Preventing radiocontrast-induced nephropathy in chronic kidney disease patients undergoing coronary angiography

Abstract

Radiocontrast-induced nephropathy (RCIN) is an acute and severe complication after coronary angiography, particularly for patients with pre-existing chronic kidney disease (CKD). It has been associated with both short- and long-term adverse outcomes, including the need for renal replacement therapy, increased length of hospital stay, major cardiac adverse events, and mortality. RCIN is generally defined as an increase in serum creatinine concentration of 0.5 mg/dL or 25% above baseline within 48 h after contrast administration. There is no effective therapy once injury has occurred, therefore, prevention is the cornerstone for all patients at risk for acute kidney injury (AKI). There is a small but growing body of evidence that prevention of AKI is associated with a reduction in later adverse outcomes. The optimal strategy for preventing RCIN has not yet been established. This review discusses the principal risk factors for RCIN, evaluates and summarizes the evidence for RCIN prophylaxis, and proposes recommendations for preventing RCIN in CKD patients undergoing coronary angiography.

INTRODUCTION

Over the past decade there has been dramatic growth worldwide in contrast-enhanced imaging services, many involving exposure to iodinated contrast media[1,2]. Acute deterioration in renal function due to contrast media administration is a well-recognized complication after coronary angiography, particularly for patients with pre-existing chronic kidney disease (CKD)[2-5]. Previous studies have shown that 12%-14% of patients who develop acute renal insufficiency during hospitalization do so after procedures involving radiographic contrast[6,7]. For patients with abnormal baseline renal function, the incidence of progressive deterioration can be as high as 42%[2,8].

Radiocontrast-induced nephropathy (RCIN), also called radiocontrast-induced acute kidney injury (AKI), is associated with increased health resource utilization, prolonged hospital stay, increased in-hospital and long-term mortality, and an acceleration in the rate of progression of CKD[6,9-13]. It has been associated with both short- and long-term adverse outcomes, including the need for renal replacement therapy (RRT), and major cardiac adverse events. Importantly, RCIN is associated with increased short- and long-term mortality[14-16].

RCIN has gained increased attention in the clinical setting, particularly during cardiac intervention, but also in many other radiological procedures in which iodinated contrast media are used. There are at least four factors explaining that the incidence of RCIN is likely to increase in the future[17,18]. First, CKD, which is a principal risk factor for RCIN, is likely to continue to grow in prevalence. Second, diabetes mellitus, which amplifies the risk for RCIN in patients with underlying CKD, is also increasing in prevalence. Third, the world population is aging, and more elderly patients are undergoing contrast-enhanced procedures. Fourth, patients with advanced CKD have been shown to be at risk for nephrogenic systemic fibrosis; a potentially debilitating disorder that is associated with the administration of gadolinium-based contrast agents[17]. As a result, many patients with CKD, who in recent years would have undergone contrast-enhanced magnetic resonance angiography, are now likely to undergo conventional angiography with iodinated contrast, thus, increasing the number of patients at risk for RCIN[18].

RCIN has been defined as the acute deterioration of renal function after parenteral administration of radiocontrast media in the absence of other causes[19]. Unfortunately, the definition of RCIN has not been consistent in the literature, which makes the comparison of data from various studies difficult. It is generally defined as an increase in serum creatinine concentration of 0.5 mg/dL (44 mmol/L) or 25% above baseline within 48 h after contrast administration[20-24]. There is no effective therapy once injury has occurred, therefore, prevention is the cornerstone for all patients at risk for AKI. There is a small but growing body of evidence showing that prevention of AKI is associated with a reduction in adverse outcomes[25]. The optimal strategy for preventing RCIN has not yet been established. This review discusses the principal risk factors for RCIN and current interventions, as well as evidence behind each intervention for preventing RCIN in patients with baseline renal dysfunction undergoing coronary angiography.

RISK FACTORS FOR RCIN

The pathophysiology of RCIN in humans is not clearly established, and is probably multifactorial. It is hypothesized that a combination of ischemia due to vasoconstriction and direct toxicity to the renal tubules mediated via reactive oxygen species leads to RCIN[26]. Intra-arterial infusion of contrast medium results in initial vasodilation, followed by vasoconstriction, accompanied by shunting of blood flow from the medulla to the cortex, with a net result of a 20% increase in blood flow to the cortex and a 40% decrease in blood flow to the medulla; the medullary ischemia that ensues is thought to contribute to tubular injury[27]. Direct cytotoxicity to the tubular epithelial cells has been demonstrated, as demonstrated by vacuolization and tubular epithelial cell death. In addition, production of reactive oxygen species and subsequent tubular damage have been demonstrated in several animal studies[27].

Many risk factors have been described for RCIN, among which pre-existing renal disease is the most important. The increasing levels of renal impairment have been associated with escalating levels of risk[28]. Among patients in the Minnesota Registry of Interventional Cardiac Procedures, RCIN was diagnosed in 22% of patients with serum creatinine > 2 mg/dL and in 30% of patients with serum creatinine > 3 mg/dL[16]. Diabetes, increased age, higher dose of contrast agent, route of contrast administration (intra-arterial vs intravenous), congestive heart failure (CHF), hypertension, periprocedural shock, baseline anemia, postprocedural drop in hematocrit, use of nephrotoxins, and nonsteroidal anti-inflammatory medications, volume depletion, increased creatine kinase-MB, and need for cardiac surgery after contrast exposure, have been associated with increased risk of RCIN[16,29]. Prior studies have shown that procedural issues such as the amount and type of contrast administered[30-32] could be the additional risk factors. Risk factor scoring (including both baseline comorbidity and procedural factors) has been used to predict the incidence of RCIN, need for RRT, and long-term mortality[33,34]. They have been developed to identify modifiable and nonmodifiable risk factors for RCIN. Mehran et al[33] have published a simple risk score of RCIN including both preprocedural and periprocedural risk factors. Mehran’s model includes CHF, hypotension, intra-aortic balloon pump, age > 75 years, anemia, diabetes mellitus, contrast volume, and estimated glomerular filtration rate (eGFR)[33]. Brown et al[34] have developed a similar model but restricted it to only preprocedural risk factors that would be accessible for clinicians to ascribe risk of RCIN prior to the case. They found preprocedural serum creatinine, CHF, and diabetes accounted for > 75% of the predictive model, whereas other factors accounting for the remainder of the risk model were urgent and emergency priority, preprocedural intra-aortic balloon pump use, age ≥ 80 years, and female sex[34,35]. Both models have identified potentially modifiable risk factors that include priority of the procedure, baseline renal function, diabetes, and contrast volume[35] (Table 1). The unfavorable prognostic implications of RCIN make preventing the condition of paramount importance, especially in the performing of some targeted interventions focus on the modifiable risk factors to reduce this risk[35,36].

