The Lack of Effect of Routine Magnesium Administration on Respiratory Function in Mechanically Ventilated Patients: Finally
The role of worsening diaphragmatic weakness as an acute cause of failure to wean from mechanical ventilation is not well defined. Several reviews have concluded that inspiratory muscle weakness may predispose to acute respiratory failure or complicate weaning from mechanical ventilation. The corollary of this would be that treatment aimed at improving respiratory muscle strength should help to hasten weaning from mechanical ventilation. We did not specifically test whether magnesium infusion would shorten the time to successful weaning but rather tested whether magnesium infusion improved respiratory muscle performance. We did, however, find that those patients who increased VC most following magnesium infusion (ie, change in VC greater than 0) tended to be more rapidly weaned from mechanical ventilation (5 ±2 days) compared with those patients who did not demonstrate an increase in VC (9 ±8 days). This, however, may merely reflect the tautology that patients wean when they improve.
Part of the effect that magnesium may have had upon the measurements of VC could be mediated by its role as a bronchodilator. However, none of our patients were known to have asthma and the rapidity of administration of magnesium in the treatment of asthma (1 g in 15 min) far exceeded our infusion rate (1 g every 160 min). Finally, it must be recognized that high serum magnesium levels may result in neuromuscular blockade with resultant diaphragmatic weakness. This effect of neuromuscular blockade is potentiated by concurrent use of aminoglycosides and may therefore pose an extra hazard in the treatment of patients in the ICU. However, none of our patients was found to have a level in the toxic range following the infusion of 6 g of magnesium.
In summary, we did not find an overall benefit to routine magnesium infusion in mechanically ventilated patients with normal serum magnesium levels. There were, however, no complications of this treatment in our selected population and with the slow infusion of 6 g of MgS04, we did not achieve clinically dangerous levels of serum magnesium. Although picking subgroups of responders within a given population is fraught with statistical hazards, we used reasonable guidelines that would have allowed us to decide if there was indeed a subgroup of “responders” to magnesium infusion. The study was based on the hypothesis that a large subgroup of patients with tissue magnesium deficiency existed and that they would have a large clinical response to magnesium replacement. We could not find this subgroup on evaluating 21 separate episodes of respiratory failure. We believe that in the absence of any contraindication, the routine administration of magnesium can be attempted in patients who are difficult to wean from mechanical ventilation, but that it is unlikely to be of much clinical benefit.