Abstract
Objectives
To examine retrospectively the relationship between acute kidney injury (AKI) and acute myocardial infarction (AMI), and the association between estimated glomerular filtration rate (eGFR) at admission and AKI outcome.
Methods
AKI was defined as an increase in serum creatinine (SCr) by ≥0.3 mg/dl within 48 h or an increase in SCr to ≥1.5 times baseline within the first 7 days of hospitalization. Patients with AMI were divided into subgroups according to their eGFR at admission and the development of AKI.
Results
This study enrolled 396 patients with AMI; 48 (12.1%) developed AKI. In-hospital mortality was 39.6% (19/48) for patients with AKI compared with 7.5% (26/348) in those without AKI (odds ratio [OR] 8.11; 95% confidence interval [CI] 4.02, 16.39). The mortality rate was 35.7% (five of 14) in the eGFR ≥ 60 ml/min/1.73m2 with AKI group (OR 6.21, 95% CI 1.50, 25.69) and 41.2% (14/34) in the eGFR < 60 ml/min/1.73m2 with AKI group (OR 12.62, 95% CI 5.54, 28.74).
Conclusions
AKI development was common and associated with mortality in AMI patients with either preserved or impaired eGFR levels.
Keywords
Introduction
Acute kidney injury (AKI; previously known as acute renal failure) is a common complication that affects patients with various clinical conditions, such as acute myocardial infarction (AMI),1–3 congestive heart failure; 4 and those undergoing cardioangiography (CAG) 5 and percutaneous coronary intervention (PCI). 6 Recent studies suggest that AKI is associated with poor outcomes and independently predicts increasing long-term mortality.1,7–11 However, few studies have investigated the early risk of AKI with AMI.
It is well known that the diagnostic procedure for AKI is not standardized. Rodrigues et al. 12 reported there are over 30 definitions of AKI in the medical literature. This variability causes confusion, making it difficult to compare results across multiple studies. The Risk, Injury, Failure, Loss, End-Stage Renal Disease (RIFLE) 13 and Acute Kidney Injury Network (AKIN) 14 criteria are most commonly used in the study of AKI syndrome. Unfortunately, the existing criteria, while useful and widely validated, are still limited. Hence, in 2012, the Kidney Disease Improving Global Outcomes (KDIGO) 15 Clinical Practice Guidelines for AKI were designed, to compile information systematically on this topic by experts in the field. Definition and staging of AKI are based on the RIFLE and AKIN criteria. In short, AKI is defined as an increase in serum creatinine (SCr) by ≥0.3 mg/dl within 48 h or an increase in SCr to ≥1.5 times the baseline score within the first 7 days. The effect of pre-existing renal dysfunction on AKI and mortality with AMI remains controversial.16–20 There is very little information about the role of impaired estimated glomerular filtration rate (eGFR) at the time of hospital admission on the mortality in patients with AMI-associated AKI (as defined by KDIGO).
The present investigation sought to evaluate the prevalence of AKI in association with AMI, and to investigate the association between a decreased admission eGFR and short-term outcome inpatients developing AKI, as defined by the KDIGO criteria, in the AMI setting.
Patients and methods
Study population
This retrospective study enrolled consecutive patients admitted to the Department of Cardiology, The Affiliated Longyan First Hospital of Fujian Medical University, Longyan, Fujian Province, China, between January 2011 and December 2012, whose discharge diagnosis was AMI (ST-elevation myocardial infarction [STEMI] or non-ST segment elevation myocardial infarction [NSTEMI]). 21 Study exclusion criteria included: lack of at least two SCr measurements at hospitalization; presence of obstructive AKI. The institutional review board of Longyan First Hospital of Fujian Medical University approved the study and all study participants provided written informed consent prior to enrolment.
Data collection
Data collection was performed using a standardized paper case-report form. Collected data included: patient characteristics, past medical history, final discharge diagnosis, electrocardiogram findings, laboratory investigations, echocardiography changes, medical therapies, use of cardiac procedures and interventions, in-hospital outcomes and overall mortality.
