Abstract
Background
Patients with chronic kidney disease (CKD) experience poor outcomes after acute myocardial infarction (AMI). We sought to compare clinical characteristics of advanced CKD, dialysis, and non-CKD patients hospitalized with AMI.
Methods
This observational study used record-linked data from the US Renal Data System and Third National Registry of Myocardial Infarction to identify 2390 dialysis patients with AMI hospitalizations between April 1998 and June 2000. Advanced CKD patients (n = 29,319) were identified by baseline creatinine ≥ 2.5 mg/dL. Clinical characteristics of CKD, dialysis, and non-CKD patients (n = 274,777) were compared using the χ2 test.
Results
Clinically significant differences among advanced CKD (dialysis, non-CKD, respectively) patients on admission were chest pain, 40.4% (41.1%, 61.6%); diagnosis other than acute coronary syndrome, 44% (47.7%, 25.8%); and ST elevation, 15.9% (17.6%, 32.5%). Inhospital adverse outcomes were mortality, 23% (21.7%, 12.6%); unexpected cardiac arrest, 8.9% (12.3%, 6%); congestive heart failure, 41% (25.8%, 21.1%); and major bleeding, 4.9% (4.4%, 3%). P < 0.001 for all comparisons. In a logistic regression model, the adjusted odds ratio for inhospital mortality for CKD (vs. non-CKD) patients was 1.44 (95% confidence interval 1.39–1.49).
Conclusions
The clinical characteristics of nondialysis-dependent, advanced CKD patients with AMI are similar to characteristics of dialysis patients, and likely contribute to poor outcomes. Intensive efforts for timely, accurate recognition of AMI in advanced CKD patients are warranted.
Introduction
Nearly 13% of the US population is diagnosed with chronic kidney disease (CKD), and prevalence increases as the population ages. Patients with CKD carry a considerable burden of cardiovascular disease (CVD) relative to patients with normal renal function; there is an independent inversely graded correlation between worsening renal function and all-cause mortality. In patients with end-stage renal disease on maintenance dialysis, CVD accounts for 43% of all deaths. All-cause mortality of dialysis patients with acute myocardial infarction (AMI) is 59% at 1 year and about 73% at 2 years. AMI in patients with nondialysis-dependent advanced CKD is also associated with poor long-term cardiovascular outcomes and survival. Minor reductions in estimated glomerular filtrates rates (eGFR) in CKD patients with AMI and left-ventricular dysfunction are associated with escalating hazards of future cardiovascular events and death.
Understanding factors underlying poor outcomes in patients with CKD is important. Much of the literature relates to reduced use of evidence-based pharmacological and interventional therapies in this population (“therapeutic nihilism”). However, presence of kidney disease may fundamentally change how patients experience AMI. Clinical characteristics of dialysis patients with AMI differ markedly from characteristics of nondialysis patients, likely contributing to poor outcomes. In particular, on presentation with AMI, prevalence of chest pain and ST-segment elevation is lower in dialysis patients.11
We aimed to compare the clinical characteristics of a large population of AMI patients with advanced CKD with characteristics of dialysis patients and patients without significant renal dysfunction. We speculated that differences in clinical attributes and presentation of patients with advanced CKD might contribute to underdiagnosis or misdiagnosis of AMI.
Methods
This retrospective observational study was a collaborative effort between the National Institute of Diabetes and Digestive and Kidney Diseases (National Institutes of Health, Bethesda, Maryland) and the Third National Registry of Myocardial Infarction (NRMI-3)/Genentech (San Francisco, California). The Institutional Review Board of Hennepin County Medical Center (Minneapolis, Minnesota) and the Office for Human Research Protections (Department of Health and Human Services, Rockville, Maryland) approved this study.
In the US Renal Data System (USRDS) database, 7245 unique dialysis patients with AMI hospitalizations were identified from April 1998 through June 2000 (corresponding to NRMI-3 enrollment), and linked with the NRMI-3 registry (which does not identify dialysis dependence). A record-linked cohort of 2390 dialysis patients was created after excluding patients transferred into or out of the index hospital. In the NRMI-3 database, 304,096 non-dialysis patients were identified; from these, advanced CKD patients (n = 29,319 with baseline, stable creatinine level > 2.5 mg/dL) were identified. All other patients were categorized as non-CKD (n = 274,777). Clinical characteristics (demographics; medical history; temporal, admission, and in-hospital variables) of the 3 groups were compared.
