Arturo J Martí-Carvajal, Daniel Simancas-Racines, Vidhu Anand, Shrikant I Bangdiwala

Coronary artery disease is a major public health problem affecting both developed and developing countries. Acute coronary syndromes include unstable angina and myocardial infarction with or without ST-segment elevation (electrocardiogram sector is higher than baseline). Ventricular arrhythmia after myocardial infarction is associated with high risk of mortality. The evidence is out of date, and considerable uncertainty remains about the effects of prophylactic use of lidocaine on all-cause mortality, in particular, in patients with suspected myocardial infarction. To determine the clinical effectiveness and safety of prophylactic lidocaine in preventing death among people with myocardial infarction. We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (2015, Issue 3), MEDLINE Ovid (1946 to 13 April 2015), EMBASE (1947 to 13 April 2015) and Latin American Caribbean Health Sciences Literature (LILACS) (1986 to 13 April 2015). We also searched Web of Science (1970 to 13 April 2013) and handsearched the reference lists of included papers. We applied no language restriction in the search. We included randomised controlled trials assessing the effects of prophylactic lidocaine for myocardial infarction. We considered all-cause mortality, cardiac mortality and overall survival at 30 days after myocardial infarction as primary outcomes. We performed study selection, risk of bias assessment and data extraction in duplicate. We estimated risk ratios (RRs) for dichotomous outcomes and measured statistical heterogeneity using I(2). We used a random-effects model and conducted trial sequential analysis. We identified 37 randomised controlled trials involving 11,948 participants. These trials compared lidocaine versus placebo or no intervention, disopyramide, mexiletine, tocainide, propafenone, amiodarone, dimethylammonium chloride, aprindine and pirmenol. Overall, trials were underpowered and had high risk of bias. Ninety-seven per cent of trials (36/37) were conducted without an a priori sample size estimation. Ten trials were sponsored by the pharmaceutical industry. Trials were conducted in 17 countries, and intravenous intervention was the most frequent route of administration.In trials involving participants with proven or non-proven acute myocardial infarction, lidocaine versus placebo or no intervention showed no significant differences regarding all-cause mortality (213/5879 (3.62%) vs 199/5848 (3.40%); RR 1.02, 95% CI 0.82 to 1.27; participants = 11727; studies = 18; I(2) = 15%); low-quality evidence), cardiac mortality (69/4184 (1.65%) vs 62/4093 (1.51%); RR 1.03, 95% CI 0.70 to 1.50; participants = 8277; studies = 12; I(2) = 12%; low-quality evidence) and prophylaxis of ventricular fibrillation (76/5128 (1.48%) vs 103/4987 (2.01%); RR 0.78, 95% CI 0.55 to 1.12; participants = 10115; studies = 16; I(2) = 18%; low-quality evidence). In terms of sinus bradycardia, lidocaine effect is imprecise compared with effects of placebo or no intervention (55/1346 (4.08%) vs 49/1203 (4.07%); RR 1.09, 95% CI 0.66 to 1.80; participants = 2549; studies = 8; I(2) = 21%; very low-quality evidence). In trials involving only participants with proven acute myocardial infarction, lidocaine versus placebo or no intervention showed no significant differences in all-cause mortality (148/2747 (5.39%) vs 135/2506 (5.39%); RR 1.01, 95% CI 0.79 to 1.30; participants = 5253; studies = 16; I(2) = 9%; low-quality evidence). No significant differences were noted between lidocaine and any other antiarrhythmic drug in terms of all-cause mortality and ventricular fibrillation. Data on overall survival 30 days after myocardial infarction were not reported. Lidocaine compared with placebo or no intervention increased risk of asystole (35/3393 (1.03%) vs 14/3443 (0.41%); RR 2.32, 95% CI 1.26 to 4.26; participants = 6826; studies = 4; I(2) = 0%; very low-quality evidence) and dizziness/drowsiness (74/1259 (5.88%) vs 16/1274 (1.26%); RR 3.85, 95% CI 2.29 to 6.47; participants = 2533; studies = 6; I(2) = 0%; low-quality evidence). Overall, safety data were poorly reported and adverse events may have been underestimated. Trial sequential analyses suggest that additional trials may not be needed for reliable conclusions to be drawn regarding these outcomes. This Cochrane review found evidence of low quality to suggest that prophylactic lidocaine has very little or no effect on mortality or ventricular fibrillation in people with acute myocardial infarction. The safety profile is unclear. This conclusion is based on randomised controlled trials with high risk of bias. However (disregarding the risk of bias), trial sequential analysis suggests that additional trials may not be needed to disprove an intervention effect of 20% relative risk reduction. Smaller risk reductions might require additional higher trials.