Domestic Ducks and H5N1 Influenza Epidemic, Thailand

Thaweesak Songserm; Rungroj Jam-on; Numdee Sae-Heng; Noppadol Meemak; Diane J. Hulse-Post; Katharine M. Sturm-Ramirez; Robert G. Webster


Emerging Infectious Diseases. 2006;12(4):575-581. 

In This Article


Detection of Influenza Viruses in Different Duck-raising Systems

Closed High-Biosecurity System. As mentioned earlier, ≈1% of every duck flock was sampled for H5N1 detection before being sent to slaughter. More than 10,000 ducks were tested during the study period. No virologic or serologic evidence of H5N1 virus infection was detected in the birds raised in this closed system in western Thailand, including Nakornpathom and Kanchanaburi provinces, despite cocirculation of H5N1 influenza viruses in other duck-raising systems in the region.

Open House System. Most farms that raised ducks with the open house system are in western Thailand, including the 4 provinces of Nakornpathom, Kanchanaburi, Suphanburi, and Rachaburi. Birds from 17 farms were tested for infection with virus; in birds from 4 (23.5%), infection with the H5N1 virus was detected.

Grazing System. In 28 (45.9%) of the 61 free-range duck flocks tested, infection with H5N1 influenza virus was detected. Investigators studied H5N1 infection in 10 flocks of grazing ducks in Ayuthdhaya, Nakornpathom, and Suphanburi provinces between February and July 2004 to determine the biologic and pathologic features of H5N1 infection in the field ( Table 1 ). No virologic or serologic evidence of H5N1 infection was detected in any of the flocks while they were located in the brooding houses. However, after they were moved outdoors to the rice fields, infection with H5N1 influenza was detected in all 10 flocks; the earliest infection was detected 12 days after the ducks left the brooding houses (flock 3, at 42 days of age). The interval between leaving the brooding houses and detection of H5N1 infection was 12−63 days. Of the 10 flocks, 3 (flocks 2, 8, and 9) showed disease signs; only a few birds (<1%) in each flock were clinically affected. However, the interval between initial detection of H5N1 viruses in the flock and culling was 5−10 days, which supports the contention that most ducks in the flocks showed no disease signs.

Serologic evaluation of the flocks showed that low titers of HI antibody were detected before culling, which indicates that an immune response had already begun without disease signs in most birds. Cloacal virus titers in individual ducks showing disease signs before culling were 2.0−3.8 log10 50% egg infectious dose (EID)50/mL which shows that virus was being shed in feces ( Table 1 ). Similar virus titers were detected in asymptomatic ducks.

Signs of disease in flocks, 2, 8, and 9 were depression, lethargy, cloudy cornea, and blindness. However, no deaths were observed in the 10 days before culling.

Backyard Ducks. Of the backyard poultry, chickens were the most frequently infected; 56% of the chicken flocks tested were positive for H5N1 influenza.[12] Ducks were the second most frequently infected; 27% of backyard duck flocks were positive for H5N1. During the second wave of H5N1 infection of poultry and humans in Thailand (August−November 2004), 47% of backyard duck flocks were H5N1 positive. During this time, scientists realized that most ducks infected with H5N1 were asymptomatic.

Pathologic Features

As previously mentioned, our laboratory received 450 sick, moribund, or dead ducks, which were studied for pathologic features of H5N1 infection. These birds had been raised in the open house system or were from backyard flocks. They exhibited signs of disease such as high fever, dyspnea, depression, and diarrhea, and nervous signs such as ataxia, incoordination, and convulsions (Figure 3A). Most had ocular and nasal discharge accompanied by conjunctivitis; 20%−100% of the birds in each flock from which these ducks originated were dead. All cloacal and tracheal swabs and tissue samples were positive for H5N1 by HI and RT-PCR (results not shown).

Figure 3.

A) A White Cherry Valley duck (Anas platyrhynchos), infected with HPAI H5N1 displays nervous signs, convulsions. B) Histopathologic features of the lung of an HPAI H5N1−infected white Cherry Valley duck; infiltration of inflammatory cells in the lung parenchyma (magnification ×100).

