Etiology and the challenge of diagnostic testing of community-acquired pneumonia in children and adolescents.

BACKGROUND: Pneumonia is the leading cause of mortality in pediatric population. The etiology of pneumonia in this population is variable and changes according to age and disease severity and where the study is conducted. Our aim was to determine the etiology of community-acquired pneumonia (CAP) in children aged 1 month to 17 years admitted to 13 Colombian hospitals. METHODS: Prospective cohort study. Hospitalized children with radiologically confirmed CAP and ≤ 15 days of symptoms were included and followed together with a control group. Induced sputum (IS) was submitted for stains and cultures for pyogenic bacteria and Mycobacterium tuberculosis, and multiplex PCR (mPCR) for bacteria and viruses; urinary antigens for pneumococcus and Legionella pneumophila; nasopharyngeal swabs for viruses, and paired serology for atypical bacteria and viruses. Additional cultures were taken at the discretion of primary care pediatricians. RESULTS: Among 525 children with CAP, 71.6% had non-severe pneumonia; 24.8% severe and 3.6% very severe pneumonia, and no fatal cases. At least one microorganism was identified in 84% of children and 61% were of mixed etiology; 72% had at least one respiratory virus, 28% pyogenic bacteria and 21% atypical bacteria. Respiratory syncytial virus, Parainfluenza, Rhinovirus, Influenza, Mycoplasma pneumoniae, Adenovirus and Streptococcus pneumoniae were the most common etiologies of CAP. Respiratory syncytial virus was more frequent in children under 2 years and in severe pneumonia. Tuberculosis was diagnosed in 2.3% of children. IS was the most useful specimen to identify the etiology (33.6%), and blood cultures were positive in 3.6%. The concordance between all available diagnostic tests was low. A high percentage of healthy children were colonized by S. pneumoniae and Haemophilus influenzae, or were infected by Parainfluenza, Rhinovirus, Influenza and Adenovirus. CONCLUSIONS: Respiratory viruses are the most frequent etiology of CAP in children and adolescents, in particular in those under 5 years. This study shows the challenges in making an etiologic diagnosis of CAP in pediatric population because of the poor concordance between tests and the high percentage of multiple microorganisms in healthy children. IS is useful for CAP diagnosis in pediatric population.

Acute Intrathoracic Tuberculosis in Children and Adolescents with Community-Acquired Pneumonia in an Area with an Intermediate Disease Burden.

Tuberculosis (TB) in the pediatric population is a major challenge. Our objective was to describe the clinical and microbiological characteristics, radiological patterns, and treatment outcomes of children and adolescents (from 1 month to 17 years) with community-acquired pneumonia (CAP) caused by TB. We performed a prospective cohort study of a pediatric population between 1 month and 17 years of age and hospitalized in Medellín, Colombia, with the diagnosis of radiologically confirmed CAP that had ≤ 15 days of symptoms. The mycobacterial culture of induced sputum was used for the bacteriological confirmation; the history of TB contact, a tuberculin skin test, and clinical improvement with treatment were used to identify microbiologically negative TB cases. Among 499 children with CAP, TB was diagnosed in 12 (2.4%), of which 10 had less than 8 days of a cough, 10 had alveolar opacities, 9 were younger than 5 years old, and 2 had close contact with a TB patient. Among the TB cases, 50% (6) had microbiological confirmation, 8 had viral and/or bacterial confirmation, one patient had multidrug-resistant TB, and 10/12 had non-severe pneumonia. In countries with an intermediate TB burden, Mycobacterium tuberculosis should be included in the etiological differential diagnosis (as a cause or coinfection) of both pneumonia and severe CAP in the pediatric population.

Induced sputum as an adequate clinical specimen for the etiological diagnosis of community-acquired pneumonia (CAP) in children and adolescents.

OBJECTIVES: This study aimed to evaluate the utility of induced sputum (IS) for the diagnosis of community-acquired pneumonia (CAP) in pediatric population. METHODS: This cross-sectional study included pediatric population aged between 1 month and 17 years who were hospitalized with a diagnosis of CAP in 13 hospitals in Colombia, in whom an IS sample was obtained. Gram staining, aerobic bacterial and mycobacterial culture tests, and polymerase chain reaction (PCR) for 6 atypical bacteria and 15 respiratory viruses were performed. We evaluated the quality of IS samples. RESULTS: IS samples were collected in 516 of 525 children included in this study. The median age was 32 months, 38.6% were younger than 2 years, and 40.9% were between 2 and 5 years. Two patients had transient hypoxemia during the procedure. The quality of the IS obtained was good in 48.4% and intermediate in 24.5%. Identification of a respiratory pathogen was achieved with an IS sample (with Gram staining, culture test, and PCR) in 372 of 516 children with CAP. CONCLUSION: Our study shows that IS is an adequate sample for the diagnosis of CAP in pediatric population that required hospitalization. The procedure was safe, well tolerated, and with better diagnostic yields compared with the rest of the samples obtained.

