What are the effects of interventions to prevent acute renal failure in persons at high risk?
Contrast Media (Low Osmolality More Effective Than High Osmolality Contrast Media)
One systematic review found that low osmolality contrast media reduced nephrotoxicity in persons with underlying renal failure needing contrast investigation compared with high osmolality contrast media. This benefit was greatest in persons with underlying renal impairment.
LIKELY TO BE BENEFICIAL
Sodium Chloride-Based Fluids
One randomized controlled trail (RCT) of persons having elective cardiac catheterization found that intravenous sodium chloride hydration reduced acute renal failure compared with unrestricted oral fluids 48 hours after catheterization. One RCT found that hydration with sodium chloride 0.9% infusion reduced contrast nephropathy compared with sodium chloride 0.45%. This effect was greater in women, persons with diabetes, and persons who received more than 250 mL of contrast. One RCT found inconclusive evidence on the effects of inpatient hydration regimens compared with outpatient hydration regimens.
Amphotericin B (Lipid Formulations May Cause Less Nephrotoxicity Than Standard Formulations)
We found no RCTs. Lipid formulations of amphotericin B seem to cause less nephrotoxicity compared with standard formulations, but direct comparisons of long-term safety are lacking. (Categorization based on consensus.)
Systematic reviews have found that N-acetylcysteine plus hydration reduced contrast nephropathy compared with hydration alone in persons at high risk of contrast nephropathy who received radiocontrast. One further RCT found that double-dose N-acetylcysteine may be more effective in preventing contrast nephropathy than single-dose N-acetylcysteine, especially in persons who received high volumes of nonionic, low osmolality contrast agent. One recent large RCT found that intravenous and oral N-acetylcysteine may prevent contrast nephropathy with a dose-dependent effect in persons treated with primary angioplasty. One small RCT evaluated N-acetylcysteine in persons having elective aortic aneurysm surgical repair and found no significant differences in the incidence of acute renal failure or mortality compared with placebo. Similarly, another RCT found no reduction in the incidence of postoperative renal dysfunction with the use of N-acetylcysteine in high-risk persons having cardiac surgery.
Contrast Media (Iso-osmolar May Be More Effective Than Low Osmolality Contrast Media)
One RCT identified by a systematic review found that nonionic iso-osmolar contrast media reduced contrast media-induced nephropathy compared with low osmolar contrast media in persons with diabetes.
Aminoglycosides (Single Dose as Effective as Multiple Doses for Treating Infection, but with Reduced Nephrotoxicity)
One systematic review and one additional RCT compared single and multiple doses of aminoglycosides and found different results for nephrotoxicity. A systematic review of persons with fever and neutropenia receiving antibiotic treatment including aminoglycosides found no significant difference in cure rates or nephrotoxicity between once daily compared with three times daily administration of the aminoglycoside. However, the RCT found that single doses of aminoglycosides reduced nephrotoxicity compared with multiple doses in persons with fever and receiving antibiotic treatment including an aminoglycoside.
Sodium Bicarbonate-Based Fluids (Limited Evidence Better Than Sodium Chloride for the Prevention of Contrast Nephropathy)
One single-center RCT in persons with stable serum creatinine found limited evidence that the administration of sodium bicarbonate solution before and after contrast exposure was more effective in preventing contrast-induced nephropathy compared with the administration of sodium chloride.
UNLIKELY TO BE BENEFICIAL
Seven RCTs compared the role of fenoldopam with placebo in preventing acute renal failure. Four small, limited RCTs suggested that fenoldopam may improve renal perfusion and creatinine clearance compared with conventional care. One large RCT, which focused on clinical outcomes in persons having invasive cardiovascular procedures, found no evidence that it is more effective than conventional care for preventing acute renal failure. One recent RCT found that low-dose fenoldopam did not reduce the need for dialysis or improve survival compared with placebo in persons with sepsis. Similarly, another RCT found that fenoldopam did not reduce the incidence of death or dialysis therapy in critically ill persons with early acute tubular necrosis. Two RCTs found no difference between fenoldopam and dopamine in incidence of acute renal failure. Fenoldopam may cause hypotension.
Small RCTs in persons with traumatic rhabdomyolysis, or in persons who had undergone coronary artery bypass, vascular, or biliary tract surgery, found that mannitol plus hydration did not reduce acute renal failure compared with hydration alone. One RCT found that mannitol increased the risk of acute renal failure compared with sodium chloride 0.45% infusion, but the difference was not significant.
Renal Replacement Therapy (Prophylactic Hemofiltration/Dialysis)
One single-center RCT in persons with baseline chronic renal dysfunction (very high risk of developing contrast nephropathy) having coronary interventions with contrast found better in-hospital and one-year survival with low-dose hemofiltration compared with saline hydration alone. However, several limitations to this study were identified. Hemofiltration is invasive, expensive, and can lead to important clinical complications, such as hypotension.
