We prospectively studied all individuals who underwent stress CMR in our unit between 2009 and 2013. The decision to refer patients for this specific test over others was based on the clinical criteria of the referring cardiologist and following current clinical practice guidelines.17 We excluded patients with classic contraindications for magnetic resonance imaging, such as claustrophobia, pacemaker or implantable cardioverter-defibrillator implantation, magnetic resonance imaging-unsafe objects, and chronic kidney disease (estimated glomerular filtration rate ≤ 30 mL/min/1.73 m2). Individuals were instructed to avoid caffeine-containing food and drinks 24hours prior to the CMR examination. The study was approved by the local research ethics committee (reference 096/2010) and all patients signed an informed consent form.

Clinical data such as cardiovascular risk factors, including body mass index, smoking history, dyslipidemia, diabetes mellitus, family history of IHD and previous cardiovascular disease; clinical parameters such as heart rate, blood pressure and baseline heart rate, and analytical measures, including low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, baseline glucose levels, glycosylated hemoglobin and estimated glomerular filtration rate, were recorded in the outpatient cardiology clinic. The HeartScore for a low-risk population (Spain) was used to estimate the cardiovascular risk of each patient. The diabetic population and those patients with known IHD were considered as being at very high risk. We calculated the risk in patients aged at least 70 years by extrapolating the HeartScore, as indicated by clinical practice guidelines.18

The stress CMR studies were carried out by a cardiologist and a radiologist on a 1.5 T magnet (MAGNETOM Symphony and MAGNETOM Aera, Siemens Healthcare GmbH, Erlangen, Germany), according to the 2008 recommendations of the Society for Cardiovascular Magnetic Resonance.19 A total dose of 0.2 mmol/kg of gadobutrol (Gadovist, Bayer Schering Pharma AG, Berlin-Wedding, Germany) was administered at 4 mL/s after maximum hyperemia induced by increasing doses of intravenous adenosine (140, 180, and 210 μg/kg/min) for 4-6minutes in order to achieve an appropriate response: an increase in heart rate of at least 10 bpm and/or decrease in systolic blood pressure by at least 10mmHg.20

The test was visually analyzed and classified as positive for ischemia if reversible perfusion defects were detected. The degree of hypoperfusion was classified as mild (1-2 affected segments), moderate (3-4 segments) or severe (more than 4 segments). As for the analysis, moderate and severe perfusion defects were jointly evaluated. In those segments with ischemic late gadolinium enhancement, ischemia was considered if there was a reversible perfusion defect greater than the scar extension.

Follow-up was scheduled every 3 to 6 months in the outpatient clinic or by telephone according to the cardiologist's criteria and patient preference until the end of follow-up or the first occurrence of an event: death from any cause, acute coronary syndrome (ACS), and/or need for revascularization. As our objective was to establish the diagnostic and prognostic values of stress CMR, we considered a combined reference pattern that included coronary angiography guided by the result of the test or the appearance of the cardiovascular events. The indication and timing of coronary angiography depended on the criteria of the cardiologist and patient preference.

Quantitative variables are expressed as mean and standard deviation or median and interquartile range in nonparametric data, and qualitative variables as number and percentage. Continuous quantitative variables were compared using the Student t test or the sum of Wilcoxon ranges in nonparametric data. Categorical variables were compared with chi-square test and Fisher's exact test. A significance level of .05 (bilateral) was established for all statistical tests.

The survival distribution related to time to an event was evaluated using the Kaplan-Meier method, analyzing the existence and degree of ischemia (mild vs moderate/severe). To compare survival curves, the log-rank test was employed. A multivariate Cox proportional hazards regression model was conducted to calculate the adjusted hazard ratios (HR) and to determine the effect of several variables on survival function. Univariate analysis was performed to select variables for the multivariate analysis. Variables with a result of P <.01 in the univariate analysis were selected for the multivariate analysis. In the multivariate analysis, a P value of <.05 was considered statistically significant. The statistical analysis was performed with Stata 13.1 (Stata Corporation, College Station, Texas, United States) and SPSS 21.0 (SPSS Inc, Chicago, Illinois, United States).