Is It Possible to Differentiate Pneumocystis jirovecii Pneumonia and Colonization in the Immunocompromised Patients with Pneumonia?

Respiratory sample staining is a standard tool used to diagnose Pneumocystis jirovecii pneumonia (PjP). Although molecular tests are more sensitive, their interpretation can be difficult due to the potential of colonization. We aimed to validate a Pneumocystis jirovecii (Pj) real-time PCR (qPCR) assay in bronchoscopic bronchoalveolar lavage (BAL) and oropharyngeal washes (OW). We included 158 immunosuppressed patients with pneumonia, 35 lung cancer patients who underwent BAL, and 20 healthy individuals. We used a SYBR green qPCR assay to look for a 103 bp fragment of the Pj mtLSU rRNA gene in BAL and OW. We calculated the qPCR cut-off as well as the analytical and diagnostic characteristics. The qPCR was positive in 67.8% of BAL samples from the immunocompromised patients. The established cut-off for discriminating between disease and colonization was Ct 24.53 for BAL samples. In the immunosuppressed group, qPCR detected all 25 microscopy-positive PjP cases, plus three additional cases. Pj colonization in the immunocompromised group was 66.2%, while in the cancer group, colonization rates were 48%. qPCR was ineffective at diagnosing PjP in the OW samples. This new qPCR allowed for reliable diagnosis of PjP, and differentiation between PjP disease and colonization in BAL of immunocompromised patients with pneumonia.

[Current characteristics of tuberculosis and human immunodeficiency virus co-infection in a cohort of hospitalized patients in Medellín, Colombia].

INTRODUCTION: Tuberculosis (TB) is an important cause of morbidity and mortality in HIV patients. It is unknown if the advent of molecular diagnostic methods and a greater availability of antiretroviral therapy (ART) in our country have changed some characteristics of the TB/HIV co-infection. OBJECTIVE: To describe the epidemiology, clinical features, diagnosis, resistance patterns, tuberculosis drug effects and mortality in co-infected patients. MATERIALS AND METHODS: Retrospective study based on the review of medical records of hospitalized co-infected adults in a university hospital in Medellín, Colombia. RESULTS: A total of 178 patients was included in the study. TB and HIV diagnosis was simultaneous in 49.4%. In the moment of TB diagnosis, the median CD4 count was 61 cells/µL (27-145). Pulmonary tuberculosis (PTB) occurred in 28% of patients, extrapulmonary (EPTB) in 23%, and mixed TB in 48.9%. The main EPTB affectations were lymphatic (55.4%), gastrointestinal (35.9%), and of the central nervous system (18.7%). Ziehl-Neelsen stain was positive in 137 patients (77%), mycobacterium culture in 121 (68%), and TB-PCR, in 85 of those patients in whom the test was done. Rifampicin resistance was detected in six cases (4.9%). Transaminases (ALT) increased in half of the patients during TB treatment, but only 10% met liver-toxicity criteria. In-hospital mortality was 11.3%. The single risk factor associated with mortality was CD4 count <50/µL (RR=3.9; 95% CI: 1.36-11.37; p=0.01). CONCLUSIONS: When it occurs as an opportunistic infection, TB usually leads to the diagnosis of advanced HIV disease. If used appropriately, TB diagnosis in these patients can be done by conventional methods. It is always necessary to monitor liver function during TB treatment and to rule out drug resistance.

Current characteristics of tuberculosis and human immunodeficiency virus co-infection in a cohort of hospitalized patients in Medellín, Colombia / Características actuales de la coinfección con tuberculosis y el virus de la inmunodeficiencia humana en pacientes hospitalizados en Medellín, Colombia

