Physiologic Rationale

Lung imaging studies in patients with ARDS show that alveoli that are poorly aerated because of exudative edema, hyaline membranes, and microatelectasis are not homogeneously distributed throughout the lung parenchyma. Instead, certain zones are relatively preserved and remain compliant, allowing them to receive disproportionately large fractions of the minute ventilation. The more diseased lung regions, located predominantly in the dependent areas of the lungs, may be poorly ventilated and yet receive much of the right ventricular cardiac output, resulting in a significant ventilation-perfusion mismatch.

Heart-lung interactions are also part of the pathology of ARDS. Laboratory research has shown that hypoxia-induced vasoconstriction leads to pulmonary hypertension. This is compounded by the dysregulation of constricting and dilating mediators, which contribute to a pathologic increase in the pulmonary vascular resistance. In severe ARDS, these effects may lead to right ventricular failure, a plausible independent predictor for death.

Theoretically, selective vasodilatation of vessels perfusing aerated lung tissue would redistribute blood from poorly ventilated regions, reducing the shunt fraction and at the same time correcting pulmonary hypertension. Improved oxygenation would reduce the mortality risk that is directly attributable to respiratory and right ventricular failure, whereas quicker resolution of ARDS would reduce the complications and morbidities associated with prolonged mechanical ventilation. Unfortunately, these are not the effects that investigators have observed in randomized clinical trials.

The following discussion focuses mainly on inhaled nitric oxide (NO), which is by far the most extensively studied inhaled vasodilator in the context of ARDS. Fewer data are available for nebulized prostaglandins, specifically prostaglandin I ₂ (PGI ₂ ; prostacyclin), prostaglandin E ₁ (PGE ₁ ; alprostadil), and prostaglandin E ₂ (PGE ₂ ; dinoprostone).

Nitric Oxide

In 1993, Rossaint et al. demonstrated in a prospective cohort of 10 patients that inhaled NO, as opposed to intravenous prostacyclin, improved oxygenation in adult patients with ARDS. This report supported the potential benefit of selective pulmonary vasodilatation. Other preclinical and clinical observational studies confirmed the effects of inhaled NO on arterial oxygenation. Added to further laboratory investigations finding additional benefits of NO on platelet and leukocyte function, these results inspired the conduct of several randomized clinical trials.

Two systematic reviews have evaluated inhaled NO in ARDS. Among the included randomized trials, the study populations varied to some extent. Most included adults with moderate to severe ARDS; however, some included children, those with less severe ARDS, or patients with a demonstrated favorable physiologic response to inhaled NO. Protocols for the dose and duration of therapy also varied from 1 to 80 ppm and less than 1 day to 28 days, respectively. One trial was a “dose-finding” study. Lastly, efforts to minimize bias ranged across the studies: 10 had concealed allocation, 5 studies blinded caregivers, and 6 reported on the use of alternative experimental therapies for ARDS.

Despite the nuances of study populations, therapeutic protocols, and methodological rigor, the results related to mortality were strikingly consistent. The relative similarity of patients, methods, and results supports the decision to statistically aggregate results for this outcome. With or without statistical pooling, a visual review of the meta-analytical results provides a strong impression ( Figure 34-1 ). The aggregate results further suggest that inhaled NO does not improve survival despite a demonstration of improved oxygenation. In fact, trends were more in keeping with increased mortality (relative risk 1.06; 95% confidence interval [CI], 0.93 to 1.22). Likewise, the pooled results suggest that inhaled NO is not beneficial in terms of duration of mechanical ventilation (mean difference 1.02 days; 95% CI, −2.08 to 4.12) or ventilator-free-days (mean difference −0.57; 95% CI, –1.82 to 0.69).

Inhaled nitric oxide ( iNO ) for acute respiratory distress syndrome ( ARDS ) and acute lung injury in children and adults. (Review)

(Copyright 2013 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.)

Both systematic reviews suggest a statistically significant increase in the risk of renal dysfunction with inhaled NO therapy in the four studies that evaluated this outcome (relative risk 1.59; 95% CI, 1.17 to 2.16). One unblinded and three blinded trials observed this effect.

The generalizability of these results to clinical practice is high. The studies included patients across the spectrum of ARDS who clinicians commonly considered (before the publication of these studies) for inhaled NO therapy. Moreover, the treatment effects were strikingly similar across studies, notwithstanding the variations in populations, drug administration protocols, and methodological quality.

