Intravenous magnesium for acute myocardial infarction

 

Abstract

Background

Mortality and morbidity from acute myocardial infarction (AMI) remain high. Intravenous magnesium started early after the onset of AMI is thought to be a promising adjuvant treatment. Conflicting results from earlier trials and meta-analyses warrant a systematic review of available evidence.

Objectives

To examine the effect of intravenous magnesium versus placebo on early mortality and morbidity.

Search methods

We searched CENTRAL (The Cochrane Library Issue 3, 2006), MEDLINE (January 1966 to June 2006) and EMBASE (January 1980 to June 2006), and the Chinese Biomedical Disk (CBM disk) (January 1978 to June 2006). Some core Chinese medical journals relevant to the cardiovascular field were hand searched from their starting date to the first-half year of 2006.

Selection criteria

All randomized controlled trials that compared intravenous magnesium with placebo in the presence or absence of fibrinolytic therapy in addition to routine treatment were eligible if they reported mortality and morbidity within 35 days of AMI onset.

Data collection and analysis

Two reviewers independently assessed the trial quality and extracted data using a standard form. Odds ratio (OR) were used to pool the effect if appropriate. Where heterogeneity of effects was found, clinical and methodological sources of this were explored.

Main results

For early mortality where there was evidence of heterogeneity, a fixed-effect meta-analysis showed no difference between magensium and placebo groups (OR 0.99, 95%CI 0.94 to 1.04), while a random-effects meta-analysis showed a significant reduction comparing magnesium with placebo (OR 0.66, 95% CI 0.53 to 0.82). Stratification by timing of treatment (< 6 hrs, 6+ hrs) reduced heterogeneity, and in both fixed-effect and random-effects models no significant effect of magnesium was found. In stratified analyses, early mortality was reduced for patients not treated with thrombolysis (OR=0.73, 95% CI 0.56 to 0.94 by random-effects model) and for those treated with less than 75 mmol of magnesium (OR=0.59, 95% CI 0.49 to 0.70) in the magnesium compared with placebo groups.

Meta-analysis for the secondary outcomes where there was no evidence of heterogeneity showed reductions in the odds of ventricular fibrillation (OR=0.88, 95% CI 0.81 to 0.96), but increases in the odds of profound hypotension (OR=1.13, 95% CI 1.09 to 1.19) and bradycardia (OR=1.49, 95% CI 1.26 to 1.77) comparing magnesium with placebo. No difference was observed for heart block (OR=1.05, 95% CI 0.97-1.14). For those outcomes where there was evidence of heterogeneity, meta-analysis with both fixed-effect and random-effects models showed that magnesium could decrease ventricular tachycardia (OR=0.45, 95% CI 0.31 to 0.66 by fixed-effect model; OR=0.40, 95% CI 0.19 to 0.84 by random-effects model) and severe arrhythmia needing treatment or Lown 2-5 (OR=0.72, 95% CI 0.60 to 0.85 by fixed-effect model; OR=0.51, 95% CI 0.33 to 0.79 by random-effects model) compared with placebo. There was no difference on the effect of cardiogenic shock between the two groups.

Authors' conclusions

Owing to the likelihood of publication bias and marked heterogeneity of treatment effects, it is essential that the findings are interpreted cautiously. From the evidence reviewed here, we consider that: (1) it is unlikely that magnesium is beneficial in reducing mortality both in patients treated early and in patients treated late, and in patients already receiving thrombolytic therapy; (2) it is unlikely that magnesium will reduce mortality when used at high dose (>=75 mmol); (3) magnesium treatment may reduce the incidence of ventricular fibrillation, ventricular tachycardia, severe arrhythmia needing treatment or Lown 2-5, but it may increase the incidence of profound hypotension, bradycardia and flushing; and (4) the areas of uncertainty regarding the effect of magnesium on mortality remain the effect of low dose treatment (< 75 mmol) and in patients not treated with thrombolysis.

 

Plain language summary

Intravenous magnesium for acute myocardial infarction

In some countries, intravenous (IV) magnesium is administered to heart attack patients in order to limit damage to the heart muscle, prevent serious arrhythmias and reduce the risk of death. Several small trials appeared to support the practice. But the authors of this review found that other trials went unpublished once they produced unfavorable results. A controversy erupted in 1995, when a large well-designed trial with 58,050 participants did not demonstrate any beneficial effect to IV magnesium, contradicting earlier meta-analyses of the smaller trials. This review includes 26 clinical trials that had randomly assigned heart attack patients to receive either IV magnesium or an inactive substance (placebo). Their results were mixed: IV magnesium reduced the incidence of serious arrhythmias, but this treatment also increased the incidence of profound hypotension, bradycardia and flushing. However, any apparent beneficial effects of magnesium may simply reflect various biases in these trials. Additionally, there was a lack of uniformity in these trials in terms of dosage and the timing of the IV magnesium regimen, which in some trials also included anti-clotting drugs. The evidence produced by this review does not support continued use of IV magnesium. Other effective treatments (aspirin, beta-blockers) should be used to treat heart attack.

