Early prediction of life-threatening arrhythmias in non-ST elevation myocardial infarction – does it change clinical practice?

The incidence of life-threatening ventricular arrhythmias (LT-VAs) among patients with non-ST elevation myocardial infarction (NSTEMI) is substantially lower (1.5–2.1%) than that among patients with ST-segment elevation myocardial infarction (STEMI) (10%).1–3 In patients with NSTEMI, data on ventricular arrhythmias are scant and limited to sub-analysis of large multicentre trials or to highly selected patient groups.2 In patients with STEMI, LT-VAs usually occur immediately after the onset of acute myocardial ischaemia or reperfusion. In contrast, LT-VAs are infrequent after NSTEMI as compared to STEMI but may occur within or after 48 h.3 LT-VAs are associated with significantly increased mortality in NSTEMI.2 Some observational studies reported that LT-VAs in the early setting of an acute coronary syndrome (ACS) might be due to reversible ischaemia. These patients seem to be at low risk for sudden death during follow-up if they are completely revascularised.4,5 However, several trials have shown that implantation of a cardioverter defibrillators in the first 40 days after an acute myocardial infarction (MI) which resulted in low left ventricular ejection fraction (LV-EF) did not improve survival.6,7 Thus, optimal therapy for patients with ventricular arrhythmias due to NSTEMI is uncertain, and strategies to improve survival are needed.2 This prospective study included 1325 consecutive patients presenting with NSTEMI in a single centre and assessed predictors of in-hospital spontaneous LT-VAs and all-cause mortality.1 In contrast to previous studies investigating this issue, the present analysis did not include LT-VAs occurring during coronary intervention. Periprocedural LT-VAs show different pathophysiologic mechanisms and differ in prognostic significance from spontaneous arrhythmias. Analyses were based on 21 (1.5%) LT-VAs and 62 (4.7%) deaths, one due to documented arrhythmia. Of the patients who survived LT-VAs, 35% subsequently died of heart failure. A Global Registry of Acute Coronary Event (GRACE) score >140 and LV-EF <35% were independent predictors of LT-VA and in-hospital all-cause mortality. The cumulative probability of in-hospital LT-VA and death was zero in patients with a GRACE score ≤140 and a LV-EF ≥35%. The message is clear but this study has several limitations including the small sample size, the single centre setting, and the low number of events. The long duration of hospitalisation (median of eight days; only 4% of patients were discharged within five days) raises the question about the applicability of the findings of this study in contemporary clinical practice. Thus, these results should be considered as generating a hypothesis and a larger study is needed for validation before definite conclusions can be drawn. Patients with STEMI receiving thrombolytic therapy have a substantially higher incidence of LT-VAs of up to 20% and a greater risk of adverse short- and long-term outcomes with these arrhythmias. In contrast, the occurrence of LT-VAs during primary angioplasty in patients with STEMI do not have an impact on the procedural success or risk of adverse in-hospital and long-term outcomes. However, LT-VAs during primary angioplasty are associated with a greater length of stay.8,9 Periprocedural LT-VAs show different pathophysiologic mechanisms compared to spontaneous arrhythmias. During primary angioplasty such arrhythmias are mainly driven by mechanisms of ischaemia and reperfusion including rapid changes in the metabolic cellular milieu that may result in regional electrophysiologic instability.10 Several clinical and angiographic variables have been identified as predictors of LT-VAs in the cardiac catheterisation laboratory during primary percutaneous coronary intervention (PCI) including history of current smoking, time from symptom onset to emergency room arrival, lack of beta-blocker therapy at admission, right coronary artery (RCA)-related infarct, and initial thrombolysis in myocardial infarction (TIMI) flow grade 0 of the infarct-related artery.10 In contrast, LT-VA in NSTEMI patients outside the cardiac catheterisation laboratory are infrequent (1.5%), are likely to occur after 48 h (about one-quarter of events), and are associated with significantly increased mortality in NSTEMI.1–3 This may suggest that other mechanisms beyond ischaemia could contribute to the pathogenesis of such arrhythmias. However, in the study by Zorzi and colleagues1 only 15 of 21 patients (71.4%) with LT-VAs had invasive management (73.3% within 24 h). Twelve patients with LT-VAs underwent coronary revascularisation by PCI (57%) and only a minority of patients with multivessel disease had complete revascularisation. Thus, ongoing ischaemia might have contributed substantially to the occurrence of LT-VAs outside the the cardiac catheterisation laboratory and beyond 48 h. Nevertheless, a GRACE score >140 and LV-EF <35% were the strongest predictors of both, LT-VAs and in-hospital death. About 25% of patients with both risk factors died