All cardiovascular events recorded in the NRMI database are in-hospital events classified by trained clinical data abstractors using standardized definitions without the aid of central adjudication. Mean (± standard deviation) was calculated for continuous variables and proportions for categorical variables. For discrete data, the Pearson chi-square test was used to detect differences among comparison groups. Continuous data were compared using 1-way ANOVA. Non-parsimonious multivariable logistic regression models were constructed to study the associations between CKD and 4 in-hospital events: death, recurrent AMI, stroke, and major bleeding. For each event, covariates were selected a priori based on previous literature and clinical significance. Model covariates included demographics (age, sex, race, payer), medical history (angina, coronary artery bypass graft [CABG] surgery, family history of coronary artery disease [CAD], congestive heart failure [CHF], chronic obstructive pulmonary disease, diabetes, hypercholesterolemia, hypertension, previous AMI, percutaneous transluminal coronary angiography [PTCA], current smoking, stroke), hospital characteristics (AMI volume, PTCA capability, geographic region, teaching status, urban vs. rural), clinical presentation (admission diagnosis, Killip class, pulse, systolic and diastolic blood pressure, electrocardiogram [ECG] results), medications (angiotensin-converting enzyme [ACE] inhibitors, aspirin, beta blockers, calcium blockers, unfractionated heparin, glycoprotein IIb/IIIa, low-molecular weight heparin), and reperfusion. The models did not include interaction terms. Where applicable, Bonferroni adjusted P values were used to control for multiple comparisons. All P values used 2-tailed tests; < 0.05 was considered significant and 95% confidence intervals (CIs) were used. Analyses were conducted using SAS, Version 9.1.3 (SAS Institute Inc, Cary, NC).
The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the paper, and its final contents. The study was performed as a deliverable under Contract No. HHSN267200715003C (National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Bethesda, Maryland). The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy or interpretation of the US government.
Results
Comparisons of dialysis (n = 2390), advanced CKD (n = 29,319), and non-CKD (n = 274,777) patients regarding demographics and medical history, admission variables, therapeutic strategies, and in-hospital events are summarized in Tables 1–4, respectively. Of note, demographic and medical characteristics of advanced CKD patients were a “blend” of high-risk features of dialysis and non-CKD patients. Demographic characteristics were similar to non-CKD patients (predominantly white and male), but cardiovascular risk factors (prevalence of diabetes, hypertension, previous smoking) were similar to dialysis patients. Advanced CKD patients with AMI were demographically distinct (Table 1): older than dialysis or non-CKD patients (mean age 75.3 ± 12 years) and predominantly white (78.6%), unlike dialysis patients, who have disproportionate representation of non-white ethnicities. Prevalence of hypertension (71.4%) and diabetes (51.9%) was similar to dialysis patients (77.7% and 58.5%) but significantly higher than non-CKD patients (56.4% and 28.2%). Notably, compared with non-CKD patients, advanced CKD patients had significantly higher prevalence of heart failure (47% vs. 17.7%), stroke (19.9% vs. 11.1%), and previous AMI (37.3% vs. 25.1%).
Table 1.
Demographic and Medical Characteristics of Dialysis, Advanced Chronic Kidney Disease, and Non-Chronic Kidney Disease Patients
| Characteristics, n (%) | Dialysis | Advanced CKD | Non-CKD | P |
|---|---|---|---|---|
| n | 2390 | 29,319 | 274,777 | |
| Demographic | ||||
| Men | 1253 (52.4) | 16,347 (55.8) | 156,295 (56.9) | <0.001 |
| Age, yrs, mean ± SD | 67.7 ± 12.0 | 75.3 ± 12.0 | 70.2 ± 14.1 | <0.001 |
| < 55 | 397 (16.6) | 2057 (7.0) | 46,695 (17.0) | <0.001 |
| 55–< 65 | 479 (20.0) | 3291 (11.2) | 46781 (17.0) | |
| 65–< 75 | 757 (31.7) | 7163 (24.4) | 64545 (23.5) | |
| ≥ 75 | 757 (31.7) | 16,808 (57.3) | 116,733 (42.5) | |
| White | 1478 (62.4) | 22,905 (78.6) | 232,071 (85.0) | <0.001 |
| African American | 588 (24.8) | 3585 (12.3) | 19,007 (7.0) | |
| Medical | ||||
| Current smoker | 258 (10.8) | 3201 (10.9) | 63,861 (23.2) | <0.001 |
| Diabetes | 1399 (58.5) | 15,226 (51.9) | 77,516 (28.2) | <0.001 |
| Past myocardial infarction | 628 (26.3) | 10,928 (37.3) | 68,914 (25.1) | <0.001 |
| Hypertension | 1857 (77.7) | 20,943 (71.4) | 155,059 (56.4) | <0.001 |
| Hypercholesterolemia | 439 (18.4) | 7160 (24.4) | 79,183 (28.8) | <0.001 |
| Family history | 297 (12.4) | 4351 (14.8) | 65,872 (24.0) | <0.001 |
| Angina | 360 (15.1) | 5161 (17.6) | 36,449 (13.3) | <0.001 |
| Congestive heart failure | 756 (31.6) | 13,775 (47.0) | 48,505 (17.7) | <0.001 |
| Percutaneous coronary intervention | 231 (9.7) | 2873 (9.8) | 27,422 (10.0) | >0.99 |
| Coronary artery bypass surgery | 482 (20.2) | 5854 (20.0) | 37,125 (13.5) | <0.001 |
| Stroke | 384 (16.1) | 5839 (19.9) | 30,552 (11.1) | <0.001 |
CKD, chronic kidney disease; SD, standard deviation.
