The Human Influenza Viruses
Structure and Replication
Classification and Nomenclature
Clinical Signs and Symptoms
Pulmonary complications
Non pulmonary complications
Diagnosis
Treatment
Transmission and Epidemiology
Antibody Responses
Antigenic Variation
Vaccination
Inactivated vaccine
Live attenuated vaccine
Influenza in Animals
Avian Influenza (Bird flu)
Swine Influenza (Pig flu)
Epizootic swine influenza
Zoonotic swine influenza
History of swine influenza
Equine Influenza (Horse flu)
The 2007 Australian equine influenza outbreak
Canine Influenza (Dog flu)
Influenza Pandemics
The 1918 Spanish Flu
The H5N1 Asian Bird Flu
Origins H5N1
Human H5N1 Infection
Treatment
Infection control
Vaccines
The 2009 H1N1 Swine Flu Pandemic
Origin of the H1N1 Pandemic
Transmission
Clinical features
Treatment
Vaccines
Bioterrorism and Influenza
The Human Influenza Viruses
Influenza or "flu" usually starts as an acute febrile illness, characterized by sore throat, muscle pain, cough and headache. Thereafter infections can vary from an ongoing cough to fulminant pneumonia, resulting in death. One central feature of influenza is its high transmissibility, leading to the epidemic and pandemic nature of the infection.
Structure and Replication
Influenza viruses are roughly spherical when seen by electron microscopy, but filamentous forms are also common. The virus is covered by a glycoprotein coat or envelope. Projecting from the surface of this envelope are multiple large glycoprotein spikes (diagram below). These are either rod shaped and called hemagglutinin (H) or mushroom shaped and called neuraminidase (N). Hemagglutinin is essential for binding of the virus to pharyngeal cells so that subsequent invasion can occur, while neuraminidase is necessary for infection from cell to cell and release of virus from infected cells.
The glycoprotein coat is wrapped around a central core, comprising viral RNA and viral protein; the nucleocapsid. The RNA is usually single stranded and separated in seven or eight separate segments (diagram below).
Each segment of RNA codes for specific viral proteins: segment 1- basic polymerase 1 (PB1); segment 2 – basic polymerase 2 (PB2); segment 3 – acid polymerase (PA); segment 4 -hemagglutinin (H); segment 5 – neuraminidase (N); segment 6 – nucleoprotein (NP); segment 7 - matrix 1 protein (M1) and matrix 2 protein (M2); segment 8 – nonstructural proteins (NS1 and NS2) and nuclear export protein (NEP).
The initial step of infection is the attachment of hemagglutinin to sialic acid sugars on the cells of the nose, pharynx and throat. Once attached, the hemagglutinin is cleaved by epithelial cell proteases and the virus is able to enter the interior of the cell by endocytosis where it is contained in a vacuole within the cell cytoplasm. The hemagglutinin fuses with the vacuole membrane and the M2 protein forms a channel through which protons move to acidy the virus core. These acidic conditions provide a trigger for the viral nucleocapsid to be released back into the cytoplasm. The nucleocapsid migrates to the cell nucleus and enters the nucleus by way of nuclear pores. Within the host nucleus there is initially transcription of proteins necessary for replication. Once the initial proteins have been produced multiple copies of viral RNA are produced which are translated into the necessary viral proteins on ribosomes within the cell cytoplasm. These viral proteins are then assembled into virus particles. Hemagglutinin and neuraminidase molecules cluster into a bulge in the cell membrane and the viral nucleocapsid also enters the membrane protrusion. The virus then buds off from the cell wrapped in the cell phospholipid membrane covered with viral hemagglutinin and neuraminidase. The final detachment from the cell requires the viral neuraminidase to cleave the cell sialic acid residues. The host cell dies once the viruses is released and the new virus particles can then attach to another cell via its hemagglutinin and so initiate another cycle of replication.
Classification and Nomenclature
Influenza viruses are part of the family Orthomyxoviridae and are separated into Influenza A, Influenza B and Influenza C, based on the characteristics in the table below.
Influenza A viruses are found in a wide variety of animal species such as humans, birds, horses, pigs and others. Influenza B viruses occur in humans only, while Influenza C occurs only in humans, dogs and pigs.
Influenza viruses can be further classified according to their hemagglutinin (H) and neuraminidase (N) surface glycoproteins. There are 16 H and 9 N known subtypes. All of these are found in birds but only H 1, 2 and 3, and N 1 and 2 are found in humans. A number of examples of H and N classification are as follows:
H1N1 - The Spanish Flu of 1918 and the 2009 Swine Flu.
H5N1 - The Asian bird flu.
H3N8 - Equine influenza.
H3N2 - Seasonal influenza.
The complete nomenclature of a particular influenza virus could be as follows:
A/Sydney/52/2007/H3N2
A= Virus type
Sydney= Place of origin
52= Strain number
2007= Year of isolation
H3N2= Virus subtype
Clinical Signs and Symptoms
After an incubation period of about 2 days, there is an abrupt onset of chills, fever, headache and generalized aches and pains. Cough, sore throat and watering eyes are also common. The fever and acute symptoms usually last for about 3 days and then defervesce, followed by a 1 to 2 week period to full recovery.
Pulmonary complications
Croup and exacerbation of asthma are common complications of influenza in children, while exacerbation of chronic lung disease (bronchitis and emphysema) is common in adults.
Primary influenza pneumonia and secondary bacterial infection are 2 well recognized complications of influenza. Primary viral pneumonia usually occurs in patients with other chronic illnesses such as cardiovascular disease, chronic bronchitis and emphysema. After the typical abrupt onset, the patient's cough and fever progress and they become short of short of breath, cyanosed and blood gases show marked hypoxia. Sputum usually shows no bacteria on gram stain and culture and the patients do not respond to antibiotics. Chest X-ray reveals bilateral diffuse involvement with no lobar consolidation. In the past, mortality rate was high, but has improved with the advent of new antiviral agents. Secondary bacterial pneumonia usually occurs following a period of initial improvement. The patient then deteriorates, with relapse of high fever and a cough productive of sputum. Physical examination and chest X- ray reveal an area of consolidation while sputum shows bacteria on gram stain and culture. Streptococcus pneumoniae, Haemophilus influenzae and Staphylococcus aureus are the predominant pathogens. A few patients who have progressive pulmonary disease may not fit neatly into either of these 2 categories, and may have a mixed viral and bacterial pneumonia or primary influenza pneumonia which only affects one lobe or segment. In immunosuppressed patients (cancer, chemotherapy, transplants and HIV), influenza pneumonia can be particularly severe with a high mortality rate.