After an overnight fast, venous blood samples (10 ml) were collected into haemogram tubes containing di-potassium ethylenediaminetetra-acetic acid (1.5–2.2 mg/ml) and biochemistry tubes. Samples were maintained at room temperature and tested within 1 h of collection. White blood cell (WBC) counts and haemoglobin levels were determined using an automated blood cell counter (XE-5000™; Sysmex, Kobe, Japan), according to the manufacturer’s instructions. Creatine phosphokinase myocardial bundle (CKMB) and SCr were quantified using a biochemical analyser (AU2700; Olympus, Tokyo, Japan). Systolic and diastolic blood pressures were measured by trained personnel using a mercury sphygmomanometer. Ejection fraction (EF) was measured and calculated using ultrasound equipment (HD7 Ultrasound System; Philips, Eindhoven, The Netherlands).
Diagnostic criteria for AKI
The SCr at hospital admission and daily during the coronary care unit stay was available for all analysed patients. eGFR was calculated using the abbreviated Modification of Diet in Renal Disease study equation. 22 Chronic kidney disease (CKD) was defined as an eGFR < 60 ml/min/1.73 m2 for >3 months. AKI was defined as increase in SCr by ≥0.3 mg/dl within 48 h or an increase in SCr to ≥1.5 times baseline within the first 7 days of hospitalization (based on the KDIGO criteria). AKI stage was defined as follows, using SCr: stage 1, 1.5–1.9 times baseline or ≥0.3 mg/dl increase; stage 2, 2.0–2.9 times baseline; stage 3, 3.0 times baseline or an increase to ≥4.0 mg/dl or initiation of renal replacement therapy. Non-AKI was defined as a change in creatinine level of <0.3 mg/dl. Patients were divided into four admission eGFR subgroups: eGFR ≥ 60 ml/min/1.73 m2 without AKI; eGFR < 60 ml/min/1.73 m2 without AKI; eGFR ≥ 60 ml/min/1.73 m2 with AKI; eGFR < 60 ml/min/1.73 m2 with AKI.
Outcomes
The primary study endpoint was in-hospital death from all causes. The secondary study endpoints were the occurrence of in-hospital complications including cardiogenic shock, major bleeding, stroke and malignant ventricular arrhythmia.
Statistical analyses
All statistical analyses were performed using the SPSS® statistical package, version 15.0 (SPSS Inc., Chicago, IL, USA) for Windows®. Continuous variables are presented as mean ± SD or median (with interquartile ranges) and categorical variables as n (%) of patients. Baseline characteristics of the groups were compared using analysis of variance for continuous variables and χ2-test for categorical variables. Univariate and multivariate logistic regression were used to identify the risk factors influencing AKI, and the association between mortality and admission eGFR and AKI development. The strength of the association between risk factors and AKI was expressed as an odds ratio (OR) with a 95% confidence interval (CI). The following baseline clinical and biochemical characteristics were considered in the multivariate procedure: age, sex, history of hypertension, diabetes mellitus, stroke, prior MI, smoking status, systolic blood pressure, heart rate, anterior MI (or not), heart failure on admission, revascularization status and complications. In addition, the following baseline laboratory tests were taken into account: haemoglobin, WBC counts, baseline SCr and eGFR in successive models. Only variables with P < 0.1 in the univariate logistic regression analyses were used in the multiple logistic regression analyses. Differences were considered statistically significant by a two tailed P-value of < 0.05.
Results
Demographic and clinical characteristics of patients (n = 396) with a discharge diagnosis of acute myocardial infarction (MI), stratified according to presence or absence of acute kidney injury (AKI).
Data presented as mean ± SD, median (interquartile range) or n (%) of patients.
Baseline characteristics of the two groups were compared using analysis of variance for continuous variables and χ2-test for categorical variables.
SBP, systolic blood pressure; HR, heart rate; bpm, beats per min; SCr, serum creatinine; eGFR, estimated glomerular filtration rate; CKMB, creatine phosphokinase myocardial bundle; WBC, white blood cell; EF, ejection fraction; NS, not statistically significant (P ≥ 0.05).
Comparison of in-hospital procedures, medication use and adverse outcomes for patients (n = 396) with a discharge of diagnosis acute myocardial infarction, stratified according to presence or absence of acute kidney injury (AKI).
Data presented as n (%) of patients.
Clinical variables of the two groups were compared using χ2-test.
ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; CAG, cardioangiography; PCI, percutaneous coronary intervention; VF/VT, ventricular fibrillation/ventricular tachycardia; NS, not statistically significant (P ≥ 0.05).
