Recurrence rate and subjective symptoms after standardized (Hamburg protocol) phototherapeutic keratectomy on recurrent corneal erosions.

PURPOSE: To evaluate the objective and subjective outcome after phototherapeutic keratectomy (PTK) on recurrent corneal erosions (Hamburg protocol). METHODS: For the standardized PTK according to Hamburg protocol a manual abrasio corneae performed with 20 % alcohol is followed by an excimer ablation depth of ≥15 µm (group1 15 µm; group 2 > 15 µm ablation depth) and 7 mm optical zone. All patients (N = 48) were invited for follow-up examinations and the evaluation of changes concerning subjective symptoms. RESULTS: A significantly reduced subjective impairment of night vision, significantly less pain and less foreign body sensations (for all p < 0.05) were noted in 48 patients that met the inclusion criteria (26 women, 22 men). In the follow-up period an improvement of corrected distance visual acuities (cdva) from 0.80 to 1.08 (group 1) and from 0.58 to 0.99 (group 2) was demonstrated. CONCLUSION: PTK (Hamburg Schema) is a safe and effective procedure to reduce subjective symptoms and improve discomfort in recurrent corneal erosion.

Burn and smoke inhalation injury in sheep depletes vitamin E: kinetic studies using deuterated tocopherols.

To test the hypothesis that burn and smoke injury will deplete tissue alpha-tocopherol and cause its faster plasma disappearance, deuterium-labeled vitamin E was administered to sheep exposed to both surface skin burn and smoke insufflation, which cause injuries similar to those of human victims of fire accidents. Two different protocols were used: (1) deuterated vitamin E was administered orally with food at time 0 (just before injury) or (2) the labeled vitamin E was administered orally with food the day before injury. The animals, which had been operatively prepared seven days before, were anesthetized and then received both 40% body surface area third-degree burn and 48 breaths of cotton smoke or sham injuries. All were resuscitated with Ringer's lactate solution (4 ml/kg/% BSA burn/24 h) and mechanically ventilated. Blood samples were collected at various times after vitamin E dosing. In both studies the depletion of plasma alpha-tocopherol was faster in the injured sheep. The sheep given deuterated vitamin E 24 h before injury had similar maximum alpha-tocopherol concentrations at similar times. The exponential rates of alpha-tocopherol disappearance were 1.5 times greater and half-lives were 12 h shorter (p < 0.05) in the injured sheep. In separate studies, various tissues were obtained from sheep that were sacrificed from 4 to 48 h after injury. The liver alpha-tocopherol concentrations in sheep killed at various times after injury seem to show a linear decrease at a rate of 0.1 nmol alpha-tocopherol/g liver per hour, suggesting that the liver is supplying alpha-tocopherol to maintain the plasma and lung alpha-tocopherol concentrations, but that this injury is so severe the liver is unable to maintain lung alpha-tocopherol concentrations. These findings suggest that alpha-tocopherol should be administered to burn patients to prevent vitamin E depletion and to protect against oxidative stress from burn injury.

The role of the bronchial circulation in the acute lung injury resulting from burn and smoke inhalation.

Smoke inhalation in burn patients is a serious medical problem around the world. Inhalation injury increases mortality in addition to increasing infections, ventilator-days, and hospital stays. There are also large numbers of patients subjected to smoke inhalation without burns from cooking fires, burning crops and forest fires. The injury results in a fall in arterial oxygenation as a result of airway blockade, increased pulmonary transvascular fluid flux and loss of hypoxic pulmonary vasoconstriction. The changes in cardiopulmonary function are mediated at least in part by reactive oxygen and nitrogen species. Nitric oxide (NO) is generated by both inducible and constitutive isoforms of nitric oxide synthase (NOS). NO combines with superoxide to form reactive nitrogen species such as peroxynitrite. These reactive nitrogen species can be detected by measuring their reaction products such as 3-nitrotyrosine. The latter is elevated in the airway following smoke/burn injury. The control of NO formation involves poly (ADP ribose) polymerase (PARP) and its ability to up-regulate the activity of nuclear transcription factors through ribosylation. Present data also support a major role for the bronchial circulation in the injury since blockade of bronchial blood flow will also minimize the pulmonary injury. The data suggest that cytotoxins or activated cells are formed in the airway and carried to the parenchyma. These materials cause the formation of oedema and a reduction of PaO(2).

Positive end-expiratory pressure ventilation increases extravascular lung water due to a decrease in lung lymph flow.

