NIH News Release
National Institute of
Allergy and Infectious Diseases

Thursday, October 11, 2001
Laurie K. Doepel
(301) 402-1663

New Study Shifts Thinking About Risk of Rare Bowel Blockage Linked to Rotavirus Vaccine

A study in this week's issue of The Lancet concludes that the risk of intussusception, a bowel obstruction, in infants vaccinated against severe rotavirus diarrhea was substantially lower than previous reports estimated. The new study is the first to examine the overall risk of intussusception in a large population during the nine-month time period when the vaccine, known as Rotashield, was in use.

Lone Simonsen, Ph.D., an epidemiologist with the National Institute of Allergy and Infectious Diseases (NIAID), and her colleagues found an overall decrease in hospitalizations for intussusception among infants less than 1 year old from October 1998 to July 1999, when the vaccine was available, relative to previous comparable time periods going back as far as 1993. They derived their data by examining hospital discharge records from 10 states — home to more than one-fourth of the 3.8 million babies born in the United States each year — in which an estimated 300,000 infants had received the vaccine.

Rotaviruses are the most common cause of severe dehydration and diarrhea in infants and young children, causing some 500,000 physician visits, 50,000 hospitalizations and 20 deaths in the United States each year. In developing countries, rotavirus diarrhea is a leading cause of infant mortality, killing 600,000 to 800,000 infants and young children annually.

After extensive clinical trials proved its safety and efficacy, the first vaccine against severe rotavirus diarrhea, developed by NIAID scientists, was licensed in the United States in August 1998. Two months later, public health officials added the oral Rotashield vaccine to the recommended immunization schedule for healthy infants. The vaccine was to be given in a three-dose series at 2, 4 and 6 months of age. Some 1.5 million doses of vaccine were administered to about 1 million U.S. infants. Hopes were high that it was only a matter of time before the vaccine would reach developing countries, where the need for it is so pressing. But in July 1999, after reports of 15 cases of intussusception among vaccinated children, Rotashield's manufacturer, Wyeth Laboratories, pulled the vaccine off the market. A few months later, U.S. public health officials removed it from the list of recommended vaccines. Though currently licensed by the Food and Drug Administration (FDA), Rotashield is not now available.

Intussusception, diagnosed most often in infancy and childhood, strikes about 2,000 infants (1 in every 1,900) in the United States in the first year of life. Its incidence begins to rise at about 2 to 3 months of life, peaks at 4 to 9 months, and then gradually declines at around 18 months. The condition often resolves by itself, but the blockage can require surgery. In the United States, intussusception is rarely fatal.

After the first reports of intussusception linked to Rotashield vaccination, the Centers for Disease Control and Prevention (CDC) conducted a series of studies to examine the evidence for a temporal association between the two. From those data, CDC initially projected there would be one extra case of intussusception for each 2,500 to 5,000 infants who received the vaccine. Based on this projection, vaccine use was halted. More recently, after further analysis of the data, the CDC has revised the risk magnitude downward by approximately half.

In contrast, the Lancet article reports finding a 4 to 10 percent decline in hospitalizations for intussusception among infants less than 1 year of age during the period of Rotashield use when compared with analogous time periods when the vaccine was not yet available.

The earlier CDC studies, explains Dr. Simonsen, extrapolated from data on the temporal association in the post-vaccination weeks to project the probable total magnitude of risk increase during infancy in the population. "Our study," she says, "looked at what actually occurred during the period of Rotashield use. We directly estimated the overall magnitude of the risk." Although the CDC studies found a strong association between Rotashield vaccination and intussusception in the weeks right after vaccination, the current study suggests that the temporal association did not translate into an overall change in the risk of intussusception hospitalizations during the first year of life.

Dr. Simonsen says these seemingly paradoxical findings can be reconciled if one assumes that the heightened risk just after vaccination was offset by a decreased risk later in infancy for those who had been vaccinated. Vaccination with Rotashield may not have changed the overall risk that an infant would get intussusception, she notes — it may have just nudged the timing of it to a narrower window of time, the couple of weeks after vaccination, the time period focused on in the CDC studies. "Our data agree with those of the CDC in that both show there is an increased occurrence of intussusception shortly after vaccination. But their estimates of the overall magnitude of the problem are not in agreement with what we found. When the observation period is extended beyond the immediate post-vaccination weeks, the increased risk to the infant population disappears."

David M. Morens, M.D., study co-author and NIAID epidemiologist, adds that the two groups both looked at the same problem, but from different angles. "When you look only at whether or not there is an association between vaccination and intussusception — and you zoom your lens in real close to the period of time right after vaccination — you see a small hill of apparent risk. But if you zoom out and look at what happens over a longer period of time, you see that beside that hill is a valley. And the valley is just as deep, or deeper, as the hill is high. What we think could be happening is that there is real risk of intussusception up front, but then a lower risk after that. So it might all even out. And in the end, the vaccine might actually prevent net cases of intussusception."

Dr. Simonsen acknowledges that their study has some limitations. Most notably, they were unable to crosscheck the medical records of the infants to verify their vaccination records and diagnoses of intussusception. But she believes scientists need to explore and fully understand information about the potential benefits and risks of Rotashield. "We want to keep the door open and learn all we can," she says, "for the sake of a fair evaluation of the Rotashield vaccine, and for the sake of any future oral vaccines that may find a very small intussusception risk to be a very large obstacle for licensing and use."

NIAID is a component of the National Institutes of Health (NIH). NIAID supports basic and applied research to prevent, diagnose, and treat infectious and immune-mediated illnesses, including HIV/AIDS and other sexually transmitted diseases, tuberculosis, malaria, autoimmune disorders, asthma and allergies.

Press releases, fact sheets and other NIAID-related materials are available on the NIAID Web site at

Reference: L Simonsen, DM Morens, A Elixhauser, M Gerber, M Van Raden and WC Blackwelder. Effect of rotavirus vaccination programme on trends in admission of infants to hospital for intussusception. The Lancet 358:1224-29 (2001).