Jody R. Murph

Cytomegalovirus reinfection in young children

OBJECTIVE To determine the frequency of reinfection with new cytomegalovirus (CMV) strains in children in group child-care environments. METHODS Ninety-two CMV strains isolated serially from children attending child care centers were analyzed. Strains were obtained from 1986 to 1994, from 37 children attending one of six centers in the area of Cedar Rapids and Iowa City, Iowa. The CMV isolates were analyzed by a polymerase chain reaction-based algorithm using primers for the a-sequence, glycoprotein B, and major immediate early (MIE) genes of human CMV. The a-sequence polymerase chain reaction products were compared on the basis of size, and products derived from glycoprotein B and MIE genes were compared according to restriction fragment length polymorphisms. RESULTS Children were between 8 months and 5 years 7 months of age at the time of CMV isolation. The number of isolates ranged from 2 to 6 per child, and the intervals between the first and last CMV isolation ranged from 11 weeks to more than 3 years. At least 7 (19%) of the 37 children had evidence of infection with more than one CMV strain. In six of these children, reinfection with distinct strains was confirmed by analysis of the MIE gene products of sequential CMV strains. CONCLUSIONS Children who attend child care centers, like adults who are immunosuppressed or have multiple sexual partners, are at risk of being reinfected with distinct CMV strains.

Intrauterine Cytomegalovirus Infection and Glycoprotein B Genotypes

Cytomegalovirus (CMV) strains display polymorphisms for the gene encoding glycoprotein B (gB; gpUL55). Recent data suggest that the gB genotype may influence the outcome of acquired CMV infections. To determine whether the gB genotype also contributes to the outcome of intrauterine infection, CMV strains were studied from 56 infants with culture-confirmed intrauterine CMV infections who were born in Iowa or Texas. CMV gB genotypes were compared with the neonatal clinical features and neurodevelopmental outcomes. Fifty-three strains (95%) could be assigned a gB genotype. The overall distribution of genotypes was as follows: type 1, 50%; type 2, 18%; type 3, 23%; and type 4, 4%. Strains with the gB 3 genotype were more common among the Iowa infants (P=.082). The gB 3 genotype was more common among infants with asymptomatic infections (P=.004), but geographic location and ascertainment biases may have accounted for these differences. The gB genotypes did not correlate with the neurodevelopmental outcome of intrauterine infection.

Congenital infections and the nervous system

Despite vaccines, new antimicrobials, and improved hygienic practices, congenital infections remain an important cause of death and long-term neurologic morbidity among infants world-wide. Important agents include Toxoplasma gondii, cytomegalovirus, Treponema pallidum, herpes simplex virus types 1 and 2, and rubella virus. In addition, several other agents, such as the varicella zoster virus, human parvovirus B19, and Borrelia burgdorferi, can potentially infect the fetus and cause adverse fetal outcomes. This article provides an overview of these infectious disorders and outlines current strategies for acute treatment and long-term management.

Cytomegalovirus transmission in a Midwest day care center: Possible relationship to child care practices

The epidemiology of cytomegalovirus (CMV) infection and transmission in a large Iowa day care center was studied. Over the 9 months of the study the overall CMV prevalence rates were 21% to 22%, with rates as high as 71% in toddlers. Titers of CMV in the urine or saliva of infected children were as high as 3 X 10(4) plaque-forming units of CMV per milliliter, similar to titers of CMV observed in some congenitally infected infants. Restriction enzyme analysis of CMV isolates from children in the center demonstrated two major clusters with similar patterns, one among 2- and 3-year-old children and another among infants. The clustering of similar CMV isolates among nonambulatory infants suggests that child care or hygienic practices may contribute to the spread of CMV infection in day care centers. Furthermore, the relatively high prevalence of CMV excretion in this center and the low seropositivity rates to CMV among adults in Iowa suggest that adults in the Midwest who have contact with children in day care centers may be at risk for primary CMV infection.

Infectious diseases and child day care

It is estimated that more than 5.3 million children attend out-of-home child day care in the United States. This includes 2.1 million children who attend approximately 63 000 licensed child day-care centers. An additional 500 000 children receive care in 105 000 regulated day-care homes. Since the total regulated child care slots available in centers and homes are only 2.6 million, some 2.7 million additional children are likely attending unregulated family day-care homes. As a result infants and preschool children are intermingled in child care facilities that often lack adequate toilet and hand-washing facilities and are frequently staffed by individuals with little or no training in the area of infection control. Placing children in out-of-home care should not compromise their health and that of the community. The risk of infection can be lessened by teaching hygiene, supervising unregulated day-care facilities and regular antibiotic use so that bacterial resistance may be prevented.