Table 1.

Risk factors for radiocontrast-induced nephropathy

Modifiable risk factors

Higher dose of contrast agent used

Congestive heart failure

Periprocedural shock

Anemia or postprocedural drop in hematocrit

Use of nephrotoxins

Use of nonsteroidal anti-inflammatory medications

Dehydration

Hypertension or blood pressure control

Diabetes mellitus or sugar control

Increased CK-MB

Urgent or emergency priority of the procedure

Need for cardiac surgery after contrast exposure

Preprocedural intra-aortic balloon pump use

Non-modifiable risk factors

Age > 75 yr

Female sex

Baseline renal function

PREVENTION OF RCIN

Although the treatment of established RCIN is limited to supportive care and dialysis, the renal injury resulting from iodinated contrast exposure is potentially preventable[18]. Multiple interventions have been investigated for their capacity to prevent RCIN. The initial steps in reducing the risk of kidney injury are looking for risk factors and reviewing the indications for the administration of contrast medium. Most risk factors can be detected by history taking and physical examination. Factors such as dehydration can be at least partially corrected before exposure to the contrast medium[37]. According to the statement of Weisbord et al[18], efforts to find effective preventive interventions for RCIN have focused on four principal strategies: (1)administration of less nephrotoxic contrast agents; (2) provision of pre-emptive RRT to remove contrast from the circulation; (3) utilization of pharmacological agents to counteract the nephrotoxic effects of contrast media; and (4) expansion of the intravascular space and enhanced diuresis with intravenous fluids.

TYPE AND VOLUME OF CONTRAST MEDIA IN RCIN

Volume of contrast media

The correlation between the amount of contrast and risk of RCIN has been reported in several studies[6,12,30,31,33,38,39]. According to McCullough et al[6], the risk of RCIN is minimal in patients receiving < 100 mL contrast. In another study in the diabetic population, RCIN developed in approximately every fifth, fourth and second patient who received 200-400, 400-600 and > 600 mL contrast, respectively[38]. In one observational study[31] of 561 patients with ST-elevation myocardial infarction who were undergoing primary angioplasty, higher amounts of contrast volume were associated with higher rates of RCIN and mortality. Kane et al[40] have determined in a multivariate analysis that the only predictor of RCIN in patients with pre-existing CKD was volume of contrast administered. They found that very-high-risk patients undergoing coronary arteriography who received 14 ± 4 mL contrast had a 4.4% incidence of RCIN; whereas those receiving 61 ± 12 mL had a 29.8% incidence. Thus, minimizing contrast media amount is the key element in preventing renal function impairment in patients undergoing cardiovascular intervention[36].

Some researchers have studied the relationship between the maximum radiographic contrast dose (MCD), baseline renal function, and risk of RCIN after angiography[8,31,40-42]. The MCD was calculated using the following formula: 5 mL contrast medium/kg body weight (maximum 300 mL) divided by serum creatinine (mg/dL)[41]. This calculated MCD was validated in 115 patients with CKD (creatinine > 1.8 mg/dL) undergoing angiography over a 10-year period[8]. Liu et al[41] have used the ratio of contrast media volume/estimated glomerular filtration rate, or V/eGFR to evaluate the RCIN in patients with ST-elevation myocardial infarction who underwent primary PCI. They reported that a V/eGFR ratio ≥ 2.39 was a significant and independent predictor of RCIN in their series. In another registry of > 16 000 percutaneous coronary interventions (PCIs) by Freeman et al[42], the strongest independent predictor of nephropathy requiring dialysis in patients undergoing elective coronary interventions was exceeding the calculated MCD. Patients who received a volume of contrast that exceeded the MCD were six times more likely to develop RCIN. Recently, Marenzi et al[31] have defined the contrast ratio as the ratio between the contrast volume administered and the MCD. They have demonstrated that patients who received more than the MCD had a more complicated in-hospital clinical course and a higher RCIN rate than patients administered less contrast volume than the MCD (35% vs 6%, P < 0.001). Serum creatinine, especially in the elderly and female populations, who have low muscle mass, is an inexact measure of renal function, therefore, it is more accurate, and should be routine, to estimate creatinine clearance by the Cockroft-Gault formula or eGFR calculated using the modification of diet in renal disease equation for adults[31]. According to Laskey et al[43], the ratio of the volume of contrast to creatinine clearance (V/CrCl) of > 3.7 is a good predictor of RCIN.

Type of contrast agents

Contrast media are classified by their osmolality: high osmolar contrast media (HOCM), 2000 mOsm/kg; low-osmolar contrast media (LOCM), 600-800 mOsm/kg; and iso-osmolar contrast media (IOCM), 290 mOsm/kg[44]. Over the past 40 years, the osmolality of available contrast media has been gradually decreased to physiological levels. In the 1950s, only HOCM (e.g., diatrizoate) with osmolality 5-8 times that of plasma were available. In the 1980s, LOCM agents such as iohexol, iopamidol and ioxaglate were introduced with osmolality 2-3 times greater than that of plasma. In the 1990s, iso-osmolar nonionic iodixanol with the same physiological osmolality as blood was developed[45]. The impact of contrast media osmolality on the incidence of RCIN had been assessed in several randomized trials. In a meta-analysis of the relative nephrotoxicity of the HOCM and LOCM reported in 1993, the pooled OR for the incidence of contrast-induced AKI events in 25 trials was 0.61 (95% CI: 0.48-0.77) times that after HOCM, indicating a significant reduction in risk with LOCM compared to HOCM[46]. In a multicenter randomized controlled trial (RCT) of 1196 patients, Rudnick et al[47] demonstrated a reduction in the incidence of RCIN (as defined by an increase in serum creatinine > 1 mg/dL at 48-72 h post-procedure) with the use of LOCM iohexol (3%), as compared with HOCM diatrizoate (7%, P < 0.002) in patients with pre-existent renal insufficiency, independent of the presence of diabetes mellitus.

More recent studies have focused on the comparative nephrotoxicity of IOCM iodixanol, and various LOCM. Several studies have evaluated whether an IOCM might provide a similar benefit over LOCM agents, but no consensus has emerged. Some randomized trials have suggested a lower incidence of RCIN with iodixanol in high-risk patients[48-52], while some other clinical trials have failed to show a benefit from use of IOCM[53-60]. A recent multicenter trial (ACTIVE trial) even demonstrated favorable results for LOCM (iomeprol) compared to IOCM (iodixanol), and the authors concluded that LOCM appeared to have a protective effect compared to IOCM (0% vs 6.9%)[61]. Of particular interest, Wessely et al[57] randomized 324 patients with CKD undergoing coronary angiography with PCI to receive either iodixanol or the low osmolar agent iomeprol. The primary endpoint was the peak increase in serum creatinine during hospitalization for PCI. They found contrast-medium-induced nephropathy rates were lower with iodixanol (22.2% vs 27.8% for iomeprol), but this difference was not statistically significant (P = 0.25). However, subgroup analysis suggested a favorable outcome regarding nephrotoxicity in patients who received higher contrast volumes (> 340 mL) in the iodixanol group compared with the LOCM group (P = 0.016).

