Nutrition has a critical role in childhood. Multiple factors can impair linear growth and neurocognitive development as described in Table 72.3 . The Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines and Pediatric Renal Nutrition Taskforce, a team of doctors and dietitians from Europe and North America, have produced clinical practice recommendations on the assessment of nutritional status and management of various aspects of the nutritional prescription, as well as the role of enteral tube feeding. Detailed guides for health care professionals, as well as children and their caregivers, are available at https://www.espn-online.org/nutrition-taskforce .

Malnutrition, which is the result of poor appetite, vomiting, and decreased intestinal absorption, is common in children with ESKF. Close monitoring is crucial, especially when the child is younger than 1 year (every 2–4 weeks), decreasing in frequency as the child grows. Height and weight are plotted on appropriate growth curves, and Z-scores are calculated. Head circumference is measured regularly in children younger than 3 years. Most younger children fail to meet the energy and protein requirements by oral feeding alone and need supplementation by nasogastric tube or gastrostomy.

Initial recommended caloric intake is the age-related dietary reference intake (DRI) but is adjusted according to weight gain. Protein intake recommendation is 100% to 120% of the age-related DRI, with somewhat higher intake in PD patients to compensate for peritoneal losses. In contrast to adults, a Cochrane review found that protein restriction has no impact on delaying progression of renal failure. ^, Dietary restrictions may include fluid, sodium, phosphorous, and potassium, and special infant formulas are adapted to the needs of patients with CKD. Human milk also has the adequate electrolyte composition for the infant with CKD, as it is relatively low in potassium and phosphorus. However, because infants’ diet is fluid based, it may ultimately cause volume overload. This requires in some cases concentrating the formula to achieve a higher caloric intake in a smaller volume (80–120 Kcal/100 mL vs. the usual 67 Kcal/100 mL). It may also influence the dialytic modality selected or the frequency of the dialysis sessions. Conversely, in several renal diseases (e.g., dysplastic kidneys, nephronophthisis, and cystinosis) patients may be polyuric due to a renal concentrating defect even when they reach ESKF. This makes water handling and nutrition possible but will require careful attention to electrolyte replacement, too.

Growth is severely impaired in children with ESKF. A longer period with ESKF and short stature at disease onset is associated with decreased final height. Short stature affects body image and may have an impact on self-esteem. In the context of chronic illness, this leads to overprotection, low expectations, and delayed independence and may subsequently affect future socioeconomic status and formation of intimate relationships.

When assessing patient height, it should be remembered that there are differences between ethnicities and growth curves, with normal values available only for some (higher income) countries.

Many factors contribute to impaired linear growth:

• •

  Low energy and protein intake: This is especially significant in early childhood.

• •

  Impaired growth hormone (GH)–IGF-1 (insulin-like growth factor 1) axis.

• •

  Additional factors include water and electrolyte imbalance, anemia, mineral bone disease, and metabolic acidosis. Correcting these abnormalities may improve the likelihood of achieving normal growth.

Beyond correction of metabolic abnormalities and providing the best possible nutritional support, treatment includes recombinant human GH (rhGH) therapy.

The use of GH in kidney failure is based on experimental and clinical evidence demonstrating that GH insensitivity can be overcome by supraphysiologic doses of exogenous GH. This treatment was proved to be effective in a study based on NAPRTCS data: Of 5122 dialysis patients, 33% received rhGH treatment. Catch-up growth was achieved in 11% of the treated dialysis patients (compared with 27% of patients in earlier stages of CKD). The mean increase was 0.5 standard deviation score (0.8 standard deviation score in earlier CKD stages). In addition, Cochrane database meta-analysis found increased height velocity (+3.9 cm/year after the first year), without advance in bone age. In the NAPRTCS 2011 report, children younger than 6 years of age had improvement in Z score of +0.63 compared with–0.1 in the untreated patients. Older children had milder improvement of +0.26 Z-scores. A consensus document from ESPN describes the indications and management of GH therapy in ESKF.

Hyperkalemia is commonly found in children with ESKF, due to decreased urinary excretion. Treatment includes dietary potassium restriction, particularly the restriction of processed foods that are likely to contain potassium additives, and the use of potassium binders such as Patiromer, sodium zirconium cyclosilicate, or sodium polystyrene sulfonate. Formula adapted for infants with CKD, low in potassium, phosphorous, and calcium, is commercially available.

Chronic metabolic acidosis is treated in children because of its impact on enzymatic functions, bone demineralization, and linear growth. The effect of acidosis on bone health and growth is multifactorial including increased osteoclastic and decreased osteoblastic activity; blunted action of the GH–IGF-1 axis on bones; decreased 1,25-dihydroxyvitamin D production, and hence less intestinal calcium absorption; and alteration of homeostatic relationships among vitamin D, parathyroid hormone (PTH), and ionized calcium. Metabolic acidosis is treated with bicarbonate salts with target bicarbonate serum concentrations of 20 to 22 mEq/L.

Hemoglobin starts to decline in children when eGFR is lower than approximately 50 mL/min/1.73 m ² . Diagnosis of iron deficiency may be challenging because many patients with ESKF have low transferrin levels due to malnutrition or chronic inflammation, causing the calculated transferrin saturation to be seemingly normal. Ferritin concentrations, by contrast, may be high as ferritin is an acute-phase reactant and increases in inflammatory states. The physiologic hemoglobin concentrations vary with age and gender, as do the recommended target levels: above the fifth percentile of the specific age. Treatment includes iron supplementation and erythropoiesis-stimulating agents (ESAs). Target iron levels are higher than in the healthy population: transferrin saturation greater than 20% and ferritin greater than 100 ng/mL (to optimize the effect of ESA). The relative ESA dose (units/kg/week) is higher in children and may even reach 1250 erythropoietin units/kg/week in infants. Other causes for failure of ESAs include osteitis fibrosa cystica, chronic inflammation, malnutrition, hemolysis, and rarely carnitine deficiency or aluminum toxicity. Studies in adults on dialysis found that excessive correction of anemia (even within the physiologic hemoglobin level) is associated with increased mortality; therefore it is recommended not to exceed hemoglobin levels of above 12 g/dL. The pathophysiology of this phenomenon is unclear and has not been validated in children.

Although the pathophysiology of CKD-mineral bone disease (MBD) is similar to that in adults, both the management and the consequences differ, as linear growth and rapid accrual of calcium into the growing skeleton is an important factor in children. ^,

Monitoring of various CKD–MBD-associated parameters includes the serial measurement of serum calcium, phosphorous, PTH, and alkaline phosphatase levels every 1 to 3 months ^, and vitamin D every 3 to 12 months. Bone mineral density may be measured by dual-energy x-ray absorptiometry (DXA); however, this modality is limited in differentiating between trabecular and cortical bone in CKD-MBD and thus does not predict fracture risk well in this population. Peripheral quantitative computed tomography is another modality to assess bone mineralization, which can distinguish between cortical and trabecular bone. Adjustment of DXA spine and whole-body measures for height Z-scores can provide clinically relevant data on bone health that can be monitored over time.

Because normal blood calcium and phosphorous concentrations and dietary requirements are age dependent, the target levels and nutritional recommendations also vary with age. Age-adjusted hypophosphatemia should be avoided as it can result in hypophosphatemic rickets, often seen in preterm infants with insufficient phosphorous intake, in children with phosphaturia due to proximal tubulopathy, or in those on intensified dialysis regimens. In CKD stage 5 the KDOQI 2017 guidelines recommend phosphorous restriction depending on blood levels. Phosphorous restriction can be challenging to achieve and maintain, particularly in teenagers who may indulge in more processed and ultraprocessed foods that have a high phosphate content and high bioavailability of the phosphate salts used in these. Of note, a person on a strict phosphate-controlled diet is likely to become calcium deficient, as a similar range of foods are rich in both calcium and phosphate. ^,

A controlled phosphate intake is expected to result in lower PTH and increased 1,25-dihydroxyvitamin D (calcitriol) levels, as well as better bone morphology and reduced risk for CKD-associated vasculopathy. ^, The first-line phosphate binders in children remain calcium-based phosphate binders. ^, The non–calcium-based phosphate binder, sevelamer carbonate or sevelamer hydrochloride, can be used as a second-line agent in those who are hypercalcemic and have comparable phosphate-binding efficacy. The use of lanthanum carbonate is not recommended in children because lanthanum deposits in bones, including growth plates, and the consequences on the developing bone are yet unknown.

