Unidade de Cuidados Intensivos Neonatais. Hospital de Dona Estefânia. Centro Hospitalar de Lisboa Central
- XI World Congress of Perinatal Medicine 2013, 19-22/6/2013, Moscovo
Introduction: World Health Organization estimates that 10% of all under-five mortality are due to neonatal sepsis. From them 42% occur in the first week of life. Sepsis is estimated to contribute for 26% of perinatal deaths worldwide. In high-income countries early-onset sepsis vary between 0.77 to 2.7/1000 live births while in low-income countries incidence is estimated, is based on positive blood cultures and vary widely from 6.5/1000 live births to 21/1000 LB.
Difficulties: There are important difficulties in comparing data. Early-onset sepsis definitions and criteria for epidemiological studies, may vary among the first 48h, the firts 72h, the first 6 days, the first week; with positive blood culture, excluding coagulase negative Staphylococcus, by 1000 live births or by 100 NICU admissions. Also there are many difficulties in getting data in low-income countries. All neonatal period may be included, as well as inborn, out born and those coming from the community. On the other way often there is under-reporting by Demographic and Health Surveys; when birth and death take place at home death in the first few days may be not recognized as a death of a live born; septic newborns may died before they get health care; inadequate laboratory resources are common.
Portuguese data: In Portugal, on the National Registry of Infection in NICUs, started on January the first 2008, among 25 714 registrations 2 690 (10.5%) were reported as having mather-related infection. The incidence was 11.8% for VLBW infants and 16.2% for ELBW.
It is very important to understand that early-onset infection is mainly due to birth canal bacteria because we have to choose the most adequate antibiotics. Moreover bacteria and their sensitivity may change over time and also from country to country.
With the policy of screening and prophylaxis to group B Streptococcus the frequency of its infection decreased. Eventhough in the Portuguese registry it is the most frequent isolate (31.3%) followed by E. Coli (26.9%). These data are similar to those find by others. In low-income countries the incidence vary among authors probably because differences in screening, population and time. However the biggest difference between low and high income countries is that, in those, the most prevalent bacteria continue to be E. Coli, with Gram negative responsible for 97,8% of mother-related infections. Mortality rates depend on gestational age and causative microorganisms.
Concerning prevention while something has been done to prevent GBS infection nothing has been done against E.coli. A prophylaxis with a regimen that supress neonatal colonization by both ampicillin-resistant E.coli and GBS might reduce rates of EOS. Eventhough screening and prophylaxis against GBS have room to be improved.
Conclusions: Screening and prophylaxis partly failed to eradicate EOS; the same policy as to GBS has to be found to E.coli; GBS carrier state should not be regarded as the only main risk for early-onset neonatal infection. Even with negative GBS screening “the” other risk factors such as maternal fever with laboratory signs of infection deserve attention and treatment
Main remarks: A newborn infant at risk of infection should be treated, preferable with ampicillin and gentamicin and therapeutics stopped on the evidence of non-infection ; Septic newborn infants must be treated; Those infants born to a mother with amnionitis must continue in utero treatment after being born.
Low-income countries have to be helped to improve perinatal care and safe procedures in order to reduce the high levels of multi-resistance; There are no accurate prescriptions for empiric treatment; it should be based on usual bacteria and its sensitivity; Early treatment should not be denied but have to be stopped as soon as no infection is proved.
Key-words: Low and high income countries, mother-related infections, pregnancy, prophylaxis, screening,