Non pulmonary complications
Inflammation of muscles (myositis) resulting in tender leg and back muscles occurs mostly in children and to a lesser extent in adults. This may be seen biochemically as an increase in serum creatinine phosphokinase and myoglobin in the urine.
Both myocarditis and pericarditis are occasionally associated with influenza infection. More importantly however is the significant risk of death in patients with influenza who have preexisting cardiac disease.
A toxic shock like syndrome is occasionally seen, particularly in severe infections such as seen in H5NI (see below). A specific Staphylococcal toxic shock syndrome may occur when secondary Staphylococcus aureus pneumonia is caused by a toxin producing strain.
Central nervous system complications are unusual and include Guillian-Barré syndrome, transverse myelitis and encephalitis.
Reyes syndrome is seen in many viral infections including influenza. It occurs exclusively in children who have been given aspirin to lower their temperature. Several days after taking the aspirin the children have a change in mental status which can vary from lethargy to delirium, coma, seizures and respiratory arrest.
Diagnosis
Nasal swabs, throat swabs, nasopharyngeal aspirates and sputum are the specimens usually submitted to the laboratory. Virus can readily be isolated from these specimens by growing the virus in embryonated eggs or tissue culture. Influenza antigen can also be detected in these specimens by immunologic testing with labeled influenza antibody (direct immunofluoresence). Culture of virus is no longer routinely used because of the need for maintaining cell lines and the lengthy turnaround times. Direct immunofluoresence is still practiced by some laboratories but together with culture is being replaced by polymerase chain reaction (PCR). PCR is now the most sensitive, specific and rapid test available for the detection of influenza nucleic acid.
Detection of antibody in the serum is not useful for diagnosis during the acute illness but can be useful for a retrospective diagnosis or for epidemiological purposes.
Treatment
Symptomatic treatment is the mainstay of influenza management. Patients should rest, drink fluids and take medications to reduce the fever and relieve the muscle aches. Aspirin should be avoided in children because of the complication of Reyes syndrome. Antibiotics are of no value and should be avoided unless secondary bacterial pneumonia is suspected. In severely ill patients supportive therapy with artificial ventilation may be required.
When using antiviral drugs a number of principles need to be considered. Most immunocompetent patients have had previous influenza and because of cross immunity will resolve their influenza infection fairly rapidly, making antiviral therapy unnecessary. It is also well established that there is no benefit of antiviral therapy if commenced after 48 hours. Antiviral therapy should therefore only be used in selected at risk patient groups and preferably within 24 hours.
There are 2 classes of antiviral drugs available: the M2 inhibitors (amantadine and rimantadine) and the neuraminidase inhibitors. The M2 inhibitors are so named because they inhibit the M2 ion channel in viral replication (see replication). These drugs are only active in vivo against influenza A. Unfortunately these agents are now seldom used as nearly all seasonal H3N2 strains are resistant. Seasonal H1N1 strains are still susceptible to these agents but nearly all pandemic H1N1/09 are resistant. As knowledge of the strain is seldom available when treatment is initiated these drugs are seldom used.
The neuraminidase inhibitors such as oseltamivir (Tamiflu) and zanamivir (Relenza) are analogues of sialic residues and inhibit the neuraminidase of both influenza A and B. Oseltamivir is taken orally and is rapidly absorbed from the gastrointestinal tract. Zanamivir is not available orally and is taken by inhalation. Both drugs are well tolerated, with gastrointestinal symptoms such as nausea and diarrhea being seen in only a small proportion of patients. Both drugs result in reduced viral shedding and reduction in symptoms if given within 36 hours of symptoms. Early treatment reduces the frequency of complications and the need for hospitalization in adults and decreases the frequency of otitis media in children.
Drug resistance to these antiviral agents has been seen. The most common cause of this resistance is a mutation in the neuraminidase gene which prevents binding of the drug. These mutations are specific to the subtype of neuraminidase: the common resistance mutation in N1 is H274Y, while the R292K is seen in N2. Other more unusual resistance mechanisms involve changes in the hemagglutinin, which results in reduced affinity of the virus for hemagglutinin, allowing easy cell to cell spread of the virus without the need for neuraminidase activity. Most H3N2 seasonal influenza strains remain sensitive to both oseltamivir and zanamivir. However nearly all seasonal H1N1 strains are resistant to oseltamivir but sensitive to zanamivir. A small number of pandemic H1N1 strains are resistant to oseltamivir (1-2%) but all remain sensitive to zanamivir.
Transmission and Epidemiology
Influenza virus is maintained in humans by spreading from person to person during an acute infection. The virus is transmitted through the air as droplets or aerosols which are generated by coughs and sneezes. Patients are usually infectious for one day before symptoms start, till about 5 to 7 days later. Direct transmission occurs when droplets of mucous are directly coughed or sneezed into another person's nose, eyes or mouth and is thought to be the main method of transmission. The virus may also be spread via the airborne route where the virus is aerosolized and suspended in the air in very small particles (0.5µm to 5µm in diameter). The water in these small particles can then evaporate and the virus can remain suspended in air for long periods. These tiny particles or droplet nuclei are inhaled into the airways of other people as the virus is dispersed by air currents within a room or over larger distances. Influenza can also survive outside the body on inanimate objects (door knobs, light switches, coins, notes, tissue paper) and on skin including hands. The virus can be transmitted to another person by touching these contaminated objects or shaking hands and then transferring the virus to the nose or mouth. Influenza virus can survive outside the body for up to 2 days, but if protected by mucous such as in tissues it may survive much longer.