Vitamin D monitoring and supplementation are important in patients at all CKD stages. The use of activated vitamin D analogs, such as alfacalcidol and calcitriol, in adults is limited to CKD stage 5D, but as recommended by KDIGO and the European Society for Pediatric Nephrology, in children their use may be considered to maintain serum calcium levels in the normal range in CKD stages 2 to 5D. ^, Treatment with activated vitamin D analogs should be started if hyperparathyroidism is persistent despite therapy with native vitamin D, aimed at normalizing serum calcium concentration and decreasing PTH (although the target PTH values at each CKD stage are not widely agreed ^, ). Finally, calcimimetics, which increase the sensitivity of the parathyroid calcium receptors, may have a role in the management of persistent hyperparathyroidism that is resistant to conventional treatments with calcium and vitamin D and should be considered before parathyroidectomy.

Deformities of the growing bone and impaired linear growth are complications that are typical in the pediatric age group ( Fig. 72.1 ), in addition to vascular calcifications (VCs) that are also well described in the adult patient population. Skeletal deformities in CKD, like the phenotype of vitamin D–deficient rickets, can be debilitating but are amenable to medical therapy with marked improvement and even complete resolution.

Severe renal osteodystrophy in a boy with end-stage renal disease with resolution obtained with medical treatment.

(A) Age 1 year: severe rickets, absorption of the metaphyseal edges of the radius and ulna. Delayed bone age (6 months). (B) Age 2 years: deformation in the radius and ulna bones, widening of the metaphyseal edges, healing rickets. (C) Age 4 years: normal bone structure. Bone age is 3 1/2 years. (D) Age 2 years: deformation with angulations of the femur, tibia, and fibula bones—healing phase. (E) Age 4 years: marked improvement, although some angulation can still be seen.

Cardiovascular disease (CVD) in children with CKD is the most important comorbidity affecting long-term survival, with a 500- to 1000-fold greater cardiovascular mortality compared with the general population. Different factors that contribute to developing CVD include hypertension, disorders in mineral metabolism, dyslipidemia, hyperuricemia, abnormal glucose metabolism, obesity, systemic inflammation, and oxidative stress. Uncontrolled hypertension affects approximately 25% of patients with CKD stages 3 and 4 and 47% with CKD stage 5. Hypertension is most commonly due to fluid overload, especially in patients with advanced stages of CKD and on dialysis, but other contributing factors are hyperactivation of the renin-angiotensin-aldosterone system (RAAS), catecholamine secretion and hypervolemia, and in some cases is caused by corticosteroids or calcineurin inhibitors used to treat the underlying kidney disease. ^, Atherosclerotic disease that involves calcification of the tunica media may be seen even in young children on dialysis, eventually contributing to left ventricular hypertrophy (LVH). Hypertension along with sustained high levels of FGF23, hypervolemia, and anemia are the drivers of LVH, either concentric or eccentric, which leads to systolic and diastolic cardiac disfunction. ^, Finally, carotid intima–media thickness, a surrogate marker of CVD, has been found to be elevated in 41.6% of children with CKD stages 3 to 5. Further details are described under transplant outcomes later.

Numerous reasons account for neurodevelopmental compromise in children with ESKF. Some pediatric kidney diseases are a part of a syndrome or systemic disease, which may have an effect on the central nervous system and psychomotor development (e.g., syndromes such as Bardet-Biedl, Joubert, trisomy 21, Galloway Mowat, Wilms tumor, aniridia, genitourinary anomalies, and intellectual disability).

Chromosomal microarrays to detect genomic imbalances were performed in children enrolled in the Chronic Kidney Disease in Children study. Genomic imbalances were detected in 7.4% of children with CKD. In the vast majority of these cases, the genomic lesion was unsuspected based on clinical assessment. Other pediatric kidney diseases can be associated with prematurity (Finnish-type nephrotic syndrome) or oligohydramnios and hypoplastic lungs and hence possible perinatal hypoxia (autosomal recessive polycystic kidney disease). Uremic toxins, HTN (or BP fluctuations during HD), and chronic anemia can cause further damage to the developing brain. Motor skills development can also be delayed because the sick infant is not free to move and play while connected to the dialysis or feeding machine. Various catheters including feeding tubes and gastrostomy may interfere with rolling over and crawling, and there is less sensory stimulation during hospitalization or dialysis. Muscle tone is diminished, and renal osteodystrophy may be painful and prevent the needed physical activity for normal development. Some medications frequently given to CKD patients, including aminoglycoside antibiotics (limited use, only in anuric patients) and furosemide, are potentially ototoxic.

Several studies over the past 30 years showed that patients with CKD have compromised cognitive functions. This can be seen even in the early stages of CKD. Specifically, studies found lower intelligence quotient (IQ) levels (∼10 points), verbal or written language difficulties, impaired memory, decreased executive functions, and inefficiency in key neurocognitive domains. Particularly affected are children who reached ESKF at a younger age, ^, children with longer dialysis vintage, ^{, ,} more severe HTN or hypertensive crisis events, ^, and children with comorbidities. Sensorineural hearing loss was found in 14% to 18%, and ischemic lesions of variable severity on magnetic resonance imaging (MRI) were seen in 18% to 33% of the patients. ^, Despite this, 61% to 79% of the children with ESKF attend regular school and have academic achievement comparable with their siblings. ^{, ,} Long-term follow-up in adults treated since childhood for ESKF found impaired schooling and lower IQ (9.2–10.4 points less than controls). Fortunately, the current ESKF treatment modality (dialysis or transplantation) was not associated with decreased IQ score. A study in children and adults younger than 30 years of age assessed neurocognitive functioning using a comprehensive battery of tests and brain structure by MRI. It showed that a combination of risk factors (including time since kidney transplantation, longer dialysis duration and late CKD onset) was significantly associated with lower intelligence and/or worse processing speed and working memory and that disrupted white matter integrity may contribute to these neurocognitive impairments. A further study from the CKiD cohort has shown that despite high rates of high school graduation, nearly 20% of patients with CKD are unemployed or receiving disability at long-term follow-up. Tailored interventions may benefit patients with CKD with lower kidney function and/or executive function deficits to optimize educational/employment outcomes in adulthood.

Quality of life of children on RRT was assessed in only a few studies. Quality of life scores are lower than the general population norms in all domains. Generally, patients perceive their quality of life as better than their parents do. This may be because patients are unaware of an alternative reality, developing a defense mechanism, or reflecting parental overprotection. The disadvantage is that parents’ lower rating may decrease the patients’ or their parents’ motivation and thus may adversely affect their outcomes. A recent study has shown that despite the presence of a chronic disease, children with CKD endorse a positive self-concept, which may predict academic success in this population.

Cardiovascular mortality is the most frequently reported cause of death: 14% to 41% of all pediatric deaths, depending on age and on population and ethnic affiliation. ^{, , , ,} To underscore the high risk for morbidity and mortality, the American Heart Association classified pediatric CKD patients in the same high-risk group as children with homozygous familial hypercholesterolemia, type 1 diabetes mellitus, and postorthotopic heart transplantation. In a single study where the authors directly reviewed all patients’ files, the report includes under the heading of CVS death, cerebrovascular accident (58% of cardiac deaths), congestive heart failure (15%), and other diagnoses that may be the consequence of HTN, heparinization, and volume overload. In addition, most of the data available are on children who reached ESKF 2 or 3 decades ago. There is marked decrease in not only the absolute death rate but also the percentage of deaths attributed to cardiac causes over the years: from 44.4% in 1972 to 1981 to 33.3% in 1992 to 1999. This was also shown from 1990 to 2010: In patients younger than 5 years at initiation of dialysis, CVS mortality decreased from 35.3% in 1990 to 1994 to 22.6% in 2005 to 2010 or, according to the USRDS, from 29% in 2005 to 2009 to 15% in 2010 to 2014 (same trend, lower percentages, in other age groups). A study from The Netherlands points to a shift from cardiovascular disease to infections as the main cause of death at long-term follow-up in patients with CKD—all these may be reflecting a more accurate diagnosis. However, more specific causes of death must be defined. This is not merely semantic, as accurate determination of causes of death can help focus efforts aimed at improving survival.

