OUP user menu

Oral acetylcysteine as an adjunct to saline hydration for the prevention of contrast-induced nephropathy following coronary angiography
A randomized controlled trial and review of the current literature

Ilan Goldenberg, Michael Shechter, Shlomi Matetzky, Michael Jonas, Miriam Adam, Hanna Pres, Dan Elian, Oren Agranat, Ehud Schwammenthal, Victor Guetta
DOI: http://dx.doi.org/10.1016/j.ehj.2003.11.011 212-218 First published online: 1 February 2004


Aims To determine laboratory and clinical benefit of oral acetylcysteine, as an adjunct to saline hydration, in chronic renal insufficiency patients undergoing coronary angiography.

Methods and results We prospectively studied 80 patients with chronic renal insufficiency (mean [±SD] serum creatinine concentration 2.0±0.39mg/dl), who underwent coronary angiography with or without intervention. Patients were randomly assigned to receive either acetylcysteine (600mg orally t.i.d.) or placebo, in addition to intravenous 0.45% saline (1ml/kg of body weight per hour), 12h prior to and after coronary angiography. There was an increase of ≥0.5mg/dl in the serum creatinine concentration 48h after coronary angiography in seven of the 80 patients (9%): in four of the 41 patients (10%) in the acetylcysteine group and in three of the 39 patients (8%) in the placebo group (P=0.52). The incidence of in-hospital adverse clinical events (acetylcysteine, 5% vs placebo, 8%, P=0.47) and the length of hospital stay [acetylcysteine, median (interquartile range) 4 (2–4) days vs placebo, 2 (2–4) days, P=0.44] did not differ significantly between the two treatment groups.

Conclusion Our findings do not support routine prophylactic administration of oral acetylcysteine as an adjunct to saline hydration for the prevention of contrast-induced nephropathy in chronic renal insufficiency patients undergoing coronary angiography.

  • Acetylcysteine
  • Coronary angiography
  • Renal insufficiency
  • Contrast-nephropathy

1 Introduction

Radiographic contrast-induced nephropathy (CIN) is a significant cause of iatrogenic renal dysfunction contributing to mortality and morbidity, involving prolonged hospitalizations and increased medical care costs.1–4Risk factors for CIN include pre-existing renal dysfunction, particularly that caused by diabetic nephropathy, congestive heart failure, reduced effective arterial volume, high-dose administration of a contrast agent, intracardiac injection and the use of concomitant drugs which impair renal responses, such as angiotensin converting-enzyme inhibitors.5–7

Recently, Tepel et al.8showed that the antioxidant acetylcysteine attenuates CIN in patients with chronic renal insufficiency undergoing computed tomography. This study, however, was conducted in patients without overt coronary heart disease (CHD). Patients undergoing coronary angiography and/or percutaneous coronary intervention (PCI) comprise a significant proportion of CIN cases, and acute renal failure after PCI portends a high mortality risk, both in-hospital and long-term.9Current prospective studies evaluating the role of acetylcysteine in patients undergoing coronary angiography, however, have yielded inconsistent data,10–16possibly due to differences in criteria for selection of study patients, non-uniform and/or incomplete hydration protocols and lack of a universal definition of CIN in terms of post-exposure serum creatinine reductions.

We therefore conducted a prospective, randomized, double-blind placebo-controlled trial to evaluate the impact of acetylcysteine administration, in addition to routine intravenous saline hydration and the use of a nonionic, low-osmolality contrast agent, on biochemical markers of renal function, and clinical events of patients with chronic renal insufficiency undergoing coronaryangiography.

2 Methods

2.1 Patients

Eighty consecutive patients who underwent coronary angiography and had serum creatinine concentrations ≥1.5mg/dl or creatinine clearance of <50ml/min (calculated on the basis of the serum creatinine concentration, body weight, age, and gender),17were prospectively studied. All patients had a known history of chronic renal failure with stable serum creatinine concentrations. Patients with acute renal failure, acute myocardial infarction requiring primary or rescue coronary intervention within less than 12h, cardiogenic shock, current peritoneal or haemodialysis, planned post-contrast dialysis, or a known allergy to acetylcysteine, were excluded from the study. Ten patients were not included in the study because of exclusion criteria: 4-acute renal failure; 4-acute myocardial infarction requiring primary coronary intervention less than 12h, 1-cardiogenic shock; and 1-current peritoneal dialysis. Forty-one patients were hospitalized for acute coronary syndromes, and 39 were electively hospitalized with the following indications: stable angina (n=20), evaluation of dilated cardiomyopathy (n=10), follow-up after heart transplantation (n=5), pre-operative assessment of coronary artery disease (n=3) and percutaneous closure of an atrial septal defect (n=1).

