Chronic remote ischemic conditioning treatment in patients with chronic stable angina (EARLY-MYO-CSA): a randomized, controlled proof-of-concept trial

The trial followed a prospective, randomized, sham-control design. It complied with the Declaration of Helsinki and was approved by the New Business and Technology Ethics Committee of Fuwai Central China Cardiovascular Hospital, Zhengzhou University (Zhengzhou, China). We followed the Consolidated Standards of Reporting Trials (CONSORT) checklist [18] to report this study (Additional file 1: CONSORT checklist). All participants signed an informed consent form. This trial was registered with the Chinese Clinical Trial Registry (ChiCTR2000038649).

Eligible participants were those (1) > 18 years old, (2) having CSA confirmed by angiography without complete revascularization (> 50% stenosis on angiography and the quantitative flow ratio < 0.75 for at least one main coronary artery or branches), and (3) with persistent angina pectoris after receiving ≥ 3-month guideline-recommended optimal medical therapy (the main drugs include β-blocker, calcium channel blocker (CCB), nitrates, ivabradine, trimetazidine, nicorandil, antiplatelet agents, angiotensin-converting enzyme inhibitor (ACEI) or angiotensin receptor blocker (ARB), and statin; all enrolled patients received the recommended dose in accordance with the guidelines, unless contraindicated). Exclusion criteria included the following: intolerance to RIC therapy; stenosis of the left main stem ≥ 50%; pregnancy or having a plan for it; a history of arterial or venous thrombosis in upper limbs; comorbidities of severe valvular disease, congenital heart disease, severe arrhythmia, aortic dissection, aortic aneurysm, cardiomyopathy, and severe uncontrolled hypertension (defined as systolic blood pressure > 180 mmHg and/or diastolic blood pressure > 110 mmHg after taking medication); life expectancy < 1 year; malignant tumor; poor compliance to RIC; patients with acute myocardial infarction or coronary revascularization in the past 1 year; use of glibenclamide [19]; and MFR ≥ 2.5 as determined using single photon emission computed tomography (SPECT).

Before randomization, each patient underwent an RIC tolerance evaluation. RIC intolerances include (1) unbearable limb pain and numbness during cuff inflation and (2) unable to sustain 35 consecutive minutes of restricted activity by RIC device. Patients with RIC intolerance will be excluded.

From October 2020 to May 2022, 275 patients with CSA who were referred to Fuwai Central China Hospital were recruited. Of these, 220 were randomized after screening according to the inclusion and exclusion criteria. Among them, two withdrew their informed consent after randomization, and 10 participants refused to return to hospital for endpoints assessment (Fig. 1). A total of 208 patients (105 in the CRIC group and 103 in the control group) completed the treatment and endpoints assessments. Randomization (1:1 ratio) was performed by an independent statistician using a computer-generated list and with stratifications by factors of sex and age ≥ 50 years.

All patients received standard optimal medical therapy according to current guidelines. For those assigned to the CRIC group, a semi-automatic RIC device with a controller and a cuff (GTHR Medical Technology company, Shenzhen, China) was used. The device has been approved by the Food and Drug Administration in Guangdong Province, China. Once the cuff was attached to the participant’s upper arm and the controller was turned on, the device automatically performed the RIC procedure. Each RIC procedure took 35 min with four cycles of cuff inflation at 200 mmHg pressure (5 min for each cycle) and 5 min intervals of relaxation between two cycles. This procedure was performed twice per day. Patients in the control group received sham RIC using a device with the identical appearance to the RIC device at a cuff pressure of 60 mmHg. One investigator held a device-use training during the trial screening phase to ensure that all participants could operate the device correctly. During follow-up, participants were asked to provide device usage records. Participants were evaluated by clinic visit or telephone at 1 and 3 months after randomization and were asked to return for face-to-face assessments at 6 months.

The primary endpoint of this trial was the change of MFR at 6 months (± 14 days) post-randomization. Secondary endpoints included 6-month changes of rest and stress MBF, Canadian Cardiovascular Society (CCS) classification-defined angina improvement, all dimensions of the Seattle Angina Questionnaire (SAQ) score, and the distance from 6-min walk test (6-MWT).