There is no recommended eGFR for initiation of dialysis in the absence of symptomatic uremia, volume overload, or growth failure refractory to medical therapy. The timing of initiation of dialysis is guided by a composite assessment of the child’s physical and psychosocial well-being, fluid status, growth, level of kidney function, and biochemical abnormalities. The median eGFR at dialysis start is currently 8.2 and 7.8 mL/min/1.73 m ² for children in registries from Europe and the United States, respectively. ^, Prospective data from large international registries in more than 21,000 children show that clinical outcomes do not improve in patients starting dialysis early. While young children with structural abnormalities of the kidney can have stable kidney function and adequate urine output for many years, uremic symptoms frequently begin when the GFR falls below 15 mL/min/1.73 m . Therefore use of available formulas (based on eGFR, proteinuria, and/or diagnosis) to prognosticate progression to ESKF and planning for initiation of dialysis when the GFR is <15 to 20 mL/min per 1.73 m ² is pragmatic.

The optimal KRT modality in children with ESKF is preemptive transplantation, given the lower morbidity, higher survival, and better quality of life after transplantation. If preemptive transplantation is not feasible, home dialysis therapies are most appropriate for children as they will allow for the child’s schooling and social life. No studies have suggested that either PD or HD has superior outcomes in children with ESKF. Rather, different dialysis modalities can be considered complementary to each other and the most suitable type of dialysis for a given patient at the time should be used. Because children with ESKF have a lifetime of KRT ahead of them, preserving dialysis access including peritoneum and vascular access is paramount. There are few medical contraindications for either HD or PD, so the choice of dialysis modality must be largely based on patient (and caregiver) choice and the availability and expertise of the treating center. The factors to consider when selecting a dialysis modality are shown in Fig. 72.2 and the advantages and disadvantages of different dialysis modalities for children are shown in Table 72.4 .

The factors to consider when selecting a dialysis modality.

Although PD is the most commonly available home dialysis modality, some pediatric centers also offer home HD. There are international differences in choice of dialysis modality with PD being twice as common than HD in Europe, whereas the ratio is reversed in Asia and the United States ^, ; overall ∼30% to 50% of children begin dialysis by HD. There is no evidence to suggest a difference either in morbidity or transplant outcome based on the initial dialysis modality in children and adolescents. Given the complexity of dialysis treatment in children, a multidisciplinary team including pediatric nephrologists, dialysis nurses, dietitians, pediatric surgeons, social workers, and psychologists are an important part of the team.

Sometimes, PD may be the only choice in locations where there are no available HD centers at all, or if the patient lives in a remote area where pediatric HD service is lacking, even in high-income countries. ^, The treatment burden of dialysis must be carefully considered. Parents of patients on KRT become high-level health care providers, are required to solve problems and seek information, and provide financial resources and practical skills. Caregivers are in continuous need of emotional, psychological, and financial support. These stresses cause fatigue, as well as disruption of work and social life. Thus once a patient approaches ESKF, the family’s abilities and socioeconomic and psychological background must be carefully evaluated by a multidisciplinary team including a nephrologist, dialysis nurse, social worker, and psychologist.

The choice of selection of HD versus PD is greatly influenced by patient age: According to the NAPRTCS 2011 report, of 927 patients initiating dialysis at the age of 0 to 1 year, 857 (92%) were on PD compared with 552 of 727 (76%) in the 2- to 5-year age group, 1373 of 2125 patients (65%) in the 6- to 12-year age group, and 1648 of 3250 (51%) in the 13- to 18-year age group who were on this dialysis mode. In Australia and New Zealand, similar proportions are seen: 89.8% of infants were initially treated with PD, with declining percentage to 34.5% of adolescents. Similar trends were noted in the United Kingdom: at age 0 to 2 years, 80% were on PD and at age 16 to 18 years only 10.4% were on PD. In the United States, there is a steady trend of increased use of HD instead of PD, from 34% of patients treated with HD in 1991 to 54% in 2010. This trend was not found in the United Kingdom, where a relatively constant percentage, around 30%, of the pediatric dialysis patient population has been on HD since 1996. Similarly, in Australia and New Zealand, 30% to 40% of all dialysis patients were on HD during the period 2006 to 2015.

Infants are a unique group for whom the widespread recommendation is to preferentially treat with PD and to initiate HD only if PD is not feasible. An ESPN/ERA registry study has shown that of the 1063 infants, 13.7% were initiated on HD rather PD. The HD and PD groups had similar characteristics at initiation, and prospective mortality and time to transplantation were the same. The proportion of patients younger than the age of 1 year at initiation of dialysis has steadily increased and doubled from 10% to around 20% of all pediatric dialysis patients from 1990 to 2010.

A well-functioning vascular access is crucial for dialysis. Recent guidelines suggest that children requiring chronic HD should start with a functioning arteriovenous fistula (AVF) where appropriate. AVFs have better patency rates, lower access–related infection rates, and decreased need for access revision compared to central venous lines (CVCs). However, CVCs remain the most common access type in all pediatric age groups, due to the perceived ease of access creation. The median access survival time is 3.14 years for AV fistula (95% CI 1.22–5.06) compared with 0.6 year for a CVC (95% CI 0.2–1.00). Additional advantages attributed to AV fistulas including better clearance and higher albumin and hemoglobin levels are less convincing. CVCs cause more damage to blood vessels and the resulting central venous stenosis may preclude future CVC insertion or ipsilateral AVF creation. This is important in children, who have a lifetime of KRT ahead of them, often also for a second dialysis period after failure of kidney transplant, and therefore blood vessel preservation is crucial. Therefore AVF is the preferred vascular access, except for children weighing less than 15 to 20 kg, but this depends on the skills of the vascular surgeons.

Clinical practice recommendations promote the use of AVFs over CVCs, but pediatric registries suggest a different picture. A survey in Europe and the International Pediatric Hemodialysis Registry found that the use of CVCs was more frequent than recommended in prevalent pediatric HD patients: 40% had an AVF or a graft versus 57% who had a CVC. ^, Only in children older than 15 years of age were there more patients with AVF as an access.

A subcutaneously tunneled, cuffed CVC may be preferred in small children or those with an anticipated short dialysis course before transplantation or transition to PD. The type and size of catheter and site of insertion are individualized for each patient. CVCs should be adjusted to patient size, aiming for the largest-bore catheter that can be accommodated in the vessel without the risk of vascular damage and stenosis ( Table 72.5 ) . The internal jugular veins (IJVs) are the first choice; the subclavian and femoral routes are associated with increased vessel stenosis and infection, respectively. CVCs are inserted in the IJV so that the tip is in the right atrium using ultrasound guidance. Ultrasound and fluoroscopic guidance are used to prevent catheter malposition.

The main CVC complications are infections and malfunction. Infections are more common in younger patients, possibly due to close proximity of the exit site to the infection source: diapers and gastrostomy tubes. ^, Infection rate is usually 1.5 to 4.8/1000 CVC days, although an exceptionally favorable report documented a rate of 0.5/1000 CVC days. Malfunction may be the result of thrombosis, fibrin sheath formation, vascular stenosis, or mechanical damage to the CVC. If local fibrinolysis fails, the CVC must be replaced. A major complication with long-term sequelae is central venous stenosis. It occurs more often in younger patients, with small-diameter veins, with prolonged use of CVCs, with many reinsertions, and possibly also depending on CVC locations. The worst outcome is for subclavian veins, followed by left IJ vein, whereas the right IJ vein has the smallest chance of becoming obstructed. Clinically, stenosis is often asymptomatic until an ipsilateral AV fistula is constructed, but sometimes superior vena cava syndrome may develop.

Infection in vascular catheters may be at the exit site, in the subcutaneous tunnel, or in the catheter. The development of biofilm within the catheter makes bacterial eradication particularly difficult. Line sepsis may present as a rigor soon after starting dialysis, fever with raised C-reactive protein, or septicemic collapse. Factors increasing the risk of catheter infection include exit site and/or tunnel infection or contamination of the hub, failure of aseptic technique, frequent need to access the catheter during dialysis and long duration of use, use of nontunneled rather than tunneled lines, immunosuppression, hypoalbuminemia, diabetes, and nasal and cutaneous colonization with Staphylococcus aureus. Postdialysis heparin or alteplase into the line decreases infection risk by decreasing clot formation, which predisposes to infection.

