CHAPTER 104: MUMPS

Mumps is an illness characterized by acute-onset unilateral or bilateral tender, self-limited swelling of the parotid or other salivary gland(s) that lasts at least 2 days and has no other apparent cause.

Mumps is caused by a paramyxovirus with a negative-strand, nonsegmented RNA genome of 15,384 bases encoding at least 8 proteins: the nucleo- (N), phospho- (P), V, matrix (M), fusion (F), small hydrophobic (SH), hemagglutinin-neuraminidase (HN), and large (L) proteins. The N, P, and L proteins together provide the polymerase activity responsible for genome transcription and replication. The viral genome is surrounded by a host cell–derived lipid bilayer envelope containing the M, F, SH, and HN proteins. The M protein is involved in viral assembly, whereas the HN and F proteins are responsible for cell attachment and entry and are the major targets of virus-neutralizing antibody. The V and SH proteins are accessory proteins, acting as antagonists of the host antiviral response; the former interferes with the interferon response and the latter with the tumor necrosis factor α (TNF-α)–mediated apoptotic signaling pathway. Because of the hypervariability of the SH gene, its nucleotide sequence is used to “genotype” the virus for molecular epidemiologic purposes. Thus far, 12 mumps virus genotypes have been assigned by SH gene sequence and are designated A–N (with the exclusion of E and M, which have been merged with genotypes C and K, respectively).

Nucleotide sequencing of clinical isolates shows that virus genotypes D and G circulate predominantly in the Western Hemisphere; genotypes F, C, and I in the Asia–Pacific region; and genotypes B, H, J, and K in the Southern Hemisphere (Fig. 104-1). Although numerous mumps virus genotypes have been identified and some vary antigenically from others, only one serotype exists, and there is no evidence to suggest that certain circulating virus strains are more virulent or contagious than others.

Mumps is endemic worldwide, with epidemics every 3–5 years in unvaccinated populations. These epidemics typically occur in locations where children and young adults congregate, such as schools, military barracks, and other institutions. In countries without national mumps vaccination programs, the estimated annual global incidence is 100–1000 cases per 100,000 population. After the introduction of mumps vaccine in the United States in 1967, the number of reported cases declined dramatically. By 2001, fewer than 300 cases were reported, representing a 99.8% reduction from prevaccine-era levels. Mumps incidence remained at historic lows in the United States until 2006, when 6584 cases were reported—the largest outbreak since 1987. At the time of the 2006 outbreak, the disease was resurging globally, even in populations with high-level vaccination coverage. The number of reported U.S. cases declined precipitously in the 2 years that followed but then spiked in 2009–2010, with focal outbreaks in New York and New Jersey, and again in 2011, with a focal outbreak in California. A recent study by the Centers for Disease Control and Prevention (CDC) showed that two-dose coverage with measles–mumps–rubella (MMR) vaccine in major U.S. cities (94.8%) remains at or very near the level needed to contain these childhood infections; however, focal areas with inadequate vaccination coverage still leave some children at risk. Sporadic, large-scale mumps outbreaks continue to be reported worldwide, sometimes in countries where the disease was once under control.

Although historically a disease of unvaccinated children, with the largest proportion of cases occurring in children 5–9 years of age during the prevaccine era, mumps now frequently occurs in older age groups—primarily college students, most of whom were vaccinated in early childhood. This shift in age distribution and the occurrence of mumps in vaccinated populations are probably the result of several coincident circumstances, including (1) situations promoting the spread of respiratory viruses among young adults (e.g., residence in college dormitories), (2) waning of vaccine immunity with time, (3) lack of endemically circulating wild-type virus to periodically boost vaccine-induced immune responses, and (4) continuing global epidemics (due to either lack of mumps vaccination programs or, where such programs do exist, low rates of mumps vaccination). The notable decline in mumps vaccination–induced immunity with time may be due to both declining titers and decreasing avidity of antibodies. The waning of mumps immunity over time is supported by studies suggesting that a third dose of MMR vaccine significantly reduces the mumps attack rate; however, these studies were not adequately controlled to rule out the possibility that the observed declines in mumps incidence were unrelated to the intervention. Therefore, the effectiveness of a third dose of MMR vaccine remains to be demonstrated.

