Heart Rate Recovery After Exercise Is a Predictor of Silent Myocardial Ischemia in Patients With Type 2 Diabetes

A total of 98 consecutive patients with type 2 diabetes and no chest symptoms were studied. They had electrocardiographic abnormalities or at least two risk factors for CAD in addition to diabetes, and presented to Toshiba Hospital between September 2005 and December 2008. Patients were excluded if they had a pacemaker, congestive heart failure, cardiomyopathy, β-blocker or digitalis therapy, congenital valvular heart disease, or left bundle branch block. As a result, 87 patients underwent a treadmill test and single-photon emission computed tomography (SPECT) with thallium scintigraphy.

Exercise stress tests were done on a motorized treadmill (Quinton Q-STRESS TM55, Cardiac Science, Bothell, WA) using the standard Bruce protocol. Patients were encouraged to perform maximal exercise. Testing was terminated after the patient reached the target heart rate (based on age) or because of fatigue, dyspnea, leg discomfort, systolic blood pressure >250 mmHg, ventricular tachycardia, or ischemic electrocardiographic changes. After peak workload was achieved, HRR was calculated as the decrease of the heart rate from its peak during exercise to that at 1 min after finishing the exercise.

Thallium SPECT imaging (Toshiba GCA-7200A, Tokyo, Japan) was performed according to the standards of the American Society of Nuclear Cardiology (7). Rest and stress images were obtained on the same day. Myocardial perfusion abnormalities were judged by two cardiologists, and SMI was diagnosed from abnormal myocardial perfusion images without associated symptoms. Patients with abnormal myocardial perfusion underwent coronary angiography (Toshiba Infinix Celeve-I INFX-8000C), and significant stenosis was defined as ≥75% diameter stenosis.

Continuous variables are presented as the mean ± SD. Differences between groups were compared with Student t test or the χ² test. Analyses were performed with StatView 5.0 software (SAS Institute, Cary, NC), and significance was accepted at P < 0.05.

The patients were divided into groups with or without SMI. Patients were 64 ± 10 years old, 21 were women (24.1%), and the mean BMI was 24.0 ± 3.2 kg/m². The mean duration of diabetes was 9.8 ± 6.6 years, 11 patients (12.6%) had a history of cardiovascular disease, and the mean HbA_1c was 7.4 ± 1.7%.

The mean resting heart rate was 83.7 ± 12.0 bpm, maximum heart rate was 138.0 ± 14.2 bpm, 1-min postexercise heart rate was 113.9 ± 14.4 bpm, 3-min postexercise heart rate was 91.9 ± 12.0 bpm, and mean maximum workload was 8.2 ± 2.1 METs. The exercise end point was leg fatigue in 29 (33%), diagnostic ST segment changes in 42 (48%), or shortness of breath in 17 (18%). Sixty patients (69%) achieved their target heart rate (≥85% of [220 − age]).

During the treadmill test, 42 patients (48%) showed ST depression, and 41 (47%) with abnormal perfusion defects on scintigraphy were diagnosed as having SMI. Thirty patients (34%) showed significant stenosis on coronary angiography.

There were no differences of clinical characteristics between the two groups (Table 1). The 1- and 3-min heart rates were similar in both groups, but the SMI group showed slower HRR (18 ± 6 vs. 30 ± 12 bpm; P < 0.0001) along with a higher resting heart rate (P = 0.01), lower maximum heart rate (P < 0.001), lower rate pressure product (P = 0.0032), and lower max METs. The number of patients who achieved the target heart rate was not significantly different between the two groups (P = 0.13).

Multivariate logistic regression analysis was performed to assess parameters significantly associated with myocardial ischemia, using the following variables: HRR, max METs, resting heart rate, maximum heart rate, rate pressure product, HbA_1c, use of sulfonamides, and history of cardiovascular disease. As a result, HRR was significantly associated with SMI (odds ratio 0.83 [95% CI 0.75–0.92], P = 0.0006) and was also significantly associated with significant angiographic stenosis, even after adjustment for the above covariates (0.84 [0.75–0.94]; P = 0.0017).

Many physicians screen asymptomatic persons by stress testing, but it has a low specificity (8). Furthermore, it has been reported that a decrease of the chronotropic reserve predicts CAD (9).

Our findings indicated that slow HRR after exercise strongly predicts SMI in patients with type 2 diabetes. Slow HRR similarly predicts myocardial ischemia at the microvascular and macrovascular levels.

Possible mechanisms of SMI include autonomic denervation of the myocardium (10–12), a higher pain threshold during exercise testing (13), higher β-endorphin levels (14), and increased production of anti-inflammatory cytokines that may block pain transmission and increase the neural activation threshold (15).

The prevalence of SMI may exceed 20% among asymptomatic patients with type 2 diabetes (3). In this study, the prevalence was a high 47%, probably because our cohort included patients with electrocardiogram abnormalities, at least two risk factors for CAD in addition to diabetes, or a history of CAD (13%).

This study had the limitations of being performed at a single institution and in a small patient population. Accordingly, the relation between HRR and SMI should be assessed by a larger study with pathophysiologic data in the future.

Although it is possible that our results were influenced by the difference of exercise parameters between the two groups, we conclude that HRR is useful for easy detection of SMI in high-risk type 2 diabetic patients to allow primary prevention and that HRR is significantly associated with SMI even after adjustment for the influence of exercise parameters.

No potential conflicts of interest relevant to this article were reported.

T.Ya. contributed to all study processes, including data research and manuscript preparation. T.Yo. contributed to discussion and wrote, reviewed, and edited the manuscript. K.M., T.L., and F.S contributed to discussion.