An AVF may be created at the wrist (radiocephalic or radiobasilic), the elbow (brachiocephalic), or by basilic vein transposition to create a brachiobasilic fistula. Distal vessels of the nondominant arm are preferred; however, the larger the vessels, the greater the chance of success (intraluminal venous diameter >3 mm is preferred). A child who has had previous central lines may need upper limb venography to confirm patency of the central vessels because ultrasound examination of the proximal vessels can be misleading, and veins cannot easily be seen under the clavicle. The average AV fistula blood flow is 800 to 1200 mL/min. In adults, this is significant compared with the normal cardiac output (CO): normal adult CO is 5.2 to 8.6 L/min and the flow through an AV fistula will increase CO by 9% to 23%. A similar flow in a child, required for adequate dialysis, may increase the CO of a child with a 0.7-m ² body surface area (BSA), whose normal CO is 2.1 to 3.5 L/min, by 23% to 57%. Theoretically, this may lead to high-output heart failure, especially in face of the high cardiac morbidity in dialysis patients; in practice, this rarely becomes clinically significant.

AVFs show an average maturation time of 8 to 10 weeks ^{, ,} and can be canulated by rope-ladder or button-hole technique. Because fistula thrombosis/malfunction is the main complication, and this is usually the result of vessel stenosis, routine surveillance of the fistula by Doppler ultrasound performed by an experienced professional should be performed. Vascular surgeons and invasive radiologists should be available to detect and treat stenosis or thrombosis promptly to salvage the AVF.

These are rarely used in children and usually a last resort option. An artificial conduit can be inserted between an artery and vein subcutaneously (graft) where it can be needled, or it may be brought out externally (shunt), where the loop can be disconnected to attach to dialysis lines. Grafts and shunts are, like AVF, also liable to stenosis (particularly at the anastomosis site) and clotting.

• 1.

  Hemodialysis machine: Pediatric hemodialysis machines must be capable of low blood flow speeds, use lines of varying blood volumes, and have precise ultrafiltration (UF) control so that they can be used even in infants; small inaccuracies in volumes removed that are negligible in adults can have a major hemodynamic effect on the small child. Automated volumetric UF systems that precisely monitor pressure across the dialysis membrane (transmembrane pressure, TMP) allow accurate fluid removal. The inherent level of UF inaccuracy mandates careful assessment of the patient’s volume status, especially in smaller children. Several options are now available for continuous online monitoring of changes in hematocrit and blood viscosity (blood volume), oxygen saturation, blood pressure, and pulse, as well as for urea clearance. Modern dialysis machines allow for both HD and hemodiafiltration (HDF) and are suitable for children above 10- to 17-kg body weight.

• 2.

  Dialyzer: The dialyzer (dialysis filter) is selected on the basis of its surface area and the priming volume ( Table 72.6 ). The dialyzer membrane surface area should be approximately equal to the patient’s BSA. Conventional hemodialysis uses a low-flux (small pore size) membrane, and solute removal is primarily by diffusion. High-efficiency hemodialysis refers to a more rapid removal of urea, potassium, and other small solutes. This is achieved by using a low-flux membrane with a high efficiency (KoA) for removal of small solutes. It is also achieved by using a larger surface area membrane and a high blood flow. Even when using high-flux dialyzers, conventional HD is limited in clearing middle-sized molecules, which are better removed by convection using HDF.

  Table 72.6

  Dialyzer Characteristics, Options, and Considerations for Pediatric Hemodialysis

  Dialyzer Characteristic	Range of Options	Considerations
  Surface area	0.2 m 2 -2.5 m 2	Larger surface gives greater capacity for clearance and ultrafiltration but requires high blood flow to achieve these advantages
  Priming volume	18 mL-140 mL	Consider along with tubing set volume to determine maximum safe extracorporeal volume, especially for smaller patients
  Membrane material	Cellulosic, modified cellulose, artificial	Greater biocompatibility with modified cellulose and further with artificial (plastic) membranes
  Sterilization	Ethylene oxide (ETO), γ irradiation, steam, electron beam	Allergic reactions associated with ETO; fewer reactions with γ and steam
  Flux	Low flux vs. high flux	Higher-flux dialyzers have higher permeability (e.g., larger pores), permitting higher rates of middle molecule clearance and higher UF rate at a given prescription
  Clearance (typically described as K D for urea [mL/min] at blood flow of 200 mL/min)	K 0 A Urea 170 mL/min-1700 mL/min	Varies with surface area, flux, membrane material
  Ultrafiltration capability (K UF mL/h/mm Hg)	K UF 7 mL/h/mm Hg-111 mL/h/mm Hg	Varies with surface area, flux, membrane material

• 3.

  Extracorporeal circuit: The tubings and hemodialyzer are selected on the basis that the child can tolerate 8% (up to a maximum of 10%) of their total blood volume (TBV, 80 mL/kg estimated dry weight) in the extracorporeal circuit ( Table 72.7 ). Lines are primed with saline. However, in the very young/small, even the smallest circuit may exceed the safe extracorporeal volume and blood priming of the circuit is needed for each dialysis session, which increases the risks of infection and sensitization before transplantation.

  Table 72.7

  Dialyzer and Tubings with Extracorporeal Priming Volumes ∗

  Name	Company	Surface Area (m 2 )	Volume (mL)
  Dialyzers
  FX Paed	Fresenius	0.2	18
  Sureflux 03L	Nipro	0.3	25
  Sureflux 05E	Nipro	0.5	35
  FX40	Fresenius	0.6	32
  Sureflux 07E	Nipro	0.7	45
  Sureflux 09E	Nipro	0.9	55
  FX50	Fresenius	1.0	53
  Sureflux 11E	Nipro	1.1	65
  Sureflux 13E	Nipro	1.3	78
  FX60	Fresenius	1.4	74
  FX80	Fresenius	1.8	95
  Tubings
  Neonatal	Gambro		33
  Pediatric	Gambro		85
  Adult	Gambro		132
  Paed AV set ONLINEplusBVM	Fresenius		111
  Paed SN set ONLINEplusBVM	Fresenius		142
  Adult HDF	Gambro		151
  Adult AV set + BVM	Fresenius		192

  AV, Arteriovenous; BVM, blood volume monitoring; HDF, hemodiafiltration; SN, single needle.

• 4.

  Water: A water treatment unit purifies municipal water containing potentially hazardous contaminants to dialysis-grade water that is safe if diffused into the bloodstream. A water treatment unit comprises filters to remove particulate matter, activated carbon for chlorine and chloramines, and water softeners to remove calcium and magnesium. Finally, a reverse osmosis unit purifies softened filtered water before it is of a quality suitable for dialysate production.

• 5.

  Blood flow rate (Q _B ) and dialysis flow rate (Q _D ): The blood pump flow rate is an important determinant of solute clearance, allowing maximum diffusion and convection; the vascular access determines the maximum achievable Q _B. Wider-bore CVCs and AVF allow higher Q _B . The blood flow rate should be up to body weight (kg) × 5 to 8 mL/min. Dialysis fluid flow rate (Q _D ) is required to be 1.5 to 2 times that of Q _B.

UF is targeted to achieve dry weight, which is the post-HD weight below which the child will become symptomatically hypotensive. Current guidelines recommend clinical assessment of fluid status, dry weight, and dietetic assessment, at least monthly, and more often in small children. Objective assessment of dry weight using body composition analysis is a useful adjust to clinical assessment. ^, The safe limit of UF is no more than 10 mL/kg/hour, and no more than 5% to 8% of body weight should be removed in one session. Adequate UF is difficult to manage for patients who have significant volume overload (a high interdialytic weight gain), have poor vascular tone (critically ill, sepsis), have cardiac dysfunction (risk for myocardial hypoperfusion), and in young children due to small intravascular volume. Careful bedside monitoring, with assistance of technology such as noninvasive blood volume monitors (that monitor circulating blood volume by determining changes in hematocrit), can help safe achievement of dry weight with appropriate UF goals. Regular assessment of UF tolerance, using frequent HD or long dialysis times to avoid excessive UF rates, is required.

