A systematic review and meta-analysis on the effectiveness of an invasive strategy compared to a conservative approach in patients > 65 years old with non-ST elevation acute coronary syndrome

Background Patients 65 years old and older largely represent (>50%) hospital-admitted patients with acute coronary syndrome (ACS). Data are conflicting comparing efficacy of early routine invasive (within 48–72 hours of initial evaluation) versus conservative management of ACS in this population. Objective We aimed to determine the effectiveness of routine early invasive strategy compared to conservative treatment in reducing major adverse cardiovascular events in patients 65 years old and older with non-ST elevation (NSTE) ACS. Data sources We conducted a systematic review of randomized controlled trials (RCTs) through PubMed, Cochrane, and Google Scholar database. Study selection The studies included were RCTs that evaluated the effectiveness of invasive strategy compared to conservative treatment among patients ≥ 65 years old diagnosed with NSTEACS. Studies were included if they assessed any of the following outcomes of death, cardiovascular mortality, myocardial infarction (MI), stroke, recurrent angina, and need for revascularization. Six articles were subsequently included in the meta-analysis. Data extraction Three independent reviewers extracted the data of interest from the articles using a standardized data collection form that included study quality indicators. Disparity in assessment was adjudicated by another reviewer. Data synthesis All pooled analyses were initially done using Fixed Effects model. For pooled analyses with significant heterogeneity (I2≥ 50%), the Random Effects model was used. A total of 3,768 patients were included, 1,986 in the invasive strategy group, and 1,782 in the conservative treatment group. Results Meta-analysis showed less incidence of revascularization in the invasive (2%) over conservative treatment groups (8%), with overall risk ratio of 0.29 (95% CI 0.14 to 0.59). Across all pooled studies, no significant effect of invasive strategy on all-cause mortality, cardiovascular mortality, stroke, and MI was observed. Only one study assessed the outcome of recurrent angina. Conclusion There was a significantly lower rate of revascularization in the invasive strategy group compared to the conservative treatment group. In the reduction of all-cause mortality, cardiovascular mortality, MI, and stroke there was no significant effect of invasive strategy versus conservative treatment. This finding does not support the bias against early routine invasive intervention in patients ≥ 65 years old with NSTEACS. Further studies focusing on these patients with larger population sizes are still needed.


Introduction
Based on the World Health Organization's Global Burden of Disease report, ischemic heart disease (IHD) is the overall leading cause of death worldwide [1]. Although the annual number of hospital discharges for acute coronary syndromes (ACS) in developed countries has declined slowly over the past two decades, the number has increased in developing countries [2]. In the Philippines, cardiovascular disease (CVD) remains the leading cause of mortality [3]. The Philippine Heart Association ACS registry reported that ACS is prevalent in the age range 51-70, with mean age group of 66 years old [3].
The most recent American College of Cardiology/American Heart Association (ACC/AHA 2014) and the European Society of Cardiology (ESC 2015) guidelines for non-ST segment elevation ACS (NSTEACS) reflect medical advancements in therapeutics and strategies of care leading to improved survival in ACS, but this was mainly observed in relatively younger individuals (<65 years of age) and in men. These guidelines emphasize intensive and early medical and interventional therapy, particularly for those at high risk [4][5][6].
The 2014 AHA/ACC NSTEACS Guidelines generally recommend that older patients with NSTEACS should be treated with goal-directed medical therapy, together with an early invasive strategy, and revascularization as appropriate [5]. The 2015 ESC Guidelines for the Management of ACS, on the other hand, recommend that decisions on elderly patients with NSTEACS should be based on ischemic and bleeding risks, estimated life expectancy, comorbidities, quality of life, patient values and preferences, and the estimated risks and benefits of revascularization [6]. Despite the guidelines, older patients are less likely to undergo procedures after an NSTEACS than younger patients due in part to patient and practitioner concerns about the increased risk of complications [7][8][9].
Due to conflicting results of studies, lack of specific recommendations from the abovementioned guidelines, and the paucity of data on early invasive strategy versus conservative treatment for NSTEACS in patients � 65 years old, this meta-analysis was conducted to focus on this special population to compare benefits and risks of early invasive therapy versus conservative management.

