Đề ôn thi thử môn hóa (130)

43  Diagnosis and Treatment of Acute Kidney Injury in Children and Adolescents

Significant FO represents the most common indication for intervention in children with
AKI.  The literature has consistently shown that
increasing magnitude of FO is associated with
adverse outcomes in a variety of populations,
including children treated with continuous renal
replacement therapy (CRRT) and other high-risk
AKI populations (e.g., ECMO, bone marrow transplant) [30–32]. Literature supports that FO >10%
should be considered a prognostic marker in ICU
patients and a marker of need for intervention.
In recent years, it has also become clear that
the development of FO may predate and delay the

835

diagnosis of AKI. At the heart of this issue is the
fact that SCr freely distributes between intracellular and extracellular spaces, resulting in inaccuracies due to fluid status. Several studies of
ICU children show that failure to account for FO
when interpreting AKI severity, as measured by
SCr rise, leads to delays in diagnosis and staging
of AKI and under-recognition of AKI incidence
and association with mortality [33–35]. Recent
studies have further cemented the concept that
FO often occurs before meeting criteria for AKI
[36–38]. The following formula is commonly utilized in the literature to correct SCr for FO:

Corrected
creatinine  serum creatinine 1   net fluid balance / TBW  

where

total body water  TBW   0.6  weight  kg 


 iagnostic Laboratory Evaluation
D
(Table 43.5)
The initial laboratory workup of a patient with
AKI seeks to identify the underlying etiology and
potentially reversible causes of AKI (e.g., hypovolemia, nephrotoxins). The initial evaluation
should minimally include electrolyte panel, SCr,
urinalysis, urine sodium, urea and creatinine, and
renal ultrasound.

Fractional Excretion of Sodium and Urea  With
renal hypoperfusion, the kidney expands the
intravascular volume by increasing sodium and
urea retention, as described above [28]. This compensatory mechanism forms the basis of the fractional excretion of sodium (FeNa) and urea
(FeUrea) calculations (Table  43.5). Both these
calculations compare urine to serum concentrations of solute, corrected for GFR.

Urinalysis and Urine Microscopy  Urinalysis
is a critical test to evaluate for hematuria, proteinuria (to rule out glomerular diseases), and/or
signs of infection. Gross hematuria and severe
proteinuria suggest glomerular disease. With
sterile pyuria, acute interstitial nephritis must
always be considered, and urinary eosinophil
testing ordered. Urine microscopy aids in diagnosing intrinsic renal disease and may reveal
muddy brown casts (acute tubular necrosis), red
blood cell casts (glomerulonephritis), pyuria or
crystals.

Renal Ultrasound  Imaging plays a small role in
diagnosing intrinsic renal disease. Ultrasound should
be considered if there is concern for obstruction or
performed with a Doppler to rule out large vessel disease (e.g., vessel thrombosis). Further information
about the chronicity of a process may be obtained by
evaluating renal size (e.g., small kidneys suggest
CKD; larger kidneys may suggest an acute process).
Biopsy  Renal biopsy is usually done to diagnose
intrinsic AKI, findings of which are not reviewed
here.


E. H. Ulrich et al.

836
Table 43.5  A proposed list of investigations for AKI

Urine testing
Urinalysis and urine culture
Dipstick testing for hematuria, proteinuria, signs of infection (leukocytes, nitrites)
Urine culture should be collected by catheterization (non-toilet-trained children) or midstream sample (toilet-trained
children)
Urine microscopy
Muddy brown or granular casts are suggestive of acute tubular necrosis
Predominant leukocytes are suggestive of acute interstitial nephritis
Urinary eosinophils for suspected acute interstitial nephritis
Red blood cell casts and/or white blood cell casts are suggestive of acute glomerulonephritis
Quantification of proteinuria
Total (including glomerular and tubular) protein using protein/creatinine ratio

Glomerular protein using albumin/creatinine ratio
Tubular protein using ß2-microglobulin
Fractional excretion of sodium

Fractional excretion of sodium  FeNa  %  

 urine Na / plasma Na 
100
 urine creatinine / plasma creatinine 

FeNa <1% (<2% for neonatesa) suggests pre-renal AKI
FeNa >2% (>2.5% for neonatesa) suggests intrinsic AKI
Fractional excretion of ureab

