The story of avian influenza in Thailand

March 2004

---Kulkanya Chokephaibulkit

On Jan. 13, a previously healthy 6-year-old boy was admitted to our hospital for pneumonia. He had had intermittent upper respiratory tract symptoms since December. The illness had been mild until four days before the admission when he developed a distinct fever spike and shortness of breath.

A pediatrician found rales on physical exam and prescribed ceftriaxone and released him. After three days of ceftriaxone, he seemed to be deteriorating with more tachypnea. A chest X-ray showed right lower lobar pneumonia with some effusion, consistent with bacterial pneumonia. The complete blood count revealed white blood count of 1,200/mm³ (N 44%, L52%, Eo 2%, M2%), and a platelet count of 89,000/mm³. He was admitted with rapidly progressing pneumonia, spreading to the right upper lobe on the next day and to the left upper lobe in the following day despite aggressive antibiotics (imipenem and azithromycin) and supportive care. He was intubated within 12 hours of admission and required ventilator support. Within two days, he developed acute respiratory distress syndrome and required 100% oxygen to keep O₂ saturation above 90%. He also had elevation of AST (790 U/L) and ALT (150 U/L), and a significant persistent proteinuria of 3+ with some red blood cell in the urine, but with a normal BUN and creatinine. Because of persistent leucopenia (nadir of 590 cell/mm³) and thrombocytopenia, the bone marrow was examined and found evidence of maturation arrest and hemophagocytosis. He was put on filgrastim that restored the leukocyte count but not platelets.

Because of the rapidly progressing pneumonia with multiple organ involvement in the healthy child, we added unusual causes like severe acute respiratory syndrome (SARS) to the differential.

We also saw the WHO alert of the human cases of avian influenza (H5N1) in Vietnam on that day. A good friend of mine just told me a few weeks previously that avian influenza was detected in limited poultry farms, but the information was not officially released. We suddenly realized this could be the first case of avian influenza in Thailand. We asked the Ministry of Health to import antivirals just in case it was H5N1 virus. Thailand does not have an antiviral drug available for influenza

Serologic tests were negative for mycoplasma and Bacillus pseudomallei antibody, sputum was negative for acid fast bacillus (AFB) stain, silver stain for PCP and also for bacterial culture, Pneumocystis carinii pneumonia (PPD) skin test was non-reactive, the urine Legionella antigen was negative, and blood and urine culture showed no growth. However, the fluorescent antibody staining of the nasopharyngeal aspirate for respiratory viruses was positive for influenza A and negative for other respiratory viruses, including parainfluenza, adenovirus and respiratory syncytial virus (RSV).

At this point we knew that the patient had a serious form of influenza A, but we never saw such a severe influenza pneumonia in a healthy child. It took the lab another two days to confirm that it was H5N1 virus. This was quite a quick turn around time considering that our lab has never set up the test for avian influenza before.

Additional history revealed that his family and neighbors raise chickens in the backyard and let them run freely in the nearby area around the house. The boy grew up with chickens and likes to play with them. About 500 meters away, his relative just started a chicken farm. It is an open farm with 300 chickens to sell for meat. During the winter months every year, many chickens would die from cold or infections. This year seemed like there were more chicken deaths than usual. All of the hundreds of chickens in his relative’s farm died in early January. Four of the five chickens in his house also died. He also carried one of the sick chickens from the farm back home. No other family members or neighbors came down with the illness like him, but some had a cold.

The child was defervescent after seven days of hospitalization with improvement of other organs involved except for his lungs. He was finally ventilated with a high frequency ventilator with 100% oxygen on day six of admission and had significant insults from barotraumas. As soon as it arrived from Bangkok (day 11), oseltamivir was prescribed for the boy, but by day 13 he died of respiratory failure.