Acute kidney injury occurred in 48 of the 396 (12.1%) patients; with 40 patients (10.1%) in stage 1 and eight patients (2.0%) in stages 2 and 3. The rate of AKI at admission stratified according to baseline eGFR status showed a significant association with eGFR (P < 0.001) (Figure 1). Univariate and multivariate analyses of risk factors associated with AKI in the acute MI setting are presented in Tables 3 and 4, respectively. The eGFR < 60 ml/min/1.73 m2 (OR 7.45, 95% CI 2.91, 19.05), baseline SCr (OR 1.10, 95% CI 1.00, 1.14) and Killip class ≥III (OR 3.79, 95% CI 1.68, 8.55) were strongly associated with AKI on multivariate analysis.
Proportion of patients (%) with acute kidney injury (AKI) in 396 patients with a discharge diagnosis of acute myocardial infarction, stratified according to their baseline estimated glomerular filtration rate (eGFR) at the time of hospital admission. Univariate risk factors for acute kidney injury (AKI) in patients (n = 396) with a discharge diagnosis acute myocardial infarction (MI). CI, confidence interval; SBP, systolic blood pressure; HR, heart rate; bpm, beats per min; eGFR, estimated glomerular filtration rate; SCr, serum creatinine; WBC, white blood cell; PCI, percutaneous coronary intervention; VF/VT, ventricular fibrillation/ventricular tachycardia; NS, not statistically significant (P ≥ 0.05). Multivariate risk factors for acute kidney injury in patients (n = 396) with a discharge diagnosis acute myocardial infarction. CI, confidence interval; eGFR, estimated glomerular filtration rate; SCr, serum creatinine.
Overall, there were 45 deaths (11.4%) during hospitalization; 26/348 (7.5%) patients in the non-AKI group compared with 19/48 (39.6%) patients in the AKI group (P < 0.001) (Table 2). In the overall sample, the OR of death in patients with AKI was 8.11 (95% CI 4.02, 16.39; P < 0.001) compared with those without AKI.
Influence of impaired admission estimated glomerular filtration rate (eGFR) on in-hospital mortality rates with and without acute kidney injury (AKI) in patients (n = 396) with a discharge diagnosis acute myocardial infarction.
Compared with group 1; odds ratios were compared using logistic regression analysis.
CI, confidence interval.
Discussion
The main finding of the study was that AKI, defined by the KDIGO criteria, was associated with in-hospital mortality in AMI in patients, with or without impaired eGFR on admission. Overall, the prevalence of AKI in the setting of AMI was 12.1% (48/396 patients). Those with AKI had a substantially higher risk of mortality than those without AKI (P < 0.001).
Patients who have an AMI are at high risk of developing AKI. For example, in a prospective study of patients with STEMI, Goldberg et al. 1 found that 9.6% of individuals experienced worsening renal injury (defined as a 0.5 mg/dl elevation of the SCr level at any point) during their hospital stay. Worsening renal function was associated with an 11.4-fold increased risk of in-hospital mortality. 1 Another study defined AKI as an absolute change in SCr levels; the resulting injury was classified as mild (0.3–0.4 mg/dl), moderate (0.5–0.9 mg/dl), or severe (1.0 mg/dl). 11 Using these criteria, AKI was identified in 19.4% (7.1% mild, 7.1% moderate and 5.2% severe) of patients. 11 A large study evaluating the influence of AKI on short-term outcome in patients with AMI observed an AKI rate of 16%. 23 When AMI is complicated by cardiogenic shock, AKI may affect over half of all patients. 24 However, these investigations did not strictly use the standardized criteria for diagnosing AKI. Some investigations used their own criteria to diagnose, while others used SCr at any timepoint or at 7 days. For future studies, it is very important that standardized AKI criteria are used, such as the KDIGO criteria, 25 to allow comparison of results between different studies.
The RIFLE and AKIN criteria also have limitations. First, despite efforts to standardize the definition and classification of AKI, there are still inconsistencies in their application. Secondly, some AKI cases may be missed. Joannidis et al. 26 directly compared the RIFLE and AKIN criteria and found that the original RIFLE criteria failed to detect 9% of cases that were detected by the AKIN criteria. However, the AKIN criteria missed 26.9% of cases detected by RIFLE. 26 Thirdly, the stages within the RIFLE criteria, containing three severity grades and two outcome criteria, are ambiguous. In addition, in the KDIGO criteria, stage 3 requires patients to reach SCr > 4.0 mg/dl, rather than an acute increase of ≥0.5 mg/dl (AKIN criteria). This change brings the definition and staging criteria to greater parity, and simplifies the criteria.
