28
5.5 Empirical antibiotics for suspected neonatal sepsis
a Neonates with signs of sepsis should be treated with ampicillin or penicillin and gentamicin as the irst line antibiotic treatment for at least 10 days.
Strong recommendation, low quality of evidence
b If a neonate with sepsis is at greater risk of staphylococcus infection e.g. extensive skin pustules, abscess, or omphalitis in addition to signs of sepsis, should be given
cloxacillin and gentamicin instead of penicillin and gentamicin.
Strong recommendation, expert opinion
c Where possible, blood cultures should be obtained before starting antibiotics. If an infant does not improve in 2–3 days, antibiotic treatment should be changed,
or the infant should be referred to a neonatologist for further management.
Strong recommendation, expert opinion
In making this recommendation, the panel considered the high antimicrobial sensitivity to third generation cephalosporins, and the lack of enough evidence to
choose between beta-lactams plus gentamicin and 3rd generation cephalosporins. In developing countries, streptococci continue to be highly sensitive to beta-
lactams, although sensitivity to gentamicin of staphylococcus aureus 40–95 and gram negative enteric bacteria 32–100 is variable. In order to guide empirical
prescribing, it is crucial to monitor changes in the pattern of causative organisms and their antimicrobial susceptibility proiles since the efects may depend on antibiotic
susceptibility patterns.
5.5.1 Evidence and summary of findings
he evidence for these recommendations are from two existing Cochrane reviews that have concluded there is not enough evidence to choose between diferent
antibiotic combinations for irst line treatment of early or late neonatal sepsis [Mtitima, 2004; Gordon, 2005]. Sepsis in neonates is classiied as early presenting
within 72 hours ater birth and late presenting ater 72 hours ater birth because this predicts aetiology. In early sepsis the most common organisms are Group B
streptococcus and gram negative enteric bacteria esp. E. coli. Late onset community acquired infections are predominantly caused by Strep. Pneumoniae, Staphylococcus
aureus, E. coli and other gram negatives.
he evidence is from four RCTs of 3rd generation cephalosporin monotherapy versus penicillin and aminoglycoside combination that were identiied. One of
these was conducted in South Africa and all others were conducted in Europe. Two trials used cefotaxime and the other two used cetazidime monotherapy as the
intervention. hree trials used ampicillin or benzyl penicillin with gentamicin as the control treatment, while one trial used penicillin with netilmicin as the control
treatment. he overall quality of evidence was graded as low
see GRADE table A7.4 .
Mortality and cure rates were similar between intervention and control groups in three of the four studies. However, there was a signiicantly higher cure rate in
the South African study. Overall, there was no diference in mortality or cure rates
29
between the intervention and control groups. One study reported that a combination of cetazidime plus ampicillin had higher cure rates than the ampicillin-gentamicin
combination.
5.5.2 Benefits and risks
Benefits Available evidence indicates no diference in cure rates and mortality between 3rd
generation cephalosporin monotherapy and penicillin-aminoglycoside combination therapy. However, efects may vary in diferent settings depending on antibiotic
susceptibility patterns. Other beneits of 3rd generation cephalosporins are good cerebrospinal luid penetration and convenient oncetwice daily administration.
Risks he risks of cetriaxone include displacement of bilirubin from albumin leading
to higher free bilirubin and risk of kernicterus and promotion of beta-lactamase resistance.
5.5.3 Acceptability and feasibility
Neonates with suspected sepsis will require admission irrespective of the antibiotic chosen. herefore diferences in the convenience of administration are less
important. Median price of ampicillin 250 mg is US 0.18vial, gentamicin 10mg ml is US 0.07ml, and cetriaxone 250 mg is US 0.44vial. Cefotaxime is more
expensive than cetriaxone.
5.6 Head or whole body cooling in management of hypoxic ischaemic
encephalopathy
Head or whole body cooling should not be done outside well-resourced, tertiary neonatal intensive care units, because there is potential for harm from this therapy
in low-resource settings.
Strong recommendation, moderate quality evidence
he panel made this recommendation on the basis of potential harm from body cooling if applied outside a neonatal intensive care unit where mechanical
ventilation, sedation and other supportive care cannot safely be provided. It noted that in low resource settings there was no evidence supporting this therapy and that
it was expensive. However, in well resourced tertiary neonatal intensive care units, treatment with whole body cooling as part of high quality intensive care support
improves survival and reduces disability for term neonates with asphyxia and hypoxic ischaemic encephalopy.
5.6.1 Evidence and summary of findings
here is moderate quality evidence that induced hypothermia cooling of newborn babies who may have sufered from a lack of oxygen at birth reduces death or
disability, without increasing disability in survivors. he damage caused by the lack