Acute Renal Failure Complications
- Author: Richard H Sinert, DO; Chief Editor: Robert E O'Connor, MD, MPH more...
Lack of a systematic definition of acute renal failure (ARF) previously led to significant confusion clinically and in the medical literature. The Acute Dialysis Quality Initiative (ADQI) group published the RIFLE classification of ARF, based on changes from the patient's baseline either in serum creatinine level, glomerular filtration rate (GFR), or urine output (UO).
The RIFLE classification of ARF is as follows :
Risk (R) - Increase in serum creatinine level X 1.5 or decrease in GFR by 25%, or UO < 0.5 mL/kg/h for 6 hours
Injury (I) - Increase in serum creatinine level X 2.0 or decrease in GFR by 50%, or UO < 0.5 mL/kg/h for 12 hours
Failure (F) - Increase in serum creatinine level X 3.0, decrease in GFR by 75%, or serum creatinine level ≥ 4 mg/dL with acute increase of >0.5 mg/dL; UO < 0.3 mL/kg/h for 24 hours, or anuria for 12 hours
Loss (L) - Persistent ARF, complete loss of kidney function >4 weeks
End-stage kidney disease (E) - Loss of kidney function >3 months
Since baseline serum creatinine level and GFRs may not be readily available, the consensus committee recommended the use of the Modification of Diet in Renal Disease (MDRD) equation to estimate the patient’s GFR/1.73 mm based on serum creatinine level, age, gender, and race. The proportional decrease in GFR should be calculated from 75 mL/min per 1.73 mm2, the agreed upon lower limit of normal.
ARF is a common entity in the emergency department (ED). Emergency physicians play a critical role in recognizing early ARF, preventing iatrogenic injury, and reversing the course of ARF.
Imaging studies in acute renal failure (ARF) are most important in the emergent workup of suspected postrenal azotemia.
The distinction between community- and hospital-acquired acute renal failure (ARF) is important for the differential diagnoses, treatment, and eventual outcome of patients with ARF.
Incidence in the United States
The annual incidence of community-acquired acute renal failure (ARF) is approximately 100 cases per 1 million population, and it is diagnosed in only 1% of hospital admissions at presentation.
Using the RIFLE classification, hospital-acquired ARF of the Risk, Injury, and Failure categories has been found in 9%, 5%, and 4% of hospital admissions, respectively, and in approximately 17%, 12%, and 7% of critical care admissions.[4, 5]
This high incidence of hospital-acquired ARF is multifactorial; it is related to an aging population with increased risks of ARF, the high prevalence of nephrotoxic exposures possible in a hospital setting, and increasing severity of illness.
Because most cases of community-acquired acute renal failure (ARF) are secondary to volume depletion, as many as 90% of cases are estimated to have a potentially reversible cause. Hospital-acquired ARF often occurs in an intensive-care-unit (ICU) setting and is commonly part of multiorgan failure.
This dichotomy in the etiology of ARF explains the increased mortality rate, dialysis requirements, and rates of progression to end-stage renal failure seen in hospital-acquired ARF compared with community-acquired ARF.
Mortality rates for ARF have changed little since the advent of dialysis at 50%. This curious statistic simply reflects the changing demographics of ARF from community- to hospital-acquired settings.
Currently, the mortality rate for hospital-acquired ARF is reported to be as high as 70% and is directly correlated to the severity of the patient's other disease processes. The mortality rate among patients presenting to the ED with prerenal ARF may be as low as 7%.
With the advent of dialysis, the most common causes of death associated with ARF are sepsis, cardiac failure, and pulmonary failure. Interestingly, patients who are older than age 80 years with ARF have mortality rates similar to those of younger adult patients.
Pediatric patients with ARF represent a different set of etiologies and have mortality rates averaging 25%.
ARF is not a benign disease. One study noted a 31% mortality rate in patients with ARF not requiring dialysis, compared with a mortality rate of only 8% in matched patients without ARF. Even after adjusting for comorbidity, the odds ratio for dying of ARF was 4.9 compared with patients without ARF.
There seems to be a stepwise relationship between the RIFLE category of renal injury and mortality. Compared with non-acute kidney injury (AKI), the relative risk (RR) of death for Risk is 2.40; for Injury, 4.15; and for Failure, 6.4.