Humans are the only natural hosts for mumps virus infection. The incubation period of mumps is ~19 days (range, 7–23 days). The virus is transmitted by the respiratory route via droplets, saliva, and fomites. Mumps virus is typically shed from 1 week before to 1 week after symptom onset, although this window appears to be narrower in vaccinated individuals. Persons are most contagious 1–2 days before onset of clinical symptoms. Inference from related respiratory diseases and animal studies indicates that primary replication likely occurs in the nasal mucosa or upper respiratory mucosal epithelium. Mononuclear cells and cells within regional lymph nodes can become infected; such infection facilitates the development of viremia and poses a risk for a wide array of acute inflammatory reactions. Classic sites of mumps virus replication include the salivary glands, testes, pancreas, ovaries, mammary glands, and central nervous system (CNS).

Little is known about the pathology of mumps since the disease is rarely fatal. The virus replicates well in glandular epithelium, but classic parotitis is not a necessary component of mumps infection. Affected glands contain perivascular and interstitial mononuclear cell infiltrates and exhibit hemorrhage with prominent edema. Necrosis of acinar and epithelial duct cells is evident in the salivary glands and in the germinal epithelium of the seminiferous tubules of the testes. The virus probably enters cerebrospinal fluid (CSF) through the choroid plexus or via transiting mononuclear cells during plasma viremia. Although relevant data are limited, typical mumps encephalitis appears to be secondary to respiratory spread and is probably a parainfectious process, as suggested by perivenous demyelination, perivascular mononuclear cell inflammation, and relative sparing of neurons. Although rare, presumed primary encephalitis has been associated with mumps virus isolation from brain tissue. Evidence of placental and intrauterine spread in pregnancy has been found in both early and late gestation.

Up to half of mumps virus infections are asymptomatic or lead to nonspecific respiratory symptoms. Inapparent infections are more common in adults than in children. The prodrome of mumps consists of low-grade fever, malaise, myalgia, headache, and anorexia. Mumps parotitis—acute-onset unilateral or bilateral swelling of the parotid or other salivary glands that lasts >2 days and has no other apparent cause—develops in 70–90% of symptomatic infections, usually within 24 h of prodromal symptoms but sometimes as long as 1 week thereafter. Parotitis is generally bilateral, although the two sides may not be involved synchronously. Unilateral involvement is documented in about one-third of cases. Swelling of the parotid is accompanied by tenderness and obliteration of the space between the earlobe and the angle of the mandible (Figs. 104-2 and 104-3). The patient frequently reports an earache and finds it difficult to eat, swallow, or talk. The orifice of the parotid duct is commonly red and swollen. The submaxillary and sublingual glands are involved less often than the parotid gland and are almost never involved alone. Glandular swelling increases for a few days and then gradually subsides, disappearing within 1 week. Recurrent sialadenitis is a rare sequela of mumps parotitis. In ~6% of mumps cases, obstruction of lymphatic drainage secondary to bilateral salivary gland swelling may lead to presternal pitting edema, associated often with submandibular adenitis and rarely with the more life-threatening supraglottic edema.

Epididymo-orchitis is the next most common manifestation of mumps, developing in 15–30% of cases in postpubertal males, with bilateral involvement in 10–30% of those cases. Orchitis, accompanied by fever, typically occurs during the first week of parotitis but can develop up to 6 weeks after parotitis or in its absence. The testis is painful and tender and can be enlarged to several times its normal size; this condition usually resolves within 1 week. Testicular atrophy develops in one-half of affected men. Sterility after mumps is rare, although subfertility is estimated to occur in 13% of cases of unilateral orchitis and in 30–87% of cases of bilateral orchitis. Oophoritis occurs in ~5% of women with mumps and may be associated with lower abdominal pain and vomiting but has only rarely been associated with sterility or premature menopause. Mumps infection in postpubertal women may also present with mastitis.

Documented CSF pleocytosis indicates that mumps virus invades the CNS in ~50% of cases; however, symptomatic CNS disease, typically in the form of aseptic meningitis, occurs in <10% of cases, with a male predominance. CNS symptoms of aseptic meningitis (e.g., stiff neck, headache, and drowsiness) appear ~5 days after parotitis and also occur often in the absence of parotid involvement. Within the first 24 h polymorphonuclear leukocytes may predominate in CSF (1000–2000 cells/μL), but by the second day nearly all the cells are lymphocytes. The glucose level in CSF may be low and the protein concentration high, a pattern reminiscent of bacterial meningitis. Mumps meningitis is a self-limited manifestation without significant risk of death or long-term sequelae. Cranial nerve palsies have occasionally led to permanent sequelae, particularly deafness. The reported incidence of mumps-associated hearing loss varies between 1 in 1000 and 1 in 100,000. In ~0.1% of infections, mumps virus may cause encephalitis, which presents as high fever with marked changes in the level of consciousness, seizures, and focal neurologic symptoms. Electroencephalographic abnormalities may be seen. Permanent sequelae are sometimes identified in survivors, and adult infections more commonly have poor outcomes than do pediatric infections. The mortality rate associated with mumps encephalitis is ~1.5%. Other CNS problems occasionally associated with mumps include cerebellar ataxia, facial palsy, transverse myelitis, hydrocephalus, Guillain-Barré syndrome, flaccid paralysis, and behavioral changes.

