E-drug: Essential diagnostics (2)
---------------------------------------------
Given that whole books have been written about this topic, I thought
that I would just add a few thoughts in point form ....
How common is bacterial pneumonia?
In a study that is likely to never be repeated, Frank Shann et al in
the 1980's published in the Lancet an article from his work in PNG
Aetiology of pneumonia in children in Goroka Hospital, Papua New Guinea.
Shann F Gratten M, Germer S, Linnemann V, Hazlett D, Payne R
To determine the aetiology of pneumonia in 83 children admitted to
Goroka Hospital, Papua New Guinea, lung aspirates and blood were
cultured for bacteria. Haemophilus infuenzae, Streptococcus
pneumoniae, or both, were isolated from 43 (52%) of the children,
other bacteria from 8 (10%), and no bacteria from 32 (39%). Of the 32
strains of H influenzae tested, 18 (56%) were non-serotypable, 8
(25%) were serotypes other than type b, and only 6 (19%) were type b.
Viruses were isolated from lung or nasopharyngeal aspirates from 18
(29%) of the 62 children for whom viral cultures were done. It seems
that, although viruses may initiate infection, death from pneumonia
in children in developing countries is often due to H influenzae, S
pneumoniae, or both. Antibiotic therapy would prevent many of these
deaths. There is an urgent need for vaccines, effective in children
less than 6 months old, that protect against all strains of H
influenzae, and S pneumoniae.
PMID: 6147602, UI: 84294391 of 100s
It shows that, at least in the population presenting to hospital,
bacteria were isolated in over 60 of children.
Based on this work, and other studies, simple clinical predictors of
pneumonia, with relatively high sensitivity and specificity and,
importantly, able to be used by a health worker who have no access to
laboratory or x-ray facilities, were developed. These clinical
indicators include cough, rapid respiratory rate and chest indrawing.
CRP
Sensitivity and predictive power.
Based on the authors evidence alone, I question the sensitivity of
CRP as a diagnostic tool for acute pneumonia. Given, as the author
states, only 2 of the 120 children classified / diagnosed with
pneumonia had a positive CRP, it seems that this test adds little to
the health worker arriving at the diagnosis of pneumonia. It would be
interesting to know how these children presented clinically
(breathing rate etc), and whether they did indeed have x-ray
findings. If their pneumonia was in fact obvious clinically or on
X-ray anyway, the CRP does not add predictive power for the diagnosis.
Even if only 20% of the 120 children classified / diagnosed with
pneumonia did indeed have a true bacterial illness requiring
antibiotics, the sensitivity of the CRP is 2/24 approx 8%. Is this
a cost effective test in developing countries??? If the percentage
is closer to 60%, as in the PNG study (admittedly hospital based),
the sensitivity is even lower.
What would be a real indication of the sensitivity of CRP, would be
to determine the percentage of those with a positive CRP amongst
those children with truly proven bacterial positive pneumonia on lung
aspirate. This is unlikely ever to be done.
IMCI
IMCI is largely intended to be used by health workers operating in a
setting where they do not have access to laboratory facilities.
Health workers using IMCI do not "diagnose" pneumonia. Pneumonia is a
"classification" based on the finding by the health worker on a
child's history (cough or difficulty breathing), and on examination
an (elevated respiratory rate and or chest indrawing).
These symptoms and signs offer what is considered to be the highest
sensitivity and specificity for a health worker to be able to
classify and treat pneumonia in the primary health care setting.
It is recognised that these symptoms could indeed by due to viral
pneumonia, or even due to asthma, or bronchiolitis. However, the
primary health care worker may not have the clinical skill or
diagnostic facilities to come to reach these diagnoses.
The IMCI algorithm provides the health worker with a tool to detect a
high percentage of children with pneumonia, a high proportion of
which will be bacterial, and treat them.
Drugs
There are two basic questions
i) to give drugs or not to give
ii) if one gives drugs, which ones does one give
The answer to the first question, to give or not to give, as the
author implies, lies in the sensitivity and specificity of the
diagnostic tools. These tools differ depending on the clinical
setting, and will be different for a village worker when compared to
the hospital situation where X-ray and CRP may be available.
The answer to the second question, what drugs to give if the decision
has been made to give, will depend on local epidemiological data.
Just some thoughts,
Cheers,
Mick
Dr Mick Creati
Pediatrician
International Health Unit, Macfarlane Burnet Centre for Medical Research
P O Box 254 Fairfield Vic Australia 3078
Telephone 613 9282 2275
Fax 613 9482 3123
Time zone: 11 hours ahead of GMT.
email <creati@burnet.edu.au>
--
Send mail for the `E-Drug' conference to `e-drug@usa.healthnet.org'.
Mail administrative requests to `majordomo@usa.healthnet.org'.
For additional assistance, send mail to: `owner-e-drug@usa.healthnet.org'.