The influence of contrast media on kidney function in patients with stable coronary artery disease

Abstract

Aims. To investigate the incidence of contrast media-induced nephropathy (CIN) in patients with stable coronary artery disease (CAD) referred for elective coronary intervention following hydration routines. The reversibility of CIN was followed in a 6 month-period. Methods and results. A total of 447 patients referred for elective coronary intervention due to suspected CAD were included. Blood samples were collected before and 24 h after intervention and medical records were obtained. Patients had no drinking fluid restrictions and were routinely treated with a 1000 ml saline infusion. All patients were invited to a 6-month examination and collection of blood samples. Results. A total of 19 patients (4.3%) developed CIN. CIN patients had a pre-investigation higher estimated glomerular filtration rate (eGRF), lower level of kidney failure and lower creatinine level than non-CIN patients. Kidney function was not normalized in CIN patients 6 months after the intervention. Two patients still met the definition of CIN. Conclusion. With no restriction in fluid intake and supplementary infusion of saline, only a few patients with stable CAD developed early indications of CIN during elective coronary interventions. Kidney function and the amount of contrast media used was not a predictor of CIN development. The induced CIN was not completely normalized in a 6-month follow-up period.

Key words::

• contrast media
• coronary artery disease
• glomerular filtration rate
• ischemic heart disease
• kidney function

Introduction

The use of invasive cardiac procedures has increased by 341% in the United States during the last two decades (1). Worldwide, in 2003, approximately 80 million doses of iodinated intravascular contrast media were prescribed (2). Because of the increasing number of diagnostic procedures and interventions, contrast-induced nephropathy (CIN) is now the third most common cause of hospital-acquired renal insufficiency (3). The development of CIN is associated with death; the 1-year-mortality rate is 22% causing 7,000 deaths per year (4). CIN is furthermore associated with later events of myocardial infarction, target vessel revascularisation, coronary artery bypass graft (CABG), haematoma formation, pseudo-aneurysms, stroke, coma, adult respiratory distress syndrome, pulmonary embolism and gastrointestinal haemorrhage (4–6).

Several pre- coronary angiography (CAG) CIN-prediction-equations have been developed. However, they are mainly developed from an unselected group of patients admitted with acute coronary syndrome (ST segment elevation myocardial infarction (STEMI) and Non-ST segment elevation myocardial infarction (NSTEMI)) needing acute invasive treatment. Three studies have previously been exploring the incidence of CIN in elective catheterisations, but the primary outcome has not been risk factors or reversibility. Chong et al. (7) found a CIN incidence of 4.5% in patients without myocardial infarction (MI) undergoing elective PCI, receiving pre-hydration with saline infusion (1 ml/kg for 6–12 h) in patients with an eGFR < 60 ml/min/1.73 m². Burchardt et al. (8) found a CIN incidence of 0.9% in patients with symptoms of ischemic heart disease undergoing elective CAG, pre-hydration treatment was saline infusion (5 h before and up to 10 h after the procedure). Kong et al. (9) found an incidence of 5.8% in patients with a creatinine < 110 μmol/l referred for elective CAG, using pre-hydration strategies with either oral tap water, intravenously saline infusion or only post-procedure tap water, in all patients independent of renal function.

The aim of the present study was to investigate the acute incidence and long-term reversibility of CIN during elective CAG and/or PCI procedures in patients with stable CAD undergoing a routine hydration programme. Furthermore, the study aimed to investigate the value of using the CIN-prediction equation developed by Cigarroa to identify patients at risk of developing CIN (10).

Methods

Study population

A total of 447 patients with established or suspected stable CAD referred to The Heart Centre, Rigshospitalet, University Hospital of Copenhagen, Denmark for elective CAG or PCI were included in a prospective cohort study from February 2009 to March 2011.

Patients were excluded if they had cancer, chronic nephropathy and dialysis, inflammatory diseases and liver disease or had an operation or acute coronary syndrome during the last 4 weeks before inclusion.

The study was performed in accordance with the Helsinki Declaration and was approved by the local scientific ethical committee (H-A-2008-117) and the Danish Data Protection Agency (2007-58-0015). All participants signed a written consent after oral and written information.

