Free radical injury is believed to be important in diaphragm dysfunction. N-Acetylcysteine (NAC) is a potent free radical scavenger shown in animal models to attenuate diaphragm fatigue; however, its effects on human diaphragm function are unknown. We assessed diaphragm function by electrophrenic twitch stimulation (PdiT) and twitch occlusion (to yield Pdimax) in four healthy subjects 35 +/- 3 yr of age (mean +/- SD). We intravenously administered NAC (150 mg/kg in 250 ml D5W) or placebo (CON) (250 ml D5W) in a randomized manner after subjects were premedicated with antihistamines. There were no significant side effects with the infusion. After infusion, we measured baseline Pdimax and PdiT at FRC. Diaphragm fatigue was then induced by subjects breathing through an inspiratory resistive load. Pdimax and PdiT were then measured at 15 to 30 min and 1, 2, 3, 4, and 20-25 h after fatigue. Times to fatigue were 13 +/- 4 min (CON) and 21 +/- 6 min (NAC) (p = 0.04). At 15 min after fatigue, PdiT was reduced to 40% (CON) compared with 30% (NAC) initial PdiT value (p = 0.05). Other twitch characteristics (maximal rate of relaxation and maximal contraction rate) were reduced to a greater degree after placebo compared with NAC. There were no significant differences in the rate of recovery between CON and NAC. Pdimax at 30 min after fatigue was significantly greater with NAC; however, at 1 h after fatigue, Pdimax for CON and NAC were not different, suggesting similar rates of recovery in high-frequency fatigue. These data suggest that NAC may attenuate low-frequency human diaphragm fatigue.

Since lung volume reduction surgery (LVRS) reduces end-expiratory lung volume, we hypothesized that it may improve diaphragm strength. We evaluated 37 patients for pulmonary rehabilitation and LVRS. Before and 8 wk after pulmonary rehabilitation, 24 patients had spirometry, lung volumes, diffusion capacity, incremental symptom limited maximum exercise test, 6-min walk test, maximal static inspiratory and expiratory mouth pressures, and transdiaphragmatic pressures during maximum static inspiratory efforts and bilateral supramaximal electrophrenic twitch stimulation measured. Twenty patients (including 7 patients who crossed over after completing pulmonary rehabilitation) had baseline measurements postrehabilitation, and 3 mo post-LVRS. Patients were 58 +/- 8 yr of age, with severe COPD and hyperinflation (FEV1, 0.69 +/- 0.21 L; RV, 4.7 +/- 1.4 L). Nineteen patients had bilateral LVRS performed via median sternotomy and stapling, and 1 patient had unilateral LVRS via thorascopy with stapling. After rehabilitation, spirometry and DL(CO)/VA were not different, and lung volumes showed a slight worsening in hyperinflation. Gas exchange, 6-min walk distance, maximum oxygen uptake (VO2max), and breathing pattern during maximum exercise did not change after rehabilitation, but total exercise time was significantly longer. Inspiratory muscle strength (PImax, Pdi(max combined), Pdi(max sniff), Pdi(max), Pdi(twitch)), was unchanged after rehabilitation. In contrast, after LVRS, FVC increased 21%, FEV1 increased 34%, TLC decreased 13%, FRC decreased 23%, and FRC(trapped gas) and RV decreased by 57 and 28%, respectively. PCO2 was lower (44 +/- 6 versus 48 +/- 6 mm Hg, p < 0.003) and 6-min walk distance increased (343 +/- 79 versus 250 +/- 89 m, p < 0.001), as did total exercise time during maximum exercise (9.2 +/- 1.9 versus 6.9 +/- 2.7 min, p < 0.01). Minute ventilation (29 +/- 8 versus 21 +/- 6 L/min, p < 0.001) and tidal volume (1.0 +/- 0.33 versus 0.84 +/- 0.25 L, p < 0.001) during maximum exercise increased whereas respiratory rate was lower (28 +/- 6 versus 32 +/- 7 breaths/min, p < 0.02). Measurements of respiratory muscle strength (PImax, 74 +/- 28 versus 50 +/- 18 cm H2O, p < 0.002; Pdi(max combined), 80 +/- 25 versus 56 +/- 29 cm H2O, p < 0.01; Pdi(max sniff), 71 +/- 7 versus 46 +/- 27 cm H2O, p < 0.01; Pdi(twitch), 15 +/- 5 versus 7 +/- 5 cm H2O, p < 0.01) were all greater post-LVRS. Inspiratory muscle workload as measured by Pdi TTI was lower following LVRS (0.07 +/- 0.02 versus 0.09 +/- 0.03, p < 0.03). On multiple regression analysis, increases in PImax correlated significantly with decreases in RV and FRC(trapped gas) after LVRS (r = 0.67, p < 0.03). We conclude that LVRS significantly improves diaphragm strength that is associated with a reduction in lung volumes and an improvement in exercise performance. Future studies are needed to determine the relationship and stability of these changes over time.