The role of cardiac pacing for prevention of reflex syncope is controversial and not always understood: reflex syncope has two components, 1) vasodilatation - a drop in arterial blood pressure, and 2) cardioinhibition - a drop in heart rate. Pacing can help with cardioinhibition.
Although vasodilation and cardioinhibition coexist in reflex syncope, one mechanism may predominate over the other, including within the same patient, and from one syncope event to another.
The benefit of pacing in this context, in patients who had a positive response to tilt table test has been examined in several randomised trials. In patients who had a positive cardioinhibitory - mostly asystolic - response, the randomised open-label VASIS-PM (1) and SYDIT (2) trials reported results that favored pacing. However, in patients who had cardioinhibitory - mostly non-asystolic - and non-cardioinhibitory response, the double-blind trials VPS2 (3) and SYNPACE (4) showed no significant benefit of pacing suggesting that the previous open-label studies had only demonstrated a placebo effect of pacing.
I- The ISSUE-3 trial
The ISSUE-2 prospective trial (5) argued that patient selection from those with suspected reflex syncope had not been sufficiently focused in on in the above-mentioned previous studies: the hypothesis was that if an implantable loop recorder (ILR) were used to identify intense bradycardia or asystole as the cause of - spontaneous, often atypical without prodrome- faints in older patients, then patients who would stand to benefit from pacing would be properly identified. The ISSUE-3 randomised controlled trial (6), that enrolled 511 patients and involved 29 centers examined the ISSUE-2 observation in a double-blind randomised-controlled trial. It focused on subjects affected by reflex syncope with a relatively high mean age (>40 years, mean 63±13 years), a history of ≥3 recurrent syncope beginning in middle or older age, and frequent injuries probably due to lack of a prodrome. They had the documentation with ILR of a mean asystolic pause of 11±4 s (range 3-44 s). Patients' implantable loop recorder readings demonstrated an asystole of >3s during spontaneous syncope or an asymptomatic pause of >6s. Young patients, who usually have a more prolonged prodrome before loss of consciousness, were not included in the ISSUE population.
Dual-chamber pacemakers were implanted in all such ILR positive patients and they were randomised to 1) pacing with a form of rate hysteresis or 2) no pacing. The results, illustrated in Figure 1, indicate that there is significant benefit of pacing (P<0.04) for these selected patients with a very substantial reduction in syncope recurrence: risk reductions were 32% absolute and 57% relative.
Figure 1. ISSUE-3 results: Time to first recurrence of syncope according to the intention-to-treat analysis. The p value was calculated at the threshold of statistical significance of 0.04.
II - Practical questions answered
Relative to these findings, here are the questions you might have, answered.
Do I have to apply ILR strategy to all patients affected by recurrent reflex syncope >40 year old ?
The ILR strategy is function of pacemaker therapy: implantation of an ILR is reasonable if the severity of the syndrome justifies the implantation of a pacemaker. The ILR will thus document an asystole.
Ultimately, pacing is effective in preventing recurrent reflex syncope however it is not always necessary: the fact that it is effective is not enough to make it necessary. The decision to implant a pacemaker needs to be undertaken while being reminded of the clinical context of syncope: event when it is recurrent, syncope is a benign condition (in terms of mortality), which frequently affects patients that are young. Thus, cardiac pacing is a last resort in highly selected patients affected by severe reflex syncopes. To that effect, the ISSUE-like patients match those defined by ESC guidelines as high-risk or high frequency, i.e. those patients who have very frequent syncope that alters quality of life, or recurrent syncope without or with very short prodromes that expose patients to risk of trauma or syncope that occurs during high risk activity (e.g. driving, machine operation, flying, competitive athletics, etc.)(7).
How many patients with suspected reflex syncope are candidates to ILR and how many patients who receive an ILR will be implanted with a pacemaker ?
Based on findings in ISSUE-3 methodology, we estimated that the patients who met the ISSUE-3 inclusion criteria were 9% of all patients affected by reflex syncope referred for evaluation in a specialised syncope facility (6).
In ISSUE-3, 18% of ILR patients had the documentation of an asystolic pause ≥3 s with syncope or ≥6 s without syncope after a mean observation of 12 months and therefore were eligible for pacing therapy. Since ILR diagnostic yield is a function of the length of observation, this rate will probably increase by prolonging the ILR follow-up. Indeed, in a recent study (8), the diagnostic yield of asystolic events on ILR rose to 40% when the observation period was prolonged to 4 years.
Is it possible to avoid ILR monitoring in patients with documentation of asystolic syncope from tilt table testing ?
We don’t know it is possible to avoid ILR monitoring in patients with documentation of asystolic syncope from tilt table testing. Due to the contrasting results of the randomised trials, the estimated benefit of pacing in cardioinhibitory tilt-positive patients is weak; divergence of opinion exists among experts. Further research is likely to impact on recommendations. However, there is strong evidence that pacing cannot be offered to patients with non-cardioinhibitory tilt-positive response such that further tests (e.g., ILR) are warranted in these patients to document the mechanism of the spontaneous reflex.
Is it possible to identify non-responders?
In ISSUE-3, despite the positive results, 25% of paced patients still had syncope recurrence at 2 years of follow-up. Owing to its sequential design, the study was underpowered to make any subgroup analysis. Future and ongoing studies will investigate whether subgroups of patients benefit more from a pacemaker.However, ISSUE -3, as well as all randomised trials, first considered syncope as the primary outcome of the study. While first event occurrence is optimal for single or rare serious outcomes, e.g., death or hospitalisation, it is not optimal for repetitive relatively benign events such as neurally mediated syncope recurrence. In the case of syncope trials, syncope burden would likely offer a better picture of the clinical benefit of pacemaker therapy. For example, in the ISSUE-2 trial (5), paced patients had only 0.05±0.15 episodes of syncope per year with a relative risk reduction of 87% compared with pre-treatment period.
The main finding of ISSUE-3 is that an overall strategy of using an implantable loop recorder, with the consequent relative certainty regarding mechanism - e.g. predominantly cardioinhibitory- likely contributed to the demonstration of the effectiveness of pacing in preventing syncope recurrence.
Dual-chamber permanent pacing is effective in preventing syncopal relapse in patients ≥40 years with severe recurrent reflex syncopes in whom a long asystolic pause has been documented.