| |
Vertically Acquired HCV: natural history
|
| |
| |
"Three Broad Modalities in the Natural History of Vertically Acquired Hepatitis C Virus Infection"
Clinical Infectious Diseases, July 1, 2005
The European Paediatric Hepatitis C Virus Networka
.....To the best of our knowledge, this is the largest group of children with vertical HCV infection described to date....We confirmed the low prevalence of clinical manifestations and the lack of any significant disease, and, therefore, focused on markers of infection....... Our findings suggest 3 broad categories of children with vertical HCV infection: these findings suggest that -20% of children may clear their infection, -50% will develop chronic asymptomatic infection (with intermittent viremia, usually normal ALT levels, and rare hepatomegaly), and -30% will develop chronic, active infection (with persistent viremia, frequent abnormal ALT levels, and, in some cases, hepatomegaly—"abnormal enlargement of the liver")...... likely indicating liver inflammation and an early stage of HCV-related liver damage... In our study, clearance occurred by 3 years of age in most children, a finding similar to that of a study of hemophilia patients who usually cleared the virus 1-2 years after infection... Among the 26 HCV-HIV-coinfected children, hepatomegaly and persistent viremia were common....
Can Markers of Infection in the First Year of Life Predict Viremia at Later Ages?
Children with a low proportion of positive PCR test results in the first year of life were significantly more likely to clear the virus than were children with a high proportion of positive PCR test results (27 children [36.5%] vs. 4 children [5.6%]; OR, 9.77; 95% CI, 2.92-32.67; P < .0001). Similarly, 24 (28.9%) of the 83 children with a low proportion of abnormal ALT levels in the first year of life cleared the virus, compared with only 1 (12.5%) of 8 children with a high proportion of abnormal ALT levels, although this difference was not statistically significant because of the small numbers involved (OR, 2.85; 95% CI, 0.32-24.99; P = .32).
Compared with children with a low proportion of positive PCR test results in the first year of life, children with a high proportion of positive PCR tests were significantly more likely to have a high proportion of positive PCR test results in years 2 and 3 (adjusted OR, 3.59; 95% CI, 1.31-9.83; P = .01) and in the longer term (i.e., in year 2 and later) (adjusted OR, 2.92; 95% CI, 1.09-7.80; P = .03), adjusting for HIV status and the proportion of abnormal ALT levels in the first year.
INTRODUCTION
With the escalating hepatitis C epidemic [1], an increasing number of women infected with hepatitis C virus (HCV) are giving birth across Europe. Although the risk of vertical (i.e., mother-to-child) transmission is low (-5%) [2], little is known about disease progression in children who do become infected. Spontaneous viral clearance, reported in adults, may occur among vertically infected children [3, 4], but this has not been reliably quantified. Accurate estimates of the clearance rate are needed to inform the management of cases of HCV infection in infants.
In adults, the course of HCV infection is slowly progressive, and many individuals are asymptomatic for decades [5]. Regardless of whether the natural history of adult HCV infection can be extrapolated to children, it is likely that serious HCV symptoms may not occur for many years in vertically infected children. This is borne out by anecdotal reports, clinical experience, and the limited research conducted to date, which indicates that clinical symptoms are rare among vertically infected children in the first 5 years of life [3, 6].
In the absence of clinical manifestations, laboratory markers of infection, such as viremia and abnormal alanine aminotransferase (ALT) levels, could be indicative of progression of infection and precede clinical symptoms. Although nonspecific, hepatomegaly can be an early indicator of liver damage and HCV-related disease progression. We have previously shown that intermittent viremia is common in vertically infected children early in life [3], but the relevance of different patterns of viremia and other markers over time as predictors of disease progression is not well understood because of a lack of prospective data for a sufficient number of children.
To further elucidate patterns of markers of vertically acquired HCV infection in subjects from birth to 16 years of age, we used data from a prospective study of HCV-infected pregnant women and their children in the health care centers of the European Paediatric HCV Network (EPHN) [7].
ABSTRACT
Background. Little is known about the natural history of vertically acquired hepatitis C virus (HCV) infection.
