Paul A. Offit, MD, Director, Vaccine Education Center at The Children’s Hospital of Philadelphia
On August 30, 2013, the Florida Department of Health was notified of a case of laboratory-confirmed pertussis in a child attending a charter school in a faith healing community in Columbia County (Mattias J, Dusek C, Pritchard PS, et al. Notes from the Field: Outbreak of Pertussis in a School and Religious Community Averse to Health Care Vaccinations—Columbia County, Florida, 2013. MMWR Morb Mortal Wkly Rep. 2014 Aug 1;63(30):655.). The school contained 316 students from pre-kindergarten through 8th grade. Immunization rates were woeful. Only 15 percent of kindergarten students were immunized against pertussis; 5 percent of 7th graders were immunized.
A few days after the first child was sickened by pertussis, his sibling became ill. Then, on Sept. 3, two additional children from the school were confirmed to have pertussis by polymerase-chain reaction. On Sept. 12, the health department declared a communicable disease emergency. Thirty-eight children were subsequently excluded from school. Eventually, 94 children and 16 teachers were infected with pertussis. Fourteen cases, including three in infants, occurred in household contacts.
The outbreak extended into the community; 109 cases were identified. Although the health department offered pertussis vaccine to members of the community at no charge, only five people chose to be vaccinated. Despite the presence of a massive pertussis outbreak, community members continued to eschew medical care in favor of their beliefs.
In a more rational world, this would not be permitted. The United States Supreme Court has made it very clear that while people can martyr themselves to their religion, they should not be allowed to martyr their children to their religion (Prince v. Commonwealth of Massachusetts, 321 U.S. 158.) Also, again in a more rational world, the state would not consider it the inalienable right of any United States citizen to catch and transmit a potentially fatal infection. Someday we may evolve to disallowing people to put their children in harm’s way in the name of their faith. Sadly, we’re not there yet.
Two post-licensure studies have reported the possible association between the quadrivalent human papillomavirus (HPV) vaccine and venous thromboembolism (VTE). VTE is a risk factor for strokes and heart attacks. The first report, from the Vaccine Adverse Event Reporting System or VAERS, suffered from the lack of an appropriate control group. The second report, from the Vaccine Safety DataLink, suffered from small numbers. In addition, it is hard to come up with a biological explanation for why a vaccine that consists of the L1 surface protein from four different strains of HPV (types 6, 11, 16, and 18) would cause something that is not a consequence of natural infection.
To get around the problem of small numbers and inadequate control groups, Nikolai Scheller and coworkers in Denmark looked at the incidence of VTE in HPV recipients (Scheller NM, Pasternak B, Svanström H, Hviid A. Quadrivalent Human Papillomavirus Vaccine and the Risk of Venous Thromboembolism. JAMA. 2014 July;312(2):187-8). Controlling for age, use of anticoagulants, pregnancy, and most importantly, use of contraceptives (which are known risk factor for VTE), these researchers studied 500,345 women who received the HPV vaccine, finding no association between the vaccine and VTE during the 42-days following vaccination.
Clinicians should be reassured that HPV vaccine is not a cause of VTE.
One of the more common questions that we’re asked at the Vaccine Education Center regards the use of rotavirus vaccines in the hospital setting. The problem is this. Currently, rotavirus vaccination is recommended to begin no later than 104 days of age. Because some children will remain in the hospital beyond that age, they may never receive rotavirus vaccine. However, both the bovine-human reassortant rotavirus vaccine (RotaTeq®) and the attenuated human rotavirus strain (RotaRix®) are live viruses raising the possibility of spread from a vaccinated to an unvaccinated child, which could be a particularly severe problem in a vulnerable population of hospitalized infants, some of whom are severely premature.
Heather Monk and colleagues at The Children’s Hospital of Philadelphia evaluated their hospital’s experience with rotavirus vaccine in the Neonatal Intensive Care Unit (NICU) (Monk HM, Motsney AJ, Wade KC. Safety of Rotavirus Vaccine in the NICU. Pediatrics. 2014 Jun; 133(6):e1555-60.) The authors performed a retrospective review of 96 children vaccinated with RotaTeq® and found that none developed symptoms attributable to vaccination within 15 days. Further, 801 infants who were not vaccinated with rotavirus vaccine, but were located near those who had been vaccinated, were also evaluated; again, none developed symptoms consistent with acquisition of vaccine virus. The authors concluded that, “Inpatient administration ensures that age-eligible infants are vaccinated regardless of hospital duration.”
