Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Nutrition and Health (including climate and ecological aspects)

Reference cut-offs to define low serum zinc concentrations in healthy 1–19 year old Indian children and adolescents

A Correction to this article was published on 06 April 2022

This article has been updated

Abstract

Background/Objectives

Population zinc (Zn) status assessment is based on serum zinc concentration (SZC) cut-offs defined by the International Zinc Nutrition Consultative Group (IZiNCG). The objective of this study is to derive reference SZC cut-offs in apparently healthy 1-19 year Indian children and adolescents using comprehensive national nutrition survey (CNNS) data, and to measure the prevalence of Zn deficiency.

Subjects/Methods

Apparently healthy children (n = 12,473) were selected from the CNNS, by including the highest 2 wealth quintiles, and excluding stunted, thin and obese children, and those with CRP > 5 mg/L, anaemia, hypo-albuminemia, diabetes, recent diarrhoea and history of smoking. The 2.5th centile of age-based distributions defined the SZC cut-offs, used to measure the prevalence of Zn deficiency in India, as against the IZiNCG cut-offs.

Results

The present study SZC cut-offs were significantly lower, by 10–18 µg/dL, than the IZiNCG cut-offs; more in adolescents. Prevalence of Zn deficiency in the entire CNNS, with these cut-offs, was 2.7 (<10 years) to 5.5 (10-19 years) times lower than with the IZiNCG cut-offs. No geographical state, nor any age group, had Zn deficiency as a serious public health problem (≥20%). In contrast, with IZiNCG cut-offs, 9-27 states (depending on age group) had a public health problem.

Conclusions

The present study reference SZC cut-offs for Zn deficiency are lower than the IZiNCG cut-offs, and their rigorous selection from a national sample makes them more appropriate for use in India. A re-examination of the global applicability of IZiNCG recommended cut-offs in other LMICs appears appropriate.

Introduction

Zinc (Zn) is an essential type-2 nutrient required for growth and general health whose deficiency manifests in impaired growth and an increased risk of infectious morbidity [1,2,3,4]. It is considered a serious public health problem, when it occurs in ≥20% of the population [5]. Serum Zn concentration (SZC) is the most widely used biomarker for assessing Zn deficiency. Age and sex-specific SZC cut-offs to assess Zn deficiency were defined by the International Zinc Nutrition Consultative Group (IZiNCG), using the National Health and Nutrition Examination Survey (NHANES)-II, conducted about 4 decades ago [5]. These cut-offs were based on the 2.5th centile of SZCs for different age groups (<10 and >10 years), stratified by sex and fasting status, and are globally applied.

However, the global application of these SZC cut-offs merits examination, especially in LMIC settings, where contextual factors, including habitual lower dietary intakes, could result in a left-shift of the normal, healthy SZC distribution [6,7,8]. Even though the SZC is tightly regulated, it can change from 3-15% (mean 9%) when Zn intakes change two-fold in children [9], by about 22% in daily circadian excursions [10] and by 11% during long-term dietary adaptation to a vegetarian diet [6, 8]. In LMIC settings, the consumption of fortified foods led to mean increase of 12 µg/dL in SZC compared to unfortified foods [11], in an otherwise healthy population. SZCs is known to respond to Zn supplementation [12, 13], and are correlated with non-specific clinical symptoms of Zn deficiency [14]. It is therefore possible that in healthy populations with lower, but still adequate, habitual Zn intakes, the SZC distribution can shift to the left, resulting in lower SZC cut-offs, which while entirely appropriate for such populations, can have substantial effects on their estimated Zn deficiency prevalence based on IZiNCG cut-offs.

While re-examining SZC cut-offs in LMICs, the IZiNCG method should be followed, using the 2.5th centile of the SZC distribution in a well-defined healthy population to derive the cut-off [5]. Thus, representative SZCs from the population are required, along with other variables to allow a careful definition of health. We selected SZC data and other biomarkers of apparently healthy Indian children and adolescents aged 1–19 years from a nationally representative Comprehensive National Nutrition Survey (CNNS) [15,16,17] to rigorously evaluate and define age- and sex-specific SZC cut-offs for Zn deficiency, and then applied these cut-offs to the entire CNNS sample, to estimate the prevalence of inflammation-corrected Zn deficiency.

Methods

The CNNS [17], a cross-sectional, nationally representative survey of Indian children and adolescents, was conducted in 30 states of India, between 2016-2018. The Population Council’s International Review Board (New York, USA) and the Ethics Committee of the Post Graduate Institute of Medical Education and Research (Chandigarh, India) gave ethical approval [17]. The survey design and sampling methodology of the CNNS are published elsewhere [15,16,17]. Briefly, a multi-stage, population proportional to size cluster sampling was done to enrol preschool (1–4 years), school age (5–9 years) children, and adolescents (10–19 years), to adequately represent the national, state, male-female, and urban-rural population. For blood sampling, 50% of all the children who completed anthropometry were contacted through systematic random sampling. Children and adolescents with physical or cognitive disabilities, chronic illness, infectious illness, fever, severe injury and pregnancy were excluded. Informed written assent (<10 years) from caregivers and consent (>10 years) from the subjects was obtained prior to data collection [17].

