Chronic renal failure has been described as a risk factor for the development of atrial fibrillation (AF). The aim of this study was to examine the association between contrast-induced nephropathy (CIN) and new-onset AF in patients with acute coronary syndromes. A total of 1,520 consecutive patients (mean age 67.1 ± 12.7 years) with acute coronary syndromes (34.4% with ST-segment elevation myocardial infarctions) who underwent coronary angiography were studied. CIN was defined as an increase in serum creatinine of 0.5 mg/dl within 72 hours of contrast exposure. The independent effect of AF history (chronic or paroxysmal AF before catheterization) on the development of CIN, as well as the independent effect of CIN on the development of new-onset AF (after catheterization, during the in-hospital phase), were tested by using different logistic regression models. One hundred thirty-nine patients (9.1%) had histories of AF before catheterization (60 with paroxysmal and 79 with chronic AF), and 56 (4.1%) developed new-onset AF after catheterization. Eighty-seven patients (5.7%) had CIN. AF history was a predictor of CIN in univariate analysis (odds ratio 2.19, 95% confidence interval 1.22 to 3.95, p = 0.007) but not in multivariate analysis, after adjusting for confounding variables (odds ratio 1.69, 95% confidence interval 0.89 to 3.22, p = 0.111). In contrast, those with CIN had an increased prevalence of new-onset AF (15.3% vs 3.4%, p <0.001). After adjusting for those variables associated with new-onset AF in the univariate analysis, CIN continued to show a significant association with new-onset AF, with a twofold increased risk (odds ratio 2.45, 95% confidence interval 1.07 to 5.64, p = 0.035). In conclusion, the development of CIN is an independent predictor of new-onset AF in the context of acute coronary syndromes.
Renal dysfunction is strongly associated with cardiovascular morbidity and mortality, especially when there is an acute coronary syndrome (ACS). Notably, some evidence has demonstrated that a small increase in serum creatinine (Cr) is a powerful predictor of cardiovascular events in patients with ACS. With the widespread use of the invasive strategy in the management of ACS, more patients undergo coronary angiography. One of the most common complications associated with coronary angiography is contrast-induced nephropathy (CIN). Although more and more is known about this complication and its consequences, nothing is known about whether there is a causal relation between CIN and atrial fibrillation (AF). In the present study, we evaluated the association of CIN with new-onset AF in patients with ACS. We hypothesized that kidney failure would be associated with a higher risk for AF and that this association would be independent of those risk factors common to both CIN and AF.
Methods
We performed a retrospective cohort study designed to examine the predictors and impact of CIN on a contemporary ACS population. From March 2008 to June 2012, a total of 1,520 consecutive patients who were admitted to our hospital with diagnoses of ACS who underwent coronary angiography were included in this registry. Patients receiving long-term dialysis and those without exact data on contrast volume were excluded from the analysis. Demographic, clinical, and angiographic data, as well as information about management and in-hospital complications, had been prospectively collected and recorded in an electronic database. All patients received a nonionic, low-osmolarity contrast agent (iohexol 300 mg iodine/ml).
CIN was defined as an increase in serum Cr of 0.5 mg/dl within 72 hours of contrast exposure. Patients with AF were divided according to the timing of AF occurrence. New-onset AF was defined as de novo AF identified for the first time after coronary angiography on electrocardiography. In contrast, the group with AF histories included those patients with documented episodes of chronic or paroxysmal AF before coronary angiography. Cardiac rhythm monitoring was performed for ≥24 hours after cardiac catheterization. A minimum of 30 seconds of documented AF was necessary to confirm the diagnosis.
The aim of the study was to analyze the possible bilateral association between AF and CIN. First we analyzed whether a history of paroxysmal or chronic AF was associated with the development of CIN and second whether CIN was associated with the development of new-onset AF.
Statistical analyses were performed with SPSS version 21.0 (SPSS, Inc., Chicago, Illinois). Categorical or dichotomous variables are expressed as absolute values and percentages and were compared using Pearson’s chi-square test. Continuous variables are described as mean ± SD. Student’s t tests were used for the comparisons of continuous variables between groups. The independent effect of AF history on the development of CIN and the independent effect of CIN on the development of new-onset AF were tested by using logistic regression models. First, we evaluated the effect of AF history (chronic or paroxysmal AF before catheterization) on the development of CIN using a multivariate logistic regression model, after adjusting by those variables associated with CIN (p <0.10) in the univariate analysis. Second, we examined the effect of CIN on new-onset AF occurrence, excluding from the analysis those patients with histories of AF. For this purpose, we used a multivariate logistic regression model, which was adjusted for those variables associated with new-onset AF (p <0.10) in the univariate analysis. Results of the regression analyses are expressed as adjusted odds ratios (ORs), and 95% confidence intervals (CIs) are presented. The discriminative power of each model was assessed by using the C-statistic. A p value <0.05 was considered statistically significant.
