Defibrillation testing : is it really necessary?

 I - What we know

The standardised requirements for cardioverter defibrillator (ICD) implantation, with or without cardiac resynchronisation therapy (CRT), include defibrillation testing (DT), which consists in the induction and termination of ventricular fibrillation (VF). 

This procedure has been set from the early days of ICD therapy in order to test the reliability of the device and to measure the defibrillation threshold. Carrying out an effective DT is considered mandatory to be consistent with the rules of good clinical practice.

However, since the implantation procedure of ICD has been markedly simplified and surgical risk today is very low, defibrillation testing can be considered as the most critical part of the implantation procedure itself.

Risks associated with DT

Although the risk associated with DT is usually low, serious complications may occur as a consequence of this practice. These complications include

• transient ischemic attack or stroke,
• cardiopulmonary arrest due to refractory VF or pulselesselectrical activity
• cardiogenic shock
• embolic events and death

This knowledge comes from small single-center retrospective surveys (1) and from anecdotal experiences. However, in the absence of data of large populations from multicentre registries, the real magnitude of intra-operative complications related to DT is still largely unknown.

Despite the fact that the standardised approach still requires a VF induction test, data coming from real world experience suggest that an increasing number of first implantation procedures are performed without any induction test at all. It seems that physicians may be concerned about practicing DT in selected patients considered at very high clinical risk. For example, in two single-centre populations, Russo et al (2) reported a lack in induction testing in 4.7% and Pires et al (3) in 24% of their patients respectively. Reasons for this lack in induction testing included

• intraoperative hypotension or hemodynamic instability
• known cavity thrombus or previous inadequate anticoagulation therapy
• recent cardiovascular accident
• severe comorbidities and absence of anesthesia support.

II - What we have learned

a) A systematic nation-wide retrospective survey

The AIAC (Associazione Italiana di Aritmologia e Cardiostimolazione) has recently conducted a systematic nation-wide retrospective survey in order to determine how often and for what reasons intra-operative DT was or was not performed, and the complication rate related to induction testing (4). An ad-hoc questionnaire was sent to all 343 centres implanting ICDs listed in the database of Italian ICD Registry of AIAC which virtually represent the totality of Italian implanting centres. The survey was limited to patients undergoing initial ICD implantation during the year 2005.

Among a total of 8820 first ICD/CRTs implanted in Italy during the year 2005, the data of 7857 (89%) implantations (38% of whom CRT) performed in the 229 centres which participated to the survey were analysed. Of these,

• 2356 (30%) were implanted without any induction test (Table 1)
• In 35 (15%) centres the induction test was performed in <25% of the patients
• while in 136 (59%) centres it was performed in >75% of the patients.

At multivariable analysis, performed in a subset of 1206 patients from 107 centres, 

• CRT device (OR 1.82) and primary prevention (OR 1.47) were independent predictors of the decision not to perform DT
• However, all together the clinical variables accounted for only 35% of the total variance whereas the remaining 65% was probably unrelated to clinical factors
• There was a total of 22 (0.4%) life-threatening complications as a consequence of the induction test: 4 deaths (0.07%), 8 cardiopulmonary arrests requiring resuscitation manoeuvres  (0.15%),  6 cardiogenic shocks  (0.11%),  3 strokes  (0.05%), and 1 pulmonary embolism  (0.02%)
• The failure rate of ICD defibrillation, requiring the use of a backup external defibrillator was 2.7% which determined a system revision (i.e additional lead insertion, etc) in 2.3% of patients.

 This nation-wide survey is the largest ever performed and covers 89% of the overall Italian first implantations during the year 2005.

The main result is that, in the real world clinical practice, DT is not performed in 30% of patients, most of these in absence of legitimate reasons, despite the fact that DT is still considered part of the standard procedure of ICD implantation.

There is a wide heterogeneity between centres and more than a quarter of Italian centres do not perform DT in 50% of their patients. These figures are much higher than those expected from the literature (2, 10) and reflect a spontaneous non-conformist opinion of several physicians that overlooks the current recommendations of ICD companies. Given the high numbers involved, this behaviour deserves explanation and actions.

b) There are few reports from small studies concerning intra-operative complications

In litterature there are few reports from small studies concerning intra-operative complications. 

