Effects of αβ-Blocker Versus β1-Blocker Treatment on Heart Rate Response During Incremental Cardiopulmonary Exercise in Japanese Male Patients with Subacute Myocardial Infarction

A simplified substitute for heart rate (HR) at the anaerobic threshold (AT), i.e., resting HR plus 30 beats per minute or a percentage of predicted maximum HR, is used as a way to determine exercise intensity without cardiopulmonary exercise testing (CPX) data. However, difficulties arise when using this method in subacute myocardial infarction (MI) patients undergoing beta-blocker therapy. This study compared the effects of αβ-blocker and β1-blocker treatment to clarify how different beta blockers affect HR response during incremental exercise. MI patients were divided into αβ-blocker (n = 67), β1-blocker (n = 17), and no-β-blocker (n = 47) groups. All patients underwent CPX one month after MI onset. The metabolic chronotropic relationship (MCR) was calculated as an indicator of HR response from the ratio of estimated HR to measured HR at AT (MCR-AT) and peak exercise (MCR-peak). MCR-AT and MCR-peak were significantly higher in the αβ-blocker group than in the β1-blocker group (p < 0.001, respectively). Multiple regression analysis revealed that β1-blocker but not αβ-blocker treatment significantly predicted lower MCR-AT and MCR-peak (β = −0.432, p < 0.001; β = −0.473, p < 0.001, respectively). Based on these results, when using the simplified method, exercise intensity should be prescribed according to the type of beta blocker used.


Introduction
Many guidelines recommend exercise training for patients with myocardial infarction (MI) [1,2] because exercise training has been associated with reduced cardiovascular mortality after acute MI [3].
When prescribing exercise for subacute MI patients, it is important to determine the optimum exercise intensity to ensure safe and effective training. The recommended technique for determining the target heart rate (HR) is by measuring HR at the anaerobic threshold (AT) via cardiopulmonary exercise testing (CPX). Exercise prescription using HR at AT is effective and safe for MI patients [4,5]. Although prescribing exercise intensity based on the AT measured by CPX is desirable, it is not easy to perform CPX for all patients due to staffing requirements and the high cost of equipment. It is recommended to first perform an exercise stress test to determine the target HR range by the Karvonen formula, if gas-exchange analysis data cannot be obtained [2]. Meanwhile, a simplified substitute for HR at AT, i.e., resting HR plus 30 beats per minute (bpm) or a percentage of predicted maximum HR (% predicted max HR), is used as a way to determine exercise intensity without gas-exchange analysis data or exercise stress test data [2,6]. This simplified method, however, is considered difficult to use if patients are undergoing beta-blocker treatment, since beta blockers reportedly affect calculations by reducing HR responses during incremental exercise [7,8]. Although the negative chronotropic effect of beta blockers is considered to differ depending on beta-blocker type (e.g., αβ-blocker or β1-blocker), previous studies have not fully evaluated the effects of beta-blocker type on HR response during incremental exercise in subacute MI patients. The effect of beta-blocker type on this simplified method of exercise prescription, therefore, remains unclear. The present study aimed to compare the effects of αβ-blockers and β1-blockers on HR response during incremental exercise in subacute MI patients, because both αβ-blockers and β1-blockers are frequently prescribed for MI patients due to their effectiveness [9][10][11]. Clarifying differences in the effects of αβ-blockers and β1-blockers on HR response would help physicians to prescribe appropriate exercise intensities for subacute MI patients who are undergoing beta-blocker treatment.

Study Design and Patients
A total of 665 consecutive acute MI patients underwent cardiac rehabilitation at St. Marianna University School of Medicine, Yokohama City Seibu Hospital, between June 1998 and March 2018. Of these, 297 were able to undergo symptom-limited CPX one month after the onset of MI, and 131 who did not meet the following exclusion criteria were enrolled in this retrospective cross-sectional study: the presence of atrial fibrillation, complex arrhythmia, or pacemaker; recent myocardial infarction; history of cardiac surgery; female sex; maximum respiratory exchange ratio (RER) during CPX < 1.10; and use of drugs other than beta blockers with positive or negative chronotropic effects (e.g., verapamil, diltiazem hydrochloride). Patients were divided into the following three groups according to the type of beta blocker used: the αβ-blocker group (67 patients), the β1-blocker group (17 patients), and the no-β-blocker group (47 patients). All patients were clinically stable and were examined while being treated with stable doses of medication.
The study protocol conformed to the ethical guidelines of the Declaration of Helsinki and was approved by the Ethics Committee of St. Marianna University School of Medicine (Chairperson of the Ethics Committee, Masashi Taki; Protocol No. 2258; date of approval by the Ethics Committee, November 30, 2012). Informed consent was obtained from all patients before conducting CPX.