The success of PD depends on a reliably functioning PD catheter. There are many different catheter designs with different intraperitoneal configurations (curled or straight), number of cuffs, and tunnel configurations. The most used catheter for children is the Tenckhoff catheter, which is surgically positioned in the pelvis via a subcutaneous tunnel by open surgical or laparoscopic technique. Pediatric PD guidelines recommend a double-cuff Tenckhoff catheter with a downward or lateral subcutaneous tunnel configuration and downward-pointing exit site. A Cochrane review has shown that there is no specific catheter insertion technique or catheter design that prevents or reduces the risk of peritonitis. However, randomized trials suggest that a downward-pointing exit site for the PD catheter is the single most important factor in preventing peritonitis, although this may not be relevant for the youngest children. A schematic representation of the catheter and cuff placement in the abdominal wall is shown in Fig. 72.3 .

A schematic representation of a peritoneal dialysis catheter and cuff placement in the abdominal wall.

Before catheter insertion, a detailed clinical examination is required to determine the position of the exit site, particularly in those who may have had prior surgery and those with abdominal stomas, and also to check for hernias, as these may need to be repaired at the same time as the PD catheter insertion. Constipation is a common issue in patients with CKD and must be addressed before PD catheter insertion; constipation is the most common cause for catheter migration and nonfunction. Preoperative antibiotic prophylaxis reduces the incidence of early-onset peritonitis. The choice of antibiotic should depend on local antibiotic sensitivity patterns: most centers in the United Kingdom use intravenous vancomycin, but routine prophylaxis with vancomycin carries a risk of vancomycin-resistant enterococci emergence, and European Best Practice guidelines recommend that a single dose of first- or second-generation cephalosporin is used just before catheter placement.

Ideally, the catheter should be left to rest for 10 to 14 days before PD initiation in order for the tunnel to heal. However, if the patient requires urgent dialysis, the catheter may be used immediately; in this situation, small dialysate volumes help to minimize intraperitoneal pressure. Standardized and rigorous exit-site care is essential for reducing the risk of exit-site infection and peritonitis. As shown by the quality improvement project, Standardizing Care to improve Outcomes in Pediatric End-stage renal disease (SCOPE), the implementation of standardized follow-up care led to a significant reduction in the average monthly peritonitis rates. Specifically in infants or children on PD, the catheter must be placed as far as possible from sources of infection such as diapers or gastrostomy tubes.

The PD prescription can be delivered by several different techniques. The PD modality is selected depending on the patient’s needs for solute and fluid clearance; the residual kidney function, if any; and its suitability for the child and family’s lifestyle and daily routine. General considerations related to KRT by PD are provided in Chapter 63. The different PD modalities are shown in Fig. 72.4 , and a brief summary of the technique and its relative advantages and disadvantages are discussed in Table 72.10 .

Schematic representation of various peritoneal dialysis regimens based on a standard adult fill volume of 2000 mL of dialysis fluid.

The fill volume in children is adjusted to the child’s body surface area, with an optimal fill volume being 800 to 1000 mL/m ² in those younger than 2 years and 1000 to 1400 mL/m ² in older children. The shaded area represents the fill volume and dwell time, and the x-axis represents a 24-hour period. PD, Peritoneal dialysis.

Although the goals of adequate solute and water removal are universal, the PD prescription must be tailored to individual needs and a “one-size-fits-all” approach cannot be considered. For all PD prescriptions, the child’s size (weight and BSA), residual kidney function, biochemistry, and peritoneal membrane function should be considered. After selecting the appropriate dialysis modality, the key aspects of the PD prescription that need to be addressed are (1) type of dialysis fluid, (2) the volume of dialysate per PD exchange (cycle), and (3) the total therapy time and number of cycles in that period. Details follow:

• 1.

  Peritoneal dialysis solutions used in children are identical to those used in adults and described in Chapter 63, although biocompatible PD fluids are favored in children. Recommendations by the European Pediatric Dialysis Working Group include the use of the lowest glucose concentration possible and fluids with reduced glucose degradation products (GDPs) content whenever possible. In the face of a paucity of studies prospectively comparing low GDP solutions or various buffers, evidence-based recommendations cannot yet be given, but it seems that correction of metabolic acidosis with pH-neutral bicarbonate-based fluids is superior to single-chamber, acidic, lactate-based solutions.

• 2.

  Volume of dialysate exchanged in each cycle is calculated on the basis of the child’s BSA; this is approximately 600 to 800 mL/m ² in those younger than 2 years and 1000 to 1200 mL/m ² (up to a maximum of 1400 mL/m ² ) in older children. The maximum safe dialysate volume per fill is determined based on intraperitoneal pressure at less than 14 cm H ₂ O in children older than 2 years and 8 to 10 cmH ₂ O in younger children. A high intraperitoneal pressure may impede lymphatic absorption (causing reduced UF), physical discomfort, dyspnea, hernia formation, and gastroesophageal reflux.

• 3.

  Total therapy time and number and duration of cycles depend on the child’s age, diet, and amount of urine passed as shorter cycles promote more UF. Infants and younger children, because of their mainly liquid diet and the need for longer dialysate drain times, need longer dialysis treatment times, usually approximately 12 hours, whereas older children do well on 8 to 10 hours of dialysis.

The duration of each PD exchange (cycle) will depend on the peritoneal membrane function. This can be measured by the Peritoneal Equilibration Test, but as a rule, infants and young children are usually high transporters (rapid small solute equilibration), whereas older children and young adults are usually low transporters (good UF with minimal reabsorption). High transporters require more frequent exchanges with shorter dwell times to avoid reabsorption and the use of icodextrin for the daytime dwell (see Fig. 72.4 ). Low transporters may require longer dwells with higher-volume exchanges to get adequate clearance.

Importantly, peritoneal membrane characteristics differ between individuals and can also change in any given individual depending on their dialysis prescription, complications of dialysis such as peritonitis, and the duration of time spent on dialysis. After a long duration of PD, particularly if highly osmolar dialysate fluids are used or if the child has suffered from peritonitis, the peritoneum can undergo sclerosis and even fibrosis so that it becomes less effective as a dialysis membrane. An understanding of peritoneal membrane characteristics allows clinicians to adjust the PD prescription to obtain optimal UF and solute clearance.

Daily monitoring of the child’s PD is now possible through use of remote patient monitoring (RPM) platforms with some dialysis machines. RPM gives the PD team an overview of how effective the dialysis program is in terms of the daily UF achieved and the child’s weight and BP. Problems with PD catheter drainage, as well as reduced UF, can be detected and the PD program is manipulated remotely to address these issues. When used effectively, RPM can lead to more individualized PD prescriptions for patients and proactive nursing care that can reduce hospital visits.

The development of new PD modalities, such as adapted automated PD, may also help achieve optimal sodium and water removal. Adapted automated PD aims to improve solute and water removal by using initial cycles with short dwells and small fill volumes to increase sodium-free UF, followed by cycles with longer dwells and larger fill volumes, which favor clearance of solutes (sodium) and uremic toxins. Pediatric studies are ongoing to confirm the advantages of this regimen.

The diagnosis of peritonitis in a child on PD should be considered in the presence of cloudy peritoneal effluent, fever, chills and rigors, abdominal pain, vomiting, and in late cases, septic shock. The child’s caregivers are trained to have a low threshold for suspecting peritonitis if any of these symptoms are present, immediately contacting their dialysis unit, and collecting a sample of peritoneal effluent that is sent for cell count, differential count, and culture to confirm the diagnosis of peritonitis. An empiric diagnosis of peritonitis is made if the dialysate effluent white blood cell (WBC) count is greater than 100/mm ³ and at least 50% of the WBCs are polymorphonuclear leukocytes (4).

In PD-related peritonitis, instillation of antibiotics into the peritoneum is the preferred route of administration because high bactericidal concentrations can be rapidly established at the site of infection and most antibiotics are absorbed from the peritoneal cavity into the bloodstream. Intravenous antibiotics may be necessary in patients with severe systemic involvement or if the patient is immunosuppressed. Antibiotic therapy must begin with a broad spectrum and include cover for both gram-positive cutaneous flora and gram-negative enteric pathogens, and choice depends on local sensitivity patterns. Once PD fluid cultures are obtained, antibiotics are adjusted according to the final culture and antibiotic sensitivity results. It is recommended that all patients receive oral antifungal therapy (nystatin) while on antibiotics. Updated guidelines on the diagnosis and management of infectious complications of PD in children have been recently published by the International Society for Peritoneal Dialysis, closely reflecting similar guidelines in adults on PD.

