Age-stratified pediatric reference values of lymphocytes in the Moroccan population

1. Introduction

Lymphocytes are a subset of white blood cells that constitute an integral part of the humoral and cellular immune defence against foreign proteins and pathogens. The number of circulating lymphocytes can either increase (lymphocytosis) during certain infections such as mononucleosis or in lymphoproliferative disorders such as acute or chronic lymphocytic leukemia, or decrease (lymphocytopenia) during viral infections such as human immunodeficiency virus (HIV) or in inherited diseases such as primary immunodeficiencies (PID) [11, 13, 17]. On one hand, many childhood and pediatric diseases and disorders can reduce lymphocyte counts and/or subsets. Examples include, PID, as well as mother-to-child transmitted HIV [21, 28]. Pediatric cancers, on the other hand, can abnormally raise lymphocyte counts [8].

Flow cytometric determination of the total number of lymphocytes and/or lymphocyte subsets plays a particularly important role in the initial assessment and diagnosis of many clinical diseases and disorders, including autoimmunity, immunodeficiencies, infections, and malignancy [17, 40]. The interpretation of routine immunophenotyping performed with flow cytometry is highly dependent on the comparison between measured laboratory values of the sample being tested and normal values obtained from healthy individuals [17, 19]. The total number of white blood cells, including lymphocytes as well as the distribution of its subsets are highly variable in the population. The first National Health and Nutrition Examination Survey in the USA demonstrated that total leucocytes were different according to the age [22]. Other studies have also demonstrated the effect of age on peripheral blood lymphocytes [10]. Another study examined the total leukocyte counts between men and women all ages from ⩽ 25 to ⩾ 71 (at 5 year intervals) and concluded that this number was significantly different between men and women at all ages examined [3]. Similar differences were observed in monocytes, and eosinophils [3]. However, the neutrophil percentage dropped for women above 50, while lymphocyte numbers increased indicating the effect of female hormones on leucocyte counts [3]. Therefore, reference values should be age-stratified.

Other well-known regulators of the immune response include environmental exposure, including prenatal exposure [7], nutrition level [27, 36] and also microbial exposure or the “hygiene hypothesis” [2, 16, 34]. Because total lymphocyte numbers, as well as its subsets, can differ between countries due to local variations in these above-mentioned conditions, multiple studies have investigated ethnic populations to establish the reference laboratory values [14, 18, 24, 29, 33, 35]. Use of domestic reference values is therefore expected to improve the accuracy of flow cytometry-based diagnosis. In a previous study, our group established the lymphocyte reference values for the adult Moroccan population [30]. The aim of this study was to establish the age-stratified normal reference values of blood lymphocyte subsets for the pediatric Moroccan population. Yet, a similar study has not been heretofore performed on this population.

2. Methods

3. Results

The absolute and relative values of the cytotoxic T

lymphocyte subset (CD3 + T cell, CD3 + CD4 + T helper cells, CD3 + CD8 + cytotoxic T cells, CD19 + B cells and CD16 + and/or CD56 + NK cells) including the median, mean and range are presented in Tables 2 and 2, respectively. The values provided in theses tables represent a 90 percent chance that 90 percent of healthy children will fall within this range.

We also compared the changes in the numbers of lymphocytes and its subsets with age. Analyses of the changes in lymphocyte subpopulation numbers in peripheral blood with age showed no statistically significant differences in CD3 + cells between group I and group II or between group III and group IV. Nevertheless, a significant age-associated reduction was observed when either group I or group II was compared to group III, group IV or group V and even when group III or group IV were compared to group V (Fig. 2A). Identical results were obtained for the CD3 + CD4 + population (Fig. 2B). Numbers of CD3 + CD8 + cells, showed a statistically significant increase of total numbers from group I to group II. However, there were no significant differences between groups I and III or between groups I and IV. Significant differences were again noted when group I was compared to group V. Group II was significantly different from groups III, IV and V. Groups III and IV also significantly differed with group V. These results indicate that CD3 + CD8 + cell numbers temporarily rise between ages 1–2 years, before decreasing at ages more than 2 years (Fig. 2C). For CD19 + B cells, we noted that except for between group I and group II, there were statistically significant differences. These results suggest that B cells decrease after age of 2 years (Fig. 2D). Finally, when NK cells number were compared across ages, we observed no significant differences between groups, except for group I versus group V, group II versus group V and group III versus group V. These results indicate that NK cell numbers decrease after 12 years age (Fig. 2E).

OPEN IN VIEWER

OPEN IN VIEWER

Comparing changes in proportions of lymphocytes and lymphocyte subsets, we did not observe any significant difference in proportions of CD3 + lymphocytes between different age groups except for a decrease from group I (0–1 year) to group III (2–6 years) (Fig. 3A). The proportions, however, significantly increased between group III (2–6 years) to group V (12 years-adult) (Fig. 3A). These results indicate that the proportion of lymphocytes dipped slightly at ages 2–6 years before increasing back to earlier levels. In contrast, proportions of CD3 + CD4 + helper T lymphocytes were significantly different in group I from the rest of the groups (Fig. 3B). These results indicate that proportions of CD3 + CD4 + helper T lymphocytes decreased between ages 0–1 years to 1–2 years. Subsequently, the proportions of this lymphocyte subpopulation did not change any further and remained constant. Relative amounts of CD3 + CD8 + cytotoxic T lymphocytes did not appear to change with age except a significant increase in the 12 years-adult group (Fig. 3C). Examination of proportions of CD3 - CD19 + B lymphocytes indicate significant differences between group I and group IV, as well as between group II and group V, group III with group IV and group V, and group IV with group V (Fig. 3D). These results indicate that there is a significant reduction in proportions of B cells around/after year 12. Proportions of NK cells showed no differences except for between group II and group III (Fig. 3E).

