key: cord-019977-kj0eaw6v
authors: nan
title: Neonatal bacterial infection: A changing scene?
date: 2005-04-14
journal: J Infect
DOI: 10.1016/s0163-4453(82)91569-9
sha: 
doc_id: 19977
cord_uid: kj0eaw6v

nan

The foetus and newborn infant have a particular vulnerability to infection of all kinds, yet most escape this hazard. Immune mechanisms are developing at each stage of existence which are appropriate to the likely challenges at such times; 1 the unusual challenge however may not be met, for the immune and inflammatory responses produced by healthy older children and adults cannot be mounted as effectively, particularly if the infant is born pre-term. 2 The risk therefore is ever present, and-at least where neonatal bacterial infection is concerned -its nature may be ever changing. The reasons for such change are sometimes obvious, sometimes obscure, and have to be sought in the maternal and infant hosts, in improved laboratory techniques, and in altering clinical practices which may impose change on host and infecting organism alike.

Records have been kept over a 5o-year period (I928-I978) at Yale of the isolates recovered from the bloodstream of infants in the first weeks of lifef1-6 and accord well with those reported intermittently from other centres. In addition to showing changes in the infecting organisms, they also reveal the growing importance of maternally transmitted intrapartum (early) infections over later infections in recent years, many of the latter being environmentally acquired. In the first quarter of the 5o-year period, two-thirds of the isolates were Gram-positive, the majority staphylococci (mostly Staphylococcus aureus) and beta-haemolytic streptococci. The introduction of Lancefield typing in the early I93OS showed these to be mainly Group A streptococci. Over the next 20 years the position was to be gradually reversed and between I958 and I9655 two-thirds of Yale's newborn bacterial isolates were Gram-negative. This time period coincided with the entry of paediatricians to newborn nurseries on a much larger scale than hitherto, with an increase in the use of antimicrobial drugs, and with the introduction of apparatus such as incubators, resuscitation and suction units, the humidification parts of which often harboured Gramnegative organisms, all capable of causing lethal disease in the infant. The last period (I966-I978) 6 has shown a decrease in Gram-negative isolates, and a rise in Gram-positive ones, the latter largely due to greatly increased isolation of Group B streptococcus. During the half century under review, mortality from neonatal bacteraemia fell from 9o per cent in the period I928-I933, ~ to z6 per cent in I966-I978 ;6 and the proportion of isolates recorded as recovered in the first 48 hours of life (early infections) increased from IO per cent to 57 per cent of the total respectively. We can only guess whether this last change was due to a growing awareness on the part of paediatricians of the possibility of maternally transmitted illness, or to a genuine increase of such infection. As the number of blood cultures drawn increased considerably over the years, the former may be more likely. On the other hand an increasing vogue for surgical induction of labour, and the use of pressure transducers for foetal monitoringfl if practised there may have introduced the organism into the amniotic fluid more frequently in recent years. A report from Hammersmith Hospital 8 suggests a further change may be under way in newborn nurseries, as the IO6 Editorial predominant isolate from cases of neonatal bacteraemia there since 1976 has been Staphylococcus epidermidis. This organism has ousted the Group B streptococcus from its prominent place in early infections at the hospital in previous yearsfl and has also displaced Gram-negative bacilli as the most important cause of later infection.

Elsewhere in this issue (p. II7) Oto discusses some of the reasons-the increasing preoccupation of neonatal intensive care units with infants of very low birth weight and the increasing complexity of their care, involving much more intravenous therapy -which may be responsible for this change. Twenty years ago the great majority of such infants would have died within 24 hours of birth, and it is likely that intrapartum infection, even as a contributory cause of their death, may have gone unsuspected and undiagnosed. Today they live, and their immaturity makes them at risk throughout the many weeks of their hospital stay. Oto also records the major infections found in such a unit over a six month period, and bacteraemia is far the commonest. Much less often seen, but next in importance, is necrotising enterocolitis. This is a condition which is known to have occurred for over one hundred years, but which reached almost epidemic proportions in certain newborn intensive care units in the late 196os and I97OS. 1° Bacteria are responsible for the progression and complications of the illness, but whether they have a primary role, or merely invade ischaemic bowel wall which has been damaged by a variety of other factors, is still unsure. The condition has certainly accompanied genuine bowel pathogens (bacterial n and viral12), other cases often occur in clusters, TM may be curtailed by infection control measures, lz and Clostridium spp. have occasionally been implicated. 14, 15 Some believe it to be part of a spectrum of disease which includes pseudomembranous or antibiotic-associated colitis. 16 Clostridium difficile and its toxin may be present in the stools of many well newborn babies, but this organism has not been implicated in the genesis of neonatal necrotising enterocolitis as it has been with pseudomembranous colitis. 17 Meningitis is now fortunately rare (o.26/iooo live births in a geographically defined population, excluding that associated with neural tube defects18), but there is still controversy about treatment of the most commonly occurring Gram-negative enteric infections, despite controlled trials. Omphalitis, the bacteriology of which is described by Brook in this issue (p. I27), is now very infrequently seen. Many newborn nurseries use some form of antiseptic cord care; and a daily antibiotic spray containing polymyxin, bacitracin and neomycin, prevents bacterial colonisation, 19 and hence infection, at this site in the great majority of babies. Improved laboratory techniques for the recognition of organisms such as Clamydia trachomatis are leading to a gradually increasing awareness of the extent of its involvement in neonatal conjunctivitis and respiratory illness, s° Bacterial infections can be permanently damaging to developing and rapidly growing organs; and constant surveillance of the patterns of infection is nowhere more important than in newborn care. Preventive measures such as handwashing and disinfection of apparatus have to be carried out to a uniformly high standard in intensive care units. This is especially important if outbreaks of infection are to be prevented among the immature residents, for the inevitable overuse of antimicrobial therapy in such units leads to a disturbance of bacterial flora which may make the hosts even more unusually vulnerable.

The development of the immune response. Characterization of the response of the human infant and adult to immunization with Salmonella vaccines

Host defences in the Human Neonate

Septicemia in the new-born. Am`7 Dis Child

Septicemia of the newborn

Septicemia of the newborn

A half century of neonatal sepsis at Yale. 1928 to I978

Nasal colonization of infants with group B Streptococcus associated with intrauterine pressure transducers

Changing blood culture isolates in a referral neonatal intensive care unit

Early neonatal bacteraemia. Comparison of group B streptococcal, other Gram-positive and Gram-negative infections

Neonatal necrotizing enterocolitis. Monographs in Neonatology. New York: Grune and Stratton

Gastroenteritis with necrotizing enterocolitis in premature babies

Association of coronavirus infection with neonatal necrotizing enterocolitis

Clustering of necrotizing enterocolitis. Interruption by infection-control measures

Outbreak of necrotising enterocolitis caused by Clostridium butyricum

Fulminant necrotising enterocolitis associated with Clostridia

Pseudomembranous enterocolitis (antibiotic-related colitis)

Clostridium difficile and the aetiology of pseudomembranous colitis

Acute bacterial meningitis in childhood. Incidence and mortality in a defined population

An investigation into the aerobic and anaerobic bacterial flora of normal and ill~low birthweight newborn babies

Prospective study of chlamydial infection in neonates