The disparate results of these previous 14 clinical trials led to a proliferation of systematic reviews and meta-analyses comparing the nephrotoxicity of IOCM iodixanol and various LOCM; five of which have now been published[32,62-66]. Among earlier meta-analyses by McCullough et al[62], pooled data from 2727 patients from 16 double-blind, RCTs have indicated that use of the IOCM iodixanol is associated with smaller rises in creatinine and lower rates of RCIN than LOCM (1.4% vs 3.5%, P < 0.001), especially in patients with CKD or CKD and diabetes mellitus. Solomon has published a systematic review of prospective, randomized, controlled studies of RCIN in 1365 renal impaired patients receiving intra-arterial doses of the IOCM, iodixanol, or other LOCM, and conducted a pooled analysis of the data from those studies to determine whether the osmolality of contrast media was predictive of RCIN incidence. This review[63] found a significant difference in RCIN rates between iopamidol and iohexol (11.3% vs 21.6%, P = 0.0001), and between iodixanol and iohexol (9.5% vs 21.6%, P < 0.0001). A multivariate logistic regression model showed that the risk of RCIN was similar with the IOCM iodixanol and the LOCM iopamidol. Heinrich et al[64] have found that iodixanol is not associated with a significantly reduced risk of RCIN compared with the LOCM. However, in patients with intra-arterial administration and renal insufficiency, iodixanol is associated with a reduced risk of RCIN compared with iohexol, whereas no significant difference between iodixanol and other LOCM was found. Reed et al[32] have compared iodixanol to several LOCM and found no difference in the incidence of RCIN when they compared iodixanol to all LOCM pooled together. However, iodixanol was associated with a lower incidence of RCIN compared with ioxaglate and iohexol, but not when compared with other LOCM. In the most recent meta-analysis, which included 36 trials encompassing 7166 patients, From et al[65] compared iodixanol to several LOCMs. They found that iodixanol had no statistically significant reduction in RCIN incidence below that observed with heterogeneous comparator agents (P = 0.11). Analysis of patient subgroups has revealed that there was a significant benefit of iodixanol when compared with iohexol alone (OR: 0.25, 95% CI: 0.11-0.55, P < 0.001) but not when compared with iopamidol. These results suggest that the LOCM agents cannot be thought of as a class when it comes to renal tolerability, and that the potential benefit ascribed to IOCM has been overestimated based on earlier trials[36]. These studies suggest a lower incidence of RCIN with the use of iodixanol compared with specific low-osmolality agents, namely iohexol and possibly ioxaglate, with no discernible difference when iodixanol is compared with iopamidol. On the basis of these data, guidelines from the American College of Cardiology/American Heart Association recommend the use of IOCM or LOCM other than iohexol and ioxaglate in patients with CKD undergoing angiography[44,67].

In summary, these data support a benefit of IOCM iodixanol compared to specific LOCM agents such as iohexol and ioxaglate among patients with CKD undergoing angiography, but not a benefit for iodixanol compared to other nonionic low osmolar agents. We suggest the use of either an iso-osmolal contrast agent or a low-molecular-weight contrast agent other than iohexol or ioxaglate.

TEMPORARY PROPHYLACTIC RRT

Hemofiltration and hemodialysis

Another important approach for prevention of RCIN is the early initiation of RRT during or after the administration of contrast. The use of RRT to prevent RCIN is predicated on the premise that rapid removal of iodinated radiocontrast material from the circulation, limiting the filtered load at the glomeruli, will decrease the risk of renal injury. Although contrast can be effectively eliminated by hemodialysis and hemofiltration[68,69], it is still controversial whether RRT is able to reduce the incidence of RCIN. Several RCTs have investigated this issue[69-79], but the results have been inconsistent.

In an effort to reconcile the disparate clinical trial findings, systematic reviews and meta-analyses have been performed to analyze the collective results of relevant studies[80,81]. Cruz et al[80] conducted a meta-analysis of blood purification therapies for the prevention of RCIN. Considering data from eight clinical trials, six of which assessed hemodialysis and two of which assessed continuous RRT, the authors found that RRT did not reduce the incidence of RCIN compared with routine preventive care. Moreover, there was considerable inter-trial heterogeneity. In sensitivity analyses that included only those studies of hemodialysis, inter-trial heterogeneity was not statistically significant, yet there was a trend toward greater risk for RCIN with hemodialysis compared with standard preventive care[80].

In a recently published meta-analysis, Song et al[81] compared the different modes of RRT in assessing the efficacy of prophylactic RRT on RCIN in 751 patients. They found considerable heterogeneity across trials (P < 0.00001). RRT reduced the risk of RCIN by 26% compared with the control group by saline infusion, but statistical significance was not reached (risk ratio, RR: 0.74, 95% CI: 0.35-1.60, P = 0.45). Subgroup analysis of modality indicated that hemodialysis was ineffective in reducing the risk of RCIN (RR: 1.21, 95% CI: 0.63-2.32, P = 0.57), while continuous RRT decreased the incidence of RCIN (RR: 0.22, 95% CI: 0.07-0.64, P = 0.006). They also analyzed the effects in the subgroups of baseline CKD stage 3 and higher. Interestingly, heterogeneity was not found in these subgroups. Patients in the studies of Marenzi et al[75,76] and Lee et al[77] represented a more severely ill population compared with other trials (CKD stage 4/5 vs stage 3). Subgroup analysis according to the CKD stage did not record heterogeneity across trials. When analysis was restricted to studies involving CKD stage 3 patients, they recorded a significant increase in relative risk of hemodialysis (RR: 1.53, P = 0.01). This finding indicated that hemodialysis was ineffective, or even harmful for prevention of RCIN in CKD stage 3 populations. However, analysis of trials with patients involving CKD stage 4/5 revealed an overwhelming favorable effect of RRT over standard treatment in reducing the incidence of RCIN (RR: 0.19, P < 0.001)[80]. Clinical trials comparing prophylactic RRT and control group for RCIN after coronary angiography procedure (baseline CKD stage 3) are shown in Table 2. Clinical trials comparing prophylactic RRT and control group for RCIN after coronary angiography procedure (baseline CKD stage > 3) are shown in Table 3.

