Prevention of contrast-associated acute kidney injury related to angiography

INTRODUCTION — The administration of radiocontrast media may lead to acute kidney injury (AKI) [1-11]. AKI is reversible in most cases, but its development may be associated with adverse outcomes [12].

The reported risk of contrast-associated acute kidney injury (CA-AKI) is much higher with procedures involving the arterial administration compared with venous administration of contrast. This difference in risk may be due to differences in patient populations (those who require arterial contrast are likely to have comorbidities that increase the likelihood of AKI) or to differences in the nephrotoxicity of intra-arterial contrast material.

This topic provides recommendations for the prevention of CA-AKI from arterial contrast administration. The prevention of CA-AKI from intravenous contrast administration is presented elsewhere. (See "Prevention of contrast-induced acute kidney injury associated with computed tomography".)

The pathogenesis, clinical features, and diagnosis of contrast-associated kidney injury are discussed separately. (See "Contrast-associated and contrast-induced acute kidney injury: Clinical features, diagnosis, and management".)

AKI and subacute kidney injury from other causes are discussed elsewhere:

●(See "Evaluation of acute kidney injury among hospitalized adult patients".)

●(See "Diagnostic approach to adult patients with subacute kidney injury in an outpatient setting".)

●(See "Etiology and diagnosis of prerenal disease and acute tubular necrosis in acute kidney injury in adults".)

EPIDEMIOLOGY — The reported incidence of contrast-associated acute kidney injury (CA-AKI) varies widely depending on the definition of AKI, the presence or absence of risk factors (primary chronic kidney disease [CKD]), the amount and type of agent administered, and the type of radiologic procedure.

According to virtually all reports, among patients who have no risk factors (particularly no CKD), the risk of CA-AKI is negligible (ie, ≤1 percent) [4,13,14].

Among at-risk patients (especially those with diabetes and CKD), the reported risk following coronary angiography with or without intervention is 10 to 30 percent [15-19]. (See 'Major risk factors' below.)

Major risk factors — Risk factors are patient related and procedure related.

Patient-related factors for CA-AKI include CKD (particularly from diabetic nephropathy), reduced renal perfusion from heart failure, hypovolemia or hemodynamic instability, and, possibly, multiple myeloma [2,7,9,13,20-23].

Procedure-related factors include the dose and type of contrast agent and the specific procedure (intra-arterial versus intravenous and interventional versus diagnostic angiography) [4,5,11,20,21,24].

Individual risk factors are discussed here. The discussion is limited to studies involving arterial administration of contrast, which is considered to be higher risk than venous administration. (See "Prevention of contrast-induced acute kidney injury associated with computed tomography", section on 'Epidemiology'.)

Chronic kidney disease — The incidence of CA-AKI is higher among patients with CKD and increases with the severity of kidney dysfunction [4,6,9,11,20,21].

The increase in risk is likely continuous with the decline in glomerular filtration rate (GFR). However, the threshold GFR at which a clinically significant risk is incurred is not well defined. One study allows an estimate of risk among patients with mild to moderate CKD [25]. In this study, among 289 patients who had an estimated GFR (eGFR) between 30 and 59 mL/min/1.73 m², the incidence of AKI was 4.2 percent following intra-arterial administration of contrast [25]. Patients with severe CKD (defined as eGFR <30 mL/min/1.73 m²) were excluded from this study.

Few studies have examined incidence in patients with severe CKD. In one study, among 124 patients with a serum creatinine ≥3.0 mg/dL (265 mmol/L) (virtually all who would have an eGFR <30 mL/min/1.73 m²), the incidence of AKI after percutaneous coronary interventions (PCIs) was 31 percent [9].

Proteinuria increases the risk further among patients with reduced eGFR [26,27]. In one prospective study of 70 consecutive CKD patients (with CKD defined by eGFR <60 mL/min/m² and/or proteinuria), AKI occurred after angiography in 62 percent of patients with proteinuria >1 gram/day versus 21 percent in patients with proteinuria <1 gram/day [27]. However, there was no between-groups difference in AKI among those that had a baseline eGFR >60 mL/min/1.73 m².