Table 4.
In-Hospital Variables and Clinical Events for Dialysis, Chronic Kidney Disease, and Non-Chronic Kidney Disease Patients
| Characteristics, n (%) | Dialysis | Advanced CKD | Non-CKD | P |
|---|---|---|---|---|
| n | 2390 | 29,319 | 274,777 | |
| Creatinine kinase-MB > diagnostic level | 1487 (68.3) | 19,451 (72.2) | 200,080 (77.9) | <0.001 |
| Troponin > diagnostic level | 1749 (84.8) | 20,365 (81.9) | 172,478 (77.7) | <0.001 |
| Ejection fraction, % | ||||
| < 40 | 663 (27.7) | 8079 (27.6) | 52,681 (19.2) | <0.001 |
| ≥ 40 | 859 (35.9) | 9096 (31.0) | 127,139 (46.3) | |
| Missing | 868 (36.3) | 12,144 (41.4) | 94,957 (34.6) | |
| In-hospital clinical events | ||||
| Mortality | 518 (21.7) | 6729 (23.0) | 34,599 (12.6) | <0.001 |
| Bleeding | 104 (4.4) | 1444 (4.9) | 8128 (3.0) | <0.001 |
| Stroke | 32 (1.3) | 412 (1.4) | 3667 (1.3) | >0.99 |
| Atrial fibrillation | 350 (14.6) | 4691 (16.0) | 33,284 (12.1) | <0.001 |
| Angina | 177 (7.4) | 2387 (8.1) | 23,719 (8.6) | 0.105 |
| Unexpected cardiac arrest | 295 (12.3) | 2614 (8.9) | 16,495 (6.0) | <0.001 |
| CHF/pulmonary edema | 617 (25.8) | 12,014 (41.0) | 57,883 (21.1) | <0.001 |
| Hypotension | 602 (25.2) | 5947 (20.3) | 40,978 (14.9) | <0.001 |
| Recurrent myocardial infarction | 40 (1.7) | 604 (2.1) | 4358 (1.6) | <0.001 |
| Shock | 151 (6.3) | 2041 (7.0) | 13,771 (5.0) | <0.001 |
| Sustained ventricular tachycardia/fibrillation | 212 (8.9) | 1806 (6.2) | 17,819 (6.5) | <0.001 |
| In-hospital outcome | ||||
| Length of stay, days, mean ± SD | 8.4 ± 8.2 | 8.3 ± 7.7 | 6.4 ± 5.9 | <0.001 |
CHF, congestive heart failure.
Advanced CKD patients (creatinine ≥ 2.5 mg/dL) comprised 9.6% of the NRMI-3 population. Using the Modification of Diet in Renal Disease (MDRD) Study equation, creatinine 2.5 mg/dL, mean age 75 years, and white ethnicity, eGFR in our study cohort predominantly was ≤ 25 mL/min/1.73 m2 for men and ≤ 19 mL/min/1.73 m2 for women. Thus, this cohort represented patients with stages 4 and 5 (nondialysis) CKD. The non-CKD cohort represented patients with normal renal function or stages 1–3 CKD.
The diagnostic suspicion for AMI or acute coronary syndrome (ACS) was significantly lower in advanced CKD patients (Table 2); only 23.1% of patients with AMI were accurately diagnosed on admission, compared with 39.6% of non-CKD patients. Among advanced CKD patients with AMI, the initial diagnosis on admission for 44% was suspected to be a condition other than ACS, compared with 25.8% of non-CKD patients. Just 40.4% of advanced CKD patients presented with chest pain as a symptom, compared with 61.6% of non-CKD patients. A greater proportion of advanced CKD than non-CKD patients were categorized as Killip class II/III on presentation (50.6% vs. 25.7%), signifying higher prevalence of CHF.
Table 2.