Clinical Characteristics
Age, body mass index (BMI), site of infarction, residual coronary artery stenosis, past medical history, complications, medications, beta-blocker dose, time between MI onset and exercise testing, and year of hospitalization were evaluated through a review of patient medical records. The ratio of administered beta-blocker dose to maximum beta-blocker dose (% maximum beta-blocker dose) was calculated by dividing the administered beta-blocker dose by the maximum beta-blocker dose for Japanese patients (e.g., carvedilol 20 mg/day, bisoprolol 5 mg/day, and atenolol 50 mg/day). Maximum serum creatine kinase-myocardial band (max CK-MB) was measured using blood samples obtained during hospitalization. Left ventricular ejection fraction (LVEF) was echocardiographically determined by a modified Simpson's method and used as the index of left ventricular systolic function at the time of hospital discharge.

Symptom-Limited CPX
All patients underwent symptom-limited CPX using a treadmill (MAT-2700, Fukuda Denshi Co., Tokyo, Japan). After three minutes of resting in a seated position, patients performed exercises with an increasing load (speed or grade) every 60 s after three minutes of warm-up activity (1 mph speed, 0% grade [12]. A 12-lead electrocardiogram was continuously monitored, and HR was measured from the R-R interval of the electrocardiogram (ML-9000, Fukuda Denshi Co., Tokyo, Japan) during the test. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured using a blood pressure cuff and an automatic blood pressure monitor (STBP-780, Colin Co., Aichi, Japan) at one-minute intervals. Oxygen uptake (VO 2 ), carbon dioxide production, minute ventilation, tidal volume, end-tidal O 2 , and end-tidal CO 2 were measured throughout the test using an aeromonitor (AE-310s, Minato Ikagaku Co., Tokyo, Japan). Expired gas was sampled using a breath-by-breath method. The endpoint of testing was determined as previously described [13]. Specifically, the appearance of a plateau in VO 2 despite an increasing exercise intensity indicated the exercise endpoint.
Ventilatory equivalents were calculated for O 2 , CO 2 , and RER using a personal computer (Model PC-9801, NEC Co., Tokyo, Japan). AT was determined by the V-slope method [14].
HR at rest (HR rest ), HR at AT (HR AT ), and HR at peak VO 2 (HR peak ) were determined from the recorded HR during the course of CPX. Increases from HR rest to HR peak (∆peak HR), from HR rest to HR AT (∆AT HR), and from HR AT to HR peak (∆AT-peak HR) were calculated for all patients.

Calculation of the Chronotropic Index
The metabolic chronotropic relationship (MCR), also referred to as the chronotropic index, was calculated as an indicator of HR response during incremental exercise at each stage (i.e., AT and peak VO 2 ) based on the relationship between HR and VO 2 during exercise [15][16][17]. The MCR, proposed by Wilkoff, allows for comparison between patients irrespective of age, physical fitness, functional capacity, exercise testing mode, or protocol. Estimated HR was calculated at AT and peak VO 2 using the following formula: where metabolic equivalents (METs) were defined as VO 2 (ml/kg/min) divided by 3.5. The chronotropic index at AT (MCR-AT) was calculated as the ratio of estimated HR AT to measured HR AT , and at peak VO 2 (MCR-peak) was calculated as the ratio of estimated HR peak to measured HR peak , as follows: MCR-peak = HR peak / estimated HR peak , MCR-AT and MCR-peak were used as indicators of HR response to incremental exercise up to the AT and peak exercise, respectively. A chronotropic index < 0.8 was previously determined to indicate chronotropic incompetence [15,16].

