This is an investigator-mediated, prospective, multicenter, randomized clinical trial comparing an invasive vs a noninvasive strategy in patients with NSTEMI, aged ≥ 70 years with frailty. Patients will be eligible for inclusion if they fulfill all 3 of the following inclusion criteria: a) NSTEMI, defined by acute chest pain, absence of persistent ST-segment elevation in the presence of interpretable repolarization and troponin elevation (according to the local laboratory troponin assay); b) age ≥ 70 years; c) frailty criteria defined by ≥ 4 points on the Clinical Frailty Scale (Figure of the supplementary material).8 The exclusion criteria will be the following: a) prior known nonrevascularizable coronary artery disease; b) significant concomitant nonischemic heart disease (ie, severe heart valve disease, cardiomyopathy); c) inability to understand/sign informed consent (patients or relatives); d) life expectancy < 12 months.

In addition to the defined inclusion and exclusion criteria, the attending cardiologist must believe that the participation of the patient in the study is reasonable. Reasons for considering participation inappropriate might be either the consideration by the attending cardiologist of the need for an invasive management strategy or any clinical factor making invasive management not an option.

This study was registered at ClinicalTrials.gov (Identifier: NCT03208153).

Patients will be randomized within 48hours of admission to undergo coronary angiography and coronary revascularization if deemed appropriate during the index hospitalization, or to be treated with medical therapy. Patients will be assigned to both treatment groups using a computer-generated randomization scheme to allocate participants in a 1:1 ratio. The randomization will be via a web site where the process will be concealed from the researchers until the interventions are assigned. Medical treatment at admission and discharge will be optimized according to the guidelines in both study arms.

The Figure shows the algorithm of the study. The invasive group will undergo coronary angiography within 72hours of admission. When percutaneous coronary intervention is performed, the type of stent implanted will left to the judgement of the treating cardiologist, although stents with proven safety using short-term antiplatelet treatment are recommended to reduce the potential bleeding risk of this population. In contrast, patients allocated to the noninvasive group will receive medical therapy only, although cardiac catheterization will be allowed if there is poor in-hospital outcome due to documented recurrent ischemia (chest pain plus either electrocardiogram changes or troponin re-elevation) or hemodynamic instability during the index admission. According to a previous a trial, the estimated crossover rate is 24%.9 Predischarge left ventricular ejection fraction will be measured in all patients by echocardiography. The standard period of dual antiplatelet therapy will be 1 year in both study arms, although in patients at high bleeding risk (in the judgement of the treating physician) or in need of oral anticoagulation therapy, 1 antiplatelet drug may be withdrawn after the first month. Detailed clinical and treatment data will be collected at admission and during hospitalization (Table 1 and Table 2).

Figure.

Algorithm of the study. ACE, angiotensin-converting enzyme; NSTEMI, non–ST-segment elevation acute myocardial infarction.

(0.16MB).

A systematic and comprehensive geriatric evaluation will be performed prior to discharge through interview with the patient and/or family/caregivers and referring to the patient's status prior to admission, as follows: a) Functional capacity for basic activities of daily living will be assessed by the Barthel Index.10 This is an ordinal scale with a total score of 0 to 100, where the intermediate ranges help evaluate the different degrees of dependency: total (0-20), severe (21-40), moderate (41-60), light (61-90), and independent (> 90). b) Instrumental activities will be evaluated with the Lawton-Brody Index.11c) Cognitive status will be evaluated with the Pfeiffer test.12d) Previous frailty will be assessed, in addition to the Rockwood scale, by the FRAIL scale, which includes evaluation of fatigue, resistance, ambulation, concomitant diseases, and weight loss.13e) After the acute clinical phase (once the patient is stabilized), the Short Physical Performance Battery will be carried out.14 This test includes: i) balance in 3 positions (feet together, semitandem and tandem), ii) walking speed (4 meters along), and iii) get up and sit in a chair 5 times. The SPPB total score is the summation of the 3 subtests, ranging from 0 (worst) to 12. A score below 10 indicates frailty and an increased risk of disability and falls. f) To assess comorbidity, the Charlson index will be applied, with a maximum score of 37 points.15 The number of medications with chronic prescription taken by the patient before admission will be also recorded. g) Nutritional risk assessment will be carried out with the Mini Nutritional Assessment-Short Form,16 whose value ranges from 0 to 14 points; scores below 11 identify patients at risk of malnutrition. h) Quality of life will be analyzed by the EQ 5D 5L test.17