PD catheter removal should be considered in case of fungal peritonitis, severe sepsis, relapsing PD-related peritonitis, or a persistently raised PD fluid WBC count despite appropriate antibiotic therapy.

Peritonitis rates are higher in infants (1/15.3 months) compared with adolescents (1/21.2 months; NAPRTCS data), probably due to close proximity to contamination sources such as diapers or gastrostomy tubes in the young ages. Higher peritonitis rates have been reported in gastrostomy-fed children on PD; hence whenever possible, a gastrostomy should be inserted before PD catheter insertion and commencing PD. ^, When this cannot be avoided, antibiotic and antifungal prophylaxis at the time of PD catheter insertion in children with a gastrostomy is essential. A report of the Italian Registry of Paediatric Chronic Dialysis found somewhat better results: 1/20.7 months in infants and 1/28.3 in older children. In a survey including 47 centers in 14 countries, 44% of the peritonitis events were due to gram-positive bacteria, 25% due to gram-negative bacteria, and 31% were culture negative. Less than half of the patients had fever higher than 38°C, less than half had abdominal pain, and 70% had marked effluent cloudiness. Of 482 children, 420 (89%) made a full recovery and 39 (8.1%) had to discontinue PD permanently due to UF problems, adhesions, uncontrolled infection, and secondary fungal infection. As such, infections are the most common reason for modality change for patients on PD. In another study, including 501 peritonitis events from 44 pediatric dialysis centers in 14 countries, significant regional variability was found for both bacteria type and sensitivity. This makes recommendations for global treatment guidelines more complex. Permanent discontinuation of PD varied considerably, from 0/18 episodes (Argentina) to 9/46 episodes (20%) in Eastern Europe. Of note, a recent Cochrane study assessing evidence in PD-associated peritonitis (including adults) concluded that the data available are poor. In general, review conclusions were based on a small number of studies with few events, in which risk of bias was generally high; interventions were heterogeneous, and outcome definitions were often inconsistent. There were no randomized controlled trials evaluating optimal timing of catheter removal and data for APD were absent.

Other infections include exit site and tunnel infections. Preventive measures include aseptic handling of the catheter, daily cleansing, immobilization, and applying topical antibiotics and intranasal mupirocin application. Interestingly, an alternative measure of preventing exit-site infection with antibacterial honey applied to the exit site found noninferior results compared with intranasal mupirocin.

Initiating dialysis in infancy is rare and remains a difficult medical and ethical situation. Most infants born with CKF have been diagnosed antenatally, so parents may have developed preconceptions about the outcome for their child, which may or may not be appropriate. These expectations, alongside the high frequency of prematurity and comorbidities, may affect the decision and technical ability to proceed with dialysis. Nonetheless, the survival of infants who initiated dialysis in the first year of life has improved, with nonrenal comorbidities being the main cause of death. PD is the KRT modality of choice for infants in whom vascular access is likely to be more difficult and serves as an essential bridge to kidney transplantation, which is only possible in most centers once the child weighs >10 kg.

Several aspects of infant PD present a technical challenge, both in terms of gaining an effective dialysis access and the day-to-day management of the dialysis prescription. PD catheter insertion in infants can be difficult and more prone to leakage or blocking by omentum. Positioning the catheter exit site above the nappy line but away from a potential future gastrostomy site, as well as other abdominal stomas, is important to prevent cross-infections and peritonitis. Infants are at a higher risk of developing gastroesophageal reflux, hernias, and peritoneal fluid leaks, so the PD prescription requires careful adjustment to achieve adequate solute and fluid clearance without causing a significant increase in the intraabdominal pressure. In addition, excess fluid removal is particularly problematic in infants, especially for those who are oligoanuric, because an infant’s diet is liquid. To achieve this lower dialyzate, fill volumes are used with short, low(er) volume exchanges and a longer total PD therapy time per 24-hour period.

Achieving optimal nutrition and growth is particularly important in infants on dialysis. Enteral tube feeding via a nasogastric tube or gastrostomy is often necessary to support nutrition and medication administration. Careful attention to the management of CKD-mineral bone disorder (CKD-MBD) is required as the growing bones of children rapidly accrue calcium and phosphate. Complication rates in infants, especially those with comorbidities, remain high with an increased risk of PD-related peritonitis, significant neurocognitive and developmental delay, poor nutrition, and cardiovascular complications.

The principles and technical aspects of CKRT are similar in children and adults. They are described in detail in Chapters 62 and 64. Dedicated machines, dialyzers, and replacement fluid are used for CKRT. The CKRT prescription is similar to that for intermittent hemodialysis but includes a set UF rate that is decided on the basis of the child’s fluid status and the volume of fluid administered. It is usually expressed in mL/kg body weight/hour or as a percentage of the total body weight. There are no robust studies in children exploring optimal CKRT dosing, but the consensus in adults is that there is no benefit to delivering high-volume CKRT reaching >25 to 30 mL/kg/hour clearance. Due to the continuous nature of CKRT, children may develop hypokalemia or hypophosphatemia, and these may require replacement in the dialysate or IV fluid.

Although CKRT shares similar principles with HD, both the blood and dialysate flows are significantly slower, resulting in a lower hourly clearance rate. This is compensated for by extending the clearance time (over 24 hours, CKRT provides solute clearance comparable with a 4-hour HD session), using hemodialysis or hemodiafiltration in the CKRT circuit. The main advantage of CKRT in the critically ill child is the ability to maintain hemodynamic stability. CKRT and PD are continuous in nature, but the former provides much greater and more controlled daily clearance rates.

Double-lumen catheters are often used, and the diameter size is selected on the basis of the patient’s BSA. One must reconcile the need for adequate blood pump flow rates with the desire to limit vessel trauma. Although femoral catheters are four times more often used than IJ ones, the latter result in a far better circuit survival. Furthermore, femoral catheters, unless located in the inferior vena cava (IVC), are significantly affected by patient movement and may require the patient to be sedated or even paralyzed for proper use. Taken together, the preferred site for vascular access is in the right IJ vein with the distal tip of the catheter placed in the right atrium of the heart. The recommended blood flow is 3 to 10 mL/kg/h with a relatively higher flow in the very young child if an adult-size device is used. Although no prospective study has randomized pediatric patients to different effluent dose targets, the common practice is to use an effluent dose of approximately 2000 mL/1.73 m ² /hour to achieve good metabolic control within 24 hours of CKRT initiation. One study demonstrated that hypotension after connection to CKRT occurred in 49.7% of connections a median of 5 minutes after beginning therapy; therefore blood flow should be increased slowly. If the circuit volume is large relative to the child’s size, albumin or blood priming may be necessary. In previous years, lactate-based fluids were used, resulting in lactic acidosis, cardiac dysfunction, and hypotension. Citrate or bicarbonate-based dialysate and replacement fluids are currently considered standard of care. Large volumes of replacement fluid, containing varying amounts of electrolytes, are administered. This requires close monitoring of the patient’s laboratory profile to ensure stable acid-base and electrolyte balance.

Activation of the clotting cascade, impacting circuit longevity, results from the contact of blood with the artificial filter, as well as from slow blood flow and small catheters used. This mandates the administration of anticoagulants: Although a single pediatric study demonstrated that heparin and citrate are equally efficacious with more bleeding episodes occurring with the former, most adult studies favor regional citrate, together with systemic calcium infusion, for better longevity of the circuit and fewer bleeding episodes. Citrate is metabolized to bicarbonate in the liver and should be cautiously used in patients with liver failure and lactic acidosis. In a multicenter pediatric survey, it was shown that citrate was used in 56% of children and heparin in 37%. In a recent retrospective study it was demonstrated that hemofilter survival was higher in the citrate group than in the heparin group. The risk of hemofilter clotting was significantly increased when heparin was used, regardless of hemofilter size and pump flow.

