Each year approximately 36 000 people in the US and approximately 500 000 people globally die from influenza-related illnesses,[1] and the elderly are disproportionately affected. During influenza epidemics, people aged ≥65 years account for >90% of the deaths from influenza and its complications.[2] However, the morbidity and mortality seen annually with seasonal influenza pale by comparison when one considers the potential of pandemic influenza.

Seasonal influenza is due to influenza A and B strains and in the US usually begins in early to mid-October with infections being observed through to May. The majority of infections in the northern hemisphere occur in mid- to late winter. In the US, the Advisory Committee on Immunization Practices (ACIP) recommends immunization against seasonal influenza for persons aged ≥50 years, individuals with chronic illnesses or conditions and healthcare workers.[3] Seasonal influenza, and thus the timing of an influenza vaccination, varies across the world and influenza is transmitted year round in the tropics.

1. Basic Influenza Genetics: Drift and Shift

Influenza viruses are constantly changing, primarily through a process called ‘antigenic drift’, which occurs when a point mutation leads to a variation in one of the influenza antigens. Immunity to key surface antigens, particularly the haemagglutinin (H) and neuraminidase (N) antigens, reduces the likelihood of infection and the seriousness of disease if it occurs.[4,5] New strains are common and vaccines must be modified on an annual basis to ensure that they provide immunity against circulating strains. The annual changes among circulating strains caused by drift, although enough to warrant a change in the vaccine, are not enough to cause a pandemic because the strains are sufficiently antigenically related to those circulating in preceding years as to ensure that a substantial degree of cross-reacting or ‘herd’ immunity generally exists within a population.[1] ‘Antigenic shift’ through genetic re-assortment between human and animal or avian strains is a less common phenomenon and results in a major change in the virus. The result is that immunity may be completely lacking and the virus then has the potential to cause a pandemic.[1,5] Influenza type B infects humans only and is therefore incapable of undergoing a shift. Pandemics are caused solely by influenza A strains.

2. Past Pandemics

Ten pandemics have been recorded in the past 300 years with 10–49 years usually passing between pandemics. The three pandemics during the 20th century were caused by H1, H2 and H3 strains.[1] The 1918–9 or Spanish influenza pandemic was estimated to have been associated with the deaths of over 500 000 people in the US and with a global mortality of 50–100 million. Calculated excess death rates during the 1918 pandemic vary from 1.2 per 1000 persons in Argentina to 236 per 1000 persons in Western Samoa.[6] The 1957–8 pandemic or Asian influenza was estimated to have been associated with the deaths of 60 000 people in the US and approximately 1 million globally. It was first identified in China and then spread globally. The 1968–9 pandemic or Hong Kong influenza was estimated to have been associated with the deaths of 34 000 people in the US. This pandemic also spread globally from Asia.[7] The world is nearly 40 years on from the last pandemic and with an average of 24 years between previous pandemics, and a novel circulating influenza A subtype, epidemiologists and public health experts are concerned that the next pandemic will occur soon.

Most pandemics are associated with death primarily of the very young, the elderly or those with chronic illnesses. These pandemics are associated with serious illness primarily among the groups most at risk of disease from seasonal influenza. However, not all pandemics are the same. The 1918 pandemic was associated with high morbidity not only in very young and very old persons but also in healthy young adults aged 15–45 years, who also experienced high mortality rates.[8] Some researchers have postulated that a robust, over-active immune response (‘cytokine storm’) may have contributed to deaths in younger persons; however, others are sceptical.[8–10] The 1918–9 Spanish influenza pandemic was of avian origin and appears to have gradually adapted to humans until it could become efficiently transmitted.[7,11] The 1957–8 Asian influenza pandemic acquired three genes from avian species and five from the then circulating human strains, and through genetic re-assortment became a novel pandemic strain.[7,12] The 1968–9 Hong Kong influenza pandemic acquired two genes through re-assortment with avian viruses and kept six from the human strains circulating at the time. The death rate from this pandemic may have been lower because the virus had a shift only in the H antigen and not in the N antigen. Pre-existing immunity to the N antigen may have modified the severity of the disease.[7,13]

