Micronutrient status during paediatric critical illness: A scoping review

Background: No evidence based recommendations for micronutrient requirements during paediatric critical illness are available, other than those arising from recommended nutrient intakes (RNI) for healthy children and expert opinion. Objectives: The objective of this review is to examine the available evidence from micronutrient status in critically ill children considering studies which describe 1) micronutrient levels, 2) associations between micronutrient levels and clinical outcome, and 3) impact on clinical outcome with micronutrient supplementation during PICU admission. Design: Scoping review. Eligibility criteria: Any study which used a qualitative and quantitative design considering causes and consequences of micronutrient levels or micronutrient supplementation during paediatric critical illness. Sources of evidence: NICE Healthcare Databases Advanced Search website (https://hdas.nice.org.uk/) was used as a tool for multiple searches, with a content analysis and charting of data extracted. Results: 711 records were identi ﬁ ed, 35 were included in the review. Studies evaluated serum micro- nutrient status was determined on admission day in majority of patients. A content analysis identi ﬁ ed (n ¼ 49) initial codes, (n ¼ 14) sub-categories and (n ¼ 5) overarching themes during critical illness, which were identi ﬁ ed as: i) low levels of micronutrients, ii) causes of aberrant micronutrient levels, iii) associations between micronutrients levels and outcome, iv) supplementation of micronutrients. Conclusion: During critical illness, micronutrients should be provided in suf ﬁ cient amounts to meet reference nutrient intakes for age. Although, there is insuf ﬁ cient data to recommend routine supple- mentations of micronutrients at higher doses during critical illness, the ‘ absence of evidence should not imply evidence of absence ’ , and well designed prospective studies are urgently needed to elucidate paediatric micronutrient requirements during critical illness. The absence of reliable biomarkers make it challenging to determine whether low serum levels are re ﬂ ective of a true de ﬁ ciency or as a result redistribution, particularly during the acute phase of critical illness. As more children continue to survive a PICU admission, particularly those with complex diseases micronutrient supplementation research should also be inclusive of the recovery phase following critical illness. To characterise the effect of thiamine on physiologic and clinical outcomes hyperlactatemia.


Introduction
Micronutrients are defined by the World Health Organisation (WHO) as ""magic wands" that enable the body to produce enzymes, hormones and other substances essential for proper growth and development. As tiny as the amounts are, however, the consequences of their absence are severe" [1]. Micronutrient deficiencies are common amongst children between 6 months and 5 years of age, with 50% being deficient in one or more micronutrients [2].
Children admitted to paediatric intensive care unit (PICU) may have an acute illness or an exacerbation of a complex chronic condition. The median age of children to a paediatric intensive care unit (PICU) is 1.9 years, with the most common indications for admission being respiratory disease, congenital heart disease, and neurologic disorders [3].
Studies investigating the impact of nutritional support on clinical outcome of paediatric critical illness are scarce and have focused on macronutrient (energy and protein) requirements [4e7]. Micronutrient deficiencies have been described in critically ill adults, with low serum levels of thiamine, folate, vitamin B12 and zinc [8]. There is a paucity of clinical data relating to micronutrient requirements in critically ill adults with expert opinions recommending providing micronutrients to dietary reference nutrient intake levels in addition to pharmacological doses of thiamine to prevent refeeding syndrome. Other expert recommendations for adults include higher doses of vitamin C, D, B12, folate, zinc and carnitine to replete low serum levels [8e10]. However, micronutrient deficiencies may precede admission or occur as a result of inadequate intake, acute stress response, increased requirements or excessive losses, all of which may impact on morbidity and compromise clinical outcomes [8,11,12]. Micronutrient pathophysiology and requirements during paediatric critical illness are not well defined, other than those based on reference nutrient intakes (RNI) for healthy children [13] and expert opinion [14,15].
The goal of this scoping review was to systematically assess and describe the published literature for micronutrient requirements in critically ill children, while identifying key themes to guide the development of a conceptual framework for future research with regards to micronutrient status [16]. The objective of the review was to examine the available evidence from micronutrient studies in critically ill children considering those studies which describe 1) micronutrient levels during critical illness, 2) associations between micronutrient levels and clinical outcome, 3) potential causes for low micronutrient levels 4) impact on clinical outcome with micronutrient supplementation during PICU admission.