Mumps pancreatitis, which may present as abdominal pain, occurs in ~4% of infections but is difficult to diagnose because an elevated serum amylase level can be associated with either parotitis or pancreatitis. An etiologic association of mumps virus and juvenile diabetes mellitus remains controversial. Myocarditis and endocardial fibroelastosis are rare and self-limited but may represent severe complications of mumps infection; however, mumps-associated electrocardiographic abnormalities have been reported in up to 15% of cases. Other unusual complications include thyroiditis, nephritis, arthritis, hepatic disease, keratouveitis, and thrombocytopenic purpura. Abnormal renal function is common, but severe, life-threatening nephritis is rare. It remains at issue whether an excessive number of spontaneous abortions are associated with gestational mumps. Mumps in pregnancy does not appear to lead to premature birth, low birth weight, or fetal malformations.

During a mumps outbreak, the diagnosis is made easily in patients with parotitis and a history of recent exposure; however, when disease incidence is low, other causes of parotitis should be considered and laboratory testing is required for case confirmation. Infectious causes of parotitis include other viruses (e.g., HIV, coxsackievirus, parainfluenza virus type 3, influenza A virus, Epstein-Barr virus, adenovirus, parvovirus B19, lymphocytic choriomeningitis virus, human herpesvirus 6), gram-positive bacteria, atypical mycobacteria, and Bartonella species. Rarely, other gram-negative or anaerobic bacteria are associated with parotitis. Parotitis can also develop in the setting of sarcoidosis, Sjögren’s syndrome, Mikulicz’s syndrome, Parinaud’s syndrome, uremia, diabetes mellitus, laundry starch ingestion, malnutrition, cirrhosis, and some drug treatments. Unilateral parotitis can be caused by ductal obstruction, cysts, and tumors. In the absence of parotitis or other salivary gland enlargement, symptoms of other visceral organ and/or CNS involvement may predominate, and a laboratory diagnosis is required. Other entities should be considered when manifestations consistent with mumps appear in organs other than the parotid. Testicular torsion may produce a painful scrotal mass resembling that seen in mumps orchitis. Other viruses (e.g., enteroviruses) may cause aseptic meningitis that is clinically indistinguishable from that due to mumps virus.

Laboratory diagnosis is primarily based on detection of viral RNA by reverse-transcriptase polymerase chain reaction (RT-PCR) or on serology. Detection of viral antigens (e.g., via mumps virus–specific immunofluorescent staining of cultured clinical specimens) is comparatively inefficient and is no longer commonly performed.

For RT-PCR-based testing, viral RNA can be extracted either directly from clinical samples or from cell cultures incubated with clinical samples. Buccal swabs appear to be the best specimens for virus detection, particularly when obtained within 2 days of clinical onset; however, mumps virus can also be detected readily in throat swabs and saliva and, in cases of meningitis, in CSF. Despite the apparent high frequency of viremia during mumps, mumps virus has rarely been detected in blood. The ability to detect viral RNA in clinical samples rapidly diminishes beyond the first week after symptom onset, and in several studies rates of virus detection were substantially lower in recipients of two vaccine doses than in unvaccinated persons or recipients of one dose. The rate of false-negative RT-PCR findings can be quite high, approaching 70% in some studies.