3. Pandemic Preparedness

Pandemic preparedness planning by the Department of Health and Human Services (DHHS) in the US and by the WHO globally depends on several assumptions derived from studying past pandemics. It is presumed that everyone will be susceptible to the pandemic strain and that the clinical attack rate will be ≈30% in the general population with half of these individuals ill and seeking medical care. Hospitalizations and deaths will depend greatly on the virulence of the virus.[14,15] Table I outlines two possible scenarios for the US, as modelled by the DHHS.[14]

Table I
figure Tab1

Number of episodes of illness, healthcare utilization and death associated with moderate and severe pandemic influenza scenariosa (modified from US Department of Health and Human Services[14])

Full size image

The typical incubation time for influenza is 2 days and people can transmit the virus 1 day prior to clinical symptoms. Two to possibly three secondary infections are likely to occur per primary case. Communities should expect two waves of pandemic illness, each lasting 6–8 weeks. Later waves in previous epidemics have tended to be more severe. The seasonality of a pandemic cannot be predicted with certainty, although in the northern hemisphere it is thought to be most likely to occur in the autumn or winter.[14]

As the review of previous pandemics in the 20th century noted (see section 2), risk groups for severe and fatal infection cannot be predicted with certainty. Although infants, young children, those with chronic illnesses, pregnant women and the elderly are generally more susceptible to influenza, the mortality among young healthy adults was very high in the 1918–9 Spanish influenza pandemic. A pandemic of any severity will have major economic and social consequences but one with high morbidity among healthy adult workers has the possibility of triggering economic and social chaos if a well thought-out plan is not in place for dealing with the social disruption that will result when 20% or more of the workforce is incapacitated. The DHHS regularly updates its website, which reviews the pandemic preparedness plans of the federal and state governments.[16] This website also delineates recommendations for businesses, schools, healthcare providers and even individuals to prepare for a potential pandemic.

4. Avian Flu — Current Threat of Pandemic Influenza

Potential pandemic strains must have the following features: (i) an antigenic make-up to which the population is immunologically naive; (ii) the ability to replicate and cause disease in humans; and (iii) efficient transmission from human to human.[7] The current influenza A/H5N1 virus meets the first two criteria but not the third at this timepoint.

While several avian influenza viruses are being monitored for their potential to cause human disease, the greatest current concern is the threat of highly pathogenic avian influenza A/H5N1. Avian influenza A/H5N1 is a recently described, new viral subtype that has spread rapidly through poultry flocks and has now infected migratory bird populations in >40 countries. This virus causes severe disease in humans and currently has a reported case-fatality rate of >50% for laboratory confirmed cases.[7,15,16] The virus is not currently able to cause a pandemic since it is not spread easily from person to person. There is, however, great concern that the virus will undergo re-assortment with circulating human strains either in a swine intermediary host or directly in a human host and change into an easily transmissible pandemic strain.[7] This would be a similar chain of events to that which led to the 1957–8 and 1968–9 pandemics. Another possibility is for the virus to continue to undergo antigenic drift and eventually mutate to a strain that is easily transmissible from person to person, which would be similar to the proposed evolution of the 1918–9 pandemic strain.

Identification of human cases of A/H5N1 influenza and deaths has steadily increased over the past 4 years. In 2003, four patients, all of whom died, were identified in China and Vietnam. As of 30 January 2008, there have been 357 documented human cases with 224 deaths, resulting in a mortality rate of >50%, in 14 countries; additional details of confirmed cases reported to the WHO are shown in table II.[17] The epidemic in birds has affected >40 countries throughout Asia, the Middle East, Africa and Europe.[15] The host range of the virus has continued to expand from waterfowl to poultry, migratory birds, swine and felids.[18] Rare cases of human-to-human transmission have also been reported.[19,20]

Table II
figure Tab2

Cumulative number of confirmeda human cases of avian influenza A/H5N1 reported to WHO as of 30 January 2008 (modified from WHO[17])

Full size image

The WHO has defined three pandemic periods with a total of six phases and has established an action plan for each phase. Table III summarizes the periods and phases as defined by the WHO.[21] We are currently in a pandemic alert period phase III. Although human cases have occurred, the virus is not easily transmissible from person to person.