Influenza spreads relentlessly by these mechanisms from person to person until a large number of people in a specific location (city, town or country) are infected; this is called an epidemic. Such epidemics characteristically reach a peak in two to three weeks and last five to seven weeks. About 10% to 20% of people who are exposed acquire the disease before the epidemic ceases; often as quickly as it commenced.
Influenza epidemics occur most often in the winter months which therefore results in a northern and a southern hemisphere influenza season each year. The reason for the winter prevalence of influenza is usually believed to be that people are indoors during the winter and are therefore in close contact with others. Other theories include longer survival on surfaces in winter and higher rates of aerosol transmission in the cold.
A single influenza strain is usually responsible for an outbreak or epidemic. In some seasons two strains within the same subtype can circulate ( 2 strains of H3N2) or two subtypes circulate together (H1N1 and H3N2).
When the seasonal epidemics spread beyond the epidemic location and progress to involve all parts of the world a pandemic is called. Pandemics are characterized by extremely rapid transmission; outbreaks occurring at different places around the world; occurrence of outbreaks in the summer; high attack rates in all age groups and; high mortality rates. Pandemics occur as a result of the emergence of a new virus to which the world population has no immunity (see below).
Antibody Responses
Before considering the genesis of epidemics and pandemics it is necessary to consider the immune response to influenza infection. Influenza infection results in the development of antibodies to the surface glycoproteins hemagglutinin and neuraminidase and to structural M and NP proteins. These antibodies begin to appear within 2 weeks of infection, with peak antibody responses occurring at 4 to 7 weeks. These antibodies slowly decrease with time but can be detected many years after infection. Antibodies to hemagglutinin protect against viral attachment (infection) and disease with the same virus. Antibodies to neuraminidase do not protect against viral attachment but reduces release of virus from cells and therefore reduces viral shedding. Hemagglutinin and neuraminidase antibodies are present in the serum (systemic antibodies) and in the cells lining the upper respiratory tract (mucosal antibody). Either mucosal antibody or systemic antibody can be protective, but optimal protection occurs if both mucosal and systemic antibodies are present. Antibodies to proteins other than hemagglutinin and neuraminidase may be partially protective but their role is not entirely clear.
Antigenic Variation
Circulating seasonal influenza viruses frequently undergo relatively minor spontaneous mutations which results in changes to small protein sequences (antigens) and is termed antigenic variation. When theses small mutations occur on the external hemagglutinin and neuraminidase glycoproteins it is termed antigenic drift. When such a new mutated antigen appears there is limited immunity within a community and an epidemic can occur. Subsequent exposure to this new antigen results in the generation of new antibody and development of immunity.
Less commonly, a completely new virus appears to which the population has absolutely no immunity. These new viruses are the cause of pandemic influenza and because of the magnitude of the antigenic change is called an antigenic shift. These large antigenic changes are not the result of a mutation in a circulating seasonal virus, but the emergence of a totally new virus. How antigenic drift, antigenic shift, population immunity and epidemics and pandemics interrelate is shown schematically in diagram below.
In 1968 the previously unseen H3N2 was introduced into a world without immunity and a pandemic ensued. The population developed immunity and the pandemic subsided. In the following years the same virus (H3N2) underwent small mutations (antigenic drift) causing epidemics of seasonal influenza to which the population developed immunity. Over the next 40 years the level of immunity to this virus became very high and epidemics became less and less severe. In this setting the emergence of a new virus such as H1N1 in a susceptible population resulted in the next pandemic.
The critical question is where do these new pandemic viruses come from? Although birds are the predominant hosts for influenza (see Influenza in Animals below), pigs are considered to play an important role in evolution of new strains and act as a link from birds to humans. The respiratory epithelium of pigs has receptors for both bird and human influenza viruses, allowing both types to cause infection at once and thus allowing reassortment of RNA fragments. The pigs therefore act as a mixing vessel for rearranging and combining different components of avian, swine and human influenza viruses. Such reassortment may give rise to appearance of a totally new virus which may be pathogenic for birds or humans or both and may result in a pandemic virus.
Vaccination
The most effective way to prevent influenza is by the annual administration of vaccine. The primary objective of vaccination is to prevent the disease impact by vaccinating certain groups and contacts who are at high risk of more severe influenza, possibly resulting in hospitalization and death. There is also some evidence to suggest that widespread vaccination of children may help to interrupt the transmission of influenza in the community.
Persons who should receive the vaccine are as follows:
- Children aged 6 months to 4 years.
- Persons aged over fifty.
- Women who will be pregnant during the influenza season.
- Adults and children with chronic lung, cardiovascular, kidney, liver, diabetes and hematological problems.
- Adults and children who are immunosuppressed, either as a result of malignancy, chemotherapy or HIV.
- Residents of nursing homes and other chronic care facilities.
- Persons who can transmit influenza to others at risk. These include health care workers, household contacts of children aged less than 5 years and adults over 50 years, household contacts of persons with medical conditions as described above.
- Children who are taking long term aspirin and therefore may be at risk for Reyes syndrome.
For global vaccine production large quantities of virus are required. This has been achieved in the past through specifically adapted strains that can grow in high yield from hen's eggs. More recently bulk culture of virus from tissue culture has been achieved. Because of the high virus mutation rate (antigenic drift) the vaccine will usually only protect against circulating virus for about 2 years. Therefore new vaccines are produced annually and are based on predictions of which virus will be circulation in the forthcoming season. Because of the large quantities of vaccine required, manufacturers require about six months of production time before the next seasonal influenza strikes. This prediction can not be entirely accurate and therefore some people who get vaccinated will still acquire influenza.
Inactivated influenza vaccine
The original vaccine licensed in the United States in 1943 was made from whole virus inactivated with formalin. These vaccines were effective but had significant adverse reactions. Since then virus has been treated with solvents to create split vaccines or with detergents to create subunit vaccines which have been better tolerated.
