Article Text

Original research
Management and outcomes over time of acute coronary syndrome patients at particularly high cardiovascular risk : the ACSIS registry-based retrospective study
  1. Tzlil Grinberg1,2,
  2. Yoav Hammer1,2,
  3. Maya Wiessman1,2,
  4. Leor Perl1,2,
  5. Tal Ovdat3,
  6. Or Tsafrir4,
  7. Yoni Kogan5,
  8. Roy Beigel3,
  9. Katia Orvin1,2,
  10. Ran Kornowski1,2,
  11. Alon Eisen1,2
  1. 1Cardiology Department, Rabin Medical Center, Beilinson Hospital, Petah Tikva, Israel
  2. 2Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
  3. 3Heart Institute, Sheba Medical Center, Ramat Gan, Israel
  4. 4Cardiology Department, Western Galilee Medical Center, Nahariya, Israel
  5. 5Cardiology Department, Assuta Medical Center, Ashdod, Israel
  1. Correspondence to Dr Alon Eisen; alon201273{at}


Objective Some patients following acute coronary syndrome (ACS) are at particularly increased risk for recurrent cardiovascular events. We aimed to examine temporal trends in the management and outcomes across the spectrum of these particularly high-risk patients.

Design and setting A retrospective study based on the ACS Israeli survey (ACSIS) registry, a multicentre prospective national registry, taking place biennially in 25 cardiology departments in Israel. Temporal trends were examined in the early (2002–2008) and late (2010–2018) time periods.

Participants Consecutive patients with ACS enrolled in the ACSIS registry were stratified according to the Thrombolysis in Myocardial Infarction Risk Score for secondary prevention (TRS2°P) to high (TRS2°p=3), very high (TRS2°p=4) or extremely high risk (TRS2°p=5–9). Patients with TRS2°p<3 were excluded. From the initial 15 196 patients enrolled, 5359 patients were eventually included.

Clinical outcome measures included 30-day major adverse cardiovascular events (MACE) and 1-year mortality.

Results Among 5359 patients (50% high risk, 30% very high risk and 20% extremely high risk), those with a higher risk were older, had more comorbidities, presented more with non-ST elevation myocardial infarction, and were treated less often with guideline-recommended pharmacotherapy and percutaneous coronary intervention. Over time, treatment has improved in all risk strata, and the rate of 30-day MACE has significantly decreased in all risk groups (from 21% to 10%, from 22% to 15%, and from 26% to 16%, in high, very high and extremely high-risk groups, respectively, p<0.001 for each). However, 1-year mortality decreased only among high and very high-risk patients, and not among extremely high-risk patients in whom 1-year mortality rates remained very high (28.7% vs 28.9%, p=1).

Conclusion Within a particularly high-risk cohort of patients with ACS, treatment has significantly progressed over almost 2 decades. While short-term outcomes have improved in all risk groups, 1-year mortality has remained unchanged in extremely high-risk patients with ACS.

  • coronary heart disease
  • ischaemic heart disease
  • myocardial infarction

Data availability statement

No data are available.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Strengths and limitations of this study

  • A multicentre prospective national registry-based retrospective study examining over two decades the treatment and outcome of consecutive acute coronary syndrome patients at particularly high cardiovascular risk .

  • The cohort at focus is unique and rarely included in clinical trials.

  • Data are complete and well documented.

  • Data on cardiovascular mortality and treatment adverse effects were lacking.

  • A causal relationship between improved treatment and a favourable cardiovascular outcome cannot be inferred due to the study’s observational nature.


Patients with acute coronary syndrome (ACS) are at increased risk for recurrent cardiovascular (CV) events.1–3 However, some patients are at particularly high, at times at the highest, CV risk.4–6 These patients are often those with a multitude of comorbidities, such as complex coronary artery disease, polyvascular disease, chronic kidney disease (CKD), diabetes and advanced age, but are also frequently the most debilitated and frail.7 8 They are also at an increased risk of having adverse events from medical or invasive treatments.9 10 Accordingly, these patients are often undertreated, although they might benefit the most from guideline-directed medical treatment (GDMT). Hence, therapeutic decision-making in these very high-risk patients is often challenging, especially in view of the paucity of guiding data, in part due to the scarce enrollment of these patients in clinical trials.11 12

