APOLLO: Worse Long-Term Outcomes in American vs European Survivors of Acute MI

BARCELONA, Spain—Patients who survive past 1 year after acute myocardial infarction (MI) are at high risk of outcomes such as death and hospitalized bleeding, according to results presented on August 31, 2014, at the European Society of Cardiology Congress. Additionally, there is substantial risk variation among the United States and several European countries, although no one country has an overall lower risk profile.

Harry Hemingway, MD, of University College London (London, England), said the “definition of ‘high-risk’ needs to be considered” for these patients as the guidelines currently focus on stable coronary populations without specifying post-MI patients. “These international comparisons provide impetus to improve the quality of whole health care systems,” he continued. “We have data to support a primary care chronic disease management approach.”

Methods
For the APOLLO study, Dr. Hemingway and colleagues looked at administrative and electronic health records from 140,887 acute MI patients who were discharged from hospitals in Sweden (n = 77,798), England (n = 7,238), France (n = 1,764), or the United States (n = 53,909) 1 year prior to study enrollment from 2002 to 2011.
At baseline, US patients were more likely to have hypertension, diabetes, and a history of heart failure, A-fib, COPD, hospitalized bleeding, renal disease, and PAD. French patients were the most likely to receive PCI (61.6%) and US patients were the most likely to receive CABG (19.7%). Although baseline medication information is missing for the United States, most patients in the other 3 countries were taking statins, beta-blockers, ACE inhibitors or angiotensin receptor blockers, and aspirin. More than half (54.4%) of French patients were on dual antiplatelet therapy, though this number was lower in both Sweden (22.9%) and England (41.0%).


Observed 3-year all-cause death and the combined endpoint of MI, stroke, and all-cause death were highest in US patients, though the differences evened out after adjustment (table 1). 

Table 1. Three-Year Event Risks by Country

 

Sweden

(n = 77,798)

United States

(n = 53,909)

England

(n = 7,238)

France

(n = 1,764)

Observed Risk

    All-Cause Death

    MI, Stroke, All-Cause Death

 

20.1%

26.9%

 

30.2%

36.2%

 

13.7%

24.1%

 

14.3%

17.9%

Adjusted Risk

    All-Cause Death

    MI, Stroke, All-Cause Death

 

11.2%

19.8%

 

12.8%

18.2%

 

8.7%

21.3%

 

12.4%

16.7%

 

Compared with Swedish patients, those in the United States were at higher risk for all-cause death even after adjustment for comorbidities (adjusted RR 1.11; 95% CI 1.05-1.18) and revascularization choice (adjusted RR 1.17; 95% CI 1.11-1.24). However, there was no difference between these countries for the combined endpoint after adjustment for the same confounders. There were no differences between England and Sweden, though French patients were at slightly lower risk for the combine endpoint compared with Swedish patients after adjustment for both comorbidities (adjusted RR 0.82; 95% CI 0.68-0.98) and revascularization choice (adjusted RR 0.78; 95% CI 0.64-0.94).

Cumulative hospitalized bleeding risk was higher in the United States (5.3%) than in Sweden (2.5%), England (3.6%), and France (2.2%). Moreover, Sweden had lower risks of bleeding compared with both the United States and England even after adjustment for both comorbidities and revascularization choice (table 2).

Table 2. Adjusted Cumulative Bleeding Risk vs Sweden

 

Adjusted RR

95% CI

United States

    Comorbidities

    Revascularization Choice

 

1.64

1.69

 

1.45-1.84

1.51-1.90

England

    Comorbidities

    Revascularization Choice

 

1.94

1.94

 

1.28-2.93

1.28-2.93

 

Dr. Hemingway said they used Sweden as the reference population because “it’s the largest sample, the data quality is the best, and it’s also the most homogeneous population.”

Missing data, he continued, was the biggest limitation of the study—medication information for the United States, cause-specific mortality data for the United States and France, and socioeconomic data for Sweden, United States, and France.  

Electronic Records Help to Study Growing Population

Studying this patient population is important because “clinical care at this point is waning… and this population is growing because more people are surviving their acute myocardial infarctions,” Dr. Hemingway said, adding that it “behooves all of us to make better use of administrative and electronic health records that exist in all of our countries.”

Panelist François Schiele, MD, PhD, of the Université de Franche Comté (Besançon, France), commented that while administrative data is helpful for generating hypotheses, “from a French point of view, any study coming from [the French database used in APOLLO] is subject to bias because we have learned in the past that these very large databases cannot be used for reliable conclusions.”

Dr. Hemingway disagreed in that “the good news is that there’s no other show in town for getting unselected populations and reliable estimates…. They do not suffer the famous biases of voluntary registries which always underestimate event rates.” Still, he agreed that “these data need to be improved and the quality needs to be examined.”  

Panel moderator, Udo Sechtem, MD, of Robert-Bosch Medical Center (Stuttgart, Germany), said the main question relates to whether observed risk differences are “due to the health system or doctor performance or … to intrinsic differences in the population, which you cannot rectify.”

Dr. Hemingway replied that since hospitalized bleeding differences cannot be attributed to background population characteristics, overall variability is most likely due to health care systems. For example, US life expectancy is lower at 65 years than in other European countries, he said, possibly because “many of those [patients] were previously uninsured and are carrying with them an extra mortality risk, whereas in some of the European populations, they have experience of long-term primary care management.”

 


Source:

Hemingway H. International comparison of outcomes among 140,880 patients stable after acute MI; real-world evidence from electronic health and administrative records. Presented at: European Society of Cardiology Congress; August 31, 2014; Barcelona, Spain. 

Related Stories:

Disclosures
  • APOLLO was funded by AstraZeneca.

Comments