Keywords

FormalPara Case Vignette

A 4-year-old girl with cerebral palsy is admitted to the Pediatric Intensive Care Unit (PICU) with severe pneumonia and potentially bacteremia. She is normally tube-fed at home and on multiple medications including anti-seizure and antispasmodic medications. She requires fluid resuscitation, norepinephrine, and invasive mechanical ventilation at admission. By day 2 of PICU admission, her serum creatinine rises from 40 to 120 μmol/L. She has not tolerated an attempt of gastric enteral feeding and is currently receiving dextrose and saline intravenous fluids. By day 5 of PICU admission, she fulfills criteria for renal replacement therapy requirement and continuous renal replacement therapy (CRRT) is initiated. Enteral feeds which were being administered at half-strength are stopped at CRRT initiation. The dietician became involved in the patient’s care on PICU day 3. She is now asking what type of formula to use while the patient is on CRRT and questions your suggestion to provide 3 g/kg/day of protein in this child with severe AKI.

1 Introduction

Malnutrition and protein-energy wasting (or muscle wasting and loss of fat) are common in children in the intensive care unit (ICU) [1]. This is due to many factors, including, but not limited to, hypercatabolism and substrate requirement in the stress-response state, baseline nutritional status prior to acute illness, and ongoing growth needs (including brain growth in infants) [2, 3]. Young children are at highest risk for protein-energy wasting [4, 5]. Children with acute kidney injury (AKI) tend to be among the most critically ill ICU patients and AKI seems to enhance the risk of protein-energy wasting. The presence of AKI and AKI-associated acidosis/uremia alters lipid, carbohydrate, and protein metabolism, leading to promotion of catabolism [2]. With renal replacement therapy (RRT), catabolism is enabled by the addition of nutritional losses caused by RRT, which may worsen preexisting and acute nutritional deficiencies. Finally, fluid overload often occurs in AKI; this poses challenges to providing nutrition. These nutrition problems overlay the fact that underfeeding is common in children admitted to the ICU and is even more prevalent in children with AKI, especially those receiving RRT [6,7,8]. In children with AKI, the high baseline risk of protein-energy wasting must be appreciated. The goal should be to provide and enhance nutrition, rather than restrict.

There are no nutrition trials in pediatric AKI. However, observational data support that malnutrition is more common in AKI and contributes to poor outcome [6, 9, 10]. An ongoing multinational pediatric ICU study showed that ICUs with dedicated programs aimed at enteral nutrition advancement had lower infection rates and that adequate protein intake was inversely related to 60-day mortality [11, 12]. Moreover, attesting to the increasingly perceived importance of adequate evaluation and provision of nutrition in critically ill children, consensus guidelines were published in 2017 by the collaboration of two major collaborative groups (American Society of Parenteral and Enteral Nutrition and the Society of Critical Care Medicine) [3]. Table 12.1 provides the non-exhaustive list of selected recommendations from this guideline. Some of the most important conclusions from this guideline was the need for more nutrition research, a more widely systematic approach to nutrition and the need for patient-individualized approach. In recent years, the importance of adequate nutrition in children with AKI is also increasingly appreciated, and is becoming the basic principle of AKI management and often considered an indication for RRT initiation. Children with AKI are indeed the types of patients that require an individualized and thoughtful approach to nutrition. This chapter is not a detailed review on nutrition in critical illness, however references provided were handpicked to provide a comprehensive reading list on this topic. This chapter provides an overview of issues regarding overall assessment of nutritional status, approach and considerations specific to nutrition in children with AKI and those treated with CRRT.

Table 12.1 Selected recommendations on nutrition in critically ill children (from the recent Society of Critical Care Medicine (SCCM) and American Society for Parenteral and Enteral Nutrition Guideline) [3]
Full size table