It has been previously demonstrated13–17 that patients who were at a higher baseline risk for major adverse cardiovascular events (MACE) according to the Thrombolysis in Myocardial Infarction (TIMI) Risk Score for Secondary Prevention (TRS2°P) derived proportionally greater benefit from improved treatment over time compared with low and intermediate-risk patients. When appropriate secondary prevention treatment and cardiac rehabilitation were provided to high-risk patients (TRS2°p≥3), a larger relative risk reduction in clinical outcomes was observed compared with lower-risk patients, despite the fact theformer were oftentimes relatively undertreated .18 19 The inverse association between patients’ CV risk and the delivery of GDMT has long been acknowledged and referred to as the ‘risk-treatment paradox’.20 21

It is still unclear whether over the years intensive treatment has been used more often in ACS patients at the highest risk for CV events and whether their clinical outcomes have improved. We aimed to examine temporal trends over almost two decades in the management and outcome across the spectrum of high-risk post ACS patients according to the TRS2°P.


A multicentre study from the ACS Israel Survey (ACSIS), a biennial prospective national registry of all patients with ACS hospitalised over a 2-month period (March to April) in 25 coronary care units and cardiology departments in all general hospitals in Israel. This registry entails demographic, clinical and angiographic data of all consecutive patients included in the survey. Data were recorded on prespecified forms for all admitted patients diagnosed with ACS. Admission and discharge diagnoses were documented by the attending physicians based on standard electrocardiographic, clinical and biochemical criteria.

Included in the study were all patients enrolled in the ACSIS registry during the years 2002–2018 who had available TRS2°P data. Excluded were patients with TRS2°p<3, considered to be at low or intermediate risk. All patients in each of the medical centres involved signed an informed consent form.

The TRS2°P is a simple risk score incorporating nine clinical characteristics, each assigned one point in the total count. These characteristics include age ≥75, diabetes mellitus, hypertension, current smoking, peripheral vascular disease (PVD), prior stroke, prior coronary artery bypass graft surgery (CABG), chronic heart failure (CHF), and CKD (defined by Modification of Diet in Renal Disease (MDRD) as <60 mL/min). This score was introduced relatively recently14 and was originally designed to predict a graded risk for MACE at 3 years postmyocardial infarction (MI). It was later validated for secondary prevention in other studies13 18 19 22–24 as an effective risk stratification tool that can distinguish a pattern of increasing benefit with the application of GDMT. In this study we suggested further stratification of this score pertaining specifically to the high-risk spectrum of patients; Patients were stratified for recurrent CV events by the TRS2°P into three groups: high-risk (TRS2°p=3), very high risk (TRS2°p=4) and extremely high risk (TRS2°p≥5).

Temporal trends were examined in the early (2002–2008) and late (2010–2018) time periods. This time partition was chosen to reflect the advancement in patients’ care after an ACS brought about by the later era, when percutaneous coronary intervention (PCI), radial approach, more efficacious antiplatelets, high-potency statins, and cardiac rehabilitation have become the standard of care. Temporal trends in management were examined by referring to all aspects of secondary prevention therapy over time, including pharmacotherapy, PCI, and cardiac rehabilitation. Another analysis of temporal trends in treatment and outcome by survey years was additionally performed (online supplemental appendix table S1) to underscore the general direction and trend of the results.

Clinical outcomes included 30-day MACE and 1-year mortality from the index hospitalisation with ACS. MACE was a composite of death, MI, unstable angina pectoris, stroke, stent thrombosis or urgent revascularisation. In-hospital and 30-day outcome data were ascertained by hospital chart review, telephone contact and clinical follow-up data. Mortality data during hospitalisation, at 30 days and 1 year were determined for all patients based on hospital charts and by matching patients’ identification numbers with the Israeli National Population Register.

Patient and public involvement

Patients or the public were not involved in the design, conduct, reporting or dissemination plans of the study.