Socioeconomic and demographic characteristics of households, anthropometric data of one child/adolescent per age group and history of morbidity in the preceding two weeks, were collected from each household; these methods are detailed elsewhere [17]. The Wealth Index was computed as described in the National Family Health Survey-4 [18]. Access to facilities like drinking water, hand washing and sanitation was categorized based on the Joint Monitoring Program guidelines of the World Health organization (WHO)/ United Nations Children’s Emergency Fund (UNICEF) [19]. Age- and sex-standardized height for age (HAZ), weight for age (WAZ) and body mass index (BMI)-for-age Z-scores were calculated using the WHO Growth References [20, 21].

Blood sample collection procedure and biomarker analysis methods are detailed elsewhere [15,16,17]. Briefly, a day prior to sample collection, parents and children were instructed to ensure overnight fasting. Venous blood samples along with information on fasting status (yes/no) and time of sample collection were obtained by trained phlebotomists, and a 4 mL aliquot of blood was collected in trace element free tubes (Red top with yellow ring, Greiner Bio One, India) and were transported to the nearest laboratory in cool bags (3L-12H-08P, PronGo). Serum was separated within 6 h from the time of sample collection. A simulation study that mimicked the field conditions showed that there was no difference in the SZC of blood samples that were either stored at 2-8 °C or at room temperature (22–30 °C), when serum was separated in less than 6 h [22]. Serum Zn was estimated by atomic absorption spectrometry with D2 correction (Perkin Elmer, Analyst 600, USA), against Zn standards (MERK, India) in a commercial laboratory (SRL Labs, Mumbai, India); the samples were routinely checked and excluded if the samples were haemolysed, had clots or when the quantity was insufficient for analysis. Instrument calibration was performed every day with Zn standards (MERK, India). In-house quality control serum samples (2 levels) were also run every day and trends assessed in a Levey-Jennings (LJ) chart prior to daily sample analysis. The coefficient of variation (CV) of the method was 6.7%. In addition, rigorous internal and external quality control procedures were implemented in the CNNS survey [15,16,17].

The selection of an apparently ‘healthy’ sample of children from the entire CNNS survey was based on a combination of biomarkers, sociodemographic and anthropometric exclusion cut-offs, and is depicted in Table 1. Briefly, a primary analytic sample (analytic sample 1, n = 12,473) was selected after applying filters (as detailed in Fig. 1) for serum CRP > 5 mg/L [23], anaemia [24], hypo-albuminemia [25], height-for-age, BMI for age, recent diarrhoea, elevated HbA1c [26], poverty and a history of smoking. Additional filters of severe wasting, severe thinness and severe underweight, recent fever, and unimproved drinking water and sanitation yielded analytic sample 2 (n = 9966). A further filter that only selected the richest wealth quintile (since from the first filter, the two top quintiles for wealth were selected) yielded analytic sample 3 (n = 6060).

Table 1 Biomarker, sociodemographic and anthropometric exclusion cut-offs for selection of apparently healthy population.
Fig. 1: Flowchart of participant exclusion.
figure 1

The participants meeting the exclusion criteria were sequentially excluded for selection of primary and the two analytical samples. Serum albumin, and glycosylated haemoglobin (HbA1c) assessments were done only for children >5 years age. CRP C-reactive protein, SZC serum zinc concentration.

Initially, we stratified subsamples by age group 1–4 y, 5–9 y, 10–14 y and 15–19 y, as well as by sex and fasting status and compared density plots of the SZC distribution across these strata. While symmetric, we smoothed the distribution of SZC in specific strata over each year of age for each sex and by fasting status, by the generalized additive model for location scale and shape (GAMLSS) with Box-Cox-Cole-Green transformation [27], to avoid undue temporal variability owing to unequal sample size across ages. Further age-specific distributions were derived by averaging the year specific distributions within age groups; the same process was repeated for the four strata that occurred by sex and fasting status. The <5th centile and >95th centile of SZC were excluded from each analytic-sample to avoid over-dispersion due to unobserved factors that could systematically influence the distribution. Next, the 2.5th and 5th centiles of the SZC distribution at yearly intervals in two age groups (1–9 years and 10–19 years) of all strata of sex- or fasting status were derived from GAMLSS. We considered the 5th centile additionally as a less rigorous biomarker than the commonly used 2.5th centile. The 95% confidence intervals of centiles were derived with normal probability distribution [28]. The estimated 2.5th and 5th centile of SZC distribution of the primary analytic sample, with their respective 95% CI, at yearly intervals were compared with the IZiNCG cut-offs. In sensitivity analyses, these cut-offs were also estimated for analytic subsamples 2 and 3, and compared with estimates from the primary analytic sample. Finally, the 2.5th or 5th centile cut-offs from the primary analytic sample (the eventual study cut-offs) were used to determine the inflammation-corrected prevalence of Zn deficiency [29], and these were compared with estimates based on the existing IZiNCG SZC cut-offs for the entire sample with paired serum Zn and CRP concentrations. The statistical analyses were conducted using Statistical software R version 4.1.0 (R Core Team, 2021, Vienna, Austria).