All patients underwent MBF quantification using SPECT at the Central China Fuwai Hospital at baseline and at 6 months (± 14 days) post-randomization. SPECT was performed with a cadmium zinc telluride camera system (Discovery NM 530c; GE, Boston, MA) using a 2-day protocol. On day 1, the patient underwent low-dose computerized tomography (CT) scanning for physical image processing correction and was administered an adenosine injection (injection rate of 0.14 mg/kg/min) after drinking 500 ml of water. The load list mode dynamic SPECT data collection started at 2 min and 50 s. At the third minute, 20 mCi 99mTc-MIBI was injected as an intravenous bolus, and dynamic SPECT data acquisition continued for 10 min. On day 2, the patient started resting list-mode dynamic SPECT data acquisition after drinking 500 ml of water; 10 s later, 20 mCi 99mTc-MIBI myocardial imaging agent was injected as an intravenous bolus, and dynamic SPECT data acquisition continued for 10 min. The corresponding left ventricle (LV) MBF quantification data were analyzed using the MyoFlowQ software [20]. The MFR was calculated as the ratio of stress to rest MBF values. According to the SPECT flow status, a polar map was converted from the rest MBF, stress MBF, and MFR using the flow diagram [21, 22]. The left ventricle myocardial blood flow restriction of each patient was classified into seven degrees: definitely normal, normal limit, mildly abnormal, moderately abnormal, ischemia, steal, and infarct. Then, the flow restriction extent was calculated as a percentage using the MyoFlowQ software. During this study, we identified the sum of the percentages of ischemia and steal flow status on the polar map as the reversible myocardial ischemia extent (RMIE) in order to more intuitively observe the improvement of myocardial ischemia (Additional file 2: Figure S1). The SPECT results were assessed by three nuclear medicine readers who were blinded to the randomization and participants’ clinical information using MyoFlowQ software and then the mean of MBF quantification was calculated. To assess the reproducibility of SPECT, intra-observer and inter-observer variability was validated in SPECT results from all 208 participants. The first nuclear medicine reader conducted a repeat MBF quantification analysis after an interval of 3 months for intra-observer variability. Inter-observer variability was performed using the results obtained by the second nuclear medicine reader who had participated in the initial analysis.

Any change in angina symptoms was assessed using the CCS classification, SAQ score, and 6-MWT by investigators blinded to the randomization and participants’ clinical information.

The CCS classification was used to assess the severity of angina pectoris based on the combination of physician’ s assessment and patient’ s self-reported symptoms. The SAQ is a 19-item questionnaire that measures five domains related to angina pectoris: angina frequency, angina stability, physical limitation, treatment satisfaction, and disease perception [23]. Each domain has a score from 0 to 100, with a higher score representing an improvement of angina pectoris. More information on CCS classification and SAQ is available in the Additional file 3. The 6-MWT was performed as follows: after resting for 5–10 min, the patient walked up and down a 30 m straight corridor for 6 min; the assessor encouraged the patient to walk as fast as possible while following directly behind; the assessor recorded the maximal distance covered by the patient in 6 min or < 6 min if the patient stopped earlier.

We evaluated the safety of CRIC treatment by assessing major adverse cardiovascular events and local adverse reactions to RIC in the limbs. Major adverse cardiovascular events included all-cause death, nonfatal myocardial infarction, heart failure, and stroke after randomization. Detailed definitions of events are provided in our previous studies [24, 25]. Local limb adverse reactions reported by participants or investigators, included skin ecchymosis, limb pain, limb weakness, and limb arteriovenous thrombosis. All adverse events were judged by an independent committee.

The sample size of the present trial was calculated based on the findings of our pilot study: there was a 0.20 difference in the 6-month MRF change from baseline across two randomized groups, with 0.15 ± 0.46 in the CRIC group and − 0.05 ± 0.41 in sham the CRIC group. Accordingly, group sample sizes of 102 per group were required to achieve 90% power to detect a difference of 0.20 in a two-sample t-test allowing unequal variance study design. The standard deviation of CRIC and sham CRIC was 0.46 and 0.41, respectively, while the alpha level was two-sided 0.05. With an expected 5% dropout rate, at least 214 patients should be enrolled. Power calculation was performed using the PASS software (version 15; NCSS, LLC, Kaysville, Utah, USA). More information on sample size calculation is available in Additional file 3.

The primary analysis was performed using the full analysis set on a modified intention-to-treat basis. Statistical analysis was performed using SPSS software (version 25.0; SPSS Inc., Armonk, NY). Categorical variables were presented as the number of cases and percentages. Comparisons between groups were performed using the chi-square test. Continuous variable data that conformed to normal distributions were presented as the mean ± standard deviation. Comparisons between groups were performed using the unpaired Student’s t-test. For repeated-measures data, an analysis of within-group measurements was performed using the paired t-test. Continuous variables that did not conform to normal distributions are expressed as median (interquartile ranges). Between-group comparison were processed using the Mann–Whitney U test. Repeated-measures data were processed using the Wilcoxon two-sample test. The Bland–Altman analysis, intraclass correlation coefficients (ICC), and coefficient of variation (COV) were used to evaluate the inter and intra-observer variability of MBF, MFR, and RMIE. P < 0.05 was considered statistically significant (two-sided). Both the primary and secondary endpoint indicators of this study relate to outcomes at follow-up, while data on any endpoint indicators were not available for patients lost to follow-up after randomization, and the data for the remaining patients analyzed met the modified intention-to-treat data set criteria.

