Pacemaker implantation following transcatheter aortic valve implantation: impact on late clinical outcomes and left ventricular function

Selected in Circulation by F. D'Ascenzi



Urena M, Webb JG, Tamburino C, Muñoz-García AJ, Cheema A, Dager AE, Serra V, Amat-Santos I, Barbanti M, Immè S, Alonso Briales JH, Benitez LM, Al Lawati H, Cucalon AM, García Del Blanco B, López J, Dumont E, Delarochellière R, Ribeiro HB, Nombela-Franco L, Philippon F, Rodés-Cabau J.


December 2013


Access Article

My Comment


The need for permanent pacemaker implantation (PPI) is one of the most frequent complications associated with TAVI. Although strong data suggest that right ventricular apical pacing has a potential negative impact, the evidence on clinical impact of PPI following TAVI remains scarce and based on small studies. Urena and colleagues conducted a multicenter study enrolling a large cohort of patients undergoing TAVI (with self-expandable or balloon-expandable valves, SEV or BEV respectively) with the aim to assess the impact of new PPI on late outcomes, on LV function, and on functional status.

Major findings

The study population consisted of 1.556 patients. 239 (15.4%) received a PPI within 30 days following TAVI (25.5% of patients treated with SEV and 7.1% treated with BEV).

  • There were no differences between PPI vs. no PPI groups in 30-day mortality or major complications following TAVI (p>.20 for all). No differences between groups were observed in death or rehospitalization due to heart failure (34.1% vs. 31.8%, HR: 1.00, 95% CI: 0.77-1.30, p=.98);
  • A lower rate of unexpected (sudden or unknown) death among patients who had a PPI within 30 days following TAVI was observed (HR: 0.31, 95% CI: 0.11-0.85, p=.023);
  • Pre-existing paroxysmal/permanent atrial fibrillation (HR: 1.76, 95% CI: 1.09-2.86, p=.021) and the lack of 30-day-PPI (HR: 3.22, 95% CI: 1.16-9.09, p=.024) were the independent predictors of unexpected death. The occurrence of new-onset persistent left bundle branch block (HR: 2.77, 95% CI: 1.09-7.07, p=.033) and a lower LV ejection fraction (EF) at baseline (5.25 for each decrease in 5%, 95% CI: 5.15-5.45, p=.001) were the independent predictors of sudden cardiac death;
  • At resting ECG, pace rhythm was observed in 66.9% of patients and it was more frequent in patients who had received SEV (72.8% vs. 46.7% in patients with BEV, p=.007);
  • LVEF significantly increased in overall population at 6-to 12-month follow-up (p<.001). Whereas LVEF increased over time in patients with no PPI, LVEF decreased at follow-up in those patients who had PPI following TAVI (p=.017), without differences between SEV and BEV groups (p=.668). A poorer evolution of LVEF in patients who needed PPI was observed in those patients who received a dual-chamber (vs. single-chamber) PPI (p=.043; p=.023 after adjusting for the presence of atrial fibrillation);
  • LVEF at baseline and the need for PPI within 30 days were the only independent predictors of LVEF decrease over time (estimated coefficient: -3.44 CI: -4.11 to -2.26, 95% CI: -4.07 to -0.44, p=.013, R2: 0.121, respectively);
  • A marked improvement in NYHA functional class was found in patients with and withouth 30-day PPI (p<.001) without differences in NYHA class changes between PPI and no PPI groups (p=.672).

My comments

The study by Urena and colleagues demonstrated in a large cohort of patients undergoing TAVI that the need for PPI periprocedurally had no impact on overall and cardiovascular death, on functional status, or on heart failure decompensation. Interestingly, the authors demonstrated that 30-day PPI was a protective factor for the occurrence of unexpected death. This finding, as suggested by authors, indirectly raises questions about the most appropriate management of new conduction disturbances that do not meet the criteria for PPI following TAVI. The authors demonstrated that PPI had a negative impact on LVEF and that this negative effect was more pronounced in those patients receiving a dual-chamber pacemaker. However, a lack of association between the negative impact of PPI on LVEF and the NYHA functional class is somewhat comfortable, even if these findings need to be confirmed in studies with a longer follow-up time.

Another stimulating key-point of the present study is the observation that, although pacing-dependency and right ventricular pacing burden were not systematically evaluated, more than one third of TAVI patients with PPI did not exhibit pacing activity during follow up and pace rhythm was more frequently observed in SEV vs. BEV patients (72.8% vs. 46.7%, p=.007). These findings suggest that a significant proportion of atrio-ventricular blocks resolve over time and this phenomenon is more pronounced in patients undergone TAVI with BEV.

Taking together, the findings of the present study suggest that further efforts are needed to tailor the specific therapy for each TAVI patient experiencing new conduction disturbances.

No comments yet!