Statistical Analysis
Results are expressed as mean ± standard error (SE). The Kolmogorov-Smirnov test was performed to assess the normality of distribution. Differences in clinical characteristics, CPX results, and MCR findings among the three groups were analyzed using one-way analysis of variance (ANOVA), χ 2 test, and Tukey's honestly significant difference test. Tukey's honestly significant difference test was used if the results of one-way ANOVA were statistically significant. Pearson's correlation coefficient and Spearman's rank correlation coefficient were used in a regression analysis to determine correlations between MCR-AT, MCR-peak, and clinical characteristics (age and BMI were excluded from correlations as the formula of MCR included these factors), and correlations in which p < 0.2 were considered statistically significant. Correlations, clinically important factors on the nominal scale, αβ-blocker treatment, and β1-blocker treatment were entered into a multiple regression model for predictions of MCR-AT or MCR-peak using the forced-entry method. For patients undergoing beta-blocker treatment, Spearman's rank correlation coefficient was used to examine correlations between MCR-AT, MCR-peak, and % maximum beta-blocker dose. p < 0.05 was considered statistically significant. All statistical analyses were performed with JMP ® Pro 13 (SAS Institute Inc., Cary, NC, USA). A priori power analysis was performed for multiple regression analysis using G*Power 3 (Heinrich-Heine-Universität, Düsseldorf, Germany). The sample size was calculated to be 123 in the regression model for MCR-peak (effect size = 0.15; α error probability = 0.05; power (1−β error probability) = 0.8; number of predictors = 11). Accordingly, the sample size of the present study was considered sufficient for conducting multiple regression analysis. Effect sizes were also calculated using G*Power 3. Table 1 shows the clinical characteristics of patients included in this study. Significant differences between αβ-blocker, β1-blocker, and no-β-blocker groups were observed in renin-angiotensin system inhibitor treatment and hospitalization before 2006 (p = 0.024 and p < 0.001, respectively). The value of % maximum beta-blocker dose was significantly lower in the αβ-blocker group compared to the β1-blocker group (p < 0.001). No significant differences were observed in other clinical characteristics between the three groups.  Table 2 summarizes the data obtained during CPX. None of the patients showed ischemic ST changes or experienced chest pain or serious arrhythmias during CPX. The AT and peak VO 2 did not differ significantly between the three groups (p = 0.297 and p = 0.276, respectively). HR AT , HR peak , ∆AT HR, and ∆peak HR were significantly higher in the αβ-blocker group compared to the β1-blocker group (p = 0.004, p = 0.002, p = 0.004, and p = 0.003, respectively). HR rest , HR AT , and HR peak were significantly lower in the αβ-blocker group compared to the no-β-blocker group (p < 0.001, p = 0.014, and p = 0.002, respectively). HR rest , HR AT , HR peak , ∆peak HR, and ∆AT-peak HR were significantly lower in the β1-blocker group compared to the no-β-blocker group (p < 0.001, p < 0.001, p < 0.001, p = 0.001, and p = 0.011, respectively). MCR-AT and MCR-peak were both significantly higher in the αβ-blocker group compared to the β1-blocker group (p < 0.001, effect size (d) = 1.143; p < 0.001, d = 1.215, respectively). However, compared to the no-β-blocker group, MCR-peak was significantly lower in the αβ-blocker group (p = 0.038, d = 0.448), although no significant difference was observed in MCR-AT between the two groups (p = 0.406, d = 0.279). Both MCR-AT and MCR-peak were significantly lower in the β1-blocker group compared to the no-β-blocker group (p < 0.001, d = 1.429; p < 0.001, d = 1.346, respectively). Comparison results of variables during cardiopulmonary exercise testing with gas analysis in the αβ-blocker group, the β1-blocker group, and the no-β-blocker group were analyzed using one-way analysis of variance and Tukey's honestly significant difference test. Values are expressed as mean ± standard deviation (SD). AT, anaerobic threshold; DBP, diastolic blood pressure; DBP AT , DBP at AT; DBP peak , DBP at peak oxygen uptake; DBP rest , DBP at rest; RER, Respiratory exchange ratio; RER AT , RER at AT; RER peak , RER at peak oxygen uptake; HR, heart rate; HR AT , HR at AT; HR peak , HR at peak oxygen uptake; HR rest , HR at rest; MCR, metabolic chronotropic relationship; MCR-AT, estimated heart rate at AT / measured heart rate at AT; MCR-peak, estimated heart rate at peak / measured heart rate at peak; Peak VO 2 , peak oxygen uptake; SBP, systolic blood pressure; SBP AT , SBP at AT; SBP peak , SBP at peak VO 2 ; SBP rest , SBP at rest; ∆AT HR, (HR AT ) − (HR rest ); ∆peak HR, (HR peak ) − (HR rest ); ∆AT-peak HR, (HR peak ) − (HR AT ); † , p < 0.05 vs. no-β-blocker group; † † , p < 0.01 vs. no-β-blocker group; ‡ , p < 0.05 vs. β1-blocker group; ‡ ‡ , p < 0.01 vs. β1-blocker group; § , p < 0.05 vs. αβ-blocker group; § § , p < 0.01 vs. αβ-blocker group.