Avian influenza

Usually, birds are asymptomatically infected with influenza A. They shed the virus in feces and secretions contaminating the feeding ground and water. Influenza A can survive many days in water, especially in cold temperatures. At times, by chance of trials and errors on mutations, some viral mutants may be pathogenic, causing death in birds. If the mutant strain is so highly pathogenic that all the susceptible hosts die out, the outbreak will stop from lack of a host to sustain the transmission. Highly pathogenic avian influenzas have a great impact on the economy in that they cause the death of commercial animals or farm animals like chickens and ducks, eg, H5N2 in America 1983, H7N1 in Italy in 1999. Millions of birds were culled to stop the outbreaks that resulted in the loss of millions of dollars.

Not normally transmitted to people

Bird flu usually does not affect people. The previous pandemic from H1N1 (Spanish flu in 1918) and H2N2 (Asian flu in 1957) were from avian virus reassorted with the human gene in pigs. Pigs can be infected by both human and avian virus, and are the animals that live closely with both human and birds in nature. In rural China, pigs are raised under chicken cages, and are fed with chicken feces. Pigs have been the “mixing vessel,” resulting in a new mutant that can infect people. It causes a serious disease if the mutant is also highly pathogenic, like the previous pandemics of H1N1 and H2N2. Efforts were put in place to discourage raising pigs with chickens and ducks.

In 1997, H5N1 was reported in Hong Kong and resulted in 18 human cases with six deaths. It carried the potential for the next pandemic of influenza. To prevent this, 1.5 million chickens were culled in three days. It was unprecedented that we could have averted a potential pandemic by that action. Humans do not have immunity to H5. With it’s high pathogenicity, if spread out, H5N1 may be just like the Spanish flu (H1N1) in 1918 that killed more than 20 million people worldwide, or Asian flu in 1957 (H2N2) that also killed millions of people. Fortunately, the H5N1 virus in Hong Kong seemed to be rarely transmissible from human to human. Less than 2% of the contacts seroconverted, and no symptomatic secondary cases among contacts were reported. This is much better than SARS, which posed a danger to health care workers.

A later molecular study found that the H5N1 isolates in Hong Kong in 1997 were purely avian gene, without human gene. This may be the reason for the ineffective transmission. But also it was the first time that avian virus “jumped” directly from birds to human, without pigs as the mixing vessel. It could be said that some hosts are genetically more susceptible to the avian strain than the normal population. There was a scary possibility that if a person got avian influenza and human influenza in the same time, that person may be a “mixing vessel” for a deadly influenza that could be easily transmitted among humans. Fortunately, H5N1 was stopped in Hong Kong, although two more cases were reported in Hong Kong in 2003. If it happened in other poor countries where the diagnostic labs were not as efficient and the economy could not afford to cull the poultry quickly, it would be like the H5N2 in poultry in Mexico that took more than three years to control and could cost many more human lives.

The spread of a highly pathogenic influenza among poultry was reported in Japan and Korea in December 2003. A large number of poultry were culled to contain the virus. Nobody knew that the virus had already been seeded in many more countries in Asia, just waiting to be reported. The first alert from WHO was on Jan. 13 — the day that we got our first patient — which reported on 14 deaths of atypical pneumonia since October. At least three of these were found to be from H5N1 virus. The deadly virus was found in poultry a week earlier causing a 100% death rate in chickens. The virus was also found in pigs and ducks. Later, the virus isolated from the human cases was found to be purely avian genes without human gene reassortment. The virus had mutated from the isolate from Hong Kong in 1997.

Disease awareness is the key to controlling the disease in humans. Most of the cases occurred before the public became aware of the disease. As of Feb. 5, 2004, there were only five confirmed cases (all died) in Thailand, and 15 confirmed cases in Vietnam (11 died). But there were more than 19 suspected cases of which 10 had died before the appropriate specimen was obtained or the final lab result was reported. All of them had symptom onset in January. After late January, people realized it was dangerous to have contact with sick or dead chickens and no new cases have been reported. The increases in numbers after today will mostly be the result of laboratory confirmation of the current suspected cases.

Fortunately, there has been no secondary cases due to low transmissibility of the current strain. Of 83 health care workers in our center who cared for the patient from the time before the awareness of the diagnosis until he died, 12 came down with a cold. One of these was infected by H3N2 virus.