Mortality rates are generally lower for nonoliguric ARF (>400 mL/d) than for oliguric (< 400 mL/d) ARF, reflecting the fact that nonoliguric ARF is usually caused by drug-induced nephrotoxicity and interstitial nephritis, which have few other systemic complications.
The following conditions should be considered in the differential diagnosis of acute renal failure (ARF):
Anemia, Sickle Cell
Congestive Heart Failure and Pulmonary Edema
Hemolytic Uremic Syndrome
Pediatrics, Diabetic Ketoacidosis
Pediatrics, Inborn Errors of Metabolism
Pediatrics, Sickle Cell Disease
Pediatrics, Urinary Tract Infections and Pyelonephritis
Renal Failure, Chronic and Dialysis Complications
Urinary Tract Infection in Females
Urinary Tract Infection in Males
Normal-range blood urea nitrogen (BUN) and creatinine levels do not reliably rule out the diagnosis of ARF. Patients with low muscle mass and/or vegetarians may have significant decreases in GFR and still remain in normal ranges for BUN and creatinine. Comparison with baseline values and trends are more important than are absolute numerical values.
Microscopic examination of urine is essential in establishing differential diagnosis for acute renal failure (ARF). Findings may include the following:
Normal urinary sediment without hemoglobin, protein, cells, or casts generally consistent with prerenal and postrenal failure, HUS/thrombotic thrombocytopenic purpura (TTP), preglomerular vasculitis, or atheroembolism
Granular casts - Acute tubular necrosis (ATN), glomerulonephritis, interstitial nephritis
Red blood cell (RBC) casts - Glomerulonephritis, malignant hypertension
White blood cell (WBC) casts - Pyelonephritis
Eosinophiluria - Acute allergic interstitial nephritis, atheroembolism
Crystalluria - Acyclovir, sulfonamides, methotrexate, ethylene glycol toxicity, radiocontrast agents (Mild crystalluria can be a normal finding)
The urea concentration correlates poorly with the GFR. Because urea is highly permeable to renal tubules, urea clearance varies with urine flow rate.
Urea is filtered freely, but reabsorption along the tubule is a function of urine flow rate. During antidiuresis with urine flow rates of less than 30 mL/h, urea clearance is as low as an estimated 30% of GFR. Under conditions of diuresis, with urine outputs greater than 100 mL/h, urea clearance can increase to 70-100% of GFR.
This information can be used clinically to help differentiate prerenal failure from other etiologies of ARF. In prerenal conditions, low urine flow rates favor BUN reabsorption out of proportion to decreases in GFR, resulting in a disproportionate rise of BUN relative to creatinine, creating a serum BUN-creatinine ratio of more than 20 in prerenal failure.
BUN concentration is dependent on nitrogen balance and renal function.
BUN concentration can rise significantly with no decrement in GFR by increases in urea production with steroids, trauma, or GI bleeding.
Tetracycline increases BUN by decreasing tissue anabolic rates.
Basal BUN concentration can be depressed severely by malnutrition or advanced liver disease.
Always first estimate basal BUN concentration when attempting to correlate changes in BUN with GFR. For example, in a patient with cirrhosis and a BUN of 12 mg/dL, a GFR in the normal range may be assumed. Only with the knowledge of a baseline BUN of 4 mg/dL does the real decrease in GFR become apparent.
Serum creatinine measurement provides the ED physician with an accurate and consistent estimation of GFR. Correct interpretation of serum creatinine measurement extends beyond just knowing normal values for the specific laboratory.
The serum creatinine level varies by method of measurement, either Jaffe or iminohydrolase. The upper limit of the normal creatinine level can be 1.6-1.9 mg/dL or 1.2-1.4 mg/dL, respectively. This becomes important when patients present with changes in creatinine measured in different laboratories.
Differing methods report markedly different results when interfacing with certain chemicals.
The Jaffe method of measuring creatinine reports falsely elevated serum creatinine in the presence of the following noncreatinine chromogens: glucose, fructose, uric acid, acetone, acetoacetate, protein, ascorbic acid, pyruvate, cephalosporin antibiotics. High levels of bilirubin cause reports of falsely low creatinine by the Jaffe method.
Extremely high glucose levels and the antifungal agent flucytosine interfere with the iminohydrolase method.
The serum creatinine level, a reflection of creatinine clearance, is a function of creatinine production and excretion rates.
Creatinine production is determined by muscle mass. The serum creatinine level must always be interpreted with respect to patient's weight, age, and sex.