Results

The sociodemographic characteristics of the primary analytic sample in three broad age groups (1–4, 5–9, and 10–19 years) are summarized in Supplementary Table 1. The male/female ratio in the 10-19 years age group (1.39) was higher compared to that in the 1–4 (1.13) and 5–9 (1.17) year age groups. The proportion of rural (~40%) and urban (~60%) participants was similar across the age groups. Approximately half of the participants were from north (23–31%) and south (21.7–23%) India, and a lower proportion from north east (19–24.4%), east (9.6–13.3%), west (11.1–13.2%), central (3.7–5.5%) India. Between 15% and 21% of the participants reported a history of fever 2 weeks preceding the survey, 8–15% were underweight (WAZ < −2SD), and 15–18% were thin (BMIZ < −2SD) among 5–19-year age group, while 9% of 1–4 years children were wasted (WHZ < −2SD). The majority of included participants had piped water source (81–82%) and improved sanitation (74–78%) facilities in their households. In the primary analytic sample, the proportion of fasting samples was low in the 1–4 years age group (18%), compared to the 5–9 (89%) or 10–19 years (91%) age groups (Table 2).

Table 2 Mean and 2.5th centiles of serum zinc concentrations (µg/dL) by age, gender and fasting status of primary analytic sample.

Across all age groups (1–4, 5–9, 10–14 and 15–19 years), the SZCs were normally distributed both for the fasted and non-fasted states (Fig. 2). In comparison to the 1-4 years age group, either the mean or the 2.5th centile of SZCs were slightly lower in the 5–9 and 10-14-year age groups regardless of fasting status; however, there was a rebound with slightly higher values in the 15–19-year boys, but such a pattern was not observed in girls (Table 2). When stratified by fasting status, the mean or 2.5th centile SZCs were slightly lower in non-fasted 1-9-year males and 1-4-year females (Table 2), but not in other age/gender groups. Although the 2.5th and 5th centile SZCs in non-fasted subjects tended to be lower in comparison to fasted subjects, they remained statistically similar across the age groups (Supplementary Fig. 1)

Fig. 2: The distribution of serum zinc concentrations (µg/dL) stratified by age, gender and fasting status.
figure 2

Distribution of SZCs in fasting (upper panel) and non- fasting (bottom panel) children and adolescents, across different age groups. Distributions are presented using nonparametric density plots.

When grouped according to IZiNCG reference age groups, namely, 1–9 years (boys and girls combined) and 10–19 years (boys and girls separately), the mean SZCs were quite similar in each group regardless of their fasting status (Table 3). The 2.5th and 5th centiles of SZC were lower (~1 µg/dL) in non-fasted compared to fasted subjects in both female and male participants. In sensitivity analyses using the three analytic subsamples, the 2.5th and 5th centile cut-offs in the various subgroups were also similar (Table 3, overlapping 95% CIs).

Table 3 Mean, 2.5th and 5th serum zinc concentrations (µg/dL) centiles of primary analytic sample and analytic samples 2 and 3, stratified by age groups and fasting status.

The 2.5th (Fig. 3) and 5th (Supplementary Fig. 2) centiles of SZC in the present study were significantly lower than the IZiNCG cut-offs; the difference was 9–10 µg/dL in <10 years age, and 13–18 µg/dL in 10–19 years old (Supplementary Table 2). In general, the 2.5th centile values were 4–5 µg/dL lower than the corresponding 5th centiles (Table 3).

Fig. 3: Comparison of 2.5th centile serum zinc concentrations from the primary analytic sample with existing IZiNCG cut-offs.
figure 3

The dots indicate 2.5th centile mean SZCs and vertical lines are 95% CI. The dashed line indicate SZC cut-offs of IZiNCG.

We then assessed the inflammation-adjusted prevalence of Zn deficiency, using the study (2.5th centile) and IZiNCG cut-offs among all the participants in the CNNS survey with valid serum Zn estimates. The weighted inflammation-adjusted national prevalence (%) of Zn deficiency with the present study cut-offs in different age groups were 2.7 to 5.5 times lower when compared to the prevalence estimated by the IZiNCG cut-offs (6.0 vs 17.4% in 1–4 year; 5·7 vs 15.8% in 5–9 year; and 5.6 vs 31.1% in 10–19-year age groups). When evaluating geographic state-based prevalence estimates of Zn deficiency with the present study cut-offs, none indicated a public health problem (≥20%), in contrast to the IZiNCG cut-offs, which suggested that 12 (for 1–4 year), 9 (5–9 year) and 27 (10–19 year) states had a serious public health problem (Fig. 4). Indeed, even when using the higher, more conservative 5th centile SZC cut-offs from the present study, only 3 states showed a prevalence estimate of ≥20% (Supplementary Fig. 3).

Fig. 4: Weighted mean (95% CI) prevalence of low SZC in Indian children and adolescent with study cut-offs (2.5th centile) and IZiNCG cut-off.
figure 4

The inflammation and fasting status–adjusted prevalence of low SZCs among Indian states. The prevalence estimates of low SZCs are based on 2.5th centiles derived in this study (new) or based on IZiNCG recommended cut-offs (old). The dot indicates the mean, and horizontal line is 95% CI. The dotted vertical lines intersecting at 20% prevalence was provided to highlight public health significance.

Discussion

This study provides reference centiles of SZC for 1–19 years old apparently healthy participants in a LMIC, selected through stringent criteria, from the nationally representative and quality-controlled CNNS survey [15,16,17]. Two study cut-offs (both 2.5th and 5th centiles) are presented for identifying low SZCs, and both were lower than the present IZiNCG recommendations for corresponding sex, age and fasting status groups; indeed, the present study centile that corresponded to the IZiNCG cut-off value was the 10th for 1-9-year children, and 24th and 30th centile for 10–19 year fasting female and male children, respectively, indicating large leftward shift has occurred in the distribution of apparently healthy subject SZCs from a LMIC population. However, since the IZiNCG considered the 2.5th centile to define low SZC or Zn deficiency, this cut-off is considered in subsequent discussions below.

