It is not true that the spectrum of organisms causing early onset sepsis
is greatly different from that causing late onset sepsis in the developing
world. Unlike developed countries, the spectrum is largely similar [1,2].
Barring the fact that meningitis is more commonly associated with late
rather than early onset sepsis, the more fulminant form of disease is in
fact early onset sepsis, mortality is higher in early onset sepsis.
Therefore, early onset sepsis is certainly not "more benign". In any case,
these contentions are of little relevance to the core clinical issue which
is that the current standard of care of probable sepsis does not
distinguish between early and late onset sepsis as far as the duration of
antibiotics is concerned. Our intervention (of shortening the duration)
therefore was pragmatic and common to both early and late onset sepsis.
It is often easy to be wise in hindsight and say that a
particular patient with probable sepsis was actually "probably not septic"
and needed no antibiotics. However, when viewed prospectively in a
real-life situation, most clinicians would treat a probable sepsis
(defined by us as persistence of clinical signs for at least 6 hours plus
positive CRP) empirically and then try to minimize the exposure to
unnecessary antibiotics, based on clinical course and culture results.
This is exactly what we attempted to do in this pilot trial.
References
1. Zaidi AK, Huskins WC, Thaver D, Bhutta ZA, Abbas Z,
Goldmann DA. Hsopital acquired neonatal infections in developing
countries. Lancet. 2005;365:1175-88.
2. NNPD network. National Neonatal Perinatal Database – report for the
year 2002-2003, New Delhi: 2005.
|