• A report (1) on 440 consecutive single-center ICD implantations showed 0.2% perioperative deaths, 0.5% difficulty in defibrillation with requirements for more than three external shocks, 0.7%  perioperative ischemic attack.
• In another single-center study (2), performed on 835 ICD implantations, there were 3 (0.35%) perioperative deaths (within 30 days from implant). It has been reported that shocks during DT may cause hemodynamic compromise (5), especially in patients with severe heart failure, as are those candidates to CRT. 

Moreover, anaesthesia has a cardiac depressive effect in adjunction of VF induction (6). Patient at implantation could have worse clinical conditions that could improve later with CRT, thus decreasing the risk of complications related to DT; for example a DT delayed up to 2 months after CRT device implant, when patient’s clinical condition improved due to CRT, showed effectiveness without compromising  safety (7).

On the other hand, few data are available regarding the risk of death due to failure of the ICD to interrupt a VF during long-term follow-up. Sudden death in patients with ICD is reported to range approximately from 1.8% to 2.6% during the 1 to 3 years of follow-up (3, 8, 9). Of these, an analysis of the mechanisms of sudden death, with data retrieved from ICD diagnostic memory, showed that only a quarter of the cases could be attributed to shock failure during VF (8).  Therefore, we can assume that the sudden death rate potentially attributable to shock failure ranged form 0.45% to 0.65% during the 1 to 3 years of follow up. This percentage is very similar to the percentage to the intra-operative deaths observed following VF induction during implantation.

In addition, there are no data that specifically demonstrates increased mortality among patients with high DT thresholds at implant and in a recent study (3) success of ICD therapies and sudden-death-free survival were similar in patients who had defibrillation threshold measurement, safety margin testing or no testing.

III - Where we go from there

Is it time to change current standard of performing DT at the time of ICD implantation? The question has been previously raised by authoritative authors (2,3,10,11,12).

There is no evidence-based answer yet. Reasons in favour and contrary to DT are summarised in Table 2. The clinical impact of DT versus no DT will remain unclear until the not negligible introperative complication rate is weighed against the long-term potential benefit of DT.

Until long-term follow-up data regarding the safety and efficacy of ICD implantation in large groups of patients in whom DT is not performed are available, DT should be considered a standard practice. However, data from literature and from the present study support the need to perform such large multicentre studies and the urgency for scientific sub-speciality societies to give precise recommendations.

Conclusions

In the real world practice, DT is not performed in a substantial number of patients, most of these in absence of legitimate reasons. This finding, much higher than that expected from the literature, reflects a spontaneous non-conformist opinion of several physicians that goes beyond the current recommendations. Given the high numbers involved, this behaviour deserves explanations and actions. The results of available literature form the background for future trials aimed to evaluate the clinical benefit of performing DT and underline the urgency for scientific sub-speciality societies to give precise recommendations.

Table 1 : Principal findings

Centres invited to participate : 343
Centres that participated : 229 (67%)
Total number of first implant procedures : 7857
- with intraoperative defibrillation test : 5501 (70%)
- without intraoperative defibrillation test 2356 (30%)
Total number of complications related with defibrillation test : 22 (0.4%)
Death : 4 (0.07%)
Cardiopulmonary arrest requiring resuscitation : 8 (0.15%)
Cardiogenic shock : 6 (0.11%)
Stroke : 3 (0.05%)
Pulmonary embolism : 1 (0.02%)

Table 2 : Advantages and disadvantages of performing DT at the time of implant

Reasons in favour of induction

• Standard practice for ICD implant
• Most device safety studies required DT at implant
• DT allows the choice of corrective  measures at implant in case of high threshold
• DT ensures that the system provides appropriate sensing of VF
• DT may include defribrillation threshold evaluation for better ICD programmin


Reasons in favour of non-induction

• No data specifically demonstrates increased mortality among patients with high DT thresholds
• Quite small probability of a high threshold and a failed implant with current technology
• The nature of defibrillation threshold is probabilistic and repeated shocks below threshold can be effective
• The shocks may cause hemodynamic compromise
• The cardiac depressive effect of anaesthesia in adjunction of VF induction
• In the great majority of patients receiving an ICD, the initial spontaneous life-threatening arrhythmia is VT and  not VF, thus a DT at implant imposing an additional risk which most would not have in their lives
• Patient at implant could have worse clinical conditions that could improve later with CRT, thus decreasing the risk of complications related to DT
• In one retrospective analysys, success of ICD therapies and sudden-death-free survival were similar in patients who had defibrillation threshold testing, safety margin testing, and no testing

The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.