CLINICAL PRACTICE
Factors Affecting Cardiac Catheterization Rates in Elders with Acute Coronary Syndromes Jin H. Han, MD, MSc, Karen F. Miller, RN, MPA, Alan B. Storrow, MD
Abstract Background: Elder patients with acute coronary syndromes (ACS) are less likely to receive cardiac catheterization. The reasons for this are unclear. Objectives: To assess whether elder patients who had a documented history of dementia, lived in extended care facilities, or had do not intubate–do not resuscitate (DNR-DNI) advance directives were less likely to receive cardiac catheterization, despite having ACS with high-risk features. Methods: This was a medical record review conducted at an urban teaching hospital. DNR-DNI status before hospitalization, extended care facility (nursing home or assisted living) residence, and a previous diagnosis of dementia were obtained from the medical record. Patients 65 years and older who presented to the emergency department with acute myocardial infarction or with unstable angina with ST segment deviation were included. Univariate and multivariate logistic regression were performed, and odds ratios (ORs) were reported with their 95% confidence intervals (CIs). Results: Of the 201 eligible patients, 66 (32.8%) patients did not undergo cardiac catheterization. In the univariate analysis, patients who had dementia, resided in extended care facilities, or were DNR-DNI were less likely to receive cardiac catheterization. Only extended care facility residence (OR, 0.18; 95% CI = 0.04 to 0.83) and DNR-DNI status (OR, 0.19; 95% CI = 0.04 to 0.92) remained significantly associated with decreased cardiac catheterization in the multivariate analysis. Conclusions: Elder patients with ACS residing in extended care facilities or who are DNR-DNI are less likely to receive cardiac catheterization. Future studies concerning the quality of ACS care for elders should take these variables into account. ACADEMIC EMERGENCY MEDICINE 2007; 14:228–234 ª 2007 by the Society for Academic Emergency Medicine Keywords: elders, acute coronary syndromes, cognitive status, functional status
E
lders with acute coronary syndromes (ACSs) are at high risk for death and adverse cardiac events.1,2 Unfortunately, treatment guidelines specific to these patients are limited. There are few data on how factors unique to elders, such as life expectancy, comorbidities, cognitive function, or functional impairment, should affect medical decision making. The Ameri-
From the Department of Emergency Medicine, Vanderbilt University Medical Center (JHH, KFM, ABS), Nashville, TN. Received June 7, 2006; revisions received August 18, 2006, and September 26, 2006; accepted September 26, 2006. Presented at the Annual Meeting of the Society for Academic Emergency Medicine, San Francisco, CA, May 2006. Dr. Han is funded by the Vanderbilt Physicians Scientist Development Grant. Reprints are not available. Contact for correspondence: Jin H. Han, MD, MSc; e-mail: jin.h.
[email protected].
228
ISSN 1069-6563 PII ISSN 1069-6563583
can Heart Association–American College of Cardiology (AHA-ACC) guidelines state that decisions on management should reflect these considerations (level C), but this recommendation is based on expert consensus, rather than on objective data. Post hoc analyses of observational studies or registries have shown that elder patients are less likely to receive cardiac catheterization and other AHA-ACC guideline recommended therapies.1,3 Although patients older than 65 years of age were included in these studies, the presence of dementia, extended care facility residence, and advance directives were not recorded.4,5 In addition, these studies did not exclude patients with these characteristics from their analyses.4,5 Elder patients currently have the highest emergency department (ED) visit rate, and the number of elders visiting EDs has trended upwards over the past decade.6 With this population expected to grow exponentially over the next several years, more elders will be diagnosed with ACS in the ED. In conjunction with cardiology and internal medicine, emergency physicians will be
ª 2007 by the Society for Academic Emergency Medicine doi: 10.1197/j.aem.2006.09.054
ACAD EMERG MED
March 2007, Vol. 14, No. 3
www.aemj.org
challenged with making important management decisions in this high-risk patient population. To understand how to optimally manage elders with ACS, a better understanding is required of how elders’ comorbidities should affect medical decision making. We sought to explore how dementia, do not resuscitate–do not intubate (DNR-DNI) status, and extended care facility residence affect the decision to perform cardiac catheterization in patients older than 65 years who have high-risk ACS.