The cause of renal insufficiency was diabetic nephropathy in 35 patients (44%), nephrosclerosis in 31 (39%), glomerulonephritis in 1 (1%), and unknown in 13 patients (16%). Indications for coronary angiography and causes of renal insufficiency were equally distributed between the two treatment groups.

The study protocol was approved by the institutional review board, and all patients signed written informed consent.

2.2 Study protocol

The study was a prospective double-blind, randomized, placebo-controlled trial. Patients were randomly assigned to receive either acetylcysteine or placebo. Acetylcysteine (600mg t.i.d.) was administered orally for a total of 48h, starting 24h before the administration of the contrast agent. All patients were treated with intravenous saline (0.45%) at a rate of 1ml/kg of body weight per hour for 12h before and 12h after administration of the contrast agent. Coronary angiography was performed with a non-ionic low-osmolality radiographic contrast agent (Iopamidol, Bracco s.p.a, Milan, Italy), which was administered in boluses of 8–15ml. The solution contained 0.755g of iopromide per millilitre, and the iodine content was 370mg per millilitre. The decision to perform PCI, deployment of stents and the use of glycoprotein IIb/IIIa antagonists, was left to the discretion of the interventional cardiologist. Serum creatinine and urea concentrations were measured 24h and immediately prior to coronary angiography, and 24, 48h and 7 days after the procedure. Serum creatinine concentration immediately prior to coronary angiography was referred to as the baseline level.

Randomization was carried out with computer-generated random numbers. Hospital pharmacists provided the attending nurse with the assigned treatment. Patients and physicians were blinded to treatment assignment.

2.3 Pre-specified end-points

The primary end-point was an acute contrast-agent-induced increase in serum creatinine concentrations of ≥0.5mg/dl 48h after administration of the contrast agent. Secondary end-points were: (1) changes in serum creatinine levels including a >25% increase above baseline, 48h after coronary angiography; a 1.0mg/dl rise; an increase to a level >5.0mg/dl; changes in serum creatinine levels after 48h and at 7 days and (2) clinical adverse events including need for dialysis; overt congestive heart failure following coronary angiography; transient hypotension (systolic blood pressure <100mmHg); peri-procedural acute myocardial infarction (diagnosed in the presence of two of the following three criteria: (a) chest pain for at least 20min; (b) elevation of creatine kinase (or the MB fraction) to >3 times that of normal; and (c) new Q waves on electrocardiography; emergency cardiac surgery; cardiac arrhythmias (defined as ventricular premature beats ≥Lown grade II); the need for intra-aortic counter-pulsation; in-hospital death and the length of hospital stay. Signs and symptoms of adverse events occurring at any stage during the study were recorded in detail on the patient's chart.

2.4 Statistical analysis

We projected the sample size from the study of Tepel et al.,8which demonstrated that the proportion developing an increase in serum creatinine concentrations of ≥0.5mg/dl 48h after administration of the contrast agent among the acetylcysteine group was P1=0.02 and the proportion among the placebo group was P2=0.21. The significance level α is set at 0.05 and the power 1-β at 0.80. We calculated that the number of patients required for an acetylcysteine versus placebo study is 40 per group.

Continuous data are expressed as mean±SD and categorical data as percentage. Continuous variables were analysed by Student t-test and categorical variables by the chi-square test. Because of skewed distribution, length of hospitalization was presented as median and inter-quartile range and was tested by Wilcoxon rank-sum test. All statistical tests were two-sided. A multiple logistic regression analysis was used to examine the effect of acetylcysteine on the primary end-point with adjustment for baseline characteristics [age (≤65 years vs >65 years), body mass index (≤25kg/m2vs >25kg/m2), systolic hypertension status (yes/no), baseline creatinine (≤2mg/dl vs >2mg/dl), diabetes status (yes/no) and amount of contrast agent used (≤150ml vs >150ml)].