Study design and data collection

Clinical history and blood samples were obtained at admission (baseline), the day after and 6 months after the investigation. During intervention (CAG and/or PCI), information about contrast use was extracted from the departments’ cardiac database PATS. Laboratory data was extracted from the Department of Biochemistry laboratory database, LABKA.

Pre-invasive treatment

Patients were treated according to the local routine hydration guidelines. Before the intervention, there were no restrictions on drinking water and the patients were given an intravenous infusion of 1000 ml saline initiated just before the procedure. Patients with serum creatinine > 100 μmol/L received a total of 2000 ml saline and 4800 mg of acetylcysteine the night before the intervention. Patients with diabetes received an intravenous infusion of glucose and insulin if found necessary. Biguanides (Metformin^®) were withheld 2 days before the intervention. There was no control of whether the patients had actually received the implemented hydration regime since the aim was to investigate its function in a routine situation. If the patient had experienced an earlier event with an allergic reaction to contrast, anti-histamine and glucocorticoids were administered in addition.

The contrast media-of-choice was Vesipaque^®, but six patients received Iomeron^®.

The use of additional medication, beta-adrenergic blockers, angiotensin-converting enzyme (ACE) inhibitors, platelet glycoprotein IIb/IIIa inhibitors, diuretics or inotropic drugs support was left to the discretion of the operating physician.

Estimated-GFR and serum-creatinine measurements

Creatinine-clearance, serum creatinine and eGFR all predict CIN equally well, and serum creatinine and eGFR are further strong independent predictors for CIN (11).

eGFR was calculated by using the re-expressed 4-variable MDRD formula, without correction for ethnicity (12):

The levels of kidney failure were defined according to the widely accepted KDOQI guidelines (13).

Table

Level 1	eGFR > 90 (ml/min/1.73 m^2)
Level 2	eGFR 60-89 (ml/min/1.73 m^2)
Level 3	eGFR 30-59 (ml/min/1.73 m^2)
Level 4	eGFR 15-29 (ml/min/1.73 m^2)
Level 5	eGFR < 15 (ml/min/1.73 m^2)

Serum was analysed immediately after blood sampling by using an automated platform (Modular P, Roche Diagnostics, Germany). Creatinine was analysed by using the enzymatic method on Roche MODULAR analytics (SWA) module P1 and P3.

Outcome variables

CIN was defined according to The European Society of Urogenital Radiology definition as an increase in serum creatinine by > 25% or 44.2 μmol/L [0.5 mg/dL] (14).

Data analysis

Continuous data were presented as mean ± one standard deviation (SD) and categorical data as absolute values and percentages. Continuous variables were compared with Mann–Whitney U test or Wilcoxon's test for paired data. Categorical variables were compared with chi-square test. Analyses were performed by using IBM SPSS statistical software version 20.

All p-values were two-sided and a p-value < 0.05 was considered significant.

Results

Patient characteristics

The baseline data for all included patients, and divided with regard to CIN development, are shown in Table I.

Table I. Baseline data of the included patients.

