DOI: 10.1097/01.CCM.0000104925.62805.D9

,  

,  

PMID: 14707600

Issn Print: 0090-3493

Publication Date: 2004/01/01


Print

How relaxed should we be with acute respiratory distress syndrome? *

Ross Freebairn

    loading  Checking for direct PDF access through Ovid

Excerpt

Neuromuscular blocking agents (NMBAs) have enjoyed mixed fortunes in the intensive care environment, and their role in the management of patients with acute respiratory distress syndrome (ARDS) remains controversial. Half a century ago, the forerunners of today’s intensive care units (ICUs) were developed to deal with poliomyelitis patients, in whom muscle paralysis was the cause rather than the solution to the problems of ventilation (1). At the same time, neuromuscular blockade was gaining popularity in the operating theater. Curare was utilized primarily to improve surgical exposure rather than to facilitate ventilation. In 1954, Beecher and Todd (2) warned of the “inherent toxicity” of curare and the associated six-fold greater mortality when used in anesthesia. The mysteries of this toxicity remained unresolved until Dripps et al. (3) published an article concerning 6,000 patients who had received curare, without a death, in 1961. The problems of postoperative hypoventilation and recurarization were recognized, and NMBA use became widely accepted in anesthetic practice.
With time, the migration of anesthetic practice to the ICU resulted in NMBAs being used in 98% of units (4). In common with many therapies imported to the ICU from other specialties, there is scant evidence of benefits accruing from the use of NMBAs in the ICU setting. In this issue of Critical Care Medicine, Dr. Gainnier and colleagues (5) demonstrate a sustained improvement in the Pao2/Fio2 ratios in a group of ventilated patients with ARDS who were managed with a 48-hr infusion of cisatracurium.
It is noteworthy that no significant changes in any of the measured variables were seen until the end of the infusion. A direct pharmacologic effect of cisatracurium would be expected to have an onset within the first 24 hrs, be sustained for the duration of infusion, and diminish after its cessation. Pharmacologic paralysis altering chest wall compliance with improvement in alveolar ventilation is one possible mechanism for increased oxygenation. Another mechanism may be reduced respiratory muscle oxygen consumption. However, improvement in oxygenation does not occur until the end of the second day, and it is sustained past the time when neuromuscular function would be expected to recover. The above mechanisms are therefore unlikely to account for the occurrence of this phenomenon. The prolonged effect parallels significant late changes from baseline positive end-expiratory pressure and plateau pressure. These changes are not seen in the control group, suggesting that after 2 days, the group receiving neuromuscular blockade had sustained less lung injury. Reduced ventilator-associated lung injury could result from decreased intrapulmonary damage or from reduction of intrapulmonary activation of circulating mediators. In either case, reduction in pathophysiologic changes could improve oxygenation and the potential for survival.
Unfortunately, increasing oxygenation does not always mean an improvement in clinical outcome. There is no data to suggest that improved arterial oxygenation can be used as a surrogate outcome measure for mortality (6). Dr. Gainnier and colleagues (5) improved oxygenation but fell short of demonstrating significant improvements in clinical outcomes. Nitric oxide, prone positioning, or extracorporeal circulation can successfully increase oxygenation in ARDS, but none of these improved survival (7–9). The intensivist must balance the possible benefits of improving oxygenation against the adverse effects associated with the means of achieving this goal. The data of Dr. Gainnier and colleagues (5) suggest another option for improving oxygenation, while avoiding the hazards of higher inspired oxygen concentrations or increased airway pressures.
It is remarkable that with only 56 patients, with variable primary diseases, the mortality differences approached statistical significance. If the 25% absolute reduction in mortality was confirmed in an appropriately powered study, the number needed to treat for each additional life saved is only four.

Related Topics

    loading  Loading Related Articles