Introduction: Tuberculosis (TB) is an important cause of morbidity and mortality in HIV patients. It is unknown if the advent of molecular diagnostic methods and a greater availability of antiretroviral therapy (ART) in our country have changed some characteristics of the TB/HIV co-infection. Objective: To describe the epidemiology, clinical features, diagnosis, resistance patterns, tuberculosis drug effects and mortality in co-infected patients. Materials and methods: Retrospective study based on the review of medical records of hospitalized co-infected adults in a university hospital in Medellín, Colombia. Results: A total of 178 patients was included in the study. TB and HIV diagnosis was simultaneous in 49.4%. In the moment of TB diagnosis, the median CD4 count was 61 cells/µL (27-145). Pulmonary tuberculosis (PTB) occurred in 28% of patients, extrapulmonary (EPTB) in 23%, and mixed TB in 48.9%. The main EPTB affectations were lymphatic (55.4%), gastrointestinal (35.9%), and of the central nervous system (18.7%). Ziehl-Neelsen stain was positive in 137 patients (77%), mycobacterium culture in 121 (68%), and TB-PCR, in 85 of those patients in whom the test was done. Rifampicin resistance was detected in six cases (4.9%). Transaminases (ALT) increased in half of the patients during TB treatment, but only 10% met liver-toxicity criteria. In-hospital mortality was 11.3%. The single risk factor associated with mortality was CD4 count <50/µL (RR=3.9; 95% CI: 1.36-11.37; p=0.01). Conclusions: When it occurs as an opportunistic infection, TB usually leads to the diagnosis of advanced HIV disease. If used appropriately, TB diagnosis in these patients can be done by conventional methods. It is always necessary to monitor liver function during TB treatment and to rule out drug resistance.

Introducción. La tuberculosis es una causa importante de morbilidad y mortalidad en pacientes positivos para el HIV. Los métodos de diagnóstico molecular y una mayor disponibilidad del tratamiento antirretroviral en el país pueden haber cambiado las características de la infección concomitante. Objetivo. Describir la epidemiología, las características clínicas, el diagnóstico, los patrones de resistencia, los efectos secundarios de los medicamentos antituberculosos y la mortalidad, en pacientes con las dos infecciones. Materiales y métodos. Se hizo un estudio retrospectivo basado en la revisión de historias clínicas de adultos hospitalizados en un hospital universitario de Medellín, Colombia. Resultados. Se incluyeron 178 pacientes en el estudio. El diagnóstico de tuberculosis e infección por el HIV fue simultáneo en 49,9 %. En el momento del diagnóstico, la mediana de CD4 fue de 61 células/ µL (rango de 27 a 145). La tuberculosis pulmonar ocurrió en 28 % de los pacientes, la extrapulmonar en 23% y la mixta en 48,9%. En la tuberculosis extrapulmonar, el compromiso fue principalmente linfático (55,4 %), gastrointestinal (35,9%) y del sistema nervioso central (18,7 %). La tinción de Ziehl-Neelsen fue positiva en 137 pacientes (77 %), en tanto que el cultivo para micobacterias lo fue en 121 (68 %). La reacción en cadena de la polimerasa para detectar la tuberculosis fue positiva en 85 de los pacientes a quienes se les hizo la prueba. Se detectó resistencia a la rifampicina en seis casos (4,9 %). Al iniciar el tratamiento antituberculoso, las transaminasas se elevaron en la mitad de los pacientes, pero solo 10 % cumplieron los criterios de hepatotoxicidad. La mortalidad hospitalaria fue de 11,3 %. El único factor de riesgo asociado con la mortalidad fue un conteo de CD4 menor de 50/µL (RR=3,9; IC95% 1,36-11,37; p=0,01). Conclusiones. Cuando la tuberculosis se presenta de manera oportunista, comúnmente lleva al diagnóstico de enfermedad avanzada por el HIV. Su diagnóstico en estos pacientes puede hacerse con los métodos convencionales. Es necesario vigilar la función hepática durante el tratamiento y excluir la resistencia a los medicamentos.

Evaluación de la concordancia entre dos métodos de lavado broncoalveolar para el diagnóstico microbiológico de la neumonía en pacientes con asistencia respiratoria mecánica / Concordance between two methods of bronchoalveolar lavage for the microbiological diagnosis of pneumonia in mechanically ventilated patients

Introducción. El diagnóstico microbiológico de la neumonía permite optimizar el uso de antibióticos en pacientes con asistencia respiratoria mecánica. Para ello se han cultivado cuantitativamente las muestras del lavado broncoalveolar broncoscópico, procedimiento que no siempre es posible. Objetivo. Evaluar la concordancia microbiológica entre muestras respiratorias tomadas porlavado broncoalveolar broncoscópico y no broncoscópico, y establecer si el uso previo de antibióticos y el momento de presentación de la neumonía pueden afectarla. Materiales y métodos. Estudio prospectivo realizado en el Hospital Universitario San Vicente de Paúl, en 38 pacientes con sospecha de neumonía y con asistencia respiratoria mecánica. En todos se practicó el lavado broncoalveolar por fibrobroncoscopia y el lavado nobroncoscópico usando un catéter telescopado de punta preformada (Balcatho). Todas las muestras fueron procesadas siguiendo protocolos microbiológicos convencionales. Resultados. Considerando el lavado broncoalveolar por fibrobroncoscopia como patrón dereferencia, los cultivos permitieron identificar el agente en 60,5 por ciento de los casos. El acuerdo diagnóstico se logró en 82 por ciento de los pacientes y 79 por ciento de los aislamientos. Utilizando el índice kappa de Cohen, la concordancia general entre los dos métodos fue 0,76 [0,60-0,93]; pero en quienes habían recibido antibióticos previos fue 0,26 [0,05-0,48], versus 1,0 en quienes no lo habían hecho (p<0,0001). La concordancia no difirió significativamente cuando se compararon los casos de neumonía temprana y tardía. Conclusiones. La concordancia general entre los dos métodos de lavado broncoalveolar es buena en pacientes con neumonía y respiración asistida mecánicamente. Sin embargo, el uso previo de antibióticos y no el momento de aparición de la neumonía, disminuye ésta significativamente.

