Cardiovascular Event Rates After Myocardial Infarction or Ischaemic Stroke in Patients with Additional Risk Factors: A Retrospective Population-Based Cohort Study

This large study with nationwide coverage of Swedish register data shows that patients with a history of MI or IS, and additional risk factors have a very high risk of suffering subsequent MACE despite previous use of standard-of-care lipid-lowering therapy (i.e. moderate- or high-intensity statins, with or without ezetimibe). This is particularly true for patients with specific additional risk factors, such as diabetes, CKD, subsequent events within 2 years, and polyvascular disease.

Patients in the prevalent cohorts may have had their qualifying MI or IS event up to 5 years before the index date, while patients in the incident cohorts were followed from the time of the qualifying event. This led to differences in MACE rates between the two cohorts, as the risk of a second event is highest in the first years after the initial event and decreases thereafter. The risk of a subsequent CV event has been reported to be higher in the first year post-MI and remain high in the following years [1], and this study aimed to describe the change in CV risk over time by producing MACE yearly rates for the incident cohorts.

In patients with an incident MI or IS, first MACE rates were highest in the first 2 years after the index event, decreasing during year 2 and remaining stable in year 3 and beyond. At 1 year after the index event, the rates were more than three times that seen after year 3 across all subgroups. Moreover, in the first year after MI or IS, the MACE yearly rates in the subgroups of patients with specific additional risk factors were almost double those of the overall incident MI or IS cohorts.

As expected, first MACE rates in the present study based on real-world clinical practice data were considerably higher than those seen in recent clinical trials. Notably, this data set was designed to have almost identical inclusion criteria to the FOURIER study [23]. To allow a comparison of real-world and clinical trial data, we conducted an ad hoc analysis of data from the FOURIER study, showing that first MACE rates in the placebo arm in patients with a history of MI were 3.4 per 100 person-years overall, and 3.8–6.3 per 100 person-years in the subgroups with additional risk factors (Table S2 in the supplementary material) (Amgen, data on file), compared with 6.2 per 100 person-years overall and 10.7–13.2 per 100 person-years in the subgroups for prevalent patients in the present study. In patients with a recent MI (within less than 1 year) in the placebo arm of FOURIER, the first MACE rate was 3.8 per 100 person-years, compared with 11.9 per 100 person-years in the incident MI cohort in the present study. Similarly, in the ODYSSEY study, 1126 MACE events (but also including non-CV mortality) were reported in 9462 patients with a recent acute coronary syndrome followed for 2.8 years, meaning an event rate of 4.3 per 100 person-years for the placebo arm [14]. First MACE rates in patients with a history of IS in the placebo arm of FOURIER were 4.0 per 100 person-years overall and 4.7–6.7 per 100 person-years in the subgroups, compared with 6.9 and 9.9–12.3 per 100 person-years, respectively, in prevalent patients in the present analysis (Table S2 in the supplementary material) (Amgen, data on file). The first MACE rate in patients with a recent IS in the placebo arm of FOURIER was 4.3 per 100 person-years, compared with 12.3 per 100 person-years in the incident IS cohort in the present study.

This very high risk of subsequent MACE in patients with MI or IS and additional risk factors indicates that there is a large unmet need with current disease management, and that additional interventions may be warranted, particularly early after an MI or IS. Indeed, this issue has been raised in European treatment guidelines, which recommend the use of aggressive secondary prevention interventions, including intense LDL-C-lowering treatment [1, 24]. This is also consistent with a recent Swedish study that compared LDL-C levels achieved 6–10 weeks after an MI to the 2019 European Society for Cardiology/European Atherosclerosis Society guidelines LDL-C goal (below 1.4 mmol/L and at least a 50% reduction in LDL-C [1]) and found that more than 80% of patients were eligible for escalated lipid-lowering therapy [25]. Ensuring that patients with ASCVD and additional risk factors are offered intensified treatment known to decrease CV risk is crucial to avoiding a recurrence of CV events. In addition, a range of other preventive measures should be considered alongside lipid control, including encouraging adherence to lipid-lowering therapy, smoking cessation, diet and lifestyle advice, and, where relevant, control of hypertension and hyperglycaemia [1, 24].

Strengths of the present study include the long follow-up period, the large sample size, and the robustness of the data sets used, which cover all inpatient and specialised outpatient visits occurring in Sweden between 2006 and 2015. The National Patient Register contains more than 99% of hospitalisations, while the Prescribed Drug Register covers all prescriptions dispensed at pharmacies. This study, therefore, provides good estimates on event rates in patients in real-world clinical practice, rather than being based on highly selected clinical trial populations. As a result, our data could serve as a valuable source for populating health economic models evaluating interventions reducing CV risk.

Limitations of the study include the small sample size of some subgroups, and the lack of data on LDL-C levels, meaning that changes in CV risk could not be linked to the level of control of hyperlipidaemia. It is, however, well documented that the use of intensive lipid-lowering therapy is associated with CV risk reduction in very high-risk patients in clinical practice [26‐28]. It should be noted that patients were included on the basis of the use of moderate- or high-intensity statins during the 1 year before the index date, but may have stopped or changed therapy by the index date. In addition, statin use, including adherence, and other concomitant therapies and interventions were not studied during follow-up, so potential treatment discontinuations and changes in type or intensity of treatment were not accounted for. It should also be noted that some very high-risk patients may reside in secondary care, such as rehabilitation facilities, and their statin use may not appear in a register of prescriptions. Furthermore, at the beginning of the study period, secondary prevention with high-intensity statins was not standard clinical practice in Sweden but became standard of care in 2012. This study was descriptive in nature. No statistical comparison of differences in rates between the subgroups was conducted, and further research to assess the statistical significance of the differences in rates might be of interest, in line with existing studies [21, 22]. It might also be of interest to investigate the relevance of other risk factors that were not included in the present analysis, such as smoking or hypertension, because of the limitations of the data. Finally, it is important to note that, as this study was conducted in Sweden, the results might not be generalisable to other countries.