In this present study, AKI development was common in patients with AMI and was associated with an increased risk of mortality. Furthermore, AKI development was associated with increased in-hospital mortality in AMI patients. Similarly, a retrospective evaluation of the incidence of AKI, based on the AKIN definition, in patients with acute coronary syndromes (ACS; 53% with STEMI, 41% with NSTEMI, and 6% with unstable angina pectoris) found that 13% developed AKI: 8.3% had stage 1, and 4.7% had stage 2 and 3 AKI. 27 In-hospital mortality was greater in patients with AKI than in those without AKI (21% versus 1%; P < 0.001). 27 There were some differences between these two studies. First, studies used different criteria (KDIGO versus AKIN) and different patient types (AMI versus ACS). This current study examined a more high-risk population, with 51.8% of patients being treated conservatively, 10.1% of patients having had a prior stroke and 16.2% having AMI complicated by cardiogenic shock. As expected, the in-hospital mortality rate in this cohort was similar as that usually reported for AMI patients, which was about 4.1–13.2%.23,28
A limited number of studies have assessed the role of admission eGFR on the outcome of AMI-associated AKI, especially in terms of short-term outcomes. For example, a study evaluating AMI patients showed that AKI, defined as a ≥25% decrease in eGFR at any timepoint during hospitalization, was associated with a higher 1-year mortality rate. 29 The conclusion was that AKI was a risk factor for 1-year mortality, independent of admission eGFR. 29 Bruetto et al. 28 reported that an admission eGFR of <60 ml/min/1.73 m2 with AKI was significantly associated with a 30-day to 1-year mortality hazard (adjusted hazard ratio 3.05, 95% CI 1.50, 6.19) using RIFLE criteria (AKI was defined as a SCr increase of ≥50% from the time of admission in the first 7 days of hospitalization). Most studies have focused on the long-term mortality but few on the short-term outcomes.11,23 This current study looked at short-term outcomes and for the first time found that an admission eGFR of <60 ml/min/1.73m2 with AKI was significantly associated with in-hospital mortality. This means that patients with either chronic or acute renal dysfunction are at a very high risk for in-hospital mortality, which might reflect the underlying mechanisms that adversely affect both cardiac and renal function.
Associations between AKI and these short-term risks after AMI have several possible explanations. First, patients who develop AKI have a higher prevalence of comorbidities such as diabetes mellitus, hypertension, and CKD, each of which may increase the risk of kidney failure and death. 2 Secondly, significant differences were observed in the type of AMI treatment received by the two groups in this study. For example, patients in the AKI group were significantly less likely to receive some pharmacological therapies or to undergo interventional procedures. Thirdly, AKI will directly contribute to distant organ dysfunction through the mechanisms of oxidative stress and inflammation. This means that AKI severity, using the KDIGO criteria, represents not only kidney damage but also an imbalance between the kidney and other organ systems. Therefore, AKI defined by the KDIGO criteria, has a greater effect on mortality.
There are several points to be considered. First, the prevalence of AKI in the AMI setting is high and clinicians need to be aware of this important complication. Secondly, uniform standards for diagnosing and classifying AKI should be adopted, which need validation before using in future studies. Thirdly, AKI is associated with the early in-hospital outcomes of AMI in patients with either nonimpaired or impaired admission eGFR. Thus, it is critical to consider measures to prevent AKI as they may represent a viable method for reducing mortality in the AMI setting.
This present study had a number of limitations. It was a single-centre, observational, retrospective cohort study and it used the KDIGO criteria to define AKI, but information on urinary output was not available in the study sample.
In conclusion, AKI occurred in 12.1% (48/396) of patients hospitalized for AMI. This common complication was strongly associated with mortality. The presence of a low eGFR level on admission in patients with AMI-associated AKI was related to a poor short-term survival prognosis. Patients with an impaired eGFR level upon admission, who develop AKI, require extensive clinical monitoring. These results strongly suggest that recognition of these risks and the adoption of appropriate strategies to avoid AKI may improve outcomes after AMI, especially for patients with CKD.
Footnotes
Declaration of conflicting interest
The authors declare that there are no conflicts of interest.
Funding
This study was supported by a grant from the Affiliated Longyan First Hospital of Fujian Medical University (no. 2012001).