The incidence and prognosis of coronary disease after acute coronary syndrome (ACS) in the short- and medium-term are among the most widely studied areas in cardiology.1–8 However, there is little information on the incidence and determinants of new adverse cardiovascular events during the long-term clinical course.9–11 Moreover, the limited information available has been obtained from studies of populations of ACS patients in whom coronary revascularization was little used9–11 even in the most contemporary series.9

On the other hand, the effect of many factors associated with the development of cardiovascular events after an ACS is often not considered according to whether the follow-up is short-to-medium term or long-term. Characterizing the clinical course after ACS in relation to major adverse cardiovascular events and identifying the precursors of such events could help to individualize clinical follow-up, assess the effectiveness of treatments, and optimize resource allocation. These aspects gain further relevance regarding forecasts of the substantial burden of morbidity and costs of ischemic heart disease12–14 and the importance of individual risk stratification in chronic disease management models.

In a cohort of ACS patients with a high rate of coronary revascularization, we studied the risk of acute myocardial infarction (AMI), stroke, or cardiovascular death during the first year of follow-up (adjusting for noncardiovascular mortality as a competing event) and compared it with the risk of these events in successive years. We also attempted to facilitate estimation of the risk of the 3 events by creating 2 intuitive clinical tools.

Table 1 shows the baseline characteristics of the 4858 patients (mean age=67.3 years; 28.7% women). Of these patients, 90.3% underwent coronary angiography and 68.6% were revascularized.

During 4.8±2.6 years of follow-up, there were 945 AMI, strokes, or cardiovascular deaths.

During the first year, 329 patients had a composite event (CIF = 7.3 [95%CI, 6.5-8.1]/100 person-years): The 329 events comprised 185 AMI, 40 strokes, and 104 cardiovascular deaths.

During the first year, 184 patients died of nonvascular causes. Thus, during the second follow-up period, the second cohort comprised the 4345 patients who survived the first year and remained event-free.

These patients were followed-up for 4.9±2.4 years. In total, 616 had a composite event (21.5 [95%CI, 18.3-24.7]/100 person-years). In the second cohort, the CIF at 1, 2, 3, 4, and 5 years of follow-up was 3.5, 6.0, 8.6, 10.1, and 13.2 events/100 person-years, respectively.

The 616 events during the second follow-up period comprised 274 AMI, 133 strokes, and 209 cardiovascular deaths.

Some characteristics were more frequent among patients who had a composite event during the first year: the patients were nearly 3 years older (73.6 years vs 70.8 years; P<.001), had a significantly higher prevalence of previous ischemic heart disease, diabetes mellitus, heart failure, and atrial fibrillation (Table 1), and more frequently had non—ST-segment elevation AMI and poor kidney function at admission. They were also less frequently prescribed acetylsalicylic acid, statins, angiotensin-converting enzyme inhibitors, or angiotensin receptor blockers.

Eight independent predictors of composite events were identified during the first year and 10 were identified during the successive years (Tables 2 and 3) (the effect of all the prognostic factors that composed the multivariable models is shown in Tables 1 and 2 of the supplementary material).

These predictors were used to design 2 intuitive quantitative tools to estimate the composite event risk during the first year (Figure 1) and during successive years (Figure 2).

Figure 1.

Scoring system and corresponding estimated risk of acute myocardial infarction, stroke, or cardiovascular death during the first year of follow-up. AMI, acute myocardial infarction; NSTEMI, non—ST-segment elevation myocardial infarction.

(0.21MB).

Figure 2.

Scoring system and corresponding estimated risk of acute myocardial infarction, stroke, or cardiovascular death during after the first year of follow-up. AMI, acute myocardial infarction; NSTEMI, non—ST-segment elevation myocardial infarction; TIA, transient ischemic attack.

(0.25MB).

The distribution of the risk scores for AMI, stroke, or cardiovascular death during or after the first year is shown in Figure 1 of the supplementary material. During the first year, the risk score values were -4 to 22, whereas in successive years they were –4 to 29.5.

We used the risk scores to stratify the risk of events into 3 groups (tertiles) and calculated the CIF.

Of the 329 composite events during the first year, 39 occurred in the first tertile or low-risk group (< 3 points; CIF=2.5 [95%CI, 1.8-3.4]/100 person-years), 76 in the second tertile or intermediate-risk group (3-6 points; CIF=4.8 [95%CI, 3.8-6.0]/100 person-years), and 214 in the third tertile or high-risk group (> 6 points; CIF=15.5 [95%CI, 13.6-17.7]/100 person-years; P<.001) (Figure 3).

Figure 3.

Cumulative incidence function of acute myocardial infarction, stroke, or cardiovascular death during the first year of follow-up stratified by risk tertiles. Adjusted for mortality as a competing event. 95%CI, 95% confidence interval; AMI, acute myocardial infarction; HR, hazard ratio.

(0.23MB).

In the cohort that had composite events in successive years, the CIF in the first tertile (low-risk, < 3 points), the second tertile (intermediate risk, 3-6 points), and the third tertile (high risk, > 6 points) was 10.7 (95%CI, 8.4-13.0)/100 person-years (n=104 events), 19.0 (95%CI, 16.0-22.0)/100 person-years (n=167 events), and 40.3 (95%CI, 36.0 to 44.5)/100 person-years (n=345 events), respectively (P<.001) (Figure 4).

Figure 4.

Cumulative incidence function of acute myocardial infarction, stroke, or cardiovascular death after the first year of follow-up stratified by risk tertiles. Adjusted for mortality as a competing event. 95%CI, 95% confidence interval; AMI, acute myocardial infarction; HR, hazard ratio.

(0.22MB).

