Contrast-induced nephrotoxicity (CIN) is a significant cause of morbidity and mortality [1]. Although the incidence in the general population is reported to be less than 2% [2], in high-risk patients, namely those with both diabetes and pre-existing renal impairment, the reported incidence ranges from 9% to 50% [2–5].

The mechanism of renal injury is poorly understood and is likely to be multifactorial. Experimental models have demonstrated that following the injection of radiographic contrast media there is transient intrarenal vasodilatation followed by a longer period of vasoconstriction [6–8]. The resulting reduction in glomerular filtration rate (GFR) is compounded by a further response likely to be mediated by oxygen free radicals and resulting in direct cellular damage [9–14]. Previous work by our group demonstrated a significant rise in urinary malondialdehyde (a urinary marker of free radical injury) concentration following infusion of low-osmolar radiographic contrast medium in humans [13].

The administration of antioxidants may therefore prevent CIN and against this background recent randomized controlled studies have compared the prophylactic administration of the antioxidant N-acetylcysteine with placebo in high-risk patients undergoing contrast-enhanced computed tomography [15] and coronary angiography [16–22]. The results of these individual studies are conflicting. Furthermore, three meta-analyses of the published literature have concluded a significant benefit in using prophylactic N-acetylcysteine in patients with chronic renal insufficiency [23–25] whilst others have failed to reach a definite conclusion [26–29].

The aims of this study were twofold: to determine whether the prophylactic use of N-acetylcysteine reduces the incidence of CIN in patients undergoing peripheral angiography, and to test the hypothesis that N-acetylcysteine prevents contrast-induced oxidative damage to the kidney, by measuring levels of urinary malondialdehyde (MDA).

Materials and Methods

Study Patients

Patients referred for noncoronary diagnostic angiography, i.e., head and neck, extremity and visceral abdominal angiography, with or without renal impairment, were considered eligible for the study. Patients with acute renal failure and patients with renal transplants were excluded from the study. All patients underwent diagnostic angiography using either a nonionic iso-osmolar or low-osmolar radiographic contrast agent.

The study protocol was approved by the Local Research and Ethics Committee and all patients gave written informed consent.

Study Protocol

The patients were randomly assigned, by selection of a sealed envelope, to receive N-acetylcysteine at a dose of 600 mg twice daily the day before the procedure and on the day of the procedure. The volume, concentration of iodine per milliliter, and type of contrast agent used varied according to the site of interest. Iopamidol, a low-osmolar non-ionic contrast medium (Niopam, Bracco) was used for lower limb angiography at an iodine concentration of 300 mg/ml, and in upper limb and mesenteric angiography at an iodine concentration of 370 mg/ml. Iodixanol, an iso-osmolar, non-ionic contrast medium (Visipaque, Amersham Health) was used for renal angiography at an iodine concentration of 320 mg/ml.

Serum creatinine was measured before the examination and 48 hr after the administration of the contrast medium. Renal impairment was defined as a serum creatinine concentration above 1.2 mg/dl (106 μmol/l). Urine samples were collected before and after the examination for measurement of MDA concentration. These data were correlated with the urine creatinine concentration. CIN was defined as an acute rise in the serum creatinine of at least 0.5 mg/dl (44 μmol/l) 48 hr after the administration of the radiographic contrast medium.

Statistical Analysis

Student’s t-test was used to compare the baseline characteristics and procedural details between the two groups. The t-test and Wilcoxon ranked nonparametric test were applied to the data relating to the serum and urinary variables.

Results

Study Patients

From October 2001 to October 2002, 116 patients were enrolled in the study. Of the 116 patients, 58 were randomized to receive prophylactic N-acetylcysteine and 58 were randomized to receive no treatment. Ten patients with incomplete study data were excluded from the analysis. The demographic data and cardiovascular history were similar between the two groups (Table 1). The proportion of patients with renal impairment was higher in the group receiving N-acetylcysteine: 50% versus 34%. The baseline mean serum creatinine concentration was similar between the two groups as was the estimated creatinine clearance calculated using the Cockcoft-Gault formula. There was no difference between the types of procedure or volume, strength and type of contrast used between the two groups (Table 2). Twenty-one patients (47%) with renal impairment received iso-osmolar contrast compared with 7 (11%) without renal impairment.