Table 2.

Clinical trials comparing prophylactic renal replacement therapy and control group for radiocontrast-induced nephropathy after coronary angiography procedure (baseline chronic kidney disease stage 3)

Ref.	Time from contrast exposure to the start of RRT (modes of RRT and duration)	No. of patients (RRT:control)	Incidence of RCIN results (RR, 95% CI)	Permanent dialysis rate of RCIN	In-hospital mortality of RCIN
Lehnert et al[70] (1998) diagnostic procedures	63 ± 6 min (HD 3 h)	30 (15:15)	8/15 vs 6/15 (RR = 1.33, 0.61-2.91)	NA	NA
Sterner et al[72] (2000) diagnostic procedures	< 3 h (HD 3 h)	32 (15:17)	6/15 vs 4/17 (RR = 1.70, 0.59-4.90)	NA	NA
Berger et al[71] (2001) diagnostic procedures	106 ± 25 min (HD 2–3 h)	15 (7:8)	3/7 vs 1/8 (RR = 3.43, 0.45-25.93)	NA	NA
Vogt et al[73] (2001) diagnostic procedures	2 h (HD 3 h)	113 (55:58)	24/55 vs 20/58 (RR = 1.27, 0.80-2.01)	3/55 vs 2/58 (RR = 1.58, 0.27-9.11)	1/55 vs 1/58 (RR = 1.06, 0.06-17.30)
Frank et al[74] (2003) diagnostic procedures	0 (HD 4 h)	17 (7:10)	NA	2/7 vs 2/10 (RR = 1.43, 0.26-7.86)	NA
Reinecke et al[78] (2007) diagnostic procedures	< 20 min (HD 4 h)	273 (135:139)	22/135 vs 10/138 (RR = 2.28, 1.12-4.64)	2/135 vs 1/137 (RR = 2.03, 0.19-22.12)	3/135 vs 3 137 (RR =1.02, 0.20-5.12)

RRT: Renal replacement therapy; HD: Hemodialysis; CVVH: Continuous venovenous hemofiltration; RCIN: Radiocontrast-induced nephropathy; CKD: Chronic kidney disease; RR: Relative risk; 95% CI: 95% confidence interval; NA: Not assessed.

Table 3.

Clinical trials comparing prophylactic renal replacement therapy and control group for radiocontrast-induced nephropathy after coronary angiography procedure (Baseline chronic kidney disease stage 4-5)

Ref.	Time from contrast exposure to the start of RRT (modes of RRT and duration)	CKD stage, No. of patients (RRT:control)	Incidence of RCIN results (RR, 95% CI)	Permanent dialysis rate of RCIN	In-hospital mortality of RCIN
Marenzi et al[75] (2003) interventional procedures	0 (CVVH 22-30 h)	Stage 4, 114 (58:56)	4/58 vs 32/56 (RR = 0.12, 0.05-0.32)	2/58 vs 11/56 (RR = 0.18, 0.04-0.76)	1/58 vs 8/56 (RR = 0.11, 0.01-0.87)
Marenzi et al[76] (2006) diagnostic and interventional procedures	0 (CVVH 18-36 h)	Stage 4, 92 (62: 30)	9/62 vs 12/30 (RR = 0.36, 0.17-0.77)	NA	3/62 vs 6/30 (RR = 0.20, 0.05-0.88)
Lee et al[77] (2007) diagnostic procedures	81 ± 32 min (HD 4 h)	Stage 5, 82 (42:40)	2/42 vs 18/40 (RR = 0.11, 0.03-0.43)	0/42 vs 5/40 (RR = 0.09, 0-1.52)	No

RRT: Renal replacement therapy; HD: Hemodialysis; CVVH: Continuous venovenous hemofiltration; RCIN: Radiocontrast-induced nephropathy; CKD: Chronic kidney disease; RR: Relative risk; 95% CI: 95% confidence interval; NA: Not assessed.

In summary, considering the greater cost and risk of RRT, temporary prophylactic hemodialysis or continuous RRT are not indicated for the prevention of contrast nephropathy in patients with stage 3 CKD. Although more data are needed in CKD, we consider the prophylactic use of hemodialysis in patients with stage 4/5 CKD when the functioning access is already available.

PHARMACOLOGICAL INTERVENTIONS

The mechanisms of pharmacological prophylaxis for CIN include antioxidant strategy, inhibition of renal vasoconstriction, and combination of these two effects. Currently, there are no approved pharmacological agents for the prevention of RCIN. Trials of pharmacological interventions, including furosemide, dopamine, fenoldopam, calcium channel blockers, and mannitol, have failed to demonstrate significant benefit for the prevention of RCIN and in some cases have been associated with harm[82-84]. Findings on the benefit of N-acetylcysteine (NAC), ascorbic acid (vitamin C), and statins are discussed below.

NAC for RCIN prevention

The rationale for the use of NAC for the prevention of RCIN relates to its capacity to scavenge reactive oxygen species, reduce the depletion of glutathione, and stimulate the production of vasodilatory mediators, including NO[18]. However, there has been ongoing debate over whether NAC is effective in preventing RCIN[18,29,36]. NAC is a potent antioxidant that scavenges a wide variety of oxygen-derived free radicals, and it may be capable of preventing RCIN by improving renal hemodynamics and by diminishing direct oxidative tissue damage.

Oral NAC treatment

The value of NAC for RCIN prevention has been the focus of many studies. The first study by Tepel et al[85] reported in 83 patients with CKD (serum creatinine > 1.2 mg/dL) undergoing applied computed tomography scanning with small amounts (75 mL) of LOCM. Administration of oral NAC at 600 mg twice daily on the day before and the day of the procedure (total dose: 2.4 g), in addition to an infusion of hypotonic saline, reduced the incidence of RCIN 10-fold[85]. Since the first publication of this initial study, many trials evaluating NAC for the prevention of RCIN have been performed and published in the literature but yielded highly conflicting results[85-100]. The clinical trials showing benefit of oral NAC for RCIN after angiography are shown in Table 4. Briguori et al[87] have concluded that the use of a double dose of NAC seems to be more protective in preventing contrast-induced renal dysfunction, especially in patients with high volumes of contrast medium. Similarly, in Marenzi’s study[93], a 2.6-fold lower incidence of RCIN was shown in patients with ST-elevation acute myocardial infarction who were treated with isotonic saline and 1.2 g IV bolus of NAC, followed by 1.2 g oral NAC after emergency coronary intervention compared with patients receiving 600 mg IV bolus of NAC before and 600 mg oral NAC after procedure. In contrast, the renoprotective effect of NAC was not supported by some other studies of orally administered NAC[94-100].