Positive end-expiratory pressure (PEEP) is used to improve gas exchange, increase functional residual capacity, recruit air spaces, and decrease pulmonary shunt in patients suffering from respiratory failure. The effect of PEEP on extravascular lung water (EVLW), however, is still not fully understood. This study was designed as a prospective laboratory experiment to evaluate the effects of PEEP on EVLW and pulmonary lymph flow (QL) under physiologic conditions. Twelve adult sheep were operatively prepared to measure haemodynamics of the systemic and pulmonary circulation, and to assess EVLW In addition, the lung lymphatic duct was cannulated and a tracheostomy performed. The animals were then mechanically ventilated in the awake-state without end-expiratory pressure (PEEP 0). After a two-hour baseline period, PEEP was increased to 10 cmH2O for the duration of two hours, and then reduced back to 0 cmH2O. Cardiopulmonary variables, QL, and arterial blood gases were recorded intermittently; EVLW was determined two hours after each change in PEEP. The increase in PEEP resulted in a decrease in QL (7 +/- 1 vs 5 +/- 1 ml/h) and an increase in EVLW (498 +/- 40 vs 630 +/- 58 ml; P<0.05 each) without affecting cardiac output. As PEEP was decreased back to baseline, QL increased significantly (5 +/- 1 vs 10 +/- 2 ml/h), whereas EVLW returned back to baseline. This study suggests that institution of PEEP produces a reversible increase in EVLW that is linked to a decrease in QL.

Role of nitric oxide in myocardial dysfunction after combined burn and smoke inhalation injury.

This study tested the hypothesis that nitric oxide (NO) synthesized from inducible NO synthase (iNOS) is responsible for the cardiac dysfunction observed after burn and smoke inhalation injury. Twelve sheep received 40% third-degree burn and smoke inhalation under halothane anesthesia. The animals were divided into two groups: a MEG group [iNOS was inhibited with mercaptoethylguanidine (MEG), a selective inhibitor of iNOS, n=6] and a control group (n=6). The control group showed a significant increase in NO(2)(-)/NO(3)(-) (NO(x)) concentration, metabolite of NO, in plasma after 24 h, whereas the MEG group did not. In the control group, cardiac depression was observed immediately after injury associated with hemoconcentration. Cardiac function returned to a normal level within 6 h following injury. In the control group cardiac dysfunction was observed again after 24 h although the hemoconcentration peaked at 24 h after injury and then began to resolve. In the MEG group, cardiac depression and hemoconcentration were not observed. The present data suggest that cardiac depression seen with this combination injury consists of two phases and that the later phase is mediated by iNOS-NO.

Pathophysiological analysis of combined burn and smoke inhalation injuries in sheep.

We investigated the pathophysiological alterations seen with combined burn and smoke inhalation injuries by focusing on pulmonary vascular permeability and cardiopulmonary function compared with those seen with either burn or smoke inhalation injury alone. To estimate the effect of factors other than injury, the experiments were also performed with no injury in the same experimental setting. Lung edema was most severe in the combined injury group. Our study revealed that burn injury does not affect protein leakage from the pulmonary microvasculature, even when burn is associated with smoke inhalation injury. The severity of lung edema seen with the combined injury is mainly due to augmentation of pulmonary microvascular permeability to fluid, not to protein. Cardiac dysfunction after the combined injury consisted of at least two phases. An initial depression was mostly related to hypovolemia due to burn injury. It was improved by a large amount of fluid resuscitation. The later phase, which was indicated to be a myocardial contractile dysfunction independent of the Starling equation, seemed to be correlated with smoke inhalation injury.

Role of nitric oxide in vascular permeability after combined burns and smoke inhalation injury.

Patients with severe burn and/or smoke inhalation injury suffer both systemic and pulmonary vascular hyperpermeability. We hypothesized that nitric oxide (NO) produced by inducible nitric oxide synthase (iNOS) plays a role in the changes in microvascular permeability seen with this injury. To test the hypothesis, we administered mercaptoethylguanidine (MEG), a selective iNOS inhibitor, to conscious sheep subjected to a combined smoke inhalation and third-degree burn injury to 40% of total body surface area. The sheep were surgically prepared for chronic study with lung and prefemoral lymph fistulas in order to estimate microvascular permeability. Both the groups and a control group of animals showed an increase in iNOS protein and message in their lungs. The control animals showed significant increases in either plasma or lymph NO2-/NO3- (NOx) concentration at 24 h after injury, with associated cardiac depression and hemoconcentration. The airway epithelium stained for nitrotyrosine. In the treatment group, NOx did not increase significantly in plasma or lymph throughout the experiment, there was no nitrotyrosine staining, hemodynamic depression was not observed, and the fluid requirement was significantly less than in the control group. Changes in pulmonary microvascular permeability were significantly suppressed by inhibition of iNOS. However, there was no significant difference between the two study groups in the microvascular permeability of burned tissue. These data suggest that NO produced by iNOS plays an important role in the changes in systemic and pulmonary microvascular permeability in combined smoke inhalation/third-degree burn injury, but does not affect the vascular permeability of third-degree-burned tissue in this type of injury.