Diabetic nephropathy with reduced estimated GFR — Among CKD patients, diabetic patients are at higher risk compared with nondiabetic CKD patients [4,7,13]. In an analysis of data from a randomized trial that included 250 patients with serum creatinine >1.5 mg/dL (133 micromol/L), a higher incidence of AKI was observed among diabetic patients compared with nondiabetic patients (33 versus 12 percent, respectively) [4].

Among patients with normal kidney function, two studies suggest that diabetes does not increase the risk of CA-AKI [4,7]. In a randomized trial that included 341 patients with serum creatinine ≤1.5 mg/dL (133 mmol/L), there was no difference in the incidence of CA-AKI between diabetic and nondiabetic patients [4]. However, the event rate was too low (ie, <1 percent) to be confident of the relative risk conferred by diabetes. In a review of 1826 consecutive patients, dialysis-requiring AKI developed following interventional coronary angiography in 19.5 percent of diabetic patients versus 12.8 percent of nondiabetic CKD patients; however, no patient who had a baseline creatinine clearance >47 mL/min developed dialysis-requiring AKI [7].

Dose and type of contrast agent — Lower doses (<125 mL) of contrast tend to be safer, though are not free of risk [4-7,20,21,28]. Very small amounts of contrast (<10 mL) have been safely used in patients with advanced kidney disease for examination of arteriovenous fistulas [29]. However, diabetic patients with a serum creatinine concentration >5 mg/dL (440 micromol/L) may be at risk from as little as 20 to 30 mL of contrast [6].

The type of administered contrast agent may alter the risk. Contrast agents are either ionic or nonionic and of variable osmolality [10,30].

Older agents (first generation) were ionic and hyperosmolal compared with plasma (1400 to 1800 mosmol/kg). These agents are rarely used today. First-generation agents are more nephrotoxic than agents that are commonly used today [4,10].

Currently used agents include:

●Low-osmolal agents – Low-osmolal agents have a lower osmolality than the first-generation radiocontrast media but still higher osmolality (500 to 850 mosmol/kg) compared with plasma. Commonly used low-osmolal agents include the nonionic agents (iohexol, ioversol and iopamidol) and the ionic agent, ioxaglate.

●Iso-osmolal agents – Only one iso-osmolal agent (iodixanol) is available. Iodixanol is nonionic and iso-osmolal with plasma (290 mosmol/kg).

The decreased risk of newer agents is likely related to the decreased osmolality and the absence of charge that characterizes newer agents. The relative contributions of charge versus osmolality are difficult to determine since most studies compared nonionic low-osmolal agents versus ionic hyperosmolal agents [4,10]. However, a benefit of lower osmolality was conclusively demonstrated in one study in which the ionic low-osmolal agent ioxaglate was less toxic compared with ionic hyperosmolal agents (1500 to 1800 mosmol/kg) [10].

Among newer agents, the iso-osmolal agent iodixanol has been purported to have a lower risk of AKI than low-osmolal agents, particularly iohexol, among high-risk patients [15,31,32]. A meta-analysis that included 25 randomized trials comparing iodixanol with a diverse group of low-osmolal agents (n = 5053, approximately 72 percent with known renal impairment or diabetes) reported a modest reduction in risk of AKI (pooled relative risk [RR] 0.80, 95% CI 0.65-0.99) with iodixanol [32]. However, the small reduction in risk is unlikely to be clinically meaningful, although it was marginally statistically significant. There was no difference between groups in the risk for renal replacement therapy (RRT), cardiovascular outcomes, or death.

It is possible that iohexol carries an increased risk of AKI compared with non-iohexol low osmolal agents [15]. In a trial (included in the meta-analysis cited above [32]) that compared iodixanol and iohexol in 129 high-risk patients with diabetes and CKD (mean serum creatinine 1.5 mg/dL [133 micromol/L]) who underwent angiography, iodixanol was associated with a lower incidence of AKI (3 versus 26 percent with iohexol) [15]. However, three trials that compared iodixanol with two other nonionic low-osmolal contrast agents (ioversol and iopamidol) found no difference between groups in the incidence of AKI [18,33,34]. In addition, a meta-analysis of 16 randomized trials also suggested that iodixanol was associated with a reduction in risk among patients with CKD who received contrast when compared with iohexol (RR 0.19, 95% CI 0.07-0.56) but not when compared with other nonionic low-osmolal contrast agents (RR 0.79, 95% CI 0.56-1.12) [31].