Methods. We performed a large, multicenter, prospective study of children born to HCV-infected women in Europe. Children were considered to be infected on the basis of ⩾2 polymerase chain reaction (PCR) test results positive for HCV RNA and/or test results positive for anti-HCV antibody >18 months after birth.
Results. Two hundred sixty-six children with vertical HCV infection were followed up until a median of 4.2 years of age (range, 3.2 months to 15.9 years of age).
Twenty-six children were coinfected with human immunodeficiency virus.
Hepatomegaly, the only clinical sign reported, was found in 10% of children and was significantly associated with a high proportion of abnormal alanine transaminase (ALT) levels (adjusted odds ratio [OR], 4.17; 95% confidence interval [CI], 1.67-10.42; P = .002). In multivariable analysis, children with a high proportion of abnormal ALT levels were -4 times more likely to have hepatomegaly than were those with a low proportion of abnormal ALT levels.
ALT levels. One hundred eighty-seven of 240 children had at least 1 ALT level determined during follow-up; the median number of ALT levels determined was 5 (range, 1-16 levels). ALT levels peaked in the first 2 years of life and decreased thereafter (figure 2). One hundred forty children had at least 1 ALT level >40 IU/L, and 82 children had at least 1 ALT level >80 IU/L; 49 children (26%) had abnormal ALT levels (>40 IU/L) each time that they were determined, and 47 (25%) had normal ALT levels each time they were determined. Sixty-two children (approximately one-third) had an abnormal ALT level >75% of the time.
Of the 26 children with HCV-HIV coinfection, 7 (27%) had hepatomegaly; of these, 6 also had splenomegaly (no other cases of splenomegaly were reported). ALT levels were 10-223 IU/L, with 5 children having abnormal results (ALT level, >40 IU/L) more than 75% of the time. The children with coinfection tended to have a higher proportion of positive PCR test results than did children with only HCV infection, although this was not statistically significant (table 1). Three children initially had PCR results negative for HCV RNA and became PCR positive at 2.2 months, 3.1 months, and 6.4 years of age. Two of 26 children (7.7%; 95% CI, 0.9%-25.1%) fit our definition of HCV clearance.
An estimated 21%-25% of children may have cleared the virus (i.e., had 2 consecutive PCR test results negative for HCV RNA, normal ALT levels, and no clinical signs) at a median age of 14.9 months.
A high proportion of positive PCR test results obtained in the first year of life was associated with a lower likelihood of clearance (OR, 9.77; 95% CI, 2.92-32.67; P < .0001) and persistent viremia in children >1 year old (adjusted OR, 2.92; 95% CI, 1.09-7.80; P = .03).
Ten children were treated with IFN at a median age of 5.4 years (range, 22 months to 11.7 years); of these, 3 achieved a sustained viral response. The analyses described above were repeated after excluding the children with HCV-HIV coinfection and the 10 children who were treated with IFN, with similar results.
Conclusions. We confirm the low prevalence of HCV-related clinical signs and symptoms among vertically infected children in the first 10-15 years of life. Approximately 20% of children appear to clear the infection, 50% have evidence of chronic asymptomatic infection, and 30% have evidence of chronic active infection. Although viremia and abnormal ALT levels were associated with hepatomegaly, further investigation is necessary before these markers can be used in the clinical management of HCV infection in children.
MORE RESULTS
Children with HCV Infection Only
Clinical symptoms.
Fifty-three (22%) of the 240 children with HCV infection only developed at least 1 clinical sign or symptom during follow-up. The most common sign or symptom was hepatomegaly, reported in 25 (10.4%) of the children at least once. Splenomegaly was reported for 2 of these 25 children at the same visits and was not reported for any other children. Of 240 children with a total time at risk of 771.2 child-years, an estimated 13% will have developed hepatomegaly by 5 years of age, and 28% will have developed it by 10 years of age. The estimated median age at first report of hepatomegaly was 7.1 months.
HCV genotype.
Of the 113 children (47%) for whom HCV genotype was determined, approximately one-half (56 children) had HCV genotype 1 and one-quarter (27 children) had HCV genotype 3.
Patterns of viremia.