We pooled study-specific estimates of the odds ratios of CIN to provide a pooled odds ratio and a confidence interval, using the Cochran–Mantel–Haenszel and maximum likelihood estimates for a combined odds ratio. We applied the PEPI (Programs for Epidemiologic Analysis) program CASECONT18to combine measures of associations, from 2×2 tables of the different studies. The programme provides the chi-square test of association and tests for heterogeneity between individual studies. If heterogeneity is significant, an alternative estimate (the Laird DerSimonian model) for a pooled odds ratio is calculated. We performed an overall analysis, as well as individual ones for studies where a full hydration protocol was employed and mean baseline serum creatinine was <2.5mg/dl, and for those which did not employ a full hydration protocol and/or where mean serum creatinine was ≥2.5mg/dl.

3 Results

3.1 Study patients

All patients enrolled were followed as required by the study protocol. No patient was lost to follow-up or prematurely stopped the study treatment. Baselinecharacteristics of the study patients and angiographic data are shown in Table 1.

View this table:
Table 1

Baseline clinical, pharmacological and laboratory characteristics of the study patientsa

Acetylcysteine group n=41Placebo group n=39P-value
Baseline clinical characteristics
Age, years71±969±100.42
Sex (male/female)35/631/80.34
Body mass index, kg/m227.3±4.628.3±5.70.38
Diabetes mellitus, n (%)16 (39)19 (49)0.25
Prior MI, n (%)22 (54)26 (67)0.16
Prior PCI, n (%)19 (46)13 (33)0.12
Prior CABG, n (%)12 (29)8 (21)0.25
Blood pressure systolic/diastolic, mmHg139±20/83±9140±20/84±100.84/0.99
Ejection fraction,%45±1442±150.35
Aspirin, n (%)37 (90)33 (85)0.33
ACE-inhibitors/ARB, n (%)27 (66)24 (62)0.43
β-blockers, n (%)33 (80)26 (67)0.12
Nitrates, n (%)32 (78)30 (77)0.55
Diuretics, n (%)23 (56)21 (54)0.50
Baseline laboratory values
Serum creatinine, mg/dl2.0±0.41.9±0.30.22
Serum urea nitrogen, mg/dl42±1336±110.05
Creatinine clearance, ml/min38±1141±130.33
  • a All plus-minus values are mean±SD.ACE indicates angiotensin converting enzyme; ACS, acute coronary syndrome; ARB, angiotensin receptor blockers; CABG, coronary artery bypass graft.

There were no significant differences between treatment groups with regard to CHD risk factors, concomitant medications, indications for coronary angiography and baseline serum creatinine or creatinine clearance. However, serum urea nitrogen levels were significantly lower in the placebo group compared to the acetylcysteine group (P=0.05) (Table 1).

PCI was performed in 36 patients (45%) for a total number of 43 lesions. Concomitant glycoprotein IIb/IIIa antagonist administration, and the amount of contrast medium employed during catheterization, were not significantly different between the two treatment groups (Table 2).

View this table:
Table 2

Cardiac catheterization data

Acetylcysteine group n=41Placebo group n=39P-value
PCI proceduresa25180.36
Stent implantationb24170.41
GP IIb/IIIa administration15100.20
Procedural success, n (%)c24/25 (96)16/18 (89)0.37
Ventriculography, n (%)5 (12)7 (18)0.34
Dose of contrast agent, ml111±43121±490.33
  • Plus-minus values are mean±SD. PCI indicates percutaneous coronary interventions.

  • a Defines the number of lesions in which coronary angioplasty was performed.

  • b Defines the number of lesions in which a stent was implanted.

  • c Procedural success was defined as a final TIMI flow of 3 with <10% residual stenosis of the artery in which angioplasty and/or stenting were performed.