Fractional excretion of urea  FeUrea   %  
b

 urine urea / plasma urea 
 100
 urine creatinine / plasma creatinine 

FeUrea <35% suggests pre-renal AKI
FeUrea >50% suggests intrinsic AKI
Blood tests
Serum creatinine
Indirect measure of GFR
Limitations include:
 Delayed marker of reduction in GFR and tissue damage. Delay can be up to 72 h following renal insult
 Affected by a number of different factors, including age, sex, diet, muscle mass, and medications
 Serum creatinine varies with fluid status. Some investigators suggest correcting serum creatinine for fluid status:

  net fluid balance  
Corrected creatinine  serum creatinine 1  

  total body water  
Further investigations for severity and etiology of AKI
Complete blood count
If concern of thrombotic microangiopathy (anemia and thrombocytopenia), send markers of hemolysis, including lactate
dehydrogenase, bilirubin, haptoglobin, blood film. Further investigations if evidence of hemolysis is observed
Sodium, potassium, chloride, bicarbonate, ionized and total calcium, magnesium, phosphate
Albumin
If concern of rhabdomyolysis, send serum creatine kinase, urine myoglobin
Tests, if abnormal, suggest acute on chronic kidney disease
Iron studies, including ferritin, iron, transferrin, total iron binding capacity, and calculation of percent transferrin saturation
(TSAT)
Intact parathyroid hormone
25-[OH]-Vitamin D3
Other tests
Diagnostic imaging
Includes renal ultrasound with Doppler
Other investigations depend on etiology including VCUG, renal MAG3 scan, and DMSA renal scan
Renal biopsy
Renal biomarkers

Neonates have reduced urine concentration and sodium retention due to relative tubular immaturity
FeNa and FeUrea will be lowered with high urine flow rates. Diuretics reduce sodium reabsorption and thereby increase
FeNa. FeUrea is less affected by diuretic therapy and may be helpful to distinguish pre-renal AKI from intrinsic AKI in
patients treated with diuretics [105]
a

b


43  Diagnosis and Treatment of Acute Kidney Injury in Children and Adolescents

 anagement of Acute Kidney
M
Injury
 KI Management Prior to Renal
A
Support Therapy
Despite research advances described above, there
remain no treatments for AKI.  Many interventional trials aimed at treating established AKI
have failed. Current management strategies are
limited to preventing and treating AKI sequelae
(FO, electrolyte abnormalities, etc.). There has
been a recent paradigm shift from reactive AKI
management to risk stratification and early identification. Targeted interventions in at-risk populations have shown some promising results for
preventing AKI.  This section focuses on risk
stratification, potential interventions to prevent
or treat AKI, and management of established
AKI. In addition to close renal function monitoring, AKI management includes optimizing nutrition, avoiding hypotension and excessive FO, and
limiting nephrotoxin exposure. Often, management decisions require a team-based collaborative approach to weigh AKI risks against benefits
of individual interventions.

I nvestigational Strategies to Risk-­
Stratify AKI
Timely identification of patients at risk for developing severe AKI, before significant SCr rise or
AKI sequelae development, is critical to allow
early intervention. Recent examples of strategies developed to achieve early AKI identification include risk stratification (e.g., renal angina

index, below), AKI biomarkers (discussed previously), and a functional assessment of kidney
function (e.g., furosemide stress test, below).
Renal Angina Index  The renal angina index is a
scoring system developed and validated to predict AKI risk in ICU children by combining
known AKI risk factors and functional evidence
of injury (Fig.  43.2) [39]. Renal angina index
derivation and validation studies showed that a
score ≥8 predicted ≥ stage 2 AKI development
on ICU day 3 [40, 41]. Combining the renal
angina index score with AKI biomarker results

837

(neutrophil gelatinase-associated lipocalin) in
children at ICU admission led to almost perfect
prediction of severe AKI on ICU admission day
3. This work demonstrates that achieving early/
timely AKI and AKI risk identification likely
requires both clinical and laboratory evidence of
kidney tissue injury [42].
Furosemide Stress Test  The furosemide stress
test is a novel measure that evaluates UO response
6  h after furosemide administration, to predict
severe AKI. In adults, patients with <200 mL of
UO in the first 2 h of the test had the highest risk
for severe AKI [43], and when the test was used
in conjunction with biomarkers, severe AKI prediction improved further [44]. Emerging single-­
center studies evaluating the furosemide stress
test in pediatric populations have shown similar
promising results [45, 46].

