body temperature (°C) on the 3rd daya 37.6 (37.0-38.8) 37.0 (36.6-38.0) 1.65 (1.05-2.59) 0.028
Treatment failureb 17.6 12.5 0.39 (0.14-1.07) 0.06
Length of hospitalizationa 11 (7-15) 7 (5-10.5) 1.09 (1.02-1.17) 0.004

PE: pneumococcal empyema

BPP: bacteremic pneumococcal pneumonia. aValues expressed as median (interquartile range)

bValues expressed as %. cRR > 50 breaths/min in the children 3-11 months of age and > 40 breaths/min in those >12 months of age.

In brief, our logistic regression analysis showed that, on hospital admission, children with PE were older and presented with a longer symptomatic period before hospitalization. In addition, tachypnea, difficulty breathing, and a leukocyte count > 15,000 cells/mm3 were more common in those children. Persistent fever was one of the most common findings during the first days of hospitalization, requiring further investigation (chest X-ray, chest ultrasound, or both). Fever (axillary temperature > 37.5°C) and lethargy at diagnosis/baseline tended to be more common in the children with BPP. The hospital stay tended to be longer for the children with PE (p < 0.001; data not shown).

In the children with PE, the most common serotypes were 14, 1, 6B, 3, 9V, 19A, and 5, accounting for 92.4% of all serotypes in those children. In the children with BPP, the most common serotypes were 14, 6B, 5, 1, 6A, 9V, and 19F, accounting for 84.9% of all serotypes in those children. In other words, serotypes 1, 5, 6B, and 14 were found in both PE and BPP, whereas serotypes 6A, 9V, 19A, and 19F were found in either PE or BPP. It is of note that serotypes 6A and 19A are not included in the 10-valent pneumococcal conjugate vaccine currently used in Brazil.

To the best of our knowledge, the present study is the first to have focused specifically on BPP and PE in children under 5 years of age. Therefore, the results presented herein cannot be compared with those of other studies. In fact, our analyses showed that demographic characteristics, clinical features, and hematologic findings (i.e., leukocyte counts) are likely to be quite different between children with BPP and those with PE.

Although our objectives and methods were different from those of François et al.,( 4 ) some of the results were similar between the two studies. François et al. studied 767 children with CAP and found that 83 (11%) had empyema.( 4 ) The children with CAP and empyema were older and had a longer symptomatic period in comparison with those with CAP without empyema.( 4 ) Interestingly, the length of hospital stay found in the present study was quite similar to that reported in a study involving 33 Brazilian children with empyema (i.e., 12 days).( 6 ) In addition, serotype 1 was one of four serotypes found in both PE and BPP (i.e., 14, 1, 6B, and 5), a finding that is consistent with those of the study by François et al.,( 4 ) which was conducted in a developed country.

In conclusion, our study investigated CAP patients under 5 years of age and showed clinical and laboratory findings at admission and during the first three days of hospitalization. Our findings can help clinicians and pediatricians to distinguish children with PE from those with BPP. Because these two critical conditions are associated with high rates of morbidity and mortality, prompt clinical, laboratory, and radiological diagnosis is crucial for appropriate management.


1. Grijalva CG, Nuorti JP, Zhu Y, Griffin MR Increasing incidence of empyema complicating childhood community-acquired pneumonia in the United States. Clin Infect Dis. 2010;50(6):805-13. [ Links ]

2. Roxburgh CS, Youngson GG, Townend JA, Turner SW. Trends in pneumonia and empyema in Scottish children in the past 25 years. Arch Dis Child. 2008;93(4):316-8. [ Links ]

3. Calado C, Nunes P, Pereira L, Nunes T, Barreto C, Bandeira T. Are there any differences in the community acquired pneumonias admitted to hospital over the past decade?. Rev Port Pneumol. 2010;16(2):287-305. PMid:20437005 [ Links ]

4. François P, Desrumaux A, Cans C, Pin I, Pavese P, Labarère J. Prevalence and risk factors of suppurative complications in children with pneumonia. Acta Paediatr. 2010;99(6):861-6. [ Links ]

5. Cardoso MR, Nascimento-Carvalho CM, Ferrero F, Berezin EN, Ruvinsky R, Camargos PA, et al. Penicillin-resistant pneumococcus and risk of treatment failure in pneumonia. Arch Dis Child. 2008;93(3):221-5. [ Links ]

6. Amorim PG, Morcillo AM, Tresoldi AT, Fraga Ade M, Pereira RM, Baracat EC. Factors associated with complications of community-acquired pneumonia in preschool children. J Bras Pneumol. 2012;38(5):614-21. [ Links ]

*Study carried out at the University of São Paulo School of Public Health, São Paulo, Brazil; the Federal University of Bahia School of Medicine, Salvador, Brazil; the Pedro de Elizalde Children's Hospital, Buenos Aires, Argentina; the Santa Casa School of Medical Sciences in São Paulo, São Paulo, Brazil; the Durand Municipal Hospital, Buenos Aires, Argentina; the Martagão Gesteira Institute of Pediatrics and Child Care, Rio de Janeiro, Brazil; the Center for Meningitis, Pneumonia, and Pneumococcal Infections, Department of Bacteriology, Adolfo Lutz Institute, São Paulo, Brazil; the Pernambuco Mother and Child Institute, Recife, Brazil; the Dr. Robert Reid Cabral Children's Hospital, Santo Domingo, Dominican Republic; the Department of Child and Adolescent Health and Development, Pan American Health Organization, Washington, DC, USA; and the Department of Pediatric Pulmonology, Federal University of Minas Gerais Hospital das Clínicas, Belo Horizonte, Brazil.

* Study carried out at the University of Southern Santa Catarina, Tubarão, Brazil.

** A versão completa em português deste artigo está disponível em

Received: July 16, 2013; Accepted: October 28, 2013

Correspondence to: Paulo Camargos Departamento de Pediatria, Faculdade de Medicina da Universidade Federal de Minas Gerais Avenida Alfredo Balena, 190, Sala 267 CEP 30130-100, Belo Horizonte, MG, Brasil Tel. 55 31 3409-9772. Fax: 55 31 3409-9664 E-mail: or

Financial support: This study received financial support from the Canadian International Development Agency (CIDA); the Pan American Health Organization; the World Health Organization Department of Child and Adolescent Health and Development; and the Boston University Applied Research on Child Health (ARCH) Project (United States Agency for International Development cooperative agreement no. HRN-A-00-960010-00).

Creative Commons License This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.