Results
The results of the regression analysis revealed no correlation between MCR-AT and Log max CK-MB, LVEF, and time between MI onset and exercise testing. MCR-peak was positively correlated with Log max CK-MB (p = 0.188), and negatively correlated with LVEF (p = 0.053). Multiple regression analysis revealed that β1-blocker but not αβ-blocker treatment significantly predicted lower MCR-AT and MCR-peak (β = −0.432, p < 0.001; β = −0.473, p < 0.001, respectively) (Tables 3 and 4). Multiple regression analysis for the prediction of MCR-AT was performed using the forced-entry method. MCR, metabolic chronotropic relationship; AT, anaerobic threshold; MCR-AT, measured heart rate at AT / estimated heart rate at AT; B ± SE, partial regression coefficient ± standard error; β, standardized partial regression coefficient; CI, confidence interval; MI, myocardial infarction. * Compared with lateral infarct. ** Compared with not undergoing beta-blocker treatment.  Multiple regression analysis for the prediction of MCR-peak was performed using the forced-entry method. MCR, metabolic chronotropic relationship; MCR-peak, measured heart rate at peak/estimated heart rate at peak; B ± SE, partial regression coefficient ± standard error; β, standardized partial regression coefficient; CI, confidence interval; LVEF, left ventricular ejection fraction; Max CK-MB, maximum value of serum creatine kinase-myocardial band; MI, myocardial infarction. * Compared with lateral infarct. ** Compared with not receiving beta-blocker treatment. Figure 1a and 1b show the correlation results between MCR-AT, MCR-peak, and % maximum beta-blocker dose. In patients undergoing beta-blocker treatment, no significant correlations were observed between MCR-AT, MCR-peak, and % maximum beta-blocker dose (p = 0.419, r = −0.089; and p = 0.760, r = −0.034, respectively).  Figure 1a shows correlation results between MCR-AT and % maximum beta-blocker dose. In patients undergoing beta-blocker treatment, no significant correlations were observed between MCR-AT and % maximum beta-blocker dose. Figure 1b shows correlation results between MCR-peak and % maximum beta-blocker dose. In patients undergoing beta-blocker treatment, no significant correlations were observed between MCR-peak and % maximum beta-blocker dose. MCR, metabolic chronotropic relationship; MCR-AT, ratio of estimated heart rate at the AT to measured heart rate at the AT; MCR-peak, ratio of estimated heart rate at peak oxygen uptake to measured heart rate at peak oxygen uptake; % maximum beta-blocker dose, ratio of administered beta-blocker dose to maximum beta-blocker dose.

Discussion
The present study compared the effects of β1-blocker versus αβ-blocker treatment on HR response during incremental CPX in subacute MI patients, using MCR as an indicator of HR response. Both MCR-AT and MCR-peak were significantly lower in the β1-blocker group than in the αβ-blocker group. Multiple regression analysis revealed that β1-blocker but not αβ-blocker treatment significantly predicted lower MCR-AT and MCR-peak. These results suggest that the negative chronotropic effect of beta blockers during incremental CPX may be stronger with β1-blockers than with αβ-blockers in subacute MI patients.

Selecting Beta-Blocker Type for Subacute MI Patients
Existing guidelines recommend administration of beta blockers to MI patients in order to reduce post-MI mortality [2,18]. However, there are no criteria for selecting a specific type of beta blocker for MI patients. A previous study reported no differences in the risk of all-cause death or MI between MI patients treated with αβ-blockers versus β1-blockers [10]. In the present study, the type of beta blocker used (i.e., αβ-blocker or β1-blocker) was determined based on clinical decisions made by physicians. Despite guideline recommendations, some patients were not treated with beta blockers. This is likely because, in Japan, the administration of beta blockers was first recommended by the Guidelines for Secondary Prevention of Myocardial Infarction in 2006 [19]. In fact, most patients who did not undergo beta-blocker treatment were hospitalized prior to 2006.