For example, GFR decreases by 1% per year after age 40 years, yet serum creatinine level generally remains stable. Balance is achieved via a decrease in muscle mass with age, which matches the fall in GF.
Men generally have a higher muscle mass per kilogram of body weight and thus a higher serum creatinine level than women.
The GFR can be estimated by the following formulas:
Cockcroft-Gault equation: GFR mL/min = (140 - Age y)(Weight kg)(0.85 if female)/(72 X Serum Creatinine mol/L
Modification of Diet in Renal Disease (MDRD) equation: GFR, in mL/min per 1.73 mm 2 = 186.3 X ((Serum Creatinine) exp[-1.154]) X (Age exp[-0.203]) X (0.742 if female) X (1.21 if African American)
The ADQI consensus committee on acute renal failure (ARF) favors the (MDRD) equation to estimate GFR.
An important consideration and limitation is that significant decrements in GFR can occur while creatinine levels remain in the normal range.
Changes in serum creatinine level reflect changes in GFR. Rate of change in serum creatinine level is an important variable in estimating GFR. Stable changes in serum creatinine level correlate with changes in GFR by the following relationships:
If creatinine 1 mg/dL is baseline for a given patient with normal GFR
Creatinine 2 mg/dL - 50% reduction in GFR
Creatinine 4 mg/dL - 70–85% reduction in GFR
Creatinine 8 mg/dL - 90–95% reduction in GFR
As suggested by these data, knowledge of a patient's baseline creatinine level becomes very important. Small changes with low baseline levels of creatinine may be much more important clinically than large changes with high basal creatinine.
Certain diseases and medications can interfere with the correlation of serum creatinine with GFR. Acute glomerulonephritis causes increased tubular secretion of creatinine, falsely depressing the rise in serum creatinine level when ARF occurs in acute glomerulonephritis. Trimethoprim and cimetidine cause decreased creatinine secretion and a falsely elevated creatinine with no change in GFR.
Cystatin C is emerging as a superior biomarker for early kidney injury. In a study of 198 patients presenting to an emergency department, serum cystatin C >1.44 mg/L alone or along with serum creatinine and estimated glomerular filtration rate has been found to be a strong predictor for the risk of acute kidney injury.
It is generated at a constant rate by all nucleated cells and is not secreted by the tubules or eliminated by other routes than renal excretion.
It does not appear to be affected by body habitus, nutritional state, or comorbid illness.
One of its principal advantages is that it identifies kidney injury while creatinine levels remain in the normal range.
Complete Blood Count
The following points should be kept in mind concerning complete blood count (CBC) results:
Leukocytosis - Common in acute renal failure (ARF)
Leukopenia and thrombocytopenia - Suggest systemic lupus erythematosus (SLE) or TTP
Anemia and rouleaux formation - Suggest multiple myeloma
Microangiopathic anemia - Suggests disseminated intravascular coagulation (DIC), TTP, or atheroemboli
Eosinophilia - Suggests allergic interstitial nephritis, polyarteritis nodosa, or atheroemboli
Coagulation disturbances - Indicate liver disease, DIC, TTP, or hepatorenal syndrome
Creatine phosphokinase (CPK) elevations are seen in rhabdomyolysis and myocardial infarction.
Elevations in liver transaminase levels are seen in rapidly progressive liver failure and hepatorenal syndrome.
Hypocalcemia (moderate) is common in acute renal failure (ARF); marked hypocalcemia is more typical of chronic renal failure.
Hyperkalemia is a common and important complication of ARF.
Urine Chemical Indices
Differentiation of prerenal azotemia from ATN takes on a special importance in early management of these patients. Aggressive fluid resuscitation is appropriate in prerenal acute renal failure (ARF). However, overly aggressive volume resuscitation in a patient with ATN who is unable to excrete the extra fluid can result in volume overload and respiratory embarrassment.
To help with the differentiation of prerenal azotemia, analysis of urine may provide important clues. Diuretics interfere with some of these indices, so collect urine prior to any considered administration of diuretics.