To account for variation in testing frequency, the proportion of PCR tests with positive results was calculated as a summary variable. Of children infected with HCV only, 105 (44%) had persistently positive results, and 17 (7%) had persistently negative results. Of the 118 children (49%) with both positive and negative PCR results, 35 (15%) initially had negative PCR results and developed persistently positive results at a median age of 3.1 months (range, 26 days to 10.6 years).
Clearance of viremia.
Children in group 2 were observed, in accordance with the study protocol, only from a median of 2.4 years of age. Because of this, clearance is reported separately by group.
In group 1, 32 (21%) of the 155 children fit our definition of clearance of viremia (95% CI, 15%-29%). An estimated 17% of the children would have experienced clearance of viremia by 2 years of age, and 24% would have experienced clearance of viremia by 3 years of age. Among those experiencing clearance of viremia, the estimated median age at the time of the first negative PCR test result was 14.9 months. For 26 of the 32 children who cleared viremia, negative antibody test results confirmed the negative PCR test results; the median age at the time of the first negative antibody test result was 12.6 months (range, 3.8 months-3.0 years). The remaining 6 children were antibody positive when last tested at 1.5-5.3 years of age. In addition, 6 children were PCR negative at the last 2 tests but not classed as having cleared viremia because ALT results were unavailable at these visits. If we assume that ALT levels in these children, had they been tested, would have been normal, the maximum estimate of clearance would be 25% (38 of 155 children) (95% CI, 18%-32%).
In group 2, 3 of the 85 older children fit our definition of clearance. With the exclusion of 1 child who was treated with IFN, our estimate of clearance was 2 (2.4%) of 85 children (95% CI, 0.3%-8.2%). This finding is in line with the finding that 4 children in group 1 (2.6%) apparently cleared their infection after 3 years.
Association of Hepatomegaly and Markers of Infection
We investigated whether the extent of viremia and abnormal ALT levels were associated with the occurrence of hepatomegaly. All 266 children were included, with adjustment for HIV status and group in multivariable analyses. A binary viremia variable was created to split the data at the median; ⩽75% of PCR test results positive was considered to be a low proportion of positive test results, and >75% of PCR test results positive was considered to be a high proportion of positive test results. A similar binary variable was created to represent a low (⩽75%) and a high (>75%) proportion of ALT levels >40 IU/L.
ALT category and HIV status were the only variables significantly associated with hepatomegaly. In multivariable analysis, children with a high proportion of abnormal ALT levels were -4 times more likely to have hepatomegaly than were those with a low proportion of abnormal ALT levels. Although hepatomegaly was significantly associated with a high proportion of positive PCR test results in univariable analysis, this effect did not remain in multivariable analysis. Adjusting for the number of PCR test results did not substantially alter the OR estimates.
For the 240 children with only HCV infection, we also examined the association of viremia and ALT levels with occurrence of hepatomegaly at each visit, allowing for repeated measurements for each child. The model that included both viremia and ALT levels showed a significant variation among the children's linear predictors and a strong within-child correlation (intraclass correlation coefficient, 0.81; 95% CI, 0.64-0.91; likelihood ratio χ2, 71.75; P < .001), suggesting that clinical and biological evidence of active infection tend to occur at the same time. The OR for the effect of positive PCR test results on hepatomegaly, adjusted for ALT level, was 10.94 (95% CI, 0.55-218.64; P = .12), the nonsignificant result probably reflecting a lack of statistical power. ALT level was not associated with hepatomegaly in this model; the OR for each unit increase in ALT level was 0.996 (95% CI, 0.98-1.01; P = .56). Adjusting for age did not alter the OR estimates.
In summary, these findings suggest that -20% of children may clear their infection, -50% will develop chronic asymptomatic infection (with intermittent viremia, usually normal ALT levels, and rare hepatomegaly), and -30% will develop chronic, active infection (with persistent viremia, frequent abnormal ALT levels, and, in some cases, hepatomegaly).