Characteristics	All patients (n = 447) Mean ± SD or number (%)	Patients developing CIN (n = 19) Mean ± SD or number (%)	No-CIN patients (n = 428) Mean ± SD or number (%)	CIN-patients vs No-CIN patients P-value
Age	66 ± 10	62 ± 9	66 ± 10	0.136
Age > 70 years	153 (34)	3 (16)	150 (35)	0.083
Female	132 (29)	8 (42)	124 (29)	0.219
Active smoker	87 (20)	4 (21)	83 (20)	0.99
BMI	27.8 ± 4.6	29.3 ± 5.7	27.7 ± 4.5	0.203
BSA	2.0 ± 0.2	2.0 ± 0.2	1.9 ± 0.2	0.312
Diabetes	107 (24)	5 (26)	102 (24)	0.910
Hypercholesterolemia	392 (89)	16 (89)	376 (89)	0.978
Hypertension	329 (74)	14 (78)	315 (74)	0.728
NYHA	1.8 ± 0.9	1.5 ± 1	1.8 ± 0.9	0.198
CCS	1.9 ± 1	2 ± 0.8	1.9 ± 1	0.961
LVEF	51 ± 12	50 ± 11	52 ± 11.8	0.632
Systolic BP	143 ± 21	148 ± 20	143 ± 21	0.245
Diastolic BP	80 ± 14	88 ± 9	80 ± 14	0.002
Pulse	65 ± 12	72 ± 11	65 ± 12	0.016
eGFR (ml/min)	77 ± 16	86 ± 7	76 ± 16	0.006
eGFR level	1.8 ± 0.8	1.4 ± 0.5	1.8 ± 0.8	0.006
Creatinine (μmol/l)	83 ± 34	60 ± 21	83 ± 34	0.0001
Creatinine 100–200 (μmol/l)	56 (12)	0 (0)	56 (13)	0.092
Creatinine > 200 (μmol/l)	6 (1)	0 (0)	6 (1)	0.603
Total volume of contrast (ml)	146 ± 94	153 ± 105	146 ± 94	0.856
V/eGFR	2.1 ± 1.6	2.3 ± 1.7	2.1 ± 1.6	0.288
MCD (ml)*	503 ± 478	1114 ± 2110	476 ± 142	0.0001
MCD exceeded	15 (4)	0 (0)	27 (6)	0.391
Prior AMI	159 (68)	12 (63)	149 (67)	0.233
Prior PCI	197 (95)	11 (58)	186 (95)	0.443
Prior CABG	91 (20)	3 (16)	88 (21)	0.934
Haemoglobin (mmol/l)	8.5 ± 1	8.4 ± 0.7	8.6 ± 1	0.201
Albumin (g/l)	44 ± 3	43 ± 5	44 ± 3	0.467
Cholesterol, total (mmol/l)	4.3 ± 1	4.2 ± 0.8	4.3 ± 1	0.738
Cholesterol, LDL (mmol/l)	2.3 ± 0.8	2.1 ± 0.8	2.3 ± 0.8	0.305
Cholesterol-lowering drugs	447 (100)	19 (100)	428 (100)	#
ACE inhibitor	162 (36)	6 (68)	156 (36)	0.666

BMI, Body Mass Index; BSA, Body Surface Area; NYHA, New York Heart Association; CCS, Canadian Cardiovascular Society; LVEF, left ventricular ejection fraction; BP, blood pressure; eGFR, estimated glomerular filtration rate; V/eGFR, volume-of-contrast to eGFR ratio; MCD, maximum contrast dose; AMI, acute myocardial infarct; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft; LDL, low-density lipoprotein; ACE, angiotensin-converting enzyme.

In the entire population, mean age was 66 years and 70% were males. About one third of the patients were above 70 years of age, diabetes was present in 24% of them and the majority had hypertension and hypercholesterolemia (Table I). Pre-investigation serum creatinine concentration was elevated (> 100 μmol/l) in 62 patients (18%) and kidney function was impaired (< 60 ml/min) in 71 patients (16%). The maximum recommended contrast (MCD) dose used according to Cigarroa (10) was exceeded in 15 patients (3.6%). All patients were treated with cholesterol-lowering drugs, and 36% with ACE-inhibitors.

Contrast-induced nephropathy

According to the definitions used, 19 (4.3%) patients developed CIN the day after the intervention (Table I). These patients were characterized by normal kidney function at baseline, evaluated by eGFR (86 ml/min, mean) and creatinine (60 μmol/l, mean), and none of them had a creatinine level between 100 and 200 μmol/l.

Patients were more often men (58%) with a mean age of 62 years and only 3 (16%) were above 70 years. The majority of the patients who developed CIN had hypertension or hypercholesterolemia and 5 (26%) had diabetes. A total of 12 patients (63%) were known with previous ischemic heart disease.

All patients were treated with a cholesterol-lowering drug and 6 of them (32%) with an ACE-inhibitor. None of the 62 patients with known impaired kidney function (creatinine > 100 μmol/l) developed CIN after the invasive procedure.