2 Assessment and Monitoring of Nutritional Status

Evaluation of the nutritional status in critically ill children is an evolving field. An international consensus group has been working on defining/evaluating effects of malnutrition in ICU children [13]. It is ideal to assess the nutritional status at admission, to elicit evidence of chronic malnutrition; nutritional issues will worsen with critical illness and may be anticipated. History may reveal past growth problems or nutrition deficits. For example, a patient who has been ill with chronic osteomyelitis with 3 months of unexplained fever should immediately trigger the possibility of chronic nutritional issues. Patients with chronic kidney disease, as another example, must be considered as being at high risk for nutrition issues, and ICU admission should include evaluation for iron deficiency and nutrition status. Physical examination (hair, eyes, skin, mouth, extremities) may reveal specific signs of malnutrition or vitamin/mineral deficiencies [2]. Height and weight (with minimal clothing) should be measured in all patients and head circumference in children <2 years old and especially in young infants. This may be challenging because of difficulties in moving patients (if bed scales are not available). Moreover, the presence of fluid overload at admission must be considered. If the patient is edematous, one must be very weary of using admission weight to identify baseline nutrition status. When admission weight is uncertain, ideal body weight may be used for nutritional calculations. An effort should be made to calculate body mass index at admission, which may reveal nutritional problems. These measures should be expressed as corrected age percentiles and z-scores. The 2006 World Health Organization growth-for-age charts have been recommended in children <2 years old and the Center for Disease Control growth charts have been recommended for older children [13]. Height/weight/body mass index z-score < −2 standard deviations from the mean are reasonable ways to screen for poor baseline nutrition status [13]. Others have suggested height-for-age or weight-for-height < 10th percentile as a proxy for poor baseline nutritional status [2]. Several nutrition scores have been published and may be considered [2, 13]. Measuring triceps skinfold thickness and mid upper arm circumference may be useful (expressed as z-scores) [2, 13]. Weight (or triceps skinfold thickness and mid upper arm circumference) may be followed during admission. However, remember that edema/fluid overload may lead to underestimating weight loss. Some ICUs may use more advanced methods to evaluate body composition, most commonly, bioimpedance analysis. Bioimpedance analysis allows for measurement of lean body mass, which may decrease substantially with illness, even though body mass index may not change [14, 15]. However, it is important to accept that bioimpedance analysis measures will be strongly affected by the presence of edema/fluid overload, and thus should be interpreted with caution.

Biochemical parameters (e.g., albumin, prealbumin, visceral proteins) traditionally used to assess nutrition status will often be altered by fluid shifts and should be cautiously interpreted. If there is suspicion for baseline chronic malnutrition, measuring trace elements or vitamin levels at admission may be helpful to guide initial supplementation plans.

As shown, properly evaluating baseline nutrition status and monitoring nutrition status can be quite tedious and may be time consuming. However, waiting 2 or 3 weeks when the patient begins to improve clinically and noting the presence of hand muscle wasting is not acceptable. Despite the challenges of monitoring nutrition status and weight in critically ill patients, routinely collecting the data is more likely to lead to more tailored nutrition than collecting no data at all. The inclusion of a dedicated critical care dietician as an integral part of the multidisciplinary care team is invaluable.

3 Nutrition Timing, Approach, and Modality in AKI

The goals of nutrition in AKI are similar to other ICU patients, including preventing protein-energy wasting, avoiding further metabolic derangements and complications, allowing growth, and reducing mortality/morbidity. AKI generally occurs early in pediatric ICU admission [16], thus nutritional problems are triggered early in critical illness. Nutrition planning in AKI should begin immediately and not be delayed while awaiting renal function improvement or global clinical improvement. As described above, it is known that children in the ICU are often underfed [17]; this problem is even more prevalent in patients with AKI, for many reasons (e.g., reduced urine output and fluid restriction; enteral feeding intolerance) [6, 7].

Choice of enteral versus parenteral nutrition should be approached similarly to other ICU patients. Enteral nutrition is preferred when possible (with acknowledged advantages like lower cost, manageability, preservation of gastrointestinal function, and integrity) [18]. AKI itself lends no contraindications to enteral feeding [11, 19, 20]. Several studies demonstrate no survival advantage of early parenteral nutrition [21]. Choice of enteral feeding route (gastric; transpyloric) should also be similar to other ICU children, though transpyloric feeding may increase risk for gastrointestinal complications [2, 22].

With hyperphosphatemia or hyperkalemia, low-potassium/phosphate formulas should be used. With continuous RRT (CRRT) or peritoneal dialysis (peritoneal dialysis), regular enteral formulas may be used once electrolyte homeostasis is achieved. However, it is important to remember that formulas and parenteral nutrition electrolyte contents may need to be changed during non-dialytic periods (e.g., stopping CRRT for a head imaging exam). In AKI, feeding should not be avoided as a means to avoid RRT [19].