These data suggest that the benefit of iodixanol compared with iohexol may represent a unique nephrotoxic effect of iohexol. However, there are no trials comparing nephrotoxicity of iohexol to other low-osmolar agents.

Specific radiologic procedure — The highest risk for AKI is associated with interventional (rather than diagnostic) coronary angiography (particularly in the setting of acute myocardial infarction) [20,35,36]. This may relate to the higher volume of contrast used in interventional procedures and hemodynamic instability associated with acute myocardial infarction.

Most studies have suggested that the risk of AKI associated with contrast computed tomography (CT) scans (ie, venous administration of contrast) is quite low, even among patients with CKD [37,38]. (See "Prevention of contrast-induced acute kidney injury associated with computed tomography", section on 'Incidence'.)

Other — Other risk factors that have been suggested by a few studies include hyperglycemia and the use of either angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs) [39,40].

Hyperglycemia may increase the risk for CA-AKI independent of a pre-existing diagnosis of diabetes mellitus. In a study of 6358 patients, the adjusted risk of AKI following angiography incrementally increased with higher glucose levels among patients without diabetes mellitus, with odds ratios (ORs) of 1.31, 1.5, 1.6, and 2.1 for glucose groups of 110 to <140, 140 to <170, 170 to <200, and >200 mg/dL [39]. There was no increase in glucose-associated risk among patients with established diabetes mellitus.

The effect of ACE inhibitors and/or ARBs on the incidence of CA-AKI is not clear. In a retrospective study of 5299 patients undergoing percutaneous interventions, compared with a propensity-matched cohort, ACE inhibitor or ARB users were more likely to develop AKI (odds ratio [OR] 1.43, 95% CI 1.06-1.94) [40].

However, in a randomized trial of 220 patients with eGFR of 15 to 60 mL/min/1.73 m², there was no difference in the incidence of CA-AKI between patients who were on ACE inhibitors and/or ARBs prior to angiography and patients who had a similarly reduced eGFR but were not on ACE inhibitors and/or ARBs [41]. In addition, it is not clear whether holding or withdrawing an ACE inhibitor and/or ARB prior to angiography provides any benefit.

PREVENTION — For all at-risk patients who are to receive intra-arterial contrast, we use preventive measures [1-3,42]. It is important to appreciate that most of the clinical trials that have evaluated these measures have used small, transient elevations in the serum creatinine concentration as endpoints (eg, ≥0.5 mg/dL [44.2 micromol/L] or ≥25 to 50 percent above baseline) [42]. Nevertheless, reports suggest that contrast-associated nephropathy, defined in this manner, is associated with significant in-hospital and long-term mortality [8,9,43]. (See 'Prognosis' below.)

Identifying patients at risk — At-risk patients include the following:

●All patients with estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m² who have significant proteinuria (defined as albuminuria >300 mg/day, which corresponds to proteinuria > 500 mg/day).

●All patients with eGFR <60 mL/min/1.73 m² and comorbidities including diabetes, heart failure, liver failure, or multiple myeloma.

●All patients with eGFR <45 mL/min/1.73/m² even in the absence of proteinuria or any other comorbidities.

●Patients who have eGFR <45 mL/min/1.73 m² and have proteinuria and diabetes or other comorbidities and all patients with eGFR <30 mL/min/1.73 m² should be considered at highest risk.

Preventive measures

Avoid volume depletion and NSAIDs — Patients who are to receive intra-arterial contrast should avoid volume depletion and withhold nonsteroidal antiinflammatory agents (NSAIDs) for 24 to 48 hours prior to the procedure. Both volume depletion and NSAIDs can increase renal vasoconstriction, which increases the risk of contrast-associated acute kidney injury (CA-AKI). (See 'Major risk factors' above.)

We do not withhold angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs). There are insufficient data to support a benefit of withholding ACE inhibitors and ARBs, and there are risks associated with resulting hypertension. (See 'Withholding ACE inhibitors and/or ARBs' below.)

Dose and type of contrast agent — We use the lowest effective dose possible of contrast and avoid performing repeated studies that are closely spaced (within 48 to 72 hours) [4-7,21,28]. (See 'Dose and type of contrast agent' above.)