Urine indices that suggest prerenal ARF include the following:
Urine specific gravity >1.018
Urine osmolality (mOsm/kg water) >500
Urine sodium (mEq/L) < 15-20
Plasma BUN-creatinine ratio >20
Urine-plasma creatinine ratio >40
Urine indices that suggest ATN include the following:
Urine specific gravity < 1.012
Urine osmolality (mOsm/kg water) < 500
Urine sodium (mEq/L) >40
Plasma BUN/creatinine ratio < 10-15
Urine-plasma creatinine ratio < 20
Calculation of Fractional Excretion of Sodium
The calculation of fractional excretion of sodium (FeNa) is as follows:
FeNa = (urine Na/plasma Na)/(urine creatinine/plasma creatinine)
If FeNa is less than 1%, this suggests prerenal acute renal failure (ARF).
If FeNa is greater than 1%, this suggests ATN.
The advantages of FeNa compared with other indices include the following:
Physiologic measure of sodium reabsorption
Measured creatinine and sodium clearances, accounting for filtration and reabsorption of sodium
FeNa increased before oliguric phase established and predictive of incipient ARF
Exceptions (intrinsic renal failure with FeNa < 1%) include the following:
Radiologic contrast–induced ATN
Myoglobinuric and hemoglobinuric ARF
Renal allograft rejection
Drug-related alterations in renal hemodynamics (eg, captopril, nonsteroidal anti-inflammatory drugs [NSAIDs])
Renal ultrasonography is the test of choice for urologic imaging in the setting of acute renal failure. It has excellent sensitivity and specificity for detecting hydronephrosis due to obstruction, and it can also give valuable information other than ruling obstruction in or out.
Bipolar renal length is easy to assess, and kidneys smaller than 9 cm suggest chronic renal failure.
Renal parenchyma should be isoechogenic or hypoechogenic when compared with that of the liver and spleen; hyperechogenicity indicates diffuse parenchymal disease.
Color Doppler allows assessment of renal perfusion and can allow diagnosis of large-vessel etiologies of ARF.
In critically ill patients, bedside ultrasonography warrants special consideration, because it can quickly diagnose treatable etiologies of the patient’s condition and give guidance for fluid resuscitation.
Chest Radiography and Electrocardiography
Obtain chest radiographs on a routine basis to look for evidence of volume overload.
Findings of lung infiltration can lead to pulmonary/renal syndromes, such as Wegener granulomatosis and Goodpasture syndrome, or evidence of pulmonary emboli from endocarditis or atheroembolic disease.
Obtain routine electrocardiograms to look for manifestations of hyperkalemia and arrhythmias, ischemia, and infarction.
Renal biopsy is often helpful in finding specific cause of intrinsic renal failure; however, it is not an ED procedure. This is reserved for evaluation of acute renal failure (ARF) when the cause cannot be determined.
Renal biopsy is especially important when glomerular causes of ARF are suspected.
It is often helpful in finding a specific cause of renal failure.
Stabilize acute life-threatening conditions and initiate supportive therapy. Watch for electrocardiographic evidence of hyperkalemia.
Emergency Department Care
Treatment of acute renal failure (ARF) ideally should begin before the diagnosis of ARF is firmly established. A high index of suspicion often is necessary to diagnose early ARF. Significant decreases in GFR frequently occur before indirect measures of GFR reveal a problem. All seriously ill medical patients (eg, elderly patients, diabetic patients, hypovolemic patients) should have ARF included early in their differential diagnosis.
Physicians can play a pivotal role in reversing many of the underlying causes and preventing further iatrogenic renal injury if ARF is recognized early. After providing an adequate airway and ventilation, focus on fluid management of the patient with ARF.
Patients with ARF represent challenging fluid management problems.
Hypovolemia potentiates and exacerbates all forms of ARF.
Reversal of hypovolemia by rapid fluid infusion often is sufficient to treat many forms of ARF. However, rapid fluid infusion can result in life-threatening fluid overload in patients with ARF.
Accurate determination of a patient's volume status is essential and may require invasive hemodynamic monitoring if physical examination and laboratory results do not lead to a definite conclusion.
Bedside ultrasonographic evaluation, including IVC measurement, may give additional useful information.
Urinary catheter placement
Urinary obstruction often is an easily reversible cause of ARF.
Placement of a urinary catheter early in the workup of a patient with ARF not only allows diagnosis and treatment of urethral and bladder outlet urinary obstruction but also allows for accurate measurement of urine output.
If available, bedside ultrasonography can quickly identify a large and distended bladder.
Routine use of urinary catheters should be tempered by consideration of the inherent risks of catheter-associated infections.