Although generally well tolerated inactivated vaccines do have some minor adverse reactions. Mild local soreness at the injection site occurs in about 60% to70% of vaccinees. Systemic reactions such as flu-like illness, fever and malaise are uncommon and occur in only about 3%. Life threatening hypersensitivity reactions are rare and are due to hypersensitivity to hens eggs. If people can eat eggs or egg products, vaccination can be safely undertaken. Guillain-Barré syndrome, a rare condition characterized by ascending paralysis of muscles is a very rare complication of influenza vaccination, occurring in 1in 100,000 vaccinations.
After vaccination, antibodies occur in about 90% of recipients. This usually occurs with one dose of vaccine if recipients are vaccinated annually, but may require two doses in previously unvaccinated people. These antibodies usually peak at 2 to 4 months and then fall quickly being barely detectable by the next influenza season. Only systemic antibodies and not mucosal antibodies are generated by the injected inactivated vaccine. These vaccine generated antibodies appear to be protective with influenza being prevented in 70% to 90% of vaccine recipients. Although the antibody response may be variable in some of the at risk group for whom vaccination is recommended, studies have shown that vaccine offers protection to the older population and at risk groups including those with HIV.
Live attenuated influenza vaccine
In order to use a live virus as a vaccine, it needs to loose its ability to cause influenza (attenuated) but still be able to generate an antibody response and provide protection against disease. This attenuation process is established by cold adaptation. Influenza virus need to be able to replicate at 37°C in order to enter the blood stream and cause a systemic infection. By producing a virus that replicates at 25°C and not at 37°C the virus may be able to enter the cells lining the upper respiratory tract but will not be able to invade. Such cold adapted influenza vaccines (CAIV) are now readily available. Hemagglutinin and neuraminidase components can be introduced by taking advantage of the segmented nature of the influenza RNA. Hemagglutinin and neuraminidase genes of the predicted next seasonal influenza strain can be inserted into the cold adapted vaccine strain ready for vaccination before the influenza season.
CAIV are given by the nasal route and are usually well tolerated in adults and children with runny nose, nasal congestion and coryza being most common. Safety of these live vaccines has also been demonstrated in some high risk or immunosuppressed groups. Shedding of CAIV does occur in recipients of the vaccine and therefore live virus could potentially be transmitted to susceptible contacts. This however occurs infrequently.
CAIV replicate in the upper respiratory tract and generates both systemic and mucosal antibodies from this site. Mucosal antibodies are induced more readily than with the inactivated vaccine, with 85% of children having mucosal responses. Adults have a lower response probably because of multiple episodes of pre vaccination influenza. Both mucosal and systemic antibodies peak at 7 to12 days after vaccination. CAIV has been shown to protect 87% to 95% of children from influenza, with slightly lower rates reported in adults.
Influenza in Animals
Influenza A viruses are known to exist in a wide variety of animal species, while Influenza B exists only in humans and Influenza C occurs in humans and swine. Influenza in animals is therefore largely due to Influenza A viruses. There are sixteen known subtypes of hemagglutinin (H1 to H16) and 9 subtypes of neuraminidase (N1 to N9) present in influenza viruses and all have been found in aquatic birds while only a few subtypes have been found in other mammals. Only H1, H2 and H3 and N1 and N2 have been found in humans and H1, H3, N1 and N2 in pigs. Within in each hemagglutinin and neuraminidase subtype there are multiple different strains of influenza. Some of these strains are exclusive to one species while others can be found in numerous animal species. It is not entirely clear how these different viruses evolved in different mammals but a putative diagram of the interspecies transmission of influenza A viruses is shown below.
Waterfowl and shore birds carrying all hemagglutinin and neuraminidase subtypes are central to Influenza A epidemiology with transmission to most mammalian species.
Although birds are the predominant hosts for influenza, pigs are considered to play an important role in evolution of new strains and act as a link from birds to humans. The respiratory epithelium of pigs has receptors for both bird and human influenza viruses allowing both types to cause infection at once and thus allowing reassortment of RNA fragments and the subsequent appearance of a new virus which may be pathogenic for birds or humans or both.
The barriers to efficient interspecies transmission particularly between birds and humans are not clearly understood. Initially it was theorized that specific hemagglutinin receptors in the upper airways of humans and birds was the primary barrier. Human influenza viruses carry hemagglutinin which binds specifically to sialic acid linked to galactose by a α-2,6 linkage (SA α-2,6) found in the epithelial cells lining the mouth and pharynx of humans. On the other hand avian viruses carry hemagglutinin which binds specifically to sialic acid linked to galactose by a α-2,3 linkage. However we now know that duck, gull and chicken viruses differ in their recognition for the SA α-2,3. Furthermore the finding of both SA α-2,6 and SA α-2,3 in the terminal bronchioles and alveoli of humans suggests the interspecies barrier may be more complex and possibly involves more structures including the mucin covering the respiratory epithelium, the neuraminidase of the virus, sialylated glycan receptors and the PB genes.
Avian Influenza (Bird flu)
Influenza is common in aquatic birds and not often associated with widespread disease and death in these animals. These strains have been referred to as low-pathogenicity avian influenza (LPAI) and these strains occasionally establish themselves in poultry. Sometimes viruses arise in poultry causing a more severe infection, resulting in high mortality. Such viruses have been called high-pathogenicity avian influenza (HPAI). From 1959 to 1990 there were 9 HPAI outbreaks recorded in different parts of the world (Europe, North America and Australia). After 1990 the number of HPAI outbreaks escalated and a further 10 outbreaks were seen including the current H5NI outbreak. In comparison to other HPAI outbreaks the current H5N1 outbreak is unprecedented both in terms of severity and spread and will be dealt with separately (see below). The commercialization and extensive, intensive poultry farming practices as well as the widespread transport of poultry is most likely the reason for the increases in HPAI outbreaks. These outbreaks cause huge economic losses to the poultry industry as well as impacting significantly on subsistence farmers in developing countries.
Swine Influenza (Pig flu)
Influenza in pigs is common throughout the world. Antibody studies indicate that about half of the global pig population has been exposed to influenza. The known swine influenza virus (SIV) strains are found in the influenza A subtypes H1N1, H1N2, H3N1, H3N2 and H2N3. Influenza C which infects man, but not birds also causes influenza in pigs.