[Concordance between two methods of bronchoalveolar lavage for the microbiological diagnosis of pneumonia in mechanically ventilated patients]. / Evaluación de la concordancia entre dos métodos de lavado broncoalveolar para el diagnóstico microbiológico de la neumonía en pacientes con asistencia respiratoria mecánica.

INTRODUCTION: Microbiological diagnosis of pneumonia allows the optimal use of antibiotics in mechanically ventilated patients. That is why samples of bronchoscopic bronchoalveolar lavage had been quantitatively cultivated, but this procedure is not always possible. OBJECTIVE: To evaluate the microbiological concordance between respiratory samples obtained by non-bronchoscopic protected bronchoalveolar lavage compared to the bronchoscopic ones, and to find out whether concordance was affected by previous use of antibiotics or the time of pneumonia onset. MATERIALS AND METHODS: Prospective study conducted at Hospital Universitario San Vicente de Paúl, in 38 patients with suspected pneumonia in mechanical ventilation. Bronchoalveolar lavage specimens were taken by two methods, the traditional one and non-bronchoscopic bronchoalveolar lavage, using a telescoping preformed tip catheter (Balcath). All samples were processed using conventional microbiologic protocols. RESULTS: Considering flexible bronchoscopy with bronchoalveolar lavage as the gold standard, cultures allowed the identification of at least one respiratory pathogen in 60.5% of cases. Diagnostic agreement was achieved in 82% of patients and 79% of microbiologic isolates. Using the Cohen's kappa coefficient, general concordance between both methods was 0.76 [0.60-0.93]; but in those who received previously antibiotics was 0.26 [0.05-0.48], versus 1.0 in those who did not (p<0.0001). Concordance did not differ significantly when cases of early or late pneumonia were compared. CONCLUSIONS: Concordance between non-bronchoscopic and bronchoscopic bronchoalveolar lavage is good in mechanically ventilated patients with pneumonia. However, the use of antibiotics previously, but not the time of pneumonia presentation, significantly decreases that concordance.

Diagnostic accuracy of bronchoalveolar lavage samples in immunosuppressed patients with suspected pneumonia: analysis of a protocol.

BACKGROUND: Fast and accurate etiologic diagnosis of pneumonia in immunocompromised patients is essential for a good outcome. Utility of bronchoalveolar lavage (BAL) samples has already been established, but studies about them are scarce and limited to few countries. We aimed to evaluate the accuracy of a diagnostic protocol, emphasizing on local epidemiology, rapidity, and yield of different techniques. METHODS: One year prospective study of 101 consecutive immunosuppressed patients admitted with suspected pneumonia to a university hospital. They all had bronchoscopic BAL (n=109) and respiratory sampling. Conventional microbiological studies, cytomegalovirus pp65 antigenemia and transbronchial biopsy (TBB), whenever considered pertinent, were done. Results were analyzed along with other diagnostic procedures, clinical course and final outcome. RESULTS: HIV/AIDS infection was the most frequent cause of inclusion (n=80). Infections accounted for 79 out of 122 final diagnoses (64.8%). Our protocol identified 60 infectious and 3 noninfectious pathologies (general yield: 51.6%). Sensitivity in pulmonary infections was 75.9% (IC95%: 64.8-84.6%), specificity 86.0% (72.6-93.7%), positive predictive value 89.6% (79.1-95.3%), negative predictive value 69.4% (56.2-80.1%), accuracy 79.8% (71.7-86.2%). Mycobacterium spp. (n=27), bacteria (n=19), Pneumocystis jirovecii (n=18) and other fungi (histoplasmosis: 6, aspergillosis: 5, cryptococosis: 3) were the most common infectious pathogens. Direct microscopy allowed an early definite/presumptive diagnosis in 36/49 fungal and mycobacterial infections (73.5%). Up to 30% of mycobacterial infections were missed. CONCLUSIONS: Systematical study of BAL samples has a high diagnostic yield in our immunocompromised patients with suspected pneumonia. As economical and epidemiological conditions of regions are different, it should be tried everywhere.