On November 3, 2003, a 48-year-old man from the Caribbean sought care at an emergency department in Westchester County, New York, USA, after an episode of near syncopy; a 2–4 week history of feverishness, cough, fatigue, and myalgia; and a 10-pound weight loss over 2 months. He had lived in the United States since 1987 and had no known medical conditions. A month earlier, he had been evaluated at a clinic, and an oral antimicrobial drug was prescribed for possible pneumonia. Eight days before the emergency department admission reported here, he had sought emergency care for unilateral conjunctivitis, eye pain, and blurred vision; the diagnosis was corneal abrasion.
Physical examination on November 3, 2003, found that the patient was afebrile, weak, and mildly tachypneic (respiratory rate 18–26 breaths/minute, room air oxygenation saturation 98%) with bibasilar inspiratory rales. Pertinent laboratory findings included mild anemia and thrombocytopenia (hemoglobin 11.9 g/dL, platelets 107 × 109/L, leukocytes 8.0 × 109 cells/L [52% lymphocytes]), mildly elevated hepatic transaminases (aspartate aminotransferase 116 U/L, alanine aminotransferase 87 U/L), and elevated creatine kinase (1,844 U/L). A chest radiograph showed a right hilar density and left lower lobe infiltrates; computed tomographic scan of the chest and abdomen showed bilateral micronodular opacities with right perihilar infiltrates and lymphadenopathy. The patient was admitted for community-acquired pneumonia and received intravenous gatifloxicin.
A tuberculin skin test was reactive (20-mm induration). HIV ELISA/Western blot test results were positive (HIV test result from 3 years earlier was reportedly negative), and CD4 count was 300 cells/μL. Treatment was switched to rifampin, isoniazid, pyrazinamide, ethambutol, and pyridoxine. Bronchoalveolar lavage (BAL) performed on November 7 yielded influenza A virus by tissue cell culture at the Westchester County Department of Laboratories and Research and was negative for Pneumocystis spp., Legionella spp., and other bacterial or viral pathogens. A second BAL and biopsy performed later during hospitalization to evaluate adenopathy indicated inflammation without definitive pathology. The lower respiratory tract disease improved after 13 days, and the patient was empirically prescribed tuberculosis treatment (directly observed therapy) and discharged while mycobacterial culture results were pending. After 8 weeks, mycobacterial culture of the BAL specimen was negative for Mycobacterium tuberculosis but yielded M. avium complex.
The patient lived in an apartment with his wife and 4 children, none of whom were sick during his illness. He denied recent travel and had not traveled outside the United States for 4 years. He worked in a cafeteria as a dishwasher and handled food and garbage until 1 month before hospitalization. He denied any known risk factors for HIV infection.
The influenza A virus isolate was difficult to grow in culture, reacted minimally with antiserum to hemagglutinin H1, and was sent to the Centers for Disease Control and Prevention (CDC) for further characterization. On March 19, 2004, CDC reported that the influenza isolate, designated A/New York/107/2003, was an LPAI A (H7N2), not subtype H1N1, virus.
An epidemiologic investigation was initiated by the Westchester County Department of Health. During 3 interviews (with a Creole interpreter), the patient denied any exposure to live or dead poultry, wild birds, or bird feces. No live poultry markets or poultry were found on the surrounding property or in the neighborhood.
Serum samples obtained during the patient’s hospitalization on November 5, 2003, and on April 4, 2004, were tested at CDC by microneutralization assay with the LPAI A (H7N2) virus from the patient. The acute-phase serum sample was negative (titer 10), but the convalescent-phase serum sample was positive (influenza [H7N2] virus neutralizing antibody titer 80), indicating seroconversion and evidence of infection with LPAI A (H7N2) virus. A confirmatory Western blot assay detected H7 hemagglutinin-specific antibody in the convalescent-phase serum sample. Testing of paired serum samples by ELISA demonstrated a 16-fold rise in H7 hemagglutinin-specific IgG. Serum samples collected from the patient’s wife and 3 of the children on April 4, 2004, were seronegative for influenza A (H7N2) neutralizing antibodies.