Results
A total of 1,520 patients were included in the present study. Baseline patient characteristics are listed in Table 1 . Overall, the mean age was 67.1 ± 12.7 years, 30.1% were women, and 25.1% had diabetes. One hundred thirty-nine patients (9.1%) had histories of AF before catheterization (60 with paroxysmal and 79 with chronic AF). Four hundred seventy-four patients (31.2%) had renal dysfunction at admission (defined as an estimated glomerular filtrate rate according to the 4-variable Modification of Diet in Renal Disease [MDRD] equation <60 ml/min/1.73 m 2 ). More than one third of patients (34.4%) had ST-segment elevation myocardial infarctions (STEMIs), and left ventricular systolic dysfunction was detected in 16.1%. More than two thirds (67.7%) underwent percutaneous coronary intervention.
| Variables | Total population (N=1,520) | CIN (N=87) | No CIN (N=1,433) | P Value |
|---|---|---|---|---|
| Age (years) | 67.1 ± 12.7 | 76.4 ± 8.9 | 66.6 ± 12.6 | <0.001 |
| Women | 30.1% | 33.3% | 29.9% | 0.494 |
| Hypertension | 57.4% | 73.6% | 56.4% | 0.002 |
| Diabetes mellitus | 25.1% | 42.5% | 24.1% | <0.001 |
| Peripheral artery disease | 7.6% | 11.5% | 7.4% | 0.162 |
| History of AF ∗ | 9.1% | 17.2% | 8.7% | 0.007 |
| STEMI | 34.4% | 44.8% | 33.8% | 0.035 |
| LVEF ≤ 40% | 16.1% | 33.8 | 15.1% | <0.001 |
| Creatinine at admission (mg/dL) | 1.1 ± 0.7 | 1.9 ± 1.5 | 1.0 ± 0.6 | <0.001 |
| PCI | 67.7% | 67.8% | 67.7% | 0.981 |
| Contrast volume (cc) | 165.0 ± 128.3 | 192.0 ± 100.0 | 163.4 ± 129.6 | 0.043 |
| Prior ACEI/ARB | 37.9% | 43.7% | 37.%5 | 0.252 |
| Prior Statin | 31.1% | 37.9% | 30.6% | 0.153 |
| CHA2DS2-Vasc (points) | 3.2 ± 1.8 | 4.7 ± 1.7 | 3.1 ± 1.7 | <0.001 |
∗ Chronic or paroxistic atrial fibrillation prior to catheterization.
Eighty-seven patients (5.7%) had CIN. Patients who developed CIN were older and had worse cardiovascular risk profiles ( Table 1 ). They were more likely to have STEMIs and depressed left ventricular ejection fractions (LVEF). They also had higher values of serum Cr at hospital presentation. As expected, contrast volumes were significantly higher in patients who developed CIN. History of AF was associated with CIN in the univariate analysis. However, after adjusting for age, diabetes, hypertension, STEMI, LVEF ≤40%, serum Cr at admission, and contrast volume ( Table 2 ), AF history was not significantly associated with CIN occurrence (OR 1.69, 95% CI 0.89 to 3.22, p = 0.111).
| Variables | OR | 95% CI | P value |
|---|---|---|---|
| Age ≥ 75 years | 2.96 | 1.80 – 4.89 | <0.001 |
| Diabetes mellitus | 1.50 | 0.89 – 2.54 | 0.127 |
| Hypertension | 1.84 | 1.04 – 3.24 | 0.036 |
| STEMI | 1.54 | 0.93 – 2.55 | 0.093 |
| LVEF ≤ 40% | 2.05 | 1.20 – 3.52 | 0.009 |
| Creatinine at admission | 1.86 | 1.43 – 2.42 | <0.001 |
| Contrast volume | 1.01 | 0.99 – 1.02 | 0.287 |
| History of AF ∗ | 1.69 | 0.89 – 3.22 | 0.111 |
∗ Chronic or paroxysmal atrial fibrillation prior to catheterization.
Among 1,381 patients without histories of AF, 56 (4.1%) developed new-onset AF after coronary angiography. These patients were older (mean age 75.6 ± 10.8 vs 65.8 ± 12.6 years, p <0.001); had higher rates of diabetes (35.7% vs 24.2%, p = 0.051), peripheral artery disease (17.9% vs 6.8%, p <0.001), STEMI (60.7% vs 34.6%, p <0.001), and LVEF ≤40% (31.4% vs 14.8%, p <0.001); and had higher values of serum Cr (mean 1.3 ± 0.8 vs 1.1 ± 0.7 mg/dl, p = 0.029) and higher CHA 2 DS 2 -VASc scores (mean 4.4 ± 1.9 vs 3.2 ± 1.7, p <0.001). Interestingly, patients who developed CIN had an increased risk for new-onset AF (15.3% vs 3.4%, p <0.001). Figure 1 shows the increased risk for new-onset AF according to impaired renal function (serum Cr) after coronary angiography. After adjusting for age, diabetes, peripheral artery disease, STEMI, LVEF ≤40%, and serum Cr ( Table 3 ), CIN continued to show a significant association with new-onset AF (OR 2.45, CI 95% 1.07 to 5.64, p = 0.035). Even after adjusting for CHA 2 DS 2 -VASc score, CIN maintained its predictive value for the development of new-onset AF (OR 2.81, 95% CI 1.35 to 5.83, p = 0.006).