Of the initial cohort of 333 patients undergoing stress CMR (76.3% men, mean age 64.6±10.6 years), 223 patients (66.7%) were identified as being younger than 70 years and 110 (33%) as being at least 70 years old. The baseline characteristics of each group are shown in Table 1. Patients aged 70 years or older were predominantly male and showed a higher prevalence of previous smoking history, hypertension, diabetes, chronic kidney disease, previous neoplasia, peripheral arterial disease, stroke, and atrial fibrillation, thus resulting in a higher cardiovascular risk HeartScore. There were no statistically significant differences in the percentages of previous coronary disease, previous myocardial infarction, or in the other clinical parameters, although low-density lipoprotein cholesterol and the percentage of active smokers were lower in the elderly population.

The most common CMR indications were suspicion of coronary artery disease in very high cardiovascular risk patients or with typical chest pain (Table 2). The adenosine dose was similar in both groups (163.2±28.8 μg/kg/min and 160.5±27.2 μg/kg/min for patients at aged least 70 years and younger than 70 years, respectively; P=.39). There were no major complications. A minor complication occurred in the group older than 70 years (transient atrioventricular block) and 3 minor complications were observed in individuals younger than 70 years (transient atrioventricular block in 2 and a chest pain episode not requiring nitroglycerine in 1).

The stress CMR in patients younger than 70 years was negative for ischemia in 159 (71.3%) and positive in 64 (28.7%). The perfusion defect was mild in 28 individuals (43.7%) and moderate/severe in 36 patients (56.3%). In those at least 70 years, the result was negative in 65 (59.1%) and positive in 45 (40.9%), the perfusion defect being mild in 17 (37.8%) and moderate/severe in 28 (62.2%). There were no significant differences in the percentages of the presence of late gadolinium enhancement or in the development of contractility changes during the exam. A positive stress CMR result was more likely with age (odds ratio, 1.04; 95% confidence interval [95%CI], 1.02-1.07; P=.03).

In patients older than 70 years, those with a positive result were more frequently men with previous MI and peripheral arterial disease. There were no statistically significant differences in the rest of the baseline characteristics (Table 3). Moreover, there were no differences in CMR parameters, even in the presence of late gadolinium enhancement, according to the result of the test (Table 4).

Follow-up was available for all patients. After a median follow-up time of 26 [18-37] months 70 events were recorded, 35 events in each group. In patients at least 70 years, there were 15 deaths (4 in the positive CMR group), 10 ACS (8 in the positive CMR group), and 10 revascularizations (7 in the positive CMR group). In those younger than 70 years, 7 deaths, 12 ACS, and 16 revascularizations were recorded. Events were more likely to occur in elderly patients than in younger patients (odds ratio, 1.05; 95%CI, 1.02-1.08; P=.04).

In the survival analysis of patients older than 70 years, there were no significant differences according to the result of the stress CMR (positive vs negative, Kaplan-Meier survival curves, Log Rank test; P=.69). Significant differences were obtained depending on the degree of positivity (hypoperfusion) of the stress CMR (mild positive vs moderate/severe positive; Kaplan-Meier survival curves; log rank test; P=.003) (Figure 1 and Figure 2). In patients with a positive stress CMR test, those with a moderate or severe degree of ischemia had a higher risk of having an event adjusted for age (older or younger than 70 years), sex and cardiovascular risk (HR, 3.53; 95% CI 1.41-8.79; P=.01). Therefore, a positive stress CMR with a moderate or severe hypoperfusion defect predicted the appearance of cardiovascular events in the follow-up in patients older than 70 years.

Figure 1.

Kaplan-Meier survival curves of elderly patients according to the result of the stress CMR. The result was classified as positive or negative for ischemia if reversible perfusion defects (hypoperfusion) were detected. CMR, cardiac magnetic resonance.

(0.09MB).

Figure 2.

Kaplan-Meier survival curves of elderly patients according to the degree of myocardial hypoperfusion in the stress CMR. Hypoperfusion was classified as mild (≤ 2 affected segments) or moderate or severe (more than 2 segments). CMR, cardiac magnetic resonance.

(0.09MB).

Finally, in patients with a positive result in the stress CMR, the degree of ischemia was confirmed as an independent predictor of events (HR 2.81; 95%CI 1.11-7.11; P=.03), independently of previous MI, late gadolinium enhancement, age, cardiovascular risk, and sex.