Lung fibrosis can be complicated by pulmonary hypertension, limiting exercise tolerance and life expectancy. Furthermore, vasodilators might cause deterioration in gas exchange. Our aim was to compare acute effects of sildenafil, nitric oxide, and epoprostenol in individuals with pulmonary hypertension secondary to lung fibrosis.

We did a randomised controlled, open-label trial, in 16 individuals admitted to our hospital with pulmonary hypertension secondary to lung fibrosis. After inhalation of nitric oxide (10–20 ppm), we assigned patients to either maximum tolerated dose of intravenous epoprostenol (mean 8·0 ng/kg per min; n=8) or oral sildenafil (50 mg; n=8). Our primary objective was to assess pulmonary vasodilative potency (decrease in pulmonary vascular resistance index) of sildenafil by comparison with inhaled nitric oxide and infused epoprostenol. Analyses were by intention to treat.

Pulmonary vascular resistance index was reduced by nitric oxide (−21·9%, 95% Cl −14·1 to −36·2), epoprostenol (−36·9%, −24·4 to −59·6), and sildenafil (−32·5%, −10·2 to −54·1). However, ratio of pulmonary to systemic vascular resistance decreased only in individuals who received nitric oxide and sildenafil. Baseline measurement of multiple-inert-gas elimination showed right-to-left shunt flow (4·8%, 0·0−28·2) and little perfusion of low ventilation(V)/perfusion(Q) areas (0·1%, 0·0−13·0). Prostacyclin increased V/Q mismatch (shunt 16·8%, 10·8−35·9; low V/Q 3·8%, 0·0−13·0) and decreased arterial oxygenation. By contrast, nitric oxide (4·5%, 0·0−18·0; 0·0%, 0·0−17·3) and sildenafil (3·3%, 0·0−11·3; 0·0%, 0·0−12·4) maintained V/Q matching, with raised arterial partial pressure of oxygen (14·3 mm Hg, −1·7 to 31·3) noted for sildenafil. We recorded no adverse events.

Sildenafil causes preferential pulmonary vasodilation and improves gas exchange in patients with severe lung fibrosis and secondary pulmonary hypertension.