Transmission of SIV occurs commonly between pigs (epizootic) and from pig directly to humans (zoonotic) which is uncommon. Neither epizootic nor zoonotic swine influenza is transmitted from human to human. Some influenza virus strains which are originally of swine origin can be transmitted from human to human as seen in the 1918 pandemic and the 2009 pandemic which will be discussed later.
Epizootic swine influenza
Transmission between infected and uninfected animals occurs readily in pigs, probably as a result of touching noses and secretions such as saliva and also via the airborne route by coughing and sneezing. The close proximity of pigs to each other during modern animal farming, coupled with cramped animal transport facilitates spread of the virus within the herd and over long distances. Infection produces fever, loss of appetite, lethargy, coughing, sneezing and shortness of breath. The death rate is low at about 1-4%. The biggest problem with swine influenza is economic loss to farmers due to substantial loss of body weight. Prevention of SIV in pigs is largely through vaccination programs, although the changing subtypes of SIVs are making vaccination less efficient. Regular disinfection of pig facilities readily inactivates virus and is an important component of control. Pigs that carry influenza should not be added to unexposed herds as exposed pigs may carry the virus for up to 3 months.
Zoonotic swine influenza
Transmission of SIV from pigs directly to humans is uncommon and is called zoonotic swine flu. There have only been about 50 recorded cases of zoonotic transmission and these strains rarely pass from human to human. When zoonotic SIV does occur in humans it is usually mild having the same symptoms as human influenza. People who have close and regular contact with pigs such as abattoir works, pig farmers and veterinarians are at greatest risk.
History of swine influenza
The first indication that swine influenza might be related to human influenza was during the 1918 pandemic when pigs developed influenza symptoms at the same time as humans. This virus in pigs was identified as HINI and was shown to be closely related to the 1918 human pandemic H1N1 strain. It is unsure whether the human virus was acquired from pigs or transferred to pigs. However it is known that influenza in pigs was not known before 1918 suggesting a human to swine transfer. Recent phylogentic studies however suggest that reassortment of human and avian influenza in swine might be the genesis of the human 1918 strain. Decedents of the 1918 human H1N1 continue to circulate in humans contributing to the seasonal influenza epidemics. In 1997 a new H3N2 emerged as a reassortment of swine, avian and human genes and is now a major cause of swine influenza in North America. Reassortment between the two SIV, H1N1 and H3N2 produced H1N2. In 2008 a new H1N1 virus appeared in pigs which went on to cause the 2009 H1N1 swine flu pandemic.
Equine Influenza (Horse flu)
Horse flu occurs globally and is caused by H7N7 and H3N8. This is a highly infectious virus and can result in 100% infection in unvaccinated horses and those with no prior exposure to the virus. For this reason most countries have wide scale vaccination programs. Some countries (Australia and New Zealand) are free of horse flu and therefore have strict quarantine programs. The 2007 Australian horse flu outbreak will be dealt with below.
After exposure to infected horses or infected urine and manure in stable waste, there is an incubation period of 1 to 5 days. Horses develop a fever, have a dry hacking cough and have a runny nose. They become extremely lethargic and stand motionless for long periods, reluctant to eat or drink. They usually recover completely in 2-3 weeks.
Records indicate that horse flu has existed for many centuries, although the virus was first isolated in 1956. A large outbreak of equine influenza was well recorded in North America in 1872. This outbreak started in Ontario, Canada in October and by the middle of December had spread across Canada, down to Florida, into Mexico and across to Cuba. The impact on American way of life was significant. The street railway industry ceased, without coal locomotives didn't run, and fires raged unchecked as horse drawn fire carts were inoperable. Cargo from trains and ships could not be transported. Even the war against the Apaches was affected as the cavalry were without mounts and guns and supplies could not be transported.
The 2007 Australian equine influenza outbreak
This influenza outbreak deserves special comment as it illustrates the rapid spread of a very contagious virus in a susceptible population. Furthermore it highlights how stringent containment measure can halt and eradicate an epidemic. Prior to this outbreak Australia was equine influenza free and as a result there was no active vaccination program. The prevention of equine influenza introduction was aimed solely at rigid quarantine measures of imported horses.
Between 17 July and 6 August 2007, Japanese racehorses were stabled on the Island of Hokkaido ready for export. There was and equine influenza outbreak on the island at that time. On 8 August 2007, 13 of these racehorses arrived in Australia with 9 going to Spotswood quarantine station in Victoria and 4 going to Eastern Creek quarantine station west of Sydney. On 17 August one horse developed symptoms of equine influenza (H3N8) at Eastern Creek followed by a second horse on 20 August. On 22 August 2 horses showed symptoms of equine influenza in Centennial Parklands in Sydney, 40 km from the quarantine station. By 25 August there were reports of equine influenza north of Sydney as far as Brisbane. By 10 October up to75000 horses had become infected, mainly along a narrow inland strip that follows a major highway between Sydney and Brisbane. On 9 December the last case of equine influenza due to an H3N8 subtype was reported. After a year of surveillance and no new reported cases Australia was again declared equine influenza free. In retrospect it appears that the virus was carried out of Eastern Creek quarantine station, probably on the boots, clothes or hands of quarantine workers and taken to a major horse show in Maitland which was held between 17 and 19 August. From here the virus was transferred by infected horses and horse products back to Sydney and north to Brisbane. Although most of the transmission was thought to due to direct contact with horses or horse product there was evidence to suggest that some cases, at isolated properties may have acquired equine influenza by windborne spread.
The symptoms were relatively mild and typical of equine influenza, with a low mortality rate and deaths mainly occurring in young foals and immunocompromised horses. There were confirmed cases of transmission to dogs probably as a result of eating infected horse meat. There was no evidence of infection in humans or birds.