In total, 208 patients completed the study consisting of 137 (65.9%) men and with a mean age of 61.99 ± 10.53 years. Among the participants, 44 (21.2%) had a history of myocardial infarction, 133 (63.9%) had undergone PCI, and 9 (4.3%) had undergone CABG. Participants in both groups had high syntax scores (22.33 ± 7.52), suggesting the complexity of their coronary lesions. The baseline characteristics were comparable between the two randomized groups (Table 1).

SPECT outcomes showed that both groups had low global rest and stress MBF values at baseline (rest: 0.84 ± 0.21 vs. 0.82 ± 0.20 mL/min/g, P = 0.423; stress: 1.11 ± 0.45 vs. 1.09 ± 0.44 mL/min/g, P = 0.828). After 6 months, the control group had no significant changes in rest and stress MBF measurements. Conversely, rest MBF values significantly decreased in the CRIC group at 6 months compared with baseline (0.80 ± 0.16 mL/min/g vs. 0.84 ± 0.21 mL/min/g; P = 0.020). Stress MBF in the CRIC group significantly increased in 6 months compared with baseline (1.26 ± 0.50 mL/min/g vs. 1.11 ± 0.45 mL/min/g; P < 0.001). The MFR increased from baseline at 6-month follow-up in the CRIC group (1.33 ± 0.48 vs. 1.61 ± 0.53; P < 0.001) but not in the control group (1.35 ± 0.45 vs. 1.31 ± 0.44; P = 0.608). The mean 6-month change of MFR was significantly greater in the CRIC group than in the control group (0.27 ± 0.38 vs. − 0.04 ± 0.25; P < 0.001). The RMIE significantly decreased from baseline at 6-months follow-up in the CRIC group (39.42 ± 28.15% vs. 29.33 ± 24.66%; P < 0.001), but not in the control group (39.61 ± 28.14 vs. 40.16 ± 29.47; P = 0.108; Table 2, Fig. 2).

Intra-observer variability yielded good concordance for MBF (bias =  − 0.02 to 0.01, ICC = 0.86 to 0.98, and COV = 8.04 to 13.67%), MFR (bias =  − 0.02 to 0.00, ICC = 0.90 to 0.92, and COV = 14.26 to 16.17%), and RMIE (bias =  − 0.36 to − 0.32, ICC = 0.98 to 0.9, and COV = 10.57 to 12.11%). The ICC showed moderate inter-observer agreement in the rest MBF of baseline (ICC = 0.74) and good in other MBF assessment (ICC = 0.88–0.95), MFR (ICC = 0.87 to 0.89), and RMIE (ICC = 0.99). The Bland–Altman analysis and COV showed good inter-observer agreement for MBF (bias =  − 0.02 to 0.02 and COV = 12.67 to 18.18%), MFR (bias =  − 0.03 to 0.01 and COV = 16.77 to 18.59%), and RMIE (bias =  − 0.50 to − 0.26 and COV = 10.86 to 11.73%; Additional file 4: Figure S2, Additional file 5: Table S1).

A significant difference in terms of 6-month angina symptoms relief according to the CCS classification between groups was observed, with a higher proportion of patients having symptom improvement in the CRIC group than in the control group (60% vs. 14.6%, P < 0.001; Table 3, Fig. 3).

The five dimensions of SAQ were not significantly different between the two groups at baseline; however, all the dimension scores were significantly higher in the CRIC group than in the control group, with the exception of disease perception score (P = 0.126). Moreover, disease perception scores significantly increased from baseline in the CRIC group at 6-month follow-up (P = 0.001; Table 4, Fig. 4).

At baseline, the 6-MWT results were comparable between the two groups (400 [320–450] vs. 405 [330–470] m; P = 0.193]. After 6 months, the distance covered over 6 min was greater among patients in the CRIC group than those in the control group (440 [400–523] vs. 420.00 [330–475] m; P = 0.016; Table 5).

During the study period, three patients in the CRIC group and four patients in the control group underwent PCI (2.9% vs. 2.3%, P = 0.720). After 6 months, a greater proportion of patients in the CRIC group exhibited a decrease in the frequency of short-acting nitrate administration (70.5% vs. 15.5%, P < 0.001; Table 6).

No cardiovascular death or heart failure events occurred in either group during the 6-month follow-up period. A few participants experienced nonfatal myocardial infarction (3 [2.9%] in the CRIC group vs. 2 [1.9%] in the control group; P = 0.509). All participants in the CRIC group tolerated RIC well. A few participants reported transient hand numbness or sweating while using the RIC device; however, complete relief was achieved immediately after the procedure. The presence of skin ecchymosis or upper extremity pain was slightly higher in the CRIC group than in the control group; however, the difference was no statistical significance (P = 0.117 and P = 0.351, respectively). Furthermore, neither group experienced arteriovenous thrombosis events during the trial (Table 7).