3.2 Primary end-point

There was an increase in serum creatinine concentration of ≥0.5mg/dl 48h after coronary angiography in seven of the 80 patients (9%): four of the 41 patients (10%) in the acetylcysteine group and three of the 39 patients (8%) in the placebo group (P=0.52). Unadjusted odds ratio (95% CI) was 1.30 (0.27–6.21). Adjustment for age, diabetes status, body-mass index, systolic blood pressure and volume of contrast did not influence these findings [odds ratio (95% CI) was 1.20 (0.25–8.35)].

None of the seven patients with the primary end-point demonstrated an increase in serum creatinine ≥0.5mg/dl on day 7 post-catheterization, compared to baseline levels.

One of the 10 patients (10%) excluded from the study had an increase in serum creatinine ≥0.5mg/dl 48h after coronary angiography.

3.3 Secondary end-points

Among all study participants there was a non-significant change in serum creatinine levels 48h after cardiac catheterization, compared to baseline levels (2.0±0.39mg/dl vs 1.98±0.37mg/dl respectively, P=0.64).

Serum creatinine did not change significantly 48h post-procedure within the placebo group compared to the acetylcysteine group (a decrease of 0.03±0.32mg/dl vs 0.01±0.36mg/dl respectively, P=0.77) (Fig. 1).

Fig. 1

Serum creatinine concentrations immediately prior to and 48h after the administration of a radiocontrast agent, in the placebo (A) and acetylcysteine (B) groups. The mean (±SD) for each treatment group [acetylcysteine (n=41), placebo (n=39)] is indicated by the heavy lines and circles.

In addition, there were no significant changes in serum urea nitrogen concentrations 48h after cardiac catheterization compared to baseline levels. In theplacebo group, the mean serum urea nitrogen concentration was reduced from 36±11mg/dl at baseline to 35±12mg/dl (P=0.58). In the acetylcysteine group, the mean serum urea concentration also decreased non-significantly from 42±13mg/dl at baseline to 40±15mg/dl (P=0.62).

Specific renal and clinical outcomes are presented in Table 3. Because different end-points were used in previous studies, we recorded renal outcomes in four different ways according to serum creatinine, and added the need for renal dialysis. Contrast-agent–induced reductions in serum creatinine 48h after cardiac catheterization were all below 1mg/dl, and did not differ statistically between the two treatment groups. The incidence of in-hospital adverse clinical events was low (five events in four patients [5% of all patients]) and did not differ significantly between the treatment groups (two events in the acetylcysteine group compared to three events in the placebo group, P=0.47). Length of hospital stay did not differ significantly between treatment groups (Table 3) and among the 10 not included in the current study because of exclusion criteria [4 (2–4) days; median (interquartile range)].

View this table:
Table 3

Laboratory and clinical end-points of study patients after coronary angiography

Acetylcysteine group (n=41)Placebo group (n=39)P-value
Serum creatinine (mg/dl)
Day 0a2.02±0.441.91±0.310.22
After 24h2.07±1.051.89±0.330.32
After 48h2.00±0.511.87±0.360.14
After 7 days2.01±0.451.87±0.310.13
Change in serum creatinine at 48hb
≥0.5mg/dl increase, n (%)4 (10)3 (8)0.52
>25% increase, n (%)4 (10)3 (8)0.52
Clinical end-pointsc
Congestive heart failure, n (%)1 (2)1 (3)0.74
Transient hypotension, n (%)1 (2)00.51
Emergency CABG, n (%)01 (3)0.48
Need for IABP, n (%)01 (3)0.48
Length of hospital stay (days)
All patients (n=80), median (inter-quartile range)4 (2–4)2 (2–4)0.44d
Elective patients (n=39), median (inter-quartile range)2 (2–2)2 (2–2)0.36d
ACS patients (n=41), median (inter-quartile range)4 (4–5)4 (3–5)0.60d
  • Plus-minus values are mean±SD. ACS indicates acute coronary syndrome; CABG, coronary artery bypass graft; IABP, intra-aortic balloon counter-pulsation; MI, myocardial infarction.

  • a Immediately before cardiac catheterization.

  • b The following laboratory end-points did not occur: ≥1.0mg/dl increase in serum creatinine at 48h; and an increase in serum creatinine to ≥5mg/dl.