AUTHOR DISCUSSION
We describe a large group of children with vertically acquired HCV infection who were prospectively observed from birth to 16 years of age. We confirmed the low prevalence of clinical manifestations and the lack of any significant disease, and, therefore, focused on markers of infection. The extent of viremia varied considerably, and we estimate that one-fifth to one-quarter of children may clear the virus. Hepatomegaly was the most common clinical manifestation, likely indicating liver inflammation and an early stage of HCV-related liver damage. Although hepatomegaly was observed in only a minority of children, it was significantly associated with a high proportion of abnormal ALT levels. Children with a high proportion of positive PCR test results were also more likely than others to have hepatomegaly, although this was not statistically significant in multivariable analyses. Hepatomegaly, viremia, and high ALT levels were found to be indicative of active infection in the same children, and low viral activity early in life was associated with clearance later on.
Our findings suggest 3 broad categories of children with vertical HCV infection: -20% with apparent clearance of the virus, reflecting acute infection; -50% with chronic asymptomatic infection, characterized by intermittent viremia, usually normal ALT levels, and rare hepatomegaly; and -30% with chronic, active infection, characterized by persistent viremia, frequent abnormal ALT levels, and hepatomegaly in some cases. Similar distinctions have been made for adult HCV infection [10].
Our estimated clearance rate, although conservative as a result of not including persistently PCR-negative children who were antibody positive beyond 18 months of age, is in line with reports by others; for example, Resti et al. [11] observed clearance in 12 (19%) of 62 children. The estimated rate of spontaneous viral clearance in adults is -15% [12]. Loss of antibody is rarely observed in HCV-infected adults who become persistently HCV RNA negative [13], but it has been reported [14]. One might expect a higher estimated clearance rate among children observed from birth than among adults, because adults clearing infection early without clinical symptoms are unlikely to ever be identified. In fact, in adult studies in which patients were followed up after a defined exposure, higher clearance estimates of -25% are reported [10].
In our study, clearance occurred by 3 years of age in most children, a finding similar to that of a study of hemophilia patients who usually cleared the virus 1-2 years after infection [15]. Early clearance explains the much lower proportion of viral clearance among our group 2 children. Although all children in our study were observed from birth, PCR tests were not routinely performed in the early 1990s; as a result, some infected children may have been missed if they were not tested in the first 2 years of life and if subsequent negative PCR test results or antibody negativity were interpreted as indicating the absence of infection [4]. However, in the group observed from birth, there were 4 children with apparent clearance after 3 years, which is in line with the clearance rate observed in group 2 with follow-up from a median age of 2.4 years.
In any multicenter study, the reliability of HCV RNA assays and potential interlaboratory variation must be considered. In a recent quality assessment among EPHN laboratories, false-positive qualitative HCV RNA assay results were rare [8], and it is unlikely that any children who experienced viral clearance in our study were incorrectly identified as having infection. False-negative results are more common, particularly at low viral loads, suggesting that some children with evidence of clearance may remain infected but with viral load below the detection limit [8].
Among the 26 HCV-HIV-coinfected children, hepatomegaly and persistent viremia were common. However, patterns of HCV viremia and ALT levels and the prevalence of clinical symptoms are difficult to interpret because of the complex interactions between the 2 viruses [16]. Although the number of coinfected children now born in Europe is very low, with the mother-to-child HIV transmission rate reduced to <2% [17], coinfection raises important issues relating to treatment with potent, hepatotoxic, antiretroviral therapies for HIV infection [18].
This analysis expands on our previous article [3]. Here, we move beyond a description of laboratory markers of infection to assess their association with clinical manifestations and to identify predictors of persistent viremia. However, further investigation of the associations we have observed is necessary before these markers can be used in the clinical management of cases of infection in children.
To the best of our knowledge, this is the largest group of children with vertical HCV infection described to date. All children were observed from birth, avoiding the referral bias present in other studies, in which infected children identified by pediatric hepatologists are more likely to be symptomatic or have elevated ALT levels. Despite a large total sample, the analyses to investigate associations between markers of infection and hepatomegaly lacked statistical power, especially in multivariable models.
Despite the difficulties of studying vertically acquired HCV infection, the EPHN provides a unique opportunity to collect standardized data for many infected children. Further analyses of this valuable data source will continue to elucidate the course of HCV infection in these children. Given the long latency time between infection and symptoms of ⩾20 years in adults [5], the lack of serious clinical manifestations among the children in our study highlights the need to investigate approaches to monitor the long-term consequences of vertically acquired HCV infection.
|
|
| |
| |
|
|
|