Preparing to scope the literature and protocol development
A scoping review was conducted in order to identify the key concepts that underpin this area of research [16]. The scoping study design was chosen because it offered a framework to identify and synthesize a broad range of evidence. The scoping review methodology provided an opportunity to focus on this complex area of paediatric critical illness and develop a conceptual framework to help identify gaps in the literature and future research priorities [17]. The Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) was used to develop and report the evidence reviewed for this study [18].

Identifying the research question
Is it possible to use the available literature to describe micronutrient levels during critical illness and provide recommendations for micronutrient requirements for critically ill children? Is there an association with micronutrient status and clinical outcomes in critically ill children? What are the current gaps in our knowledge and how could these be addressed?
Using the PRISMA-ScR checklist [18] and other published work [16,17,19] an a priori scoping review protocol was developed which included 1) the research question 2) eligibility criteria of the studies be to included, 3) information sources to be searched, 4) description of a full electronic search strategy, 5) data charting process with data items included, 6) critical appraisal and synthesis of the data in order to answer the questions posed. For the purpose of this review, critically ill children were defined as children between the ages of >37 weeks gestational age and 18 years of age in PICU.

Data sources e stage 1
After finalising the objectives and research questions a literature search was completed to identify relevant studies. NICE Healthcare Databases Advanced Search website (https://hdas. nice.org.uk/) was used as a tool for multiple searches within multiple databases including the PsycInfo, Cumulative Index to Nursing and Allied Health Literature (CINAHL) and Medline. PubMed, the Cochrane Library and NHS Evidence were also searched, with searches adapted for each database. Forward and backward citation searching was completed on studies exploring What this study adds: The results of this scoping review suggest there are numerous gaps in knowledge particularly relating to; the interpretation of plasma/serum levels of individual micronutrients during critical illness the causality of associations between micronutrient levels and clinical outcomes the impact on clinical outcome with micronutrient supplementation during PICU admission There is insufficient data to recommend routine supplementations of micronutrients in doses above recommended nutrient intakes levels during critical illness, however, well designed prospective studies are urgently needed to elucidate micronutrient requirements in children during critical illness.

What we know:
There are no evidence based micronutrient requirements during paediatric critical illness. Serum levels of micronutrients (vitamins and trace elements) may be affected by critical illness and inflammation. There is no evidence that supplementation of micronutrients improves clinical outcomes. micronutrients in paediatric critical illness, with searches completed until October 2019. The rationale for this time period was to not set a time limit to ensure that as much evidence as possible was captured.

Search strategy e stage 2
A search strategy was devised with the assistance of an information specialist for PubMed using key words from paediatric critical care articles and modified for additional electronic data bases (Supplementary File: Table 1).

Study selection e stage 3
After screening titles and abstracts and deletion of duplicates, full text articles were reviewed for eligibility. Where the same cohort of children were included in multiple articles they were only counted once [20e22]. Inclusion criteria were; any study which used a qualitative and quantitative design considering micronutrient supplementation or characterisation of serum/plasma micronutrient levels during various phases of critical illness in children published in the English language; based on human subjects and published up to October 2019. References from the bibliographies of studies included were hand searched for additional studies which may fulfil the inclusion criteria. Exclusion criteria which were considered not to be in scope of this review included; not English language or that were completed in other healthcare settings (e.g. ward or community environment), children with inherited metabolic disorders and levels described in fluids other than serum/plasma. Micronutrients were defined as chemical element or substance required in trace amounts for the normal growth and development [1,23,24]. Micronutrients excluded were those routinely measured as part of daily clinical biochemical monitoring such as sodium, potassium, calcium, chloride, magnesium, phosphorus [25] vitamin D (as this was classified a hormone) [26], and micronutrient levels not measured using plasma or serum samples [11]. Children with major burns were also excluded from this review as they are not commonly managed within a PICU but a dedicated burns unit, although there may be exceptions to this. In addition, micronutrient status and requirements of this cohort of patients is expertly reviewed within ESPEN endorsed recommendations for nutritional therapy in major burns [27].