Research question
Among elderly patients aged � 65 years old with NSTEACS, how effective is invasive strategy compared to conservative treatment in preventing major adverse cardiovascular events (MACE)?

General
To determine the effectiveness of invasive strategy compared to conservative treatment in reducing MACE among elderly patients with NSTEACS.

Specific
Among patients � 65 years old with NSTEACS, to determine the effectiveness of invasive strategy compared to conservative treatment, in 6 months (short-term) to

Study registration
Prior to the conduct of the research, the study was registered and approved by the Committee on Research (CORES) of Manila Doctors Hospital.

Criteria for considering studies for this review
The studies included were RCTs that evaluated the effectiveness of invasive strategy compared to conservative treatment among patients � 65 years old diagnosed with NSTEACS. Studies were included if any of the outcomes assessed were: death, cardiovascular mortality, MI, stroke, recurrent angina, and need for revascularization.

Definition of terms
1. Invasive strategy or early invasive strategy-Routine early cardiac catheterization (within 48-72 hours of initial evaluation) followed by percutaneous coronary intervention (PCI) or coronary artery bypass graft (CABG) depending on the coronary anatomy within the patient's hospital admission for the index event on top of continuing medical therapy.
2. Conservative treatment-Initial optimal medical management, with cardiac catheterization reserved for patients with recurrent ischemia at rest or after a non-invasive stress test, followed by revascularization if the anatomy is suitable.

Search methods for identification of studies
Systematic computerized search (S1 Appendix) was performed using the Pubmed, Google Scholar, and Cochrane databases. MESH and free text of the following main key terms were used: "randomized controlled trials", "elderly", "non-ST elevation acute coronary syndrome", "invasive strategy", "conservative management", "invasive strategy versus conservative strategy", "major adverse cardiovascular events", "all-cause mortality", "cardiovascular mortality", "myocardial infarction", "stroke", "recurrent angina", "need for revascularization". The last search was done on September 1, 2019. Eligibility assessment was performed independently in a standard manner by three reviewers. The literature search identified 151 relevant articles. Of the 151 articles, 34 were excluded due to different intervention since they did not involve comparing invasive versus conservative management in ACS. Among the 117 articles left, 30 articles were excluded due to different outcomes being assessed (i.e. arterial access site-related outcomes, risk for CABG, risk for heart failure, abnormal tissue perfusion). Among the 87 articles left, 51 articles subsequently were excluded due to different methods (i.e. observational studies, post-hoc analysis, coststudy, adherence study, registry data, outcome research study). Furthermore, among the 36 articles left, 23 were excluded due to different population (i.e. involved patients who have STEMI, post-MI, stable IHD, sepsis). Thirteen articles were then fully reviewed for eligibility. Among the 13 articles reviewed, 2 articles were possibly eligible but these are ongoing RCTs (MOSCA-FRAIL and DEAR-OLD Trials) with no published data yet at time of writing; hence, excluded [10,11]. Among the remaining 11 articles, 2 articles were excluded due to different intervention while 2 articles were excluded due to different population. One article was possibly eligible but did not report the event rates per treatment group (details for the titles of the studies and reasons for exclusion are listed in S2 Appendix). To access needed data in this particular study, correspondence with the author via email was done, but with no reply from the author until the time of writing. Six articles were subsequently included in the meta-analysis (Fig 1).

Assessment of risk bias of included trials
Three independent reviewers extracted the data of interest using a standardized data collection form (S3 Appendix) and individually appraised each trial. The reviewers discussed the quality of included trials, outcomes to be collected, and risks of bias. Disparity in assessment was settled by an independent adjudicator. The assessment of random sequence generation, allocation concealment, incomplete outcome data, blinding of participants and personnel, blinding of outcome assessment, and intention-to-treat analysis was done using the quality scale for meta-analytic review, the Cochrane Collaboration Tool for Risk of Bias.

