EFFECTIVENESS OF NITRIC-OXIDE INHALATION IN SEPTIC ARDS

Study objective: To evaluate the percentage of nitric oxide (NO) respo nders in septic shock patients with ARDS. Additionally, to investigate long-term NO effects on cardiac performance and oxygen kinetic patter ns in NO responders vs nonresponders. Design: Prospective cohort study . Setting: ICU of a university hospital. Patients: Twenty-five consecu tive patients with a diagnosis of septic shock and established ARDS re quiring inotropic and vasopressor support. Interventions: After diagno sis of ARDS, NO was administered at 18 or 36 ppm. Patients demonstrati ng a NO-induced rise of arterial oxygen tension of 20% or more and/or a fall in mean pulmonary artery pressure of 15% or more were grouped a s NO responders; others were grouped as nonresponders. Measurements an d results: Ten patients (40%) were NO responders, while 15 patients (6 0%) were nonresponders. Mortality was 40% in NO responders and 67% in nonresponders (NS). NO responders developed a significantly lower mean pulmonary artery pressure (28+/-6 vs 33+/-6 mm Hg; p<0.05), lower pul monary vascular resistance (PVR: 258+/-73 vs 377+/-163 dyne . s . cm(- 5). m(-2); p<0.05), and higher PaO2/FIo(2) ratio (192+/-85 vs 144+/-74 mm Hg; p<0.05) within the study period. In responders, NO-induced aft erload reduction resulted in increased right ventricular ejection frac tion (RVEF: 40+/-7 vs 35+/-9%; p<0.05), significantly higher cardiac i ndex (CI: 4.5+/-1.1 vs 4.0+/-1.2 L . min(-1). m(-2): p<0.05) and oxyge n delivery (Do(2): 681+/-141 vs 599+/-160 mL . min(-1). m(-2); p<0.05) compared with nonresponders. In NO nonresponders, RVEF was correlated with PVR, CI, Do(2), mixed venous oxygen saturation (SvO(2)), and oxy gen extraction ratio (O(2)ER) (r=+/-0.60 to +/-0.69; p<0.05). No signi ficant correlation between RVEF and any of these parameters was observ ed in responders. SvO(2) (75+/-7 vs 69+/-8%; p<0.05) and O(2)ER (0.24/-0.06 vs 0.27+/-0.06; p<0.05) were significantly different between re sponders and nonresponders, while no difference in oxygen consumption was observed (161+/-41 vs 153+/-43 mL . min . m(-2)). Conclusions: Inh aled NO is effective in only a subgroup of septic ARDS patients, with a higher, but insignificantly different percentage of survivors in the responder group. NO responders were characterized by increased RVEF a ccompanied by higher CI, Do(2), and lower O(2)ER. In nonresponders, RV EF remained depressed, with a close correlation between RVEF and CO as well as Do(2) and O(2)ER. Thus, nonresponders seem to suffer from imp aired cardiac reserves and correspondingly lower oxygen transport vari ables.