Life threatening infections labelled swine fluBMJ 2010; 340 doi: https://doi.org/10.1136/bmj.c137 (Published 13 January 2010) Cite this as: BMJ 2010;340:c137
All rapid responses
The ongoing pandemic of novel swine-origin influenza A virus (S-OIV)
creates significant impact on health care services. Recent studies could
demonstrate a substantial effect on intensive care units in Australia and
Zealand . To determine the duration of antiviral treatment and
infectious patients, monitoring the infectivity is a key issue.
Obviously virus isolation should be the ultimate answer, but even for
patients it remains time and resource consuming, furthermore adequate
laboratory capacities are not broadly available. Since rapid antigen
tests lack sensitivity , tests using nucleotide amplification
polymerase chain reaction (PCR), are the cornerstone of up to date
For patients with life threatening pulmonary infections rapid conformation
diagnosis is essential.
In an outpatient setting the issue of the optimal sample site has been
investigated extensively . In summary nasal swabs are better tolerated
obtain nearly equal results for sensitivity and specificity, when compared
nasopharyngeal sampling. For patients with severe respiratory failure,
requiring mechanical ventilation such data does not exist.
To date, 11 patients with severe respiratory failure due to an
S-OIV were admitted to our department, four of them requiring extra
corporal membrane oxygenation. Initially results of standard diagnostic
were inconsistent. By means of PCR S-OIV could be detected in
nasopharyngeal samples in 1 of 3 patients on admission, whereas PCR
positive results in all 3 patients when samples were obtained by
bronchoalveolar lavage (BAL). Under antiviral treatment with the
neuraminidase inhibitor Oseltamivir, PCR turned negative in all patients
nasopharyngeal samples were taken, but still remained positive in BAL
For infection control in an intensive care unit these findings might be
important. In contrast to seasonal influenza A H1N1 (sH1N1), which
preferentially in epithelial cells of the upper respiratory tract  and
can be monitored by nasopharyngeal swaps,
S-OIV seems to affect also lower respiratory epithelium.
In part the different receptor specificities of S-OIV compared to sH1N1
may be an explanation. Receptor specificity of influenza viruses is
by the viral glycoprotein hemagglutinin, which binds to receptors with
terminal sialic acids.
Similar differences have been observed with avian influenza A H5N1 virus
(aH5N1). aH5N1 preferentially binds to receptors with alpha 2-3 linked
acids, whereas sH1N1 has a higher affinity to receptors with terminal
2-6 linked sialic acids , which are predominantly distributed on
cell of the upper respiratory tract.
For patients with acute respiratory distress syndrome BAL-samples should
considered preferentially for PCR diagnostic, because nasopharyngeal
samples may not be sufficient to guide the decision making process
especially in S-OIV infections.
1 Moreno RP, Rhodes A, Chiche JD . The ongoing H1N1 flu pandemic and
intensive care community: challenges, opportunities, and the duties of
scientific societies and intensivists. Intensive Care Med 2009;35(12):2005
2 Chan KH, Lam SY, Puthavathana P, Nguyen TD, Long HT, Pang CM, Chan
KM, Cheung CY, Seto WH, Peiris JS. Comparative analytical sensitivities of
rapid influenza A antigen detection test kits for detection of influenza A
subtypes H1N1, H3N2 and H5N1.
J Clin Virol 2008;38(2):169-71.
3 Sung RY, Chan PK, Choi KC, Yeung AC, Li AM, Tang JW, Ip M, Tsen T,
EA. Comparative study of nasopharyngeal aspirate and nasal swab specimens
for diagnosis of acute viral respiratory infection. J Clin Microbiol
4 Stevens J, Blixt O, Glaser L, Taubenberger JK, Palese P, Paulson
IA. Glycan microarray analysis of the hemagglutinins from modern and
pandemic influenza viruses reveals different receptor specificities. J Mol
5 Li IW, Chan KH, To KW, Wong SS, Ho PL, Lau SK, Woo PC, Tsoi HW,
Cheng VC, Zheng BJ, Chen H, Yuen KY. Differential susceptibility of
cell lines to swine-origin influenza A H1N1, seasonal human influenza A
H1N1, and avian influenza A H5N1 viruses. J Clin Virol 2009;46(4):325-30.
Competing interests: No competing interests
The Royal College of General Practitioners (RCGP) has been working
with NHS Direct (NHSD) during the current swine flu (influenza A/H1N1)
pandemic, by helping to quality assure the National Pandemic Flu Service
(NPFS). NPFS was not designed as a diagnostic service, but as a facility
provide prompt assessment, advice and early treatment where indicated, and
to relieve pressure on GP and hospital services.
