[Predictive factors for invasive pnuemococcal disease: a case-control study].

BACKGROUND: Streptococcus pneumoniae causes significant morbidity in children, but data on the incidence of invasive pneumococcal disease in Spain are scarce. The objectives of this study were: 1) to describe the clinical and epidemiological features of invasive pneumococcal disease in our health district and 2) to determine factors predictive of invasive pneumococcal disease in febrile children seen at a hospital Emergency Department.
MATERIAL AND METHODS: Design. Observational, retrospective, case-control study, from 1 October, 1992 to 31 March, 2001. LOCATION: Community Hospital in the north of Spain. Entry criteria for cases: febrile children under 14 years of age, seen at the Emergency Department during the study period, with growth of S. pneumoniae in the blood culture. Eligibility criteria for controls: febrile children under the age of 14 years seen at the Emergency Department during the study period with no bacterial growth in the blood culture. The first eligible child seen after each case was included as a control. STATISTICAL ANALYSIS: descriptive analysis of patients with invasive pneumococcal disease and univariate analysis of each variable in relation to the dependent variable (blood culture positive for S. pneumoniae); multivariate analysis was performed using logistic regression techniques.
RESULTS: Seventy-six cases of invasive pneumococcal disease were studied. The mean incidence of invasive pneumococcal disease (cases/100,000 children/year) was 174.1 for children under 24 months of age, 38.9 for children aged 24-59 months, and 5.9 for children older than 59 months. The incidence of pneumococcal meningitis in children under 24 months of age was 14.8 cases/100,000 children/ year. The most common diagnoses were occult bacteremia (64.5 %), pneumonia (17.1 %), and meningitis (9.2 %). Mortality was 1.3 %. A total of 56.5 % of the S. pneumoniae strains showed penicillin resistance (11.8 % high-grade resistance) and 12.2 % showed cefotaxime resistance. Predictive factors for invasive pneumococcal disease were temperature greater than or equal to 39 degrees C (OR: 2.09; 95 % CI:91-4.79), generalized malaise (OR: 2.61; 95 % CI: 1.1-6.21), age between 6 and 36 months (OR: 4.06; 95 % CI: 1.79-9.21), and absolute neutrophil count (ANC) greater than or equal to 10,000 cells/mm3 (OR: 8.16; 95 % CI: 3.54-18.79).
CONCLUSIONS: 1. The incidence of invasive pneumococcal disease in our health district is high and is greater than that reported for other European regions. 2. In contrast, the incidence of pneumococcal meningitis is similar to that in other European countries. 3. The most frequent diagnosis was occult bacteremia. 4. In the case-control study, four variables showed significant independent association with the risk of invasive pneumococcal disease: temperature greater than or equal to 39 degrees C, general malaise, age between 6 and 36 months, and an ANC greater than or equal to 10,000 cells/mm3. The most powerful predictor of invasive pneumococcal disease in our series was ANC.