Table III
figure Tab3

WHO-defined pandemic phases[21]

Full size image

One of the important recommendations for the inter-pandemic and pandemic alert periods is to identify novel influenza strains with pandemic potential at an early stage. This is of utmost importance for vaccine research to progress and have any chance of producing an effective vaccine and vaccination strategy.

5. Influenza Prevention and Vaccination

Hand hygiene, cough etiquette and contact avoidance are all helpful in preventing the transmission of influenza. This is basic information that everyone of all ages and professions should be aware of and follow.[22] Although controversial, quarantine of individuals and potentially infected persons and groups may delay a pandemic and may be an important measure to try and gain more time to effectively deal with a pandemic, but is not thought to be able to prevent a pandemic.[23] However, since influenza can be transmitted prior to the onset of symptoms, quarantine is unlikely to ever be fully effective.

The primary way to reduce morbidity and mortality from influenza is implementation of an effective vaccination programme. This is true not only for pandemic influenza but also for seasonal influenza. Two meta-analyses of influenza vaccination in the elderly have demonstrated a reduction in all-cause mortality among community-living elderly of about 50% and an even greater benefit has been found for those living in long-term care facilities.[24–27] Recent studies have suggested that some of this decrease in mortality is related to a decrease in myocardial infarction following influenza vaccination.[28,29] Other studies have shown an increase in both myocardial infarction and stroke after acute respiratory infections and it is hypothesized that the inflammation associated with influenza and other infections alters the probability of a vascular event although the mechanism for this is not yet known.[30] Despite the high risk of significant illness from seasonal influenza and the known efficacy of vaccination, annual influenza immunization rates among high-risk groups remain distressingly low. National surveys in the US over the past decade found that only 55–65% of individuals aged >65 years report receiving an annual influenza vaccination.[31]

Vaccination of healthcare workers is also highly effective for preventing illness in the high-risk individuals they care for.[32–34] However, despite recommendations that all healthcare workers be vaccinated, the percentage vaccinated is variable and very institution dependent. Rates of <15% have been reported by some institutions, and in the US the majority of healthcare workers do not receive an annual influenza vaccination.[2,3,34–36] Immunization rates among healthcare workers can be improved when institutions make immunization of healthcare workers a priority and implement immunization campaigns.[34,35,37] The Virginia Mason Medical Center in Seattle, WA, USA, recently mandated influenza vaccination for all employees and achieved a 98% rate of vaccination.[2] The US Government is beginning to realize that this is an important safety issue and 22 states now require influenza immunization for healthcare workers in acute and/or long-term care facilities.[2] Most recently, the Infectious Diseases Society of America and other professional societies have called for influenza immunization of healthcare workers or informed declination in recognition of the pivotal role healthcare workers occupy and the importance of maintaining an effective medical response when a pandemic occurs.[38]

Table IV summarizes the current ACIP recommendations on who should receive influenza immunization. The ACIP is gradually moving toward a recommendation of universal annual vaccination.[39] This would also have the favourable effect of increasing demand for a vaccine between pandemics and would likely result in increased manufacturing capability and an improvement in the ability of the US to respond to a pandemic. A 5-year pilot study of universal vaccination in Ontario, Canada, resulted in a doubling of vaccination rates in high-risk individuals along with the development of vaccine distribution and administration programmes that were felt to improve pandemic preparedness.[39]

Table IV
figure Tab4

US Advisory Committee on Immunization Practices recommendations for influenza vaccination (2005–6)

Full size image

Current vaccines against seasonal influenza are the trivalent inactivated vaccine and the live attenuated vaccine, which are developed each year against the expected predominant influenza strains. These vaccines are generally well tolerated and safe. Occasional injection-site reactions with the trivalent inactivated vaccine and transient upper respiratory symptoms with the live attenuated vaccine occur.[40] The trivalent inactivated vaccine is indicated for those aged >6 months, and the live attenuated vaccine for healthy individuals aged 5–49 years.[3]

6. Pandemic Preparedness and Vaccination

Vaccination holds the greatest promise of averting the mortality seen in past influenza epidemics and pandemics. Although vaccines were available during the 1957–8 and 1968–9 pandemics they were used too late to have a significant impact.[41] The current avian flu threat has been recognized relatively early, allowing significant vaccine research to occur and an assessment of our capabilities for immunizing those at high risk in the event of a pandemic. The EU (under the European Medicines Evaluation Agency), the US (under the National Institute of Allergy and Infectious Diseases) and private vaccine manufacturers are working on creating and testing vaccines against the A/H5N1 strain of influenza.[42]

