Characteristics of Pandemic H1N1 Influenza in Inpatient and Outpatient Populations

Background: In April 2009, a new strain of influenza was discovered in two children from California. Though clinical features of inpatients with pandemic H1N1 (pH1N1) have previously been reported, questions remain regarding the clinical and epidemiologic features of this pandemic.

Objective: To describe the clinical and epidemiologic characteristics of patients presenting to the Johns Hopkins Hospital (JHH) with influenza A during the first wave of pH1N1.

Methods: The first case of pH1N1 at JHH was reported during the first week of May.  Peak incidence during wave 1 occurred the week of June 26, 2009.  This case series describes all inpatients and outpatients who were diagnosed at JHH with influenza A from June 1, 2009 to August 15, 2009. Using a retrospective chart review approach with a standardized data abstraction form, patient information including demographics, past medical history, presenting symptoms and outcomes was collected by three reviewers.  All testing was performed at JHH using direct fluorescence antibody detection, shell vial culture, and virus culture.  Confirmation of pandemic strains of H1N1 was performed by the Maryland State Health Department. 

Results:  A total of 172 patients (50 inpatients and 122 outpatients ) met inclusion criteria.  133 (77.3%) were African-American and 90 (52.3%) were male.  Patients  5-65 years of age made up 76.7% of all patients.  Asthma was the most common underlying co-morbidity, however 47 patients (27.3%) had no significant past medical history.  Diagnoses of sickle cell anemia and diabetes were significantly associated with admission to the hospital (p=0.007 and p=0.001, respectively).   143 patients (83.1%) met the CDC definition of influenza-like illness (ILI).  Among the 50 inpatients, 17 (34.0%) were admitted to the ICU.  The number of co-morbid conditions did not correlate with admission to the hospital or to the ICU.  Among 37 inpatients who received antivirals >48 hours after symptom onset, 9 (24.3%) developed bacterial pneumonia, 4 (10.8%) developed bacteremia, and 2 (5.4%) died.  Among the 13 inpatients receiving antivirals <48 hours after symptoms onset, 1 (7.7%) developed bacterial pneumonia.  No bacteremia and no death occurred in this group.

Conclusions:   This represents the largest case series of inpatients and outpatients reported to date. Pandemic H1N1 influenza remains a mild illness for most patients.  Sickle cell patients and those with diabetes may be more likely to require inpatient admission, but there were no significant associations between number of co-morbidities and ICU admission.  A significant portion of patients (26.9%) did not meet the CDC definition of ILI, emphasizing the need for clinical judgment in diagnosis.  The difference in outcomes between those who began antiviral treatment <48 hours from symptom onset and those in whom treatment was delayed highlights the importance of prompt antiviral therapy in cases where influenza is suspected.