Data extraction e stage 4
Data extraction was completed using a two stage process, with a data extraction template (Microsoft 2010, Redmond, WA) used to capture the study design, results and conclusions was created, followed by a content analysis.

Collating, summarizing and reporting the results -stage 5
Data synthesis was completed using an established content analysis approach [28]. Content analysis was chosen as it is a suitable technique for reporting common issues reported in data [29]. Using this approach, the descriptive aspects about the population of interest, methodology, outcomes and any key findings were coded. A content analysis was completed by selecting, coding and creating initial codes, sub-categories and overarching themes/ categories. These codes were then grouped into a number of categories and then grouped again into sub-categories. The key overarching categories from this process were developed into a conceptual framework using overarching themes.

Results
711 records were identified, of which 361 were duplicates. Following removal of duplicate records, abstracts and titles of 334 records were screened for inclusion. The full texts of 91 articles were reviewed for eligibility, of which 35 were included in the review (Fig. 1).

Content analysis: conceptual framework and overarching themes
A content analysis identified 49 initial codes, 14 sub-categories and 5 overarching themes (Table 1), which were identified as: i) Low levels of micronutrients during critical illness ii) Causes of low micronutrient levels during critical illness iii) Associations between micronutrients levels and outcome during critical illness iv) Supplementation of micronutrients during critical illness v) Micronutrient levels unchanged or high during critical illness These were used to develop a conceptual framework for factors impacting on micronutrient status during critical illness (Fig. 2).

Study characteristics
The studies included in the scoping review examined micronutrient status in 1855 infants and children in 35 studies. All but one study reported serum or plasma levels of vitamins and trace elements. Although ascorbate levels (a vitamin C salt) levels were reportedly low in the cerebrospinal fluid of children with traumatic brain injury (TBI) [30], this was excluded as it was not serum/ plasma micronutrient levels. Studies included in the scoping review investigated serum/plasma levels, reporting micronutrient status was determined on admission day in majority of patients. There were a number of studies describing low serum/plasma micronutrient levels during critical illness including thiamine, riboflavin, folate, vitamin B6, vitamin B12, vitamin A, b-carotene, zinc, selenium, iron and chromium [31e41], as well as studies describing serum/plasma levels unchanged or high during critical illness vitamin E, vitamin B6, copper and manganese [38,42e44]. A number of these studies reported factors associated with serum/ plasma levels of micronutrients levels including the use of continuous renal replacement therapy, cardiac surgery and systemic inflammatory response [21,22,30,31,37,38,42,44e55]. Associations between micronutrients levels and morbidity e.g. multiorgan failure, lactic acidosis, sepsis and mortality e.g. thiamine deficiency were also reported [40,41,46,53,56]. There were only two studies considering micronutrient supplementation in critically ill children e.g. zinc and whey protein, zinc, glutamine, selenium and metaclopramide [21,22,46] (Tables 2 and 3).