Admission Variables for Dialysis, Chronic Kidney Disease, and Non-Chronic Kidney Disease Patients
| Characteristics, n (%) | Dialysis | Advanced CKD | Non-CKD | P |
|---|---|---|---|---|
| n | 2390 | 29,319 | 274,777 | |
| Presentation | ||||
| Chest pain | 963 (41.1) | 11,605 (40.4) | 166,036 (61.6) | <0.001 |
| Killip class | ||||
| No congestive heart failure | 1382 (57.8) | 14,005 (47.8) | 200,126 (72.8) | <0.001 |
| Rales, jugular vein distension | 610 (25.5) | 9705 (33.1) | 48,347 (17.6) | |
| Pulmonary edema | 367 (15.4) | 5127 (17.5) | 22,237 (8.1) | |
| Cardiogenic shock | 31 (1.3) | 482 (1.6) | 4067 (1.5) | |
| Clinical factors | ||||
| Diagnosis on admission | ||||
| Myocardial infarction | 467 (19.8) | 6682 (23.1) | 107,642 (39.6) | <0.001 |
| Rule out myocardial infarction | 554 (23.5) | 6934 (24.0) | 63,358 (23.3) | |
| Unstable angina | 214 (9.1) | 2565 (8.9) | 30,603 (11.3) | |
| Other | 1125 (47.7) | 12,721 (44.0) | 70036 (25.8) | |
| Pulse, bpm, mean ± SD | 94.7 ± 23.6 | 93.5 ± 24.5 | 88.5 ± 24.8 | <0.001 |
| Systolic BP, mmHg, mean ± SD | 142.7 ± 39.7 | 141.9 ± 36.3 | 143.6 ± 32.5 | <0.001 |
| First 12-lead ECG results | ||||
| ST elevation | 418 (17.6) | 4641 (15.9) | 88,793 (32.5) | <0.001 |
| ST depression | 636 (26.7) | 7640 (26.2) | 75,526 (27.6) | <0.001 |
| Nonspecific | 1079 (45.3) | 13,186 (45.2) | 101,704 (37.2) | <0.001 |
| Q wave | 129 (5.4) | 1647 (5.6) | 23,938 (8.8) | <0.001 |
| LBBB | 194 (8.2) | 3205 (11.0) | 17,655 (6.5) | <0.001 |
| Normal | 158 (6.6) | 1866 (6.4) | 21,566 (7.9) | <0.001 |
| Other | 615 (25.8) | 7893 (27.1) | 53,137 (19.4) | <0.001 |
| ST elevation/LBBB | 589 (24.6) | 7564 (25.8) | 104,176 (37.9) | <0.001 |
| Location of infarction | ||||
| Anterior/septal | 390 (16.3) | 5005 (17.1) | 62,228 (22.6) | <0.001 |
| Inferior | 414 (17.3) | 4639 (15.8) | 77,677 (28.3) | <0.001 |
| Posterior | 47 (2.0) | 570 (1.9) | 11,305 (4.1) | <0.001 |
| Lateral | 220 (9.2) | 2796 (9.5) | 32,963 (12.0) | <0.001 |
| Right ventricle involvement | 11 (0.5) | 100 (0.3) | 1696 (0.6) | <0.001 |
| Unspecified/other | 1510 (63.2) | 19,046 (65.0) | 126,638 (46.1) | <0.001 |
| Q wave | 528 (22.1) | 6205 (21.2) | 97,442 (35.5) | <0.001 |
ECG, electrocardiogram; LBBB, left bundle-branch block.
Admission ECG data demonstrated clinically relevant differences between study cohorts. ST-segment elevation occurred about twice as frequently in non-CKD as in CKD patients (32.5% vs. 15.9 %). Left bundle branch block (LBBB) prevalence, however, was higher in advanced CKD patients (11% vs. 6.5%), likely reflecting advanced conduction system disease in older patients with a greater burden of comorbid CVD. Not surprisingly, a higher proportion of advanced CKD and dialysis patients (about 45% each) had nonspecific changes on admission ECG, likely reflecting nonspecific repolarization abnormalities related to left-ventricular hypertrophy or electrolyte fluxes.
A similar and significantly lower number of advanced CKD and dialysis patients (9.4% and 8.9% respectively) than non-CKD patients (22.5%) were considered eligible for acute coronary reperfusion based on NRMI criteria (Table 3). Of these, 884/2754 advanced CKD patients (32.1%) and 43,814/61,843 non-CKD patients (70.8%) received immediate reperfusion. Thus, among patients meeting NRMI eligibility criteria, only half as many advanced CKD as non-CKD patients received immediate coronary reperfusion. If NRMI exclusions were eliminated, and all CKD patients with ST-segment elevation AMI (STEMI) on first ECG, time from symptom onset to evaluation < 12 hours, and no contraindications to reperfusion were included, the proportion of advanced CKD patients eligible for immediate reperfusion minimally increased from 9.4% to 10.2%.
Table 3.