The stability of the primary analytic sample cut-offs to sensitivity analyses with other analytic samples, with even more stringent exclusion criteria, inspire confidence in the findings. There is also external validity to the proposed SZC cut-offs in this study (56 µg/dL), since this was close to the 2.5th centile reported in under-five Australian children (59 µg/dL) [5, 30]. Direct comparisons of present study SZC cut-offs with other studies (compiled in Supplementary Table 3) is problematic, due to differences in reported exclusion criteria (sometimes even undefined), different use of reference threshold centiles, and different age groups studied (children, adolescents and adults). Further, none of the collated studies excluded subjects with inflammation. Even so, most studies reported SZC cut-offs lower than the IZiNCG recommendations.

Using the present study cut-offs resulted in substantially lower national prevalence of Zn deficiency than defined using the IZiNCG cut-offs, as we have previously reported [15]. Thus, no geographic state of India had a prevalence that suggested a significant public health problem (≥20%), in contrast to 12 (1–4 y), 9 (5–9 y) and 27 (10–19 y) states, when the IZiNCG cut-offs were used. We believe that the present estimates reflect the true burden of risk of Zn deficiency in India, with several important implications: first, a substantially lower deficiency burden than that previously estimated [15], changes the picture from a serious public health problem requiring universal supplementation to one that might warrant only targeted interventions, with a significantly lower economic burden. Second, the significantly lower prevalence of Zn deficiency in India might be an additional explanation for the lack of efficacy of Zn supplementation on linear growth in Indian infants, even with a high stunting prevalence of about 60% at 18 months [31]. Further, Zn is a type 2 nutrient, whose deficiency leads to impaired growth to conserve the body zinc [1, 2]. Therefore, an association of zinc deficiency with prevalence of stunting is expected. However, in a previous study prevalence of low SZC was not associated with stunting in <5 years children in India, after adjusting for confounders[15]. In addition, SZCs are also not associated with HAZ of under 5 year old children in the CNNS data (Supplementary Fig. 4), from which healthy children were selected for the present study. These findings lend support to a much-needed policy shift to provide a medley of nutrients through diverse food, rather than single nutrient fortification of staple foods or supplementation.

The stringent recommendations of IZiNCG for sampling serum Zn were followed while assessing the SZC in the present study [32]. The normal distribution of SZCs across age groups and quality control reaffirm the robustness of the SZC values used here. The cut-offs are based on a fairly large sample size with adequate representation by sex, age groups, rural-urban differentiation and regional representation. In fact, the sample size of the primary analytic sample (12473, 1-19 years) was higher than that of NHANES II data analyzed by IZiNCG (11859, 3–74 years), and with a narrower dispersion of age. It is also noteworthy that the exclusion criteria adopted in this study were more stringent compared to IZiNCG [5], wherein anaemic, obese and diabetic subjects were additionally excluded [33, 34], while inflammation was accounted for by CRP evaluation [23]. Further, we also excluded subjects with poor socioeconomic indicators or under-nutrition based on WHO and/or UNICEF criteria [18,19,20,21].

The SZCs were lower in fed compared to fasted subjects, as well as greater in boys compared to girls beyond the age of 10 years; this observation is consistent with earlier literature [5, 35], and corroborates our previous analyses [15]. However, in contrast to the linear age-dependent increase in SZCs (between 6–20 year of both male and female subjects) in the NHANES-II data considered by IZiNCG while setting the SZC cut-offs [4], the mean SZCs were similar across age groups in this study. In agreement, more recent NHANES (2011-2014) [35] data also have shown no significant differences in SZCs in those aged 6–9 years or >10 years despite an age-dependent increase in Zn intakes. Similarly, no significant association of age with SZC was found in US [36] or Iranian/Bangladesh children and adolescents [37, 38]. In support, the 2.5th centiles of the SZC distribution calculated from healthy subjects of a later NHANES (2011-2014) survey are 2 µg (<10 years) and 7–8 µg (>10 years) lower than the IZiNCG cut-offs [35].

