ARTICLE IN PRESS

Available at www.sciencedirect.com

Travel Medicine and Infectious Disease (2008) 6, 190–194

1477-8939/$ - see frdoi:10.1016/j.tmaid

�Corresponding au

E-mail address: j

journal homepage: www.elsevierhealth.com/journals/tmid

COMMENTARY

Responding to bird influenza: Past, present and future

Joyshri Sarangi�, Chloe Sellwood, Jonathan S. Nguyen-Van-Tam

Pandemic Influenza Office, Health Protection Agency Centre for Infections, 61 Colindale Avenue,London NW9 5EQ, England, UKAvailable online 23 May 2008

Introduction

In contrast to influenza B, which is mainly responsible foroutbreaks among school children or in nursing homes, andinfluenza C, one of the many viruses responsible for thecommon cold, influenza A has the capacity to causepandemics. This is associated with a change in thehaemagglutinin (H) antigen on the virus surface, with orwithout an associated change in the neuraminidase (N)antigen. While most avian influenza strains (influenza A)circulate naturally in wild waterfowl causing few or nosymptoms, highly pathogenic variants (HPAI) are capable ofcausing high mortality in domestic poultry.1,2 Global concernover the public health impact of avian influenza is based ontheir ability to cross the species barrier into humans, theresulting clinical infection, and their potential to cause thenext influenza pandemic.3,4

This paper provides a commentary on the human healththreat posed by influenza A in terms of its pandemicpotential, based on previous history, current concernsrelated to recent avian influenza, and preparations for afuture influenza pandemic.

The pandemic potential of influenza A based onpast history

Influenza A has been responsible for at least five humanpandemics, the most well documented being those of the20th century, in 1918/19, 1957/58 and 1968/69. These areknown to have been due to influenza A subtypes H1N1,5,6

H2N27 and H3N2,8 respectively. Retrospective serological

ont matter & 2008 Elsevier Ltd. All rights reserved..2008.03.002

thor.

oyshri.sarangi@hpa.org.uk (J. Sarangi).

analysis indicates that A/H2N2 was probably responsible fora pandemic in 1889, and that a mild pandemic in 1900 mayhave been caused by A/H3N8.9,10 Additionally, influenzaA/H1N1 re-emerged in 1977, but instead of replacing H3N2is currently co-circulating with it; a recognisable pandemicdid not ensue on this occasion. At least two of the 20thcentury pandemics (in 1918 and 1957) were thereforeassociated with the re-emergence of viruses similar to thosewhich had circulated previously, a process known asantigenic recycling. According to this theory, it could beconcluded that it is possible that the next pandemic will bederived from a H2 virus rather than HPAI A/H5N1, which iscurrently causing global concern.11–14

In order for an influenza A virus to be capable of causinga pandemic, it must fulfil four criteria: it must representa ‘new’ subtype where the haemagglutinin antigen isunrelated to its immediate (pre-pandemic) predecessor(for example, different to the currently circulatingA/H1N1 and A/H3N2); there must be little or no pre-existingimmunity in the population; it must cause significant clinicalillness; and it must be able to spread efficiently from personto person.15

At the time of writing, A/H5N1 fulfils three of thesecriteria, but is not yet capable of spreading efficiently fromperson to person. It is regarded as a strong contender to giverise to the next human influenza pandemic.16,17

The current concern over avian influenza

International aspects

During the first outbreak of HPAI A/H5N1 in Hong Kong in1997 there were 18 confirmed human cases with 6 deaths,while the 2003 Netherlands HPAI A/H7N7 outbreak was

ARTICLE IN PRESS

Responding to bird influenza: Past, present and future 191

associated with 83 mild human cases and one death in animmuno-compromised vet. In both outbreaks, all cases werelinked with exposure to infected poultry. Widespread cullingof poultry in both incidents helped to bring the outbreaksunder control and undoubtedly prevented the further spreadof the disease to other birds and humans.18–20