Table 4.

Clinical trials showing benefit of oral N-acetylcysteine for radiocontrast-induced nephropathy after angiography

Ref.	NAC dosing regimen (cumulated NAC dose)	No. of patients (NAC:control)	Hydration protocol	Contrast media type	Results
Shyu et al[88] (2002) coronary angiography	400 mg po bid before and after the procedure (1.6 g)	121 (60:61)	0.45% saline for 12 h pre- and 12 h postprocedure	LOCM iopamidol	3.3% vs 24.6% (P < 0.001)
Diaz-Sandoval[89] (2002) coronary angiography	600 mg po bid × 2, 1 dose before and 3 dose after the procedure (2.4 g)	54 (25:29)	0.45% saline for 2-12 h pre-and 12 h postprocedure	LOCM ioxilan	8.0% vs 45% (P = 0.005)
Kay et al[86] (2003) coronary angiography	600 mg po bid × 2, before and after the procedure (2.4 g)	200 (102:98)	0.9% saline for 12 h pre- and 6 h postprocedure;	LOCM iopamidol	3.9% vs 12.2% (P = 0.03)
MacNeill et al[90] (2003) coronary angiography	600 mg twice daily × 5 doses (3 g)	43 (21:22)	0.45% saline at a rate of 1 ml/kg per hour for 12 h for in-patients and 2 mL/kg per hour for 4 h for daycare patient	LOCM iopromide	5% vs 32% (P = 0.046)
Efrati et al[91] (2003) coronary angiography.	1000 mg po bid × 2, before and after the procedure (4 g)	49 (24:25)	0.45% saline hydration 1 mL/kg per hour for 12 h before and 12 h after coronary angiography	LOCM iopromide	0% vs 8%
Miner et al[92] (2004) coronary angiography	2000 mg po either 2 or 3 (4 g or 6 g)	180 (95:85)	0.45% intravenous saline	LOCM iohexol	9.6 % vs 22.2% (P = 0.04)
Briguori et al[87] (2004) cardiac/peripheral angiography	standard-dose 600 mg bid × 2 (2.4 g)	224 (110:114)	0.45% saline hydration 1 mL/kg per hour for 12 h before and 12 h after angiography	LOCM iobitridol	11.0% vs 3.5% (P = 0.038)

NAC: N-acetylcysteine; LOCM: Low-osmolar contrast media.

IV NAC treatment

NAC for RCIN prevention is usually started the day before the procedure requiring contrast agent, thus, it is not possible in situations requiring urgent catheterization[101]. Therefore, IV NAC treatment is suggested as a possible alternative in situations where oral NAC is unable to be given in advance. Another reason is the controversial data obtained from orally administered NAC[102]. As in the studies with oral NAC, the results with IV NAC have also been inconsistent[102-110]. The clinical trials comparing IV NAC and control groups for RCIN after angiography are shown in Table 5.

Table 5.

Clinical trials comparing IV N-acetylcysteine and control after angiography procedure

Ref.	NAC dosing regimen (cumulated NAC dose)	No. of patients (NAC:hydration)	Hydration protocol	Contrast media type (mean dose)	Results
Baker et al[103] (2003) coronary angiography	150 mg/kg over 30 min before and 50 mg/kg over 4 h after (200 mg/kg)	41:39	0.9% NaCl 1 mL/kg per hour for 12 h pre- and post-procedure	IOCM iodixanol (253 mL)	5% vs 21% (RR = 0.28, P = 0.045)
Kefer et al[105] (2003) coronary angiography	1200 mg 12 h before 1200 mg immediately after the procedure (2.4 g)	53:51	0.9% NaCl 200 mL 12 h pre- and D5W 20 mL/h for 12 h pre and post-procedure	LOCM iopromide/iohexol (199 mL)	3.8% vs 5.9 % (P = 0.98)
Rashid et al[107] (2004) peripheral angiography	2 doses of 1000 mg at 6-12 h before and after the procedure (2 g)	46:48	2 doses of 0.9% NaCl 500 mL over 4-6 h at 6-12 h before and after the procedure	LOCM iohexol 143 mL)	pts. CrCl < 70 mL/min 7.7% vs 8.8% (P = 1.0)
Webb et al[109] (2004) coronary angiography	500 mg in D5W/0.9% NaCl 50 mL for 15 min as a bolus within 1 h before the procedure (500 mg)	242:245	D5W/0.9% NaCl 50 mL as a bolus within 1 h before the procedure and 200 mL 0.9% NaCl	LOCM ioversol (136 mL)	23.3% vs 20.7% (P = 0.51)
Kotlyar et al[110] (2005) cardiac or peripheral angiography	Gr 1: 300 mg × 2 (600 mg)	41:19	0.9% NaCl 1 mL/kg per hour	LOCM iopromide	0% vs 0 %
	Gr 2: 600 mg × 2 (1.2 g)
Carbonell et al[104] (2007) coronary angiography	600 mg twice daily (2.4 g)	107:109	0.45% intravenous saline	LOCM iopromide (193 mL)	10.3% vs 10.1% (P = 0.5)
Koc et al[102] (2010) coronary angiography and/or PCI	600 mg twice daily before and on the day of the coronary procedure (total = 2.4 g)	80:801:602	0.9% saline	LOCM iohexol 138 ± 47 mL)	2.5% vs 16.3% vs 10.0% (P = 0.012)

¹High-dose hydration group: IV 0.9% saline 1 mL/kg per hour before, on and after the day of coronary procedure (5334 ± 783 mL);

²Standard hydration group: IV 0.9% saline 1 mL/kg per hour for 12 h (1893 ± 270 mL). NAC: N-acetylcysteine; RCIN: Radiocontrast induced nephropathy; RR: Relative risk; CrCl: Creatinine clearance; NaCl: Sodium chloride; D5W: 5% dextrose; IOCM: Iso-osmolar contrast media; LOCM: Low-osmolar contrast media; PCI: Percutaneous coronary intervention.

Systemic reviews and meta-analyses of the efficacy of NAC

A variety of factors may contribute to these inconsistencies of efficacy of NAC to prevent contrast-induced AKI. These may include the definition of contrast-induced acute renal failure, baseline risk for acute renal failure (e.g., severity of renal dysfunction, heart failure, and proportion with diabetes), NAC dose and route of administration (e.g., oral or IV), IV hydration protocols, amount and type of contrast given, and type of procedure performed[21,111,112].