Discussion
The results of this scoping review reveals there are numerous gaps in knowledge relating to 1) the interpretation of serum levels of individual micronutrients during critical illness, 2) the causality of associations between micronutrient levels and clinical outcomes, and 3) the impact on clinical outcome with micronutrient supplementation during PICU admission. In the early phase of critical illness, aberrant serum micronutrient levels may be due to 1) redistribution from central circulation to tissues and organs during the acute phase inflammatory response to critical illness, 2) micronutrient losses due to exudative or stomas losses, 3) reduced stores of enzyme co-factors due to increased requirements during illness and 4) low endogenous levels due to pre-existing diseases or dysregulated renal excretion [8,40,41,57e60] (Fig. 3).
Despite the availability of well defined serum/plasma reference ranges, cautious interpretation of low levels is recommended during critical illness, particularly when there is an infection or systemic inflammatory response [11,59]. To counter this effect other quantification methods have been developed including, measuring co-enzyme activity using high performance liquid chromatography [61], and assays to measure erythrocyte concentration [62e64]. Studies of critically ill adults have demonstrated erythrocyte concentration of riboflavin and vitamin B6 status may be a more reliable indicator of micronutrient status when compared to plasma [62e64], although in contrast erythrocyte concentration has not been shown to be a reliable measure of vitamin C and E status [64].
However, there is limited knowledge regarding the effect of micronutrient supplement on intracellular signalling pathways [70], and as such whether goals of restoring serum levels to within normal range would be of clinical benefit during the acute phase of critical illness. Studies examining factors associated with redistribution of metal ions in response to infection suggest that low levels of extra-and intracellular iron [71,72] and zinc [73] during infection may be a protective mechanisms against pathogens [71,72]. In vitro studies of activated macrophages infected with fungal pathogens demonstrate Zn sequestration into intracellular niches or binding to proteins (such as albumin) in order to reduce availability [73]. Several micronutrients are essential for mitochondrial function and acts as cofactors for energy metabolism or anti-oxidant pathways, and inadequate reserves may hinder mitochondrial bioenergetics. However, supplementation may not always be of benefit, and although there are no studies describing vitamin E or C  The aim of this study was to characterise the relationship selenium, glutathione status and multiple organ failure in critically ill children. The concentrations of serum selenium and reduced and total glutathione were determined at admission and at day 5. The results showed that selenium was almost 20% lower in patients with multi-organ failure as compared to patients with zero or single organ failure (p < 0.0001). Low concentration of serum selenium as well as a high-reduced fraction of glutathione (GSH/tGSH) was associated with multiple organ failure (p < 0.001 and p < 0.01) respectively. A correlation between low serum selenium concentration and highreduced fraction of glutathione (GSH/ tGSH) was also seen (r ¼ À0. 19 and p ¼ 0.03). The serum selenium concentrations in the pediatric reference group in a selenium poor area were age dependent with lower concentrations in infants as compared to older children (p < 0.001). Almost half of the patients had a PICU-stay >5 days and these patients showed an increase in selenium of 14% from admission to 5 days. Children undergoing treatment with continuous renal replacement therapy (CRRT) showed an increase in selenium over 5 days.
A low serum selenium concentration was associated with the development of multiple organ failure.   The aim of this study was to explore the association of blood Zna d Se levels and immunomodulators in critically ill children. Children who had two organs affected have levels of zinc (median is 56.0 mg/dL) lower than that in patients in whom one organ was affected (median is 82.0 mg/dL). Selenium levels (median is 133.0 ng/mL) were lower in patients in whom one-organ was affected (median is 143.0 ng/mL). Levels of zinc, selenium, and prolactin in patients with sepsis (medians were 77.0 mg/dL, 142.0 ng/mL, and 18.2 ng/mL) were lower than that in patients without sepsis (medians were 81.0 mg/dL, 160.0 ng/mL, and 30.2 ng/ mL). Zinc was significantly inversely correlated with organ failure injury (OFI) score (p ¼ 0.047), and PRL was significantly inversely correlated with OFI score (p ¼ 0.049). There was no correlation between selenium and OFI score. Zinc was significantly inversely correlated with PELOD score (P ¼ 0.026), and PRL was significantly inversely correlated with PELOD score (p ¼ 0.039). There was no correlation between selenium and PELOD score.   . There was no difference in median (interquartile range) change in lactate from T0 to T24 between thiamine-treated cases and controls (À9.0, À17.0 to À5.0 vs À7.2, À9.0 to À5.3 mmol/L, p ¼ 0.78), with both groups exhibiting a rapid decrease in lactate. There were also no differences in secondary outcomes between groups.
Treatment of pediatric septic shock with thiamine was followed by rapid improvement in physiologic and clinical outcomes after prolonged hyperlactatemia.
Continuous renal replacement therapy amino acid, trace metal and folate clearance in critically ill children. The aim of this study was to characterise the prevalence of vitamin A deficiency in critically ill children with sepsis and clinical outcomes. Vitamin A deficiency was found in 94 (58.8%) subjects in the study group and 6 (12.2%) subjects in the control group (P < 0.001). In septic patients, 28-day mortality and hospital mortality in patients with vitamin A deficiency were not significantly higher than that in patients without vitamin A deficiency (P > 0.05). However, vitamin A levels were inversely associated with higher PRISM scores in septic children with VAD (r ¼ -0.260, P ¼ 0.012). Vitamin A deficiency was associated with septic shock with an unadjusted odds ratio (OR) of 3.297 (95% confidence interval (CI), 1.169 to 9.300; P ¼ 0.024). In a logistic model, vitamin A deficiency (OR, 4.630; 95% CI, 1.027e20.866; P ¼ 0.046), procalcitonin (OR, 1.029; 95% CI, 1.009 e1.048; P ¼ 0.003), and the Pediatric Risk of Mortality scores (OR, 1.132; 95% CI, 1.009e1.228; P ¼ 0.003) were independently associated with septic shock.
The prevalence of vitamin A deficiency was high in children with sepsis.  [36]. Levels were unchanged [42] Riboflavin/Vitamin B2 (p) Riboflavin acts in synergy with a number of other B vitamin plays a role in energy metabolism and production, in addition to red blood cell formation [96].