Therapeutic Strategies for Dialysis, Chronic Kidney Disease, and Non-Chronic Kidney Disease Patients
| Characteristics, n (%) | Dialysis | Advanced CKD | Non-CKD | P |
|---|---|---|---|---|
| n | 2390 | 29,319 | 274,777 | |
| Medications on admission | ||||
| Aspirin | 1629 (68.2) | 20,924 (71.4) | 225,340 (82.0) | <0.001 |
| Beta blockers | 1044 (43.7) | 12,654 (43.2) | 144,295 (52.5) | <0.001 |
| ACE inhibitors | 478 (20.0) | 7878 (26.9) | 69,201 (25.2) | <0.001 |
| Antiplatelet | 202 (8.5) | 2822 (9.6) | 49,833 (18.1) | <0.001 |
| Heparin (unfractionated) | 1062 (44.4) | 14,229 (48.5) | 180,697 (65.8) | <0.001 |
| Heparin (low molecular weight) | 239 (10.0) | 3739 (12.8) | 30,564 (11.1) | <0.001 |
| Glycoprotein 2B/3A antagonist | 113 (4.7) | 1365 (4.7) | 42,101 (15.3) | <0.001 |
| Reperfusion eligible | 212 (8.9) | 2754 (9.4) | 61,843 (22.5) | <0.001 |
| No immediate reperfusion strategy | 121 (57.1) | 1870 (67.9) | 18,029 (29.2) | <0.001 |
| Thrombolysis | 56 (26.4) | 609 (22.1) | 25942 (41.9) | |
| Primary PCI | 33 (15.6) | 260 (9.4) | 17,265 (27.9) | |
| Immediate CABG | 2 (0.9) | 15 (0.5) | 607 (1.0) | |
| Contraindications (non-reperfused) | n = 2225 | n = 27,583 | n = 207,035 | |
| Symptom onset to door > 12 hrs | 131 (5.9) | 1949 (7.1) | 18,927 (9.1) | <0.001 |
| Contraindications* | 420 (18.9) | 5361 (19.4) | 28,225 (13.6) | <0.001 |
| Major organ failure | 332 (14.9) | 2215 (8.0) | 2813 (1.4) | <0.001 |
| Quality of life | 257 (11.6) | 4396 (15.9) | 18,940 (9.1) | <0.001 |
| History of CVA | 140 (6.3) | 2218 (8.0) | 10663 (5.2) | <0.001 |
| Recent surgery/trauma | 138 (6.2) | 1198 (4.3) | 9374 (4.5) | 0.012 |
| Traumatic CPR | 60 (2.7) | 410 (1.5) | 3534 (1.7) | 0.002 |
| Severe hypertension | 29 (1.3) | 352 (1.3) | 1469 (0.7) | <0.001 |
| Active internal bleeding | 152 (6.8) | 2124 (7.7) | 8495 (4.1) | <0.001 |
| Medications at discharge† | n = 1872 | n = 22,590 | n = 240,178 | |
| Aspirin | 1238 (66.1) | 15,096 (66.8) | 188,005 (78.3) | <0.001 |
| Beta blockers | 1052 (56.2) | 11,953 (52.9) | 150,496 (62.7) | <0.001 |
| ACE inhibitor (EF < 40%) | 198 (40.1) | 2648 (43.1) | 27,516 (62.4) | <0.001 |
| Procedural interventions | ||||
| Coronary angiography | 621 (26.0) | 4909 (16.7) | 91,297 (33.2) | <0.001 |
| Coronary revascularization‡ | 290 (12.1) | 2726 (9.3) | 66,944 (24.4) | <0.001 |
| Percutaneous procedures§ | 198 (8.3) | 1722 (5.9) | 45,277 (16.5) | <0.001 |
| CABG (excluding immediate) | 96 (4.0) | 1054 (3.6) | 22,748 (8.3) | <0.001 |
| Intra-aortic balloon pump | 50 (2.1) | 651 (2.2) | 11,799 (4.3) | <0.001 |
| All catheterizations | 772 (32.3) | 6315 (21.5) | 139,078 (50.6) | <0.001 |
| All PTCA | 283 (11.8) | 2471 (8.4) | 76,936 (28.0) | <0.001 |
| All CABG | 100 (4.2) | 1115 (3.8) | 24,304 (8.8) | <0.001 |
ACE, angiotensin-converting enzyme; CABG, coronary artery bypass surgery; CPR, cardiopulmonary resuscitation; CVA, cerebrovascular accident; EF, ejection fraction; PCI, percutaneous coronary intervention; PTCA, percutaneous transluminal coronary angioplasty.
Active internal bleeding, history of CVA, recent surgery/trauma, intracranial neoplasm, or severe uncontrolled hypertension.
Excludes patients who did not survive to discharge.
Elective PTCA and/or CABG (excluding immediate).
Includes elective or rescue PCI.
In-hospital rates of coronary angiography and revascularization were also higher in non-CKD than in advanced CKD patients (coronary angiography, 33.2% vs. 16.7%; PTCA, 28% vs. 8.4%; CABG, 8.8% vs. 3.8%). Evidence-based medical therapies were used less commonly in advanced CKD than in non-CKD patients. Interestingly, utilization rates of medical therapies were similar in advanced CKD and dialysis patients. On admission, aspirin was administered to 71.4% of advanced CKD, 68.2% of dialysis, and 82% of non-CKD patients. ACE inhibitors for patients with ejection fraction < 40% were administered at discharge to 43.1% of advanced CKD, 40.1% of dialysis, and 62.4% of non-CKD patients.
Although higher prevalence of adverse clinical outcomes in advanced CKD than in non-CKD patients is not surprising, the similar prevalence in dialysis patients is noteworthy (Table 4). In-hospital mortality for advanced CKD patients was 23%, highest of the 3 groups and almost twice as high as for non-CKD patients (12.6%). The rate of unexpected cardiac arrest was intermediate in advanced CKD relative to dialysis and non-CKD patients (8.9% vs. 12.3% vs. 6% respectively). Rates of pulmonary edema/CHF (41%) and major bleeding (4.9%) were highest among advanced CKD relative to dialysis and non-CKD patients.