Differences in HR Response During CPX Among Subacute MI Patients Undergoing αβ-Blocker Versus β1-Blocker Treatment
HR responses during incremental exercise are greatly affected by cardiac autonomic nerve activity. When exercise intensity is low to moderate, HR response is considered to be mainly affected by decreased parasympathetic nervous system activity; when exercise intensity is moderate to high, HR response is considered to be mainly affected by increased sympathetic nervous system activity [20]. Increases in HR from rest to maximal exercise result from a transition in workload from a 4:1 parasympathetic/sympathetic balance to a 1:4 parasympathetic / sympathetic balance [20]. Among MI patients, the autonomic nerve balance is considered to trend more toward increasing sympathetic nervous system activity due to unbalanced cardiac autonomic modulation (i.e., increased sympathetic outflow and reduced parasympathetic activity) [21]. Beta blockers decrease sympathetic nervous system activity by inhibiting the interactions between noradrenaline and beta receptors. β1-blockers selectively inhibit the interaction between noradrenaline and β1 receptors in the heart, while αβ-blockers non-selectively inhibit the interactions between noradrenaline and β1 receptors. The effect of αβ-blockers to decrease cardiac sympathetic nervous system activity is thus considered to be lower than that of β1-blockers. In the present study, MCR-AT and MCR-peak were significantly lower in the αβ-blocker group than in the β1-blocker group. A chronotropic index < 0.8 has been previously determined to indicate chronotropic incompetence [15,16]. In the present study, the values of MCR-AT and MCR-peak were 0.85 ± 0.07 and 0.90 ± 0.08, respectively, in the αβ-blocker group, and 0.77 ± 0.07 and 0.78 ± 0.11, respectively, in the β1-blocker group. These results show that the β1-blocker group had a low chronotropic index, while the αβ-blocker group did not, suggesting that the effect of β1-blockers decreasing HR response during incremental CPX may be stronger than with αβ-blockers.

Effects of αβ-Blocker Versus β1-Blocker Treatment on HR Response During CPX in Subacute MI Patients After Adjusting for Multiple Factors
Previous studies suggested that several factors other than beta-blocker treatment could affect HR response during exercise, including age, body weight, functional capacity, cardiac function, high max CK-MB, diabetes mellitus, and heart failure after MI [22][23][24][25][26]. The present study used MCR as an indicator of HR response, because MCR allows for the assessment of HR response without the effects of age, physical fitness, or functional capacity [15,16]. In order to adjust for other several factors, multiple regression analysis was performed to predict MCR. The analysis revealed that β1-blocker but not αβ-blocker treatment significantly predicted lower MCR-AT and MCR-peak.

Beta-Blocker Doses
In the present study, the mean dose of carvedilol was 5.48 ± 2.99 mg/day in the αβ-blocker group, and mean doses of bisoprolol and atenolol were 2.50 ± 1.40 and 33.33 ± 12.91 mg/day, respectively, in the β1-blocker group. A previous study reported that Japanese MI patients were treated with a mean carvedilol dose of 3.48 ± 1.61 mg/day before hospital discharge and 7.71 ± 4.58 mg/day at 12 months after MI onset, or a mean bisoprolol dose of 1.26 ± 0.57 mg/day before hospital discharge and 2.47 ± 1.20 mg/day at 12 months after MI onset [11]. Another study in Japan reported that atenolol was started at a low dose and titrated up to the target dose of 25-50 mg/day at one month after MI onset [27]. In the present study population (i.e., MI patients at one month after MI onset), although the mean doses of carvedilol and atenolol were considered clinically appropriate, the mean dose of bisoprolol was slightly higher than that noted in previous study populations. To examine the effect of beta-blocker doses on HR response, we also examined the correlations between MCR-AT, MCR-peak, and % maximum beta-blocker dose. No significant correlations were observed between MCR-AT, MCR-peak, and % maximum beta-blocker dose. The negative chronotropic effect of beta blockers depends on the concentration of these medications in the blood. In addition to dose, several clinical factors can affect the blood concentration of beta blockers, such as body weight, renal function, liver function, and genetic polymorphism [28][29][30]. In clinical practice, physicians determine appropriate doses for each MI patient based on these factors. In other words, optimum beta-blocker doses for blood concentration are different for each patient. Thus, no significant correlations were observed between MCR-AT, MCR-peak, and % maximum beta-blocker dose in the present study. However, from the results of the present study, we could not conclude that there is no relationship between the dose of a β-blocker and HR response. Because it is difficult to assess whether each patient was on a low, moderate, or high dose relative to the individual response to treatment, the effect of beta-blockers on each patient was not assessed in the present study.