Zinc (p)
Zinc is a trace element involved in numerous functions including antioxidant function, and during the acute inflammatory response with zinc redistribution of zinc in tissues involved 11e24 mmol/l 7.1 mmol/l (4.6, 7.8)

Selenium(s)
Selenium is a trace element is involved in anti-oxidant, immunological and endocrine pathways, in addition to helping to maintain membrane and assist in thyroid production [50].

Copper (s)
Copper is required for redox pathway, energy production, glucose and cholesterol metabolism [101].

No change ≈ High
Low serum copper was reported in critically ill children with oxidative stress [44] Levels unchanged in critical ill children [49] High in children requiring renal replacement therapy [38]

Iron (p)
Iron is required for the normal development of red blood cells and cognitive development. Iron deficiency anaemia affects children in particular [102].
All ages 22e184 mg/dl 4e33 mmol/l 3.9 mmol/l (3.8, 4.5) Low Serum iron levels were significantly lower in critically ill children [52] and neonates [43] on admission Chromium(s) Chromium has been suggested to be required for carbohydrate and lipid metabolism by enhancing the effectiveness of insulin [103].

mg/L
Levels not reported Low Low levels of chromium [38] were reported in patients on CVVH

Manganese (s)
Manganese is required as a co-enzyme for a number of enzymes, including macronutrient metabolism, bone formation and oxidative function [104].
9e24 nmol/l Levels not reported No change z High Levels were unchanged in children with oxidative stress [44] High levels in children requiring renal replacement therapy [38] Studies describing outcomes associated with micronutrient levels during critical illness Mortality risk Low thiamine levels were associated with increased mortality risk in malnourished patients [83] Studies describing between micronutrients levels and morbidity during critical illness Inadequate intake Low micronutrient intake from enteral feeds were; low weight for age, fluid restriction, disease severity, the use of alpha-adrenergic drugs and renal replacement therapy [55] Low thiamine levels are associated with malnutrition [42] Low plasma selenium levels are associated with malnutrition [37] A case study of child with autism and a severely restricted vegan diet admitted to the PICU was associated with thiamine and vitamin B6 deficiency [36]. Inflammatory response Low serum levels of thiamine [31], iron, zinc and selenium [11,37,56] and zinc [48,49] are associated with the magnitude of acute phase inflammatory response Oxidative stress With increasing oxidative stress plasma levels of six micronutrients decreased vitamin A, C, E, b, selenium, copper and zinc [44].

Medication
Low thiamine levels was not associated with diuretic use [47] CCRT Low levels of selenium, zinc and chromium [38] Dilution effects of dialysate in renal replacement therapy result in losses of selenium, thiamine, folate, pyridoxine and vitamin C and accumulation of manganese [80].