The primary endpoint will be the number of days alive out of hospital during the first year. The coprimary endpoint will be the time to the first occurrence of 1 of the events included in the combined endpoint (cardiovascular death, myocardial infarction or postdischarge revascularization). Secondary endpoints will include time to occurrence of all-cause death, myocardial infarction, rehospitalization for cardiac and extracardiac causes, bleeding episodes, and stroke. Only bleeding type ≥ 2 according to the Bleeding Academic Research Consortium definition will be considered.18 The occurrence of procedure-related myocardial infarction will be evaluated by measuring troponins 12hours after revascularization. Procedure-related infarction will be defined by troponin elevation > 5 x 99th percentile after percutaneous coronary intervention and > 10 x 99th percentile after coronary surgery, plus data suggestive of acute ischemia according to clinical symptoms, electrocardiogram, angiography or imaging. Functional capacity (Barthel index), instrumental activities (Lawson-Brody index), cognitive capacity (Pfeiffer) and quality of life (EQ 5D5L) will be reassessed at 6 months, also as secondary endpoints. Follow-up will include a clinical visit or telephone contact at 3 months, 1 year, and yearly until 3 years.

This is an investigator-driven initiative under the auspices of the Spanish Society of Cardiology and the working groups of Interventional Cardiology and Geriatric Cardiology. A steering committee will be responsible for overseeing the scientific and operational aspects of the study. Patients and investigators will not be masked to treatment allocation, but clinical events will be assessed by a blinded clinical event committee to prevent bias. Likewise, a data safety monitoring board, unaware of the patients’ treatment allocation, will be responsible for making recommendations to the steering committee regarding endpoints and any potential significant patient safety-related observations.

There are no data on the optimal management of frail patients with NSTEMI. In a previous study in elderly patients with NSTEMI and comorbidities, patients who underwent a conservative strategy remained alive out of hospital for a mean of 273 days (standard deviation 123 days) during the first year after discharge.9 Assuming that frail patients might have a fairly similar profile and considering an increase of 20% in the proportion of days alive out of hospital (55 days) with an invasive strategy, we estimate a sample size of 178 patients (89 per arm) with an estimated power of 80%, 2-sided alpha level of .05, and 10% of losses of follow-up. The achievement of the sample size will require a multicenter approach. We estimate at least 10 participating hospitals.

All statistical comparisons will be made under the intention-to-treat principle. Results will be presented as frequencies or mean (standard deviation), as appropriate. Between-group comparisons will be performed using the t test or Fisher exact test.

Patient follow-up will be censored at the time of death or at the end of the study. The primary endpoint, as a continuous variable, will be compared between the 2 groups using the ANOVA test. Regarding the coprimary endpoint and secondary endpoints, the effect of the invasive strategy on the clinical events will be depicted by a Kaplan-Meier method and assessed using a Cox regression model. The hazard ratio and 95% confidence interval will be calculated. A prespecified subgroup analysis according to comorbidities (Charlson index) will be conducted.

A 2-sided P value of < .05 will be considered statistically significant.

The estimation of the sample size is speculative in part, since there are no previous data on the prognostic impact of the management strategy in frail patients with NSTEMI. Therefore, we had to use the information from our previous study in older patients with comorbidities and NSTEMI to estimate the sample size, even though frailty was not measured in the study.9