HPAI A/H5N1 disappeared after the Hong Kong outbreak,until it re-emerged in south-east Asia in late 2003. Sincethen the virus has infected and killed poultry and wildwaterfowl across Southeast Asia, parts of Africa and Europe.The first human cases of the current outbreaks werereported in Vietnam and China in December 2003, andthere have been a further 359 laboratory-confirmed cases,of which 226 were fatal (as on 17 February 2008).21 Themajority of these cases have been attributed to directexposure to infected birds, but in certain instances, forexample in Indonesia and Thailand, the possibility of limitedperson to person spread cannot be excluded.22

The UK public health response to avianinfluenza incidentsHPAI A/H5N1 first reached the UK in March 2006 when a wildmigratory swan, found dead in Fife, Scotland, testedpositive for the virus. In February and November 2007 HPAIA/H5N1 caused two separate large outbreaks in poultry inSuffolk in the East of England. In the February outbreak,anti-viral medication was offered to approximately 480workers involved in the culling of an additional 50 000birds on the farm.23 Outbreak management in Novemberinvolved the culling of 28 000 birds. Subsequently, A/H5N1has been detected in wild birds in the UK, where tendead mute swans have been identified as A/H5N1 positivein Dorset in south-west England during January–February2008.

Additionally, there have been two recent poultry out-breaks of low pathogenic A/H7 viruses in the UK. In April2006, A/H7N3 was found on a poultry farm in Norfolk, East ofEngland, and during May 2007 A/H7N2 was found at premisesin North Wales and north-west England, which wereepidemiologically linked. There were four confirmed humancases of A/H7N2 infection, including three hospitalisations,during this incident, but person-to-person transmission wasnot confirmed.

Given the evidence cited above, it is clear that, whilstA/H5N1 may be the subtype of influenza A currently causinggreatest concern, this should not be at the expense of otherinfluenza A viruses (notably A/H7 viruses) which also havethe capacity to infect and cause serious illness in humans,and the potential to produce a human pandemic.

Clear algorithms and guidance procedures have beendeveloped by the Health Protection Agency (HPA) to ensurethat the human health aspects of such avian influenzaincidents in England are managed in accordance withrelevant, evidence-based procedures: http://www.hpa.org.uk/infections/topics_AZ/influenza/avian/documents/HPUGuidance2409072.pdf.

Although the risk of acquisition of A/H5N1 influenza bytravellers returning from countries currently experiencingoutbreaks of avian influenza remains low, all countries nowneed to prepare for the potential occurrence of a smallnumber of human cases of HPAI A/H5N1, imported through

foreign travel. Guidelines for the reporting and investigationof travellers with suspect infection returning to the UKare also available on the HPA website: http://www.hpa.org.uk/infections/topics_AZ/influenza/avian/documents/A3Respillnessalgorithm030208.pdf.

Preparations for a future influenza pandemicWhatever the eventual influenza A subtype which gives riseto the next pandemic, countries across the world have beenplanning and preparing based on the phase-specific WorldHealth Organization Pandemic Flu Plan.24

The current epidemiological situation places the world inWHO Alert Phase 3 (see Table 1), but there is clear potentialfor escalation to WHO phase 6 when there is ‘‘increased andsustained transmission in the general population’’. In theUK, phase 6 has been subdivided into four UK alert levelswithin the UK contingency planning guidance, ‘Nationalframework for responding to an influenza pandemic’.25 Thisdocument, published jointly by the Department of Healthand the Cabinet Office in November 2007, describes theGovernment’s strategic approach for responding to aninfluenza pandemic. It provides background informationand guidance to public and private organisations developingresponse plans and is associated with supporting guidanceincluding an ethical framework and operating guidance foradult social care, ambulance services, community and acutehealthcare.

The UK strategy for mitigating the impact of pandemicinfluenza consists of pharmaceutical interventions togetherwith a broad range of non-pharmaceutical measures.