We use the iso-osmolal agent, iodixanol, or nonionic low-osmolal agents, such as iopamidol or ioversol, rather than iohexol. We do not use high-osmolal agents (1400 to 1800 mosmol/kg). Nonionic, iso- or low-osmolal agents are safer than ionic high-osmolal agents. (See 'Dose and type of contrast agent' above.)

The iso-osmolal agent, iodixanol, may be safer than iohexol among at-risk patients but has been associated with only a slight, if any, benefit compared with other nonionic low-osmolal agents. (See 'Dose and type of contrast agent' above.)

Our approach is consistent with the American College of Cardiology/American Heart Association (ACC/AHA) guidelines on percutaneous coronary intervention (PCI), which were revised to suggest the use of either an iso-osmolal contrast agent or a low-molecular-weight contrast agent other than iohexol or the ionic low-osmolal agent, ioxaglate [44]. The 2012 Kidney Disease: Improving Global Outcomes (KDIGO) guidelines also recommended low-osmolal or iso-osmolal rather than high-osmolal contrast agents but stated that the work group found no reliable evidence upon which to base a recommendation for either low- versus iso-osmolal agents [45].

Fluid administration — For all at-risk patients undergoing procedures involving intra-arterial contrast administration, if there are no contraindications to volume expansion, we administer intravenous isotonic saline prior to and continued for several hours after contrast administration.

Our preferred protocols are as follows:

●Outpatients ─ We give 3 mL/kg over one hour preprocedure and 1 to 1.5 mL/kg/hour during and for four to six hours postprocedure, with administration of at least 6 mL/kg postprocedure.

●Inpatients ─ We give 1 mL/kg/hour for 6 to 12 hours preprocedure, intraprocedure, and for 6 to 12 hours postprocedure.

Rationale — Intravenous volume administration prior to intravascular contrast administration for patients at risk for contrast-associated nephropathy is the standard of care despite an absence of adequately designed randomized trials demonstrating benefit [45]. Further studies that compare prophylactic fluid administration to no fluid administration are required to determine the role for prophylactic fluid administration, particularly in the highest-risk patients.

Only a few randomized trials have examined the effects of intravenous fluids in the absence of other interventions (such as forced diuresis) [25,35,36,46,47]. The following three randomized trials demonstrated benefit, but all were limited for various reasons:

●In a small, randomized trial of 53 unselected patients undergoing nonemergency cardiac catheterization, intravenous saline decreased the risk of CA-AKI compared with unrestricted oral hydration (4 versus 35 percent) [46]. The incidence of CA-AKI was higher than anticipated in the oral hydration group, and the study was stopped early by the safety monitoring board because of the high rate of acute kidney injury (AKI) in this group [46]. Early termination of the trial limits confidence in the result. Ionic hyperosmolar-based contrast media, which is more nephrotoxic than agents commonly used today, was used in all subjects in this trial.

●In two trials of 408 and 216 patients with acute myocardial infarction undergoing PCI, most of whom had normal kidney function, intravenous saline reduced the risk of CA-AKI compared with no saline (11 versus 21 percent and 20 versus 35 percent, respectively) [35,36]. In one trial, the in-hospital mortality rate was lower with intravenous saline (2.8 versus 9.3 percent in control) [36]. Rates of ventricular arrhythmias, pulmonary edema, dialysis, mechanical ventilation, and other adverse events were also lower in patients who received saline.

However, in both trials, the incidence of AKI, which was reduced with fluid administration, may have been related to a reduction in episodes of hypotension with intravenous fluids rather than reflecting a direct protective role against CA-AKI in all high-risk groups. In one trial in which 28 percent of the randomized population crossed over to the other group, a significant number of individuals allocated to no fluid crossed over to fluid because of concerns of hypotension [35].

In addition to studies cited above, a non-placebo-controlled randomized trial (POSEIDON) provides indirect evidence of a dose-response benefit of fluid administration [48]. In POSEIDON, an aggressive fluid replacement protocol guided by left ventricular end-diastolic pressure (LVEDP) was compared with standard intravenous fluid administration among patients with eGFR <60 mL/min/1.73 m² and other risk factors [48]. All patients received intravenous isotonic saline 3 mL/kg for one hour prior to cardiac catheterization. LVEDP was determined in all patients prior to administration of contrast.