The disparate results of these clinical trials have led to a proliferation of systematic reviews and meta-analyses comparing the overall prophylactic efficacy of NAC. To date, at least nine meta-analysis of NAC have shown beneficial treatment effects in reducing RCIN[113-121]. However, six meta-analyses were inconclusive[122-127]. In the largest meta-analysis that included 41 studies and encompassed 3393 patients, Kelly et al[118] found that oral or IV NAC significantly lowered the risk for RCIN by 38% when compared with hydration controls with saline alone (RR: 0.62, 95% CI: 0.44-0.88). Trivedi et al[121] conducted a meta-analysis of 16 trials that utilized high-dose NAC, defined as a daily dose > 1.2 g or one dose of > 600 mg within 4 h of contrast administration with a sample size of 1677 patients. Their results suggested that high-dose NAC was associated with a lower risk of RCIN compared with controls.

In summary, clinical data regarding the efficacy with NAC for RCIN prevention remains debatable. However, considering the very low toxicity and cost of this drug we recommend the use of oral NAC at a dose of 1.2 g twice daily on the day before and day of the procedure to patients at risk for contrast nephropathy.

Ascorbic acid for RCIN prevention

The rationale for the use of vitamin C for the prevention of RCIN relates to its antioxidant property; it has been shown to ameliorate renal damage in experimental postischemic stress, cisplatin, and aminoglycosides injury[128,129].

Spargias et al[130] conducted a randomized, double-blind, placebo-controlled trial of ascorbic acid in 231 patients with serum creatine > 1.2 mg/dL who underwent coronary angiography and/or intervention. Ascorbic acid, 3 g at least 2 h before the procedure and 2 g in the night and the morning after the procedure, or placebo was administered orally. RCIN occurred in 11 of the 118 patients (9%) in the ascorbic acid group and in 23 of the 113 patients (20%) in the placebo group (OR: 0.38, P = 0.02). However, in another study by Boscheri et al[131], 143 consecutive patients with CKD (serum creatinine > 120 μmol/L) referred for coronary angiography intervention were randomly assigned to receive 1 g ascorbic acid or placebo in adjunct to saline hydration before and after angiography. They found there was no significant difference between the two groups (vitamin C 6.8%, placebo 4.3%, P < 0.05). No patient required dialysis. The authors concluded the prophylactic use of ascorbic acid in patients with renal dysfunction exposed to contrast dye is not justified. In the most recent randomized, double blinded, placebo-controlled, single-center study by Brueck et al[132], the prophylactic administration of the antioxidants NAC or ascorbic acid, along with prehydration, was not associated with a significantly reduced risk of RCIN compared with placebo in patients with chronic renal insufficiency. Thus, the role of ascorbic acid for RCIN prevention remains unclear.

Statins for RCIN prevention

The rationale for the use of statins for the prevention of RCIN relates to its antioxidative and anti-inflammatory properties. Given the potential role of oxidative stress in the pathophysiology of RCIN, statins might reduce contrast media nephrotoxicity by removing free radicals[133-135].

The possibility that statins might be beneficial in reducing the incidence of contrast-induced acute renal failure has been examined in several observational studies[136-138] and RCTs[139-142]. The earliest study by Attallah et al[130] reviewed a database of 1002 patients with renal insufficiency undergoing coronary angiography. Incidence of RCIN in patients receiving simvastatin or atorvastatin 24-72 h before catheterization was significantly lower than in those not receiving it. Khanal et al[137] similarly reported a significantly lower incidence of RCIN, based on a large database review of 29 409 patients undergoing emergency and non-emergency PCI, among patients receiving statins. Patti et al[138] prospectively evaluated 434 patients undergoing PCI to determine statin benefit in prevention of RCIN and long-term outcomes over 4 years. Statin-pretreated patients had a significantly lower incidence of RCIN and improved long-term outcomes including a significant decrease in cardiac death at 4 years. However, in the first prospective, randomized, double-blind, controlled study (PROMISS trial) to evaluate the use of a statin for RCIN prevention, short-term simvastatin pretreatment at high dose did not prevent renal function deterioration after administration of contrast medium in patients with baseline renal insufficiency undergoing coronary angiography[139]. Zhang et al[143] performed a systematic review and meta-analysis of published human cohort studies and RCTs to determine whether the administration of statins is protective against RCIN and to assess the magnitude of their effect on RCIN. In this most recently published systemic review, the investigators performed qualitative analysis of the cohort studies and quantitative analysis of the RCTs to estimate the pooled RRs for preventive effect of statins. Among six cohort studies, four showed chronic statin pretreatment had a preventive effect against RCIN. From six RCTs, 1194 patients were included in the meta-analysis. Under the fixed-effects model, an insignificant protective trend toward decreased incidence of RCIN with periprocedural short-term high-dose statin treatment was seen (RR: 0.70, 95% CI: 0.48-1.02). Current data are not conclusive as to whether statins are protective for RCIN due to the inherent limitations of the included studies[144]. In the future, large well-designed studies are needed to address the effect of this drug and its longer-term clinical outcomes.

Periprocedural hydration for preventing RCIN

The rationale for the prevention of RCIN by periprocedural hydration is through blocking its two complementary pathophysiological processes[144]. First, expansion of the intravascular space is thought to blunt the vasoconstrictive effect of contrast on the renal medulla. Second, intravascular fluids are believed to attenuate the direct toxic effect of contrast agents on tubular epithelial cells. The optimal hydration solution to prevent contrast nephropathy is unclear.

The optimal method and composition of the fluid administered remains to be established. To date, there are four types of periprocedural hydration: oral fluids, IV 0.45% saline, IV normal saline, and IV sodium bicarbonate. The positive effect of adequate periprocedural hydration in reducing rates of RCIN was first established in a randomized study by Solomon et al[83] in 1994. In that study, 78 patients with stable chronic renal failure (mean serum creatinine concentration 2.1 mg/dL) about to undergo coronary angiography were randomized to one of three regimens: (1) 0.45% (half-isotonic) saline at a rate of 1 mL/kg per hour for 12 h before and 12 h after the angiogram; (2) 0.45% saline plus 25 g of mannitol infused intravenously during the 1 h before the procedure; and (3) 0.45% saline plus 80 mg furosemide infused intravenously during the 30 min before angiography. The incidence of acute renal failure (defined as an increase in serum creatinine of at least 0.5 mg/dL) was lowest in the group treated with saline alone. However, this trial did not include a control group of patients who did not receive IV fluid. In another randomized study by Trivedi et al[145] who demonstrated a significantly lower incidence of RCIN in patients undergoing non-emergency coronary angiography who received IV normal saline (1 mL/kg per hour for 24 h starting 12 h before contrast exposure) compared with a protocol of unrestricted oral fluids (3.7% vs 34.6%, P < 0.005). Trivedi’s study showed IV normal saline was better than unrestricted oral fluids. The randomized comparison of two hydration regimens in a total of 1620 patients undergoing PCI in a study by Mueller et al[146] showed the superiority of isotonic vs half-isotonic saline in reducing RCIN (0.7% vs 2%, respectively). Mueller’s study showed the effect of tonicity of IV fluids on the development of RCIN. On the basis of this study, it has generally been accepted that isotonic saline is superior to hypotonic saline for the prevention of RCIN.