The discriminative ability of the multivariable models was similar to that of the risk scores for the first year (0.74 [95%CI, 0.71-0.76] vs 0.73 [95%CI, 0.71-0.76]) and for successive years (0.69 [95%CI, 0.67-0.72] vs 0.68 [95%CI, 0.65-0.70]).

The discriminative ability of the risk score for AMI, stroke, or cardiovascular death during successive years remained virtually unchanged when the second follow-up period was restricted to 1, 3 or ≥ 5 years of follow-up (C statistic between 0.67 and 0.69).

The C statistic was 0.67 for the subgroups with LVEF < 35% and 35% to 45%, and 0.68 for subgroups with LVEF > 45% or unavailable LVEF, which are similar to the C statistic for the total group.

Both risk scores had good calibration (Hosmer-Lemeshow, P≥.44) (Figures 2 and 3 of the supplementary material).

This study shows that about 1 in 4 patients who are discharged from hospital after surviving an ACS will experience AMI, stroke, or cardiovascular death within the following 5 years. The risk of experiencing 1 of these events is particularly high in the first year, as shown by the finding that 34.8% (n=329) of the total number of events (n=945) occurred during the first 5±2 years of follow-up.

The results also show that the risk of AMI, stroke, or cardiovascular death decreases in successive years, but considerable residual risk remains such that 1 in 5 patients free of composite events during the first year will experience an event during the following 5 years.

Between 2006 and 2011, a Swedish study included more than 90 000 patients with AMI, of whom around 50% had undergone revascularization. One year after discharge, 18.3% of the patients experienced AMI, stroke, or cardiovascular death. At 3 years of follow-up, 20% of the patients who remained free of events during the first year experienced an event.9 However, between 2003 and 2004, the REACH registry20 included 21 890 patients with a history of ACS. Patients were followed-up until 2009. The cumulative incidence of AMI, stroke, or cardiovascular death was around 6% at 1 year and around 16% at 4 years. Similarly, between 2005 and 2010, Rapsomaniki et al21 studied more than 15 000 patients with ACS in the United Kingdom and found that the cumulative incidence of AMI, stroke, or cardiovascular death was 7.3%, 12.3%, and 17.7% at 1, 2, and 3 years, respectively. Thus, our results are in line with those recently obtained in different settings. All these results suggest that there is a high risk of recurrence of major cardiovascular complications after surviving the hospital phase of ACS.

Although there has been a recent decrease in mortality from ischemic heart disease,2,12–14 admissions for ACS will increase in Spain in the near future.12 The main cause of this increase will be the growing numbers of elderly people, which will account for up to 60% of all ACS patients by 2049.12 In addition to a greater mortality burden, the economic burden will also increase,22 because ACS care consumes large quantities of resources. At the beginning of the last decade, the direct costs alone of health care during the first year after an ACS were more than € 1 billion.22 Therefore, the management of chronic heart disease is a challenge for all the health services.

To meet this challenge, resources should be appropriately allocated to secondary prevention to reduce the high risk of future cardiovascular complications. Thus, it is essential to have data on the clinical course of the disease in its various stages and on the precursors of future adverse events, because the correct identification of the prognostic determinants would allow preventive therapies to be increased and allocated to high-risk patients, who are the most likely to benefit.23–28

In the present series, the predictors of AMI, stroke, or cardiovascular death comprised the interaction of classic risk factors (age, smoking, and diabetes mellitus), markers of atherosclerotic burden (previous ischemic heart disease, previous stroke or transient ischemic attack, peripheral artery disease, and renal impairment), and indicators of the extent of coronary artery disease (non—ST-segment elevation AMI and heart failure). Six of the 10 determinants of prognosis after the first year were also determinants during the first year, although the magnitude of the individual effect of these determinants varied according to the period studied. This study also shows that the proposed scoring systems can be used at the bedside or in the office to quantify the individual risk of AMI, stroke, or cardiovascular death.

All the predictors of events identified in our registry have already been previously described.9,11,20,29 The fact that smoking was associated with events during the first year but not during successive years may be because smoking cessation reduces or stops atherosclerotic plaque destabilization caused by smoking. However, this mechanism must remain speculative because no data were available on smoking after discharge. The reason for a history of atrial fibrillation being a predictor of events in the first year alone may be because the antithrombotic treatment strategy for these patients is not well defined and the effects are noticeable during the first months after discharge.

The scoring system for predicting events at 1 year of follow-up had acceptable discriminative power and good calibration. In contrast, the scoring system for predicting events in successive years had moderate discriminative power (0.68). Aside from these issues, we attempted to provide a simple method to identify prognostic determinants such that clinicians can become familiar with these determinants and incorporate them into clinical decision-making, thus helping them to stratify risk to guide prognostic strategies of proven efficiency23–28 and optimize follow-up strategies in cardiac rehabilitation programs.27

Although current financial constraints make it difficult to improve health resource allocation, such improvement is essential to address the vulnerability factors for various diseases. Individual stratification of long-term prognosis is needed to efficiently manage patients with chronic diseases. This study provides information on the magnitude of the impact of coronary disease and its predictors in a contemporary cohort of ACS patients. The scoring systems proposed in this study are not intended to replace clinical judgment, but to help the clinician to make well-reasoned balanced clinical judgments to optimize patient management and resource use.

The risk of the recurrence of cardiovascular events remained elevated in this contemporary cohort of ACS patients with a high rate of coronary revascularization. Based on a group of readily available predictors, risk can be easily quantified with acceptable predictive ability. Individual stratification of patients into groups at low or high risk of long-term events facilitates efficient resource use and the development of evidence-based chronic disease management programs.

None declared.