Enhanced efficacy with MgSO4 due to an additive, alternative, or dual mode of action?

Pharmacological conversion of atrial fibrillation–atrial flutter (AF–AFl) is still an important therapeutic modality in daily patient care. Frequently used drugs are the class III anti-arrhythmics dofetilide, sotalol, and ibutilide. By blocking (specific) potassium currents, they will prolong the duration of the action potential (APD) providing less space for re-entry to manoeuvre. Their effectiveness is still not optimal and is also hampered by their pro-arrhythmic risk due to dose-dependent lengthening of the (ventricular) repolarization times. Their use is therefore restricted, including hospitalizations when therapy is started, as was the case in this single site (Hartford Hospital) retrospective investigation (2000–08) about the effectiveness of dofetilide to acutely convert AF–AFl. 1 To protect their patients for dofetilide-induced torsade de pointes (TdP) arrhythmias, some ( n = 50) from the 160 eligible patients with AF–AFl received adjunctive magnesium sulfate (Mg). The evaluation demonstrated that a single or repeated bolus of Mg (mean ± 3 g MgSO 4 ) did not only protect for drug-induced adverse effects (0/50 vs. 1/110 TdP occurrences), but also improved the efficacy of dofetilide to suppress/prevent AF–AFl: 38% without vs. 50% with Mg, P < 0.05. This important improvement with this combination (doubling of the odds), which was recently also shown for ibutilide and Mg by the same group, 2 should now be considered to become the standard regimen for the acute chemical cardioversion of AF–AFl. Not only is the therapeutic dose of dofetilide–ibutilide free of pro-arrhythmic risk, but the conversion rate with Mg even higher. It is of interest to elucidate whether higher dosages of the class III anti-arrhythmics can now be applied in order to enhance efficacy further. Magnesium is a mineral that is involved in transmembrane and intracellular modulation of specific ion channels and ion transport processes. Overt Mg deficiency (<1.8 mg/dL) has been associated with ventricular arrhythmias, and correcting plasma levels of magnesium is certainly warranted. In post-coronary artery bypass grafting patients, for example, AF incidence was reduced by more than 50%: 35% in the untreated vs. 16% in the treated group (Mg 2.5 g for 3 days) when the transient decrease in Mg concentrations post-surgery was corrected. 3 Other effective actions include arrhythmias in the setting of heart failure or acute ischaemic syndromes. In patients with normal magnesium (normal values between 1.8 and 2.6 mg/dL) in whom ventricular arrhythmias develop, administering MgSO 4 (2 g bolus followed by an infusion 2–4 mg/min) has been effective against drug-induced TdP and digitalis intoxication. Nowadays, Mg is likely the first choice of therapy for drug or inherited long-QT syndrome. 4 Whether these effects of Mg reflect replenishment of depleted intracellular Mg stores or an anti-arrhythmic effect of Mg itself (innate effect) is still unclear. To use Mg concomitantly with a Class III drug that is administered to convert AF–AFl is less accepted but certainly a strategy of interest. In this patient population, magnesium concentrations were within physiological ranges (Table 1 in Coleman et al.1 ). The protective action of Mg against TdP arrhythmias is associated with prevention or suppression of early afterdepolarizations (EADs). 5 These EADs, often seen in Phase 3 of the action potential, facilitate (repetitive) triggering of ectopic beats that are at the basis of TdP. There are a number of inward currents that can be responsible for these EADs: L-type calcium window current, sodium late current and the transient inward current generated by the natrium–calcium exchange. Magnesium has been implicated to block some if not all of these currents. Most likely is, however, the L-type Ca-channel because Mg is also called the ‘nature's physiological calcium blocker’. 6 Besides EADs, Mg administration has also been proved effective against delayed afterdepolarizations (DADs) and related triggered activity that occur in the setting of ouabain intoxication. 7 Also its effect against the generation of DADs may be explained by the block of the currents listed above. As mentioned, the effectiveness of Class III drugs in terminating AF–AFl is not optimal and largely depending on the duration of atrial fibrillation. In general, conversion rates for AF are in the order of <40%, whereas the addition of Mg improves this value to >50%. The question then arises how Mg is contributing to the improved conversion rate. Is this due to an additive effect on the already blocked repolarization currents by dofetilide or ibutilide or is it by an alternative mechanism, being suppression of EAD- or DAD-dependent triggered activity by blocking the L-type calcium window current. So what is the evidence that EADs or DADs may be part of the AF initiation and/or perpetuation mechanism? In a recent review, 8 the existing cellular data have been examined, ample cellular evidence provided, but still the authors concluded rather cautiously ‘that triggered activity participates in some as yet unspecified way in the occurrence of atrial fibrillation’. As mentioned, Mg may affect the APD and thereby making re-entrant wavelets less likely to participate. Low Mg will prolong APD, whereas higher Mg concentration will shorten APD. Both the inward rectifier as the slow and delayed component of the delayed rectifier, key components of repolarization (times), are modified by the Mg levels. Either way, innate Mg therapy should then already been able to terminate AF. In two independent mega analysis reports 9 , 10 concerning Mg therapy for the acute management of rapid AF, this kind of evidence was provided by showing that Mg-treated patients had a higher change to regain sinus rhythm than the placebo group or than patients receiving calcium channel blockers. The latter observation points to a dual effect of Mg therapy. Administration of Mg is not completely without risk. The most common reported side effects are transient sensation of warmth and flushing. Electrophysiologically, AV-conduction problems have been reported, whereas neuromuscular depression was seen with high dosages. So in conclusion, adjunct Mg administration does not only protect against drug-induced TdP but also enhances the efficacy of the Class III anti-arrhythmic considerably. The mechanism behind this favourable Mg action still has to be clarified, but might involve actions against re-entry and triggered activity.