Strategy

To cope with the problem, we set up a strategy. Our limitations are the resource for diagnostic labs and antiviral drugs. Every Asian country is trying to stockpile oseltamivir or amantadine.

We learned from the previous cases in Hong Kong and in Thailand that in all of the cases of avian influenza patients have come down with fever and pneumonia. And the cases with poorer outcomes are those with old age, pneumonia, leucopenia, and delays in treatment. It has been noted that the antiviral may be helpful if started early, within 48 hours of illness, and oseltamivir is the only one that had evidence of the prevention of lower respiratory tract complications. There have been no studies of antiviral treatment in the later course of illness, nor are there studies of antiviral treatment in cases of severe influenza.

Our strategy starts with screening the patient who deserves a work-up and consideration for treatment. Only those with febrile respiratory tract infection and with significant exposure to sick or dead chickens within the past seven days meet the criteria for further work ups. Those who do not meet the criteria will be followed up with symptomatic treatment only. The work up includes chest X-ray and rapid test by fluorescent monoclonal antibody staining for influenza A or the enzyme immunoassay test for influenza A antigen, whatever is available. Only those with a positive rapid test or with an abnormal chest X-ray will be put on oseltamivir. Further tests to confirm influenza A and typing takes a few more days by PCR and viral culture.

The isolation strategy for infection control is also important. The patients are hospitalized in a special designated ward for strict droplet and contact precautions in a private room that is well equipped, ready for intubation and ventilation. Although the contagiousness of the H5N1 virus is very low, health care workers feel unsure of keeping the patient on the regular ICU. This setting with a separate ward may not be able to be accomplished in most small hospitals in Thailand, and therefore severe cases need to be referred.

Sick personnel are not allowed to care for the patient. We need to do our best not to let the H5N1 virus get into the human host simultaneously with human influenza virus, with the chance of a reassortment and the potential for a hybrid human-bird virus with high transmissibility to humans.

Influenza vaccine for health care workers

We encourage all health care workers to get influenza vaccination. The vaccine currently available with H3N2 and H1N1 cannot protect against H5N1, but can prevent HCW from coming down with influenza A, to spread to other patients specially the patient with avian influenza. We must prevent the chance of humans becoming the “mixing vessel” for the virus, either in the health care personnel or in the patient.

Although the past experience showed a rare chance of human-to-human transmission of H5N1, we follow-up all the personnel that have been exposed to the patients. Anyone who comes down with a respiratory illness must be checked for influenza. Of the 83 health care workers exposed to the above patient, 12 came down with a cold or flu-like symptoms. One of those had the H3N2 virus. All have been well after seven days of the last exposure to the patient.

We are also aware of the interim guidelines to enhance the surveillance and control of avian influenza in the United States, published on Feb. 3, 2004 at www.cdc.gov/flu/han020302.htm. The infection control precautions are similar to that instituted for SARS. Although we did not follow the guidelines for airborne precaution, we avoided using aerosol-generating procedures around the patient.

There are four antiviral agents available to treat influenza. Amantadine and rimantadine are the old drugs that inhibit uncoating steps of the virus by blocking ion channel activity of the viral M2 protein. Both are effective against influenza A, but not B. The therapeutic efficacy was not significant in most studies and may induce resistance rapidly. They have a much better efficacy of around 70%-90% for prophylaxis. The bad news is that the isolates from Vietnam were found to have genotypic resistance to amantadine and rimantadine.

The new antiviral class, neuraminidase inhibitors (NI), gain more attention for the treatment of influenza. NI are sialic acid analogues, the target site for cleavage by neuraminidase before the progeny virus can be released from the infected cells. They are effective for both influenza A and B. There are two drugs in this class, oseltamivir in oral tablet, and zanimivir (Relenza, GlaxoSmithKline) in inhalation form. In uncomplicated cases with treatment initiation within 48 hours of illness, both reduced symptoms by approximately 1.5 days. However, oseltamivir reduced influenza associated lower respiratory tract complication, hospitalization, and antibiotic used. Both NI are also effective for prophylaxis with the efficacy of around 80%. Oseltamivir is well tolerated with some mild gastrointestinal adverse events, and zanimivir may induce bronchospasm.

