Feasibility, safety and outcomes of contrast-free cardiac resynchronisation therapy device implantation: a retrospective cohort comparison

Left ventricular lead (LVL) implantation can be challenging in Cardiac Resynchronization Therapy (CRT). Therefore, contrast is commonly used to visualize the coronary venous system. However, contrast use is correlated with substantial risks such as contrast-induced nephropathy and anaphylactic reaction. Case studies suggest feasibility of LVL implantation without coronary sinus (CS) venography. We aimed to compare the procedure characteristics, outcome, and safety of contrast-free and contrast-aided LVL implantation. 346 LVL implantations performed between 2017 and 2019 were analysed. 167 were attempted without contrast (Intervention Group: IG), and 179 were fully performed with contrast (Control Group: CG). The data was collected and analysed retrospectively from a single centre. These two groups were compared in an Intention-To-Treat (ITT) analysis. An As-Treated (AT) analysis was performed to compare procedures fully performed without contrast (NCG), crossover procedures (COG) and the control group (CG). The intervention duration, fluoroscopy duration, and radiation dose were evaluated as procedure characteristics. The primary outcome was procedural success. The secondary outcomes were LVL position, LVL threshold value, QRS duration and LVEF evolution. Perioperative complications and lead revisions were surrogates for the intervention’s safety. The overall LVL implantation success rate was 97.7%. Of the contrast-free procedures 62.9% were successful, the remainder required crossover to contrast use compared to 96.6% with contrast use. The ITT and AT analyses yielded similar results. Contrast-free interventions were associated with shorter procedure (100.8 ± 41.0 vs. 131.1 ± 50.0 min, p < 0.01), less fluoroscopy (15.7 ± 11.2 vs. 26.0 ± 17.5 min, p < 0.01) and smaller radiation doses, (475.3 ± 422.7 vs. 897.3 ± 779.1 cGy.cm2, p < 0.01). The intraoperative LVL threshold values were equivalent (0.9 ± 0.6 in the NCG vs. 1.0 ± 0.6 V in the CG, TOST-p < 0.01), and contrast use did not significantly influence the follow-up LVL threshold (1.1 ± 0.5 vs. 1.2 ± 0.9 V (p = 0.62), QRS shortening (10.5 ± 28.3 vs. −13.2 ± 26.3ms, p = 0.16) or LVEF increase − (2.3 ± 8% vs. 4.1 ± 8.4%, p = 0.40). More CS dissections were observed during the contrast-aided procedures (5% in the CG versus 3.2% in the COG versus 0% in the NCG, p = 0.03). However, the perioperative complication rates (pericardial effusion, ICU admission, death, mechanically induced arrhythmia, and pneumothorax) did not significantly differ (p > 0.05 for each). In contrast-free LVL implantation failure cases, intraoperatively crossing over from the NCG to CG did not result in longer procedures (131.8 ± 46.8 in the COG vs. 131.1 ± 50.0 min in the CG, p = 0.90). Although more adverse events were observed in the CG, the difference was not statistically significant (6.5% in the COG vs. 12.3% in the CG, p = 0.20). Contrast-free LVL implantation was associated with shorter intervention durations, less fluoroscopy and lower radiation doses. Fewer intraoperative complications were observed without contrast, and the postoperative complication rate was similar to contrast-aided procedures. This technique was successful in almost 2/3 of the cases. Where a crossover to contrast use was required, the initial contrast-free approach did not result in a statistically significant increase in clinically relevant adverse events compared with fully contrast-aided procedures. It is therefore reasonable to consider an initial contrast-free approach for CRT implantation. Central Illustration: Feasibility, procedural characteristics, and lead positioning of contrast-free LV lead implantation.