Ximeng Yang

Hypoxia-induced intrapulmonary arteriovenous shunting at rest in healthy humans

Intrapulmonary arteriovenous (IPAV) shunting has been shown to occur at rest in some subjects breathing a hypoxic gas mixture [fraction of inspired oxygen (FI(O(2))) = 0.12] for brief periods of time. In the present study we set out to determine if IPAV shunting could be induced at rest in all subjects exposed to hypoxia for 30 min. Twelve subjects (6 women) breathed four levels of hypoxia (FI(O(2)) = 0.16, 0.14, 0.12, and 0.10) for 30 min each in either an ascending or descending order with a 15-min normoxic break between bouts. Saline contrast echocardiography was used to detect IPAV shunt and a shunt score (0-5) was assigned based on contrast in the left ventricle with a shunt score ≥ 2 considered significant. Pulmonary artery systolic pressure (PASP) was determined using Doppler ultrasound. The total number of subjects demonstrating shunt scores ≥ 2 for FI(O(2)) = 0.16, 0.14, 0.12, and 0.10 was 1/12, 7/12, 9/12, and 12/12, respectively. Shunt scores were variable between subjects but significantly greater than normoxia for FI(O(2)) = 0.12 and 0.10. Shunt scores correlated with peripheral measurements of arterial oxygen saturation (SpO(2)) (r(w) = -0.67) and PASP (r(w) = 0.44), despite an increased shunt score but no increase in PASP while breathing an FI(O(2)) = 0.12. It is unknown how hypoxia induces the opening of IPAV shunts, but these vessels may be controlled via similar mechanisms as systemic vessels that vasodilate in response to hypoxia. Despite intersubject variability our results indicate significant IPAV shunting occurs at rest in all subjects breathing an FI(O(2)) = 0.10 for 30 min.

Effect of initial gas bubble composition on detection of inducible intrapulmonary arteriovenous shunt during exercise in normoxia, hypoxia, or hyperoxia

Concern has been raised that altering the fraction of inspired O₂ (Fi(O₂)) could accelerate or decelerate microbubble dissolution time within the pulmonary vasculature and thereby invalidate the ability of saline contrast echocardiography to detect intrapulmonary arteriovenous shunt in subjects breathing either a low or a high Fi(O₂). The present study determined whether the gaseous component used for saline contrast echocardiography affects the detection of exercise-induced intrapulmonary arteriovenous shunt under varying Fi(O₂). Twelve healthy human subjects (6 men, 6 women) performed three 11-min bouts of cycle ergometer exercise at 60% peak O₂ consumption (Vo(2peak)) in normoxia, hypoxia (Fi(O₂) = 0.14), and hyperoxia (Fi(O₂) = 1.0). Five different gases were used to create saline contrast microbubbles by two separate methods and were injected intravenously in the following order at 2-min intervals: room air, 100% N₂, 100% O₂, 100% CO₂, and 100% He. Breathing hyperoxia prevented exercise-induced intrapulmonary arteriovenous shunt, whereas breathing hypoxia and normoxia resulted in a significant level of exercise-induced intrapulmonary arteriovenous shunt. During exercise, for any Fi(O₂) there was no significant difference in bubble score when the different microbubble gas compositions made with either method were used. The present results support our previous work using saline contrast echocardiography and validate the use of room air as an acceptable gaseous component for use with saline contrast echocardiography to detect intrapulmonary arteriovenous shunt during exercise or at rest with subjects breathing any Fi(O₂). These results suggest that in vivo gas bubbles are less susceptible to changes in the ambient external environment than previously suspected.

Normal pulmonary gas exchange efficiency and absence of exercise-induced arterial hypoxemia in adults with bronchopulmonary dysplasia

Cardiopulmonary function is reduced in adults born very preterm, but it is unknown if this results in reduced pulmonary gas exchange efficiency during exercise and, consequently, leads to reduced aerobic capacity in subjects with and without bronchopulmonary dysplasia (BPD). We hypothesized that an excessively large alveolar to arterial oxygen difference (AaDO2) and resulting exercise-induced arterial hypoxemia (EIAH) would contribute to reduced aerobic fitness in adults born very preterm with and without BPD. Measurements of pulmonary function, lung volumes and diffusion capacity for carbon monoxide (DLco) were made at rest. Measurements of maximal oxygen consumption, peak workload, temperature- and tonometry-corrected arterial blood gases, and direct measure of hemoglobin saturation with oxygen (SaO2) were made preexercise and during cycle ergometer exercise in ex-preterm subjects ≤32-wk gestational age, with BPD (n = 12), without BPD (PRE; n = 12), and full term controls (CONT; n = 12) breathing room air. Both BPD and PRE had reduced pulmonary function and reduced DLco compared with CONT. The AaDO2 was not significantly different between groups, and there was no evidence of EIAH (SaO2 < 95% and/or AaDO2 ≥ 40 Torr) in any subject group preexercise or at any workload. Arterial O2 content was not significantly different between the groups preexercise or during exercise. However, peak power output was decreased in BPD and PRE subjects compared with CONT. We conclude that EIAH in adult subjects born very preterm with and without BPD does not likely contribute to the reduction in aerobic exercise capacity observed in these subjects.