221 Investigation of Inpatients with 2009 Influenza A (H1N1) Infection in a New York City Hospital

Friday, March 19, 2010
Grand Hall (Hyatt Regency Atlanta)
Mahlet Tadele, MD , Montefiore Medical Center of Albert Einstein College of Medicine, Bronx, NY
Sharon Leung, MD , Montefiore Medical Center of Albert Einstein College of Medicine, Bronx, NY
Anna Kaltsas, MD , Montefiore Medical Center of Albert Einstein College of Medicine, Bronx, NY
Brian Currie, MD , Montefiore Medical Center of Albert Einstein College of Medicine, Bronx, NY
Background: The epidemiology of severe infection with 2009 pandemic influenza A (H1N1) is under active investigation. Previous reports have identified bacterial coinfection as a risk factor; however there has been no investigation of viral coinfection.

Objective: Characterization of clinical presentation, laboratory findings and outcomes of 31 adult pts with confirmed influenza A (H1N1) infection admitted to a New York City hospital.

Methods: Retrospective chart review of 31 adult pts admitted to a NYC hospital from May 15 to June 22, 2009 with confirmed H1N1 by RT-PCR (NYC Department of Health reference lab) and by positive viral shell vial culture (Montefiore Medical Center laboratory). All cultures were taken at time of admission. All pts were treated with oseltamivir within a few hrs of presentation except one. Data from ICU (n=13) and non-ICU (n=18) was compared.

Results: Mortality and coinfection were only observed among ICU pts (39% and 77% respectively). There was a significant coinfection rate among ICU admitted pts and within 7 days of hospitalization; p=0.000. The odds of death in pts with influenza A (H1N1) with bacterial or viral coinfection compared to pts without coinfection was 25 times higher (adjusted for age and sex). No pre-existing conditions were identified as significant predictors of death or ICU admission, however trend toward obesity was noted with more than half of the pts having BMI > 30. Sensitivity of rapid influenza test in pts with coinfection was very low (45.5% vs. 79% overall sensitivity). There was no significant difference in duration of symptoms between ICU and non ICU pts, and all ICU pts were directly admitted to ICU.

Table 1 ICU vs. non ICU pts with influenza A (H1N1) infection
VariablesICU H1N1Non ICU H1N1OR (95%CI)P value
21-507/1311/18 0.727
>506/137/18 0.727
Death5/130/18 0.008
Male Sex6/134/183.0(0.66-13.5)0.247
Days of symptoms to emergency room (ED)2.462 0.47
Acute renal failure8/131/1823.8(2.8-178)0.003
Co-infections10/130/18 0.000
  Methicillin resistant staph aureus2/13   
  Methicillin sensitive staph aureus 2/13   
  Group A streptococcus 1/13   
  Other virus (adenovirus, parainfluenza) 2/13   
  Coagulase negative staph 1/13   
Gram negative Bacilli 2/13   
Coexisting conditions    
  (asthma, chronic obstructive lung disease, obstructive sleep apnea)7/1310/18 1
  myocardial infarction5/133/18 0.228
  diabetes mellitus type 26/136/18 0.71
  hypertension5/138/18 1
Pneumonia on admission9/135/185.8(1.27-26.8 )0.033
Body mass index (BMI) >306/1310/181.5(0.42-5.69)0.732
Rapid Flu positive6/1318/18 0.001

Conclusions: The high mortality rate in patients with influenza A (H1N1) infection (≈39%) is possibly related to bacterial and viral coinfection. Active surveillance, better initial diagnostic tools for use in critically ill patients, and routine screening of ICU pts for coinfection may identify targets for potential therapeutic intervention.