4 Energy Provision

Without appropriate energy intake, nutrient utilization is suboptimal; overfeeding is also associated with complications. Although nutrient utilization in AKI is thought to be abnormal, how it impacts on caloric needs is unclear [20]. The gold standard method to estimate energy needs is measuring resting energy expenditure. This is most possible in ventilated children, using a metabolic cart and indirect calorimetry (measures oxygen consumption and carbon dioxide production to estimate resting energy expenditure) [23, 24]. There may be concern for resting energy expenditure measurement validity during RRT, due to bicarbonate fluxes at the hemofilter (may affect expired carbon dioxide measurements which are used for estimating resting energy expenditure), but how much RRT affects resting energy expenditure measurement is unclear. Measuring resting energy expenditure in the ICU is feasible but is not commonly done, requires expertise and equipment, and is associated with limitations [20]. Despite these limitations and challenges, current recommendations are to use indirect calorimetry to prescribe energy intake if at all possible [3]. There is extensive literature on different prediction equations for estimating resting energy expenditure in children [24]. Despite the fact that these equations may often not be accurate or precise, they are commonly used and accepted as a surrogate for measured resting energy expenditure. The Caldwell-Kennedy equation (Table 12.2, recommended by the Kidney Disease: Improving Global Outcomes guidelines) was shown to provide the least biased resting energy expenditure estimation in ICU children [19, 24]. Others have suggested using the Schofield equation [2].

Table 12.2 Proposed suggestions on nutrition in children with acute kidney injury
Full size table

While ideal energy requirements remain controversial, some authors recommend caloric intake 20–30% above requirements estimated using prediction equations. This likely provides adequate calories in most children with AKI without significant overfeeding risk. Energy provision should include lipids, protein, and carbohydrates, using insulin as needed to maintain tight glucose control (Table 12.2) [19, 24]. With peritoneal dialysis, the dietician should consider dialysis fluid glucose load when calculating energy intake [2]. If overfeeding is a concern for a child on CRRT, calories from citrate anticoagulation administered may also be included in carbohydrate energy intake calculation.

5 Protein Intake

As in adults, critically ill children with AKI have increased protein catabolism, abnormal protein production, and increased protein turnover (amino acids excessively released from skeletal muscle and extracted by other organs), resulting in a negative nitrogen balance [20, 23]. A recent systematic review of 18 studies reflecting over 2000 critically ill children found that a protein intake >1.1–1.5 g/kg/day was significantly associated with reduced mortality [25]. Measuring amino acid levels as a measure of adequate intake is not useful; serum concentrations do not reflect total body stores or utilization. Nitrogen balance measurement is one way proposed to evaluate protein status [14]. It requires measurement of all nitrogen intake (mainly from nutrition) and all nitrogen output (including urine, stool, skin, other fluid losses, and, of course, RRT losses). A negative nitrogen balance implies catabolism and a positive nitrogen balance implies anabolism; the reason it only implies anabolism is because this measure does not reflect amino acid utilization. Ensuring adequate total energy (caloric) intake will promote better amino acid utilization/anabolism [14]. It is challenging to decrease protein breakdown, but optimizing protein synthesis to preserve skeletal muscle mass and achieve positive nitrogen balance may be enhanced by increasing protein (amino acid) nutrition [4]. Some guidelines suggest protein intakes for critically ill children: 0–2 years, 2–3 g/kg/day; 2–13 years, 1.5–2 g/kg/day; and 13–18 years, 1.5 g/kg/day [26]. Whether these guidelines are valid in AKI is unknown. Most recent guidelines suggest protein intake in critically ill children of 1.5 g/kg/day (Table 12.1) [3]. Given that negative nitrogen balance often occurs in ICU children, despite protein intake similar to those guidelines [27], these protein requirements may not be sufficient in children with AKI.

The challenge of providing adequate protein intake increases with RRT due to losses of amino acids through the hemofilter. Several studies have evaluated nitrogen balance, amino acid concentrations, and clearance in children treated with CRRT. Overall, they show that nitrogen balance is frequently negative using standard protein prescriptions and that amino acid clearance is substantial (Table 12.3). During CRRT, approximately 10–20% of delivered amino acids are estimated to be lost [28,29,30,31]; however, this estimate was based on studies using CRRT clearance of 1.5–2 L/1.73 m2/h. With higher clearance, amino acid losses will be greater. For these reasons, in children with AKI, providing ~2 g/kg/day of protein with close monitoring of acid-base status is advisable. In children treated with CRRT, protein intake should be increased by at least 30% (~3 g/kg/day) and increased more with higher clearance.