Using non-CKD patients as the reference group, the adjusted likelihood of in-hospital mortality for advanced CKD patients was somewhat lower (odds ratio 1.44, 95% confidence interval 1.39–1.49) than for dialysis patients (1.55, 1.39–1.74; Table 5). However, the adjusted likelihood of recurrent AMI (1.30, 1.19–1.42) and of major bleeding (1.62, 1.53–1.73) was higher.
Table 5.
Adjusted Odds Ratio and 95% Confidence Intervals of In-Hospital Clinical Events
| Death | Recurrent AMI | Stroke | Major Bleeding | |
|---|---|---|---|---|
| Dialysis | 1.55 (1.39–1.74) | 1.14 (0.82–1.57) | 0.94 (0.66–1.34) | 1.41 (1.16–1.73) |
| Advanced CKD | 1.44 (1.39–1.49) | 1.30 (1.19–1.42) | 0.94 (0.85–1.05) | 1.62 (1.53–1.73) |
| Non-CKD | Reference | Reference | Reference | Reference |
| c-statistic | 0.82 | 0.61 | 0.67 | 0.65 |
| P* | <0.001 | <0.001 | 0.52 | <0.001 |
AMI, acute myocardial infarction.
Type 3 test for group variables.
Discussion
Observational data from this large cohort of AMI patients with advanced CKD highlight the salient clinical characteristics of these high-risk patients, their striking differences relative to patients without advanced CKD, and their remarkable similarity to dialysis patients. Albeit well established that advanced CKD patients have poor outcomes after AMI, these data draw attention to key clinical characteristics that likely contribute significantly to poor outcomes. Moreover, these data demonstrate that patients with advanced CKD, comprising a significant proportion of patients with AMI, arguably carry the highest risk for in-hospital cardiovascular morbidity and mortality.
Prevalence of CKD in AMI patients has been variably reported using registry data, contingent upon the defining criteria. Han et al studied CKD patients with non-ST-elevation AMI (NSTEMI) in the CRUSADE registry; using a cut-off of creatinine > 2 mg/dL, they reported prevalence of almost 15%. Prevalence of CKD using eGFR values is predictably higher. Using the ACTION registry, Fox et al estimated that nearly 30% of patients with STEMI and 43% with NSTEMI had CKD stage 3 or higher (eGFR < 60 mL/min/1.73 m2). Similarly, Szummer et al reported moderate or higher CKD (eGFR < 60 mL/min/1.73 m2) in about 24% of NSTEMI patients from the SWEDEHEART registry. NRMI abstractors used a cut-off of stable creatinine > 2.5 mg/dL, thus identifying the largest reported national cohort of patients with advanced CKD and AMI (comprising 9% of the entire population). Imputation using the MDRD Study equation suggests that eGFR in our study population corresponded to stages 4 and 5 of the National Kidney Foundation Kidney Diseases Outcomes Quality Initiative CKD definition.
This study offers the unique opportunity to simultaneously compare characteristics of advanced CKD patients with dialysis and non-CKD patients. Several clinically pertinent themes emerge. Despite high pretest probability of CAD based on demographics and cardiovascular risk factors, clinician diagnostic suspicion and accuracy for the diagnosis of AMI were markedly low in advanced CKD patients; 44% of those presenting with AMI were diagnosed on admission with a condition other than ACS. Several factors likely contributed to this initial diagnostic challenge.
Firstly, only 40% of advanced CKD patients with AMI complained of chest pain as the presenting symptom. Similar findings have been reported in the dialysis population; these observations are concordant with data from Sosnov et al, who reported that patients with renal disease experiencing AMI were significantly less likely to report chest pain and more likely to report shortness of breath. The authors suggested that kidney disease alters the clinical presentation of AMI. High prevalence of diabetes in this population may contribute, masking “typical” symptoms of AMI. Nonetheless, this is an important factor for clinicians to reconcile in the care of this high-risk population.
Second, in our study a higher proportion of advanced CKD patients (again similar to dialysis patients) than non-CKD patients had clinical evidence for CHF during their index presentation. Shortness of breath was the index symptom for most of these patients (congruent with data from Sosnov et al), which was likely not recognized by clinicians as an “anginal equivalent.” Data from the CRUSADE and ACTION registries also note a high prevalence of CHF in this population. Among patients with AMI and left-ventricular dysfunction, Avanekar et al reported a higher proportion with reduced GFR values presenting with Killip class > I, congruent with our data. Notably, in-hospital CHF rates were also significantly higher in advanced CKD patients in our study (41% vs. 25.8% in dialysis and 21.1% in non-CKD patients). After adjustment for demographic characteristics, Fox et al reported about a 3-fold higher likelihood of developing in-hospital CHF in stage 4 CKD patients than in non-CKD patients.