The public health response was immediate and very successful and initially involved a national ban on all horse movement and horse events. When it became apparent that the outbreak was restricted to NSW and Queensland these bans were lifted in the other states. Perhaps the most significant intervention was the division of NSW and Queensland into 4 zones depending on the degree of influenza activity in each zone. The Special Restricted Zone (Purple Zone) had the highest number of infected horses and no horse movement was allowed within or in or out of this zone. The Protected Zone (Green Zone) had no cases and free movement of horses was allowed within this zone. Buffer zones were also created around the Purple Zone and an intense vaccination program was instituted in this zone.
The cost of this outbreak was enormous with $400 million being attributed to loss of betting revenue, $100 million for containment and $288 for government assistance packages. This does not take account the unknown costs, such as cost to the horse breeding industry, horse exportation industry and cost to employees.
Canine Influenza (Dog flu)
Influenza in dogs is acquired from horses (H3N8). The source of most outbreaks has been traced to racetracks, particularly where greyhound racing also occurs. The method of transfer is thought to include the aerosol route and ingestion. Once exposed there follows a short incubation period of 2 to 5 days, followed by symptoms varying from cough and nasal discharge to severe secondary bacterial pneumonia which has a 50% mortality rate. H5N1 has also been reported in dogs and is due to eating infected duck.
Influenza pandemics
This pandemic started in June 1917 and continued through till 1920 and spread to every continent. The total number of deaths is not entirely clear but it is estimated that from 50 to 100 million people died, which was about 3% of the global population at the time. Some countries experienced higher death rates than the average (Japan 5%, Indonesia 5% and British Somaliland 7%). About 500 million people were infected which therefore reflects a mortality rate of 10% to 20% of those infected. It has been labeled the greatest infectious disaster in history.
The symptoms of infection were particularly severe, initially commencing as a viral pneumonia but then progressing to involve other major organs. Secondary bacterial pneumonia was also responsible for large proportion of the deaths. Hemorrhage from mucous membranes (nose stomach, intestine) was well described by doctors at the time probably as a result of what we now know as disseminated intravascular coagulation. Evidence indicates that as a consequence of over stimulation of the immune system by the infection, a cytokine storm resulted, contributing significantly to the death rate.
There was a second deadlier wave of the pandemic, which appeared in August of 1918, with cases initially being reported in Sierra Leone, France and the United Sates.
World War 1 undoubtedly contributed to the pandemic. Malnutrition amongst the general population and troops and subsequent decreased immunity led to much higher death rates Furthermore the massive troop movements, the confined quarters of troops and the movement of displaced persons contributed to the rapid and extensive spread of the virus. The number of new cases dropped quickly after the second wave, almost to zero. The sudden cessation of the pandemic is difficult to explain but may be due to a number of variables, including increased immunity of the population (herd immunity), earlier medical diagnosis and better treatment, or mutation of the virus to a less lethal strain.
The geographic origin of the pandemic is not clear. It did not originate in Spain but was called the Spanish Flu. Spain was neutral during WW1 and there was no censorship of reporting the illness and its high mortality rate. The increased coverage from Spain and that King Alfonso became very ill gave the impression that Spain was severely affected.
There has always been speculation about the origins of The Spanish Flu, but recent evidence suggests that the 1918 pandemic strain originated from a H1N1 swine virus which itself was possibly derived from a H1NI avian virus. It is therefore suggested that shortly before the start of the 1918 pandemic H1N1 crossed the species barrier from swine to humans. These results further confirm that swine are necessary for genetic reassortments and adaptation of avian viruses to humans.
The H5N1 Asian Bird Flu
Birds are the central reservoirs for influenza A viruses and all 16 H types and 9 N types are found in birds. Outbreaks of influenza in poultry and birds as a result of low pathogenicity avian influenza (LPAI) and high pathogenicity avian influenza (HPAI) is described above (see Avian Influenza). The current HPAI, H5N1 outbreak which began in 2003 is however unprecedented in terms of bird mortality and spread. It has been seen throughout Asia, Europe and Africa and has caused huge economic losses, particularly in the poultry industry. The combination of a highly pathogenic virus and commercialsed large scale poultry industry has undoubtedly contributed to this. The current H5N1 virus has been transmitted to other mammals (tigers, leopards, dogs and cats) by eating infected birds or poultry and also to humans, with fatal consequences. Attempts have been made to eradicate the virus by massive culling, but this has proved unsuccessful. The reasons for this failure includes the large number of backyard poultry farms for which there is little surveillance; the live poultry markets which amplifies and spreads influenza through birds and bird cages and poultry manure used as fertilizer. Furthermore H5N1 can be maintained in ducks as asymptomatic carriage and continue to be transmitted from these asymptomatic animals.
Origins of H5N1
The current H5N1 virus was first detected in geese in China (Guangdong Province) in 1996. From this point on multiple reassortments of the virus occurred in ducks and poultry, until a specific genotype (genotype Z) emerged in 2003, as the dominant strain in terrestrial poultry in southern China, Vietnam, Thailand, Cambodia and Indonesia These H5N1 viruses continue to undergo mutations and reassortments as the virus spreads around the globe. The 1997 outbreak of bird flu in Hong Kong which resulted in 18 human deaths was due to a separate side branch lineage involving reassortment of the 1996 goose strain and other avian viruses found in quail. A sentinel event in our understanding of H5N1 spread occurred when more than 6000 migratory waterfowl died at Qinghai Lake in Western China in May 2005. This lake is a major breeding ground for migratory birds that may then fly onto Siberia, India and Southeast Asia. The lineage of this virus (clade 2.2) was distinct from that of southern China (clade 2.1). Furthermore the H5NI viruses that spread to Mongolia, Siberia, India, Iraq, Turkey, Central Europe, Italy, Nigeria and Egypt were all similar and closely related to the Qinghai Lake virus. This type of evidence supports the theory that after 2005 the geographic expansion of H5N1 was largely due to bird migration and that amplification occurred regionally by local poultry.