Data analysis
Review Manager 5.3 was used to analyze the data. Analysis of dichotomous data was done using risk ratio, 95% confidence interval, and Mantel-Haenszel method with fixed effects model when there was no evidence of statistical heterogeneity. A random effects model was used in the presence of heterogeneity. Heterogeneity between trials was tested using a standard Chi-square test and I 2 statistics. The p-value of <0.10 was considered to be statistically significant and I 2 of �50% is considered to have high heterogeneity.

Description of studies
Six randomized controlled trials involving a total of 3,768 patients met the inclusion criteria. The invasive strategy group was composed of 1,986 patients while the conservative treatment group was consisted of 1,782 patients.
The data on population characteristics, intervention type, and measured outcomes were extracted from each trial ( Table 1). The summary of numerical data extracted from each trial was illustrated in S4 Appendix. Four of the trials included patients with NSTEACS aged � 70 years while two trials included patients � 65 years old [10,15]. The studies compared the effectiveness of early invasive strategy (treatment group) versus optimum medical treatment (control group) in the management of NSTEACS in patients � 65 years old.
In the control group all the trials used standard medical treatment [12][13][14][15][16][17]. In the treatment arm, four trials specified the time to intervention (4-72 hours) [12,14,15,16]. One of the six studies defined the time to intervention as coronary angiography and, if appropriate, revascularization within 7 days from admission for the index event [17]. Only one study did not specify the time to intervention but only mentioned "during initial admission" [13]. All of the trials included CABG as part of the intervention when indicated [12][13][14][15][16][17]. The study by Tegn et al.  2012) also reported that 860 out of 1,316 (65%) underwent PCI but also just stated that 71% had either PCI or CABG. Emails were sent to the authors to possibly obtain and clarify these information.
All trials assessed the outcome of all-cause mortality. Cardiovascular death was assessed by two trials [15,17]. Four trials reported the outcome of myocardial infarction [12,14,15,16]. Three trials assessed the outcome of stroke [12,14,15]. The outcomes of revascularization were reported by three studies [14][15][16]. The events of recurrent angina were assessed only by one study [15].
The Cochrane collaboration tool was used to assess the risk of bias. The random sequence generation, allocation concealment, incomplete outcome data, blinding of participants and personnel, blinding of outcome assessment, and intention-to-treat analysis were evaluated for each trial. All included trials were assessed to have low risk for bias ( Table 2). Five funnel plots (Figs 2-6) for the presence of publication bias were illustrated in the following pages. There was no funnel plot for the outcome of recurrent angina. A symmetrical plot may suggest either a low probability or absence of publication bias. However, for our meta-analysis, the power of the funnel plot to detect true publication bias may be low because the trials included were less than 10.

