Skip to main content

The Wide World of Vaccines

May 2019

A Success Story: Vaccine Modified to Reduce Febrile Reactions – The Safety Net System Works if We All do Our Part

Christopher J. Harrison, MD, FAAP, FPIDS
 | Director, Infectious Diseases Research Laboratory | Director, Vaccine and Treatment Evaluation Unit | Professor of Pediatrics, UMKC School of Medicine

Anti-vaccine proponents claim safety signals (i.e., adverse effects (AEs)) are taken less seriously by public health/medical professionals and vaccine manufacturers compared to parents. A series of recent events resulted in a publication and show that the system works,1 i.e., a vaccine’s manufacturing process was reviewed, bench laboratory investigations were performed and the manufacturing process was changed. This was followed by a clinical study that showed the revamped manufacturing process effectively and significantly reduced the AE of interest.

Seasonal influenza vaccines do not undergo clinical trials before being put into use. There simply is not sufficient time each year to figure out which strains will likely circulate, produce enough virus, combine three to four virus strains, and produce the millions of doses needed each year. The vaccine batches undergo quality inspections and review, but not clinical testing.

Several years ago, excess high fevers and febrile seizures occurred for one manufacturer’s influenza vaccines used mostly in Australia, fever in ~40% of those <3 years old and 35% in 3-5-year-olds. Similarly, severe fever, axillary temperature >101.3°F/38.5°C, occurred in ~15% of those <3 years old and nearly 10% of 3-5-year-olds. Febrile seizures were also noted in <3 years olds.2-4 These rates mimicked those of whole-virion influenza vaccines of 40 years ago. Introduction of newer split-virion vaccines had dramatically reduced fever and febrile seizure rates. A detergent in concentrations of 0.5 – 2.5% is usually used to split the whole virions into fragments for use in the split vaccines.

This unusually high AE rate was not acceptable and led to a detailed analysis of the data and the hypothesis that something about that year’s vaccine, and perhaps the manufacturing process, led to excess febrile reactions that seemed to mimic systemic innate immune responses.

One early study showed that cells from children who had IIV3-related febrile seizures that year produced increased proinflammatory cytokine/chemokines when incubated with the implicated IIV3, compared with previous IIV3 (had not produced high fever rates). The company then analyzed the vaccine and noted somewhat large lipid fragments in the vaccine. They hypothesized that these fragments were recognized by the innate immune system, leading to cytokine release and febrile reactions.

This makes sense because one method to enhance desired immune responses to vaccines is to use a lipid adjuvant, e.g., the bivalent HPV vaccine and the new zoster vaccine. However, too much lipid or too large size of the lipid vesicles might enhance adverse effects more than desired responses.

With no reliable animal model for influenza vaccine AEs, the company developed a novel in vitro model to report excess innate responses to manufactured vaccine. This reporter-system showed that increasing the amount of detergent (breaking whole virions and their lipid components into smaller fragments) induced much less innate cytokine/chemokine response in the test tube.

The company then made new vaccines using higher detergent concentrations and tested the vaccine in a stepwise fashion, first in older and then younger children. Voila! The febrile AE rate dropped to 7% (2.5% severe) in the younger group and 4.9% (2.1% severe) in the older age group.

This was a combined effort of families and medical professionals to report the AEs, the public health authorities to bring the problem to light, and the manufacturer to investigate the issue and develop a solution, restoring confidence in the vaccine for all stakeholders. It is also evidence that the system works. So let’s all stay vigilant so if problems do arise with any vaccine or any medication, we can report it in order for the system to be able to respond.


1. Modification of the Vaccine Manufacturing Process Improves the Pyrogenicity Profile of Inactivated Influenza Vaccines in Young Children. Sawlwin DC, Jones AG, Albano FR. Vaccine 37 (2019) 2447–2454. 

2. The Safety of Influenza Vaccines in Children: An Institute for Vaccine Safety White Paper. Halsey NA, Talaat KR, et al. Vaccine 2015;33(Suppl 5):F1–F67. 

3. Therapeutic Goods Administration. Investigation into Febrile Reactions in Young Children Following 2010 Seasonal Trivalent Influenza Vaccination. Status Report as at 2 July 2010. Updated 24 September 2010. < au/sites/default/files/alerts-medicine-seasonal-flu-100702.pdf> [accessed 14 February 2019].  

4. Centers for Disease Control and Prevention. Update: Recommendations of the Advisory Committee on Immunization Practices (ACIP) Regarding Use of CSL Seasonal Influenza Vaccine (Afluria) in the United States during 2010–11. MMWR 2010;59:989–92.


Figure. Increased detergent concentration in manufacturing process reduced febrile adverse effects of influenza vaccine.

Wide World of Vaccines Graph.png