Human H5N1 Infection
As discussed above the 1997 Hong Kong outbreak was the first reported case of human H5N1 infection and death. Live poultry markets were thought to be the source and massive culling followed this outbreak. It was not until February 2003 that the next case of human H5N1 occurred (acquired in China) and subsequently further cases followed the spread in poultry in China, Vietnam, Thailand, Cambodia and Indonesia. Poultry infected with H5N1 become very sick and large numbers of virus particles can be found in infected organs as well as in feces and other bird secretions. Most cases of H5N1 in humans are as direct result of handling, slaughtering, preparing for consumption and eating infected, uncooked poultry. It is believed that the respiratory route is the main route of transmission but the intestinal route is also a possible route.
In spite of the heavy exposure that humans have in China and South East Asia, there is very little H5N1 infection in humans. Furthermore, through serological studies in villages with heavily infected poultry there is very little evidence of even asymptomatic human H5N1 infection. The reasons for this are not clear, but are possibly related to the absence of avian virus receptors in the human pharynx and hence the inability of this virus to bind to pharyngeal receptors; a step essential for initiating infections. Human to human infection is yet to be conclusively proven.
The average age of those infected with H5N1 is 18 years, with a range of 3 months to 75 years. After exposure symptoms start to occur after an incubation period of 2 to 4 days. Initial symptoms are fever, cough and shortness of breath, with chest X-ray showing a bilateral, diffuse or patchy infiltrates typical of a primary viral pneumonia. There has been little evidence of secondary bacterial pneumonia both clinically and at post mortem. The ability of H5N1 to cause pneumonia in a large proportion of those infected appears to be due to the presence of avian receptors in human bronchiolar and alveolar cells. Non respiratory symptoms such as diarrhea, vomiting and abdominal pain are also common. Once pneumonia is present there is often a rapid respiratory progression requiring mechanical ventilation. This is followed by multi-organ failure, including heart and kidneys with death ensuing within 9 days of onset of symptoms. Neurological involvement is unusual. High cytokine levels have been noted in the serum of patients with H5N1, indicating that a cytokine storm may be responsible for death in a large number of cases.
By December 2010 there were 510 recorded cases of human H5N1, with 303 deaths; a staggering 59% mortality rate. The age of those dying is also very different from seasonal influenza; 50% of deaths are less than 20 years and 89% of deaths are less than 40 years. The reasons for the atypical age distribution are uncertain, but may reflect high exposure in the young age group or acquired immunity in the older age group. The pathogenesis of the high mortality rates are not entirely clear, but appear to be a combination of high and prolonged viral replication, wide dissemination of the virus and a cytokine storm.
Diagnosis of H5N1 is not always straightforward. The initial clinical and radiological findings are not specific for H5N1 influenza and a history of close contact with poultry, in a country with known H5N1, should be carefully sought. Even the rapid respiratory progression and multi-organ failure are not unique, but suspicion of H5N1 should be high and treatment initiated before a laboratory diagnosis is made. The best specimens for detection of the virus are nasopharyngeal aspirates and throat and nose swabs. The most rapid tests for detection of human influenza viruses are antigen tests of the respiratory specimens or of urine. However the sensitivity of these specimens is low. Influenza viruses can be isolated by culture in embryonated eggs or in cell lines, but this methodology is time consuming and of little value in the clinical context. PCR is now the method of choice for the detection of all human influenza virus infections. However PCR is only available in sophisticated laboratories where turnaround times of 24 to 48 hours may be achieved. The detection of antibody is again not useful for diagnosis in the clinical setting as both acute and convalescent titers are required for diagnosis. Antibody levels are however invaluable for epidemiological studies.
Treatment
In view of the rapid progression of infection and the severity of H5N1, it is likely that such patients will proceed to the intensive care unit prior to a diagnosis being made. Supportive care with artificial ventilation will probably be required. Hypotension as a result of infection and a cytokine storm may also require blood pressure support. Antiviral therapy should be instituted, although the efficacy of these agents in H5N1 infection has not been clearly defined. Most H5N1 strains from Southeast Asia are resistant to M2 inhibitors, while strains from Eurasia and Africa appear susceptible. Most strains appear susceptible to oseltamivir and zanamivir in vitro as well as showing efficacy in murine models of H5N1 infection. There is no conclusive evidence of the efficacy of neuraminidase inhibitors in patients with confirmed H5N1 infection. The limited clinical evidence however does not suggest a significant clinical impact on the mortality of H5N1 infections. This may be due to a number of factors which include late presentation, the inability of antiviral drugs to modulate the cytokine storm, and the oral route of oseltamivir in severely ill patients with deranged drug absorption. In spite of the absence of efficacy on mortality, there has been some evidence to suggest that there may be a response in some patients, when oseltamivir is started later in the course of infection. The use of inhaled zanamivir in severely ill patients on artificial ventilation is technically difficult; drug distribution is unknown and is therefore not recommended. Finally, the emergence of resistance during oseltamivir therapy has been recognized. As a result of these observations it has been suggested that in H5N1 infection double the dose of oseltamivir be given for a longer duration. Amantadine is not recommended for first line therapy but can be given in combination with oseltamivir if the strains are known.
Infection Control
At this point in time it is believed that H5N1 is poorly transmissible from human to human. But the close contact and heavy exposure required to treat very ill patients with high viral loads in intensive care units closely parallels the exposure seen in poultry workers in South East Asia. Full barrier and airborne precautions are required when managing potential H5N1 patients in hospitals. This will include thorough hand washing, gowns, gloves, face shields and masks designed to prevent viral transmission. Patients should be managed in negative pressure isolation rooms to prevent spread to other wards and patients in the hospital. Although the respiratory route is deemed to be the most likely potential source of transmission the high number of viral particles in feces makes the gastrointestinal tract another potential source of transmission.
If there has been potential transmission to family or hospital staff prior to full isolation being instituted then a prophylactic course of a neuraminidase inhibitor is appropriate. Widespread use of these drugs for potential transmission in casual acquaintances or hospital staff is unwarranted because of the potential for resistance development.
