Endothelial cells release endothelium-derived relaxing factor (EDRF) in a variety of vascular beds, including the pulmonary circulation. However, the role of EDRF-mediated pulmonary-artery relaxation in chronic hypoxic lung disease is unknown.

We studied endothelium-dependent relaxation mediated by EDRF in vitro in pulmonary arteries that had been obtained from 22 patients undergoing heartlung transplantation for end-stage chronic obstructive lung disease. Control pulmonary arteries were obtained from 15 patients undergoing lobectomy for lung carcinoma who did not have evidence of other chronic lung disease. The responses of all vascular rings (external diameter, 1.2 to 3.4 mm) to the endothelium-dependent vasodilators acetylcholine and adenosine diphosphate were studied immediately after lung excision.

Pulmonary arterial rings from the patients with chronic lung disease developed a greater tension (2.19±0.16 g) in response to phenylephrine (10^–6 M) than the rings from control patients (1.28±0.18 g, P<0.05). Inhibition of EDRF synthesis by treatment with N^G-monomethyl-L-arginine (10^–4 M) eliminated this difference, increasing the tension in the rings from the controls (P<0.01) but not in those from the patients with chronic lung disease. Rings from control patients relaxed in response to cumulative doses (10^–10 to 10^–5 M) of acetylcholine (maximal relaxation, 81.3±3.9 percent) and adenosine diphosphate (maximal relaxation, 85.3±2.6 percent). By contrast, rings from patients with chronic obstructive lung disease achieved only 41.3±4.8 percent of maximal relaxation in response to acetylcholine (n = 32) and 49.4±5.5 percent in response to adenosine diphosphate (n = 24) (P<0.001, as compared with control rings). Rings from both the controls and the patients with chronic lung disease relaxed similarly in response to the endothelium-independent vasodilator sodium nitroprusside (10^–4 M). There was an inverse correlation between the degree of intimal thickening and the level of maximal relaxation of the rings from the patients with chronic lung disease (r = -0.60, P<0.001). Maximal relaxation was also related directly to the partial pressure of arterial oxygen before transplantation (r = 0.68, P<0.01) and inversely to the partial pressure of arterial carbon dioxide before transplantation (r = -0.55, P<0.01), but not to the forced expiratory volume in one second (r = 0.19, P not significant).

Endothelium-dependent pulmonary-artery relaxation in vitro is impaired in arteries from patients with end-stage chronic obstructive lung disease. Such impairment may contribute to the development of pulmonary hypertension in chronic hypoxic lung disease. (N Engl J Med 1991; 324:1539–47.)