Patient outcomes are largely dependent on the underlying disease state and comorbid conditions. According to the CKRT Registry, overall mortality was 42%, with higher rates in children with liver failure or liver transplant, pulmonary disease, or lung transplant and in stem cell transplant recipients, ranging from 55% to 69%. Fluid overload at the initiation of CKRT is an independent risk factor for mortality: patients with >20% fluid overload were 8.5-fold more likely to die. Of note, this incremental increased risk of mortality was greater than that conferred by multiorgan failure (defined as receiving one vasoactive medication and invasive mechanical ventilation), sepsis, or malignancy, none of which were modifiable risk factors. Patients with >20% fluid overload demonstrated similar illness severity to those with 10% to 20%, inferring that the most severely fluid-overloaded group was not sicker and the excessive fluid load may have been iatrogenic. A prospective multicenter study in children, addressing the effect of early initiation of CKRT on outcome measures, including mortality, is currently required.

As of January 2024, 1208 children were listed in the United States for a kidney, out of a total of 88,695 patients listed (1.4%). Fig. 72.6 and Table 72.11 show data on age, race, gender, and ESKF etiology at time of transplant in the United States. Of the 27,332 kidney transplants performed in 2023, 789 were in children (2.8%), a significantly higher proportion than those waitlisted. Whereas there was an increase in living kidney donation to pediatric patients between 1987 and 2001, with a peak of 64% of kidney transplants, there has been a gradual decrease in this trend since 2001 in the United States, falling to only 31.5% by 2017.

One-year adjusted all-cause mortality rates in incident pediatric patients with end-stage kidney disease by (A) age with comparison to young adults (aged 0–29 years) and (B) modality (aged 0–21 years only), 2005 to 2009, and 2010 to 2014. HD, Hemodialysis; PD, peritoneal dialysis; Tx, kidney transplantation.

From 2017 USRDS Annual Data Report. Chapter 7: ESKF among children, adolescents and young adults. Am J Kidney Dis. 2018;71(S1):S383–S416.

Changes in allocation policies of deceased donor kidneys, which favor pediatric patients and shorten waiting times, were in part responsible for this decrease in living donation. The most current deceased donor allocation policy in the United States allocates the top 20% of kidneys in the kidney donor profile index to candidates in the top 20% of expected post-transplant survival. Priority is given to pediatric candidates, after candidates who were prior living donors, or candidates with zero-HLA mismatch kidneys, and calculated panel-reactive antibodies (PRAs) 98% and up. ^,

In Europe, data collected from 32 countries demonstrated that 43% of pediatric kidney transplants were from living donors. Most countries implement an allocation scheme, which prioritizes pediatric patients on the waiting list, in particular with regard to younger donors. Eurotransplant doubles the points for HLA-antigen mismatches and provides bonus points for waiting time for pediatric patients (<16 years at the time of listing in Europe, <18 years in Germany). Scandiatransplant gives priority to pediatric recipients if a donor is <40 years old. The Swiss Organ Allocation System gives preference to ABO-compatible candidates <20 years if a deceased donor is <60 years.

In Europe, pediatric kidney transplant rates vary widely between 0 and 13.5 per million children, higher in areas with a higher gross domestic product or a high pediatric priority policy. Access to transplantation differed according to ethnic groups, with black and Asian children less likely to receive a kidney transplant than white patients, mostly due to decreased living donation. Only a few reports emanate from middle- and low-income countries. The available data suggest that kidney transplant rates are 0 to 4 per million children.

Preemptive kidney transplantation, meaning no dialysis performed before transplantation, is the modality used in 22% of pediatric ESKF patients in the United States and 27% in Europe. ^, In contrast, rates of preemptive transplant are less than 6% in adult recipients. There is a wide disparity in European countries in their overall rates of preemptive transplantation ranging from <5% to >60%. Worldwide, preemptive transplantation is mostly from living donors. While socioeconomic factors that limit access to preemptive transplantation can confound the results, studies suggest advantages to preemptive transplants. In Eurotransplant data, preemptive transplants have a higher acute rejection-free proportion at 3 years post transplant (52% vs. 37%) and improved 6-year allograft survival when compared with dialysis patients (82% vs. 69%). In a recent meta-analysis of 22 studies ( n = 22,622), both overall graft loss relative risk (RR; 0.57, 95% CI: 0.49–0.66) and acute rejection risk (RR: 0.81, 95% CI: 0.75–0.88) were lower in preemptive kidney transplants compared with transplantation after dialysis.

Kidney transplantation in infants younger than the age of 1 year is uncommon, reported at 0.8% of pediatric transplants in the NAPRTCS database. As noted earlier, deceased donor kidneys from younger donors are often allocated preferentially to children. However, utilization of deceased donor kidneys from infants and small children holds unique challenges as they are more prone to vasospasm and graft thrombosis, urinary complications, and possibly glomerular hyperfiltration. Such grafts do well when transplanted en bloc into adult recipients. Recent data have shown good results in children receiving kidneys from small pediatric donors, either a single kidney or both donor kidneys en bloc. One study from Europe reported a significant increase in allograft size after transplantation and superior graft function in the recipients of pediatric kidneys.

Although allocation systems usually try to match pediatric recipients with the best brain-dead donors, a recent study showed excellent 3-year graft survival in children receiving a kidney after even circulatory death of the donor.

Living donor kidney transplantation offers better graft survival than deceased donor transplantation, though other factors, such as donor age, HLA matching, and sensitization, play a significant role. Kidney grafts from middle-aged living donors afford improved long-term allograft survival compared with younger deceased donor kidneys. However, data from the Collaborative Transplant Study showed that well-matched deceased donor kidneys compare favorably with living donor grafts with many HLA mismatches. The question of whether it is advantageous to perform living or deceased donor transplantation first has been studied, as most pediatric kidney transplant patients will eventually require a second transplant. For most candidates, living donor transplantation as the first treatment option provides the best long-term benefit, except possibly for children who are highly sensitized, for whom deceased donor transplantation first, while they have priority as children on the waiting list, may be an advantage.

When the patient approaches KDOQI stage 4 CKD, preparation of the patient for transplant and evaluation of potential living donors should be pursued. However, graft survival is not indefinite and no benefit has been demonstrated for transplantation before reaching ESKF. Preemptive transplantation is associated with better patient and graft survival when compared with children who receive dialysis for longer than 1 year. Other considerations, particularly morbidity on dialysis, neurocognitive development, and decreased linear growth, also favor transplantation as the first modality of KRT. Certain situations may preclude preemptive transplantation, such as active nephrotic disease, which is also a hypercoagulable state, presenting a high risk of thrombosis during transplantation. To be a candidate for transplantation without dialysis, the patient must have some minimal residual kidney function and urine volume.

Two age groups require special consideration. Small children may have inferior early results due mainly to technical difficulties and graft thrombosis, though long-term results are best in these recipients—the infant paradox. Some specialized centers, with greater experience in this age group, have demonstrated excellent results. The alternative, remaining on dialysis for a prolonged period, also carries a high morbidity and mortality, particularly in this young age group. Most units refer young patients for transplantation when they reach a weight close to 10 kg or length >65 cm, but the risks and benefits must be carefully considered in each case. The other potentially complex age group is adolescents. While surgical issues are simpler in this age group compared with younger children, there is a higher rate of acute rejection episodes, shorter graft survival, and lower creatinine clearance in this group—the adolescent paradox. This is mostly due to nonadherence, which is more frequent in teenagers and may manifest as missed doses of immunosuppressive medications, more extensive “drug holidays,” missed clinic visits, and other high-risk behaviors.

Some clinical situations may present a temporary or permanent contraindication to transplantation. Malignancy should postpone plans for transplantation until complete remission is achieved, chemotherapeutic medications are discontinued, and no relapse is evident. High-dose immunosuppressive therapy after transplantation may increase the risk of cancer recurrence due to enhanced development of micrometastases. The period between completion of therapy for cancer and listing for transplant will also depend on the type of malignancy and its characteristics. Immunosuppressive medications also impair the body’s ability to fight infection; therefore active infection is a temporary contraindication to transplantation and latent infections should be sought out and addressed before proceeding to transplant. Severe comorbidities or profound neurodevelopmental delay may render a child unsuitable for transplant. However, intellectual disability does not impact patient or graft survival and should not be a barrier to transplantation in itself. ^, Each child should be evaluated individually to assess the impact of transplantation on life expectancy, quality of life, and rehabilitation, taking into account the wishes of the family. Recalcitrant nonadherence to medication or dialysis schedule, diet, and clinic appointments may also be a temporary contraindication to transplantation.