Effects of intervention on outcomes of interest
A. All-cause mortality. The follow-up periods of each RCT in this study were summarized in Table 1. For the outcome of all-cause mortality, the mean duration of follow-up is 4  (Fig 7). The pooled analysis of all-cause mortality showed no significant effect of  invasive strategy on the outcome with an overall risk ratio of 0.69 (95% CI 0.39 to 1.23) with significant heterogeneity (p value of 0.00001, I 2 = 91%) using the Random Effects model. Initial analysis using the Fixed Effects model was illustrated in Fig 8. Analysis showed that the study by Puymirat et al. achieved results that were strongly different from other studies in terms of the outcome of all-cause mortality. Hence, a pooled analysis that excluded this study was attempted but the results were still not significant as illustrated in Fig 9. B. Cardiovascular mortality. The mean duration of follow-up for cardiovascular mortality was 8 years. A total of 156 among 802 (19%) patients � 65 years old with NSTEACS developed the outcome of cardiovascular mortality in the Invasive Strategy Group; while 170 among 783 (22%) patients died of cardiovascular cause in the Conservative Group (Fig 10). The pooled analysis of cardiovascular mortality showed no significant effect of invasive strategy on the outcome with an overall risk ratio of 0.86 (95% CI 0.67 to 1.10).
C. Myocardial infarction. The average follow-up for the outcome of MI was 2 years. In the Invasive Strategy Group, there were 89 events of MI among a total of 926 (10%) patients; while there were 142 among 912 (16%) patients in the Conservative Group ( Fig  11). The pooled analysis showed that invasive strategy showed no significant effect on the outcome versus conservative treatment in preventing MI with an overall risk ratio of 0.63 (95% CI 0.39 to 1.04) with significant heterogeneity (p value of 0.07, I 2 = 60%) using the  (Fig 13). In the Invasive Strategy Group, there were 13 events of stroke among 874 (2%) patients; while there were 24 among 858 (3%) patients in the Conservative Group. The pooled analysis showed that early invasive strategy showed no significant effect on outcomes of stroke with overall risk ratio of 0.52 (95% CI 0.26-1.03, I 2 = 0%). E. Need for revascularization. The average duration of follow-up for this outcome was also 2 years. In elderly patients with NSTEACS, there were a total of 10 patients among 435 (2%) who needed revascularization in the Invasive Group while there were 34 patients among 441 (8%) in the Conservative Group (Fig 14). The pooled analysis for need for revascularization showed statistically significant benefit with an overall risk ratio of 0.29 (95% CI 0.14 to 0.59) with no significant heterogeneity (p value of 0.0006, I 2 = 3%).
F. Outcome for recurrent angina. Among the six trials, only one trial assessed the outcome of recurrent angina [15]. The follow-up period for this outcome was 1 year. An invasive strategy showed no significant effect on the outcomes of recurrent angina (RR 0.81, 95% CI 0.45-1.46, p = 0.49).