Pharmaceutical interventions

The main defence against pandemic influenza is a vaccinematched to the pandemic strain. However, the developmentof such a vaccine would involve a time delay which at bestwould be 4 months from identification of the strainresponsible for the pandemic, and more realistically 6months until there were sufficient stocks available to startusing the vaccine. Therefore, in the interim, the neurami-nidase inhibitor antiviral drugs will be used to try to mitigatethe impact of the pandemic. For this purpose the UK hasalready purchased a stockpile of oseltamivir (Tamiflus)sufficient to treat one-quarter of its population. The WHOhas also acquired oseltamivir in readiness to try to containthe first outbreaks of human pandemic influenza at thesource—likely to be a resource-poor country in SoutheastAsia—by treating cases and offering post-exposure prophy-laxis on a wide scale: WHO draft plan for rapid responseand containment (http://www.who.int/csr/disease/avian_influenza/guidelines/RapidResponse.pdf).

Neuraminidase inhibitors work by inhibiting viral replica-tion, in particular release of newly formed virions from aninfected host cell. Thus, they need to be taken as early aspossible following infection, certainly within the first 48 h ofbecoming infected, in order to be effective. However,modelling studies suggest that additional public healthbenefits will accrue if symptomatic people are treatedwithin 12 h of symptom onset. This is a considerable logisticchallenge, even in resource-rich settings. Unfortunately,countries lacking the resources to procure antiviral drugs

ARTICLE IN PRESS

Table 1 WHO international phases and UK alert levels.

WHO phases UK alert levels

Inter-pandemic periodPhase 1 No new influenza virus subtypes have been detected in humans. An influenza

virus subtype that has caused human infection may be present in animals. Ifpresent in animals, the risk of human infection or disease is considered to below.

Phase 2 No new influenza virus subtypes have been detected in humans. However, acirculating animal influenza virus subtype poses a substantial risk of humandisease.

UK not affected OR UK hasstrong travel/tradeconnections with affectedcountry Or UK affected

Pandemic alert periodPhase 3 Human infection(s) with a new subtype, but no new human-to-human

spread, or at most rare instances of spread to a close contact.

Phase 4 Small cluster(s) with limited human-to-human transmission but spread ishighly localised, suggesting that the virus is not well adapted to humans.

UK not affected Or UK hasstrong travel/tradeconnections with affectedcountry

Phase 5 Large cluster(s) but human-to-human spread still localised, suggesting thatthe virus is becoming increasingly better adapted to humans, but may notyet be fully transmissible (substantial pandemic risk).

Or UK affected

Pandemic periodPhase 6 Pandemic phase: increased and sustained transmission in the general

population.UK Alert Levels

1 1 Virus/cases only outsidethe UK.

2 Virus isolated in the UK.3 Outbreak(s) in the UK.4 Widespread activity acrossthe UK.

Past experience suggests that a second, and possibly further, wave of illnesscaused by the new virus is possible 3–9 months after the first wave hassubsided depending on seasonality. The second wave may be as, or moreintense than, the first

Post-pandemic periodReturn to inter-pandemic arrangements

J. Sarangi et al.192

and pandemic vaccine once available will be forced todepend on traditional public-health countermeasures andantibiotics to treat secondary bacterial complications.

Non-pharmaceutical measures

Whilst a new pandemic might be containable at sourcethrough rapid, virtually immediate, application of a combi-nation of stringent social distancing measures, area quar-antine and geographically targeted antiviral prophylaxis,the logistics of such an exercise will prove immenselychallenging in practice.26

Containment measures to prevent a pandemic spreadingto and across the UK are unlikely to be effective assimultaneous, multiple importations of the disease wouldbe expected, and if antiviral stocks were used to try to

‘stamp out sparks’ they would be rapidly depleted to littleeffect. Statistical modelling analysis has indicated thatborder restrictions and/or internal travel restrictions areunlikely to delay spread of the pandemic by more than 1–2weeks if 90% effective, and by only 2–3 weeks unless morethan 99% effective.27