Statistical analysis

Patients’ characteristics were presented as n (%) for categorical variables, and as median (IQR) for continuous variables. The three groups (high, very high, and extremely high) were tested with a χ2 test for trend for comparison of categorical variables, and with Kendall rank correlation for continuous variables. The two groups (early vs late period) were tested with χ2 for categorical variables and with Mann-Whitney-Wilcoxon test for continuous variables. Survival curves were presented, and the Kaplan-Meier log-rank test was used to test the variable of interest on survival. To evaluate the probability of 30-day MACE a logistic regression model was performed adjusted for TRS2°P study group, time period, and their interaction (risk group * Period). Using the same adjustments, the probability of 1-year mortality was evaluated with a Cox proportional hazard model. Multivariate analyses for the outcomes of interest were performed for the first time period. Covariates that were found statistically significant in the univariate analysis and had less than 10% missing data were selected for the model. All tests were conducted at a two-sided alpha level of 0.05 which was considered statistically significant. No statistical adjustments were performed to compensate for missing data. Statistical analysis was performed using R software (Vienna, Austria, 2020).


A total of 15 196 patients were enrolled in the ACSIS registry during the years 2002 to 2018. Among them, 14 427 patients had available TRS2°P data and 5359 patients had TRS2°p≥3 and were included in the study (online supplemental appendix figure S1). Of the patients included, approximately 50% (n=2710) were classified as high risk, 30% (n=1557) as very high risk and 20% (n=1092) as extremely high risk (figure 1). Baseline characteristics of patients are presented in table 1. Compared with high-risk and very high-risk patients, extremely high-risk patients were older (mean age 77 years vs 69 years and 75 years, respectively), from a lower educational background, and had more comorbidities including hypertension (97%), diabetes mellitus (81%), dyslipidaemia (82%), CHF (51%), CKD (54%), PVD (48%), prior MI (65%), prior revascularisation by CABG or PCI (47% and 52%, respectively) and prior stroke (39%).

Table 1

Baseline characteristics by TIMI risk score for secondary prevention (TRS2°P)

Figure 1

Patient distribution by the TIMI risk score for secondary prevention (TRS2°P). Risk factors: age ≥75 years, diabetes mellitus, hypertension, current smoking, peripheral arterial disease, prior stroke, prior coronary artery bypass graft surgery, chronic heart failure and chronic kidney disease (eGFR by MDRD <60 mL/min). High risk=3 risk factors; Very high risk=4 risk factors; extremely high risk ≥5 risk factors. eGFR, estimated glomerular filtration rate; MDRD, Modification of Diet in Renal Disease; TIMI, Thrombolysis in Myocardial Infarction.

Extremely high-risk patients had already been taking secondary prevention pharmacotherapy to a larger extent compared with the other groups. Concerning the index ACS (table 2), on hospital admission, extremely high-risk patients presented more often with non-ST elevation MI. They also presented with a higher heart rate and a lower blood pressure, a higher Killip class with a reduced ejection fraction, a higher GRACE score and peak troponin I level, and a greater proportion of atrial fibrillation on ECG.

Table 2

Characteristics of the index ACS and in-hospital complications by TIMI risk score for secondary prevention (TRS2°P)

Still, they had undergone fewer coronary angiographies and PCI’s (66% vs 86% and 41% vs 59% in extremely high vs high-risk patients, respectively, table 3).

Table 3

Interventional treatment and pharmacotherapy of index ACS and clinical outcomes at 30 days and 1 year

Extremely high-risk patients developed more in-hospital complications including pulmonary oedema, new-onset atrial fibrillation, acute renal failure, bleeding requiring blood transfusions, and moderate-to-severe mitral regurgitation (table 2). At hospital discharge, extremely high-risk patients were prescribed secondary prevention therapy to a lesser extent (table 3), with less aspirin, fewer potent P2Y12 inhibitors (merely 2% received prasugrel and 9% ticagrelor), and slightly fewer statins and ACE inhibitors (ACEi)/ angiotensin receptor blockers (ARBs). Referral to cardiac rehabilitation at 30-day post-MI was also lower accordingly.

Compared with high and very high-risk patients, the clinical outcomes of extremely high-risk patients were worse (table 3), with a higher 30-day MACE (15.5%, 18.6% and 21.3%, respectively), and 1-year mortality (12.8%, 18.9% and 28.8%, figure 2).

