“Are CRT upgrade procedures more complex and associated with more complications than de novo CRT implantations?” A single centre experience

Cardiac resynchronisation therapy (CRT) is most effective in heart failure patients with left bundle branch block as it specifically aims to resynchronise the delayed electrical activation of the left ventricle [1‐2]. However, CRT is indicated in a much broader spectrum of patients with heart failure [3]. A large European survey among 141 hospitals demonstrated that approximately 25 % of CRT implantations are upgrade procedures [4]. Patients with an existing cardiac implantable electronic device (CIED) have a class I indication for upgrading to CRT when they experience a high percentage of ventricular pacing and remain in New York Heart Association (NYHA) class III to ambulatory IV with a left ventricular ejection fraction (LVEF) ≤ 35 %. Other reasons for upgrading to CRT include (i) worsening of heart failure (class) in patients with pre-existing wide QRS complexes, (ii) electrical remodelling due to heart failure or antiarrhythmic drug therapy with widening of the QRS complex, and (iii) evolving CRT guidelines.

Previous literature has shown that cardiac device replacement or upgrade procedures are associated with more complications and are more complex than de novo implantations due to the possibility of venous thrombosis and the risk of damaging or extracting old leads [5, 6]. The fear of a complicated procedure might prevent upgrading to CRT in patients with a clear indication. Moreover, reluctance about the need for an upgrade procedure in the future might make clinicians more willing to implant a de novo CRT device in heart failure patients with only a class IIb indication for CRT but a clear indication for a conventional (single or dual chamber) device.

We aimed to evaluate procedural data and periprocedural and long-term complication rates of upgrade procedures compared with de novo CRT implantations in our university medical centre.

Our single-centre experience did not find any significant differences in either procedure length, radiation duration, total hospitalisation, or in periprocedural and long-term device-related complication rates when comparing upgrade procedures with de novo CRT implantations.

Previously, the RAFT study showed that CRT-D reduced rates of death and hospitalisation for heart failure when compared with ICD therapy only. However, device-related hospitalisation was significantly higher for the CRT-D group compared with the ICD-only group [7]. The increased risk of complications of a triple chamber device compared with conventional single and dual chamber pacemakers or ICDs was also highlighted by a large nationwide cohort of 5918 Danish patients who underwent a CIED implantation, in which patients receiving a CRT device showed the highest complication rate [8]. Therefore, the decision to implant a triple chamber device needs to be carefully considered.

For patients with a class IIb indication for CRT but a clear indication for a conventional ICD the alternative to a de novo CRT implantation is to await the potential need for an upgrade to CRT. However, as was shown by Poole et al., CRT upgrade is a high-risk procedure as well, with an 18.7 % risk of any complication during a follow-up of six months [5]. The fear of complications associated with an upgrade procedure might prevent the upgrade to CRT in patients with a clear indication. This was suggested by the results of Scott et al. who performed a single-centre retrospective observational study evaluating the implantation of ICDs for a period of 5 years. In this period 549 new ICDs were implanted of which 73 % single or dual chamber ICDs and 27 % CRT-D. When they applied alternative indication criteria (LVEF ≤ 30 %, QRS ≥ 130 ms, NYHA I–IV), 42.6 % of the ICD recipients met the criteria for CRT at initial implant. However, the upgrade rate to CRT at 5 years was only 5.1 % [9].

We showed a comparable complication rate between upgrade and de novo CRT implantation procedures. This is in line with the results of the European Survey which showed that there were no significant differences in complication rates when comparing 692 upgrades with 1675 de novo CRT procedures at 141 centres in Europe [4]. However, this was a randomised trial and such trials commonly report less complications than in a real-life setting due to more strict patient selection and more experienced operators. On the contrary, the previously mentioned Danish population-based cohort of CIED implantations, including 445 CRT-D implantations, showed that CRT upgrade procedures or lead revisions were associated with more complications than CRT de novo implantations [8]. As both our study and the Danish cohort study reflect the results of a real-life setting the differences are not likely to be based on patient selection. In our retrospective study, patient selection was equal for both de novo and upgrade procedures in that it followed standard clinical care in which the benefit of CRT is balanced against the risk of complications. It might be that the operators in our tertiary centre were more experienced in implanting CRT devices. Although most procedures in the Danish cohort were also performed by operators with an annual volume of at least 75 ICD or pacemaker implantations, it is not clear whether the experience in implanting a CRT device was as high as in our centre [8]. Another point of concern for many clinicians is the fear of a more complex procedure when a single or dual chamber pacemaker or ICD needs to be upgraded, compared with a de novo implantation of a CRT device, due to the possibility of damaging or extracting old leads. Four upgrade patients experienced lead dysfunction after the procedure, which is comparable with numbers in previous literature [5, 10]. These were three old leads (two right atrial leads and one right ventricular lead) and one newly placed right atrial lead. The dysfunction of the old leads was seen 2–12 months after upgrade procedure. Moreover, this was done due to damage to the old lead during the procedure in only one of 16 upgrade patients who received a completely new system. Therefore, our data seem to indicate that when an upgrade procedure is performed by an experienced operator the risk of damaging old leads is small. Furthermore, we found no significant difference in procedure duration comparing upgrade procedures with de novo implantations, but there was a slight trend towards an even shorter procedure length for upgrade procedures. This suggests comparable complexity as one would expect the upgrade procedure to require a smaller amount of time, as in most upgrades less leads are added than implanted during a de novo implantation. Complicating factors such as subclavian vein thrombosis only slightly increased the median procedure duration compared with standard procedures (159 (95) minutes versus 139 (69) minutes). Moreover, in our upgrade group the procedure success was 98.5 % and did not significantly differ for the randomly selected de novo CRT implantation procedures. A meta-analysis of 9082 patients in 25 CRT trials showed an implantation success rate of 94.4 % indicating that the success rate of upgrade procedures was most definitely not lower compared with de novo procedures [11].

Decisions on device type (conventional or triple chamber pacemaker or ICD) can be difficult in patients with an indication for a conventional pacemaker or ICD without a clear class I or IIa indication for CRT. Based on fulfilment of indication criteria for CRT the chance of benefit from CRT in these patients is low and the complication rate increases with the number of leads implanted. However, when a conventional pacemaker or ICD is opted for the need for upgrading can eventually arise. This means the patient is exposed to two procedures, where it is generally believed that an upgrade procedure is more complex and associated with a higher complication rate. This single-centre experience of a Dutch university hospital shows neither a higher complexity nor a higher complication rate of upgrade procedures to CRT compared with de novo CRT implantations. Therefore, CRT upgrade procedures should be performed when indicated without fear for a higher complication rate. However, as upgrading means an additional procedure is needed, future research should focus on identifying patient characteristics that predict the necessity for upgrading to CRT, especially for those patients in need of upgrading within the life expectancy of the initial device.

Due to the retrospective design of the study only complications properly documented in the electronic patient file were identified and not all baseline data were available for all patients. However, the advantage of the retrospective design is that it reflects a real-life setting of these device implantations. Furthermore, although none of the baseline characteristics showed a confounding effect on the risk of periprocedural complications, it is possible that some data that were not collected confounded the results, for instance steroid use. Finally, as this study shows the results of a single-centre experience, the results can only be applicable for a comparable centre.

In a device implantation centre with experienced operators, CRT upgrade procedures are not associated with more complications nor do they seem more complex compared with de novo CRT implantations.