This paper advances our understanding of the potential risks of live, attenuated rotavirus vaccines in a NICU setting, but will be further supported by vaccine-shedding studies in those who did or did not receive vaccine in that setting.
Ten studies have now evaluated whether influenza vaccine administered during pregnancy increases the risk of adverse birth outcomes. (See references 1-10 in the journal article cited.) All have found that the influenza vaccine is safe for the unborn child. One weakness of these studies is the lack of a large cohort of women who were vaccinated during the first trimester.
To address this lack of evidence, James D. Nordin and coworkers evaluated 57,554 matched vaccinated and unvaccinated pregnant women, including 16,240 women who were vaccinated in the first trimester (Nordin JD, Kharbanda EO, Benitez GV, et al. Maternal Influenza Vaccine and Risks for Preterm or Small for Gestational Age Birth. J Pediatr. 2014 May;164(5):1051-7.) Data were collected from seven Vaccine Safety DataLink study sites.
The authors found that influenza vaccine given during any trimester did not increase the risk for either premature or small for gestational age births. Similarly, they did not find that influenza vaccination protected against these adverse outcomes.
Influenza vaccine is recommended for pregnant women because they are at increased risk of severe and occasionally fatal influenza infection when compared with women of the same age who are not pregnant. Still, however, only about one-third of women receive this vaccine. Maternal reluctance might in part be related to fear of an adverse outcome in their unborn child. The study by Nordin and colleagues, in concert with the many other studies that have examined this issue, should help to further allay those fears.
Pre-school children are at risk for severe influenza infections; especially those in child care, where the risk of acquisition is greater. To protect against influenza in this setting, the state of Connecticut passed a law in September 2010 requiring all children between 6 months and 5 years of age who attend licensed preschool or child care facility in the state to receive at least one dose of influenza vaccine each year. Connecticut became the second state – after New Jersey – to implement this policy.
After the regulation was put in place, immunization rates among Connecticut children between 6 months and 5 years of age increased from 67.8 percent during the 2009-2010 season to 84.1 percent during the 2012-2013 season. Among 11 Emerging Infections Program (EIP) sites surveyed, Connecticut had the greatest percentage increase among children ≤ 4 years of age (Hadler JL, Yousey-Hindez K, Kudish K, et al. Impact of requiring influenza vaccination for children in licensed child care or preschool programs—Connecticut, 2012-2013 influenza season. MMWR Morb Mortal Wkly Rep. 2014 Mar 7;63(9):181-5.).
Connecticut’s action lessened the incidence of influenza disease among young children in their state. During the 2007-2008 influenza season, Connecticut was third highest in its incidence of influenza-associated hospitalizations in children ≤ 4 years of age (58.6 per 100,000). During the 2012-2013 season, Connecticut dropped to seventh among states (51.5 per 100,000) and was one of only two states to record a decrease in incidence (12 percent) among children in that age group. Maryland was the other, with a 6 percent decrease.
Not surprisingly, greater immunization rates led to less hospitalization and suffering.
In the 1960s, parents in the United States had only one way to legally withhold vaccines for their children: medical exemptions. Beginning in the 1970s, however, parents pushed for other legislation to exempt their children from vaccines, even if there wasn’t a medical contraindication to receiving them. As a consequence, 41 states have religious exemptions to vaccines and 20 states have philosophical exemptions. The results have been predictable. As has been shown in several studies, states or regions with high numbers of vaccine exemptors are more likely to suffer outbreaks of pertussis and measles. Those with lower rates of exemptors suffer lower rates of disease.
To examine recent legislative efforts to expand or restrict exemptions to vaccination, Saad Omer and colleagues examined bills introduced to state legislations between 2009 and 2012 (Omer SB, Peterson D, Curran EA, Hinman A, Orenstein WA. Legislative Challenges to School Immunization Mandates, 2009-2012. JAMA 2014 Feb 12;311(6):620-1). Thirty bills were introduced into the 30 states that didn’t have a personal belief exemption; all were defeated. Five bills were introduced restricting personal belief exemptions in states that already had them; three passed (Washington, California, and Vermont).
The trend is apparent. One can only assume that state legislators — realizing that personal belief exemptions have only increased the risk of disease, suffering, and hospitalizations in their states — have started to push back on legislation aimed at expanding exemptions and adopt legislation intended to restrict them. Children in these states will clearly benefit from this trend.