Fluid Management
It is essential, but often challenging, to try to distinguish fluid responsive causes from fluid non-­
responsive causes of AKI (see Classification and
Etiology of AKI). Patient weight should be measured daily and fluid balance should be ­monitored
at least twice daily. After fluid resuscitation, fluid
management should be critically assessed to
avoid FO. With the exception of hypovolemia, a
safe initial approach to fluid management is to
replace insensible losses (~400  mL/m2 of body
surface area) and outputs (i.e., urine, gastrointestinal) to maintain euvolemia. Adult studies have
not shown clear benefit of colloid vs. crystalloid
(e.g., saline) solutions [47].
Diuretic use may be considered in oliguric
AKI.  KDIGO recommends not using diuretics
for the prevention or treatment of AKI, except
for the management of FO. Higher UO facilitates
nutrition administration. Loop diuretics are typically used first (e.g., furosemide; in some centers, bumetanide) due to their rapid effectiveness,
potency, and long history of use. When there
is evidence that diuresis is effective and allows
maintenance of desired fluid balance and nutrition, a second diuretic targeting different parts
of the renal tubule (e.g., distal tubule-targeted
diuretic, like thiazides or metolazone) may be


E. H. Ulrich et al.

838

Risk

Injury

Risk criteria

Score

ICU admission

1

Solid organ or stem cell transplantation

3

Mechanical ventilation or vasoactive support

5

SCr/Baseline

% Fluid
overload

Score

Decreased or no change

<5%

1

>1×–1.49×

5–10%

2

1.5×–1.99×

10–15%

4

≥2×

>15%

8
RAI index
Risk score × injury score
(1–40)
Renal angina = RAI
index ≥8

Fig. 43.2  Renal angina index [41]. Renal angina index
(RAI) is used to prognosticate the risk of developing
severe AKI (≥stage 2) 72 h later. RAI is calculated 12 h
following admission to pediatric intensive care unit (ICU).
Patient characteristics are assigned a score for “risk” (0, 1,

3, or 5). Elevation in SCr or fluid overload (%) is assigned
a score for “injury.” Baseline SCr is defined as the lowest
SCr measured 3 months prior to ICU admission; baseline
SCr was back-calculated when not available. The highest

SCr between admission to ICU and 12 h after admission
was used. When there was discrepancy between the score
for SCr/Baseline or fluid overload, the worse score was
used. The “risk” and “injury” scores are multiplied to
achieve the “RAI Index,” and RAI ≥8 defines renal
angina. In addition to being an important prognostic
marker, the finding of renal angina is strongly associated
with worse outcomes

considered. However, with reduced GFR of AKI,
most current diuretics are limited in their ability
to reach tubules to exert their effect. Diuretics
also have adverse effects, and many studies in
adults have shown no benefit of diuretic use in
AKI on time to AKI recovery or mortality. Thus,
when using diuretics in AKI, frequent reassessment of benefit (negative balance, nutrition) vs.
risk should be performed.

managed by restricting fluid. Hyperkalemia is
managed by reducing intake, correcting acidosis, and increasing elimination or intracellular
shifts using diuretics, cation exchange resins
(e.g., polystyrenes), beta-2 adrenergic receptor
agonists (e.g., albuterol, salbutamol), or insulin
with dextrose. When electrocardiogram changes
are present or with severe hyperkalemia, cardioprotection with calcium gluconate is critical

to preventing life-threatening arrhythmias. In
refractory cases or those with a high potassium
load (i.e., rhabdomyolysis), RST may be needed
(see Timing and Modality of Renal Support
Therapy). Hypocalcemia and hyperphosphatemia are treated with dietary adjustments and/or
phosphate binders. It is important to appreciate
that in severe hyperphosphatemia, together with
severe hypocalcemia, RST may be the best treat-