METHODS Study Design This was a retrospective cohort study that used chartreview methodology. The study was approved by the local institutional review board. Study Setting and Population This study was conducted at a single academic, tertiary care center, which has an annual ED census of 50,000 patients per year. Of patients presenting to the ED, approximately 10% are self-pay, and 45% have Medicaid insurance. The hospital has a cardiology fellowship with a full-service cardiac catheterization laboratory staffed 24 hours per day, 7 days per week. At this hospital, patients with ST elevation myocardial infarction (MI) preferentially go to the cardiac catheterization laboratory.7 We used the hospital information system to obtain medical record numbers of patients given codes from the International Classification of Diseases, Revision 9 of acute myocardial infarction or unstable angina (410.xx, 411.xx) between July 1, 2002 and June 30, 2005. Patients 65 years and older who had so-called hard signs of ACS were included in this analysis to remove any ambiguity about who should receive cardiac catheterization. These were patients with 1) a typical rise or fall in cardiac biomarkers of necrosis or 2) ST segment deviation not known to be old and suggestive of ischemia in the 12lead electrocardiogram. Patients with elevated cardiac troponins secondary to chronic kidney disease were included but required a typical rise or fall in cardiac biomarkers to meet inclusion criteria.8,9 Patients were excluded if they had non-ACS presentations or had ACS secondary to other illnesses. Our definition of ACS presentations was broad because elders with ACS are more likely to have atypical presentations.10,11 However, we excluded patients who presented to the ED with gastrointestinal bleeding, fever, or trauma as documented in the medical record. In addition, we excluded patients with ACS secondary to demand ischemia; some examples of excluded patients were those diagnosed with pneumonia (focal infiltrate on chest radiograph), pulmonary embolism (chest computed tomography angiography), sepsis (fever or positive blood cultures), pneumothorax (chest radiograph), intracranial hemorrhage (head computed tomography), intraabdominal pathology requiring urgent surgical intervention, or gastrointestinal bleeding by history or physical examination. Study Protocol Preceding the start of the study, reviewers were trained in data abstraction by using the electronic medical record
229
system and were trained in the study definition of ACS. All reviewers had clinical experience in the ED and had previous experience with data abstraction using medical records. Two reviewers (KFM and ABS) were blinded to study hypothesis. Two reviewers (KFM and JHH) classified each patient as ST elevation MI, non-ST elevation MI, or high-risk unstable angina based on either ST segment deviation on the 12-lead electrocardiogram or a characteristic rise or fall in cardiac-biomarkers of necrosis.8 There was disagreement in 7.0% of the patients’ diagnoses; these disagreements were adjudicated by a third reviewer (ABS). Independent and Dependent Variables Patient demographics, history, place of residence, advance directives, disposition, vital signs, symptoms, diagnostic tests, treatment, and discharge diagnosis were collected onto an electronic case report form by a single reviewer (KFM). All data elements were verified for accuracy. The independent variables of interest were a history of dementia that was documented in the medical record, DNR-DNI before hospitalization, and extended care facility residence. Because of the small sample sizes, nursing home (n = 12) and assisted living residence (n = 7) were combined and defined as extended care facilities. History, place of residence, and DNR-DNI status were initially obtained from the electronic problem list, which is updated after each clinic or hospital visit. The presence of these variables was then confirmed by using clinic, history and physical, and discharge notes in the electronic medical record. Electrocardiogram findings were interpreted by all reviewers when determining ACS status. If ST segment deviation was present, it was characterized as old or new on the basis of previous electrocardiograms. If no comparisons were available, then it was classified as ST segment deviation, no comparisons. The primary outcome variable was whether or not cardiac catheterization was performed. Cardiac catheterization was typically reported in the discharge note; therefore, it was not feasible to blind for this outcome in all cases. Data Analysis For continuous variables, medians and interquartile ranges were reported; comparisons were performed by using the Wilcoxon rank-sum test. Proportions were calculated for categorical variables, and comparisons were performed by using chi-square analysis or Fisher’s exact test. Initially, a univariate logistic regression was performed, testing the association between cardiac catheterization and the following covariates: dementia, extended care facility residence; DNR-DNI status; 12lead electrocardiogram findings; clinical presentation; and past history, including cardiac risk factors. For the purpose of the logistic regression analysis, ST elevation and ST depression were collapsed into dichotomous variables. Patients with ST elevation or depression that was new or with no comparisons were considered to have ST elevation or depression not known to be old. All variables with p < 0.15 were placed in a multivariate logistic regression analysis. By using stepwise selection and using a p = 0.15 as selection criteria, a final model was developed. Age was not placed in the multivariate model
230
Han et al.
Table 1 Reviewer Diagnosis of Patients Who Did and Did Not Receive Cardiac Catheterization Variable
No Cath, n (%)
Cath, n (%)
STEMI NSTEMI High-risk unstable angina
4 (6.1) 56 (84.9) 6 (9.1)
40 (29.6) 72 (53.3) 23 (17.0)
p-value