Plausible explanations for the substantial difference in SZCs reference cut-offs in Indians compared to IZiNCG, could be due to higher habitual Zn intakes in reference US populations. Indeed, dietary Zn intakes of 14-18 year old male adolescents in US was 14.4 ± 0.62 mg/d [35], while in India the intakes are half of this level; at 7.7 to 8.5 mg/d (SD 2.77 mg) [39]. A systematic review showed that Zn intakes are indeed associated with SZCs, and that doubling of Zn intakes results in 9% higher SZCs [9], which should translate to similar reduction in SZCs. The current study SZC cut-off (56 µg/dL) is about 25% lower than IZiNCG cut-off (74 µg/dL) [5] or 14% lower compared to the SZC cut-off derived from the later NHANES 2011-14 survey (66 µg/dL) [35]. Dietary Zn bioavailability is another important factor, as Swedish adult volunteers who switched their diets from mixed to lacto-vegetarian, but kept their Zn intake constant, showed a reduced plasma Zn concentration (by 11-13%, and expected due to the vegetarian diet) and reduced faecal and urinary excretion of Zn after 3 months [6]. It is important to note that after this initial fall, the plasma Zn concentration remained static at the lower value thereafter for 12 months without any apparent changes in the subjects’ general health, showing that there is a range of plasma (or serum) Zn concentrations that are compatible with good health. Therefore, the lower distribution of SZCs in apparently healthy children in the present study, with resultant lower cut-offs, should not be construed to be a “lowering of the standard”, as it was defined on stringently selected healthy children, and the observed left shift of the SZC distribution is a reflection of lower intakes, and perhaps long-term successful adaptation to typical vegetarian diets in India by optimizing the absorption and increasing the whole body Zn retention [40,41,42]. If this were to be true, a shift in Zn turn-over rates for avid retention of body Zn is expected, which needs to be measured. Nevertheless, the dietary Zn intakes are not correlated with SZC in US population [35], which could be due to a higher flux of absorbed Zn through faecal and urinary Zn excretion, after SZC reaching their highest possible homoeostatic values: a potential ceiling effect. For example, the Institute of medicine (IOM, USA) estimated daily average requirement (EAR) for Zn is 9.4 and 6.8 mg/day for men and women respectively [43]. However, the mean daily per capita Zn intake in the USA (or any HIC) is 13.2 mg/day [44]. This is almost double the requirement for a woman, and certainly much higher than the requirement for men.

There may also be an alternative constitutional explanation relied on differences in body composition. Although body weight is not directly associated with SZC values [35], it is thought that the exchangeable Zn pool (EZP), a more sensitive measure of Zn status, is associated with body weight, more so the lean body mass [45, 46]. The Indian (S Asian) population have both lower weight and historically have had a low lean mass [47, 48], compared to Western populations. Further, plasma Zn was one of 5 variables, including weight, age, sex and total absorbed Zn, that appeared to predict the EZP in adults and children [49]. Similar findings have been found elsewhere; in measurements of EZP in younger and older Korean women, the mean EZP and mean SZC moved in the same direction between groups [50]. It is more likely that the lower SZC cut-offs in the present study could be a reflection of cumulative effect of all these factors, such as lower Zn intakes, limited bioavailability and lower lean body mass, which is still compatible with good health.

It is worth pointing out that the cut-offs of the present study are closer to SZC cut-offs (although 4-6 µg/dL higher) that have been proposed to predict clinical signs associated with Zn deficiency. In a systematic review among subjects with Zn deficiency induced by either extreme acute dietary Zn restriction (<1 mg/d) or with genetic mutations that restrict intestinal Zn absorption, Wessels et al. [14] reported that a plasma Zn cut-off of 50 µg/dL predicted the appearance of clinical signs of Zn deficiency with 92% specificity and 82% sensitivity, while a value of 60 µg/dL predicted this with 75% specificity and 88% sensitivity. It should be noted that those restriction studies where clinical signs were either not reported or did not appear, were excluded. There is external validity to the present study cut-offs as well, since they are close to the 2.5th centile SZCs (57.5 µg/dL) of ≥10 year old subjects in the NHANES 2011-14 survey, where no clinical signs were reported [35].

The following limitations merit consideration. First, our stringent exclusion criteria to define health resulted in 60% loss of sample, but this was due to the abundant caution used in selecting a apparently healthy population for defining cut-offs. Second, the sample of non-fasting subsamples was quite low in the higher age groups. Third, while we excluded subjects with high CRP to account for inflammation, the survey did not measure α1-acid glycoprotein, which is a marker of chronic inflammation. Finally, the survey study design did not include infants (aged 0–12 months).

In conclusion, this study provides age-, sex- and fasting status-specific SZC cut-offs for diagnosing Zn deficiency in 1-19-year-old children and adolescents, suitable for use in India or similar LMICs. However, considering the statistical approach used in the study, the proposed cut-offs should be applicable in assessing the risk of Zn deficiency in populations, and should not be used in clinical settings. Given the low (uniformly <20%) estimated prevalence of Zn deficiency in Indian children, there is no need to initiate universal staple food fortification to increase Zn intakes. Based on the quite substantial differences compared with current IZiNCG cut-offs, we conclude that that the one-size-fits-all approach needs a thorough re-evaluation with empirical data, and suggest systematic studies in different settings to understand contextual differences.

Data availability

The Ministry of Health and Family Welfare (MoHFW), Government of India, owns the CNNS data.

Change history

References

  1. Golden MH. Specific deficiencies versus growth failure: type I and type II nutrients. SCN N. 1995;12:10–14.

    Google Scholar 

  2. King JC, Shames DM, Woodhouse LR. Zinc homeostasis in humans. J Nutr. 2000;130:1360S–1366S. https://doi.org/10.1093/jn/130.5.1360S

    CAS  Article  PubMed  Google Scholar 

  3. Yakoob MY, Theodoratou E, Jabeen A, Imdad A, Eisele TP, Ferguson J, et al. Preventive zinc supplementation in developing countries: impact on mortality and morbidity due to diarrhea, pneumonia and malaria. BMC Public Health. 2011;11:S23 https://doi.org/10.1186/1471-2458-11-s3-s23

    Article  PubMed  PubMed Central  Google Scholar 

  4. Imdad A, Bhutta ZA. Effect of preventive zinc supplementation on linear growth in children under 5 years of age in developing countries: a meta-analysis of studies for input to the lives saved tool. BMC Public Health. 2011;11:S22 https://doi.org/10.1186/1471-2458-11-S3-S22

    Article  PubMed  PubMed Central  Google Scholar 

  5. Brown KH, Rivera JA, Bhutta Z, Gibson RS, King JC, et al. International Zinc Nutrition Consultative Group (IZiNCG) technical document #1. Assessment of the risk of zinc deficiency in populations and options for its control. Food Nutr Bull. 2004;25:S99–203.