Electrolyte Management
Electrolyte management involves managing
acute disturbances and anticipating potential
problems. In oligoanuric AKI, the most common electrolyte disturbances are hyponatremia,
hyperkalemia, hypocalcemia, and hyperphosphatemia. Hyponatremia occurs due to sodium
and water retention, and this is most commonly


43  Diagnosis and Treatment of Acute Kidney Injury in Children and Adolescents

ment option as infusing large amounts of IV calcium in patients with severe hyperphosphatemia
may cause unwanted diffuse calcium-phosphate
crystal formation. Also, when administering
sodium bicarbonate therapy to correct acidosis or hyperkalemia, it is important to know the
patient’s calcium concentration, which will drop
with bicarbonate infusions. Overall, it is crucial
to be aware of all electrolyte abnormalities and
anticipate the effects from treating one electrolyte
abnormality on homeostasis of other electrolytes.

839

 iming and Modality of Renal
T
Support Therapy

Treating a patient with AKI with RST, even
temporarily, is a decision carrying tremendous
weight for the child, family, and healthcare team.
There remains significant equipoise about the
optimal timing of initiation of RST and the best
modality to use in AKI.
Traditional indications for RST in AKI are
well entrenched in the minds of nephrologists
and intensivists, including severe electrolyte or
Pharmacological Therapy
metabolic disturbances (especially hyperkalemia,
Historically, vasodilators were felt to be reno-­ severe metabolic acidosis, or severe hyperphosprotective. Low-dose dopamine causes vaso- phatemia with hypocalcemia), uremia (with uredilation of the renal vasculature and temporary mic pericarditis or encephalopathy, more typically
increased natriuresis and GFR.  However, many seen with severe CKD), symptomatic FO, and
randomized controlled trials (RCTs) have shown removal of a dialyzable toxin that is contributing
that dopamine does not prevent or treat AKI, to AKI.  However, as our understanding of AKI
and some studies have shown that it can cause moves beyond a binary model of “failure” or “not
tachyarrhythmias and ischemia. Other vasodila- failure” to a graded level of injury and dysfunction,
tors, including fenoldopam (dopamine receptor the optimal timing of initiation of RST becomes
agonist) and natriuretic peptides, are not cur- less clear. Several large RCTs have compared
rently recommended for AKI prevention or treat- early vs. delayed RST initiation in ICU adults and
ment. Vasopressors, however, are recommended produced conflicting results. In a single-­center trial
for patients with fluid-responsive hemodynamic (Zarbock et al.) comparing RST initiation within
compromise, to maintain renal perfusion.
8  h of stage 2 AKI (“early”) vs. within 12  h of
There is emerging, tenuous evidence that ade- stage 3 AKI (“late”), “early” RST was associated

nosine receptor antagonists (e.g., theophylline, with higher 90-day survival (difference in hazard
caffeine, aminophylline) may prevent or reduce ratio of 15%, p = 0.03) [53]. The second trial was
severity of AKI in some children. These drugs a multicenter study (Gaudry et al.) including only
have been studied mostly in neonates or children patients with stage 3 AKI; patients with “delayed”
having cardiac surgery in RCTs and observa- RST (requiring additional criteria, such as severe
tional studies, with conflicting results [48–50]. hyperkalemia, for >72  h to initiate RST) had no
The latest KDIGO guideline suggests that the- significantly different 60-day mortality vs. the earophylline may reduce AKI or AKI severity in lier RST group [54]. Most recently, a multicenter
neonates with hypoxic-ischemic encephalopathy. RCT (Barbar et al.) including patients with septic
However, there is no strong evidence supporting shock and severe AKI compared “early” (within
routine use of adenosine receptor antagonists in 12 h) vs. “late” RST initiation (>48 h) [55]. This
AKI.  Dexmedetomidine, an alpha-2 adrener- trial was halted due to findings of futility of early
gic receptor agonist, may reduce post-cardiac RST.  It thus remains unclear if earlier RST inisurgery AKI rates, but more research is needed. tiation improves patient outcomes. Notably, early
Rasburicase, a urate oxidase commonly used to RST was not found to be significantly associated
reduce urate levels in tumor lysis syndrome, may with adverse events in these studies. A larger mulalso benefit patients with severe hyperuricemia ticenter study is currently underway to compare
in specific AKI settings (e.g., rhabdomyolysis; accelerated vs. standard CRRT initiation in adults
hemolytic uremic syndrome) [51, 52].
(ClinicalTrials.gov NCT02568722).


