New Exposure Location for Hantavirus Pulmonary Syndrome Case

California, USA, 2018

Anne M. Kjemtrup; Sharon Messenger; Amy M. Meza; Tina Feiszli; Melissa Hardstone Yoshimizu; Kerry Padgett; Sunita Singh


Emerging Infectious Diseases. 2019;25(10):1962-1964. 

In This Article

The Study

The adult patient (age 35–40 years) sought care at a local urgent-care facility in April 2018 for a 2-day history of fever (38.8°C), chills, muscle aches, nausea, dizziness, shortness of breath, and cough. Initially diagnosed with a viral illness, the patient returned 2 days later with worsening symptoms, including nuchal rigidity and photophobia. The patient was referred to a local emergency department because of concerns about meningitis.

At hospital admission, a computed tomography scan of the lungs showed bilateral perihilar air space consolidation, increased lung density (ground glass), and moderate bilateral pleural effusion. Clinical laboratory values demonstrated thrombocytopenia (≈46,000 cells/mL [reference 150,000–400,000 cells/mL]), elevated creatinine (1.5 mg/dL [reference 0.6–1.3 mg/dL]), elevated leukocytes (12,400 cells/mL [reference 3,900–11,700 cells/mL]), neutropenia (8.9% [reference 45%–70%]), and lymphopenia (17.3% [reference 18%–48%]). One day later, the patient was transferred to a hospital with a higher acute care level, intubated, and placed on extracorporeal membrane oxygenation.

Serum collected 4 days after illness onset tested negative at the local public health laboratory for influenza A and B. At a commercial laboratory, serum test results were urine antigen–negative for leptospirosis and legionella, antibody-negative for coccidioidomycosis, and PCR-negative for Yersinia enterocolitica but were IgM- and IgG-positive for hantavirus. The California Department of Public Health Viral and Rickettsial Disease Laboratory tested an additional serum sample collected 7 days after illness onset, confirmed the positive serologic results, and detected SNV RNA by using a previously described RT-PCR.[7] The patient remained on extracorporeal membrane oxygenation for 10 days, was extubated on day 18, and was released 20 days after illness onset.

The case-patient lives and works on a farm at ≈20 m elevation in Santa Cruz County, along the north coast of California. An interview with the family initially suggested a rural work exposure in San Mateo County, north of the farm residence, at ≈400 m elevation, where both the case-patient and a family member worked outdoors in a dusty, rodent-infested environment 18 days before illness onset. The family did not recall a substantial rodent exposure on the farm except for the case-patient cleaning a shed ≥1 week before illness onset.

The California Department of Public Health Vector-Borne Disease Section collaborated with county vector-control agencies to evaluate the case-patient's place of residence, farm, and rural workplace for potential exposure to SNV. At the farm, rodent access, feces, and nesting material were present in multiple outbuildings and structures in and around the home. Of 105 Sherman traps set, 19 rodents were captured (18% trap success) from the farm, including 18 deer mice and 1 Western harvest mouse (Reithrodontomys megalotis). Rodents were anesthetized, bled through a retro-orbital blood collection technique, and humanely euthanized. Five (28%) of the deer mice and the harvest mouse were serologically positive for SNV, including 1 deer mouse from inside the shed that the case-patient cleaned and 1 from the basement of the house. Blood from 4 of the 5 deer mice and the harvest mouse were positive for SNV by RT-PCR.

The rural San Mateo County workplace location could not be investigated directly; however, trapping and habitat evaluation were conducted at public areas near the worksite. Rodents captured in 35 of 100 traps (35% trap success) included 15 parasitic mice (P. californicus) and 20 piñon mice (P. truei) but no deer mice. One piñon mouse tested serologically positive for SNV, but no viral RNA was detected by RT-PCR.

We conducted phylogenetic analysis to compare the case-patient's isolate to other California hantavirus sequences, including those from the farm where the case-patient lived and worked. Because no PCR-positive rodents were collected near the rural worksite, archived sequences from SNV-positive deer mice collected in previous years (2014, 2016, and 2018) from 2 different sites in the same county as the rural worksite (San Mateo County) were included in our analysis. We found that the SNV glycoprotein sequence from the case-patient was genetically related most closely to the hantavirus sequences recovered from the case-patient's farm (Figure). The sequences from the 2 sites in San Mateo County each form separate monophyletic clades that cluster together, despite collections over several years, and are distinct from all samples from Santa Cruz County. Thus, exposure most likely occurred at the farm where the case-patient lived and worked. Although the type of exposure of opening poorly ventilated outbuildings and performing activities that raise dust is typical for hantavirus exposure, the geographic location in this coastal California county has not been previously implicated in SNV exposure leading to HPS. Follow-up visits by county vectorborne disease officials provided information to the family on rodent exclusion and other prevention measures to reduce the risk for subsequent exposure to SNV.


Phylogenetic tree of hantavirus Gn glycoprotein sequences from isolates collected in California, USA, and reference sequences. The hantavirus sequence from the case-patient described in this study (gray box) is shown in comparison to sequences from the case-patient farm in Santa Cruz County and archived samples from neighboring San Mateo County (bold). Dotted lines indicate general geographic origins of California sequences. Representative reference sequences of hantaviruses were downloaded from Genbank (accession numbers included in taxon labels). H indicates sequences from human cases; all other sequences are from small rodents. The tree was reconstructed by analysis of 848 bases of the glycoprotein precursor gene by using the neighbor-joining method and employing the Hasegawa-Kishino-Yano model to estimate genetic distances with Geneious 10.0 ( We estimated support for relationships by using a nonparametric bootstrap analysis (1,000 replicates). Nodes with bootstrap percentages >50% are indicated. Similar tree topologies were generated from maximum-likelihood (RAxML) and Bayesian (Mr. Bayes) phylogenetic analyses (data not shown). Genbank accession numbers: Case H-18-0783, MK386451; Pm-18-0062, Santa Cruz, MK386452; Pm-18-0064, Santa Cruz, MK386453; Pm-18-0070, Santa Cruz, MK386454; Pm-18-0072, Santa Cruz, MK386455; Pm-14-0146, San Mateo, MK386456; Pm-14-0147, San Mateo, MK386457; Pm-14-0152, San Mateo, MK386458; Pm-14-0155, San Mateo, MK386459; Pm-16-0128, San Mateo, MK386460; Pm-16-0328, San Mateo, MK386461; Pm-18-0080, San Mateo, MK386462; and Pm-18-0085, San Mateo, MK386463. Scale bars represent the genetic distance (nucleotide substitutions per site).