ARTICLE
Tuberculosis Screening in Internationally Adopted
Children: The Need for Initial and Repeat Testing
Indi Trehan, MD, MPH, DTMHa,b, Jareen K. Meinzen-Derr, PhDa, Linda Jamison, RNa, Mary Allen Staat, MD, MPHa,c
aDepartment of Pediatrics andcInternational Adoption Center, University of Cincinnati, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio;bDepartment of
Pediatrics, Washington University and St Louis Children’s Hospital, St Louis, Missouri
The authors have indicated they have no financial relationships relevant to this article to disclose.
What’s Known on This Subject
Several studies have been conducted to assess the prevalence of TB infection in inter-nationally adopted children and have shown that these children are at high risk for TB.
What This Study Adds
Retesting internationally adopted children for TB after they have been home for a period of time identifies additional children who have TB infection and would have otherwise been missed.
ABSTRACT
OBJECTIVE.Because most internationally adopted children come from areas of high tuberculosis prevalence, an initial tuberculin skin test is recommended after arrival to the United States. We evaluated whether repeat testing of childrenⱖ3 months after arrival to the United States would identify additional children with latent tubercu-losis infection.
METHODS.Internationally adopted children who were seen at our International Adop-tion Center and had a tuberculin skin test within 2 months of arrival to the United States were eligible for the study. Children not diagnosed with tuberculosis with initial testing were retested at least 3 months later. The prevalence of tuberculosis on arrival and after repeat testing was determined, and potential risk factors for infection were examined.
RESULTS.Of the 527 internationally adopted children with an initial tuberculin skin test completed, 111 (21%) had evidence of latent tuberculosis infection. Repeat tuber-culosis testing was complete for 191 internationally adopted children (46.9% of those who had an initially negative tuberculin skin test). Latent tuberculosis infec-tion was found in 20% of those who were retested. No children were found to have active tuberculosis disease. Children with an initially positive tuberculin skin test result had slightly higher weight-for-agezscores at their initial clinic visit, whereas those whose tuberculin skin test result was positive after repeat testing had slightly lower weight-for-agezscores. A strong correlation between BCG immunization and tuberculin skin test result was observed.
CONCLUSIONS.Latent tuberculosis infection is common in internationally adopted chil-dren. A high proportion of internationally adopted children had an initially false-negative tuberculin skin test. Repeat tuberculosis testing of all internationally adopted children with an initially negative tuberculin skin test should be the stan-dard of care for identifying tuberculosis infection and preventing tuberculosis disease in this high-risk population.Pediatrics2008;122:e7–e14
N
EARLY 21 000 CHILDRENwere adopted into the United States in 2006 from countries around the world,⬎3 times as many as 15 years earlier.1Initial health screening of internationally adopted children (IAC) is recommended,2but screening often varies by provider and by the country of origin of the child. Primary care providers and subspecialists have few evidence-based guidelines available to direct optimal diagnostic and therapeutic strategies for newly IAC.
The majority of IAC come from developing countries in which tuberculosis (TB) is endemic.1TB rates in the United
States are highest among foreign-born individuals.3,4Children are a particularly high-priority group for treatment as
a result of their increased risk for severe disease and their lifetime risk for reactivation of tuberculosis.5Case reports
of children,6including IAC,7serving as index cases for widespread transmission of TB further highlight the need for
aggressive screening for TB in IAC.
Screening for TB by using a tuberculin skin test (TST) is recommended for all immigrants from high-prevalence countries shortly after arrival into the United States.4For IAC, it is suggested that the TST be repeated once the child
www.pediatrics.org/cgi/doi/10.1542/ peds.2007-1338
doi:10.1542/peds.2007-1338
Portions of this work were presented at the annual meeting of the Pediatric Academic Societies; May 17, 2005; Washington, DC.
Key Words
tuberculosis, screening, international adoption, BCG, malnutrition
Abbreviations
IAC—internationally adopted children TB—tuberculosis
TST—tuberculin skin test LTBI—latent tuberculosis infection NTM—nontuberculous mycobacteria PCP—primary care provider CDC—Centers for Disease Control and Prevention
WHO—World Health Organization OR— odds ratio
CI— confidence interval
Accepted for publication Jan 4, 2008
Address correspondence to Mary Allen Staat, MD, MPH, Cincinnati Children’s Hospital Medical Center, Division of Infectious Diseases, 3333 Burnet Ave, Cincinnati, OH 45229-3039. E-mail: [email protected].
is better nourished if malnutrition is initially suspected,2
although no specific evidence justifying this recommen-dation has been published. Although some evidence suggested that the receipt of the BCG vaccine should be considered when interpreting TST results,8
consen-sus guidelines2,9–11 continue to recommend that TST
results be interpreted without regard to BCG history, a recommendation that continues to be validated by recent studies.12
Although screening for TB shortly after adoption from a high-prevalence country is important for diagnosing both active and latent TB infection (LTBI),13there is the
possibility that relying only on early screening alone may not identify all children with TB. Factors such as undernourishment, concomitant infections, live vac-cines, and immunosuppression have been linked to an anergic response to tuberculin, thereby limiting the de-layed-type hypersensitivity reaction necessary to elicit a positive TST result.11 Furthermore, because TB has a
relatively long incubation period, the TST has poor sen-sitivity in identifying those with very recent exposure to TB.10 Thus, we hypothesized that repeating the TST
months later for IAC who had an initially negative TST result immediately after entry into the United States may identify additional IAC with TB. Although current recommendations do suggest repeat screening at a later time for children with suspected malnutrition,2 these
guidelines do not quantify the number of cases of LTBI that may be missed with initial screening or which fac-tors may be related to missed cases of TB. Although a nonspecific booster response to common mycobacterial antigens has been reported as a result of repeated TST, BCG vaccination, or exposure to nontuberculous myco-bacteria (NTM),14 this response is less likely when the
interval between tests exceeds 2 months.15For this study,
we sought to determine both the proportion of IAC with TB diagnosed on initial health screening shortly after ar-rival into the United States and also to determine whether additional infected children could be identified by repeat testing months later. We also examined whether specific factors such as age, birth country, BCG history, and nutri-tional status were associated with the TST result.
METHODS
Study Setting
This study was conducted at Cincinnati Children’s Hos-pital Medical Center’s International Adoption Center. The International Adoption Center was established in November 1999 and provides both preadoption counsel-ing and postadoption medical and developmental eval-uation of IAC. Families are encouraged to have a post-adoption evaluation performed within the first 2 weeks after arrival of their adopted child to the United States. At that visit, a comprehensive medical and developmen-tal evaluation is completed, and screening tests, includ-ing an HIV test and TST, are performed. A follow-up postadoption visit is strongly encouraged for all IAC 6 months after the initial visit, unless medical, social, or developmental concerns warrant an earlier visit. The Cincinnati Children’s Hospital Medical Center
institu-tional review board approved this study by using a lim-ited data set agreement for this research.
Study Population
IAC who were evaluated at the International Adoption Center at Cincinnati Children’s Hospital Medical Center for an initial postadoption visit between November 1999 and June 2004 were included in this study when they were seen within 2 months after their arrival to the United States and had their first postadoption TST during this visit. Demographic and anthropometric information was collected routinely on all children, and the child’s medical history was obtained through examination of medical charts from the child’s birth country and by interviewing the child’s adoptive parents. Immunization and medical charts were reviewed for identification of documentation of BCG vaccine, and the child’s entire body was examined for the characteristic BCG scar. For the purposes of this study, children who had either a BCG scar or written documentation of BCG vaccination were considered to have evidence of BCG immuniza-tion. Children who were known to have received a measles-containing vaccine ⱕ6 weeks before the visit did not have a TST and, therefore, were not eligible for the study because measles vaccines can temporarily sup-press tuberculin reactivity.11
Tuberculin Skin Testing
The Mantoux method of tuberculin skin testing was used for all children.16 Five tuberculin units of purified
protein derivative (0.1 mL; Aplisol [Parkedale Pharma-ceuticals Inc, Rochester, MN]) were injected intrader-mally into the volar or dorsal aspect of each child’s forearm10by 1 of 3 trained advanced practice nurses at
our International Adoption Center. Only children whose TST was read by a physician or a nurse 48 to 72 hours after placement were included in the analysis. A TST result was considered positive when there was at least 10 mm of induration, consistent with recommendations for recent immigrants from high-prevalence countries.10All
children with a positive TST result had chest radiographs performed for assessment for active TB disease. When no signs of TB were demonstrated by chest radiograph or physical examination, the child was given a diagnosis of LTBI and a 9-month course of isoniazid was prescribed. For children who had a negative initial TST result, detailed written instructions were sent to the child’s adoptive parents and designated primary care provider (PCP) recommending a repeat TST at least 3 months after the initial TST, either in their PCP’s office or during a follow-up visit to the International Adoption Center. Anthropometric data were again recorded when this repeat TST was performed. TSTs that were performed or read by the PCP were obtained via telephone calls and letters to the PCP’s office.
Statistical Analysis
database. Other relevant data from the child’s chart in-cluded each child’s age in days, birth country, time elapsed since adoption, the presence or absence of evi-dence suggesting receipt of BCG vaccination (scar or vaccination record),17 and weight at the time of initial
evaluation in our clinic. Each child’s weight was con-verted to z scores on the basis of Centers for Disease Control and Prevention (CDC) growth charts18–21and on
the World Health Organization (WHO) growth charts.22
The weight-for-age z scores were used as a marker for potential effects of malnutrition on growth. Birth country was also analyzed as a possible confounder because of potentially varying rates of TB exposure in each coun-try, which may also be a predictor of the rate of BCG immunization in the population. For birth countries with ⱕ10 children, IAC were grouped by geographic region rather than by country for maintaining ano-nymity.
Multivariable regression analyses were also performed to investigate the possible association between TST result and malnutrition while accounting for birth country and BCG vaccination as possible confounders. For these anal-yses, a weight-for-agezscore⬎2 SDs below the mean was used as a marker for malnutrition. Because of the relatively small numbers of IAC from most individual countries, this regression analysis focused on the subset of 432 IAC from Russia, China, Guatemala, Kazakhstan, and South Korea. Because Guatemala had the highest proportion of children with a positive initial TST result (32% at initial visit), this country was used as the reference group in this multivari-able regression when analyzing country of origin as a pos-sible confounder.
All statistical analyses were conducted by using Excel 2003 (Microsoft Corp, Redmond, WA) and SAS 9.1 (SAS Institute, Inc, Cary, NC). Means with SDs and medians with ranges are reported for continuous variables. Crude associations between TST results and possible continu-ous covariates were tested by using either the Student’s
ttest or the Wilcoxon rank-sum test as appropriate. The Pearson’s 2test was used to test associations between
TST result and categorical variables. Logistic regression was used to model the relationship between TST result and covariates and/or confounders. Results from the logistic models are reported as odds ratios (ORs) with 95% confidence intervals (CIs).
RESULTS
Study Population
A total of 769 IAC were seen for initial visits at the International Adoption Center between November 1999 and June 2004. Of those seen, 549 (71%) had their first TST performed at the center within 2 months of arrival to the United States and, thus, met eligibility criteria for inclusion in the study (Table 1). The mean age of study children was 23 months, and 54% were female. Eligible children came from 29 different countries, with 81% coming from Russia, China, Guatemala, Kazakhstan, or South Korea, an adoption pattern very representative of that seen nationally.1 The majority (81%) of IAC had
evidence of BCG vaccination, with children from South Korea being notable outliers with only 15% with evi-dence of BCG. Very few children (n⫽8) had documen-tation of receipt of multiple BCG vaccinations with dates (1 child had 3 documented vaccinations, the rest had 2). These children were from Kazakhstan (n⫽3), Eastern Europe (n⫽2), Russia (n ⫽2), and China (n ⫽1). A large number of children were also noted to have an initial weight that fell ⱖ2 SDs below their age- and gender-normalized mean on the CDC growth charts.18
None of the children tested positive for HIV infection.
Initial TST Results
The mean time elapsed between arrival to the United States and the initial TST for the 549 IAC who were eligible for the study was 12 days. Of the 549 children tested 22 (4%) did not have their TST read within 72 hours and were excluded from further analyses, and 111 (21%) had a TST result ofⱖ10 mm of induration (Tables 2 and 3, Fig 1). Additional evaluation (physical
exami-TABLE 1 Study Population
Country or Region of Origin No. of IAC (% of Total)
Mean Age at Adoption, mo
Female, % Weight⬎2 SDs Below Mean,
%18
Evidence of BCG (Scar or Vaccine
Record), %
Russia 193 (35) 26 42 38 93
China 110 (20) 17 92 34 80
Guatemala 69 (13) 15 43 14 78
Kazakhstan 41 (8) 25 44 20 100
Eastern Europea 36 (7) 32 44 17 69
South Korea 33 (6) 9 42 6 15
Ukraine 20 (4) 43 30 45 90
Other Asia and Pacific Rimb 20 (4) 20 55 20 53
India 12 (2) 23 58 67 100
Africac 9 (2) 62 78 0 56
Other Latin America and Caribbeand 6 (1) 63 67 17 83
Total 549 23 54 29 81
aIncludes Albania, Bulgaria, Moldova, Romania, and Slovakia.
bIncludes Azerbaijan, Cambodia, Japan, Nepal, the Philippines, the Republic of Georgia, Thailand, and Vietnam. cIncludes Ethiopia, Ghana, Liberia, South Africa, and Zambia.
nation and chest radiograph) showed that all of these children had LTBI (ie, none had active TB) and were begun on the recommended 9-month course of isoniazid therapy.2Of the remaining children, 416 (79%) had a
TST result of⬍10 mm induration; for those for whom data were available, 92% had no induration and 8% had 1 to 9 mm of induration. LTBI was diagnosed most commonly in IAC from Guatemala (32%), Africa (22%), and Russia (22%). Because of the relatively small num-ber of children whose TST was not read in the appropri-ate time frame, additional analyses focused on the 527 children who did have their TST read appropriately.
For the 527 children included in the study, the mean (SD) age of adoption was 22.9 (25.0) months, and the median age (range) was 13.5 (1.2–200.0) months. The median (range) age of adoption in months with a posi-tive TST result (n⫽ 111) compared with those with a negative TST result (n⫽416) was not significantly dif-ferent (14.8 [3.8 –181.0] vs 13.1 [1.2–200.0];P⫽.32).
The mean (SD) weight-for-age z score in the study population was⫺1.44 (1.34). The median (range) was
⫺1.32 (⫺7.00 to 3.90). Malnutrition (defined as azscore less than ⫺2.0) was seen in 158 (30%) children. The median (range) weight-for-age z score for IAC with a positive TST result was slightly higher than those with a negative TST result (⫺1.13 [⫺5.40 to 1.31] vs ⫺1.38 [⫺7.00 to 3.94]; P ⫽ .06). Children with an initially negative TST result were more likely to be malnour-ished, compared with children who had an initially
pos-itive TST result (31% vs 22%;P⫽.06). We did not find institutionalization status to be related to TST result, as has been postulated by others.13IAC who had lived in an
orphanage or hospital at any time had a positive TST result 19.7% of the time, those who had lived in an orphanage for at least 6 months were positive 19.5% of the time, and those who had resided in a foster home were positive 24.1% of the time.
Multivariable regression analysis of children from the 5 countries with the most IAC in our study (Table 4) also showed that malnourished children were less likely to have a positive TST result at the initial visit, compared with children with azscore of⫺2.0 or more (OR: 0.58 [95% CI: 0.33–1.02];P⫽.06). We also analyzed these data using WHO z scores and found no significant change in the effect sizes (ORs) reported in the logistic regression models. Children with initial weight-for-agez
scores that were⬎2 SDs below the mean were still less likely to have had a positive TST result at the initial visit, with virtually the same effect sizes (OR: 0.53 instead of 0.58). Children who were older than 10 years were
First postadoption TST performed at International Adoption Center within 2 mo of arrival in USA
n = 549
Positive TST
n = 111 (21%)
Negative TST
n = 416 (79%)
TST not read in 48–72 h
n = 22 (4%)
Second postadoption TST Performed at least 3 mo later
n = 203 (49%)
Positive TST
n = 38 (20%) nNegative TST = 153 (80%)
Second postadoption TST not Performed at least 3 mo later
n = 213 (51%)
TST read in 48–72 h
n = 527 (96%)
TST not read in 48–72 h
n = 12 (6%)
TST read in 48–72 h
n = 191 (94%)
FIGURE 1
Initial and repeat TST results.
TABLE 3 Number of Children With Complete Testing and TST Results According to BCG Status
BCG Status Initial TST Performed
Repeat TST Performed
n Positive, %
n(Follow-up Rate, %)
Positive, %
Evidence of BCG 424 25 152 (48) 24
No Evidence of BCG 103 4 39 (39) 3
Total 527 21 191 (46) 20
TABLE 2 Number of Children With Complete Testing and TST Results According to Birth Country
Country or Region of Origin
Initial TST Repeat TST
n Positive, %
n(Follow-up Rate, %)
Positive, %a
Russia 185 23 83 (58) 27
China 108 19 34 (39) 15
Guatemala 67 33 14 (31) 21
Kazakhstan 40 20 13 (41) 15
Eastern Europe 33 12 15 (52) 7
South Korea 32 6 10 (33) 0
Ukraine 20 40 5 (42) 40
Other Asia and Pacific Rim 19 5 6 (33) 17
India 11 9 7 (70) 14
Africa 7 29 2 (40) 0
Other Latin America and Caribbean
5 20 2 (50) 50
Total 527 21 191 (46) 20
TABLE 4 Multivariable Logistic Regression Analysis of Risk Factors for Initially Positive TST Results Among IAC From 5 Countries (Nⴝ432)
Risk Factor OR 95% CI P
Initial weight-for-agezscore less than⫺2.0
0.58 0.33–1.02 .0600
BCG scar or documentation 15.30 3.30–70.10 .0004 Country
Guatemalaa 1.00
Russia 0.52 0.27–0.99 .0470
China 0.46 0.22–0.97 .0400
Kazakhstan 0.37 0.14–0.95 .0380
South Korea 0.50 0.09–2.80 .4300
omitted from this analysis because no WHOzscore was available for them.
Children with evidence of BCG vaccination were more likely to have a positive TST result than those without evidence of vaccination (OR: 15.3 [95% CI: 3.3–70.1]; P ⫽ .0004). These point estimates did not change significantly when the analysis excluded chil-dren for whom multiple BCG vaccinations were docu-mented. For children who had a documented date for their BCG, the median duration between BCG and TST was not significantly different for children with a posi-tive TST result (391 days;n⫽79) compared with those with a negative TST result (377 days;n⫽245;P⫽.96).
Repeat TST Results
A repeat TST at least 3 months after the initial TST was recommended for IAC whose initial TST result was nega-tive or not read. Despite numerous reminders to adopnega-tive parents and primary care providers, only 191 (46%) of the 416 IAC with an initially negative TST result had a repeat TST performed and read at least 3 months after their initial test (Fig 1). Of these, 38 (20%) were found to be positive for LTBI. Thirty-five (92%) of these 38 IAC had TST indu-rations that increased byⱖ10 mm at repeat testing; the other 3 had indurations that increased by 2 mm, 5 mm, and 9 mm. Whereas the average (SD) weight-for-age z
score at the initial visit for this subset of children with a repeat TST was⫺1.70 (1.50), the average (SD)zscore at the time of repeat TST was⫺0.85 (1.10).
Children with a repeat TST (n ⫽191) had a slightly higher median (range) age of adoption in months com-pared with children (n⫽225) who did not have a repeat TST (14.1 [3.9 –101.0] vs 12.5 [1.2–200.0];P⫽.01]. IAC who had a repeat TST were slightly more likely to be male (51% vs 41%; P ⫽ .05) and more likely to be malnourished (37% vs 26%; P ⫽ .02) compared with those who did not return for a repeat TST.
The mean (SD) age of adoption for the 191 children who had a repeat TST performed and read in the appro-priate time frame was 22.1 (18.6) months. The median (range) was 14.1 (3.9 –101.0) months. The median (range) age of adoption in months in the group with a positive TST result (n⫽38) compared with those with a negative TST result (n⫽153) was not significantly dif-ferent (15.2 [5.6 –97.2] vs 14.1 [3.9 –101.0];P⫽.68).
We compared the relationship between weight at the initial postadoption visit with the result of a repeat TST for IAC for whom this repeat TST was indicated. The mean (SD) weight-for-age z score for children with a repeat TST was ⫺1.70 (1.49). The median (range) was
⫺1.57 (⫺7.0 to 3.9). Malnutrition (z score less than
⫺2.0) was seen in 70 (37%) children. In contrast to the results observed with initial TST, the group of children who had a positive repeat TST result had a lower median (range) weight-for-age zscore at the initial visit, com-pared with children who had a negative repeat TST result (⫺1.99 [⫺4.7 to ⫺0.09] vs ⫺1.55 [⫺7.0 to
⫺3.94];P⫽.007). Children with a positive repeat TST result were more likely to be malnourished at the initial visit, compared with children who had a negative repeat TST result (50% vs 33%; P ⫽.06). These results were
again very similar when using WHO growth parameters instead of CDC parameters.
Of the 191 IAC who had a repeat TST performed after having an initially negative TST result, 175 (91.6%) had an initial TST with no induration, and 31 (17.7%) of these children had a repeat TST result that was positive. In contrast, 16 (8.4%) of the 191 initially had 1 to 9 mm of induration, and 7 (43.8%) of these had a positive repeat TST result (P⫽.03).
Multivariable regression analysis (Table 5) was per-formed using the subset of IAC from Russia, China, Guatemala, and Kazakhstan (n⫽144), because no IAC from South Korea had a positive TST result at the fol-low-up visit. In contrast to what was found with the initial TST, malnourished children were more likely to have a positive repeat TST result compared with children who were not malnourished, although this finding was not statistically significant (OR: 1.7 [95% CI: 0.75–3.8];
P ⫽ .20). Children with evidence of BCG vaccination were again more likely to have a positive repeat TST result than those without evidence of vaccination (OR: 7.1 [95% CI: 0.90 –56.0];P⫽.06). There did not seem to be any relationship between having a positive repeat TST result and country of birth.
DISCUSSION
TB remains a tremendous public health threat around the world. In our increasingly globalized society, the ease with which individuals cross borders and oceans from areas of high TB endemicity to areas of low ende-micity has never been higher. Identifying and treating immigrants who are infected with TB on their arrival in a low-endemicity area is an opportunity and an obliga-tion of which clinicians must be aware, not only to protect the health of the immigrant but also to limit the spread of TB to the domestic population.23 It has been
recognized for some time now that a large group of such immigrants, who can be easily targeted for screening and treatment, are the rapidly expanding population of IAC.24This study aimed to provide data for the
develop-ment of evidence-based guidelines on how to screen this population of children for TB. Whereas previous studies examined the rates of TB in IAC, we believe that this study provides some unique insights to help guide screening for TB in IAC.
TABLE 5 Multivariable Logistic Regression Analysis of Risk Factors for Repeat Positive TST Results Among IAC From 4 Countries (Nⴝ144)
Risk Factor OR 95% CI P
Initial weight-for-agezscore less than⫺2.0
1.70 0.75–3.80 .20
BCG scar or documentation 7.10 0.90–56.00 .06 Country
Guatemalaa 1.00
Russia 1.10 0.27–4.60 .87
China 0.66 0.13–3.40 .62
Kazakhstan 0.52 0.07–3.90 .53
Some of the earliest data on TB screening among IAC were presented by Hostetter et al,25,26who showed a TB
prevalence of⬃3% in their series. It is not entirely clear why our results differ from the results of these 2 studies. One reason may be that the countries of origin differ remarkably between our cohort and theirs; for example, 41% of their subjects were from South Korea, a country with a relatively low prevalence of TB. Another reason may be who read the TST. In out study, we included only IAC whose TST was read by a health care provider. This has been shown to increase significantly the accu-racy of TST interpretation, although even trained heath care providers are also subject to “underreading” of the TST.27,28 In the Hostetter series, it was not described
whether the TST was read by a health care provider. Three other published studies used IAC that were from similar birth countries and adopted during a similar time period.19,20,29Saiman et al29showed a rate of LTBI of 19%
in their population, similar to our rate. In their study, it was stated that a health care provider read each child’s TST. Miller et al,19,20in contrast, reported rates of positive
TST results of 6% to 7% in their studies; however, they did not specify who read the TST. A more recent study by Mandalakas et al13 found a prevalence of 11.9% in
880 IAC who were evaluated between 1986 and 2001; however, the origins of the children evaluated was dif-ferent from in our study and what is seen nationally, with a higher proportion of children from South Korea (18.4%) and India (12.1%) and a lower proportion from China (10%), which may account for the overall lower prevalence that they observed. These methodologic dif-ferences may explain the difference in observed LTBI prevalence between our study and the others. At the same time, we also acknowledge that in our study the repeat TSTs were placed and read by either the Interna-tional Adoption Center or the PCP and that this variabil-ity may also have led to underreading of the repeat TSTs. These previous studies also did not comment on when (relative to their arrival to the United States) their children were tested for TB. In our series, we included only children who were initially tested within 2 months after arrival. This allowed us the unique opportunity to retest at a later date many children whose TST result was initially negative to assess the role that repeat testing should play in the care of IAC. To our knowledge, this is the first study to explore the issue of retesting children with an initially negative TST result and the risk factors related to TST results.
Our most striking observation was that a large per-centage of children with LTBI are not identified by an initial TST performed shortly after arrival. When those with an initially negative TST result were retested at least 3 months after their initial test, 20% had a positive repeat TST result. Presumably, these children were not exposed to TB in the United States but instead at this later date were better able to mount an appropriate delayed hypersensitivity response to the TST. The hy-pothesis that this is perhaps a result of improved nutri-tion is supported by our data showing that those with an initially positive TST result had a higher weight-for-agez
score (⫺1.13 vs ⫺1.38; P ⫽ .06). This is further
sup-ported by examination of the initial weights of IAC who had a repeat TST performed after having had an initially negative TST result: in these children, the weight-for-age
zscores were lower in those with a repeat TST result that was positive (⫺1.99 vs⫺1.55;P⫽.007), suggesting that they did indeed have LTBI at the time of initial testing but were unable initially to mount a sufficient immune response to have a positive TST result. Nevertheless, because weight is not an ideal proxy for immune status or the ability to mount a response to the TST and be-cause a large number of IAC with weights in the refer-ence range did have a positive TST result on repeat screening, we believe that guidelines for the care of IAC should include recommendations for retesting all IAC with an initially negative TST result, not just those con-sidered “malnourished,” at least 3 months after their initial test.
A limitation of our study was the relatively low rate of repeat TST performed in IAC whose initial TST result was negative or not read (Fig 1), despite recommending repeat testing to adoptive parents and their PCPs. Even among the highly motivated population of adoptive par-ents who brought their children to our International Adoption Center despite the time and commitment re-quired, only 49% followed up with a repeat TST after an initially negative TST result. Extrapolating our finding that 20% of IAC who had a repeat TST performed had LTBI to those in whom a repeat TST was not performed or not read would give us an overall prevalence of LTBI of almost 37% in our population of IAC. Of these, only 57% would have been discovered without repeat test-ing.
Other factors that are postulated to be related to initial and repeat TST result were also examined. Some physicians and adoption professionals recommend against TST for IAC until after 1 year of age because of concerns about the link between BCG and TST; how-ever, we found no significant correlation between the age of adoption or age at testing and the likelihood of having a positive TST result, either initially (P⫽.32) or at follow-up (P⫽.68). We also observed that IAC from South Korea were the only children (among the 5 most represented countries) to be statistically less likely to have a positive TST result in our logistic regression anal-ysis.
The final variable of interest to us was the association between BCG immunization status and TST result. Al-though BCG continues to provide valuable and cost-effective protection against TB for children in high-prev-alence countries worldwide,30,31the interpretation of TST
in children with BCG continues to be controversial, with some investigations showing an association8,32and
oth-ers not showing an association.12,17,33–35In IAC, our study
found a strong association between BCG immunization status and a positive TST result, similar to that seen by Saiman et al29; however, our results still do not clarify
whether the positive TST results observed are a result of BCG immunization or a response to previous exposure to mycobacteria (either TB or NTM).
similar to ours with regard to repeat TB testing in an immigrant population,36664 Indochinese refugees had a
TST placed and 46% were found to have an initially positive TST result. A repeat TST was placed 60 days later in the 217 with initially negative TST results, and 43% converted. The average age was 21 years with a range of 1 to 80 years, far older than our population. No signifi-cant association was found with age or BCG vaccination, and the role of malnutrition was not examined. In a different population of Canadian students, in which many of the students had documentation of BCG vacci-nation, 2-step testing was performed to examine whether a booster effect would be observed.37 Overall,
there were 1542 participants, and 6.6% had a positive TST result. Of the 1435 students with an initially nega-tive TST result, a second test was placed and 5.2% had a positive reaction. Unlike our study, they found a signif-icant association with older age. Similar to our study, however, a significant association with a large initial reaction and previous BCG vaccination was found. In a final study,38 69 TST-negative healthy volunteers
re-ceived BCG vaccine and had serial TSTs performed to determine the prevalence, persistence, and boosting of TST reactions. Overall, 10% had persistent reactions
ⱖ15 mm, and 3% had boosting ofⱖ15 mm 1 to 3 years later. Although this study does demonstrate the effect of BCG on the TST response, it is unclear how to use this information in a population of BCG-vaccinated children who are at high risk for TB exposure, infection, and disease.
New technologies using interferon-␥ release assays show promise in helping to resolve this limitation of the TST.39 Until these or other screening methods become
standard for TB detection in children, we continue to support the current practice guidelines, which recom-mend that a history of BCG not be a contradiction to placement of a TST and that interpretation of the TST should not be influenced by a history of BCG immuni-zation.2,9
CONCLUSIONS
Our experience with TB over several years at our Inter-national Adoption Center shows a high rate of LTBI in IAC. Furthermore, a large percentage of LTBI goes un-diagnosed because of failure to repeat the TST for chil-dren whose initial TST result was negative or not read. Additional investigation is warranted to inform better whether TB screening guidelines for immigrant children and adults in general4,40,41should include a
recommen-dation for repeat testing when the initial TST result is negative or not read in these high-risk populations.
ACKNOWLEDGMENTS
We thank Marina Bischoff, Vanessa Florian, Emilie Grube, Tyler Browning, Elizabeth Roberts, Rotimi Oku-nade, and Kristen Frommeir for assistance with this study. We are indebted to all of the wonderful children and their families who participated in this study, helping to improve the health of internationally adopted chil-dren in the future.
REFERENCES
1. US Department of State. Immigrant visas issues to orphans coming to U.S. Available at: http://travel.state.gov/family/ adoption/stats/stats㛭451.html. Accessed December 19, 2007 2. Pickering LK, Baker CJ, Long SS, McMillan JA, eds.Red Book:
2006 Report of the Committee on Infectious Diseases. 27th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2006 3. Nelson LJ, Schneider E, Wells CD, Moore M. Epidemiology of
childhood tuberculosis in the United States, 1993–2001: the need for continued vigilance.Pediatrics.2004;114(2):333–341 4. Taylor Z, Nolan CM, Blumberg HM; American Thoracic
Society; Centers for Disease Control and Prevention; Infectious Diseases Society of America. Controlling tuberculosis in the United States. Recommendations from the American Thoracic Society, CDC, and the Infectious Diseases Society of America [published correction appears inMMWR Morb Mortal Wkly Rep.
2005;54(45):1161].MMWR Recomm Rep.2005;54(RR-12):1– 81 5. Horsburgh CR Jr. Priorities for the treatment of latent tuber-culosis infection in the United States. N Engl J Med. 2004; 350(20):2060 –2067
6. Phillips L, Carlile J, Smith D. Epidemiology of a tuberculosis outbreak in a rural Missouri high school. Pediatrics. 2004; 113(6). Available at: www.pediatrics.org/cgi/content/full/113/ 6/e514
7. Curtis AB, Ridzon R, Vogel R, et al. Extensive transmission of
Mycobacterium tuberculosis from a child. N Engl J Med. 1999; 341(20):1491–1495
8. Bozaykut A, Ipek IO, Ozkars MY, Seren LP, Atay E, Atay Z. Effect of BCG vaccine on tuberculin skin tests in 1– 6-year-old children.Acta Paediatr.2002;91(2):235–238
9. American Thoracic Society, Centers for Disease Control and Prevention. Diagnostic standards and classification of tubercu-losis in adults and children.Am J Respir Crit Care Med.2000; 161(4 pt 1):1376 –1395
10. American Thoracic Society. Targeted tuberculin testing and treatment of latent tuberculosis infection.MMWR Recomm Rep.
2000;49(RR-6):1–51
11. Pediatric Tuberculosis Collaborative Group. Targeted tuberculin skin testing and treatment of latent tuberculosis infection in chil-dren and adolescents.Pediatrics.2004;114(4 suppl):1175–1201 12. Sleiman R, Al-Tannir M, Dakdouki G, Ziade F, Assi NA, Rajab
M. Interpretation of the tuberculin skin test in Bacille Calmette-Guerin vaccinated and nonvaccinated school chil-dren.Pediatr Infect Dis J.2007;26(2):134 –138
13. Mandalakas AM, Kirchner HL, Iverson S, et al. Predictors of
Mycobacterium tuberculosisinfection in international adoptees.
Pediatrics. 2007;120(3). Available at: www.pediatrics.org/cgi/ content/full/120/3/e610
14. Menzies D. Interpretation of repeated tuberculin tests: boost-ing, conversion, and reversion.Am J Respir Crit Care Med.1999; 159(1):15–21
15. Cauthen GM, Snider DE Jr, Onorato IM. Boosting of tuberculin sensitivity among Southeast Asian refugees. Am J Respir Crit Care Med.1994;149(6):1597–1600
16. Centers for Disease Control and Prevention, Division of Tubercu-losis Elimination. Tuberculin skin testing. Available at: www.cdc. gov/tb/pubs/corecurr/Chapter4/Chapter㛭4㛭Skin㛭Testing.htm. Ac-cessed December 19, 2007
17. Santiago EM, Lawson E, Gillenwater K, et al. A prospective study of bacillus Calmette-Guerin scar formation and tubercu-lin skin test reactivity in infants in Lima, Peru.Pediatrics.2003; 112(4). Available at: www.pediatrics.org/cgi/content/full/112/ 4/e298
Pediatrics. 2000;105(6). Available at: www.pediatrics.org/cgi/ content/full/105/6/e76
20. Miller L, Chan W, Comfort K, Tirella L. Health of children adopted from Guatemala: comparison of orphanage and foster care.Pediatrics.2005;115(6). Available at: www.pediatrics.org/ cgi/content/full/115/6/e710
21. Miller LC, Tseng B, Tirella LG, Chan W, Feig E. Health of children adopted from Ethiopia.Matern Child Health J. 2007; Aug 22
22. World Health Organization. The WHO child growth standards. Available at: www.who.int/childgrowth/standards/en/index. html. Accessed December 19, 2007
23. Nelson LJ, Wells CD. Tuberculosis in children: considerations for children from developing countries.Semin Pediatr Infect Dis.
2004;15(3):150 –154
24. Lange WR, Warnock-Eckhart E, Bean ME.Mycobacterium tuber-culosisinfection in foreign born adoptees.Pediatr Infect Dis J.
1989;8(9):625– 629
25. Hostetter MK, Iverson S, Dole K, Johnson D. Unsuspected infectious diseases and other medical diagnoses in the evalua-tion of internaevalua-tionally adopted children.Pediatrics.1989;83(4): 559 –564
26. Hostetter MK, Iverson S, Thomas W, McKenzie D, Dole K, Johnson DE. Medical evaluation of internationally adopted children.N Engl J Med.1991;325(7):479 – 485
27. Ozuah PO, Burton W, Lerro KA, Rosenstock J, Mulvihill M. Assessing the validity of tuberculin skin test readings by trained professionals and patients.Chest.1999;116(1):104 –106 28. Carter ER, Lee CM. Interpretation of the tuberculin skin test
reaction by pediatric providers.Pediatr Infect Dis J.2002;21(3): 200 –203
29. Saiman L, Aronson J, Zhou J, et al. Prevalence of infectious diseases among internationally adopted children. Pediatrics.
2001;108(3):608 – 612
30. Soysal A, Millington KA, Bakir M, et al. Effect of BCG vaccination on risk ofMycobacterium tuberculosis infection in children with household tuberculosis contact: a prospective community-based study.Lancet.2005;366(9495):1443–1451
31. Trunz BB, Fine P, Dye C. Effect of BCG vaccination on child-hood tuberculous meningitis and miliary tuberculosis worldwide: a meta-analysis and assessment of cost-effective-ness.Lancet.2006;367(9517):1173–1180
32. Young J, O’Connor ME. Risk factors associated with latent tuberculosis infection in Mexican American children.Pediatrics.
2005;115(6). Available at: www.pediatrics.org/cgi/content/ full/115/6/e647
33. Karalliedde S, Katugaha LP, Uragoda CG. Tuberculin response of Sri Lankan children after BCG vaccination at birth.Tubercle.
1987;68(1):33–38
34. al-Kassimi FA, Abdullah AK, al-Orainey IO, et al. The signifi-cance of positive Mantoux reactions in BCG-vaccinated chil-dren.Tubercle.1991;72(2):101–104
35. Young TK, Mirdad S. Determinants of tuberculin sensitivity in a child population covered by mass BCG vaccination.Tuber Lung Dis.1992;73(2):94 –100
36. Morse DL, Hansen RE, Swalbach WG, Redmond SR, Grabau JC. High rate of tuberculin conversion in Indochinese refugees.
JAMA.1982;248(22):2983–2986
37. Menzies R, Vissandjee B, Rocher I, St Germain Y. The booster effect in two-step tuberculin testing among young adults in Montreal.Ann Intern Med.1994;120(3):190 –198
38. Hoft DF, Tennant JM. Persistence and boosting of Bacille Calmette-Guerin-induced delayed-type hypersensitivity.Ann Intern Med.1999;131(1):32–36
39. Detjen AK, Keil T, Roll S, et al. Interferon-gamma release assays improve the diagnosis of tuberculosis and nontubercu-lous mycobacterial disease in children in a country with a low incidence of tuberculosis.Clin Infect Dis.2007;45(3):322–328 40. Centers for Disease Control and Prevention. Recommendations
for prevention and control of tuberculosis among foreign-born persons: report of the Working Group on Tuberculosis among Foreign-Born Persons.MMWR Recomm Rep. 1998;47(RR-16): 1–29
DOI: 10.1542/peds.2007-1338
2008;122;e7
Pediatrics
Indi Trehan, Jareen K. Meinzen-Derr, Linda Jamison and Mary Allen Staat
Initial and Repeat Testing
Tuberculosis Screening in Internationally Adopted Children: The Need for
Services
Updated Information &
http://pediatrics.aappublications.org/content/122/1/e7
including high resolution figures, can be found at:
References
http://pediatrics.aappublications.org/content/122/1/e7#BIBL
This article cites 29 articles, 3 of which you can access for free at:
Subspecialty Collections
http://www.aappublications.org/cgi/collection/adoption_sub Adoption
alth_sub
http://www.aappublications.org/cgi/collection/international_child_he International Child Health
following collection(s):
This article, along with others on similar topics, appears in the
Permissions & Licensing
http://www.aappublications.org/site/misc/Permissions.xhtml
in its entirety can be found online at:
Information about reproducing this article in parts (figures, tables) or
Reprints
http://www.aappublications.org/site/misc/reprints.xhtml
DOI: 10.1542/peds.2007-1338
2008;122;e7
Pediatrics
Indi Trehan, Jareen K. Meinzen-Derr, Linda Jamison and Mary Allen Staat
Initial and Repeat Testing
Tuberculosis Screening in Internationally Adopted Children: The Need for
http://pediatrics.aappublications.org/content/122/1/e7
located on the World Wide Web at:
The online version of this article, along with updated information and services, is
by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.