Clinical Application of Study Findings
In clinical practice, a simplified method of determining exercise intensity without gas-exchange analysis data or exercise stress test data is to use a substitute for HR at AT, i.e., resting HR plus 30 bpm or % predicted max HR. [2,6]. This simplified method is considered difficult to use in patients undergoing beta-blocker treatment, based on reports from previous studies that beta blockers affect calculations by reducing HR responses during incremental exercise [7,8]. Although the negative chronotropic effect of beta blockers is considered to differ depending on beta-blocker type (e.g., αβ-blocker or β1-blocker), previous studies have not fully evaluated the effects of beta-blocker type on HR response during incremental exercise in subacute MI patients. The effect of beta-blocker type on this simplified method of exercise prescription, therefore, remains unclear. Our previous study [31] examined the effect of αβ-blocker treatment on HR response during incremental exercise in subacute MI patients, because αβ-blockers are frequently prescribed for MI patients due to their effectiveness [9]. αβ-blocker treatment was found to have no effect on HR response to exercise up to the AT in subacute MI patients [31], suggesting that exercise intensity can be prescribed using the simplified method (i.e., resting HR plus 30 bpm) in subacute MI patients regardless of αβ-blocker use. Although β1-blockers are also frequently used in MI patients [10,11], the effect of β1-blockers on HR response during incremental exercise in subacute MI patients remains unclear. The present study compared the effects of β1-blocker versus αβ-blocker treatment on HR response during incremental exercise in subacute MI patients, and found that both MCR-AT and MCR-peak values were significantly lower in the β1-blocker group than in the αβ-blocker group. These results suggest that, when prescribing exercise intensity for subacute MI patients in clinical practice based on calculations of resting HR plus 30 bpm or % predicted max HR, we may need to use a lower target HR/predicted max HR value for those who are undergoing β1-blocker treatment, compared to those who are undergoing αβ-blocker treatment. In order to determine an appropriate target HR value for subacute MI patients undergoing β1-blocker treatment, further evaluation of training effectiveness and safety is needed. A previous study presented a prediction formula for maximum HR among patients with coronary heart disease receiving beta-blocker treatment (i.e., predicted maximum HR = 164 − 0.7 × age) [6]. However, this prediction formula was not specific for each type of beta blocker. Prediction formulas for maximum HR in MI patients treated with each type of beta blocker need to be established in future studies.

Limitations
The present study has several limitations. First, the influence of potential factors (e.g., body weight, renal function, liver function, genetic polymorphism) on the blood concentration of beta blockers was not examined in the present study. Second, all patients underwent symptom-limited CPX using a treadmill. The effects of beta-blocker type on HR response during other symptom-limited CPX (e.g., on a bicycle ergometer) have not yet been clarified. Third, this study was analyzed retrospectively. Hence, the effect of several biases such as an imbalance in the number of patients and a difference in the treatment of MI between the groups cannot be completely excluded in this study. A future randomized controlled trial will be needed to verify the results of the present study. Finally, this study is affected by selection bias as patients who were unable to undergo symptom-limited CPX on a treadmill due to frailty were excluded from the study. As a result, the population in this study comprised lean and young patients. A future study will be needed to verify the results of the present study in a broader age range of patients.

Conclusions
Our data suggest that the effects of β1-blockers to decrease HR response during incremental CPX may be stronger than that of αβ-blockers in subacute MI patients. Therefore, exercise intensity based on a simplified substitute for HR at AT (i.e., resting HR plus 30 bpm or % predicted max HR) without gas-exchange analysis data or exercise stress test data should be prescribed according to the type of beta blocker used in subacute MI patients.