  • c The following pre-specified clinical end-points did not occur: need for dialysis; peri-procedural myocardial infarction; major arrhythmias; and in-hospital death.

  • d By Wilcoxon rank-sum test.

3.4 Meta-analysis of clinical trials

We identified seven randomized prospective clinical trials in which acetylcysteine was used for prevention of contrast nephropathy in patients with chronic renal insufficiency undergoing coronary angiography, all of which were published between 2000 and 2003 in peer-reviewed journals.14–16The meta–analysis comprised 1023 patients who participated in those trials and our current study. Results of the meta-analysis are presented in Table 4A and B. Differences between studies are noted mainly with regard to mean baseline serum creatinine; hydration protocols; and definition of CIN.

View this table:
Table 4

Summary of current data on acetylcysteine for the prevention of contrast-induced nephropathy in patients undergoing coronary angiography

StudyPatients nDesignBaseline serumcreatininea24-h hydrationprotocolDefinition of CIN (increase in serumcreatinine)Odds ratio (95% CI)
A.Prospective trials in which a beneficial effect was not demonstrated
Boccalondo et al.15183Non-randomized, open label1.8±0.6Employed≥0.5mg/dl1.14 (0.43–2.98)
Durham et al.1379Randomized, double blind2.2±0.4Employed≥0.5mg/dl1.27 (0.40–4.03)
Brigouri et al.11183Randomized, open label1.5±0.4Employed>0.25%0.95 (0.61–1.48)
Allaqaband et al.14b123Randomized, open label2.0±0.6Employed≥0.5mg/dl1.23 (0.34–4.51)
Goldenberg et al.c80Randomized, double blind2.0±0.4Employed≥0.5mg/dl1.30 (0.27–6.21)
Meta-analysis6480.95 (0.59–1.51)
B.Prospective trials in which a positive protective effect was demonstrated
Diaz-Sandoval et al.1054Randomized, double blind1.6±0.05Not employed>25%0.21 (0.06–0.80)
Kay et al.16200Randomized, double blind1.4±0.44Not employed>25%0.32 (0.10–0.96)
Shyu et al.12121Randomized, double blind2.8±0.80Employed≥0.5mg/dl0.13 (0.08–0.20)
Meta-analysis3750.14 (0.07–0.32)
  • a Plus-minus values are mean±SD (mg/dl).

  • b Study patients were also randomized to therapy with fenoldapam.

  • c Current study.

Overall, treatment with acetylcysteine was associated with a favourable effect [OR 0.53 (95% CI 0.35 to 0.78)]. However, no beneficial effect of acetylcysteine was demonstrated in five of the studies comprising 648 patients (Table 4A) [OR 0.95 (95% CI 0.59 to 1.51)], while a beneficial effect was shown in three of the studies comprising 375 patients (Table 4B) [OR 0.14 (95% CI 0.07 to 0.32)].

4 Discussion

The main finding of the current study is that prophylactic oral administration of acetylcysteine provides no additional benefit to 0.45% saline hydration in patients with mild to moderate renal insufficiency undergoing coronary angiography with or without intervention.

To date, the value of the antioxidant acetylcysteine for the prevention of CIN following coronary angiography has been evaluated in seven prospective clinical trials, including the present study.10–16A review of the designs and outcomes of these studies (Table 4A and B) demonstrates that despite the seemingly contradictory results, a consistent pattern emerges, whereby acetylcysteine, when administered as an adjunct to a uniform 24-h hydration protocol, provides no added benefit against CIN (Table 4A). A modest beneficial effect (a definition of CIN as a >25% increase above a baseline mean serum creatinine of 1.3mg/dl [Kay et al.16] or 1.6mg/dl [Diaz-Sandoval et al.10] is lower than the definition of a 0.5mg/dl increase in serum creatinine employed in most other trials) was apparent when the drug was employed together with a non-uniform or incomplete hydration protocol (Table 4B). In addition, benefit of acetylcysteine as an adjunct to 24-h saline hydration was noted in patients with more severe renal dysfunction enrolled in the study of Shyu et al.12and in the patient group undergoing computed tomography reported by Tepel et al.8In the latter group, acetylcysteine significantly reduced the incidence of CIN only in patients with a baseline serum creatinine concentration of >2.5mg/dl (P=0.02), while in patients with lower baseline serum creatinine concentrations no significant benefit was observed (P=0.19).8In lieu of our observation, no beneficial effect was demonstrated in a meta-analysis of the studies8,11,13–15where full hydration protocols were employed in patients with mean serum creatinine <2.5mg/dl [0.95 (95% CI 0.59–1.51)] (Table 4A).

