RCT within registry as a new way for medical device industry to generate evidence

This week a new study was published in the NEJM, which utilized established patient registry, as a framework for a randomized controlled trial.

Götberg et al. conducted a multicenter, randomized, controlled, open-label clinical trial using the Swedish Coronary Angiography and Angioplasty Registry for enrollment. A total of 2037 participants with stable angina or an acute coronary syndrome who had an indication for physiologically guided assessment of coronary-artery stenosis were randomly assigned to undergo revascularization guided by either iFR or FFR. The primary end point was the rate of a composite of death from any cause, nonfatal myocardial infarction, or unplanned revascularization within 12 months after the procedure.

Study showed that an iFR-guided revascularization strategy was noninferior to an FFR-guided revascularization strategy with respect to the rate of major adverse cardiac events at 12 months. Study also showed that iFR might be less expensive and more convenient for patients. 

The study utilized patient registry to perform RCT. This is a very interesting way of generating evidence with multiple benefits: 

• cost of procedure is covered by the health care system
• there is no need to set up infrastructure for the study, as data are collected within registry’s framework
• results are much closer to the “real world” vs conditions of typical RCTs
• ability for rapid enrollment of patients
• study is powered sufficiently for desired outcomes
• good ability to control for confounders
• ability to identify even rare events

One of the potential drawbacks for commercial sponsor of such studies is the lack of the control over the protocol and execution of the study. General weaknesses include potential variability in definitions of variables and outcomes and sub-optimal data quality of the collected data. 

With all potential limitations, this is really an elegant and effective way of generating evidence. In the world, when decision makers require more and more evidence and industry cannot cope with financial pressure and does not effectively respond to the demand, registry-based studies (when suitable) can be an effective solution.

This is not the first example of such approach and even not the first example of using SWEDEHEART in such a way. The same registry in Sweden was used before to run TASTE trial (Thrombus Aspiration during ST-Segment Elevation Myocardial Infarction), which did not find any effect of aspiration thrombectomy in addition to PCI vs PCI along for STEMI. 

Currently, there are number of studies ongoing with use of SWEDEHEART registry including:

• Ffr-gUidance for compLete Non-cuLprit REVASCularization (FULL REVASC)
• Determination of the Role of Oxygen in Suspected Acute Myocardial Infarction by Biomarkers (DETO2X-bio)

Those interested can read articles explaining the concept:

• James et al: Registry-based randomized clinical trials—a new clinical trial paradigm