Discussion
Meta-analysis of data from the six trials included in this study showed that an early invasive strategy appears to be beneficial in terms of reduction of need for revascularization in suitable patients � 65 years old with NSTEACS. This finding implies that more patients in the conservative group clinically worsened during their course in the ward, requiring revascularization. It is also possible that early anatomic definition of the diseased coronaries may help the attending physician optimize an appropriate evidence-based management of the patient. The studies that evaluated the outcomes of revascularization stated that the indications for revascularization in the conservative group were: positive pre-discharge stress test, poor in-hospital outcomes, recurrent ischemia, reinfarction, malignant ventricular arrhythmias, refractory angina, and heart failure [14][15][16]. Some patients who subsequently required revascularization could have probably been better off with an early invasive approach. However, it is important to note that although there was a significant difference in the need for revascularization between the invasive (2%) and conservative (8%) groups, the rates of revascularization for both groups are low; hence, an implication that an invasive strategy is not a commonly applied management for this population limiting its broad application. Management strategies in patients > 65 years old with NSTEMI For the outcomes of death and MI, an invasive strategy showed no significant effect on the outcomes versus conservative treatment with significant heterogeneity. The possible sources of heterogeneity for the outcomes of death and MI may be the small number of events and sample sizes. In three out of the six studies, the patient population � 65 years old was just a subgroup analysis of the total population [12,13,17]. Hence, the population in the subgroup analysis may not be powered enough to detect the differences in the intervention and outcomes of interest. In addition, there were differences in age cutoffs. Two studies had age cutoffs of 75 years [12,15]; two had cutoffs of 65 years [12,17]; while the remaining two had age cutoffs of 70 and 80 years [14,16]. Possible clinical differences in outcomes may exist within the age brackets of this particular population. Furthermore, there were differences in follow-up periods, ranging from 3 months to 15 years, which may have contributed to the heterogeneity since shorter follow-up would mean lesser chance of catching the outcome of interest but longer follow-up would mean higher probability of the event occurring [12][13][14][15][16][17]. Likewise, the application of invasive strategy did not lead to reduction of cardiovascular mortality and stroke. Recurrent angina was assessed only in one study [15], which also showed no significant findings in favor of invasive strategy.
Overall, this study does not support the relatively conservative tendency when dealing with patients � 65 years old with NSTEACS in real-life clinical setting. The advance-aged population is considered a high-risk group wherein more than half the mortality in NSTEACS occur and a more aggressive approach in suitable patients may be more appropriate and beneficial particularly in reducing need for revascularization [5].
Frailty is common among older adults and is defined as a "state of reduced physiological reserve and increased vulnerability for poor resolution of homeostasis after a stressor event" [18]. A recent meta-analysis showed that frailty was associated with increased risk of cardiovascular diseases and was also associated with a 3-fold increase in risk for cardiovascular death [19]. The mechanism was said to be multi-factorial, likely owing to vascular, cardiac, cellular, bio-humoral, and endocrine alterations in the frail condition [19]. In addition, most especially during an ACS that is a stressor event, the increase in inflammation and thrombosis places the older patients at a more vulnerable state. Hence, a more aggressive approach may be suitable for appropriate, high cardiovascular risk patients. Among people who die of ischemic heart disease, 83% were >65 years of age [1]. This mortality rate is expected to increase in the forthcoming decades due to improving life expectancy of the elderly. Age is one of the most important predictors of risk in NSTEACS. Each 10-year increase in age results in a 75% increase in hospital mortality in ACS patients [20]. Despite the relatively higher risk in this age group, older ACS patients are under-represented in clinical trials such that subjects older than 75 years of age account for less than 10%, and those older than 85 years account for less than 2% of all NSTEACS subjects [7]. This highlights the need for more clinical trials and studies in this age group.
Data from the CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes with Early Implementation of the American College of Cardiology/ American Heart Association Guidelines) registry showed that NSTEMI patients aged � 65 years who experienced an in-hospital major bleed had a 33% increased risk of 30-day mortality [21]. However, the advancement of equipment and technique has made PCI safer for even very elderly patients (� 90 years of age) with high success rates and declining major bleeding risk [22].
The length of dual antiplatelet therapy (DAPT) for older patients is a concern especially for those who underwent PCI. However, there are still limited data on the optimal DAPT duration for patients for whom bleeding and ischemic risks should be considered. A recent cohort study (data obtained from the RENAMI Registry) evaluated the benefit-to-risk ratio of different DAPT durations (< 12 months, 12 months, and > 12 months) in ACS patients who underwent PCI [21]. The study showed that DAPT of at least 12 months is more beneficial than shorter DAPT in reducing net adverse clinical events (NACE), a combination of MACE and major bleeding complications. However, the result was different for the subgroup population > 75 years of age in this study since a reduced benefit was noted for prolonged DAPT > 12 months. The higher risk for harm than benefit was noted for this particular group most likely due to impaired renal function, peri-, and post-procedural bleeding [21]. Nevertheless, this issue may be addressed by using "PRECISE-DAPT" Score, a simple five-item tool that guides the optimal DAPT duration after PCI [22]. The study recommended that their finding of less favorable risk-benefit ratio for prolonged DAPT in those > 75 years old should be confirmed by RCTs.

Summary and conclusion
Results of this meta-analysis showed that there was a significant reduction in the need for revascularization in the invasive strategy group compared to the conservative treatment group in NSTEACS patients � 65 years old. This finding does not support the bias against early routine invasive intervention in this patient population.
Although an early invasive strategy may be favorable in the need for revascularization among those � 65 years old presenting with NSTEACS, the certainty of benefit versus risk still needs to be supported by larger clinical trials and registries with uniform age cutoff for this population, particularly � 65 years old, to provide high generalizability and statistical power. Current risk scoring systems such as the GRACE (Global Registry of Acute Coronary Events) Score, TIMI (Thrombolysis in Myocardial Infarction) Risk Score, and CRUSADE Bleeding Score are recommended in the initial evaluation of elderly patients presenting with NSTEACS. A special risk scoring may be developed to more accurately identify those who are suitable for an early invasive strategy, with an expected larger outcome and survival benefit.