Enhanced pulmonary expression of endothelin-1 in an ovine model of smoke inhalation injury.

Recent studies suggest a role of endothelin-1 (ET-1) in mediating airway inflammation and lung injury. The aim of this study was to assess the immunohistochemical expression of ET-1 in the lung following smoke inhalation injury. ET-1 immunoreactivity was assessed in normal sheep (N = 4) and in sheep at 1 (N = 2), 6 (N = 3), 12 (N = 3), and 24 (N = 3) hours after inhalation injury. In normal animals, ET-1 expression was limited to the basal cell layer of the tracheal epithelium, main bronchi, and associated mucous glands. One hour after injury, ET-1 immunoreactivity was enhanced in upper airway epithelium and mucus glands with new expression in bronchioles. Airway smooth muscle, vascular tissue, and alveolar duct smooth muscle cells expressed moderate levels of ET-1 at 12 and 24 hours. ET-1 immunoreactivity was absent in areas of parenchymal edema and inflammation. The pattern of ET-1 expression following inhalation injury suggests that this peptide may contribute to the airway inflammation, mucus secretion, pulmonary hypertension, increased airway resistance, and decreased lung compliance, which are evident in our ovine model of inhalation injury.

Both dermal matrix and epidermis contribute to an inhibition of wound contraction.

Contracture is a major detriment to functional recovery from large wounds. To determine the relative value of dermal replacement and epidermal coverage in inhibiting wound contraction, five full-thickness wounds (all 5 x 5 cm2) were placed on the back of 8 swine and treated in the following manner: (1) open wound, (2) porcine acellular dermis (analogous to AlloDerm for human use), (3) porcine acellular dermis with epidermal autograft placed 7 days postwounding, (4) porcine acellular dermis with immediate epidermal autograft, and (5) conventional-thickness autograft. Scar dimensions and punch biopsies were taken at days 14 and 30 postwounding. The planimetry results demonstrated that wound contraction was significantly greater with the open wounds (group 1) than all other wounds with a dermal substitute. Furthermore, wounds with initial epidermal coverage had significantly less contraction than unepithelialized wounds (14.8 +/- 1.1 cm2 at day 14 in wound group 2 vs. 20.4 +/- 0.6 cm2 in wound group 4; p < 0.05). Biopsy results revealed that wounds with initial epithelial coverage had the least amount of inflammation. These findings suggest that both dermal matrix and epidermal coverage contribute to an inhibition of wound contraction and that prompt epithelial coverage appears to impede contraction by reducing inflammation.

Effects of an L-selectin antibody on the pulmonary and systemic manifestations of severe smoke inhalation injuries in sheep.

Sheep were treated with either lymphocyte adhesion molecule (LAM)1-3, an antibody against L-selectin, (40 mg 1 hour before smoke inhalation and 35 mg 24 hours after smoke inhalation; n = 6) or equivalent volumes of 0.9% saline solution (n = 6). After the smoke inhalation injuries, the PaO2/FIO2 ratio declined in both groups until 40 hours after the injuries, when a trend toward improvement was noted in the group that received LAM1-3. Lung lymph flow increased in both groups until 36 hours after the smoke inhalation injuries and then significantly decreased in the group that received LAM1-3. Forty-eight hours after the smoke inhalation injuries, there was a significant decrease in the ratio of wet-dry lung weight and in preservation of the reflection coefficient in the group that received LAM1-3 (P < .05). Histopathologic examination showed no differences between the groups in the pulmonary morphology associated with smoke inhalation. A reduction in splanchnic blood flow was noted in the control group (P < .05); this reduction was attenuated by treatment with LAM1-3. The delayed pulmonary effects and improved splanchnic blood flow suggested that LAM1-3 attenuated the development of a systemically induced secondary lung injury rather than of the primary lung injury associated with smoke inhalation.