In the LVEDP-guided group, patients received 5 mL/kg/hour if LVEDP was lower than 13 mmHg, 3 mL/Kg/hour if LVEDP was between 13 and 18 mmHg, and 1.5 mL/Kg/hour if LVEDP was greater than 18 mmHg. The control group received 1.5 mL/kg/hour. Both groups received intravenous fluid at rates defined above throughout the procedure and for four hours afterward.

CA-AKI occurred less frequently in the LVEDP group compared with control (6.7 versus 16.3, respectively [relative risk (RR) 0.4, 95% CI 0.2-0.8]). Three patients in each group stopped intravenous fluids early because of dyspnea.

By contrast to trials cited above, a single-center, randomized trial (AMACING) found no benefit of intravenous saline compared with no saline in preventing AKI among 603 patients with eGFR between 30 and 59 mL/min/1.73 m² [25]. Adverse events including heart failure, hyponatremia, and arrhythmia were more common in the group receiving intravenous fluid (5.5 versus 0 percent in the group not receiving fluid).

However, the population was not a high-risk group for CA-AKI, as evidenced by the overall rate of AKI of only 2.7 percent; only 48 percent of procedures involved intra-arterial administration of contrast, and 65 percent had only mild chronic kidney disease (CKD; defined by an eGFR 46 to 59 mL/min/1.73 m²). Among 289 patients who received intra-arterial contrast, only six in each group developed AKI, and no patient required dialysis [49].

Isotonic saline — Isotonic saline appears to be better than more hypotonic fluids (ie, one-half isotonic saline) [50,51]. In a randomized trial of 1620 patients, compared with half-isotonic saline, isotonic saline reduced the risk of CA-AKI (2 versus 0.7 percent) [51]. The benefit of isotonic saline was greater in diabetic patients (5.5 percent versus 0) and those given >250 mL of contrast (3 versus 0 percent). However, there was no difference among those with significant kidney dysfunction (serum creatinine >1.6 mg/dL [>141 micromol/L], 14 and 17 in the saline and one-half normal saline groups, respectively). While this is an important limitation since it is patients with underlying kidney dysfunction who are at increased risk for CA-AKI, the study was underpowered with regard to subgroups; thus, a small benefit of saline may have been missed in this group.

Saline versus bicarbonate — We recommend saline rather than bicarbonate. Both are effective, but bicarbonate provides no additional benefit to saline, needs to be compounded, and is more expensive.

Sodium bicarbonate has been compared with isotonic saline in a number of randomized trials and meta-analyses with conflicting results [16,52-65]. The most definitive data are from a subsequently published randomized trial that included 4993 high-risk patients undergoing scheduled angiography that found that both treatments were associated with similar outcomes [66]. All patients had stable kidney function with eGFR 15 to 44.9 mL/min/1.73 m² with or without concomitant diabetes mellitus or eGFR 45 to 59.9 mL/min in the setting of diabetes mellitus. Eighty-one percent of patients had diabetes mellitus.

Using a two-by-two factorial design, patients were assigned to receive 1.26 percent sodium bicarbonate or 0.9 percent sodium chloride and either oral acetylcysteine (1200 mg one hour before and one hour after angiography and then twice daily for four days) or matched placebo capsules. Hydration was administered according to protocol-specified ranges of 1 to 3 mL/kg per hour for 1 to 12 hours before angiography (total 3 to 12 mL/kg before angiography), 1 to 1.5 mL/kg per hour during angiography, and 3 mL/kg per hour for 2 to 12 hours after angiography (total 6 to 12 mL/kg after angiography).

Similar rates of AKI (9.5 versus 8.3 percent [odds ratio (OR) 1.16, 95% CI 0.96-1.41]), need for dialysis at 90 days (1.3 versus 1.2 [OR 1.09, 95% CI 0.65-1.81]) or persistent kidney impairment by 90 days (1.1 versus 1.0 [OR 1.10, 95% CI 0.64-1.91]), and death (2.4 versus 2.7 [OR 0.87, 95% CI 0.61-1.24]) were seen in the bicarbonate and saline treatment groups. There was no interaction between sodium bicarbonate and acetylcysteine in the analysis of primary or secondary endpoints. The results regarding the effects of acetylcysteine are discussed below. (See 'Acetylcysteine' below.)

Oral salt loading — We do not use oral salt loading in place of intravenous volume administration. The benefit of oral hydration or salt loading for the prevention of CA-AKI is not known. However, two small trials have suggested that oral salt may provide a benefit comparable with intravenous fluids [67,68].