A serologic diagnosis of mumps is typically made by enzyme-linked immunosorbent assay (ELISA). The data must be interpreted with caution. In vaccinated persons with mumps, IgM is typically absent; thus, a negative IgM result in a vaccinated person does not rule out mumps. In addition, regardless of vaccination status, IgM may not be detectable if serum is assayed too early (prior to day 3 of symptom onset) or too late (beyond 6 weeks after symptom onset) in the course of disease. Reliance on a rise in IgG titer in paired acute- and convalescent-phase sera also is problematic: IgG titers in convalescent-phase sera may be only nominally greater than those in acute-phase sera. Thus, at present, the capacity of RNA or viral antigen detection to confirm cases is much greater than that of serologic testing. Traditional and labor-intensive serologic tests such as complement fixation, hemagglutination inhibition, and virus neutralization are now performed only rarely. The main downside to replacement of these functional serologic assays with the more rapid ELISA method is the latter’s detection of all virus-specific antibodies, including those that are nonneutralizing (i.e., nonprotective). Thus, an individual who is seropositive by ELISA may lack protective levels of antibody. While there is a strong association between the presence of mumps virus neutralizing antibody and protection from disease, an absolute antibody titer predictive of serologic protection is lacking; in this respect, mumps differs from other respiratory infections, such as measles.

Vaccination is the only practical control measure. Nearly all developed countries use mumps-containing vaccines, but in many countries mumps is not a notifiable disease and vaccination is often voluntary. However, where used, mumps vaccination has had a tremendous impact, with reductions in incidence and morbidity typically exceeding 90%. Despite the tremendous success of mumps vaccination programs, large mumps outbreaks continue to occur globally, even in settings of high-level two-dose vaccine coverage. Whereas outbreaks historically involved young (often unvaccinated) children in primary and secondary schools, more recent outbreaks have predominantly involved young adults, particularly on college and university campuses. While primary and secondary (waning-immunity) vaccine failures have been hypothesized to be factors in mumps outbreaks in several countries, in some countries other factors may have played a role, such as lack of compliance with the recommended vaccine schedule, changes to vaccination schedules resulting in missed cohorts, or changes in population demographics, such as large-scale immigration.

In the United States, the benefit-cost ratios for mumps vaccination alone are >13 for direct costs (e.g., medical expenses) and >24 for societal costs (including productivity losses for patients and caregivers). Several mumps virus vaccines are used throughout the world; in the United States, only the live attenuated Jeryl Lynn strain is used. Current recommendations are that mumps vaccine be administered as part of the combined trivalent measles-mumps-rubella vaccine (M-M-R^® II) or the quadrivalent measles-mumps-rubella-varicella vaccine (ProQuad^®). Monovalent vaccine is no longer produced for the U.S. market but is available in other countries.

Before administering mumps-containing vaccine, physicians should always consult the latest recommendations from the Advisory Committee on Immunization Practices (ACIP). Current recommendations for children specify two doses of mumps-containing vaccine: the first dose given on or after the first birthday and the second dose administered no earlier than 28 days after the first. In the United States, children often receive the second dose between the ages of 4 and 6 years.

In 2009, the ACIP revised its recommendations for evidence of mumps immunity in health care personnel to include (1) documented administration of two doses of a preparation containing live mumps vaccine, (2) laboratory evidence of immunity or laboratory confirmation of disease, or (3) birth date before 1957. For unvaccinated health care personnel born before 1957 who lack laboratory evidence of mumps immunity or laboratory confirmation of mumps, health care facilities should consider two doses of MMR vaccine separated by the appropriate interval; during a mumps outbreak, vaccination of these individuals is recommended.

Mumps vaccine contains live attenuated virus. It is not recommended for pregnant women, for individuals who have had a life-threatening allergic reaction to components of the vaccine, or for people in settings of clinically significant primary or secondary immunosuppression. (For details, see the ACIP guidelines on the CDC’s website [www.cdc.gov/vaccines/pubs/acip-list.htm].) Occasionally, febrile reactions and parotitis have been reported soon after mumps vaccination. Allergic reactions after vaccination (e.g., rash and pruritus) are uncommon and are usually mild and self-limited. More serious complications, such as aseptic meningitis, have been causally associated with certain vaccine strains but not with the Jeryl Lynn strain.

Immunity to mumps is associated with the development of neutralizing antibody, although a specific correlate of protection has not been established. Seroconversion occurs in ~95% of recipients of the Jeryl Lynn strain; however, vaccine efficacy is ~80% for one dose and 90% for two doses. Recent data indicate declining seropositivity rates with time since vaccination. Studies are under way to assess the value of including a third dose in the immunization schedule. Although it is generally accepted that mumps virus is serologically monotypic, antigenic differences between virus isolates have been detected. It is unclear whether such differences can lead to immune escape. The role of the cellular arm of the immune response is unclear, but there is evidence that it may help limit virus spread and complications.

Acknowledgment

The authors thank and acknowledge Dr. Anne Gershon, the author of this chapter in earlier editions.