Sodium bicarbonate for RCIN prevention

One major underlying hypothesis for application of sodium bicarbonate is that the alkalinization of tubular fluid diminishes the production of free oxygen radicals and protects the kidney from oxidant injury[39,147]. Pretreatment with sodium bicarbonate is more protective than sodium chloride in animal models of acute ischemic renal failure[148]. Based on the above theory, the first study of sodium bicarbonate for RCIN prevention by Merten et al[149] was done in 2004. This study involved 119 patients with stable serum creatinine levels (≥ 1.1 mg/dL) scheduled for contrast administration (coronary angiography, computed tomography, or radiographic procedures) and randomized to receive 3 mL/kg per hour 5% dextrose with 154 mEq/L

sodium chloride or 5% dextrose with 154 mEq/L bicarbonate for 1 h before contrast administration, followed by an infusion at 1 mL/kg per hour for 6 h after contrast administration. The incidence of RCIN was 1.7% in patients receiving bicarbonate, compared with 13.6% (P = 0.02) in patients receiving saline[149].

Since the first publication of this initial study by Merten et al[149], many trials evaluating IV sodium bicarbonate for the prevention of RCIN have been performed and published in the literature, but yielded highly conflicting results[53,150-158]. Over the past 8 years, at least 12 clinical trials comparing the development of RCIN following the administration of either isotonic bicarbonate or isotonic saline have been published in the peer-reviewed literature; eight demonstrating a lower incidence of RCIN with bicarbonate administration[108,149-155], and four showing no significant benefit[53,156-158]. The clinical trials showing benefit of bicarbonate over saline for RCIN are shown in Table 6.

Table 6.

Clinical trials showing benefit of IV bicarbonate over saline to prevent radiocontrast-induced nephropathy after angiography

Ref.	Inclusion criteria	No. of patients	Hydration protocol	Contrast media type	Results: RCIN in bicarbonate group vs saline group	Dialysis and death rate
Merten et al[149] (2004) CT/coronary angiography	SCr ≥ 1.1 mg/dL	119	0.9% NaCl 1 mL/kg per hour for 12 h pre- and post-procedure	Iopamidol	1.7% vs 13.6% (P = 0.02)	Dialysis rate 0%
Ozcan et al[152] (2007) coronary angiography/PCI	SCr > 1.2 mg/dL	264	0.9% NaCl 200 mL 12 h pre- and D5W 20 mL/h for 12 h pre and post-procedure	Ioxaglate	4.5% vs 13.6% (P = 0.036)	Dialysis rate 1%
Briguori et al[150] (2007) coronary/peripheral angiography	SCr ≥ 2.0 mg/dL or eGFR < 40 mL/min per 1.73 m2	326	0.9% NaCl 500 mL over 4-6 h at 6-12 h before and after the procedure	Iodixanol	1.9% vs 9.9% (P = 0.019)	Dialysis rate 1%
Masuda et al[151] (2007) emergency coronary angiography/PCI	SCr ≥ 1.1 mg/dL or eGFR < 60 mL/min	59	D5W/0.9% NaCl 50 mL as a bolus within 1 h before the procedure and 200 mL 0.9% NaCl	Iopamidol	6.7% vs 34.5% (P = 0.01)	Dialysis rate 7%; death rate 3%
Recio-Mayoral et al[108] (2007) emergency PCI	None	111	Sodium bicarbonate 5 mL/kg per hour 1 h before the procedure and 1.5 mL/kg per hour for 12 h after the procedure	Iomeprol	1.8% vs 21.8% (P < 0.001)	Dialysis rate 4%; death rate 4.5%
Pakfetrat et al[153] (2009) coronary angiography/PCI	SCr > 1.2 mg/dL	192	Bicarbonate in dextrose infusion, normal saline infusion alone or combined with oral acetazolamide before procedure	Iodixanol	4.2% vs 12.5% (P < 0.001)	NA
Tamura et al[154] (2009) elective coronary angiography	SCr > 1.1 to < 2.0 mg/dL) undergoing an elective coronary	144	Single-bolus intravenous administration of sodium bicarbonate (20 mEq) immediately before contrast exposure	Iopamidol	10.3% vs 10.1% (P = 0.5)	NA
Ueda et al[155] (2011) emergent coronary procedures	SCr > 1.2 mg/dL	59	A bolus intravenous injection of 154 mEq/L of sodium bicarbonate or saline at the dose of 0.5 mL/kg, before CM, followed by infusion of 154 mEq/L sodium bicarbonate at 1 mL/kg per hour for 6 h in both groups	Iopamidol	3.3% vs 27.6% (P = 0.01)	NA

RCIN: Radiocontrast induced nephropathy; NaCl: Sodium chloride; D5W: 5% dextrose; PCI: Percutaneous coronary intervention; NA: Not assessed.

In the most recently published study by Ueda et al[155], 59 patients with mild CKD were scheduled at admission to undergo an emergency coronary procedure. The patients were randomized to receive a bolus IV injection of 154 mEq/L sodium bicarbonate (n = 30) or sodium chloride (n = 29) at a dose of 0.5 mL/kg, before contrast administration, followed by infusion of 154 mEq/L sodium bicarbonate at 1 mL/kg per hour for 6 h in both groups. In the sodium bicarbonate group, the serum creatinine concentration remained unchanged within 2 d of contrast administration (from 1.32 ± 0.46 to 1.38 ± 0.60 mg/dL, P = 0.33). In contrast, it increased in the sodium chloride group (1.51 ± 0.59 to 1.91 ± 1.19 mg/dL, P = 0.006). The incidence of RCIN was significantly lower in the sodium bicarbonate group than in the sodium chloride group (3.3% vs 27.6%, P = 0.01).

In contrast to the above findings, several recently published studies have shown no significant benefit of isotonic bicarbonate over saline for lowering incidence of RCIN[53,156-158]. In the largest randomized trial to date by Maioli et al[157], comparing isotonic saline (1 mL/kg per hour, started 12 h before and extended 12 h after the procedure) against isotonic sodium bicarbonate in dextrose (3 mL/kg per hour for 1 h before, followed by 1 mL/kg per hour for 6 h after contrast administration) given in addition to NAC in 502 CKD patients (eGFR < 60 mL/min) scheduled for non-emergency coronary angiography, there was no superiority of either treatment modality[157]. In addition, in a large retrospective analysis of the general population of patients undergoing coronary angiography, use of sodium bicarbonate alone was associated with an increased risk of contrast nephropathy compared with no treatment; whereas NAC alone or in combination with sodium bicarbonate was not associated with any significant difference in the incidence of contrast nephropathy[159].