Acetylcysteine — We do not give acetylcysteine prior to angiography. Meta-analyses examining acetylcysteine have yielded conflicting results [42,58,66,69-86]. In general, modest benefits were noted in meta-analyses that did not account for a large degree of heterogeneity between studies [87]. However, the largest randomized trial, which was published after the meta-analyses, did not find improved outcomes with oral acetylcysteine in 4993 high-risk patients undergoing scheduled angiography [66].

Patients were assigned to receive sodium bicarbonate or sodium chloride and either oral acetylcysteine (1200 mg one hour before and one hour after angiography and then twice daily for four days) or matched placebo capsules. The details of this trial are described above. (See 'Saline versus bicarbonate' above.)

Similar rates of AKI (9.1 versus 8.7 percent [OR 1.06, 95% CI 0.87-1.28]), need for dialysis at 90 days (1.2 versus 1.2 [OR 0.97, 95% CI 0.58- 1.60]) or persistent kidney impairment by 90 days (1.0 versus 1.1 [OR 0.96, 95% CI 0.56-1.66]), and death (2.7 versus 2.4 [OR 1.10, 95% CI 0.78-1.57]) were seen in the acetylcysteine and placebo treatment groups. There was no interaction between sodium bicarbonate and acetylcysteine in the analysis of primary or secondary endpoints.

We also do not use intravenous acetylcysteine, due to risk of serious adverse events [88-90]. The benefit of intravenous acetylcysteine remains uncertain, and comparison of the various trials is difficult because of differences in patient populations and dosing or lack of an adequate control group [88-90]. In one trial, 7 percent of patients who received high doses of intravenous acetylcysteine developed anaphylactoid reactions [89].

OTHER UNPROVEN INTERVENTIONS — A variety of other interventions have been tried, including remote ischemic preconditioning (RIPC), atrial natriuretic peptide, statins, and ascorbic acid.

Remote ischemic preconditioning — RIPC is a method by which the deliberate induction of transient nonlethal ischemia of an organ protects against subsequent ischemic injury of another organ. Some, but not all, studies have suggested that RIPC prior to cardiac surgery protects against acute kidney injury (AKI). (See "Possible prevention and therapy of ischemic acute tubular necrosis".)

RIPC may also protect against contrast-associated acute kidney injury (CA-AKI) [91,92]. These results require confirmation in larger randomized trials before RIPC can be recommended as a preventive measure for CA-AKI [93].

Prophylactic hemofiltration and hemodialysis — We do not use routine hemofiltration or hemodialysis for the prevention of CA-AKI in patients with chronic kidney disease (CKD).

A 2012 meta-analysis that included eight studies of hemodialysis and three studies of hemofiltration/hemodiafiltration showed no benefit of renal replacement therapy (RRT) [94].

There is no indication for prophylactic dialysis for the prevention of volume overload from intravascular contrast administration in dialysis-dependent patients [95].

Furthermore, there are no studies that support immediate dialysis after intravascular contrast media administration in order to preserve residual kidney function or limit the risk of allergic or toxic reactions to contrast media in hemodialysis patients [95-97]. Whereas some clinicians try to perform a hemodialysis treatment within 24 to 36 hours after intravascular contrast media exposure, others wait 48 to 72 hours, until the next scheduled hemodialysis treatment [98]. We support this latter approach, except in hemodialysis patients who are significantly volume overloaded at the time of contrast administration.

Withholding ACE inhibitors and/or ARBs — We do not withhold angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs) prior to angiography. Some studies have suggested that patients who are on an ACE inhibitor or ARBs are at higher risk for CA-AKI compared with those who are not [40]. (See 'Other' above.)

However, it is not clear whether holding or withdrawing an ACE inhibitor and/or ARB prior to angiography provides any benefit. In one study, 220 patients who were on ACE inhibitors or ARBs and had an estimated glomerular filtration rate (eGFR) of 15 to 59 mL/min/1.73 m² were randomly assigned prior to angiography to either an ACE inhibitor/ARB discontinuation group, in which the ACE inhibitor or ARB was held 24 hours prior to procedure, or to a control group, in which these agents were continued [41]. There was no significant difference in the incidence of CA-AKI between patients who had the ACE inhibitor and/or ARB withdrawn and those who did not.