Table 12.3 Studies on losses from renal replacement therapy of selected nutrition analytes
Full size table

Peritoneal dialysis is commonly used to treat severe AKI in children, particularly infants. Peritoneal dialysis is known to cause transperitoneal protein losses in children on chronic dialysis [2, 32]. These losses are proportionally higher in younger infants [32]. Recommendations for protein intake in children treated with chronic peritoneal dialysis (ranging from 1.3 in older children to 1.8 g/kg/day in infants [32]) are based on the needs of non-critically ill children and likely inadequate in the critical illness setting. Thus, similar to children treated with CRRT, protein intake should approach at least 3 g/kg/day in children treated with acute peritoneal dialysis.

6 Electrolytes, Trace Elements, and Vitamins

AKI may lead to hyperkalemia, hyperphosphatemia, hypocalcemia, hypermagnesemia, and metabolic acidosis. Clearly, these laboratory parameters must be monitored and nutrition adjusted accordingly. In children with nephrotoxic AKI or with some congenital renal malformations, potassium, sodium, or phosphate losses may predominate, as a result of predominant tubular dysfunction. Children treated with CRRT very often become hypokalemic, hypophosphatemic, and hypomagnesemic mainly due to enhanced clearance. These should be expected and treated promptly; if phosphate has dropped below normal levels, increase supplementation immediately, do not wait. With frequent monitoring (at least once daily in AKI and twice daily in CRRT or PD) of these electrolytes, there is little reason that patients should have life-threatening electrolyte complications.

Little data exist on trace element metabolism in AKI. However, an excellent resource on evaluation and measurement of trace elements and vitamins in critically ill children is available [33]. In adults with AKI, plasma concentrations of trace elements (e.g., selenium, chromium, manganese, copper, zinc) and water-soluble vitamins (e.g. thiamine, folate) may be reduced [31]. However, the reasons for this are unclear, and may include altered protein-binding with critical illness, fluid loss, poor nutritional intake, and baseline nutritional deficiencies [34]. With prolonged AKI or prolonged CRRT, it may be useful to measure plasma concentrations, to help guide supplementation. Trace elements (especially chromium, selenium) and folate clearance on standard CVVHD prescriptions may be substantial (Table 12.3) [31]. Due to protein binding, trace element clearance may be higher with CVVH and losses may be significant with prolonged CRRT. It is likely that there is clearance of other water-soluble vitamins like thiamine [34]. Thiamine clearance should be considered in certain inborn errors of metabolism. While there is no evidence to support this, in cases where thiamine is especially important for a given inborn error of metabolism or is part of the treatment, we have increased thiamine intake by ~30–50% in children treated with CRRT, with monitoring of serum levels. There is no data available to guide supplementation of water-soluble vitamins and trace metals in acute peritoneal dialysis or HD. However, there are likely losses of some water-soluble vitamins with peritoneal dialysis and trace elements in both modalities. A study recently showed that children treated with CRRT became carnitine deficient, presumably due to clearance (carnitine is a water soluble, small molecule), suggesting the need for research to quantify clearance and evaluate supplementation strategies [35]. Given that the effects of increasing trace elements and/or water-soluble vitamin intake are unclear, it is recommended to provide the recommended daily allowances with the exception of vitamin C (should be lower). However, in children treated with RRT (especially CRRT) for >1 week, monitoring plasma concentrations of these elements may help guide supplementation. Finally, especially in patients with underlying chronic kidney disease, patients with prolonged AKI should be considered for vitamin D and/or activated vitamin D supplementation.

Conclusion

Suggested recommendations for feeding children with AKI are shown in Table 12.2. Feeding patients with acute kidney injury (AKI), especially when treated with continuous renal replacement therapy (CRRT), remains a delicate yet challenging task for intensive care unit (ICU) clinicians. Indeed, the AKI process itself is accompanied by inherent metabolic and physiological disturbances necessitating careful implementation of ICU feeding protocols. Nutrition provision for these patients is a crucial part of their care and institutions with standardized nutrition programs and care guidelines are more likely to achieve nutrition provision goals [11]. Healthcare providers should also be prepared to recognize signs and symptoms of nutritional toxicities and deficiencies and monitor mineral, protein, and fluid intake while patients are beginning or weaning from CRRT, as therapy becomes intermittent, and in other therapy changes.