A comprehensive evaluation of the pediatric kidney transplant candidate is crucial to achieve the best results. The goals of this workup are to provide an optimal plan for each individual patient, reduce complications, and increase long-term patient and graft survival. The main elements of the pretransplant evaluation are summarized in Table 72.12 . Some of the important issues are discussed in the following sections.

Transplantation immunobiology is covered in detail in Chapter 68. The importance of HLA matching between donor and recipient was recognized in the 1960s and 1970s and became the cornerstone of deceased donor allocation policy. A clear association between number of HLA mismatches and graft survival was shown in several large studies. With the introduction of more potent immunosuppression, the significance of HLA matching may have diminished, while other factors such as peak PRAs, cold ischemic time, and donor age retained their relative importance. However, other recent publications have shown that HLA mismatches continue to adversely affect renal allograft outcomes in the present era, as shown by a study from ANZDATA that included children. Studies that examined the effect of HLA matching specifically in children have also shown that an increase in number of HLA mismatches is associated with decreased graft survival. New allocation algorithms that have been implemented prioritize better HLA-matched kidneys in addition to favoring pediatric patients. Both donor age and HLA matching are important in determining deceased donor graft survival and may be offset by each other.

Most pediatric patients with CKD will need to undergo more than one transplant over their lifetime, and HLA matching of the first kidney graft may have an impact on subsequent transplants. More HLA or eplet mismatches at first transplant are associated with HLA sensitization, longer waiting time for a second transplant, and decreased graft survival. However, the benefits of better HLA matching at first transplant on lifetime with graft function, although significant, are relatively small.

In addition to HLA matching, PRAs are measured in renal transplant candidates and, if positive, should be characterized by solid-phase assays. The presence of DSA is associated with increased risk of acute antibody-mediated rejection (ABMR), as well as chronic AMR and decreased graft survival. When DSAs are further characterized, patients with complement-binding DSA after transplantation are at highest risk of graft loss and ABMR. The deleterious impact of DSA has been demonstrated in children, even when the direct crossmatch between donor and recipient is negative by flow cytometry, though not in all series. ^, Although good HLA matching without DSA is desirable, not all patients will find an appropriate donor. Annual reports show that about 5% of children listed for deceased donor transplantation have PRA >80%. Desensitization protocols, including treatment with combinations of intravenous immunoglobulin (IVIG), plasma pheresis, and rituximab, have shown acceptable intermediate-term outcomes in adults undergoing living or deceased donor transplantation. ^, There are remarkably few case series and clinical trials of desensitization in children. One study used IVIG and rituximab desensitization with alemtuzumab induction in highly sensitized children who underwent deceased donor kidney transplantation. Although a higher rate of acute rejection was seen in the sensitized patients, there was no significant decrease in 6-year graft survival and no increase in infectious complication.

De novo DSA appears in 35% of children in the first years after transplantation, with a high prevalence of DSA to class II and specifically HLA-DQ antigens. These antibodies are associated with increased risk of ABMR and graft dysfunction, regardless of whether they develop in the first year after transplantation or later. Serial monitoring of DSA in children after transplant may help to identify patients at risk for adverse outcomes and guide tailoring of immunosuppression, though benefits are not clear if no ABMR is present and allograft function remains stable.

In the past, ABO incompatibility was considered a contraindication to transplantation because of the high risk of hyperacute ABMR. However, in the past few years there have been reports of successful ABO-incompatible renal transplantation in children. ABO-incompatible transplantation in children is rare and usually involves patients with blood type O or B and low titers of anti-A antibodies, receiving an organ from blood type A donors. ABO-incompatible transplantation is more common in Japan, where most transplants are from living donors. Treatment of these patients with a desensitization protocol has shown long-term outcomes equivalent to the general pediatric transplant population. In Europe, data from the Collaborative Transplant Study, which included children, reported that 1420 ABO-incompatible living donor kidney transplants were performed in 2005–2012, with similar long-term patient and graft survival. A slight increase in death due to infection was seen in the first year after ABO-incompatible transplantation. Treatment protocols initially used pretransplant plasmapheresis in order to remove anti-A/anti-B antibodies, together with splenectomy or rituximab. Newer regimens include immunoadsorption, using either blood group–specific columns or nonspecific columns, to remove the antibodies, together with rituximab to prevent ongoing antibody production after transplantation. Addition of other antibody induction in these patients did not result in altered graft survival. The risk of ABMR is associated with higher titers of anti-A/B isoagglutinins, and the intensity of desensitization may be modifiable according to pretreatment titers, thus potentially opening this option in deceased donor transplantation as well. Immunoadsorption therapy for ABO-incompatible renal transplantation has been associated with surgical bleeding in some cases. Late ABMR in the setting of ABO incompatibility is less frequent than AMR in HLA sensitization. ^{, ,} Despite more intense immunosuppressive regimens, a study in adults did not demonstrate a higher risk of cancer in patients undergoing ABO-incompatible transplantation.

The risk of acute rejection is highest in the immediate post-transplant period, which is why in addition to higher doses of maintenance immunosuppressive drugs, many transplantation protocols include a biological induction agent. These antibody preparations are aimed at depleting T cells or preventing their activation and are administered in the perioperative period.

Antilymphocyte preparations are T-cell depleting-agents that can be polyclonal or monoclonal. The polyclonal antibody antithymocyte globulin (ATG) contains antibodies to a wide variety of lymphocyte antigens in serum of animals (most commonly rabbit) immunized with human lymphocytes.

The anti-CD52 antibody preparation alemtuzumab binds to the CD52 receptor, found on both T and B cells, as well as monocytes and natural killer cells, causing cell lysis. IL-2 receptor antibodies such as basiliximab block T-cell activation without depletion.

In the United States 94.3% of pediatric kidney transplant recipients reported some induction use in 2020. ATG is the most frequently used antibody induction in the United States, particularly in steroid-free protocols and in highly sensitized patients. ^, In NAPRTCS data, T-cell depleting agent usage is steadily increasing to reach 63.6% in 2018 and IL-2-RA therapy remained stable at 35.1%. However, induction therapy is used in only approximately 50% of patients in European countries reporting to the Cooperative European Pediatric Renal Transplant Initiative (CERTAIN), with IL-2 receptor antagonists therapy being the most common.

Alemtuzumab induction has been described in several small series of children, enabling steroid avoidance or tacrolimus monotherapy, with good results. Alemtuzumab has also been used for highly sensitized children in combination with desensitization. Outcomes were similar to nonsensitized patients who received IL-2 receptor blockers. A significant reduction in lymphocyte counts was observed in the alemtuzumab group up to 1 year from transplantation. Concerns regarding the risk for post-transplant lymphoproliferative disease (PTLD) have been raised regarding the use of lymphocyte-depleting agents, particularly alemtuzumab. In a large OPTN database study, T-cell depleting agents, alemtuzumab and ATG, were compared with nondepleting induction agents or no induction, and the incidence of PTLD without induction (0.43%) was shown to be not different from that with basiliximab (0.38%) or alemtuzumab (0.37%) but was slightly higher in patients who received ATG (0.67%). The population in the study had received higher doses of ATG than employed today, higher doses of maintenance immunosuppression, and less antiviral prophylaxis, suggesting that the ATG induction may not currently be a significant risk factor for PTLD. In two recent retrospective multicenter studies, low rATG induction dose ≤7.5 mg/kg (NAPRTCS data) or ≤4.5 mg/kg (Pediatric Nephrology Research consortium data) provided safe and effective outcomes in low immunologic risk pediatric cohorts. ^,

Pediatric studies using the IL-2 receptor blocker basiliximab as induction therapy combined with triple-drug maintenance immunosuppression showed a good safety profile but minimal or no additional benefit compared with no antibody induction. ^, A large ANDATA analysis showed that induction with IL-2 receptor antibodies in pediatric patients was associated with at least a 40% reduction in the rate of acute rejection, though no significant decrease in graft loss was found. Propensity scoring was used to account for confounding factors.

Retrospective data from the OPTN database found that lymphocyte-depleting induction (ATG or alemtuzumab) is more effective at lowering acute rejection rates in African-American pediatric kidney transplant recipients, but no difference was found in other ethnic groups.