Figure 2

Kaplan-Meier curves for 1-year survival by the TIMI risk score for secondary prevention (TRS2°P). TIMI, Thrombolysis in Myocardial Infarction.

Comparing the early and late time periods (online supplemental appendix table S2), patients in the late period had more comorbidities, yet presented less often with heart failure and an advanced Killip class. They underwent significantly more coronary angiographies and PCI’s and experienced fewer in-hospital complications. The utilisation of guideline-recommended pharmacotherapy at hospital discharge across all TRS2°P groups has become more prevalent over time, including potent P2Y12 inhibitors (which were entirely not in use in the early period), statins, and beta-blockers, and so has cardiac rehabilitation referral at 30-day follow-up (table 4). Yet, for the most part, the absolute percent of these GDMT measures was still lower with higher TRS2°P scores. It is noteworthy that contrary to the other risk groups, extremely high-risk patients did not demonstrate enhanced use of aspirin and ACEi/ARBs over time.

Table 4

Temporal trends in GDMT by discharge among the entire cohort and by TIMI risk score for secondary prevention (TRS2°P)

With respect to clinical outcomes over time (table 5), the rate of 30-day MACE has decreased among all high-risk-strata comparing the early to the late period (20.5% to 10.5%, 22.3% to 14.8% and 25.9% to 16.4%, in high risk, very high risk and extremely high-risk groups, p<0.001 for each, respectively), driven by almost all of its individual components. One-year mortality rate has decreased as well (online supplemental appendix figure S2), however, exclusively among high and very high-risk patients (15.5% to 9.8%, p<0.001; 21.7% to 15.9%, p=0.005) but not among extremely high-risk patients (28.7% to 28.9%, p=1, figure 3). The rates of 30-day rehospitalisation, recurrent MI, and angina have also declined over time.

Table 5

Temporal trends in clinical outcomes among the entire cohort and by TIMI risk score for secondary prevention (TRS2°P)

Figure 3

Kaplan-Meier curves for 1-year survival by time periods stratified by the TIMI risk score for secondary prevention (TRS2°P). TIMI, Thrombolysis in Myocardial Infarction.

In a logistic regression analysis adjusted to the risk level and time period, the odds for 30-day MACE decreased by 49% between the early and late time periods (adjusted OR 0.51, 95% CI 0.44 to 0.60, p<0.001), with no significant interaction between risk-level (TRS2°P) and time period. In a Cox regression analysis, the probability of 1-year mortality decreased by 25% between the early and late time periods (HR 0.75, 95% CI 0.66 to 0.86, p<0.001). The interaction between risk level and time period was significant only for the extremely high-risk group (p=0.002). When Cox regression analysis was performed for each risk level, 1-year mortality was lower over the late period among both high and very high-risk but not among extremely high-risk patients (online supplemental appendix table S3). Multivariate analysis for the predictors of 30-day MACE and 1-year mortality in the early time period was additionally performed (online supplemental appendix table S3).


In this study from a national ACS registry, several findings were demonstrated. Within a population of ACS patients who were at particularly high risk for recurrent CV events, the higher the risk, the older and more comorbid were the patients, and these were treated less commonly with GDMT (angiography, pharmacotherapy and cardiac rehabilitation). Accordingly, a graded risk for worse clinical outcomes (30-day MACE and 1-year mortality) was evident with higher TRS2°P scores. Nevertheless, over time, the utilisation of GDMT at hospital discharge has become more prevalent in all high-risk groups. Clinical outcomes have similarly improved among all high-risk groups throughout time periods with a decrease in 30-day MACE and 1-year mortality, with the exclusion of extremely high-risk patients who did not demonstrate improved 1-year mortality over time. This benefit was observed despite the fact that compared with patients in the early period, patients included in the late period had more baseline comorbidities that were unadjusted for in the analysis, therefore implying an even larger positive trend in outcome.

The TRS2°P was designed originally in the TRA2°P-TIMI 50 trial14 to predict CV death, MI or stroke at 3 years post-MI. Since then, several studies have validated its utility as a risk prediction model in other,16 22 23 25 at times higher-risk populations,4 24 as well as for other prespecified endpoints13 and for a shorter term. It has been substantiated not only for risk-stratification but also for distinguishing a pattern of increasing benefit from optimal treatment with increasing TRS2°P scores.18 19 In keeping with prior evidence, in this study, we demonstrated that for a particularly high-risk cohort of patients following ACS, further stratification of the TRS2°P was predictive of not only a greater risk for recurrent CV events and death but also of which patients will develop hospital complications such as heart failure, atrial fibrillation, and bleeding.