The issue of whether to withhold ACE inhibitors and ARBs prior to contrast procedure is not resolved, and further study of this important issue is warranted.

Statins — We do not initiate statins solely for the prevention of CA-AKI. A possible exception is patients who are likely to be started on statins prior to discharge, such as those with acute myocardial infarction, among whom it is reasonable to start the statin prior to angiography.

Some studies [99-105], though not all [106], have suggested that statins may reduce the risk. A meta-analysis of eight studies (n = 5024) did not show a conclusive benefit of statins plus intravenous saline compared with saline alone [65]. However, analysis of five studies (n = 1477) revealed that statins given with N-acetylcysteine and intravenous saline reduced the risk of CA-AKI compared with N-acetylcysteine and intravenous saline alone (relative risk [RR] 0.52, 95% CI 0.29-0.93). Most of the statin studies were performed in relatively low-risk patients. In addition, some studies suggest that acute administration of rosuvastatin is associated with an increase in normal eGFR, which confounds interpretation of AKI studies [107].

Statins merit further study for the prevention of contrast-associated nephropathy.

Diuretics — We do not routinely administer prophylactic diuretics or mannitol for prevention of AKI. These agents do not appear to be beneficial for the prevention of CA-AKI [108-110]. However, diuretics may be required to treat volume overload.

It is possible that forced diuresis may provide benefit in preventing AKI if volume depletion can be avoided. This was tested in studies that utilized a proprietary fluid management device that matches fluid replacement to urine output [111-113]. Forced diuresis and matched volume repletion decreased the incidence of AKI after angiography compared with standard overnight hydration. The potential value of the fluid management device is that volume infusion and urine output are per hour significantly greater than with conventional hydration regimens without the risk of volume overload.

However, high-volume forced diuresis raises some concerns, even in the setting of carefully matched volume replacement. As an example, volume overload may ensue if isotonic saline is used to replace urine, which, following the administration of loop diuretics, generally has the tonicity of approximately half normal saline. In one of the trials cited above, such an increase in intravascular volume may have been sufficient to increase the eGFR in patients undergoing forced diuresis [113].

Additionally, arrhythmia-inducing electrolyte abnormalities such as hypokalemia may result from forced diuresis in a vulnerable population. Additional studies that address these issues are necessary before forced diuresis and matched volume replacement are used clinically.

Other — Other preventive measures that have been studied include oral sodium citrate, atrial natriuretic peptide, ascorbic acid, trimetazidine, inhibitors of vasoconstriction, and diuretics:

●Oral sodium citrate – One randomized trial has demonstrated a benefit of oral sodium citrate (5 g in 200 mL water) one hour before and four hours after angiography [114]. The risk of CA-AKI was lower among patients who received oral citrate compared with placebo. This result requires confirmation by other trials before such an approach is recommended.

●Atrial natriuretic peptide – Atrial natriuretic peptide (anaritide) has been considered for prophylaxis in high-risk patients since its administration has been beneficial in animal models of CA-AKI [115]. However, no benefit was observed in a multicenter, prospective, double-blind, placebo-controlled, randomized trial [116].

●Ascorbic acid – Data are insufficient to support the use of ascorbic acid for prevention of CA-AKI. One randomized trial suggested a benefit of ascorbic acid in prevention of contrast-mediated nephropathy [117]. However, a second large, well-designed trial found that ascorbic acid did not provide added benefit to a prophylactic regimen of isotonic saline plus acetylcysteine among patients at high risk [53]. A meta-analysis of six studies did not conclusively demonstrate a benefit of ascorbic acid [65].

●Trimetazidine – Trimetazidine, a cellular anti-ischemic agent, provided added protection to isotonic saline from contrast-mediated nephropathy in an initial, small, randomized, prospective study [118]. Further study in a larger number of patients is required to better characterize the effectiveness of this agent.

●Inhibitors of vasoconstriction – Multiple agents that prevent vasoconstriction, including theophylline or aminophylline, nifedipine, captopril, prostaglandin E or I2, low-dose dopamine, and fenoldopam, have been studied for prevention of CA-AKI [119-130]. Some [126,128,131], but not all [129,130,132], studies suggest that vasodilator agents may reduce the risk, although additional studies are required before any such agents are recommended in this setting.

