gms | German Medical Science

Background: Intravascular imaging-guided percutaneous coronary intervention (PCI) has advantages over angiography-guided PCI, but randomised controlled trials may be limited by patients with a lower risk profile.

Objective: The aim is to compare international studies with either real-world data or data from clinical trials comparing IVUS-, OCT- and angio-guided PCI with regard to basic patient characteristics

Methods: Patients undergoing at least one PCI in the inpatient setting, with or without imaging, were included. The baseline characteristics of patients in four different studies were compared. Of these four studies, two were retrospective studies using insurance claims of Medicare (US RWE) and German Statutory Health Insurance (MADEIRA) respectively. The third study is a randomised open-label trial from 38 centres in Europe (OCTOBER), and the fourth is a randomised open-label trial using data from 9 centres in South Korea (OCTIVUS).

Results: Patients from the real-world study populations are on average older than those from randomized open-label trials and have a more balanced gender distribution (in the US RWE more than in MADEIRA). The prevalence of hypertension differs significantly between the study populations, but is higher in the real-world data than in the other clinical trials. Patients in the real-world data have more pre-existing conditions such as diabetes mellitus and especially peripheral vascular disease. Around 20% of the real-world populations had a history of myocardial infarction (MI), compared with almost 30% of the OCTOBER and around 7% of the OCTIVUS cohort. More patients in the US RWE data had prior coronary artery bypass grafting (CABG) than in MADEIRA and OCTIVUS, while OCTOBER patients had more prior PCI and fewer strokes. The rates of unstable angina were similar in all studies, whereas stable angina was more than three times higher in the OCTOBER than in the MADEIRA population. Non-ST-elevation myocardial infarction (NSTEMI) differs only in the US RWE data, but very clearly.

Implication for research and/or (healthcare) practice: The study populations differ in many characteristics. While the real-world data studies show very similar baseline characteristics, the randomised open-label trials data differ, especially with regard to fewer pre-existing conditions. Regional differences within the different data groups show an increased level of pre-existing conditions in US RWE data compared to MADEIRA data, but this is not constant across all characteristics.