A Zoonotic Henipavirus in Febrile Patients in China

To the Editor:

Figure 1. Figure 1. Genetic and Epidemiologic Characterization of LayV.

Panel A shows the genome organization of Langya henipavirus (LayV). The diagram is drawn to scale, and the scale bar is shown at the bottom. The abbreviation nt denotes nucleotides. Panel B shows the phylogenetic analysis of LayV. The tree was constructed on the basis of the complete amino acid sequence of RNA-directed RNA polymerase (the L protein) from LayV and other recognized species in the family Paramyxoviridae with the use of the maximum likelihood method. Recognized species in the family Paramyxoviridae were included in the analysis. LayV sequences obtained from humans and shrews are marked with red and blue, respectively. The scale bar denotes the number of nucleotide substitutions per site. Bootstrap testing (1000 replicates) was performed, and the bootstrap values are indicated. CedV denotes Cedar virus, DARV Daeryong virus, GAKV Gamak virus, GhV Ghana virus, HeV Hendra virus, HeV-g2 HeV genotype 2, MojV Mojiang virus, NiV Nipah virus, and SDQD Shandong Qingdao. Panel C shows the provincial location, occupation, sex, and temporal pattern of patients with confirmed LayV infection.

The Hendra virus and the Nipah virus, which belong to the genus henipavirus in the family Paramyxoviridae, are known to infect humans and cause fatal disease; however, other related henipaviruses have been detected in bats, rodents, and shrews.^1-4 During sentinel surveillance of febrile patients with a recent history of animal exposure in eastern China, a phylogenetically distinct henipavirus, named Langya henipavirus (LayV), was identified in a throat swab sample from one patient by means of metagenomic analysis and subsequent virus isolation. The genome of LayV is composed of 18,402 nucleotides with a genome organization that is identical to that of other henipaviruses (Figure 1A).¹ LayV is most phylogenetically related to Mojiang henipavirus, which was discovered in southern China (Figure 1B).³

Subsequent investigation identified 35 patients with acute LayV infection in the Shandong and Henan provinces of China, among whom 26 were infected with LayV only (no other pathogens were present). These 26 patients presented with fever (100% of the patients), fatigue (54%), cough (50%), anorexia (50%), myalgia (46%), nausea (38%), headache (35%), and vomiting (35%), accompanied by abnormalities of thrombocytopenia (35%), leukopenia (54%), and impaired liver (35%) and kidney (8%) function. A serosurvey of domestic animals detected seropositivity in goats (3 of 168 [2%]) and dogs (4 of 79 [5%]). Among 25 species of wild small animals surveyed, LayV RNA was predominantly detected in shrews (71 of 262 [27%]), a finding that suggests that the shrew may be a natural reservoir of LayV. (Additional details of the study are provided in the Supplementary Methods section in the Supplementary Appendix, available with the full text of this letter at NEJM.org.)

Although the current study does not fulfill Koch’s postulates, the following findings from the patients with acute LayV infection suggest that LayV was the cause of febrile illness: LayV was the only potential pathogen detected in 26 of the 35 patients (74%) with acute LayV infection; in paired serum samples that were obtained from 14 patients during the acute and convalescent phases of infection, the IgG titers in 86% of the convalescent-phase samples were 4 times as high as those in the acute-phase samples; viremia was associated with acute LayV infection; and the patients with pneumonia had higher viral loads than those without pneumonia (mean [±SD] log₁₀-transformed copies per milliliter, 7.64±0.98 vs. 4.52±1.13). Although human-to-human transmission has been reported for the Nipah virus,⁵ we found no obvious spatial or temporal aggregation of human cases or the assigned haplotypes on the basis of three common single-nucleotide polymorphisms (Figure 1C). There was no close contact or common exposure history among the patients, which suggests that the infection in the human population may be sporadic. Contact tracing of 9 patients with 15 close-contact family members revealed no close-contact LayV transmission, but our sample size was too small to determine the status of human-to-human transmission for LayV. The potential cross-reaction with Mojiang virus should be assessed to improve serologic testing.

In our study, a newly identified henipavirus of probable animal origin was associated with febrile illness, a finding that warrants further investigation to better understand associated human illness.

Xiao-Ai Zhang, Ph.D.
Hao Li, Ph.D.
Beijing Institute of Microbiology and Epidemiology, Beijing, China

Fa-Chun Jiang, B.S.
Qingdao Municipal Center for Disease Control and Prevention, Qingdao, China

Feng Zhu, Ph.D.
Duke–National University of Singapore Medical School, Singapore, Singapore

Yun-Fa Zhang, B.S.
Jin-Jin Chen, M.Sc.
State Key Laboratory of Pathogens and Biosecurity, Beijing, China

Chee-Wah Tan, Ph.D.
Duke–National University of Singapore Medical School, Singapore, Singapore

Danielle E. Anderson, Ph.D.
Peter Doherty Institute for Infection and Immunity, Melbourne, VIC, Australia

Hang Fan, Ph.D.
Beijing Institute of Microbiology and Epidemiology, Beijing, China

Li-Yan Dong, M.Sc.
Qingdao Municipal Center for Disease Control and Prevention, Qingdao, China

Chang Li, Ph.D.
Changchun Institute of Veterinary Medicine, Changchun, China

Pan-He Zhang, M.Sc.
Yue Li, B.S.
Heng Ding, B.S.
Li-Qun Fang, Ph.D.
Beijing Institute of Microbiology and Epidemiology, Beijing, China
[email protected]

Lin-Fa Wang, Ph.D.
Duke–National University of Singapore Medical School, Singapore, Singapore
[email protected]

Wei Liu, M.D.
Beijing Institute of Microbiology and Epidemiology, Beijing, China
[email protected]