    Article  Google Scholar 

  6. Srikumar TS, Johansson GK, Ockerman PA, Gustafsson JA, Akesson B. Trace element status in healthy subjects switching from a mixed to a lactovegetarian diet for 12 mo. Am J Clin Nutr. 1992;55:885–90. https://doi.org/10.1093/ajcn/55.4.885

    CAS  Article  PubMed  Google Scholar 

  7. Saunders AV, Craig WJ, Baines SK. Zinc and vegetarian diets. Med J Aust. 2013;199:S17–21.

    Article  Google Scholar 

  8. King JC, Brown KH, Gibson RS, Krebs NF, Lowe NM, Siekmann JH, et al. Biomarkers of Nutrition for Development (BOND)-Zinc Review. J Nutr. 2015;146:858s–885s. https://doi.org/10.3945/jn.115.220079

    Article  PubMed  Google Scholar 

  9. Moran VH, Stammers AL, Medina MW, Patel S, Dykes F, Souverein OW, et al. The relationship between zinc intake and serum/plasma zinc concentration in children: a systematic review and dose-response meta-analysis. Nutrients. 2012;4:841–58. https://doi.org/10.3390/nu4080841

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  10. King JC. Yet again, serum zinc concentrations are unrelated to zinc intakes. J Nutr. 2018;148:1399–401. https://doi.org/10.1093/jn/nxy190

    Article  PubMed  Google Scholar 

  11. Shah D, Sachdev HS, Gera T, De-Regil LM, Peña-Rosas JP. Fortification of staple foods with zinc for improving zinc status and other health outcomes in the general population. Cochrane Database Syst Rev. 2016;6:Cd010697. https://doi.org/10.1002/14651858.CD010697.pub2

  12. Lowe NM, Fekete K, Decsi T. Methods of assessment of zinc status in humans: a systematic review. Am J Clin Nutr. 2009;89:2040s–2051s. https://doi.org/10.3945/ajcn.2009.27230G

    CAS  Article  PubMed  Google Scholar 

  13. Hess SY, Peerson JM, King JC, Brown KH. Use of serum zinc concentration as an indicator of population zinc status. Food Nutr Bull. 2007;28:S403–429. https://doi.org/10.1177/15648265070283S303

    Article  PubMed  Google Scholar 

  14. Wessells KR, King JC, Brown KH. Development of a plasma zinc concentration cutoff to identify individuals with severe zinc deficiency based on results from adults undergoing experimental severe dietary zinc restriction and individuals with acrodermatitis enteropathica. J Nutr. 2014;144:1204–10.

    CAS  Article  Google Scholar 

  15. Pullakhandam R, Agrawal PK, Peter R, Ghosh S, Reddy GB, Kulkarni B, et al. Prevalence of low serum zinc concentrations in Indian children and adolescents: findings from the Comprehensive National Nutrition Survey 2016–18. Am J Clin Nutr. 2021;114:638–48. https://doi.org/10.1093/ajcn/nqab066

    Article  PubMed  Google Scholar 

  16. Sachdev HS, Porwal A, Acharya R, Ashraf S, Ramesh S, Khan N, et al. Haemoglobin thresholds to define anaemia in a national sample of healthy children and adolescents aged 1-19 years in India: a population-based study. Lancet Glob Health. 2021;9:e822–e831. https://doi.org/10.1016/S2214-109X(21)00077-2

    Article  PubMed  PubMed Central  Google Scholar 

  17. CNNS. Ministry of Health and Family Welfare (MoHFW), Government of India, Comprehensive National Nutrition Survey (CNNS) National Report. New Delhi. In, 2019.

  18. NFHS. International Institute for Population Sciences (IIPS) and ICF. In: National Family Health Survey (NFHS-4), 2015–16: India. Mumbai: IIPS, 2015–16.

  19. WHO/UNICEF (2008) Joint Monitoring Programme (JMP) for Water Supply, Sanitation and Hygiene. JMP methodology 2017 update & SDG baselines [Internet]. Geneva (Switzerland) and New York (USA): JMP; 2018. Available from: https://washdata.org/sites/default/files/documents/reports/2018-04/. Accessed on 3rd March 2020.

  20. WHO. Growth reference data for 5–19 years: BMI-for-age (5–19 years) [Internet]. Geneva (Switzerland): WHO. Available from: https://www.who.int/toolkits/growth-reference-datafor-5to19-years/indicators/bmi-for-age. Accessed on 3rd March 2020.

  21. WHO. World Health Organization child growth standards: length/height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index-for-age: methods and development. 2006.

