Mumps outbreak 2006: Evaluating vaccine efficacy complicated by disease’s characteristics

June 2006

Between Jan. 1 and May 18, 11 states have reported 2,597 cases of mumps.

Eight states, including Illinois, Iowa, Kansas, Missouri, Nebraska, Pennsylvania, South Dakota and Wisconsin, reported mumps outbreaks with ongoing local clusters of cases; and three states, Colorado, Minnesota and Mississippi, reported cases associated with travel from an outbreak state, according to a recent Morbidity and Mortality Weekly Report (MMWR).

Most mumps cases, 1,487, were reported in Iowa. Other states with high case tallies included Kansas (371), Illinois (224), Nebraska (201), and Wisconsin (176).

According to the MMWR, “of the 2,597 cases reported overall, 1,275 (49%) were classified as confirmed, 915 (35%) as probable, and 287 (11%) as suspect; for 120 (5%) cases, classification was unknown.” Eleven of the potentially infected patients had traveled on 33 flights, exposing more than 500 passengers to mumps.

The greatest number of cases was in 18- to 24-years-olds, but the age range of patients spanned from 1-year to 96-years-old. Twenty percent of the total in Iowa attended college. Twenty-eight of the 363 male patients developed orchitis, and four of the 1,254 patients had encephalitis.

The sparing of younger people would suggest that there might be waning vaccine immunity. Alternatively, the younger people may have been more likely to get two doses of vaccine, which was recommended in 1989.

One would expect those older than age 17, who were less likely to receive two doses of vaccine, would be more vulnerable. However, 42 of 154 (51%) affected individuals did report receiving two doses. In an outbreak at a summer camp in New York State, 12 out of 31 cases (39%) received two doses. (MMWR. 2006;55(07);175).

Although the vaccine was first licensed in 1967, it was not recommended until more than a decade later, which may have resulted in many of the older group not getting vaccine. However, only 74 patients in the Iowa outbreak reported that they had not received vaccine.

Thus there is concern that the immunity conferred by the vaccine may have waned. In highly-immunized school populations, there was a tendency for greater risk with increasing interval since immunization (Arch Ped Adol Med. 1995;149:774). In a second less well-immunized population, waning immunity was not noted. (J Infect Dis. 1994;169;77) In yet another study, researchers noted two doses conferred five times the protection of a single dose. (J Pediatr. 1991;119;187).

Natural vs. vaccine protection

It is clear that the antibody response following natural infection far exceeds that which follows vaccination and that immune individuals are boosted following exposure [Figure 1] (N Engl J Med. 1968;279;1357). With the widespread use of vaccine and decreased opportunities for boosting, perhaps, there is waning immunity. The epidemiologic evidence, however, is not definitive.

Initial estimates of vaccine efficacy are difficult to reconcile with the observation that in one of the recent outbreaks, the vaccine efficacy was only about 80% (Arch Ped Adol Med. 1995;149:774; J Ped. 1999:119;187). Under experimental conditions, monovalent mumps vaccine was reported to have had an efficacy of about 95% (N Engl J Med. 1968;278:227).

Our own evaluation of the antibody response to this vaccine was consistent with this estimate but we also found that assessing antibody response subsequent to the four-week customary interval following vaccination resulted in a somewhat greater seroconversion rate. Five of nine patients who had not seroconverted at week four had converted at eight months post-immunization (Am J Dis Child. 1969;118;435).

In a study evaluating the seropositivity of children two- to 19-months post immunization with a single dose of measles-mumps-rubella (MMR), 98.3% of the more than 300 infants tested at a variety of sites in the community were seropositive (JAMA. 1983;250;1409). If we assume that about 2% of children failed a single dose and about 4 million are immunized annually, 80,000 susceptibles would be added each year to those with no immunity to mumps.

Given that the vaccine was not recommended routinely until about 10 to 15 years post-licensure, one can readily accept how the introduction of mumps virus could result in outbreaks.

What is disconcerting is that so many of the cases in the Iowa epidemic had received two doses!

Accumulations of susceptibles as may occur in schools or colleges are like tinder waiting to be ignited. Last summer, there was an outbreak in a summer camp in New York that apparently was introduced in the United States by a British camp counselor. The United Kingdom has had a massive mumps outbreak in recent years (MMWR. 2006; 55(07);173).

Many of those who are infected with mumps and do not develop obvious parotitis may shed virus and are contagious [Figure 2]. What is more, virus can be isolated from the pharynx two days before the onset of parotitis [Figure 3]. Thus, isolation of cases of parotitis in an outbreak will not check the spread of disease but might slow the spread. (N Engl J Med. 1968:279;1357).

Similarly, if one tries to evaluate vaccine efficacy on the basis of cases of parotitis without laboratory testing of those who are exposed or those who do not develop parotid swelling, one will miss the subclinical, potentially infectious cases. If one assumes that subclinical cases are the same in the unvaccinated and the vaccinated, then presumably this error will cancel out in the calculation of efficacy. That is if one invokes Brunell’s Law of Multiple Errors, ie, if you make enough they will cancel one another out.

Although as a rule of thumb, one assumes that those born prior to 1957 are probably immune and do not require immunization (JAMA. 1967;199;113). Thus more of these older adults may be susceptible than they are for more contagious illnesses, eg, rubella and measles. Adult mumps cases are not pleasant. Spontaneous abortion following infection during the first trimester in a pregnant woman, pancreaitis, mastitis, ophoritis and orchitis are some of the complications. In an epidemic, however, immunizing health care workers born prior to 1957 is recommended. CDC soon will issue new recommendations.

In examining patients for parotitis, it is important to recognize that the uncinate lobe extends below and behind the ear lobe and this may be a useful physical finding. In addition, a good presumptive test is the serum amylase, but we do not know the specificity of the test.

Isolation of mumps virus from respiratory secretions and/or serologic testing are recommended for confirmation of cases.

As is true of other vaccines without written records, we are uncertain of the number and type, monovalent or MMR, the individual received, so interpreting vaccine evaluations is difficult. It also is difficult to evaluate foreign efficacy data where the Urabi rather than the Jerl Lynn strain of mumps vaccine is used.

Immunizing children against childhood diseases clearly has cut down on the morbidity caused in childhood, but the concern has been that we may be delaying the onset of some of these diseases. This might occur as a result of declining immunity, or because of the continuing threat of importations. Thus far we have done very well and those of us who had seen these diseases before these vaccines can be well satisfied of the result and perhaps a bit disappointed as we no longer will be needed to identify them.

For more information:

• CDC. Measles, mumps, and rubella – Vaccine Use and strategies for elimination of measles, rubella, and congenital rubella syndrome and control of mumps: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR. 1998;47(RR-8);1-57.