Thirdly, our study found a noteworthy difference in ECG findings for patients with and without significant renal disease. ST-segment elevation was present in only 15.9% of advanced CKD patients presenting with AMI, compared with 32.5% of non-CKD patients. Using a combination of ST-segment elevation and LBBB, 25.8% of advanced CKD patients met criteria for STEMI compared with 37.9% of non-CKD patients. The proportions of advanced CKD and dialysis patients with ST-segment elevation/LBBB on initial ECG were identical. Although the possibility that diagnostic ST-segment changes were masked by prominent repolarization changes (related to left-ventricular hypertrophy or electrolyte fluxes) cannot be discounted, absence of ECG evidence of transmural ischemia in most patients with renal disease raises the interesting possibility of fundamental differences in the pathophysiology of AMI. This observation deserves future prospective evaluation.
An important implication of ECG as a diagnostic tool in AMI is its direct role in triaging for acute coronary reperfusion, a pivotal therapy impacting future outcomes. Based on NRMI criteria, we found a significant difference in patients meeting eligibility criteria for immediate coronary reperfusion strategies. Only 9.4% of advanced CKD patients met criteria, of whom 32.1% received some form of immediate coronary reperfusion. In comparison, 22.5% of non-CKD patients were eligible, of whom 70.8% received immediate coronary reperfusion therapy. Rates of immediate coronary reperfusion were identical for advanced CKD and dialysis patients. In both groups, immediate reperfusion therapy was contraindicated for a significant number of patients due to major organ (renal) failure (8% and 14.9% respectively) or major bleeding (7.7% and 6.8% respectively). Significantly lower rates of non-immediate coronary angiography and revascularization in advanced CKD patients (lower than in dialysis patients) likely reflect hesitancy to subject advanced CKD patients to the risk of contrast nephropathy. Fox et al and Szummer et al reported a similar trend: an inverse gradient between advancing degrees of CKD and likelihood of coronary revascularization among patients with AMI.12 Comparable to findings in our study, Fox et al noted that NSTEMI patients with stage 4 CKD were least likely to undergo coronary revascularization (40% less likely than non-CKD patients), followed by patients with stage 5 CKD or on dialysis (20% less likely). Similarly, use of other medical therapies considered evidence-based (including aspirin, beta blockers, ACE inhibitors in patients with left-ventricular dysfunction) was considerably lower in advanced CKD than in non-CKD patients, but similar to dialysis patients. This observation has been consistent in multiple studies.
Unadjusted in-hospital mortality rates were highest for patients with advanced CKD, likely attributable to differences in clinical characteristics (particularly older age) and therapeutic strategies. Fox et al similarly reported the highest unadjusted hazard for in-hospital mortality in NSTEMI patients with stage 4 CKD, but lower adjusted mortality rates compared with dialysis patients. Importantly, in-hospital rates of unexpected cardiac arrest in advanced CKD patients with AMI were significantly higher than in non-CKD patients (8.9% vs. 6%). This observation deserves further evaluation. In a cohort of 19,440 patients undergoing cardiac catheterization, Pun et al reported a progressive increase in rate of sudden cardiac death associated with worsening eGFR values, suggesting that CKD patients in general have a greater predisposition to sudden death, likely due to structural cardiovascular changes.
Increased rates of in-hospital bleeding (4.9%) relative to non-CKD and dialysis patients (3% and 4.4%, respectively) emphasize the higher propensity for bleeding in advanced CKD patients. These findings are consistent with data from studies that identified low eGFR values as associated with increased risk of bleeding. Among stage 4 CKD patients, NSTEMI patients have a 3-fold higher, and STEMI patients about a 2-fold higher, adjusted risk of major bleeding. Platelet dysfunction related to advanced renal disease likely plays a role.
This study has inherent limitations due to its retrospective, observational nature, with potential for biases and unmeasured cofounders. Lack of eGFR data is also a limiting factor. In conclusion, the clinical characteristics of CKD patients with AMI likely contribute to poor cardiovascular outcomes. Advanced CKD patients with AMI are high-risk, clinically similar to dialysis patients with AMI, but distinctly different from patients without significant CKD. Clinicians must be vigilant in recognizing AMI in patients with advanced CKD, maintaining a greater diagnostic suspicion and awareness of the clinical presentation and profile of AMI in patients with advanced renal disease.