To detect factors of importance for the development of CIN, differences in variables between patients who did and did not develop CIN were compared. Patients who developed CIN had a normal and better kidney function at baseline compared with patients who did not develop CIN (Table I). They had a significant higher eGFR (p = 0.006), level of kidney failure (p = 0.006) and a lower serum creatinine level (p < 0.0001). The total volume of contrast used and contrast volume per eGFR unit was identical in the two groups. Calculated maximum tolerable contrast dose was higher in CIN-patients (p < 0.0001) due to better kidney function at baseline.

However, exceeded MCD was only detected in a small group of patients in the non-CIN group (n = 6). Diastolic blood pressure and pulse was higher in the patients who developed CIN.

Reversibility of CIN

To investigate the reversibility of contrast-induced kidney impairment, the patients were reinvestigated 6 months after CAG, and approximately 50% of the patients had a 6-month follow-up visit (n = 252). Patients who met for 6-month control were compared with patients who did not. We tested for all parameters in Table I. The only significant parameters were total volume of contrast (p = 0.002 (131 ml ± 85 ml vs. 159 ml. ± 99 ml.)), volume of contrast to eGFR ratio [p = 0.002 (1.8 ± 1.3 vs. 2.3 ± 1.8)] and prior PCI [p = 0.002 (94 patients (90%) vs. 103 patients (100%))] for patients completing baseline and the day after, in contrast to patients completing baseline, the day after and 6 months after, respectively.

The 19 patients who developed CIN were divided into two subpopulations: one with CIN-patients completing baseline and the day after (n = 8) and one with CIN-patients completing baseline, the day after and 6-month control (n = 11). The two subpopulations were comparable for all parameters in Table I, except for the fact that the 6-month follow-up patients were more often treated with an ACE-inhibitor. Key variables are shown in Table II.

Table II. Baseline characteristics of patients developing CIN divided into follow-up status.

Characteristics	Patients completing baseline and the day after (n = 8) Mean ± SD	Patients completing baseline, the day after and 6-month control (n = 11) Mean ± SD	P-value
eGFR (ml/min)	87 ± 5	85 ± 8	0.442
eGFR level	1.25 ± 0.5	1.45 ± 0.5	0.492
Creatinine (μmol/l)	52 ± 24	66 ± 17	0.395
ACE-inhibitor	0 (0)	5 (46)	0.018

ACE, Angiotensin-converting enzyme; eGFR, estimated glomerular filtration rate

For the 11 patients who developed CIN and had completed baseline, the day after and 6-month follow-up, there was a trend towards normalization in overall kidney function at 6-month follow-up. The overall kidney function was not as good as at inclusion. There was a significant difference in all three kidney parameters: creatinine was higher, eGFR were lower and therefore, the kidney function level was lower (Table III). Furthermore, 2 patients (18%) still met the diagnostic criteria for CIN at 6-month follow-up.

Table III. Patients completing baseline and 6-month follow-up.

Characteristics	Blood samples from baseline (n = 11) Mean ± SD	Blood samples from 6-month follow-up (n = 11) Mean ± SD	P-value
eGFR (ml/min)	85 ± 8	79 ±.12	0.018
eGFR level	1.45 ± 0.5	1.8 ± 0.8	0.046
Creatinine (μmol/l)	66 ± 17	75 ± 22	0.011

eGFR, estimated glomerular filtration rate.

Of the patients without procedure-related development of CIN, a total of 14 (5.5%) had a deterioration of kidney function in the follow-up period and thereby fulfilled the criteria for CIN. In Table IV, baseline values and kidney function parameters the day after and at 6-month control are shown. All parameters in Table I plus kidney function variables the day after (eGFR, eGFR level and creatinine) were compared between patients developing late CIN (n = 14) and patients without CIN (n = 238). Only potassium at baseline was different, and patients developing late CIN had a lower potassium pre CAG/PCI than patients without CIN (p = 0.049 (3.9 ± 0.3 vs. 4.2 ± 0.6)).

Table IV. Patients developing late CIN (n = 14).