 

摘要

背景

靜脈注射鎂離子於急性心肌梗塞

急性心肌梗塞後(awte myocardial mfarction, AMI)之死亡率及罹病率仍然很高，在急性心肌梗塞發生的初期給予靜注鎂離子似乎是一個有效的輔助療法，但早期研究尚未獲得一致的共識，根據現有資料來作系統性的後設分析是有其必要性的。

目標

實驗靜注鎂離子與安慰劑對早期死亡率及罹病率有何差異。

搜尋策略

我們搜尋了CENTRAL (The Cochrane Library Issue 3, 2006)，MEDLINE (1966年1月至2006年6月) and EMBASE (1980年1月至2006年6月)，和Chinese Biomedical Disk (CBM disk)(1978 年1月至2006年6月)資料庫，也針對一些和心血管領域相關的核心中文期刊，自發刊日至西元2006前半進行人工搜尋相關文章。

選擇標準

所有比較靜注鎂離子與安慰劑效果的隨機對照研究，在標準治療外不論有無使用血栓溶解劑，只要有紀錄其急性心肌梗塞後35天內的死亡率及罹病率即可被接受。

資料收集與分析

由兩位獨立檢閱者評定試驗品質，並用標準方式篩選資料， 如果適當的，治療效果將由勝算比(Odds rotio, OR)表示，如果發現異質性結果，其臨床及方法學將被提出來探討。

主要結論

在早期死亡率方面有不一致的結果， 一個固定作用的統合分析(fixedeffect metaanalysis)顯示在靜注鎂離子與安慰劑兩組沒有差異(OR0.99， 95%CI為0.94 – 1.04)，而另一個隨機作用的統合分析(randomeffects metaanalysis)卻顯示了在靜注鎂離子這組有顯著的下降(OR 0.66，95%CI為0.53 – 0.82)；根據治療時機去作分層分析(<6小時， >6小時)則發現不論在固定或隨機的後設分析，靜注鎂離子皆無明顯效果。 在層次的分析中， 和安慰劑一組比較， 靜注鎂離子的病患如果沒接受血栓溶解劑(OR 0.73， 95%CI為0.56 – 0.94， 隨機作用模型)， 或是接受少於75 mmol的鎂離子(OR 0.59，95%CI為0.49 – 0.70)， 其早期的死亡率較為減少。針對次要結果的設後分析則沒有異質性的結果，其結果顯示接受靜注鎂離子和安慰劑一組比較有較低的心室顫動發生率(OR 0.88， 95%CI為0.81 – 0.96)， 但發生顯著低血壓(OR 1.13， 95%CI為1.09 – 1.19) 及心跳緩慢(OR 1.49， 95%CI為1.26 – 1。77) 的機會則較高，對於心臟阻滯則沒有差異(OR 1.05，95%CI為0.97 – 1.14)。對於有異質性證據的結果， 固定作用和隨機作用的後設分析皆顯示， 靜注鎂離子和安慰劑比較，能減少心室心搏過速(OR 0.45， 95%OR為0.31 – 0.66，由固定作用模型；OR 0.40，95%CI為0.19 – 0.84，由隨機作用模型)，和需要治療的嚴重心率不整或Lown 2 – 5的發生率(OR 0.72，95%CI為0.60 – 0.85，由固定作用模型； OR 0.51， 95%CI為0.33 – 0.79，由隨機作用模型)。針對心因性休克的則兩組無顯著差異。

作者結論

由於實驗設計可能存在出版偏差和對於治療明顯的異質性, 謹慎地解釋研究結果是非常必要的。從這裡回顧的證據來看, 我們認為: (1) 靜注鎂離子不太可能在早期治療, 晚期治療, 或已接受血栓溶解療法的病人上, 同樣有效得降低死亡率 (2) 靜注高劑量鎂離子(>75 mmol)將不太可能減少死亡率 (3) 靜注鎂離子也許能減少心室顫動, 心室心搏過速, 需要治療的嚴重心率, 或Lown 2 – 5的發生率, 但它也可能增加嚴重低血壓, 心跳緩慢和潮紅的發生 並且(4) 關於靜注低劑量(>75mmol)鎂離子的能否降低死亡率依然不是很確定

翻譯人

本摘要由臺北榮民總醫院楊尚峰翻譯。

此翻譯計畫由臺灣國家衛生研究院(National Health Research Institutes, Taiwan)統籌。

總結

在一些國家, 對心臟病發作患者執行靜脈注射鎂離子是為了減少心肌的損傷, 防止嚴重的心率不整和減少死亡風險, 有幾個小型試驗支持這樣的做法, 但這篇文章的作者發現其它幾個結果較不討喜的試驗並未被發表, 1995年爭論爆發, 一個共計有58,050個參加, 設計良好的大型試驗發表, 顯示靜注鎂離子並沒有提供任何好處, 和一些早期的小型研究的後設分析形成相反的結果, 這個回顧包括了26個臨床試驗, 針對心臟病發作患者, 隨機分為接受靜注鎂離子或是安慰劑的兩組, 他們的研究結果結果是混雜的: 靜注鎂離子減少了嚴重心率不整的發生, 但這種治療也增加了嚴重低血壓症, 心跳緩慢和潮紅的機會. 但是, 所有靜注鎂離子的好處也許只是反映實驗中各種各樣的誤差. 另外, 這些試驗所使用靜注鎂離子的劑量和治療時機缺乏一致性, 在一些試驗還包含使用抗凝結的藥物. 由這篇回顧的結果來看並不支持對靜注鎂離子的使用. 其它有效的治療(阿斯匹靈, betablockers) 應該被用來治療心臟病發作。