Although the pathogenesis of CIN is not fully understood, it appears to be due to medullary ischaemia caused by decreased renal blood flow resulting from an imbalance of vasodilative and vasoconstrictive factors. Infusion of radiographic contrast agents, with the attendant increases in osmotic load and viscosity, increases the hypoxia of the renal medulla.19Hypoxia may not be tolerated if renal circulation or responses are compromised, as in CHD patients with lower effective circulatory volume due to cardiac dysfunction and concomitant use of diuretics or angiotensin-converting enzyme inhibitors, which impair renal responses. The antioxidant acetylcysteine has been suggested to attenuate CIN through scavenging oxygen free radicals generated as a result of renal tubular toxic damage.20These effects, however, may be less pronounced in CHD patients in whom volume repletion assumes an important role.

Our data include patients’ clinical outcomes, which have not been reported in previous acetylcysteine studies.8,10–16Thus, in our patients, where the use of careful hydration was maintained, changes in serum creatinine concentration after coronary angiography were not associated with decreased procedural success, in-hospital adverse events or an increase in the length of hospitalization. There was also no long-term decrease in renal function and none of the patients required dialysis after exposure to the contrast agent. We were unable to demonstrate any beneficial effect of acetylcsteine despite the fact that, based on its pharmacologicproperties,21we administered acetylcysteine at a dose of 600mg three times a day, compared with lower doses in previous trials.

Several limitations should be noted. The current study protocol excluded patients with acute myocardial infarction, requiring primary or rescue coronary intervention within the first 12h, or cardiogenic shock, and therefore the effect of acetylcysteine was not explored in these patient subsets. However, we did show that careful preparation 12h prior to and after the procedure is safe in hospitalized patients with acute coronary syndrome, and that even these high-risk patients received no additional benefit from acetylcysteine administration. Although the relatively small sample size calls for caution in interpreting the study results, this sample size was predetermined from a power calculation based on the findings of Tepel et al., which resulted in an expected 9% overall event rate. Therefore, the absence of any trend toward benefit with acetylcysteine treatment makes it unlikely that the study overlooked a significant beneficial effect of the drug in the current patient population.

The relative low dose of contrast agent used may also be a limitation of this study. Indeed, much higher loads are expected in cases of ad hoc PCI and multi-vessel PCI. In addition, the mild renal impairment demonstrated in our study population, and the exclusion of high-risk patients, such as the very elderly and dehydrated patients, active intervention and shock, should also be emphasized as limitations in the current study.

In CHD patients with mild to moderate renal insufficiency, hydration with saline was shown to reduce significantly the rate of contrast-induced nephropathy.22–25In a recently published retrospective study of 7586 patients undergoing PCI without routine saline hydration, an increase in serum creatinine of >0.5mg/dl or ≥25% from baseline after PCI, highly correlated with mortality and the need for dialysis during the index hospitalization.9In this study 96% of patients had mild to moderate renal insufficiency. Similarly, during the current 12-month study period only 8% of patients referred to our Heart Institute for cardiac catheterization had a serum creatinine concentration >2.5mg/dl. Therefore, routine saline hydration should continue to be the mainstay of therapy for the prevention of CIN in most CHD patients with renal insufficiency undergoing coronary angiography.

Acetylcysteine is an inexpensive, well-tolerated drug devoid of significant side effects. However, its role as an adjunct to saline hydration in patients with mild to moderate renal insufficiency appears to be limited. Benefit may be more pronounced in conditions where complete hydration is not possible (e.g. emergency coronary angiography or symptomatic congestive heart failure), or in patients with more severe renal dysfunction (serum creatinine >2.5mg/dl).


View Abstract