Characteristics	Mean ± SD or number (%)
Age	68 ± 7.5
Age > 70 years	1 (7)
Female	5 (39%)
Active smoker	2 (14)
BMI	28 ± 4.6
BSA	1.9 ± 0.2
Diabetes	2 (14)
Hypercholesterolemia	12 (86)
Hypertension	9 (64)
NYHA	2 ± 0.8
CCS	1.5 ± 1
LVEF	55 ± 7
Systolic BP	144 ± 24
Diastolic BP	78 ± 10
Pulse	66 ± 11
Baseline eGFR (ml/min)	78 ± 15
Baseline eGFR level	1.6 ± 0.7
Baseline creatinine (μmol/l)	75 ± 20
Baseline creatinine 100–200 (μmol/l)	1 (7)
Baseline creatinine > 200 (μmol/l)	0 (0)
Day 2 eGFR (ml/min)	60 ± 20
Day 2 eGFR level	1.8 ± 0.6
Day 2 creatinine (μmol/l)	78 ± 15
6-month eGFR (ml/min)	60 ± 21
6-month eGFR level	2.6 ± 0.8
6-month creatinine (μmol/l)	100 ± 26
Total volume of contrast (ml)	122 ± 75
V/eGFR	1.6 ± 0.8
MCD (ml)*	501 ± 98
MCD exceeded	0 (0)
Prior AMI	6 (43)
Prior PCI	7 (50)
Prior CABG	1 (7)
Baseline haemoglobin (mmol/l)	8 ± 0.9
Baseline albumin (g/l)	44 ± 2.3
Baseline cholesterol, total (mmol/l)	4.2 ± 0.9
Baseline cholesterol, LDL (mmol/l)	2.1 ± 0.8
Cholesterol-lowering drugs	14 (100)
ACE inhibitor	2 (14)

BMI, Body Mass Index; BSA, Body Surface Area; NYHA, New York Heart Association; CCS, Canadian Cardiovascular Society; LVEF, left ventricular ejection fraction; BP, blood pressure; eGFR, estimated glomerular filtration rate; V/eGFR, volume-of-contrast to eGFR ratio; MCD, maximum contrast dose; AMI, acute myocardial infarct; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft; LDL, low-density lipoprotein; ACE, angiotensin-converting enzyme.

Discussion

The present study investigated whether a hydration regime implemented in daily routine coronary angiographic practice could eliminate the development of CIN. A total of 447 patients with established or suspected stable CAD referred for elective coronary investigation were included, and approximately 50% of the patients had a 6-month follow-up visit to investigate the reversibility of induced kidney malfunction.

The patients had no drinking water restrictions and were routinely receiving intravenous fluid during and after the investigation. The study demonstrated that only a small fraction of patients with stable CAD referred for elective CAG and/or PCI developed CIN in this routine kidney protection programme. It was surprising that these patients all had a pre-investigation normal kidney function at a higher level than the remaining group. Moreover, none of these patients exceeded the recommended maximal contrast dose during the investigation. In opposition, it is interesting that no patients with known reduced kidney function developed further deterioration during the contrast investigation. This could be due to the programme of no drinking water restrictions before the procedure and saline infusion during and after the intervention and due to the fact that the maximal recommended contrast dose was exceeded only in very few patients.

The incidence of CIN was 4.3% in the present population. At the 6-month follow-up visit, there was still a significant impairment in kidney function in CIN patients and two patients still met the criteria for CIN. Furthermore, 14 patients developed late CIN and these patients were no different from non-CIN patients in baseline data or in data from the blood samples drawn the day after the investigation. This is the first study, to our knowledge, to investigate the reversibility of the kidney impairment in CIN in patients with stable CAD. In the present study, the kidney impairment does not seem to be fully reversible within 6 months.

According to the guidelines, the optimal time period to detect CIN is 48–72 h after contrast injection (14). However, since all patients referred for elective coronary angiography and percutaneous coronary balloon angioplasty are discharged from hospital within 24 h after the procedure, late CIN cannot be identified. We, therefore, decided that we wanted to detect early developed CIN and the persistent of this and late developed CIN by an additional 6 months follow-up. By this method we can potentially have missed some CIN developed after 24 h.

Kong et al. have compared orally vs. intravenous pre-hydration for preventing CIN in 120 patients with established or suspected stable CAD, normal kidney function (creatinine < 110 μmol/l), those undergoing elective CAG and/or PCI with a CIN incidence of 5.8% (9).