This study focused on a population who is at the highest risk for secondary CV events, far exceeding the risk attributed to very/extremely high-risk patients reported in official guidelines,26 27 and much higher than the ‘average’ ACS patients who are usually included in clinical trials. This patient cohort is of special interest since only sparse evidence11 28 is available regarding the recommended optimal treatment of these patients, who pose a therapeutic challenge due to their complex comorbidities and who are frequently managed conservatively (angiography and high potency platelets abandoned).29 In fact, we hardly found any studies addressing this patient group specifically, and the majority of data was derived from subgroup analyses of other populations.18 19 30

Indeed, the study patients had significant comorbidities, with the majority of patients having diabetes mellitus, about a half with prior MI, a third smokers, and almost all having hypertension. They developed considerable in-hospital complications and an overwhelming rate of MACE and mortality. Nonetheless, over two consecutive time periods, they exhibited improved clinical outcomes, mainly lower 30-day MACE, 30-day mortality and 1-year mortality, in association with better implementation of discharge GDMT.

While 30-day MACE has improved among all risk groups over time periods, 1-year mortality has not changed among extremely high-risk patients. Although there could be several plausible explanations for this finding, it is probably associated with the high burden of comorbidities, which outweighed the benefit GDMT might have had on their CV risk. Supporting this hypothesis is the aforementioned improvement in 30-day MACE. Still, this finding could simply be due to chance. Overall, for the entire high-risk patient cohort who is the subject of this study, clinical outcomes were favourable despite their multitude of CV risk factors.

Still, the question of whether GDMT should be more often applied in the post-ACS patients from the highest risk spectrum cannot be answered by this study, owing to its retrospective observational nature and the lack of consideration of net clinical benefit. Randomised controlled studies will be needed to bridge this knowledge gap. However, this study supports and serves as a starting point for such randomised trials to be conducted.

This study is multicentre and based on a large prospective national registry with pre-specified parameters and endpoints. Data are complete and well documented on prespecified forms by several teams from different medical centres. However, there are limitations to this study. First, data referring to medication use pertained only to discharge and 30-day follow-up, hence we cannot infer medication adherence. Second, no data were available with respect to the PCI procedure including the specific number, location, and type of stents deployed, coronary anatomy, and lesion severity. Third, high-sensitivity cardiac troponin was only available for the late time period, which may have affected the results. Forth, the proportion of missing values was significant for some of the variables (online supplemental appendix table S4). Fifth, no data were collected regarding the cause of death, thus CV mortality was not reported. Finally, based on this study we cannot deduce a causal association between increased GDMT and a favourable CV clinical outcome. Moreover, there are probably other important contributing factors not accounted for, such as enhanced treatment adherence, better pharmacological treatment of associated highly prevalent comorbidities, and a greater frequency of post-discharge community monitoring, among other factors.


Within a particularly high-risk cohort of patients with ACS, the use of GDMT has become more prevalent over almost two decades. While short-term outcomes have improved in all high-riskgroups, 1-year mortality has improved only in the high and very high-risk groups. While a causal relation cannot be inferred, GDMT should not be denied in these high-risk patients, and efforts should be made to develop strategies to intensify treatment with closer follow-up.

Data availability statement

No data are available.

Ethics statements

Patient consent for publication

Ethics approval

Each medical centre involved received the approval of its institutional review board in compliance with the Declaration of Helsinki.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Contributors AE and TG contributed to the conception of the work. AE, YH and TG contributed to its design. MW, LP, TOv, YK, RB and KO contributed to the acquisition of data for the work. TOr contributed to the statistical analysis. TG and AE drafted the manuscript and, together with RK, contributed to the interpretation of data. All authors critically revised the manuscript, gave final approval, and agreed to be accountable for all aspects of the work ensuring integrity and accuracy. TG is responsible for the overall content as guarantor.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.