Systematic reviews and meta-analyses comparing the effectiveness of bicarbonate and saline

The optimal hydration solution to prevent contrast nephropathy is unclear and the disparate results of these clinical trials, therefore, many systematic reviews and meta-analyses have been performed to analyze these collective results[160-171]. Of the 12 published meta-analyses, most have suggested a significant benefit of using sodium-bicarbonate-based hydration for prophylaxis of RCIN[160-169], although the magnitude of the benefit may have been overestimated by earlier studies. Two meta-analyses have shown no significant benefit[170,171]. Meta-analyses comparing the effectiveness of bicarbonate and saline to prevent RCIN are shown in Table 7.

Table 7.

Meta-analysis comparing the effectiveness of bicarbonate and saline to prevent radiocontrast-induced nephropathy

Ref.	Year of publication	No. of patients	No. of trials	Relative risk (95% CI) of RCIN of bicarbonate therapy compared with saline	Study heterogeneity and publication bias
Joannidis et al[160]	2008	2043	9	0.45 (0.26-0.79)	Heterogeneity detectable and publication bias was present
Hogan et al[161]	2008	1307	7	0.37 (0.18-0.714)	Evidence of heterogeneity
Ho et al[162]	2008	573	4	0.22 (0.11-0.44)	Significant heterogeneity
Meier et al[163]	2009	2633	17	0.52 (0.34-0.80)	Evidence of heterogeneity and publication bias
Navaneethan et al[166]	2009	1854	12	0.46 (0.26-0.82)	Heterogeneity and publication bias were detectable
Kanbay et al[165]	2009	2448	17	0.54 (0.36-0.83)	There are study heterogeneity and publication biases
Zoungas et al[168]	2009	3563	23	0.62 (0.45-0.86)	Evidence of heterogeneity and publication bias was present
Hoste et al[167]	2010	3055	18	0.66 (0.45-0.95)	Evidence of heterogeneity and publication bias was present
Trivedi et al[171]	2010	1090	10	0.57 (0.38-0.85)	No evidence of heterogeneity and no publication bias
Kunadian et al[170]	2011	1734	7	0.33 (0.16-0.69)	Heterogeneity and publication bias were detectable
Brown et al[164]	2009	1994	10	0.65 (0.40-1.05)	Significant heterogeneity
Brar et al[169]	2009	2290 (large trials, 1145; small trials, 1145)	14 (large trials 3; small trials 11)	large trials: 0.85 (0.63-1.16); small trials: 0.50 (0.27-0.93)	Evidence of publication bias; heterogeneity accounted for by trial size

RCIN: Radiocontrast-induced nephropathy; 95% CI: 95% confidence interval.

Hoste et al[167] completed a comprehensive meta-analysis including 18 trials with a total of 3055 patients. The aggregate result demonstrated a benefit favoring sodium bicarbonate (RR: 0.66, 95% CI: 0.45-0.95,). This effect was most prominent in coronary procedures and in patients with CKD. In their subgroup analysis of nine published papers and nine unpublished abstracts, they found published papers demonstrated a beneficial effect, while abstracts did not. They also detected significant clinical and statistical heterogeneity between studies.

In another recent meta-analysis of 14 trials, encompassing 2290 patients, Brar et al[169] also found significant heterogeneity among the trials (P heterogeneity = 0.02, I² = 47.8%), and they found the heterogeneity was largely accounted for by trial size (P = 0.016). Therefore, the authors segregated the trials into two groups. Three trials were categorized as large (n > 1145) and 11 as small (n < 1145). Among the large trials, the incidence of RCIN for sodium bicarbonate and sodium chloride was 10.7 and 12.5%, respectively; the RR and 95% CI was 0.85 and 0.63-1.16, without evidence of heterogeneity (P = 0.89, I² = 0%). However, the pooled RR among the 11 small trials was 0.50 (95% CI: 0.27-0.93) with significant between-trial heterogeneity (P = 0.01, I² = 56%) and the small trials were more likely to be of lower methodological quality. As a result, the authors concluded there was no evidence of benefit for hydration with sodium bicarbonate compared with sodium chloride for the prevention of RCIN. The benefit of sodium bicarbonate was limited to small trials of lower methodological quality.

In summary, at this time, data on the comparative effectiveness of bicarbonate and saline for the prevention of RCIN are insufficient to warrant a recommendation for the routine use of a specific isotonic IV fluid. However, in the clinical setting when patients requiring emergency coronary procedures, there would not be enough time to administer sufficient IV fluids for hydration. The idea of a single bolus of sodium bicarbonate may be helpful.

CONCLUSION

RCIN is a predictable and perhaps partially preventable complication. Reasonable steps should be taken to minimize risk. Novel diagnostic and therapeutic approaches are needed to manage the ever-increasing numbers of patients with baseline renal dysfunction undergoing coronary angiography. Clinicians should carefully evaluate the risks and benefits of prophylactic measures and apply them to individual patients. Large clinical trials that are based on clinically plausible effect sizes and examine serious, adverse, patient-centered outcomes are needed to define better the clinical utility of pharmaceutical agents for the prevention of this complication. Our current recommendations can be summarized as follows: (1) all patients should be stratified for RCIN risk prior to contrast exposure; (2) temporary withdrawal of the drugs which may affect the renal function; (3) dose of contrast medium should be limited to the minimum volume required to provide adequate clinical information; (4) high-risk patients should receive prophylaxis for modifiable factors; (5) consider using iso-osmolar or a nonionic, LOCM other than iohexol or ioxaglate; (6) unless contraindicated, consider using oral high-dose NAC at 1.2 g bid on the day before and the day of contrast exposure; (7) using periprocedural hydration, if no evidence of frank heart failure, either by isotonic saline at a dose of 1 mL/kg per hour for 12 h preceding and 12 h following the administration of contrast medium, or 154 mEq/L bicarbonate at a dose of 3.0 mL/kg for 1 h preceding and 1 mL/kg per hour for 6 h after contrast administration; (8) in an emergency setting, where preparation of the patient with IV hydration is not feasible, consider administration of isotonic sodium bicarbonate with high-dose oral NAC (2.4 g) 1 h before contrast; (9) consider the prophylactic use of hemodialysis in patients with stage 5 CKD, provided that a functioning access is already available; and (10) follow-up the serum creatinine 24-72 h after contrast exposure in high-risk patients.