The effect of inducible nitric oxide synthase (iNOS) inhibition on smoke inhalation injury in sheep.

Recent studies on smoke inhalation injury have been focused on nitric oxide (NO) as an essential factor of progressive lung injury. We studied the effects of inducible nitric oxide synthase (iNOS) inhibition on inhalation injury in sheep. Sheep (n = 14) were prepared surgically for chronic study. After recovery period, the sheep received 48 breaths of cotton smoke. The animals were then randomised into two groups: MEG group [30 mg/kg mercaptoethylguanidine (MEG), selective inhibitor of iNOS and peroxynitrite scavenger, was given 1 h after injury and then 8 h for 41 h, n = 7] and control group (0.9% NaCl, n = 7). All animals were ventilated mechanically, and airway blood flow was measured using colored microspheres. In the control group, following significant increase in airway blood flow, deterioration in the PaO2/FiO2 ratio was observed. Whereas in the MEG group, it was not observed. In addition, the MEG group did not show significant increase in pulmonary vascular resistance and intrapulmonary shunt fraction. Lung wet/dry ratios, a marker of pulmonary edema, were significantly lower in the MEG group. At 48 h after injury, lung tissue-conjugated dienes, an index of lung oxidative tissue injury, were significantly lower in the MEG group than in the control group. Our data suggest that 1) iNOS-NO produced in the airway circulation plays a major role on the significant increase in airway blood flow, which may contribute to the spread of injury from injured airway to the lung parenchyma; 2) iNOS-NO induced in the pulmonary circulation contributes to the loss of hypoxic pulmonary vasoconstriction; and 3) iNOS-NO plays an important role on the lung oxidative tissue injury.

The effect of alpha(2) agonist-induced sedation and its reversal with an alpha(2) antagonist on organ blood flow in sheep.

We investigated changes in cardiac output and organ blood flow induced by medetomidine in sheep and determined changes in cardiac output and organ blood flow after reversal of medetomidine-induced sedation by atipamezole. Eight sheep were chronically instrumented. Medetomidine was infused IV to target plasma levels of 0, 0.8, 1.6, 3.2, 6.4, and 12.8 ng/mL for 25 min each, followed by a 5-min infusion of atipamezole. Hemodynamic values and organ blood flow (using colored microspheres) were measured just before medetomidine infusion (baseline), at the end of each medetomidine infusion step, and 30 min after the administration of atipamezole. Medetomidine (12. 8 ng/mL) decreased cardiac output from 6.3 +/- 1.0 to 3.2 +/- 0.7 L/min (P < 0.0001) and increased systemic vascular resistance from 1310 +/- 207 to 3467 +/- 1299 dynes. s(-1). cm(-5) (P < 0.0001). Blood flow decreased in the cerebral cortex from 1.29 +/- 0.40 to 0. 66 +/- 0.12 mL. g(-1). min(-1) (P < 0.0001), left ventricle from 2. 11 +/- 0.61 to 1.40 +/- 0.40 mL. g(-1). min(-1) (P < 0.0001), kidney from 8.28 +/- 3.17 to 6.07 +/- 2.65 mL. g(-1). min(-1) (P < 0.0001), skin from 0.09 +/- 0.04 to 0.05 +/- 0.02 mL. g(-1). min(-1) (P < 0. 0001), intestine from 0.56 +/- 0.13 to 0.27 +/- 0.07 mL. g(-1). min(-1) (P < 0.0001), and skeletal muscle from 0.28 +/- 0.15 to 0.04 +/- 0.01 mL. g(-1). min(-1) (P < 0.0001). Blood flow in the liver (hepatic artery) increased from 0.05 +/- 0.03 to 0.24 +/- 0.16 mL. g(-1). min(-1) (P < 0.0001). After atipamezole infusion, cardiac output and systemic vascular resistance returned to baseline, but the cerebral cortex, left ventricle, and renal blood flows remained below baseline at 0.89 +/- 0.22, 1.37 +/- 0.50, and 6.25 +/- 2.76 mL. g(-1). min(-1), respectively; skeletal muscle blood flow increased above baseline to 0.44 +/- 0.27 mL. g(-1). min(-1), spleen blood flow decreased below baseline to 1.65 +/- 0.61 mL. g(-1). min(-1) (P < 0.0001), and liver, intestine, and lung blood flows returned to baseline values. In conclusion, medetomidine decreased and redistributed organ blood flow in sheep. Atipamezole reversed the medetomidine-induced hemodynamic changes, but redistributed blood flow from the brain, heart, and kidney to the skeletal muscle.

Hypertonic saline dextran produces early (8-12 hrs) fluid sparing in burn resuscitation: a 24-hr prospective, double-blind study in sheep.

OBJECTIVE: Resuscitation of large burn injuries must quickly restore and maintain cardiovascular function and fluid balance while minimizing secondary edema-related damage. We tested the hypothesis that two 4-mL x kg(-1) doses of hypertonic saline dextran (HSD; 7.5% NaCl/6% dextran-70) can produce prolonged reduction in fluid requirements after burn injury. DESIGN: Prospective, pseudo randomized, double-blind study. SETTING: Animal research laboratory. SUBJECTS: Female adult Merino sheep (n = 12). INTERVENTIONS: Sheep were given a 40% total body surface area full-thickness flame burn under halothane anesthesia. One hour after the burn, the conscious animals received an initial dose of 4 mL x kg(-1) HSD (n = 6) or normal saline (NS; NaCl 0.9%) (n = 6) intravenously during 30 mins. This was followed by lactated Ringer's solution, infused to a target urine output of 1 mL x kg(-1) x hr(-1) throughout the 24-hr study. A second 4-mL x kg(-1) dose of HSD or NS was started at 12 hrs, and infused during 5 hrs. MEASUREMENTS AND MAIN RESULTS: Hourly urine output measurements were used to guide the infusion rate of the lactated Ringer's. The initial infusion of HSD 1 hr after the burn injury promptly restored cardiac index, promoted diuresis, and reduced fluid requirements compared with the NS controls (73% reduction for HSD relative to NS at 8 hrs). Subsequent rebound fluid accumulation resulted in similar net fluid balances in both groups within 12 hrs after the burn. The second dose of HSD, given at 12 hrs, was without effect on hemodynamics and fluid balance. CONCLUSIONS: We conclude a considerable initial, but not sustained fluid-sparing effect of early HSD, and no effect of a late, slowly infused HSD dose in this two-dose regimen.

Noradrenaline and nomega-monomethyl-L-arginine (L-NMMA): effects on haemodynamics and regional blood flow in healthy and septic sheep.

This prospective, non-randomized, controlled experimental study looks at the effects of N(omega)-monomethyl-L-arginine (L-NMMA) on haemodynamics, oxygen transport and regional blood flow in healthy and septic sheep, and compares these effects with those of noradrenaline (NA; norepinephrine). All sheep were chronically instrumented. Six sheep received L-NMMA (7 mg.kg(-1).h(-1)), six sheep received NA, and seven sheep received the carrier alone (0.9% NaCl). The NA dosage was continuously and individually adjusted to achieve the same increase in blood pressure as observed in matched sheep of the L-NMMA group (non-septic phase). Treatment was discontinued after 3 h. Sepsis was initiated and maintained by a continuous infusion of live Pseudomonas aeruginosa. After 24 h of sepsis, the sheep were again challenged over a treatment period of 3 h with their previously assigned drug (septic phase). During the non-septic phase of the experiment, NA and L-NMMA both caused an increase in mean arterial pressure (MAP) through vasoconstriction. Ater 24 h of sepsis, all sheep developed a hyperdynamic circulatory state. While L-NMMA caused an increase in MAP through intense vasoconstriction, NA caused MAP to increase through a further elevation of the cardiac index. The NA dosage needed was significantly higher in the septic phase compared with the non-septic phase, reflecting a reduced vascular responsiveness to catecholamines during sepsis. Renal blood flow remained unchanged during either treatment in both the non-septic and the septic phases. Nevertheless, urine output increased during NA treatment in both the non-septic and the septic phases, while L-NMMA caused urine output to increase only under septic conditions.

Role of L-selectin in physiological manifestations after burn and smoke inhalation injury in sheep.

The effects of a monoclonal antibody against L-selectin [leukocyte adhesion molecule (LAM)1-3] on microvascular fluid flux were determined in conscious sheep subjected to a combined injury of 40% third-degree burn and smoke inhalation. This combined injury induced a rapid increase in systemic prefemoral lymph flow (sQlymph) from the burned area and a delayed-onset increase in lung lymph flow. The initial increase in sQlymph was associated with an elevation of the lymph-to-plasma oncotic pressure ratio; consequently, it leads to a predominant increase in the systemic soft tissue permeability index (sPI). In an untreated control group, the increased sPI was sustained beyond 24 h after injury. Pretreatment with LAM1-3 resulted in earlier recovery from the increased sPI, although the initial responses in sQlymph and sPI were identical to those in the nontreatment group. The delayed-onset lung permeability changes were significantly attenuated by pretreatment with LAM1-3. These findings indicate that both leukocyte-dependent and -independent mechanisms are involved in the pathogenesis that occurs after combined injury with burn and smoke inhalation.