NPFS uses algorithm and associated protocols which were developed
advice from a very wide range of clinical and remote assessment services.
These are built around available national guidelines on the assessment and
management of feverish and respiratory symptoms in adults and children.
They include the identification of ‘red flag’ symptoms indicating a need
urgent 999 referral or GP assessment.
Over the past few months, NPFS has dealt with over two million
which aim to confirm whether a patient has influenza-like symptoms and
could thereby be authorised to receive antiviral medicine. This has led to
collection of over a million courses of antiviral medication, many being
collected by a “flu friend”. NPFS also gives good practice advice
symptom control and case isolation, as well as emphasising the need to
further health advice or prompt medical assessment in the case of
illness or other concerns.
NPFS has facilitated the early treatment of flu-like symptoms. Aiming
reduce complications and hamper illness transmission in the community
through medication and good practice advice, data on the part that
medications have had to play in this is encouraging. (1)
Houlihan and colleagues (2) draw attention to potential risks of
dealing with a
flu pandemic with a national service, citing six cases in Newcastle when
diagnosis and management were delayed by an average of three days
following an initial, incorrect diagnosis of swine flu, during a time when
swab positive H1N1 prevalence was 6.5% compared with a national
prevalence of 11.8%. They state that this shows that a concise history
cover travel, immunosuppression and drug exposure comprehensively and
that current local epidemiological data should influence the
and application of algorithms.
Their article also emphasises that algorithms do not replace
history taking, clinical acumen, and laboratory support in establishing
accurate diagnoses. While this view is probably shared by most doctors
engaged in day-to-day clinical practice, NPFS algorithms and protocols are
unable to take account of local epidemiological data for a number of
including the very nature of a pandemic illness with changing prevalence.
While any delay in correct diagnosis and treatments is regrettable, such
delays may be attributable to various factors. The Department of Health
and NPFS emphasise that NPFS users and patients should seek advice from
their doctor if symptoms worsen.
In keeping with other “triage” assessment systems, the NPFS is unable
the acumen of a medical consultation. The sequence of questions in the
algorithm is designed to first identify possible symptoms of severe or
threatening conditions, where service users are advised to call for
services assistance if indicated. Approximately 5% of assessments have
outcome. It then seeks to confirm if there are symptoms defining a flu-
illness, and gives advice and authorises antivirals accordingly. Users are
reminded of the purpose of the service and advised to seek advice from the
patient’s GP or NHS Direct as appropriate. Such referrals or “hand-offs”
include seeking the immediate or urgent advice of the patient’s own GP in
significant proportion of assessments.
Call agents mostly have no healthcare background, but have been a
part of the NPFS, and deserve recognition. They work to carefully scripted
algorithms and protocols. Agents, algorithms and protocols are very
monitored and quality assured by NPFS managers and NHS direct, with
support available at all call centres. The DH, Health Protection Agency,
agencies, and GP liaisons for the contact centres are all involved with
quality assurance of the NPFS. Algorithms and protocols have been updated
to “fine-tune” assessments, including the addition of advice about
Questions about immunosuppression and serious or chronic medical
conditions have been included from the start.
This letter is intended to give an outline of the strategy and impact
NPFS. As the number of assessments now appears to be declining, the
of a rapid triage system may diminish. Although the risks of the current
pandemic continue, with the A/H1N1 vaccination programme under way, it
now seems that we may soon be able to focus on preparation for future
pandemics. This will continue to be enhanced by sharing the experiences of
the current pandemic and learning from them.
John Gration GP Gloucestershire GL6 RCGP Pandemic Flu Liaison GP
Maureen Baker Pandemic Flu Lead RCGP, 14 Princes Gate, Hyde Park, London
SW7 1PU email@example.com
Cecily Cook Clinical Assurance Director NPFS, Regional Director of Nursing
(South West) NHS Direct, 8th Floor, Riverside House, 2a Southwark Bridge
Road, London SE1 9HA
Kate Adams GP Hackney, London N1 RCGP Pandemic Flu Liaison GP
Competing Interests: The RCGP has costs relating to NPFS liaison re-
by the DH. NHSD is commissioned by the DH to provide the NPFS.
Weber JT, Nicoll A, Bridges CB, Clancio BC. Neuraminidase inhibitors
pandemic A/H1N1 flu. BMJ 2010;340:c130
Houlihan CF, Patel S, Price DA, Valappil M, Schwab U. Life threatening
infections labelled swine flu. BMJ 2010;340:c137
The RCGP received funds from the
Department of Health to quality
assure the NPFS telephone triage
Competing interests: No competing interests