Finally, we compared our median reference values to that previously reported for studies performed in Cameroon, Turkey, USA and the Netherlands [9, 29, 31, 33]. We observed that the median value of CD3 + lymphocytes, as well as CD4 + , CD8 + and CD19 + cells were higher in the Moroccan samples than the corresponding medians reported in the Cameroonian [29], Turkish [9] and USA [33] studies, especially in the age groups 0–1 years and 1–2 years (Table 3). For NK cells, the medians for the Moroccan population closely approximated that of Cameroon and USA at ages 0–1 years (Table 3). All three were lower than the Turkish population (Table 3). At 1–2 years the Moroccan population and the Turkish population were similar for median numbers of NK cells, and were higher than the Cameroonian and the USA populations (Table 3). At age 2–6 years, the Moroccan median was the highest amongst the four groups, but again by 6–12 years it dropped below that of Cameroon and Turkey to a level similar to that of USA (Table 3). However, all parameters (CD3 + , CD4 + , CD8 + , CD19 + , as well as NK cells) were lower in the Moroccan population in comparison with Cameroonian, Turkish and USA populations, after age 12 years (Table 3). We also compared the tolerance intervals between our values with a study from the Netherlands [31] -the only one study that explored tolerance intervals-. Out of 50 parameters (10 parameters each for 5 age groups), 16 parameters (32%) showed more than 57% difference (57% to 476%) in the means (Table 4). Noticeably, means of absolute counts and relative proportions of B cells were consistently higher in all age groups in Morocco (55–79%), except $>$ 12 years group (Table 4). The means of relative proportions of NK cells were higher in the Netherlands study across all age groups (Table 4). We observe also concordance of the means, 5 th and 95 th percentile values

between the two populations for relative proportions of CD3 + , CD4 + and CD8 + as well as absolute counts of CD4 + (Table 4).

The reference values of peripheral blood lymphocyte subsets were determined by flow cytometry immunophenotyping to aid in the diagnosis of the various immunological defects and dysfunction such as HIV infections [11, 39] or PID [12, 25, 26] in the Moroccan population. Paediatric HIV patients have reduced CD4 + counts [1]. Untreated HIV causes progressive CD4 + loss [6]. PID encompasses heterogenous, mostly inherited disorders that affect the development, function or both of several lymphocyte subsets [25]. For example, in PID, mutations may result in the loss of expression of markers such as CD19 [37]. Additionally, PID can be caused by mutations in genes, that affect the terminal differentiation of a specific lineage of lymphocytes, such as mutations in BTK that leads to the near complete loss of B cells [5, 15]. Such changes can be revealed by flow cytometric analyses and careful comparison with reference values. Individuals affected by HIV can develop immunological diseases, infections and oncological complications, as well as several other common morbidities. PID patients are high susceptible to routine and/or recurrent infections of a variety of organs including the skin, throat, ears, lungs, urinary or intestinal tracts [20]. These patients are also at a higher risk of failing to clear the infections or unusually severe degree of infection. An early diagnosis of HIV or PID by flow cytometry and comparison with reference values is critical for preventing dangerous disease-associated morbidity and mortality [20].

Since lymphocyte numbers and subsets are known to vary regionally, in a previous study our group established lymphocyte reference values for the adult Moroccan population [30]. In this study we established lymphocyte reference values for the children Moroccan population at ages 0–1 years, 1–2 years, 2–6 years, 6–12 years and $>$ 12 years. Results of lymphocytes immunophenotyping with flow cytometry can be influenced by many variables such as the anticoagulant used, sample preparation, monoclonal antibodies used, selected staining protocol and technical platform. In this study, samples were treated according to the single platform method-lyse no-wash procedure and the gating strategy recommended by the manufacturer. Eighty-three individuals were carefully selected, to meet the standards of healthy subjects. Our results demonstrate that the CD8 + T cell numbers increased after early infancy [16]. This may result from increased antigen exposure in the first months of life. The decreasing total numbers of T and B cells at adult age observed in our studies is also consistent with previous reports [4, 9, 32, 38]. NK cell numbers, in contrast, appear to be stable from 0–1 year till 6–12 years, similar to that reported by Ikinciogullari et al. [9]. In contrast to total numbers, the percentage values of lymphocyte subsets did not change substantially with age. This result confirms previous reports that change in absolute count of lymphocytes are not consistent with the changes in their percentages [4].

Importantly, we observed increased median values of CD3 + and CD4 + cells at 0–1 years and 1–2 years of age in Moroccan population compared to that of Cameroon, Turkey and the USA [9, 29, 33]. These differences may be due to different antigen exposure. It has been reported that children who encounter more antigens in their environment present with a higher count of lymphocytes [24]. Alternatively, environmental factors, genetics and/or nutritional status may contribute to these differences [2, 16, 27, 34, 36].

5. Conclusion

Our study quantifies the changing counts and percentages of lymphocytes subpopulation in healthy Moroccan children from birth to adulthood. These reference values should be useful for Moroccan laboratories for the correct interpretation of lymphocytes immunophenotyping in the diagnosis of HIV, PID and other diseases characterized by lymphocytosis or lymphocytopenia in pediatric patients.