Acknowledgments
Paul Frederick is employed by ICON Clinical Research, a contract research organization that receives research funding from Genentech, Inc. The other authors report no conflicts of interest. The authors thank USRDS colleagues Shane Nygaard, BA, for manuscript preparation, and Nan Booth, MSW, MPH, ELS, for manuscript editing.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Coresh J, Selvin E, Stevens LA, et al. Prevalence of chronic kidney disease in the united states. JAMA. 2007;298:2038–47. doi: 10.1001/jama.298.17.2038. [DOI] [PubMed] [Google Scholar]
- 2.Fox CS, Muntner P, Chen AY, et al. Use of evidence-based therapies in short-term outcomes of ST-segment elevation myocardial infarction and non-ST-segment elevation myocardial infarction in patients with chronic kidney disease: A report from the national cardiovascular data acute coronary treatment and intervention outcomes network registry. Circulation. 2010;121:357–65. doi: 10.1161/CIRCULATIONAHA.109.865352. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Go AS, Chertow GM, Fan D, et al. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med. 2004;351:1296–305. doi: 10.1056/NEJMoa041031. [DOI] [PubMed] [Google Scholar]
- 4.US Renal Data System. USRDS 2010 annual data report: Atlas of end-stage renal disease in the united states. bethesda, md: National institutes of health, national institute of diabetes and digestive and kidney diseases; 2010. [Google Scholar]
- 5.Herzog CA, Ma JZ, Collins AJ. Poor long-term survival after acute myocardial infarction among patients on long-term dialysis. N Engl J Med. 1998;339:799–805. doi: 10.1056/NEJM199809173391203. [DOI] [PubMed] [Google Scholar]
- 6.Wright RS, Reeder GS, Herzog CA, et al. Acute myocardial infarction and renal dysfunction: A high-risk combination. Ann Intern Med. 2002;137:563–70. doi: 10.7326/0003-4819-137-7-200210010-00007. [DOI] [PubMed] [Google Scholar]
- 7.Anavekar NS, McMurray JJ, Velazquez EJ, et al. Relation between renal dysfunction and cardiovascular outcomes after myocardial infarction. N Engl J Med. 2004;351:1285–95. doi: 10.1056/NEJMoa041365. [DOI] [PubMed] [Google Scholar]
- 8.Fox KA, Bassand JP, Mehta SR, et al. Influence of renal function on the efficacy and safety of fondaparinux relative to enoxaparin in non ST-segment elevation acute coronary syndromes. Ann Intern Med. 2007;147:304–10. doi: 10.7326/0003-4819-147-5-200709040-00005. [DOI] [PubMed] [Google Scholar]
- 9.Han JH, Chandra A, Mulgund J, et al. Chronic kidney disease in patients with non-ST-segment elevation acute coronary syndromes. Am J Med. 2006;119:248–54. doi: 10.1016/j.amjmed.2005.08.057. [DOI] [PubMed] [Google Scholar]
- 10.Sosnov J, Lessard D, Goldberg RJ, et al. Differential symptoms of acute myocardial infarction in patients with kidney disease: A community-wide perspective. American Journal of Kidney Diseases. 2006;47:378–84. doi: 10.1053/j.ajkd.2005.11.017. [DOI] [PubMed] [Google Scholar]
- 11.Herzog CA, Littrell K, Arko C, et al. Clinical characteristics of dialysis patients with acute myocardial infarction in the United States: A collaborative project of the United States Renal Data System and the National Registry Of Myocardial Infarction. Circulation. 2007;116:1465–72. doi: 10.1161/CIRCULATIONAHA.107.696765. [DOI] [PubMed] [Google Scholar]
- 12.Szummer K, Lundman P, Jacobson SH, et al. Influence of renal function on the effects of early revascularization in non-ST-elevation myocardial infarction: Data from the swedish web-system for enhancement and development of evidence-based care in heart disease evaluated according to recommended therapies (SWEDEHEART) Circulation. 2009;120:851–8. doi: 10.1161/CIRCULATIONAHA.108.838169. [DOI] [PubMed] [Google Scholar]
- 13.National Kidney F. K/DOQI clinical practice guidelines for chronic kidney disease: Evaluation, classification, and stratification. American Journal of Kidney Diseases. 2002;39:S1–266. [PubMed] [Google Scholar]
- 14.McCullough PA, Sandberg KR, Borzak S, et al. Benefits of aspirin and beta-blockade after myocardial infarction in patients with chronic kidney disease. Am Heart J. 2002;144:226–32. doi: 10.1067/mhj.2002.125513. [DOI] [PubMed] [Google Scholar]
- 15.Pun PH, Smarz TR, Honeycutt EF, et al. Chronic kidney disease is associated with increased risk of sudden cardiac death among patients with coronary artery disease. Kidney Int. 2009;76:652–8. doi: 10.1038/ki.2009.219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Mehran R, Pocock SJ, Nikolsky E, et al. A risk score to predict bleeding in patients with acute coronary syndromes. J Am Coll Cardiol. 2010;55:2556–66. doi: 10.1016/j.jacc.2009.09.076. [DOI] [PubMed] [Google Scholar]
- 17.Subherwal S, Bach RG, Chen AY, et al. Baseline risk of major bleeding in non-ST-segment-elevation myocardial infarction: The CRUSADE (can rapid risk stratification of unstable angina patients suppress ADverse outcomes with early implementation of the ACC/AHA guidelines) bleeding score. Circulation. 2009;119:1873–82. doi: 10.1161/CIRCULATIONAHA.108.828541. [DOI] [PMC free article] [PubMed] [Google Scholar]