Renal replacement therapy
The principal methods of renal replacement therapy (RRT) are intermittent hemodialysis (IHD), continuous venovenous hemodiafiltration (CVVHD), and peritoneal dialysis (PD). Each has advantages and limitations.
IHD is widely available, has only moderate technical difficulty, and is the most efficient way of removing a volume or solute from the vascular compartment quickly. Unfortunately, dialysis-associated hypotension may adversely affect remaining renal function, particularly in patients who are hemodynamically unstable. This is one reason CVVHD is widely recommended in this setting.
Continuous RRT techniques are more expensive, associated with increased bleeding risk, and not universally available; however, in addition to avoiding hypotension, they are believed to achieve better control of uremia and clearance of solute from the extravascular compartment. CVVHD may also preserve cerebral perfusion pressure more effectively. Although several studies have sought to directly compare CVVHD with IHD, no study has shown a convincing advantage of one therapy over the other.
Peritoneal dialysis is inexpensive, widely available, and does not result in hypotension. However, it is not capable of removing large volumes of fluid or solute. Its use may be most common in children and in developing countries.
Indications for and timing of initiation of RRT are also important and somewhat controversial subjects.
Widely accepted indications for initiation of RRT include the following:
Hyperkalemia (K + >6.5 or rising)
Symptomatic uremia (pericarditis, encephalopathy, bleeding dyscrasia, nausea, vomiting, pruritus)
Uremia (BUN >100)
Severe dysnatremia (< 115 or >165), and dysthermia may also be appropriate indications for RRT.
Significant intoxications with a dialyzable agent (eg, methanol, ethylene glycol, theophylline, aspirin, lithium) may be the strongest single indication for emergent dialysis, because other effective therapeutic interventions are available for most of other complications of ARF. Volume overload can be treated with nitrates and phlebotomy; hyperkalemia can be treated with calcium, insulin, glucose, bicarbonate, binding resins, and beta-agonists.
Note that, in light of little evidence of effectiveness, the possible adverse effects of the ion-exchange resin, sodium polystyrene sulfonate, in sorbitol should be considered. There is emerging concern about use of this time-honored, but scientifically unproven, management of hyperkalemia.
The timing of initiation of RRT in the absence of the aforementioned indications is controversial, although the consensus that RRT itself contributes to the resolution of ARF may be growing.
Intensity of RRT is another area of active controversy and research; some studies suggest that more is better. In a study of CVVH intensity in which patients with ARF were randomly given standard or supernormal levels of ultrafiltration, the patients with more intense RRT had significantly lower mortality rates. A second randomized trial compared daily IHD with traditional, every-other-day IHD in patients with ARF and found that the mortality rate (28% vs 46%) and speed of renal recovery (9 d vs 16 d) were significantly improved. However, before these studies, no significant evidence indicated that increased dialysis dosage improved outcomes.
Most cases of acute renal failure (ARF) in inpatients are secondary to iatrogenic causes. Be especially careful in prescribing potential nephrotoxins (eg, radiocontrast agents, aminoglycosides, NSAIDs) to patients predisposed to ARF (eg, dehydration, CHF, diabetes mellitus, chronic renal failure, elderly patients).
Diuretics and vasodilators are used commonly to treat acute renal failure (ARF). Unfortunately, in large, randomized studies, these agents have failed to prove effective.
Atrial natriuretic factor has been tested in a randomized, double-blind study in ARF but also failed to improve the course of the disease.
Calcium channel blockers have been shown in animal models to be protective in ARF if given before renal insult. Their only benefit in humans is preventing ARF in renal transplant patients receiving cyclosporine.
Infusion of mannitol is reported to be protective of myoglobinuric ARF if given within 6 hours of rhabdomyolysis. In addition, mannitol infusion has been shown to decrease the rate of ARF if given before cardiothoracic surgery and radiocontrast agents. However, no controlled studies have shown any benefit to mannitol infusion in patients with established ARF. In fact, mannitol given in high doses has been associated with ARF. Significant risks of prescribing large doses of mannitol to patients with ARF include fluid overload and hyperkalemia.
Low-dose dopamine is a potent vasodilator, increasing renal blood flow in ARF, and acts as a dopamine agonist. Unfortunately, most clinical studies fail to show that it improves recovery or mortality rates. In the majority of ARF studies, dopamine was associated only with an increase in urine output. Current recommendations for dopamine favor its use in patients with ARF and concomitant hypodynamic heart failure. Balance benefits of diuretic action with proarrhythmic side effects.
