Epidemiology of Early-onset Bacterial Neonatal Infections in Madagascar.

BACKGROUND: Little is known about early-onset neonatal bacterial infections (EONBI) in Madagascar. Our aim was to determine their epidemiology to improve their management. METHODS: Inborn neonates at risk for EONBI and admitted in the neonatal unit of 2 tertiary hospitals in Antananarivo, Madagascar, were included in a prospective study from April 2012 to March 2013. Using a clinical algorithm, blood culture, gastric fluid culture and C-reactive protein dosage were performed in newborns at high risk of infection, that is, peri partum fever, prematurity <35 weeks' gestation or birth weight <2000 g, or presenting with clinical signs of infection. EONBI was defined as a bacteremia occurring within the first week of life. RESULTS: Among 307 neonates, 75 (24.4%) had an EONBI caused by 1 (n = 59) or 2 (n = 16) bacteria (91 isolates). Gram-negative bacteria were predominant (n = 62, 82.7%), including Enterobacter cloacae (n = 26), Klebsiella pneumoniae (n = 14), Escherichia coli (n = 7) and Proteus mirabilis (n = 2). Group B Streptococcus, Acinetobacter baumanii and Enterococcus sp. represented 3.6%, 8.2% and 12.1% of the isolates, respectively. All E. cloacae and 12/14 (85.7%) K. pneumoniae were extended-spectrum ß-lactamase producers. At all, 41/91 (45.1%) bacteria were multidrug-resistant (MDR) and 34/75 (45.3%) newborns had an EONBI caused by an MDR bacteria. Neonatal asphyxia was the only factor associated with multidrug resistance (odds ratio: 4.52; CI: 1.20-16.94; P = 0.025). The EONBI-related mortality (n = 20/75, 26.7%) rose up to 38.2% (n = 13/34) in case of MDR bacteria. CONCLUSIONS: The epidemiology of EONBIs in Madagascar is comparable to that found in many low-income countries. Prevention, including improvement of hygiene during resuscitation for neonatal asphyxia, is likely to be more effective in reducing EONBI-related morbidity and mortality than using new antibiotics to counter resistance.

Prevalence of Mycoplasma pneumoniae Infection in Malagasy Children.

BACKGROUND: Childhood community-acquired pneumonia is a leading cause of childhood morbidity in low-income countries. The etiologic agents are usually Staphylococcus aureus, Streptococcus pneumoniae and Mycoplasma pneumoniae. M. pneumoniae was recognized as a cofactor in asthmatic disease. High asthma prevalence was reported in Madagascar. Our aim was to clarify the prevalence of M. pneumoniae infection in this country and its relationship with asthma. METHODS: A prospective study was conducted in 351 children (from 2 to 16 years of age) from January 2012 to December 2014. According to the clinical symptoms, children were enrolled in 3 groups: "control group" (CG, n = 106), "asthma group" (n = 129) and "pneumonia group" (n = 116). The IgG and IgM M. pneumoniae status was evaluated by an enzyme-linked immunosorbent assay. Clinical signs of infection, socioeconomic data and antimicrobial treatment were recorded. RESULTS: The overall prevalence of M. pneumoniae infection was 18.2%. The multivariate analysis demonstrated that M. pneumoniae infection was significantly more frequent in the CG [pneumonia group vs. CG: odds ratio = 0.45 (0.21-0.91), P = 0.037 and asthma group vs. CG: odds ratio = 0.39 (0.18-0.87), P = 0.021]. The C-reactive protein value was significantly higher in children with M. pneumonia-positive serology (85 vs. 61 mg/L, P = 0.03). Of note, 99 (41%) children received antibiotics before attending. CONCLUSIONS: We report a prevalence of 18.2% for M. pneumoniae infection in children in Madagascar. The prevalence of M. pneumoniae infection was higher in the control patients than in asthmatic ones.