Table 1 Demographic and baseline characteristics of the patients
Full size table
Table 2 Procedural details of the patients
Full size table

The mean time to collection of the urine samples for MDA and creatinine analysis was similar between the two groups.

Effect on Serum Creatinine and Urinary Malondialdehyde Concentrations

The overall incidence of CIN was 2.8% (n = 3) and in patients with pre-existing renal impairment the incidence was 4.4% (n = 2). All 3 patients were in the group receiving N-acetylcysteine. Two of the patients had pre-existing renal impairment and one patient had normal renal function prior to angiography. There was no significant difference in serum creatinine concentration or urinary MDA/creatinine ratio between the two groups before or after angiography (Tables 3, 4).

Table 3 Changes in serum creatinine concentrations and MDA/Cr ratios in patients receiving N-acetylcysteine (mean ± SD)
Full size table
Table 4 Changes in serum creatinine concentrations and MDA/Cr ratios in patients not receiving N-acetylcysteine (mean ± SD)
Full size table

Discussion

This study demonstrates that the incidence of CIN in patients attending for diagnostic peripheral angiography is low, i.e., 2.8%. Even in patients with pre-existing renal impairment the incidence of CIN was 4.4%.

The reasons for the low incidence of CIN are unclear. CIN is thought to result from complex changes in intrarenal blood flow causing free radical formation and subsequent tubular damage. Many ways of preventing CIN have been tried, of which hydration [30], the use of low-osmolar and iso-osmolar contrast agents [4, 31–35] and prophylactic N-acetylcysteine [16–24] have been shown to be of benefit but the data are inconclusive.

In this study, intravenous hydration was not part of the study protocol in order to represent the reality of peripheral angiographic procedures, which are often performed as an emergency or as a day case.

The type of contrast agent may have influenced the incidence of CIN. The use of low-osmolar contrast agents is associated with a lower incidence of CIN [4, 31–35], although these studies still report an incidence of 12–50%. More promising results have been reported with the use of iso-osmolar contrast agents, which theoretically may not cause volume depletion and subsequent alteration in intrarenal blood flow [35, 36]. Aspelin et al. showed an increase in the post-angiography creatinine concentration of 0.5 mg/dl in 3% of patients receiving the iso-osmolar agent iodixanol, compared with 26% of patients receiving the low-osmolar agent iohexol [35]. Chalmers and Jackson also reported an incidence of CIN of 3.7% in patients receiving iodixanol compared with 10% in the iohexol group [36]. In the current study, 46.6% of patients with pre-existing renal impairment received iodixanol and the incidence of CIN was 4.4%. Furthermore, all 3 patients who developed CIN had received iopamidol. However, on subgroup analysis there was no difference in mean creatinine or MDA/Cr ratio after angiography according to the type of contrast agent used.

This is contrary to our previous findings in which infusion of intra-arterial iopamidol did result in a significant rise in urinary MDA levels suggesting that part of the injury in CIN is free-radical-induced [13]. On this basis the current study was designed to test the hypothesis that the antioxidant N-acetylcysteine may prevent CIN. However, no benefit was observed and all 3 patients who developed CIN had received N-acetylcysteine. These results are in keeping with other studies [19–22] and it is interesting that in one of these studies [21] an iso-osmolar contrast agent was used. Furthermore, four recent meta-analyses of all the trials that have evaluated the use of prophylactic N-acetylcysteine have concluded that there is marked heterogeneity in the available data, with some trials reporting a significant benefit and others reporting no renoprotective effect, so that the routine use of N-acetylcysteine cannot be supported [26–29].

However, the current study has its limitations. The power calculations were based on a higher incidence of CIN and therefore the sample size may be too small. Furthermore this study did not evaluate the clinically relevant parameters of outcome such as mortality, morbidity, and length of hospitalization.

Clearly the etiology of CIN is more complex than currently understood. Large studies are needed to clarify the pathophysiology before useful preventive measures can be introduced. Based on the results of our study there is no role for the routine prophylactic administration of N-acetylcysteine in patients undergoing peripheral angiography with either low-osmolar or iso-osmolar contrast agents.