PROGNOSIS — In most cases, contrast-associated acute kidney injury (CA-AKI) is reversible; the estimated glomerular filtration rate (eGFR) recovers in 5 to 10 days. However, in patients who have severely reduced eGFR at baseline, the creatinine may not return to baseline values. Even if the creatinine returns to baseline, the development of CA-AKI has been associated with short- and long-term adverse outcomes. (See "Contrast-associated and contrast-induced acute kidney injury: Clinical features, diagnosis, and management", section on 'Management'.)

However, there are no randomized trials that have proven that specific interventions to prevent AKI decrease mortality or major cardiovascular events.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Chronic kidney disease in adults".)

SUMMARY AND RECOMMENDATIONS

●Patients with near-normal kidney function are at low risk for contrast-associated acute kidney injury (CA-AKI), and few precautions are necessary, other than avoidance and/or correction of volume depletion. Procedures involving intra-arterial contrast are higher risk than those involving intravenous contrast. (See 'Epidemiology' above and 'Dose and type of contrast agent' above.)

●Patients at risk for CA-AKI include those with estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m², who also have proteinuria (defined as albuminuria >300 mg/day which corresponds to proteinuria >500 mg/day), diabetes, or other comorbidities including heart failure, liver failure, or multiple myeloma. Patients are also at risk who have eGFR <45 mL/min/1.73 m², even in the absence of proteinuria, diabetes, or other comorbidities. Patients who have eGFR <45 mL/min/1.73 m² and have proteinuria and diabetes or other comorbidities and all patients with eGFR <30 mL/min/1.73 m² should be considered at highest risk. (See 'Major risk factors' above.)

●For all at-risk patients, we use the following preventive measures:

•We recommend the use of iodixanol or nonionic low-osmolal agents, such as iopamidol or ioversol, rather than iohexol (Grade 1B). We do not use high-osmolal agents (1400 to 1800 mosmol/kg). (See 'Dose and type of contrast agent' above.)

•We use lower doses of contrast and avoid repetitive, closely spaced studies (eg, <48 hours apart). (See 'Dose and type of contrast agent' above.)

•Avoid volume depletion and nonsteroidal antiinflammatory drugs (NSAIDs). (See 'Avoid volume depletion and NSAIDs' above.)

•For patients at highest risk, defined as those who have eGFR <45 mL/min/1.73 m², proteinuria, and diabetes or other comorbidities, and all patients with eGFR <30 mL/min/1.73 m², in the absence of contraindications to volume expansion, we recommend intravenous fluids prior to and continued for several hours after contrast administration (Grade 1B). While no placebo-controlled studies have proven a benefit of prophylactic intravenous fluid in these risk groups, indirect data support its use.

In addition, for all at-risk patients in the absence of contraindications to volume expansion, we suggest intravenous fluids prior to and continued for several hours after contrast administration (Grade 2C). We recommend isotonic saline rather than bicarbonate to prevent CA-AKI (Grade 1B). Bicarbonate provides no additional benefit to saline, needs to be compounded, and is more expensive. (See 'Fluid administration' above.)

Timing and rate of administration are independent of fluid type and vary between inpatients and outpatients:

-Among outpatients, we give 3 mL/kg over one hour preprocedure and 1 to 1.5 mL/kg/hour during and for four to six hours postprocedure, with administration of at least 6 mL/kg postprocedure, regardless of fluid type.

-Among inpatients, we give 1 mL/kg/hour for 6 to 12 hours preprocedure, intraprocedure, and for 6 to 12 hours postprocedure. (See 'Fluid administration' above.)

•For all at-risk patients, we suggest that acetylcysteine not be given either orally or intravenously (Grade 2B). Although meta-analyses have been inconsistent and marked by heterogeneity, a large randomized trial found similar outcomes in patients receiving acetylcysteine compared with placebo. (See 'Acetylcysteine' above.)

•We do not use mannitol or other diuretics prophylactically. Diuretics have not been shown to be effective in preventing CA-AKI. However, diuretics may be used to treat volume overload if present. (See 'Diuretics' above.)

•We do not perform prophylactic hemofiltration or hemodialysis after contrast exposure to prevent CA-AKI. (See 'Prophylactic hemofiltration and hemodialysis' above.)