Implementation of school closures has also been con-sidered as a non-pharmaceutical intervention to reduce thespread of pandemic influenza. Although this has had variedsuccess in the past, recent modelling work indicates thatschool closure during the peak of a pandemic might reduceclinical peak attack rates in children by up to 40% and slowthe epidemic spread to some degree. However, it has littleimpact on overall attack rates. Even with school closure it isimportant to ensure that children do not then meet ininformal networks outside of school, thereby negating anypositive effect.28

ARTICLE IN PRESS

Responding to bird influenza: Past, present and future 193

Therefore, individual and household interventionsfocussed on reducing local transmission of influenza maybe more effective at reducing the rate of spread. Thesemeasures include hand washing; voluntary isolation of cases;effective handling of contacts; as well as reducing non-essential travel.

This strategy is supported by the plans and specificguidance documents produced by Government departmentsand other organisations which play a crucial role inpreparedness and response arrangements for pandemicinfluenza. These include the infection control guidanceprepared by the Health Protection Agency (HPA) and theDepartment of Health (DH) working with various NHS andnon-NHS sectors29–31:

�

Guidance for pandemic influenza: infection control inhospitals and primary care settings � Clinical guidelines for patients with an influenza like

illness during an influenza pandemic

� Pandemic flu guidance for funeral directors � Pandemic flu guidance for cleaning staff and refuse

collectors in non-health care settings

� Pandemic flu guidance for the fire and rescue service � Pandemic flu guidance for the hospitality industry � Pandemic flu guidance for the police service

Relevant weblink: http://www.dh.gov.uk/en/Publication-sandstatistics/Publications/PublicationsPolicyAndGuidance/DH_080734.

Learning from our experience of previous pandemicsand outbreaks of avian influenza currently occurring acrossthe globe enables us to prepare for the next pandemic.Although much can be learnt from relevant historyand science, the specific impact of a pandemic is unpre-dictable and uncertain. Therefore it is essential thatresilience, flexibility and targeted resource is built intothe pandemic influenza preparedness planning across allsectors of society.

Conflict of interest

None.

References

1. Gilsdorf A, Boxall N, Gasimov V, et al. Two clusters of humaninfection with influenza A/H5N1 virus in the Republic ofAzerbaijan. Euro Surveill 2006;11:122–6 [Medline].

2. Euro Surveillance. Tenth EU country detects highlypathogenic avian influenza in wild birds. Euro Surveill 2006;11:E060309.3.

3. Smith GJ, Naipospos TS, Nguyen TD, et al. Evolution andadaptation of H5N1 influenza virus in avian and human hosts inIndonesia and Vietnam. Virology 2006;350:258–68 [CrossRef][Medline].

4. World Health Organization Global Influenza Program Surveil-lance Network. Evolution of H5N1 avian influenza viruses inAsia. Emerg Infect Dis 2005;11:1515–21 [Medline].

5. Taubenberger JK, Reid AH, Krafft AE, et al. Initial geneticcharacterization of the 1918 ‘‘Spanish’’ influenza virus. Science1997;275:1793–6 [Abstract/Free Full Text].

6. Shope RE. Swine influenza. III. Filtration experiments andaetiology. J Exp Med 1931;54:373–80 [Abstract].

7. Chu CM, Shao C, Hou CC. Studies of strains of influenza virusisolated during the epidemic in 1957 in Changchun. Vop Virus1957;B:278–81.

8. Chang WK. National influenza experiences in Hong Kong, 1968.Bull World Health Organ 1969;41:349–51 [Medline].

9. Rekart M, Rupnik K, Cesario TC, et al. Prevalence ofhemagglutination inhibition antibody to current strains of theH3N2 and H1N1 subtypes of influenza: a virus in sera collectedfrom the elderly in 1976. Am J Epidemiol 1982;115:587–97[Abstract/Free Full Text].

10. Davenport FM, Minuse E, Hennessy AV, et al. Interpretations ofinfluenza antibody patterns of man. Bull World Health Organ1969;41:4553–60.