Fenoldopam is a potent dopamine A-1 receptor agonist that increases blood flow to the renal cortex and outer medulla and evidence to date suggests that it reduces mortality and provides renal protection in critically ill patients with or at risk of renal failure. Because it is titratable and it reliably controls severe hypertension, fenoldopam may be ideal for treating hypertensive emergencies where the affected end organ is the kidneys.
Patients with nonoliguric (rather than oliguric) ARF have better mortality and renal recovery rates, prompting many to recommend diuretics in oliguric ARF. Unfortunately, randomized double-blind controlled trials fail to show benefit. Studies conclude that diuretics are useful only in management of fluid-overloaded patients and venodilators and dialysis are more effective interventions for this indication.
Consider consultation with a nephrologist or critical care specialist in patients with newly diagnosed ARF.
Inpatient Care and Transfer Considerations
Patients with acute renal failure (ARF) should generally be admitted to an inpatient setting; intensive care will be appropriate for many of them.
Transfer patients with significant ARF to a facility with capability for hemodialysis on a 24-hour basis.
Deterrence, Prevention, and Patient Education
It is important to recognize renal failure early as well as risk factors for renal injury and to avoid interventions that may iatrogenically induce renal failure.
Stress to patients that progressive renal failure is a silent disease. Symptoms of uremia occur only with advanced, generally irreversible renal failure. The only way for patients to reliably follow the course of their disease is through regular checkups with their physician.
[Guideline] Bellomo R, Ronco C, Kellum JA, et al. Acute renal failure - definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care. 2004 Aug. 8(4):R204-12. [Medline]. [Full Text].
Nee PA, Bailey DJ, Todd V, Lewington AJ, Wootten AE, Sim KJ. Critical care in the emergency department: acute kidney injury. Emerg Med J. 2015 May 12. [Medline].
Uchino S, Bellomo R, Goldsmith D, Bates S, Ronco C. An assessment of the RIFLE criteria for acute renal failure in hospitalized patients. Crit Care Med. 2006 Jul. 34(7):1913-7. [Medline].
Ostermann M, Chang RW. Acute kidney injury in the intensive care unit according to RIFLE. Crit Care Med. 2007 Aug. 35(8):1837-43; quiz 1852. [Medline].
Bagshaw SM, George C, Dinu I, Bellomo R. A multi-centre evaluation of the RIFLE criteria for early acute kidney injury in critically ill patients. Nephrol Dial Transplant. 2008 Apr. 23(4):1203-10. [Medline].
Ympa YP, Sakr Y, Reinhart K, et al. Has mortality from acute renal failure decreased? A systematic review of the literature. Am J Med. 2005 Aug. 118(8):827-32. [Medline].
Ricci Z, Cruz D, Ronco C. The RIFLE criteria and mortality in acute kidney injury: A systematic review. Kidney Int. 2008 Mar. 73(5):538-46. [Medline].
Bongiovanni C, Magrini L, Salerno G, Gori CS, Cardelli P, Hur M, et al. Serum Cystatin C for the Diagnosis of Acute Kidney Injury in Patients Admitted in the Emergency Department. Dis Markers. 2015. 2015:416059. [Medline].
[Guideline] Papnicolaou N, Francis IR, Casalino DD, Arellano RS, Baumgarten DA, Curry NS, et al. ACR Appropriateness Criteria renal failure. [online publication]. Reston (VA): American College of Radiology (ACR). 2008. [Full Text].
Santhanakrishnan A, Nestle TT, Moore BL, Yoganathan AP, Paden ML. Development of an accurate fluid management system for a pediatric continuous renal replacement therapy device. ASAIO J. 2013 May-Jun. 59(3):294-301. [Medline]. [Full Text].
Nikibakhsh AA, Mahmoodzadeh H, Vali M, Enashaei A, Asem A, Yekta Z. Outcome of immediate use of the permanent peritoneal dialysis catheter in children with acute and chronic renal failure. Iran J Pediatr. 2013 Apr. 23(2):171-6. [Medline]. [Full Text].
Sterns RH, Rojas M, Bernstein P, Chennupati S. Ion-exchange resins for the treatment of hyperkalemia: are they safe and effective?. J Am Soc Nephrol. 2010 May. 21(5):733-5. [Medline].