In 1999, the simian-human reassortant rotavirus vaccine (RotaShield®) was taken off of the market because it was found to be a rare cause of intussusception, an intestinal blockage. The attributable risk was 10 per 100,000 vaccinees. At the time, it was hard to understand why RotaShield had caused this problem. All existing evidence didn’t support the notion that natural rotavirus infection caused intussusception. So why should a vaccine, which replicates at the intestinal mucosal surface much less efficiently than natural virus, cause a problem that isn’t caused by natural virus?
In 2006, the pentavalent bovine-human reassortant rotavirus vaccine (RotaTeq®) was licensed and recommended for all infants. Since then, about 50 million doses have been administered in the United States. Similarly, in 2008, the monovalent attenuated human rotavirus vaccine (RotaRix®) was also licensed and recommended for universal use; about 10 million doses have now been administered. With millions of children now vaccinated with these two newer rotavirus vaccines, two recent papers have now addressed the question of whether RotaTeq and RotaRix are also rare causes of intussusception.
Yih and coworkers evaluated data from a program called PRISM (Post-Licensure Rapid Immunization Safety Monitoring), which is directed by the Centers for Biologics and Research of the FDA (Yih WK, et al. Intussusception risk after rotavirus vaccination in U. S. infants. New Engl J of Med. 2014 Feb 6;370:503-12). These researchers found that the attributable risk for intussusception following RotaTeq was 1.5 per 100,000 vaccinees (about seven-fold less than that found after RotaShield). The number of children inoculated with RotaRix was too small for adequate evaluation using PRISM.
A second article in the same journal evaluated the risk of intussusception following both RotaTeq and RotaRix using a different data set: the Vaccine Safety DataLink (Weintraub ES, et al. Risk of intussusception after monovalent rotavirus vaccination. New Engl J of Med. 2014 Feb 6;370:513-9). These investigators found that RotaRix was also a rare cause of intussusception with an attributable risk of 5.3 per 100,000 (about half that of RotaShield); RotaTeq was not associated with intussusception in this study.
What do these data teach us? The findings with RotaRix are particularly instructive. RotaRix is a live weakened virus originally obtained from a young boy with rotavirus diarrhea in Cincinnati in 1989. If RotaRix is a rare cause of intussusception, it raises the question of whether natural rotavirus infection is also a rare cause of intussusception. This question can be answered by evaluating the incidence of intussusception in the United States since the introduction of RotaTeq and RotaRix. Because the incidence of natural rotavirus infections has decreased dramatically, we have essentially replaced natural rotavirus with vaccine viruses. The CDC recently presented data showing that the incidence of intussusception has stayed about the same. This means that, in all likelihood, the attributable risk of intussusception following vaccination is about the same as that following natural infection.
If parents are concerned about whether rotavirus vaccination causes intussusception, they should be reassured that it does not appear that rotavirus vaccines have altered the incidence of intussusception in the United States.
In 2009, an unusual side effect was observed in a group of European and Scandinavian people who received a swine flu vaccine that contained the adjuvant, squalene. About 1 of every 55,000 people who received the vaccine developed narcolepsy: a disorder characterized by excessive sleepiness during the day, a sudden loss of muscle tone and strength, and rapid eye movement abnormalities.
Several important facts are known about narcolepsy. It’s caused by a loss of the 70,000 neurons in the hypothalamus that make a peptide called hypocretin, which is responsible for wakefulness. Also, the disease is far more common in people who have a particular genetic background (HLA-DQA1).
Alberto De la Herrán-Arita and colleagues at Stanford University School of Medicine appear to have figured out what happened (De la Herrán-Arita AK, Kornum BR, Mahlios J, et al., CD4+ cell autoimmunity to hypocretin/orexin and cross-reactivity to a 2009 H1N1 influenza A epitope in narcolepsy. Sci Transl Med.2013 Dec 18;5(216):216ra176). The Stanford researchers found a peptide on the influenza hemagglutinin that mimics hypocretin. So when people were immunized with the swine flu vaccine, those with HLA-DQA1 were more likely to develop an immune response (T cells) against their own hypocretin than those who didn’t get the vaccine or those who didn’t have that genetic susceptibility.
More research is required. People in the United States, for example, received an unadjuvanted vaccine and didn’t develop narcolepsy. It would be interesting to know whether the same side effect would have occurred in Europe and Scandinavia if their vaccine had not contained squalene as an adjuvant.
We welcome your input. Please contact us with story ideas, questions or other comments: