Published papers

Objective: A noninvasive bronchoscopic microsampling (BMS) probe was developed to sample biochemical constituents of the epithelial lining fluid in small airways. Design: Observational, controlled study. Setting: Intensive care unit of academic medical center. Patients and Procedure: BMS was applied in a control group of seven patients who had hemoptysis or small solitary peripheral nodules but no hypoxemia or other signs of acute inflammation and in four patients with acute respiratory distress syndrome (ARDS), to test whether BMS can ascertain the presence of acute pulmonary inflammation without complications. Measurements and Results: Complications, including a significant decrease in arterial oxygen saturation, were observed neither during nor after BUS. In the ARDS group, albumin, lactate dehydrogenase, interleukin-6, basic fibroblast growth factor, and neutrophil elastase concentrations in epithelial lining fluid were significantly higher (p < .0001, p = .012, p < .0001, p < .0001, and p < .0001, respectively) than in the control group. Serial BMS was safely performed in one patient with ARDS, allowing us to observe a correlation between changes in the concentration of inflammation-related biochemical markers and clinical course of the disease. Conclusions: These results suggest that BMS is safe and useful to monitor pulmonary biochemical events in ARDS.

Background: Because it is postulated that gut is important via bacterial translocation in the development of the systemic inflammatory response and multiple organ dysfunction, the preservation of gut integrity is a therapeutic goal for physicians who care for critically ill patients. The aim of the current study was to evaluate whether epidural anesthesia prevented gut injury and subsequent translocation of endotoxin during acute progressive hypoxia in rabbits. Methods: After the placement of an epidural catheter, 18 male rabbits, anesthetized with buprenorphine-midazolam, were allocated randomly to two groups: 0.5% lidocaine (group E) and saline (group C) groups. The solutions (0.4 ml/kg) were injected epidurally, followed by continuous infusion (0.1 ml·kg-1·h-1) during the study period. Portal blood flow, portal endotoxin concentrations, and intramucosal pH (pHi) of the ileum were measured at baseline and during two stages of progressive hypoxia (fraction of inspired oxygen [FIO2] = 0.15 and 0.10). Results: In both study groups, the portal blood flow was preserved to a similar extent during acute hypoxia. However, phi was reduced to a lesser extent in group E (7.33 ± 0.12 versus 7.22 ± 0.12 at an FIO2 of 0.15 and 7.13 ± 0.15 versus 7.03 ± 0.12 at an FIO2 of 0.10; mean ± SD, P < 0.01), concurrently with lower portal endotoxin concentrations (P < 0.05) compared with group C. Conclusions: The current study showed that epidural anesthesia slowed the progression of intestinal ischemia during acute hypoxia, subsequently preventing translocation of endotoxin through the gut mucosa.

Background: Heme oxygenase 1 (HO-1), induced by a variety of stressors, provides endogenous carbon monoxide (CO) and bilirubin, both of which play consequential roles in organs. The current study aimed to examine whether induction of HO-1 and its by-products modulated endothelial interaction with circulating leukocytes and platelets evoked by sevoflurane anesthesia in vivo. Methods: Rats, pretreated with or without hemin, were anesthetized with sevoflurane in 100% O2, and lungs were mechanically ventilated. Platelets labeled with carboxyfluorescein diacetate succinimidyl ester and leukocyte behavior in mesenteric venules were visualized during sevoflurane anesthesia at 1,000 frames/s using intravital ultrahigh-speed intensified fluorescence videomicroscopy. To examine the mechanisms for the effects of HO-1 on leukocyte and platelet behavior, these studies were repeated with superfusion of either CO, bilirubin, or N-nitro-L-arginine methyl ester (L-NAME). Results: As reported previously, the elevation of sevoflurane concentration evoked adhesive responses of leukocytes, concurrent with platelet margination and rolling. Pretreatment with hemin, a HO-1 inducer, prevented such sevoflurane-elicited changes in the microvessels. These changes were restored by zinc protoporphyrin IX, a HO inhibitor, and repressed by CO but not by bilirubin. During sevoflurane anesthesia, however, nitric oxide suppression by L-NAME deteriorated microvascular flows irrespective of the presence or absence of the HO-1 induction. Conclusions: These results indicate that endogenous CO via HO-1 induction attenuates sevoflurane-induced microvascular endothelial interactions with leukocytes and platelets, although local nitric oxide levels appear to dominate microvascular flow in situ.

We examined the alterations of peripheral vascular responses following ischemic insult during perioperative period of esophagectomy. Increase of palm blood flow after vascular occlusion, i.e., reactive hyperemia (RH), measured by near-infrared spectroscopy (NIRS) was employed to assess forearm vascular responses. The measurements of RH were performed in esophagectomized patients (n=12) before induction of anesthesia and postoperatively until the next day of extubation in comparison with normal volunteers (n=11). After esophagectomy, the RH, which was comparable with those in volunteers, was depressed by 50% on 1 POD, and did not recover until the third POD. In particular, patients receiving laparoscopy-assisted surgery showed bess decrease of RH than those receiving the standard open laparotomy. These results suggest that vascular responses to increase blood flow against ischemic insult is depressed following esophagectomy.