Current inactivated influenza vaccines use proven technology and there is a large worldwide experience and familiarity with this approach.[40] Currently, the manufacturing capacity is for egg-based inactivated vaccines, although even at current maximum production, influenza vaccine supply is expected to fall far short of projected demand in the event of a pandemic, and would allow only 1–2% of the population to be immunized.[40,42,43] Other approaches that have the potential to allow a larger number of doses to be produced in a shorter period of time include peptide-based vaccines, DNA vaccines and live attenuated vaccines. Additional technologies such as mammalian cell culture techniques and effective adjuvants that allow dose sparing are being investigated. Finally, an approach known as a universal vaccine approach, utilizing highly conserved and antigenic influenza proteins that are subject to less mutational pressure, is being studied and may offer a potential breakthrough.[44]

Current research involving A/H5N1 inactivated vaccines suggest that these vaccines may be less immunogenic than seasonal influenza vaccines.[40,43,45–48] A population with a complete lack of immunity to a novel virus would probably require at least two doses of vaccine to induce protective immunity.[40] A requirement for two doses would make rapid implementation of an effective vaccine strategy during a pandemic impossible. Current trials of candidate vaccines do not include the elderly, a group that can be expected to be hit hard by a pandemic and who may be more difficult to immunize effectively.[49]

Strategies to improve the most efficient use of a limited supply of vaccine during a pandemic include:

  • A pre-event strategy that would include the A/H5N1 vaccine in annual vaccinations to prime the population. This may avoid the need for a two-dose schedule during a pandemic and potentially allow lower doses of the vaccine to be administered, meaning more people could be immunized with a limited amount of vaccine.[40]

  • Use of adjuvants that increase immunogenicity and are therefore ‘dose sparing’.[40,43]

  • Use of alternative routes of administration, such as intradermal administration, which may be more immunogenic and dose sparing.[50]

Research is ongoing regarding each of these strategies.

In April 2007, the US FDA announced licensure of the first A/H5N1 vaccine for use in humans.[51] This vaccine is derived from the A/Vietnam/1203/2004 influenza virus and is the vaccine studied by Treanor and colleagues.[46] The vaccine is given in two 90 µg doses 28 days apart. Forty-five percent of 103 healthy adults between the ages of 18 and 64 years responded with levels of antibodies expected to reduce the risk of acquiring A/H5N1 influenza.[46] The vaccine has been purchased by the federal government for inclusion in the national stockpile for distribution by the DHHS if needed.

Live attenuated influenza virus vaccine (LAIV) is highly immunogenic in unprimed populations and would potentially allow the use of a single low dose of vaccine.[40] LAIV would be expected to induce mucosal immune responses and therefore may be more effective in reducing virus transmission than inactivated vaccine.[40] Additional information regarding the safety and immunogenicity of LAIV in the very young and in individuals aged >50 years is needed prior to consideration of LAIV as the primary vaccine against a pandemic strain of influenza. Additionally, while concern exists among some that LAIV could theoretically be transmitted to others and may be able to undergo re-assortment to become more pathogenic,[40] there is no evidence to date of this occurring. LAIV for novel influenza A subtypes are undergoing clinical trials.

It is important to note that the capacity to manufacture a vaccine can be increased by increasing demand for an inter-pandemic influenza vaccine. Current worldwide demand is reportedly too weak for many manufacturers to make large investments in preparations for the scale of production needed in the event of a pandemic.[41]

7. Conclusion

Current influenza vaccination rates remain low, particularly among those most vulnerable and those most likely to transmit infection to others (healthcare workers). The development of a vaccine that could achieve cross-protective immunity against different influenza A variants, including the A/H5N1 variant and other circulating strains, would be an enormous breakthrough. The global community must intensely gear up vaccine production and mass immunization programmes would have to be implemented worldwide in the event of a pandemic. In the meantime, we can each do our part in saving lives by promoting seasonal influenza vaccination particularly for those at high risk, including the elderly, and for healthcare workers who are likely to transmit the virus to those most vulnerable persons.