Current guidelines recommend the use of coronary physiology, particularly fractional flow reserve (FFR) and instantaneous wave-free ratio, to guide revascularization in angiographically intermediate coronary stenoses. Hyperemia can be induced by continuous intravenous infusion of adenosine or alternatively by an intracoronary bolus administration of adenosine. Based on a large multicenter registry, we compared the results of intracoronary and intravenous adenosine for FFR measurement. 

A total of 3,378 pressure index measurements of 2,000 consecutive patients were collected in a prospective, large, multicenter registry of 8 interventional cardiology centers in Germany, the Fractional Flow Reserve Fax Registry F(FR²). We compared measurement results, revascularization decisions, and complications of FFR measurements performed with intravenous and intracoronary adenosine administration. Data were compared with the Wilcoxon rank sum test with continuity correction and distributions with the two-sample Kolmogorov-Smirnov test. 

Of the 2,960 FFR measurements collected in the registry, hyperemia was induced by continuously infusing adenosine in 384 (13.0%) measurements and by administering an intracoronary bolus of adenosine in 2,556 (86.4%) measurements. 

FFR values (0.84±0.08 vs. 0.87±0.08, p<0.001) and their distribution (p<0.001) differed significantly between lesions assessed with intravenous adenosine compared to intracoronary adenosine (Figure 1a). However, visually estimated diameter stenosis (75±11% vs. 58±14%, p<0.001) and their distribution (p<0.001) also differed significantly between these lesions (Figure 1b). In a multivariable analysis including mode of adenosine administration, diameter stenosis, sex, age, and site, the mode of adenosine administration was no independent predictor of measured FFR. Adding the information of whether intravenous or intracoronary adenosine was used for FFR measurements resulted in a change-in-estimate of -1.27% for a linear regression model predicting FFR from diameter stenosis. Hence, no confounding effect of the method adenosine was administered on FFR measurement can be detected.

Complications were rare irrespectively of whether hyperemia was induced by intravenous infusion or an intracoronary bolus of adenosine and included prolonged chest pain, third-degree atrioventricular block, coronary artery occlusion, and coronary artery dissection (Table 1). In one patient, FFR measurement using intracoronary adenosine resulted in a dissection of the left main coronary artery and was fatal. 

An intracoronary bolus of adenosine, which already appears to be the preferred method in clinical practice in Germany, seems to be a reliable and safe alternative to an intravenous infusion of adenosine to induce hyperemia for FFR measurement. 


Table 1

Complication	Intravenous Adenosine (n=328)	Intracoronary Adenosine (n=1465)
Prolonged Chest Pain	1 (0.3%)	0 (0.0%)
3rd Degree AV Block / Asystole	1 (0.3%)	21 (1.4%)
Coronary Artery Occlusion	2 (0.6%)	0 (0.0%)
Coronary Artery Dissection	0 (0.0%)	3 (0.2%)

Figure 1