11. Laurel N, Marine WM. Recycling of Asian and Hong Konginfluenza A virus hemagglutinins in man. Am J Epidemiol 1973;97:44–9 [Abstract/Free Full Text].

12. Masurel N, Heijink RA. Recycling of H1N1 influenza A virus inman—a haemagglutinin antibody study. J Hyg London 1983;90:397–402 [Medline].

13. Dowdle WR. Influenza A virus recycling revisited.Bull World Health Organ 1999;77:820–8 [1969;41:349–51(Medline)].

14. Nguyen-Van-Tam JS, Hampson AW. The epidemiology andclinical impact of pandemic influenza. Vaccine 2003;21:1762–8 [CrossRef][Medline].

15. Nguyen-Van-Tam JS, Sellwood C. Avian influenza and the threatof the next human pandemic. J Hosp Infect 2007;65(S2):10–3[sciencedirect].

16. Kandun IN, Wibisono H, Sedyaningsih ER, et al. Three Indonesianclusters of H5N1 virus infection in 2005. N Engl J Med 2006;355:2186–94 [Abstract/Free Full Text].

17. Normile D. Avian influenza. Human transmission but nopandemic in Indonesia. Science 2006;312:1855 [Medline].

18. Du Ry van Beest Holle M, Meijer A, Koopmans M, De Jager CM,Van de Kamp EEHM, Wilbrink B, et al. Euro Surveill 2005;10(12):264–8.

19. Webster RG, Hay AJ. The H5N1 influenza outbreak inHong Long: a test of pandemic preparedness. In:Nicholson KG, Webster RG, Hay AJ, editors. Textbook ofinfluenza. London: Blackwell Scientific Publications; 1998.p. 561–5.

20. Buxton Bridges CB, Katz JM, Seto WH, et al. Risk of influenza A(H5N1) infection among healthcare workers exposed to patientswith influenza A (H5N1), Hong Kong. J Infect Dis 2000;181:344–8 [CrossRef][Medline].

21. Zhu QY, Qin ED, Wang W, et al. Fatal infection withinfluenza A(H5N1) virus in China. N Engl J Med 2006;354:2731–2.

22. Kandun IN, Wibisono H, Sedyaningsih ER, et al. Three Indonesianclusters of H5N1 virus infection in 2005. N Engl J Med 2006;355:2186–94.

23. Editorial Team. Confirmed H5N1 avian influenza out-break on a poultry farm in England. Euro Surveill 2007;12:E070208.2.

24. World Health Organization. WHO global influenza preparednessplan. The role of WHO and recommendations for nationalmeasures before and during pandemics. Geneva: WorldHealth Organization; 2005 [Document ref: WHO/CDS/CSR/GIP/2005, 5].

25. UK Department of Health and Cabinet Office. Pandemicinfluenza—a national framework for responding to an influenzapandemic. November 2007.

26. Ferguson NM, Cummings DAT, Cauchemez S, Fraser C,Riley S, Meeyai A, et al. Strategies for containing anemerging influenza pandemic in Southeast Asia. Nature 2005;437:209–14.

ARTICLE IN PRESS

J. Sarangi et al.194

27. Pitman RJ, Cooper BS, Trotter CL, Gay NJ, EdmundsWJ. bmj.com 2005; bmj.com/cgi/doi/10.1136/bmj.38573.696100.3A.

28. Ferguson NM, Cummings DAT, Fraser C, Cajka JC, Cooley PC,Burke DS. Nature 2006;442:448–52.

29. Health Protection Agency. Health Protection Agency pandemicinfluenza contingency plan, 2006.

30. Department of Health, Health Protection Agency. Guidance forpandemic influenza: infection control in hospitals and primarycare settings, 2005.

31. British Thoracic Society, British Infection Society, HealthProtection Agency for the Department of Health. Clinicalguidelines for patients with an influenza like illness during aninfluenza pandemic, 2006.