At necropsy, gross lesions were detected, including ecchymotic or petechial hemorrhage of leg and footpad; serous fluid surrounding the heart, pancreas, liver, and abdomen; cyanosis of the oral cavity; and mild pleural effusion. On histopathologic examination, the most striking lesions were found in the lung, with extensive pneumonia and severe pulmonary edema with hyaline material in the alveolar space and slight mononuclear infiltration in the area surrounding congested vessels (Figure 3B). Nonsuppurative encephalitis with perivascular cuffing of mononuclear cells and gliosis were detected in the brains of ducks that displayed nervous signs. Hyaline degeneration and necrosis of myocardium with mononuclear infiltration were detected predominantly in dead ducks from fast-growing breeds such as the Pekin and white Cherry Valley ducks. Necrotizing pancreatis with mononuclear infiltration was detected in all affected ducks. Most affected ducks exhibited focal hepatitis, tubulonephritis, splenic lymphoid depletion or necrosis, and enteritis. Virus antigen was detected by immunohistochemical tests in all organs tested, including trachea, lung, liver, pancreas, rectum, bursa of Fabricius, spleen, brain, heart, and kidney (Figure 4).

Figure 4.

Immunohistochemistry of an HPAI H5N1−infected white Cherry Valley duck (Anas platyrhynchos). The viral antigen is detected in myocardial cells and lymphoid cells (arrow) (A) and renal tubular cells (B) (magnification ×100). The primary antibody used for immunohistochemistry in this study was a mouse anti−avian influenza H5 antibody (Magellan Biotechnology, Chunan, Taiwan).

Experimental Infection of Khaki Campbell Ducks

Because culling of all H5N1−positive ducks was mandated in Thailand, we could not determine the natural outcome of infection in birds raised in the open on rice fields. Therefore, khaki Campbell ducks were experimentally infected with 4 representative H5N1 viruses isolated in Thailand in 2004 and 2005. All animal experiments were performed in biosafety level 3+ facilities. All 4 viruses caused the deaths of infected ducks; however, their degree of death varied ( Table 2 ). The most lethal virus tested was A/duck/Thailand/71.1/2004, which caused death in 10/10 of the infected khaki Campbell ducks, a lethality rate comparable to that previously reported for Mallard ducks.[16] Also tested was a human virus isolated in 2004, A/Thailand/MK2/2004, which resulted in the death of 2/10 khaki Campbell ducks. Of the two 2005 viruses tested, 1 caused very slight disease and resulted in only 1/10 deaths (A/quail/Thailand/551/2005) whereas the other (A/duck/Thailand/144/2005) resulted in 5/10 deaths. Ducks inoculated with A/Thailand/MK2/04 shed virus for the longest period of time (day 10 postinfection), whereas the 2005 virus isolates were shed only until day 8 postinfection. These results indicate that the H5N1 avian viruses recently isolated in Thailand can cause death in khaki Campbell ducks; however, several infected ducks remained completely healthy with no signs of disease throughout the study.

Current Status of Duck Raising in Thailand

As of October 2005, the government of Thailand forbids the practice of raising ducks in open fields and moving grazing ducks from 1 region to another. Farmers who do so are subject to fines and other punishments. Additionally, they receive no compensation if they raise ducks in the open free-range system, and the ducks become infected with H5N1. Farmers were initially compensated for the culling of their ducks. Duck raising is now confined to the high-biosecurity system.

After a lull of almost 1 year, a case of human H5N1 infection was reported in Thailand in October 2005. The report was preceded by the illegal grazing of 3 flocks of 3,000 to 5,000 free-range ducks in rice fields in the area (Kanchanaburi Province). Although no direct contact between the grazing ducks and backyard chickens was known, within 2 weeks of the arrival of the ducks, chickens in the area began dying, and a person who had direct contact with the diseased chickens died of H5N1 infection. Approximately 500 backyard chickens were culled in the village. Sequence analysis of the human isolate and avian isolates (duck and chicken) from this area would be essential to confirm the epidemiologic link between these cases and, coupled with the chronology of events, to assess whether free-grazing ducks were indeed the source of infection for this outbreak.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: