INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The next wave of human immunodeficiency virus (HIV) infection and AIDS is expected to occur in Asia (32). While estimates of the potential impact that the epidemic may have on the pediatric population in the developing world have been made (2, 29), most studies on the development of the normal immune system in relation to HIV infection (10, 13-14, 19, 25-27), the course of HIV infection (3, 4, 6, 8, 21, 23), as well as on antiretroviral treatment (ART) (5, 22, 31) and Pneumocystis carinii pneumonia prophylaxis (PCPP) (7, 20, 30) in the pediatric population were done in Western countries. Data on maturational changes in the peripheral lymphocyte subsets among pediatric patients have been scarce. We therefore believed that there is an urgent need to study the underlying parameters that have been proposed to be used to monitor and guide therapy for HIV-positive children.

Using quality-controlled and standardized flow cytometry, we have previously found significantly lower peripheral CD4⁺ T-lymphocyte (CD4) values but higher natural killer (NK) cell values, in terms of both percent peripheral lymphocytes and absolute cell counts, in our adult HIV-negative population (17). Subsequently, we proposed a separate staging system based on peripheral CD4 cells which can be used to monitor HIV-infected adult Chinese individuals and assessed the potential impact that this may have on the use of ART and PCPP in our HIV-infected population (18).

In the present study, we examined the maturational changes in peripheral lymphocyte subsets that have been shown in previous studies to be pertinent to the monitoring of HIV-infected pediatric patients. In anticipation of future studies of HIV-infected pediatric patients in Asia, we also highlight the immunological differences that may exist between individuals in various age groups in different geographic populations.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Study population. Healthy, full-term ethnic Chinese infants with normal spontaneous delivery whose cord blood was used to screen for congenital syndromes (glucose-6-phosphate dehydrogenase deficiency, hypothyroidism) were included in the study (group A). Ethnic Chinese children in other age groups (groups B to E) were recruited in an anonymous fashion from part of a long-term hepatitis B vaccine study cohort in Hong Kong who had been given a course of vaccine (at a time schedule of 0, 1, and 6 months) and who were monitored at regular intervals. Those who were excluded had a history of infection in the past 4 weeks (including infections with viral, bacterial, fungal, or parasitic pathogens), had been hospitalized within the past 3 months, or had taken antibiotics or other medications (steroids, nonsteroidal anti-inflammatory agents, or other cardiovascular drugs). Informed consent was obtained from the participants. EDTA-anticoagulated whole-blood samples were collected between approximately 9:00 a.m. and 12 noon and were then transported and analyzed in the immunocytometry laboratory on the same day. A simultaneous sample was obtained and analyzed with an automated hematological instrument for absolute cell counts and percent lymphocytes with leukocyte differential (17). All children tested were HIV type 1 (HIV-1) and HIV-2 seronegative, as determined by enzyme-linked immunosorbent assay.

Flow cytometry. A standardized flow cytometry method with lysed whole blood and a panel of two-color combinations of fluorescein isothiocyanate- and phycoerythrin-conjugated monoclonal antibody reagents obtained from a single manufacturer (Becton Dickinson, San Jose, Calif.) was used to determine the expression of each antigen or antigen combination (17). Data acquisition was performed on configured FACScan flow cytometers. Appropriate quality control procedures were done as described previously (17, 18). Absolute cell counts (numbers of cells per microliter) were obtained by multiplication of the percentage of lymphocytes by the leukocyte differential obtained from the simultaneous blood sample analyzed with an automated hematological instrument (Coulter MAXM; Coulter Corp., Miami, Fla.).

Statistical analysis. Patients were stratified by four age groups for statistical analysis: group A, newborn to less than age 1 year; group B, age 1 to 2 years; group C, age 5 to 7 years; group D, age 9 to 12 years. Means, standard deviations, and 95% confidence intervals (CIs) were calculated, and between-age-group comparisons were done by using Student's t test assuming unequal variance. Values obtained from a previously published study with adults (17) were used for comparison. Two-sided P values were calculated to test for statistically significant differences, and P values of <0.01 or <0.05 were tabulated.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

A total of 212 children were recruited into the study. The study population consisted of 101 (47.6%) females and 111 (52.4%) males. The sex and age distributions of the subjects in the various age groups are shown in Table 1. No significant difference between the sexes within each age group could be detected for any lymphocyte marker antigen. The results for all age groups were subsequently analyzed without further breakdown into different sexes.

Table 2 shows the changes in peripheral white cell differential with age. It can be seen that the total white cell count dropped precipitously from birth (15.1 cells/µl; 95% CI, 12.7 to 17.5) and during the first 5 years of life (7.8 cells/µl; 95% CI, 7.1 to 8.5), which stabilized by about the age of individuals in group C, before again declining during adolescence. Analysis of the white cell differential showed that there was a striking increase in percent lymphocytes between birth (26.7%; 95% CI, 19.2 to 34.2) and the second year of life (57.5%; 95% CI, 51.8 to 63.2), followed by a drop in both the percentage and the absolute numbers of lymphocytes in early childhood (age groups B and C) and adolescence (age group D). This pattern was almost mirrored by a concomitant drop (from 64.3% [95% CI, 56.6 to 72.1] to 35.9% [95% CI, 30.0 to 41.8]) and then a rise in percent granulocytes, although the changes in absolute numbers of granulocytes were less marked except for the change during infancy (age group A [9,813 cells/µl; 95% CI, 7,614 to 12,012] to age group B [3,556 cells/µl; 95% CI, 2,695 to 4,416]). Changes in percent monocytes were much less marked, with only differences between age groups A versus B and D versus adults just reaching statistical significance (P < 0.05).

Table 3 shows the changes in lymphocyte subsets in the different age groups. Examination of the changes showed that the percentage of CD3⁺ T lymphocytes remained remarkably steady throughout the whole growth period from the age of individuals in group A to adulthood, while the changes in absolute numbers appeared to reflect the changes in total peripheral lymphocyte counts. Percent B cells (CD19⁺ CD3), however, showed a marked rise in infancy (from age group A [17.3%; 95% CI, 10.1 to 24.6] to age group B [24.5%; 95% CI, 22.2 to 26.7]), followed by a gradual decline, which occurred most significantly during adolescence (from age group D [14.9%; 95% CI, 13.7 to 16.0] to adulthood [11.1%; 95% CI, 10.6 to 11.6]). The percent NK cells (CD3 with CD16⁺ and/or CD56⁺ cells), on the other hand, displayed a pattern of change almost opposite that for percent B cells, while the absolute numbers of NK cells stayed at relatively constant levels in all age groups after the apparent initial drop from birth.

The percentage of T-helper cells (CD3⁺ CD4⁺) at first showed a small decline, followed by a more marked decline from age group B (41.5%; 95% CI, 39.1 to 43.9) to age group C (34.4%; 95% CI, 32.1 to 36.7) and then a slight increase during adolescence, before reaching adult levels (36.4%; 95% CI, 35.4 to 37.4). On the other hand, peripheral T-suppressor cells (CD3⁺ CD8⁺) showed a gradual increase among individuals of all age ranges before adulthood (29.7%; 95% CI, 28.7 to 30.7), while the absolute numbers rose and peaked immediately after infancy (1,516 cells/µl; 95% CI, 1,243 to 1,788), before declining to adult levels. The T-helper cell/T-suppressor-cell ratio (HS ratio), which combines the changes in both T-helper and T-suppressor cells, remained fairly constant among individuals of all age ranges studied, apart from a significant decline which occurred from age group B (1.6; 95% CI, 1.4 to 1.8) to age group C (1.3; 95% CI, 1.1 to 1.4).

Results of examination of the activation markers (HLA-DR⁺, CD38⁺) in CD8⁺ cells are shown in Table 4. Breakdown analysis of the subgroups showed that the percent CD8⁺ CD38⁺ cells decreased during infancy (from age group A [30.1%; 95% CI, 26.2 to 34.0] to age group B [24.6%; 95% CI, 22.6 to 26.6]) and rose again during late childhood (age groups C to D); this was probably related to childhood infections. The percent HLA-DR⁺ CD8⁺ cells, however, showed a different pattern of change, with only one significant rise during late childhood (from age group C [10.8%; 95% CI, 8.9 to 12.6] to age group D [17.2%; 95% CI, 15.2 to 19.2]). The changes in absolute numbers of cells for both activation markers also appeared to reflect the changes in absolute total peripheral lymphocyte numbers.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Present guidelines on ART and PCPP in the HIV-infected pediatric population rest heavily on estimation and enumeration of peripheral CD4 cells as well as other cell markers of HIV disease (4-9, 23). These recommendations have been based on previous studies of the normal maturational process in infancy and childhood (1, 11, 13, 18, 33, 34) and studies of the uninfected as well as the infected babies of HIV-infected mothers (3, 16, 19, 22, 25). In the present study, we examined the maturational changes in peripheral circulating lymphocyte subsets that occurred in a Chinese pediatric population with the specific purpose of monitoring the progression of HIV disease.

There are several limitations to our study: first, we studied only the HIV-negative pediatric population and have not yet obtained sufficient data for HIV-infected pediatric patients. It may also be possible that a proportion of our study population was at a window period of infection and, therefore, that HIV was not detected by our enzyme-linked immunosorbent assay screening. The latter possibility is, however, unlikely to be significant because an unlinked anonymous community-wide screening program showed that the seroprevalence of HIV in the general population is <0.1%. With an impending maternal screening program and the availability of highly active ART, we also expect that the number of HIV-infected pediatric individuals will take some time to accumulate. Second, variations in flow cytometry methods which may lead to erroneous interpretation of test results have been well described (12, 25), although we endeavored to overcome this problem by adopting standardized and quality-controlled laboratory procedures. Third, our study cohort consisted of a group who had been given hepatitis B vaccine whose long-term effects on the immune system with respect to the HIV disease markers are practically unknown, although we assume that the effects should be minimal. The findings of a smaller-scale study with a group of pediatric patients who were admitted for nonacute surgical procedures (data not shown) and whom we tested by the same laboratory procedures were very similar to those that have been described here. This gave us confidence that our results are probably reliable. Also, the possibility of the occurrence in our study subjects of other subacute or subclinical infections which might affect our results cannot be completely excluded. Another weakness is the limited number of study participants in groups A and B, as previously published work (10,14) has shown that marked changes occur within the period of life defined by the ages of age groups A and B. The changes that occur in individuals whose ages were covered by age groups A and B are also very important for understanding pediatric HIV disease. Future studies are required to expand our knowledge of the exact immunological changes in individuals in these age groups.

Previous studies have examined possible differences in lymphocyte markers between populations, among Hispanics (10), and among a group predominantly comprising African Americans (22) and found no significant differences between absolute CD4⁺-lymphocyte values among Caucasians, African Americans, and Hispanics in the study populations. Tollerud et al. (28) also did not find any significant differences in absolute CD4⁺-lymphocyte numbers, CD4⁺ percentages, or CD4⁺:CD8⁺ ratios between Caucasian and African-American teenagers. A study in Europe detailed the changes in the percentages and absolute counts of CD4 and CD8 cells in the first 4 years of life (14). Our previous study showed that both CD4⁺-cell percentages and absolute counts are substantially lower and that NK-cell percentages and absolute counts are higher in our indigenous Chinese HIV-non-infected as well as HIV-infected populations (17, 18).

A comparison of the results obtained from the present study with those obtained for the Caucasian population by similar flow cytometric technologies and with monoclonal antibodies was done (28, 31, 34, 35) (Table 5). We are aware that it may not be appropriate to directly compare the results of two studies, particularly because our group A included individuals from birth to age 1 year, whereas the study with a Caucasian population included only cord blood, and our Chinese group B included those ages 1 to 2 years, whereas Caucasian group B included infants who were <1 year of age. These groups are not completely the same and should not be compared as though they are. Given these caveats, a few very notable and interesting differences still appear. For total white cell counts, the downward trend from cord blood samples to adult samples was apparent in both populations (13). This was also true for the apparent initial rise in percent lymphocytes from birth to the first year of life, before a continuous decline to adulthood. Further comparison of the T-cell subsets revealed that while the percentage of CD3⁺ T cells remained stable throughout the maturational process in both populations, the percentages of CD3⁺ CD4⁺ cells (T-helper cells) were apparently lower for Chinese pediatric individuals from the first year of life onward to adulthood. Such a phenomenon (i.e., lower cell percentages) was also seen for the percent CD3⁺ CD8⁺ cells (T-suppressor cells) as well as the percent B cells (CD19⁺ CD3 cells). However, for NK cells, the reverse was true, namely, that except from birth to the first year of life, the percentage of NK cells was consistently higher in Chinese individuals from the first year of life onward compared with those in their Caucasian counterparts. While this observation may be due to innate differences in genetic constitutions, it is also possible that there were immune stimuli and/or antigens (e.g., environmental mycobacteria) that led to such marked differences. If the latter were the case, the stimuli would have to be chronic ones so that the effects were seen over all age groups studied; the stimulus could also have been a single episode whose effects were long lasting. In the absence of pointers from white cell differential or activation markers, it is impossible to postulate the likely causal step at this stage. A study which examined adults of different racial backgrounds in the United States provided similar findings, although it was not clear whether those individuals were ethnic Chinese or recent immigrants (24). Our study has confirmed the findings of the previous study and further elucidated the actual components involved in the maturational process.

The clinical relevance of these differences between Chinese and Caucasian pediatric populations delineated by the present study (if they do exist) is not known. If Chinese norms for the pediatric population are lower, it may mean that one can safely start PCPP when the CD4 count reaches a lower point or a high risk of P. carinii pneumonia still occurs when the CD4 count reaches the same breakpoints that are described for the Caucasian pediatric population. The present study does not address this question, but it does suggest that a potential for a different CD4 breakpoint for risk might exist in the Chinese pediatric population. Further studies are needed to answer that question.

In conclusion, since present ART and PCPP regimens rely heavily on assessment of peripheral lymphocytes for adjustment of treatment strategies, it is important that the maturational changes and differences detected in the present study be applied conscientiously to the HIV-infected Chinese pediatric population. In particular, as the HIV infection and AIDS epidemic establishes itself in Asia, it would be most relevant to extend the study to the HIV-infected pediatric population to see how HIV affects the different lymphocyte subsets, as has already been done for the Chinese adult population (18). Also, as seen from our findings in the present study, it would be prudent to establish normative ranges for HIV-infected pediatric subjects separately from those for HIV-infected adults.