  22. Abraham RA, Agrawal PK, Acharya R, Sarna A, Ramesh S, Johnston R, et al. Effect of temperature and time delay in centrifugation on stability of select biomarkers of nutrition and non-communicable diseases in blood samples. Biochem Med. 2019;29:020708. https://doi.org/10.11613/bm.2019.020708

    Article  Google Scholar 

  23. McDonald CM, Suchdev PS, Krebs NF, Hess SY, Wessells KR, Ismaily S, et al. Adjusting plasma or serum zinc concentrations for inflammation: biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia (BRINDA) project. Am J Clin Nutr. 2020;111:927–37. https://doi.org/10.1093/ajcn/nqz304

    Article  PubMed  PubMed Central  Google Scholar 

  24. Garcia-Casal MN, Pasricha S-R, Sharma AJ, Peña-Rosas JP. Use and interpretation of hemoglobin concentrations for assessing anemia status in individuals and populations: results from a WHO technical meeting. Ann NY Acad Sci. 2019;1450:5.

    Article  Google Scholar 

  25. Kim S, McClave SA, Martindale RG, Miller KR, Hurt RT. Hypoalbuminemia and clinical outcomes: what is the mechanism behind the relationship? Am Surg. 2017;83:1220–7. https://doi.org/10.1177/000313481708301123

    Article  PubMed  Google Scholar 

  26. Expert panel on integrated guidelines for cardiovascular health and risk reduction in children and adolescents: summary report. Pediatrics. 2011;128:S213–56. https://doi.org/10.1542/peds.2009-2107C

  27. Cole TJ, Green PJ. Smoothing reference centile curves: the LMS method and penalized likelihood. Stat Med. 1992;11:1305–19. https://doi.org/10.1002/sim.4780111005

    CAS  Article  PubMed  Google Scholar 

  28. Chakraborti S, Li J. Confidence interval estimation of a normal percentile. Am Statistician. 2007;61:331–6. https://doi.org/10.1198/000313007×244457

    Article  Google Scholar 

  29. Ghosh S, Kurpad AV, Sachdev HS, Thomas T. Inflammation correction in micronutrient deficiency with censored inflammatory biomarkers. Am J Clin Nutr. 2020;113:47–54. https://doi.org/10.1093/ajcn/nqaa285

    Article  Google Scholar 

  30. Karr M, Mira M, Causer J, Earl J, Alperstein G, Wood F, et al. Age-specific reference intervals for plasma vitamins A, E and beta-carotene and for serum zinc, retinol-binding protein and prealbumin for Sydney children aged 9-62 months. Int J Vitam Nutr Res. 1997;67:432–6.

    CAS  PubMed  Google Scholar 

  31. Radhakrishna KV, Hemalatha R, Geddam JJ, Kumar PA, Balakrishna N, Shatrugna V. Effectiveness of zinc supplementation to full term normal infants: a community based double blind, randomized, controlled, clinical trial. PloS one. 2013;8:e61486. https://doi.org/10.1371/journal.pone.0061486

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  32. IZiNCG. International Zinc Nutrition Consultative Group. Technical Brief No. 2. Assessing population zinc status with serum zinc concentration, 2nd Edn. 2012.

  33. Di Martino G, Matera MG, De Martino B, Vacca C, Di Martino S, Rossi F. Relationship between zinc and obesity. J Med. 1993;24:177–83.

    PubMed  Google Scholar 

  34. Houghton LA, Parnell WR, Thomson CD, Green TJ, Gibson RS. Serum zinc is a major predictor of anemia and mediates the effect of selenium on hemoglobin in school-aged children in a nationally representative survey in New Zealand. J Nutr. 2016;146:1670–6. https://doi.org/10.3945/jn.116.235127

    CAS  Article  PubMed  Google Scholar 

  35. Hennigar SR, Lieberman HR, Fulgoni VL 3rd, McClung JP. Serum zinc concentrations in the US population are related to sex, age, and time of blood draw but not dietary or supplemental zinc. J Nutr. 2018;148:1341–51. https://doi.org/10.1093/jn/nxy105

    Article  PubMed  Google Scholar 

  36. Lin CN, Wilson A, Church BB, Ehman S, Roberts WL, McMillin GA. Pediatric reference intervals for serum copper and zinc. Clin Chim Acta. 2012;413:612–5. https://doi.org/10.1016/j.cca.2011.12.005

    CAS  Article  PubMed  Google Scholar 

  37. Azemati B, Khoramdad M, Qorbani M, Rastad H, Shafiee G, Heshmat R, et al. Percentile values of serum zinc concentration and prevalence of its deficiency in Iranian children and adolescents: the CASPIAN-V study. J Pediatr Endocrinol Metab. 2020;33:525–31. https://doi.org/10.1515/jpem-2019-0520

    CAS  Article  PubMed  Google Scholar 

  38. Barman N, Salwa M, Ghosh D, Rahman MW, Uddin MN, Haque MA. Reference value for serum zinc level of adult population in Bangladesh. Ejifcc. 2020;31:117–24.

    CAS  PubMed  PubMed Central  Google Scholar 

  39. National Nutrition Monitoring Bureau. Technical Report No. 27, Diet & nutritional status of urban population in India and prevalence of obesity, hypertension, diabetes and hyperlipidemia in urban men and women. A brief NNMB report on urban nutrition. Hyderabad: National Institution of Nutrition; 2017.

