Clinical Utility of Inhaled Nitric Oxide in ARDS

ABSTRACT & COMMENTARY

Exogenous administration of inhaled nitric oxide (no), an endogenous endothelial-derived smooth muscle relaxant, has demonstrated improved oxygenation and reduced pulmonary artery (PA) pressures in patients with the acute respiratory distress syndrome (ARDS), primary pulmonary hypertension of the newborn, and primary pulmonary hypertension in adults. The proposed mechanism of these effects is presumably due to vasodilatation of the pulmonary vessels, improving circulation to the better ventilated areas of the lung and shunting blood away from the poorly ventilated areas. To date, the limitations of published studies has been that relatively few patients were reported, only the acute physiologic effects of the inhaled NO were observed, and there were no concurrent placebo groups.

The design of this trial addressed those limitations. A prospective, double-blinded, placebo-controlled, phase II study was conducted at 30 academic, teaching, and community hospitals. ARDS patients were eligible to be enrolled if the onset of their disease was within 72 hours of randomization and if the etiology of ARDS was presumed to be pneumonia, aspiration, trauma, massive blood transfusion, pancreatitis, or major abdominal, chest, or orthopedic surgery. Since NO was expected to affect lung function, patients whose underlying condition wouldn't be altered by improved lung function were excluded (e.g., patients with head injury, refractory hypotension, and vasopressor use, or multiple organ dysfunction syndrome).

Patients were randomized to receive placebo (nitrogen gas) or one of five doses of NO: 1.25, 5, 20, 40, or 80 parts per million (ppm). Standardized ventilator management was used to optimize compliance, end-inspiratory plateau pressure, and oxygenation. Acute changes in oxygenation and PA pressures following treatment gas initiation were observed, as well as intensity of mechanical ventilation and the oxygenation index. The dose effects of inhaled NO on overall mortality and on the number of days alive on and off mechanical ventilation were compared.

Over a 14-month period, 177 patients were enrolled-57 of whom received placebo and 120 one of the doses of NO. The groups were well matched with respect to etiology of ARDS and baseline respiratory dysfunction. Ventilator settings were held constant for the first four hours of the study protocol, unless the patient's oxygenation deteriorated, so that the responders could be more easily identified. Responders were defined as patients who demonstrated greater than 20% increase over baseline PaO₂.

The proportion of responders in each inhaled NO group was about 60% but varied over time. At four hours, 24% of the placebo patients were responders. The oxygenation response in the first four hours for all the NO groups was significant. The acute change in PA pressure was evaluated in the 119 patients who had PA catheters; there was a significant drop of 2 mmHg in the NO group and no change in the placebo group. Over the seven-day study period, there was no significant reduction in level of FiO₂, inspiratory plateau pressure, or pulmonary artery wedge pressure.

There was significant improvement in PaO₂/FiO₂ ratio and reduction of mean PA pressure on day 1 in the NO group. Significant reduction in the NO group of the oxygenation index ([FiO₂ x mean airway pressure]/PaO₂), which is intended to reflect the intensity of mechanical ventilation, was seen on days 1, 2, and 3. Overall mortality was 30% for the placebo and the pooled NO groups, and there were no significant changes in survival across treatment groups.

The most common adverse events in all patients were pneumothorax (16%), cardiac arrest (12%), infection (10%), hypotension (9%), and renal failure (7%); these events occurred at similar rates in both groups. Seven adverse events in four patients receiving NO were thought to be possibly related to study drug: myopathy and agitation (NO @ 1.25 ppm); abnormal liver enzymes (NO @ 1.25 ppm); apnea, lung hemorrhage, and coagulation disorder (NO @ 40 ppm); and renal dysfunction (NO @ 40 ppm). (Dellinger RP, et al. Crit Care Med 1998;26:15-23.)

COMMENT BY DOREEN M. ANARDI, RN

Dellinger and colleagues and the Inhaled Nitric Oxide in ARDS Study Group are to be complimented on their excellent study design and execution of this technically difficult study. They achieved their aims of demonstrating that NO can be delivered safely and in a placebo-controlled fashion to patients with ARDS. The inhaled NO was well tolerated in doses between 1.25 ppm and 40 ppm. At 80 ppm, the patients had elevated NO₂ and methemoglobin levels, and, while these levels were not unsafe, they would bear watching if doses that high or higher were used. Although a significant improvement in oxygenation was seen in the first four hours of NO administration, these improvements decreased over time and did not translate into dramatic reductions in ventilator support.

Although the study was not designed with sufficient power to demonstrate significant benefit for any outcome, it is impossible not to note the lack of difference in mortality between the patients receiving placebo and those receiving NO. It is discouraging when yet another ARDS intervention that seems like it should "work" doesn't show benefit in the most looked at outcome of all-survival. This study design tried to identify patients at enrollment who would benefit from improvement in lung function. The problem with ARDS patients is that they are at increased risk of developing infections and other organ system dysfunctions every day that they live in the ICU, which may negate the temporary improvement in oxygenation seen from NO.

Although this is a cumbersome, costly, laborious technology, its use may rest in its ability to temporarily improve oxygenation, reducing the amount of inspired oxygen and iatrogenic pulmonary injury related to intense mechanical ventilation in certain patients with ARDS (i.e., getting them over a "hump"). The reduction in mean PA pressure may improve right ventricular function. It is possible that the non-smooth muscle effects of NO, inhibition of platelet aggregation, and anti-inflammatory properties may be revealed in a larger phase III trial. Whether such a trial should be conducted is already being debated.