Increased risk of influenza among vaccinated adults who are obese

Abstract

Background: Influenza infects 5-15% of the global population each year, and obesity has been shown to be an independent risk factor for increased influenza-related complications including hospitalization and death. However, the risk of developing influenza or influenza-like illness (ILI) in a vaccinated obese adult population has not been addressed.

Objective: This study evaluated whether obesity was associated with increased risk of influenza and ILI among vaccinated adults.

Subjects and methods: During the 2013-2014 and 2014-2015 influenza seasons, we recruited 1042 subjects to a prospective observational study of trivalent inactivated influenza vaccine (IIV3) in adults. A total of 1022 subjects completed the study. Assessments of relative risk for laboratory confirmed influenza and ILI were determined based on body mass index. Seroconversion and seroprotection rates were determined using prevaccination and 26-35 days post vaccination serum samples. Recruitment criteria for this study were adults 18 years of age and older receiving the seasonal trivalent inactivated influenza vaccine (IIV3) for the years 2013-2014 and 2014-2015. Exclusion criteria were immunosuppressive diseases, use of immunomodulatory or immunosuppressive drugs, acute febrile illness, history of Guillain-Barre syndrome, use of theophylline preparations or use of warfarin.

Results: Among obese, 9.8% had either confirmed influenza or influenza-like-illness compared with 5.1% of healthy weight participants. Compared with vaccinated healthy weight, obese participants had double the risk of developing influenza or ILI (relative risk=2.01, 95% CI 1.12, 3.60, P=0.020). Seroconversion or seroprotection rates were not different between healthy weight and obese adults with influenza or ILI.

Conclusions: Despite robust serological responses, vaccinated obese adults are twice as likely to develop influenza and ILI compared with healthy weight adults. This finding challenges the current standard for correlates of protection, suggesting use of antibody titers to determine vaccine effectiveness in an obese population may provide misleading information.

Figure 1

Influenza HAI and MN antibody titers for influenza confirmed and ILI participants and uninfected controls. Pre and post influenza vaccination HAI titers of participants with influenza confirmed and ILI against (a) A/California/pdm2009, (b) A/Texas/50/2012 and (c) A/Switzerland/9715293/2013. Pre and post vaccination titers were not statistically different among healthy weight (white with open circles), overweight (checkered with black circles) or obese (grey with grey circles) participants. Pre and post influenza vaccination HAI titers in participants with ILI (open circles and grey boxes) and demographically matched uninfected controls (black circles and black boxes) for (d) A/California/pdm2009, (e) A/Texas/50/2012 and (f) A/Switzerland/9715293/2013. Pre and post vaccination titers were not statistically different between ILI participants and non-illness reporting matched controls. Pre and post influenza vaccination HAI (g, h, i) and MN (j,k) titers in participants with confirmed influenza (open circles) and matched uninfected controls (black circles). Pre and post HAI or MN vaccination titers were not statistically different between confirmed influenza participants and matched controls. a,b: n=14 for healthy weight, n=16 for overweight, n=44 obese; c: n=9 healthy weight, n=9 for overweight, n=25 for obese. d, e: n=74; f: n=43, g, h, j, k: n=10; I: n=7.