  40. Hunt JR, Beiseigel JM, Johnson LK. Adaptation in human zinc absorption as influenced by dietary zinc and bioavailability. Am J Clin Nutr. 2008;87:1336–45. https://doi.org/10.1093/ajcn/87.5.1336

    CAS  Article  PubMed  Google Scholar 

  41. Griffin IJ, Hicks PD, Liang LK, Abrams SA. Metabolic adaptations to low zinc intakes in premenarcheal girls. Am J Clin Nutr. 2004;80:385–90. https://doi.org/10.1093/ajcn/80.2.385

    CAS  Article  PubMed  Google Scholar 

  42. Roohani N, Hurrell R, Kelishadi R, Schulin R. Zinc and its importance for human health: An integrative review. J Res Med Sci. 2013;18:144–57.

    PubMed  PubMed Central  Google Scholar 

  43. Institute of Medicine (IOM): Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. National Academies Press (US), Washington (DC), 2001.

  44. Wessells KR, Brown KH. Estimating the global prevalence of zinc deficiency: results based on zinc availability in national food supplies and the prevalence of stunting. PloS One. 2012;7:e50568. https://doi.org/10.1371/journal.pone.0050568

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  45. Pinna K, Woodhouse LR, Sutherland B, Shames DM, King JC. Exchangeable zinc pool masses and turnover are maintained in healthy men with low zinc intakes. J Nutr. 2001;131:2288–94. https://doi.org/10.1093/jn/131.9.2288

    CAS  Article  PubMed  Google Scholar 

  46. Miller LV, Hambidge KM, Naake VL, Hong Z, Westcott JL, Fennessey PV. Size of the zinc pools that exchange rapidly with plasma zinc in humans: alternative techniques for measuring and relation to dietary zinc intake. J Nutr. 1994;124:268–76. https://doi.org/10.1093/jn/124.2.268

    CAS  Article  PubMed  Google Scholar 

  47. Pomeroy E, Mushrif-Tripathy V, Cole TJ, Wells JCK, Stock JT. Ancient origins of low lean mass among South Asians and implications for modern type 2 diabetes susceptibility. Sci Rep. 2019;9:10515. https://doi.org/10.1038/s41598-019-46960-9

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  48. D’Angelo S, Yajnik CS, Kumaran K, Joglekar C, Lubree H, Crozier SR, et al. Body size and body composition: a comparison of children in India and the UK through infancy and early childhood. J Epidemiol Community Health. 2015;69:1147–53. https://doi.org/10.1136/jech-2014-204998

    Article  PubMed  Google Scholar 

  49. Miller LV, Hambidge KM, King JC, Westcott JE, Krebs NF. Predictors of the size of the exchangeable zinc pool differ between children and adults. J Nutr. 2017;147:187–94. https://doi.org/10.3945/jn.116.239541

    CAS  Article  PubMed  Google Scholar 

  50. Kim J, Paik HY, Joung H, Woodhouse LR, King JC. Plasma zinc but not the exchangeable zinc pool size differs between young and older Korean women. Biol Trace Elem Res.2011;142:130–6. https://doi.org/10.1007/s12011-010-8758-2

    CAS  Article  PubMed  Google Scholar 

Download references

Acknowledgements

HSS and AVK are recipients of the Wellcome Trust/Department of Biotechnology India Alliance Clinical/Public Health Research Centre Grant # IA/CRC/19/1/610006. RP, BK, GBR, and HR are supported by the Indian Council of Medical Research, Govt of India.

Author information

Authors and Affiliations

Authors

Contributions

HSS, AVK, SG, and RP conceptualized the manuscript, and RP wrote the first draft and revised under the supervision of HSS and AVK. SG led the statistical analysis with guidance from HSS and AVK. HSS, AVK, RP, SG, BK, GBR and HR contributed to the data interpretation and had access to all the data. All authors reviewed and approved the submitted manuscript. HSS and AVK had final responsibility for the decision to submit for publication.

Corresponding authors

Correspondence to Anura V. Kurpad or Harshpal S. Sachdev.

Ethics declarations

Competing interests

HSS designed the draft protocol of the CNNS with consultancy support from UNICEF, India. HSS, UK, and AVK were members of the Technical Advisory Committee of the CNNS, constituted by the Ministry of Health and Family Welfare of the Government of India, to oversee its conduct and analysis. HSS is a member of the WHO Nutrition Guidance Expert Advisory Subgroup on Diet and Health and Guideline Development Group on the use and interpretation of haemoglobin concentrations for assessing anaemia status in individuals and populations, member of the World Health Organization Nutrition Guidance Expert Advisory Subgroup on Diet and Health and Expert group on nutrient requirements: setting calcium, vitamin D and Zn nutrient intake values for children aged 0-4 years. He is also member of Expert Groups of the Ministry of Health and Family Welfare on Nutrition and Child Health. SD was involved in the CNNS study implementation. There were no other conflicts to declare. The views expressed here by the authors are in their individual capacity and should not be construed as views or recommendations of the institutions the authors belong to.

Ethical approval

The CNNS was conducted after obtaining due International Ethical approval from the Population Council’s International Review Board, New York, USA and National Ethical approval from Post Graduate Institute of Medical Education and Research, Chandigarh, India (Ref. 14).

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Pullakhandam, R., Ghosh, S., Kulkarni, B. et al. Reference cut-offs to define low serum zinc concentrations in healthy 1–19 year old Indian children and adolescents. Eur J Clin Nutr 76, 1150–1157 (2022). https://doi.org/10.1038/s41430-022-01088-4

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41430-022-01088-4

Further reading

Search

Quick links