during hospitalisation, while all patients with GRACE score <140 and LV-EF >35% survived until discharge. Current guidelines suggest that patients treated with early revascularisation are at low risk for developing LT-VAs, with 80% occurring during the first 12 h after onset of symptoms.5,11 Accordingly, routine monitoring of the patients beyond 24–48 h is not warranted. Patients with NSTEMI should be hospitalised for at least 24 h after successful stenting of the culprit lesion.5 Accordingly, in many institutions, telemetry and ECG-monitoring are focused on the initial 48 h of hospitalisation. Based on the data of the present study with 24% of life-threatening arrhythmic events occurring more than 48 h and one out of 21 events occurred more than five days after admission, the authors suggest that telemetry and monitoring should be continued beyond 48 h. However, only one-half of the patients with LT-VAs had coronary revascularisation and only a minority of patients with multivessel disease had complete revascularisation. Thus, such patients with a additional high risk profile (as expressed by high GRACE risk score and low LV-EF) might remain at high risk of LT-VAs even beyond 48 h of hospitalisation. Therefore, in most patients with NSTEMI successfully treated with early revascularisation, routine monitoring beyond 24–48 h is not warranted. However, high-risknon-ST-elevation acute coronary syndrome (NSTE-ACS) patients including those without successful complete revascularisation remain at high risk even beyond 48 h. The study by Zorzi and colleagues1 does not provide information about treatment of patients including use of angiotensin converting enzyme (ACE) inhibitors, mineralocorticoid-receptor-antagonists, ß-blockers, and diuretics (associated with hypokalaemia and arrhythmias). Additionally, prevalence of hypokalaemia was double in patients with LT-VA (28.6% vs 14.0%). The significance of this finding remains unclear. The best management strategy (including monitoring and treatment) for those patients remains uncertain and novel strategies including complete revascularisation, devices (e.g. wearable automatic defibrillator vest, subcutaneous implantable defibrillator) and drugs are awaited eagerly and should be tested in clinical trials. LT-VAs in NSTEMI patients outside the cardiac catheterisation laboratory are associated with significantly increased mortality in NSTEMI patients.2,3 In the study by Zorzi and colleagues1 seven of the 20 patients (35%) who survived the LT-VAs subsequently died of heart failure. Additionally, high-risk NSTEMI patients characterised by a high GRACE score (>140) and low LV-EF (<35%) are not only at high risk of LT-VAs but also at very high risk of early death. A longer follow-up beyond hospitalisation would be needed to fully determine the independent effect of occurrence of LT-VAs on outcome in these high-risk patients. However, the data presented by Zorzi and colleagues1 clearly elucidated that LT-VAs influence outcome for such high-risk NSTE-ACS patients. This single-centre study demonstrates that risk stratification at admission of a consecutive series of NSTEMI patients undergoing either invasive or conservative treatment strategies can accurately identify a subset of patients at high risk of LT-VAs or death. The tools used for risk stratification are simple and could easily be included in daily practice. If the results can be validated in a large prospective study, the combination of the GRACE score and LV-EF can be recommended for prediction of risk for LT-VAs (beyond prediction of recurrent MI and death) (Figure 1). Early prediction of in-hospital life-threatening ventricular arrhythmias in patients with non-ST elevation acute myocardial infarction (NSTEMI). GRACE: Global Registry of Acute Coronary Event; LV-EF: left ventricular ejection fraction If the results can be validated in a large randomised prospective study, the combination of the GRACE score and LV-EF can be recommended for identification of high risk patients including prediction of risk for LT-VAs. In the absence of knowledge about the best management strategy (including monitoring and treatment) for those patients, novel strategies including revascularisation, devices and drugs should be tested in clinical trials. Until clear evidence is available, we should manage patients according to current guidelines including early revascularisation, optimal medical therapy and appropriate device utilisation. In selected patients, a prolongation of the monitoring period beyond 24–48 h (until appropriate management strategies are applied) would be reasonable. Finally, the good news of this study: LT-VAs occurred in only 0.2% and there were no in-hospital deaths in NSTEMI patients when the GRACE score was below 140 and LV-EF was above 35%. One could argue that these patients do not need admission to intensive or coronary care units, and simple electrocardiogram monitoring may be sufficient. However, as mentioned above, confirmation of these results is needed to recommend this strategy for clinical practice. None declared. This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.