Vaccines
The types of vaccines for influenza are discussed above. There are specific vaccination dilemmas that apply to H5N1. We have already noted that H5N1 has wide antigenic diversity and it will be difficult to know which one of these viruses will become the pandemic virus in humans. Currently the WHO has identified a number of vaccine candidates among the wide array of H5N1 types. Although live attenuated vaccines are now commonly used with seasonal influenza there is a theoretical risk against their use prior to the emergence of a true pandemic strain. Such a live attenuated H5N1 could undergo reassortment with a human influenza strain resulting in a pandemic with an entirely different virus. However a vaccine given during the course of a pandemic should not pose the same problems. Even if a suitable vaccine can be found the production of enough vaccine for the entire world will be impossible with current production methods.
The 2009 H1N1 Swine Flu Pandemic
This pandemic is usually referred to as the swine flu but will be referred to here as pandemic H1N1/09. The outbreak began in the state of Veracruz, Mexico in March 2009, but the first isolation of the virus was from 2 cases in Southern California in mid April 2009. Within 6 weeks of the initial identification the virus had spread to 48 countries with 12515 cases and 91 deaths reported by the World Health Organization (WHO) on 25 May 2009. Thereafter the pandemic continued to spread globally but by November 2009 the number of cases started to decrease and by May 2010 the pandemic was in steep decline. On August 2010 the WHO announced the end of the H1N1 pandemic after it had resulted in the deaths of more than 18000 people.
Origin of the H1N1 Pandemic
At the end of the 1990s a triple reassortment H1N1 had become enzootic in pigs in North America. This swine influenza virus contained genes from avian (PB2 and PA), human (PBI) and swine (H, NP, NS, M, and N) influenza viruses. There were sporadic cases of human infection with this virus but all patients recovered. The pandemic H1N1 virus emerged from this enzootic swine virus when the neuraminidase (N) and matrix proteins(M) were replaced from a Eurasian H1N1 swine influenza virus. Based on genomic studies it appears this pandemic H1N1/09 first evolved around September 2008 and circulated in humans for several months before the first cases were detected. The genetic diversity of pandemic H1N1/09 is low and suggests that transmission to humans resulted from single molecular event.
Transmission
The pandemic H1N1/09 virus sustains relatively high human to human transmission and does not require ongoing contact with swine. The incubation period ranges from 2 to 7 days and viral shedding can occur from 1 day prior to symptoms till symptoms have resolved. The virus is believed to be transmitted through the air by aerosols produced by coughing or sneezing or through contact with saliva, nasal secretions, feces or blood.
Clinical Features
Pandemic H1N1/09 is generally no more severe than seasonal influenza. The age range of those infected is 0 to 93 years, but the mean age is only 19 years. Sixty four percent are aged between 10 to 29 years and only about 1% are over 60 years. This relative sparing of adults older than 60 years is probably due the presence of cross protective antibodies generated by exposure to antigenically similar viruses much earlier in life. The clinical features are similar to seasonal influenza, with cough and fever being the predominant signs. Gastrointestinal symptoms such as vomiting and diarrhea were however more common than with seasonal influenza. As with seasonal influenza, patients with complications and severe influenza were more likely to have underlying medical conditions such as chronic lung disease and cardiovascular problems. However pregnancy and obesity were specific risk factors for severe influenza due to pandemic H1N1/09.
Treatment
As with seasonal influenza, treatment is largely symptomatic, with antiviral therapy only being used in severe cases. As discussed above amantadine resistance is now so widespread in seasonal H3N2 that it can no longer be used. Resistance to the neuraminidase inhibitor oseltamivir is low in seasonal H3N2 (1-4%) and hardly seen with zanamivir. Seasonal H1N1 strains remain susceptible to amantadine but are nearly all resistant to oseltamivir, which is due to a mutation at H274Y in the neuraminidase. There is little resistance to zanamivir in seasonal H1N1. Since the 2009 pandemic seasonal H1N1 appears to have disappeared and is believed to no longer be circulating. In spite of the large amounts of oseltamivir used during the H1N1 pandemic only about 1% of pandemic H1N1 strains show evidence of resistance to oseltamivir with no evidence of zanamivir resistance. The H274Y mutation is also responsible for oseltamivir resistance in pandemic H1N1. Treatment is indicated in all hospitalized patients and those at risk for complications and severe influenza infection.
Vaccines
The first vaccine for pandemic H1N1 was released in October 2009, 6 months after the first isolation of the virus. The vaccine produces a strong immune response that should be protective against pandemic H1N1 infection. The vaccine is safe with no increase in side effect profile compared to seasonal vaccination. The groups targeted to receive the pandemic H1N1 vaccine are similar to those receiving seasonal influenza vaccinations, with the addition of pregnant women and morbidly obese patients. The current seasonal influenza vaccine now also contains components of the pandemic H1N1/09 strain.
Bioterrorism and Influenza
Influenza does not currently appear in the CDC bioterrorism disease category list. However a highly pathogenic influenza virus (H5N1) with high transmissibility could pose an even more deadly bioterrorist weapon that the currently listed category A infections (Anthrax, Botulism, Plague, Smallpox, Tularemia and Hemorrhagic fevers).
Influenza has a number of characteristics that make it a serious bioterrorist threat. Influenza is easily disseminated by aerosols and could be spread by paint sprayer or crop duster. Influenza is easily and rapidly transmissible from person to person. Influenza can be lethal, as demonstrated by the 1918 Spanish flu and the current H5N1 Asian bird flu (see above) which has a mortality rate of 59%. A bioterrorist initiated outbreak would be detected late, as an initial cluster of cases would not prompt an immediate investigation. The short incubation period of 1 to 4 days would ensure rapid spread. Eradication of a deliberately released strain is impossible due to reservoirs becoming established in birds and swine. Influenza is easily accessible by scientists as it is not controlled as are the infectious diseases in category A. With current technology it would be fairly simple to genetically engineer a highly transmissible and lethal influenza strain.
Even though not currently listed by the CDC, the extract below clearly indicates that influenza has been a bioterrorist concern to the CDC for some time.
"While previous influenza pandemics were naturally occurring events, an influenza pandemic could be started with the intentional release of a deliberately altered influenza strain." – Gensheimer KF, Meltzer MI, Postema AS, Strikas RA. Influenza Pandemic Preparedness. CDC Emerging Infectious Diseases Volume 9 No 12 December 2003.