Theophylline or Aminophylline
One systematic review found that theophylline or aminophylline attenuated the degree of increase in serum creatinine after radiocontrast administration compared with control. However, no patient-centered outcomes were affected. In many RCTs included in the review, the hydration status of persons was unclear. One included RCT found that, in persons with adequate intravenous hydration who required radiocontrast investigations, theophylline did not prevent radio-contrast-induced nephropathy compared with placebo. One RCT found no significant reduction in renal impairment after elective coronary artery bypass surgery with theophylline compared with hydration alone, but the study may have been underpowered to detect a clinically important difference.
One large RCT found no significant difference between natriuretic peptides and placebo in the prevention of acute renal failure induced by contrast media. Subgroup analysis in another large RCT in persons with early acute renal failure found that atrial natriuretic peptide reduced dialysis-free survival in nonoliguric persons compared with placebo. One small RCT in persons with postcardiac surgical heart failure requiring inotropic and vasoactive support found limited evidence that prolonged infusion of human recombinant atrial natriuretic peptide reduced the proportion of persons requiring dialysis (before or at day 21) and the composite outcome of dialysis or death (before or at day 21) compared with placebo.
LIKELY TO BE INEFFECTIVE OR HARMFUL
Calcium Channel Blockers (for Early Allograft Dysfunction)
One RCT found no significant difference between isradipine and placebo in preventing early allograft dysfunction in persons receiving cadaveric or living renal transplant. One systematic review of persons with cadaveric renal transplant found limited evidence from heterogeneous RCTs that calcium channel blockers given in the perioperative period reduced post-transplant acute tubular necrosis, although it found no significant effect on graft loss, need for hemodialysis, or mortality. We found no RCTs assessing the effects of calcium channel blockers in preventing other forms of acute renal failure. Calcium channel blockers are associated with hypotension and bradycardia.
Two systematic reviews and one subsequent RCT found no significant difference between dopamine and placebo in the development of acute renal failure, the need for dialysis, or death. Dopamine is associated with serious adverse effects, such as extravasation necrosis, gangrene, and conduction abnormalities.
One systematic review and one subsequent RCT found that adding loop diuretics to fluids was not effective and may actually increase the possibility of acute renal failure compared with fluids alone in persons at high risk of acute renal failure.
What are the effects of treatments for critically ill persons with acute renal failure?
LIKELY TO BE BENEFICIAL
Renal Replacement Therapy (Reduced Mortality Compared with Low Dose)
One RCT found that high-dose continuous renal replacement therapy (hemofiltration) reduced mortality compared with low-dose continuous therapy. A small prospective study found that intensive (daily) intermittent hemodialysis reduced mortality in persons with acute renal failure compared with conventional alternate-day hemodialysis. A subsequent small three-arm RCT found no significant difference in survival at 28 days between early, low-dose hemofiltration; early, high-dose hemofiltration; and late, low-dose hemofiltration.
Loop Diuretics (unclear if Continuous infusion More Effective Than Bolus Injection)
We found no RCTs comparing continuous infusion versus bolus injection of loop diuretics in critically ill persons with acute renal failure.
Renal Replacement Therapy (Unclear Whether Continuous or Intermittent Renal Replacement Therapy More Effective)
One systematic review found no significant difference between continuous and intermittent renal replacement therapy in mortality, renal death, or dialysis dependence in critically ill adults with acute renal failure.
Albumin Supplementation Plus Loop Diuretics (Intravenous)
We found no RCTs on the effects of intravenous albumin supplementation plus loop diuretics in critically ill persons with acute renal failure.
Dialysis Membranes (Unclear if Synthetic or Cellulose-Based Membranes More Effective)
Two systematic reviews provided inconclusive evidence of the effects of synthetic membranes on mortality in critically ill persons with acute renal failure compared with cellulose-based membranes.
UNLIKELY TO BE BENEFICIAL
One large and two small RCTs in critically ill persons with oliguric renal failure found no significant difference between loop diuretics and placebo in renal recovery, the number of days spent on dialysis, or mortality. Loop diuretics have been associated with ototoxicity and may lead to volume depletion.
LIKELY TO BE INEFFECTIVE OR HARMFUL
One systematic review found no significant difference in mortality or need for dialysis between dopamine and placebo. One additional RCT found that low-dose dopamine did not reduce renal dysfunction compared with placebo. Dopamine has been associated with important adverse effects, including extravasation necrosis, gangrene, and conduction abnormalities.
RCTs found no significant difference between atrial natriuretic peptide, ularitide (urodilatin), and placebo in dialysis-free survival in oliguric and nonoliguric persons with acute renal failure. One of the RCTs found that atrial natriuretic peptide may reduce survival in nonoliguric persons.