Cardiac surgery
In case study reports thiamine deficiency was associated with lactic acidosis in septic shock [32], extracorporeal membrane oxygenation [33], haematological malignancies [65] and following cardiac surgery [34]. Low selenium levels have been described in children requiring cardiopulmonary bypass [54] (continued on next page)

L.V. Marino et al. / Clinical Nutrition xxx (xxxx) xxx
administration on mitochondrial function in critical illness, effect of vitamin E and C supplementation on athletes has been studied, with unexpected results [74]. Athletes who took vitamin C [74] or C and E [75], to reduce the effect of oxidative stress following endurance exercise appeared to have an unintended consequences with reduced mitochondrial bioenergenesis arising from reduced perioxisome proliferator activated receptor gamma co-activator, nuclear respiratory factor and mitochondrial transcription factor A. This prevented exercise induced expression of cytochrome C and reduced the maximal rate of oxygen consumption hampering cellular adaptation to endurance training [74,75]. During critical illness, mobilisation of stores may occur but if there are inadequate stores to meet increased demands during critical illness or the recovery phase, conditional deficiency may occur. Particularly as micronutrient levels available in current enteral or parental nutrition support, are based on RNI for healthy children (Table 4). This has been characterised in a study considering micronutrient intake during paediatric critical illness from enteral feeds. Low micronutrient intake from this route of nutrition support was associated with low weight for age, fluid restriction, diseases severity, the use of alpha-adrenergic drugs and renal replacement therapy [55]. As such, low serum levels, especially during the recovery phase in the absence of systemic inflammation, may therefore be reflective of relative deficiency states particularly in children pre-admission [76e78]. This may also be true for those with increased losses arising from thermal injuries [27], ostomies [79] or from renal replacement therapy [38,80]. Equally low levels may arise as a result of a dilution effects of dialysate or crystalloids used in medical and surgical procedures such as cardio-pulmonary bypass [11,54]. Following a period of critical illness, micronutrients available in enteral feed or parenteral micronutrient mixes may not replete losses during rehabilitation, possibly resulting in growth failure and poor metabolic, physiological and immune resilience [81,82].
Some studies have described impaired outcomes associated with low levels of micronutrients. Association studies included low levels of thiamine and an increased risk of mortality in malnourished patients [83]. Other studies described inverse relationship between low levels of vitamin A [39], iron [52] zinc [48,50,52] and severity of injury scores e.g. the higher the severity of injury the lower the serum levels, and multi-organ failure [53]. However, the impact of aberrant levels on short or longer term clinical and nutritional outcomes are not known, as the majority of studies describe measures on admission to intensive care. As such there is a paucity of evidence to suggest any causal link between low serum levels of various micronutrients during the acute phase of critical illness and the impact on clinical outcomes [12].
Only two studies have been completed considering micronutrient supplements, one considering intravenous zinc [46], which was safe to give, and the other as part of the CRISIS trial were enteral zinc, glutamine, whey protein, selenium and metaclopramide were administered. In the CRISIS study there was no difference reported in the clinical outcomes compared to the placebo, however children recruited into the treatment arm had low severity of illness scores who may not have required supplementation [21,22]. Neither study demonstrated any clear benefits with respect to outcomes including reduction in length of hospital stay or survival. Part of the consideration for future trials should be around methodological design including adequately powered studies in addition to appropriate patient recruitment.
Many of paediatric studies completed have small numbers with short duration and therefore make interpretation and comparison of studies challenging [12]. At present there are no specific recommendations for micronutrient supplementation in the acute phase of critical illness [40] and pharmacological use of micronutrients remains controversial due to reports of toxicity [11,84]  Multi-organ failure Low plasma selenium levels are associated with multi-organ failure [53] Lymphopenia Low zinc levels were also found in patients with lymphopenia [51]. In case study reports thiamine deficiency was associated with lactic acidosis in septic shock [32], extracorporeal membrane oxygenation [33], haematological malignancies [65] and following cardiac surgery [34].
Severity of illness score Low serum iron levels were associated with severity of illness score critically ill children [52] but not in neonates [43]. Serum copper levels were not associated with severity of illness score [43,52].
Low serum vitamin A levels in children with septic shock are associated with severity of injury scores [39]. Low plasma selenium levels are associated with severity of risk scores [48,50,52].