In parallel with these systematic reviews, data on the long-term quality of life outcomes and costs of inhaled NO have emerged. Using the dataset of a previously published trial of inhaled NO in ARDS, Angus et al. performed a cost-effectiveness analysis suggesting that inhaled NO did not modify long-term outcomes or posthospital discharge costs. In a separate retrospective analysis of the same dataset, Dellinger et al. reported on the long-term pulmonary function of ARDS survivors who had participated in the trial. At 6 months, the 51 survivors treated with inhaled NO (compared with 41 who were not) had a greater mean (standard deviation [SD]) (1) total lung capacity (TLC; 5.54 [1.42] vs. 4.81 [1.0], P = 0.026), (2) percentage of predicted forced expiratory volume in 1 second (FEV ₁ ; 80.2 [21.2] vs. 69.5 [29.0], P = 0.042), (3) percentage of predicted forced vital capacity (FVC; 83.8 [19.4] vs. 69.8 [27.4], P = 0.02), (4) percentage of predicted FEV ₁ /FVC (96.1 [13.8] vs. 87.9 [19.8], P = 0.03), and (5) percentage of predicted TLC (93.3 [18.2] vs. 76.1 [21.8], P < 0.001). Most recently, Medjo and colleagues reported on a prospective observational study of inhaled NO in 16 children with ARDS who were compared with historic controls. Although oxygenation improved for up to 4 hours with inhaled NO, values had returned to baseline 24 hours after the onset of therapy and survival was not improved.

In summary, current clinical trials do not support a role for inhaled NO in the routine management of patients with acute lung injury and ARDS. In fact, meta-analyses suggest this approach to patient care is more likely to cause harm.

Prostaglandins

Bearing the same physiologic rationale as inhaled NO in ARDS, three vasodilating prostaglandin molecules are a focus of interest in ARDS research: prostaglandin I ₂ (PGI ₂ ), alprostadil (PGE ₁ ), and dinoprostone (PGE ₂ ). In addition, PGI ₂ blocks platelet aggregation and neutrophil migration, and PGE ₂ has anti-inflammatory properties. For these reasons, many investigators have hypothesized that nebulized prostaglandins would serve as selective vasodilators; therefore, they would be useful adjuncts in the context of ARDS.

The body of literature evaluating a role for inhaled prostaglandins in the management of patients with ARDS is limited. Dahlem et al. reported that among 14 children with ARDS randomized to nebulized prostacyclin or placebo, oxygenation did improve with prostacyclin (median change in oxygen index −2.5, interquartile range −5.8 to −0.2), but mortality was unchanged. Other uncontrolled trials led to similar results. In a dose-finding study, Van Heerden et al. treated nine adult patients who had ARDS with inhaled prostacyclin. The partial pressure of oxygen in arterial blood (Pa o ₂ )/fraction of inspired oxygen (F io ₂ ) increased, but prostacyclin had no effect on hemodynamic variables or on platelet function. Sawheny et al. treated 20 patients with ARDS and elevated pulmonary arterial pressures with PGI ₂ . The mean Pa o ₂ /F io 2 ratio increased from 177 (SD 60) to 213 (SD 67), but the partial pressure of carbon dioxide in arterial blod (Pa co ₂ ), peak and plateau airway pressures, systemic blood pressure, and heart rate did not significantly change. Using a different prostaglandin, Meyer et al. treated 15 adult patients with acute lung injury with inhaled PGE ₂ . The mean Pa o ₂ /F io ₂ ratio increased from 105 (standard error [SE] 9) to 160 (SE 17) ( P < 0.05) after 4 hours and to 189 (SE 25) ( P < 0.05) after 24 hours.

In contrast, Camamo et al. reviewed the charts of 27 patients treated with PGI ₂ or PGE ₁ (alprostadil) for a primary or secondary diagnosis of ARDS and found no statistically significant improvement in oxygenation. Likewise, in a prospective uncontrolled trial of nebulized PGI ₂ to 15 consecutive patients with ARDS and severe hypoxemia, Domenighetti et al. found no improvement in oxygenation.

Comparisons of Inhaled Nitric Oxide and Prostaglandin

Comparisons between nebulized PGI ₂ and inhaled NO suggest that these agents have similar effects. Walmrath et al. individually titrated doses of both agents sequentially. The effects on pulmonary arterial pressure and distribution of blood flow were nearly identical. Torbic et al. compared the effects and costs of inhaled NO and PGI ₂ in 105 patients. There was no difference in the change in Pa o ₂ /F io ₂ , duration of mechanical ventilation, and intensive care unit and hospital lengths of stay. The authors did observe that inhaled NO was 4.5 to 17 times more expensive than nebulized PGI ₂ .