Key Point:

Methylprednisolone does not appear to improve outcomes in patients with severe, persistent ARDS.

ORLANDO, FLA—Methylprednisolone appears to be no more effective than placebo in patients with the adult respiratory distress syndrome (ARDS), investigators from ARDSNet reported at the recent annual meeting of the American Thoracic Society.¹ In fact, corticosteroid use may increase the duration of intensive care in ARDS patients, said Leonard D. Hudson, MD, one of the ARDSNet investigators.

ARDSNet, the large network of clinical centers established by the National Institutes of Health, has performed—and continues to undertake—a number of landmark studies on ARDS. Also presented at the meeting were results from several other ARDSNet trials; among the topics addressed were factors that increase the risk of death and the impact that the ARDSNet findings about low tidal volumes have had on physicians’ practice.

THE STEROID STUDY

The corticosteroid study had a simple goal: to determine whether methylprednisolone administration could reduce mortality and morbidity in patients with severe, persistent ARDS. The primary end point was the proportion of patients who were alive and at home 60 days after study entry.

Although final results of the study are still pending, preliminary data were presented by Dr. Hudson, a Professor of Pulmonary and Critical Care Medicine at the University of Washington. He noted that these data must be interpreted with caution because more than one quarter of the 180 patients originally enrolled in the study were still in the hospital two months later. “At 60 days, 40 patients remained in the hospital off the ventilator, and 12 patients remained in the hospital on the ventilator,” Dr. Hudson explained.

To be included in the study, patients had to have ARDS for seven to 28 days. ARDS was considered to be present if, within a 24-hour interval, a patient who was receiving positive pressure ventilation through an endotracheal tube had:

Following enrollment, the patients were randomized to receive methylprednisolone or placebo. Methylprednisolone was delivered initially as a loading dose of 2 mg/kg of predicted body weight, followed by the same amount in divided doses for two weeks and 1 mg/kg in divided doses for one week. The drug was then tapered.

When the primary end point was analyzed, no difference between the two study groups could be found. By day 60, 26 patients in each group had died. In addition, 20 patients in each group still required hospitalization but not mechanical ventilation; nine patients given placebo and three methylprednisolone recipients remained hospitalized and mechanically ventilated. Thirty-six patients in the placebo group and 40 patients in the methylprednisolone group had been discharged home.

The methylprednisolone recipients did fare better with regard to several secondary outcomes, including the median number of ventilator-free days at study day 28, the median number of organ failure–free days at day 28, the risk of cardiovascular failure at day 28, and the median change in bronchoalveolar lavage markers of inflammation and fibroproliferation between baseline and day 7. Also, the overall 60-day mortality rate was only 29%, surprisingly low for ARDS patients, Dr. Hudson remarked.

However, methylprednisolone treatment was associated with significantly more days of intensive care by day 28. And although the two groups had similar overall adverse event rates, a trend toward an increased risk of neuromyopathy was identified in the methylprednisolone group. By contrast, there were more cases of suspected and probable pneumonia and of septic shock in the placebo group.

According to Dr. Hudson, closer analysis of the results revealed a useful lesson: The percentage of survivors in each group who were no longer receiving mechanical ventilation was significantly different at day 60—65% in the placebo cohort and 80% in the methylprednisolone recipients—which suggested that the corticosteroid might be beneficial in some patients. Yet, considerably more of the patients in the methylprednisolone group who were taken off the ventilator had to be put back on it between baseline and day 60—19, versus four in the placebo group.

Based on the findings to date from this ongoing study, Dr. Hudson offered this conclusion: “At this time … we are unable to recommend methylprednisolone for persistent ARDS based upon our primary outcome variable.”

PREDICTORS OF DEATH

To identify baseline clinical variables that independently predict 28-day mortality in patients with ARDS or acute lung injury (ALI), Lanken and colleagues performed a secondary analysis of 902 patients from several earlier ARDSNet trials. In this group, related Dr. B. Taylor Thompson, an Associate Professor of Medicine at Harvard Medical School in Boston, 12 baseline variables predicted 28-day mortality: age, number of days in the hospital before study entry, cause of ALI/ARDS, number of organ failures, radiologic lung injury score, tidal volumes (high versus low), PaO₂/FiO₂ ratio, urine output, fluid balance, and albumin concentration.

An elevated level of plasminogen activator inhibitor 1 (PAI-1) was associated with increased mortality in a large controlled study of ALI/ARDS patients by Ware et al. The median baseline PAI-1 value was 90 ng/mL among the patients who died but only 66 ng/mL among the survivors. Elevated PAI-1 levels were also associated with decreases in the number of ventilator-free and organ failure–free days. Rising PAI-1 concentrations are a marker for impaired fibrinolysis, which may contribute to multisystem organ failure by triggering widespread fibrin deposition in the lungs and other organs, Dr. Thompson suggested.

LOW TIDAL VOLUMES

What impact have previous ARDSNet findings had on clinical practice? To address this question, Dr. Thompson described an analysis by Brower et al, which found that among the hospitals in ARDSNet, the mean tidal volume prescribed for mechanically ventilated ALI/ARDS patients fell significantly—from 10.4 to 7.9 mL/kg—in the two years after a 2000 study demonstrated that lower tidal volumes are associated with better outcomes.

The 2000 study has been criticized by some because it compared only low and high tidal volumes; it might have been better, these critics maintain, if a control arm had allowed some patients to received unrestricted best current care. According to Dr. Thompson, the addition of this control arm would have done nothing to improve the study’s safety in terms of mortality, but it would have increased the study’s size, cost, and complexity.

—Timothy Begany

Reference
1. Bernard GR, Hudson LD, Thompson BT. NIH NHLBI ARDSNetwork: Report on ongoing clinical trials. Presented at: annual meeting of the American Thoracic Society; May 24, 2004; Orlando.

Return to table of contents