There was no statistical difference in patients’ demographics, clinical or procedural data and the total volume of contrast used in the three groups. There was no statistical difference between oral use and intravenous pre-hydration treatment.

Chong et al. found an incidence of CIN = 4.5% in 1097 patients without MI referred to elective CAG and/or PCI (7). CIN-incidence was 4.3%, 2.4% and 25.9% measured 48 h after catheterisation in patients with an eGFR > 60 ml/min/1.73 m², eGFR = 30–60 ml/min/1.73 m² and eGFR < 30 ml/min/1.73 m², respectively. Only patients with an eGFR < 60 ml/min/1.73 m² received pre-hydration.

Previous studies have found a 3% CIN incidence in unselected patients with CAD, measured on the days after the intervention. The acute incidence of CIN may be as high as 15% in unselected patients with both stable and unstable CAD, normal and impaired kidney function referred to emergent and elective CAG and/or PCI (4,15–18).

From the literature, there seems to be an association between impaired kidney function and CIN development, if the patients have a very low kidney function (eGFR < 30 ml/min). However, based on the results of the present study and the study by Chong et al. (7) it seems that baseline kidney function is not a good predictor of the development of CIN in patients with stable CAD receiving elective invasive procedures.

The present low incidence of CIN in patients with both normal and reduced kidney function may be attributed to the renal prophylaxis used in these elective patients. Pre-treatment with saline and/or no drinking water restrictions seems important, as the incidence of CIN in the present study and in the study by Chong et al. (7), Burchardt et al. (8) and Kong et al. (9) is identical and low.

Studies investigating CIN in an unselected group of patients, but mostly in patients with STEMI or patients who had urgent catheterisations, found that age, low LVEF, hypotension (and in some studies hypertension), chronic kidney disease, dehydration, volume of contrast media, low albumin level and anaemia predispose to CIN (1,4,7,10,19–23). Patients with STEMI or NSTEMI have an impaired haemodynamic function, thereby worsening the effect of contrast on the kidneys.

In the present study, only high diastolic blood pressure and high pulse were associated with CIN.

Studies have found that in 80% of patients developing CIN, creatinine will rise within the first 24 h (22,24). There is no recommendation for when to measure kidney function after contrast intervention to detect CIN. Since most of the present patient populations are discharged from the hospital on the same day or on the day after the investigation, not all cases of CIN will be detected.

It is interesting that we found some patients had developed late CIN, and that the condition persisted and was detected at the 6-month follow-up. Therefore, blood samples should be collected for several days after the procedure to diagnose all cases of CIN. We have used traditional measurement parameters of kidney function. However, there is an ongoing discussion on whether new markers of kidney function could detect minor kidney injuries and hereby be a better measurement of contrast effect on kidney function.

It has been suggested that Cystatin C and neutrofil gelatinase-associated lipocalin (NGAL) could be superior to serum creatinine for the early diagnosis of CIN in patients who underwent coronary catheterization (25). Whether the use of these new parameters would alter the present findings is unanswered, but it is not likely.

This study had a limitation in that not all patients had a 6-month follow-up visit. However, this decline in numbers of patients having a follow-up visit in such a clinical registration study population is not unusual due to factors like long travel distances, age, co-morbidity, patient disability and death.

In the present study, patients who had not developed CIN after a coronary intervention procedure had exceeded the total contrast dose according to the formula used by Cigarroa (10). It could, therefore, not be used in our population to predict which of the patients will develop CIN. The formula has been used in patients undergoing primary PCI where 45.1% exceeded MCD and developed CIN and 4.5% developed CIN without exceeding MCD (10).

Conclusion

Among patients with stable CAD referred for elective CAG and/or PCI, only a few developed early indications of CIN when following a hydration regime for their daily routine. There was no association between impaired kidney function and chronic kidney disease at inclusion and the development of early CIN after contrast administration. The contrast-induced kidney impairment does not seem to be reversible for all patients. The programme of no restrictions in drinking water consumption and addition of intravenous saline during and after intervention could be the background for the low incidence of early CIN in the present population.

Declaration of interest: The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.