840

E. H. Ulrich et al.

using PD to treat AKI.  Recent data suggest a
preferential shift toward use of CRRT. Today, PD
is often preferred in small infants, where vascular
access remains challenging [58]. PD is commonly used in children undergoing cardiac surgery for the additional benefit of abdominal
decompression (to reduce venous pressure and
improve renal perfusion) [5, 59]. Finally, for
patients with some primary renal diseases (e.g.,

glomerular diseases, hemolytic uremic syndrome) that do not require a critical care setting,
PD is preferred because it is well tolerated and
allows vessel preservation [57].
A number of studies have evaluated PD in
Modality of RST
infants and small children undergoing cardiac
There are several RST modalities for AKI treat- surgery. Some centers place “prophylactic” PD
ment commonly used in children, including peri- catheters in high-risk patients at the time of cartoneal dialysis (PD), intermittent hemodialysis diac surgery, and some observational studies have
(IHD), continuous renal replacement therapy shown benefit with this approach [60], including
(CRRT), and also sustained low-efficiency daily earlier negative fluid balance and improved clinidialysis (SLEDD). Many factors contribute to cal outcomes. A single-center RCT showed supeRST modality choice, including patient size, ease rior fluid removal with PD compared to standard
of access, comorbidities, and center experience dose diuretics [61]. Several observational studies
and resources. Decisions regarding modality, par- have shown benefit of earlier PD initiation folticularly pertaining to FO, will be discussed here. lowing high-risk cardiac surgery. Prophylactic
SLEDD, which uses conventional HD machines PD is generally defined as PD initiation in
to administer IHD over prolonged periods (e.g., patients without FO or reduced UO; a single6–12  h), will not be discussed. The technical center RCT did not show improved outcomes
aspects of each modality in AKI will be discussed [62], while another study showed >40% reduced
later in this section.
30- and 90-day mortality in the prophylactic arm
[63]. At this time, there is consensus that PD is
Peritoneal Dialysis  Historically, PD was the the preferred modality for infants and small chilpreferred RST modality for AKI in children due dren following cardiac surgery. However, it is not
to ease of use and availability (see Chap. 1). A known which patients should have prophylactic
major advantage of PD is the lack of need for vas- PD catheter insertion at the time of surgery. As
cular access, which can be very challenging in well, there is equipoise regarding early PD initiachildren. PD is well suited in children compared tion compared to standard use. In North America,
to adults because the peritoneal membrane sur- other pediatric populations are predominantly
face area is larger relative to patient weight, treated with CRRT, unless limited by vascular
enabling more efficient clearance. PD is less pro-­ access.
inflammatory and promotes hemodynamic stability because it provides physiologic continuous Intermittent Hemodialysis  Although there has
RST [57]. Disadvantages of PD include inconsis- been significant shift away from PD toward
tency of fluid removal and solute clearance, CRRT, the use of intermittent hemodialysis
slower solute clearance (vs. IHD or CRRT), and (IHD) in AKI has remained relatively constant
PD catheter site post-insertion leaks, especially for the treatment of life-threatening hyperkalein very edematous patients. Frequency of these mia or acute poisoning (with or without AKI).

complications is likely lower in centers primarily However, several studies have shown that with
In children, there are no such trials. However,
many observational studies have identified that
higher FO at CRRT initiation is associated with
poorer outcomes. In one multicenter study, the
highest mortality (>65%) was seen with patients
with ≥20% FO at CRRT initiation; this mortality
is 8.5 times higher vs. patients with <20% FO
[56]. Consensus opinion is that RST should be
considered in patients with 10–20% FO. Despite
the known risk of FO in ICU children, there
remains equipoise on whether earlier intervention in children with severe FO improves
outcomes.