Polymorphisms within human cytomegalovirus chemokine (UL146/UL147) and cytokine receptor genes (UL144) are not predictive of sequelae in congenitally infected children

Human cytomegalovirus (HCMV) viral chemokine, UL146, and TNF alpha-like receptor UL144 genes show a high degree of hypervariability in clinical isolates. These proteins are predicted to be immune modulators and may contribute to the pathogenesis of HCMV infections. We analyzed the UL146 and UL144 genetic variation of 51 HCMV isolates from congenitally infected children and 13 isolates from children in childcare. There was no statistically significant correlation between UL146 and UL144 genotypes and HCMV disease and/or sequelae. However, there were some groups that had a relatively large proportion of asymptomatic outcomes. These included UL146 group 8 (7/8 asymptomatic) and UL146 group 10 (3/3 asymptomatic). UL144 group B had 11/15 (73%) asymptomatic. UL146 and UL144 genes remained stable in serial isolates from children in daycare for intervals up to three years. These results indicate that most UL146 and UL144 genotypes do not predict clinical sequelae following congenital HCMV infections.

Characterization of Human Cytomegalovirus Strains by Analysis of Short Tandem Repeat Polymorphisms

ABSTRACT Human cytomegalovirus (HCMV) strains display genetic polymorphisms, and these polymorphisms can be analyzed to study viral transmission and pathogenesis. Recently, short tandem repeat (STR) length polymorphisms have been identified in the HCMV genome. We assessed the utility of STRs in characterizing HCMV strains and found that a multiplexed PCR assay using primers based upon these STRs accurately maps HCMV strains. Using primers for 10 microsatellite regions, the STR profiles of 44 wild-type and 2 laboratory strains of HCMV were characterized. The results of STR analysis were compared with those for strain characterization using nucleotide sequencing and restriction fragment length polymorphism analysis. In each instance, STR analysis accurately and specifically identified strains that were indistinguishable or distinct by conventional molecular analysis. Analysis of short tandem repeats also detected polymorphisms that supported simultaneous excretion of two HCMV strains. These results indicate that STR analysis allows rapid, precise molecular characterization of HCMV strains.

Sabin inactivated trivalent poliovirus vaccine: first clinical trial and seroimmunity survey

The most widely used poliovaccine in the United States contains the three live attenuated strains originally produced by Sabin. An inactivated (“killed”) formulation of this trivalent polio vaccine has now been prepared. Before testing this new vaccine, we assessed the polio-virus immune status of 39 healthy adult males between the ages of 20 and 44 years and found that 69% had detectable (titer 1:4) neutralizing antibody to all three types of poliovirus, whereas 31% lacked antibody to 1 or more types even though they had a history of childhood polio immunization. Of interest, the lowest levels of neutralizing antibody were found among young adults in their late 20s, 2 of whom lacked antibody to all 3 polio types. When the Sabin inactivated trivalent poliovirus vaccine was initially administered to 12 seropositive volunteers, all responded with rising titers of neutralizing antibody that persisted for at least 18 months (range, 1:249 to 1:4948). The new vaccine was also given to a second group of 9 individuals with little or no detectable neutralizing antibody to at least one poliovirus type and again all vaccinees manifested a humoral immune response to poliovirus. Except for transient local tenderness at the injection site, no untoward reactions to immunization were observed. Thus, this Phase I study (1) confirmed earlier reports that titers of poliovirus antibody may decline to undetectable levels by early adulthood and (2) demonstrated that adults previously immunized with poliovirus vaccine responded rapidly to all 3 poliovirus types (within 7 days) upon reimmunization with Sabin inactivated trivalent vaccine whether or not there was preexisting detectable antibody.

Survival of cytomegalovirus in paper diapers and saliva

We studied the survival of human cytomegalovirus inoculated experimentally into saliva and urine-soaked paper diapers. Infectious virus could be recovered from saliva maintained at room temperature or 37°C for up to 2 hours after inoculation. Cytomegalovirus survived in paper diapers for periods as long as 48 hours, with the quantity of virus in urine remaining relatively constant for the initial 12 hours. These observations provide further support for the concept that fomites may have a role in the transmission of cytomegalovirus infections.

Risk of cytomegalovirus infection among educators and health care personnel serving disabled children.

To determine the risk of cytomegalovirus (CMV) infection for personnel who provide services to young disabled children, we studied the prevalence of CMV infection among such children and determined the seroconversion rate among exposed personnel. The prevalence of CMV excretion was 9.8% among children aged 0 to 5 years in a University-based outpatient program vs. 3.3% in 3- to 5-year-old children attending community-based preschools. Initial serologic studies of personnel demonstrated no differences in CMV seropositivity rates among staff with occupational child contact vs. staff without such contact (40% (40 of 99) vs. 34% (26 of 77] (P = 0.37). However, 21 of the 31 personnel 40 years and older who had occupational child contact were seropositive vs. 10 of 26 personnel of comparable age who had no occupational child contact (P = 0.026). During a 1-year follow-up, 2 of 86 (2.3%) susceptible personnel seroconverted. Rates were 4.4% (2 of 45) among staff with occupational child contact vs. no seroconversions (0 of 41) for those without (P = 0.27). These results indicate that the risk of CMV infection for personnel who work with disabled children is low. However, we cannot exclude the possibility that there may be a small cumulative risk of CMV infection that may exceed that of adults who do not have occupational contact with children.