Studies describing supplementation of micronutrients during critical illness
Zinc Two studies considered micronutrient supplementation in critically ill children. One considered the safety of intravenous zinc supplementation in children, which was reported as well tolerated but not associated clinical outcomes were reported [46].
Enteral whey, zinc, selenium, metaclopramide Critically ill children expected to require >72 h PICU stay were randomised to receive enteral whey, zinc, selenium and metaclopramide, there was no difference reported in the clinical outcomes compared to the placebo [21,22].   In the early phase of critical illness, aberrant serum micronutrient levels may be due to 1) redistribution from central circulation to tissues and organs during the acute phase inflammatory response to critical illness, 2) micronutrient losses due to exudative or stomas losses, 3) reduced stores of enzyme co-factors due to increased requirements during illness and 4) low endogenous levels due to pre-existing diseases. Adapted with permission from Casaer M et al. [8]. and potentially unintended consequences as described in adult athletes [74,75]. However, this may not always be the case particularly during the recovery phase of critical illness [38,40,80]. With high numbers of critically ill children surviving a period of critical illness [85], nutritional rehabilitation, will require careful consideration. For nutrition rehabilitation to be successful sufficient energy, protein and also micronutrients are required [81,86]. Sub-clinical or clinical micronutrient deficiencies may impede nutritional recovery particularly when anabolism has been restored [77]. Future research should focus on what are the best methods to evaluate micronutrients requirements during critical illness, what micronutrient supplementation (if any) should be provided during critical illness and which cohorts of critically ill children may benefit the most from such supplementation. It may be possible to translate the principles within the World Health Organisation (WHO) recommendations for the management of severe malnutrition, into a conceptual framework for nutrition requirements for critically ill children (Fig. 4) [86].
However, the main challenge for the critical care community is to design research studies which address issues relating to clinical tests of biomarkers using serum or plasma samples which have sufficient sensitivity and specificity, optimal supplementation doses/duration, clinical and nutrition outcome measures [11], particularly as there may be some cohorts of critically ill child for who would benefit from additional micronutrient supplementation. In this scoping review the following factors were identified, which would need to be included within a conceptual framework for determining micronutrient requirements during critical illness (Fig. 2); 1) micronutrition malnutrition prior to admission to PICU arising from acute or chronic diseases increased requirement or reduced intake [87,88], 2) duration/severity of critical illness and increased micronutrient turnover e.g. oxidative pathways [44]. 3) use of medications or treatment factors impeding absorption or increasing micronutrient losses and associated transport issues e.g. renal replacement therapy [89], 4) redistribution, sequestion or dilution of serum levels of micronutrients and biomarkers to accurately identify the reason for low/high serum values [80], 5) abnormal losses of micronutrients as exudates from wounds, drains and stomas impact on micronutrient status [27,90].
At present routine extra supplementation of micronutrients during critical illness to correct low levels in the absence of clear functional pathophysiology as to why levels are low is not recommended [12,40]. However, as survivorship of paediatric critical illness increases ensuring micronutrition malnutrition is addressed to ensure nutritional rehabilitation is achieved post-discharge, is important.

Limitations
This is a scoping review to present the current range of evidence specific to micronutrient status in serum/plasma of critically ill children in young survivors of intensive care. The results should not be generalised beyond this paper other than about the quality of the available evidence. Children with major burns were also excluded from this review as they are not commonly managed within a PICU but a dedicated burns unit, although we acknowledge, they represent a cohort of children with significant nutrition risk [27]. A significant issue within this review was the lack of evidence regarding micronutrient status in critically ill children during the acute phase and into rehabilitation, making it difficult to provide any recommendations. Given this, it was not possible to synthesis results or reliably estimate prevalence and impact of micronutrient status during the acute phase and whether micronutrient supplementation would have any beneficial effect.

Conclusion
During critical illness, micronutrient should be provided in sufficient amounts to meet reference nutrient intakes for age. Although, there is insufficient data to recommend routine supplementations of micronutrients at higher doses during critical illness, the 'absence of evidence should not imply evidence of absence', and well designed prospective studies are urgently needed to elucidate paediatric micronutrient requirements during critical illness. The absence of reliable biomarkers make it challenging to determine whether low serum levels are reflective of a true deficiency or as a result redistribution, particularly during the acute phase of critical illness. As more children continue to survive a PICU admission, particularly those with complex diseases micronutrient supplementation research should also be inclusive of the recovery phase following critical illness.