Acute renal failure is characterized by abrupt and sustained decline in glomerular filtration rate,1 which leads to accumulation of urea and other chemicals in the blood. Most studies define it biochemically as a serum creatinine of 2 to 3 mg per dL (180 to 270 μmol per L), an elevation of more than 0.5 mg per dL (40 μmol per L) over a baseline creatinine below 2 mg per dL, or a twofold increase of baseline creatinine. A recent international, interdisciplinary, consensus panel has classified acute renal failure according to a change from baseline serum creatinine or urine output. The three-level classification begins with risk, defined by a 50 percent increase in serum creatinine or a urine output of less than 0.5 mL per kg per hour for at least six hours; and it concludes with failure, defined by a threefold increase in serum creatinine or a urine output of less than 0.3 mL per kg per hour for 24 hours.2 Acute renal failure is usually additionally classified according to the location of the predominant primary pathology (prerenal, intrarenal, and postrenal failure). Critically ill persons are clinically unstable and at imminent risk of death, which usually implies that they need to be in, or have been admitted to, the intensive care unit (ICU).
Incidence and Prevalence
GENERAL RISK FACTORS
Risk factors for acute renal failure that are consistent across multiple causes include older age; hypovolemia; hypotension; sepsis; pre-existing renal, hepatic, or cardiac dysfunction; diabetes mellitus; and exposure to nephrotoxins (e.g., aminoglycosides, amphotericin, immunosuppressive agents, nonsteroidal anti-inflammatory drugs, angiotensin-converting enzyme inhibitors, intravenous contrast media).4–8
RISK FACTORS/ETIOLOGY IN CRITICALLY ILL PERSONS
Isolated episodes of acute renal failure rarely occur in critically ill persons, but are usually part of multiple organ dysfunction syndromes. Acute renal failure requiring dialysis rarely occurs in isolation (less than 5 percent of persons). The kidneys are often the first organs to fail.9 In the perioperative setting, acute renal failure risk factors include prolonged aortic clamping, emergency rather than elective surgery, and use of higher volumes (greater than 100 mL) of intravenous contrast media. One study (3,695 persons) using multiple logistic regression identified the following independent risk factors: baseline creatinine clearance below 47 mL per minute (0.78 mL per second; odds ratio [OR] = 1.20; 95% confidence interval [CI], 1.12 to 1.30), diabetes (OR = 5.5; 95% CI, 1.4 to 21.0), and a marginal effect for doses of contrast media above 100 mL (OR = 1.01; 95% CI, 1.00 to 1.01). Mortality of persons with acute renal failure requiring dialysis was 36 percent while in the hospital.5
Prerenal acute renal failure is caused by reduced blood flow to the kidney from renal artery disease, systemic hypotension, or maldistribution of blood flow. Intrarenal acute renal failure is caused by parenchymal injury (i.e., acute tubular necrosis, interstitial nephritis, embolic disease, glomerulonephritis, vasculitis, or small vessel disease). Postrenal acute renal failure is caused by urinary tract obstruction. Observational studies (in several hundred persons from Europe, North America, and West Africa with acute renal failure) found a prerenal cause in 40 to 80 percent, an intrarenal cause in 10 to 50 percent, and a postrenal cause in the remaining 10 percent.7,8,10–13 Prerenal acute renal failure is the most common type of acute renal failure in persons who are critically ill.7,14 Intrarenal acute renal failure in this context is usually part of multisystem failure, and most commonly because of acute tubular necrosis resulting from ischemic or nephrotoxic injury, or both.15,16
One retrospective study (1,347 persons with acute renal failure) found that mortality was less than 15 percent in persons with isolated acute renal failure.17 One recent prospective study (more than 700 persons) found that, in persons with acute renal failure, overall mortality (72 percent in ICU versus 32 percent in non-ICU; P = .001) and the need for dialysis (71 percent in ICU versus 18 percent in non-ICU; P < .001) were higher in an ICU than in a non-ICU setting, despite no significant difference between the groups in mean maximal serum creatinine (5.21 ± 2.34 mg per dL [460 ± 210 μmol per L] in ICU versus 5.82 ± 3.26 mg per dL [510 ± 290 μmol per L] in non-ICU).18 One large study (more than 17,000 persons admitted to Austrian ICUs) found that acute renal failure was associated with a greater than fourfold increase in mortality.19 Even after controlling for underlying severity of illness, mortality was still significantly higher in persons with acute renal failure (62.8 percent) than in persons without (38.5 percent), suggesting that acute renal failure is independently responsible for increased mortality even if dialysis is used. However, the exact mechanism that leads to increased risk of death is uncertain.
A systematic review including 80 articles and a total of 15,897 persons with acute renal failure from 1970 to 2004 found mortality unchanged, at about 50 percent, and exceeding 30 percent in most studies.20 An observational study including 54 sites and 23 countries screened 29,269 persons and found that 1,738 (5.7 percent) had severe acute renal failure warranting renal replacement therapy. Overall hospital mortality among persons with severe acute renal failure was 60.3 percent (95% CI, 58.0 to 62.6 percent).21