However, none of the antiviral drugs had evidence of efficacy in treatment of severe influenza, especially when used for symptom onset beyond 48 hours, or in preventing serious complications. In most developing countries, antiviral agents for influenza are not available and quite expensive. With this outbreak, many countries stockpiled the drugs causing the shortage for those that needed them. Antiviral prophylaxis is particularly important for the cullers or those with farmers with sick chickens.

What we need to do next

The first thing to do is to create public awareness of the disease and educate about prevention. Culling birds and poultry that may contract the disease is the best way to stop the chain of transmission. So far, all the cases of H5N1 in humans were from direct and indirect contact with sick or dead chickens. No other bird has been reported to transmit H5N1 to human although the possibility is there. The H5N1 virus was also found in some dead birds in Thailand. Protective devices are important for cullers and those who are exposed to the carcasses. The public panic has had such a strong impact that few new cases have been reported after the culling was initiated.

We need to watch the virus for its evolution. Sophisticated laboratories are needed for the task. We need to eliminate any chance that would offer the virus to reassort to be a new strain with virulence and transmissibility. Recent reports found H5N1 in pigs in Vietnam. This is of concern because pigs can be very effective mixing vessels for the human and avian virus.

We need to work hand in hand with other countries to share the information and facilities. An effective collaboration will help push forward the research for diagnosis, treatment and prevention of this deadly virus. Vaccine development needs resources and good collaboration.

The lesson learned

Open information is needed, especially from authorities. We need more communication between veterinarians and physicians. The earlier we identify the problem, the fewer chickens are culled. A good surveillance system and labs for both human and bird testing are needed. We need a good rapid test for H5N1 to be used in humans and birds. Chicken farming should be set up in a closed system. This is an opportunity to improve the standard of chicken farming. Finally, we need more understanding of the virus and its evolution to find a way to prevent a new outbreak. Next time, we may face an even more deadly and highly contagious virus for humans.

Editor’s note: This editorial ran in Infectious Disease News’ sister publication, Infectious Diseases in Children in March. The editorial was written by Dr. Chokephaibulkit at the request of Philip Brunell, MD, IDC chief medical editor.

For more information:

• Update: Isolation of avian influenza A (H5N1) viruses from human beings — Hong Kong, 1997-1998. Ann Emerg Med. 1998;31:647-648.
• CDC. Update: isolation of avian influenza A (H5N1) viruses from humans — Hong Kong, 1997-1998. MMWR. 1998;46:1245-1247.
• Webster RG. Influenza: an emerging disease. Emerg Infect Dis. 1998;4:436-441.
• Yuen KY, Chan PKS, Peiris M, et al. Clinical features and rapid viral diagnosis of human disease associated with avian influenza A H5N1 virus. Lancet. 1998;351:467-471.
• Uyeki TM. Influenza diagnosis and treatment in children: a review of studies on clinically useful tests and antiviral treatment for influenza. Pediatr Infect Dis J. 2003;22:164-177.
• Kaiser L, Wat C, Mills T, et al. Impact of oseltamivir treatment on influenza-related lower respiratory tract complications and hospitalizations. Arch Intern Med. 2003;163:1667-1672.
• Hayden FG, Atmar RL, Schilling M, et al. Use of the selective oral neuraminidase inhibitor oseltamivir to prevent influenza. N Engl J Med. 1999;341:1336-1343.
• Monto AS, Pichichero ME, Blanckenberg SJ, et al. Zanamivir prophylaxis: an effective strategy for the prevention of influenza types A and B within households. J Infect Dis. 2002;186:1582-1588.

• Kulkanya Chokephaibulkit is associate professor of pediatrics in the Department of Pediatrics at Siriraj Hospital at Mahidol University, Bangkok, Thailand.