Selective inhibition of inducible nitric oxide synthase: effects on hemodynamics and regional blood flow in healthy and septic sheep.

OBJECTIVES: To investigate the effects of S-ethylisothiourea (S-EITU) on hemodynamics, oxygen transport, and regional blood flow in healthy and septic sheep. DESIGN: Prospective, randomized, controlled experimental study with repeated measures. SETTING: Investigational intensive care unit at a university medical center. SUBJECTS: Eleven healthy, female adult sheep of the Merino breed, divided into a control group (n = 5) and into a group treated with S-EITU (n = 6). INTERVENTIONS: All sheep were chronically instrumented. After a 5-day recovery period, they were randomly assigned to either control or S-EITU groups. While control sheep received only saline, S-EITU was administered in increasing doses of 1, 3, and 9 mg/kg/hr over 1 hr each (nonseptic phase). After 2 days of recovery, a continuous infusion of live Pseudomonas aeruginosa (2.5 x 106 colony-forming units/min) was started in all sheep and maintained for the remainder of the experiment. After 24 hrs of sepsis, the sheep again received their assigned treatment (septic phase). In both the nonseptic and septic phases, the sheep received colored microspheres through a left atrial catheter to allow analysis of regional blood flows. All animals were autopsied at the end of the experiments, and organ probes were removed for blood flow analyses. MEASUREMENTS AND MAIN RESULTS: The administration of S-EITU caused a dose-dependent vasoconstriction in the nonseptic phase. After 24 hrs of Pseudomonas infusion, all sheep developed a hyperdynamic circulatory state, with increased cardiac indices and reduced arterial pressures and systemic vascular resistances. Oxygen extraction decreased significantly, preventing an increase in oxygen consumption, despite an increased oxygen delivery. The hyperdynamic circulation was dose dependently reversed by S-EITU, causing an increase in arterial pressure by peripheral vasoconstriction. Sheep in the control group showed a continuation of the hyperdynamic circulation. The effects of S-EITU on hemodynamics and regional blood flows were comparable under septic and nonseptic conditions. CONCLUSIONS: With the inducible form of nitric oxide synthase expressed under septic, but not under nonseptic conditions, S-EITU was expected to have vasoconstrictive properties only in the septic phase. It produced a comparable vasoconstriction during the nonseptic phase of the experiment. Thus, either S-EITU does not selectively block the inducible nitric oxide synthase in sheep, or other vasodilators besides nitric oxide play an important role in septic vasodilation.

Bronchoscopic lavage with perfluorocarbon decreases postprocedure hypoxemia in an ovine model of smoke inhalation.

OBJECTIVE: Bronchoscopy and lavage are used to confirm diagnosis and can be therapeutic in patients suffering inhalation injury. Lavage is traditionally performed using saline, which is unfortunately associated with profound transient hypoxemia. Perfluorocarbons, having a high gas solubility for oxygen and carbon dioxide, increase oxygenation when instilled into the airway. We hypothesized that the use of perfluorocarbons for bronchoscopic lavage would attenuate this transient hypoxemia. METHODS: Sheep were prepared for chronic study. They were insufflated with cotton smoke and then randomized to receive a lavage with 200 mL of perfluorocarbon or saline at 2, 6, 12, and 24 hours after injury. RESULTS: All animals had a steady and significant decline in their pre- to post-PaO2/FiO2 (P/F) ratio. At 2, 6, and 12 hours, the saline lavage group had a significant decrease in their P/F ratio (485+/-32 to 212+/-37 mm Hg, 439+/-22 to 170+/-40 mm Hg, and 381+/-48 to 184+/-59 mm Hg). This decrease in P/F ratio was not observed in the perfluorocarbon group (474+/-19 to 459+/-29 mm Hg, 424+/-32 to 387+/-43 mm Hg, and 366+/-50 to 357+/-67 mm Hg). CONCLUSION: These findings indicate that perfluorocarbons attenuate the transient hypoxemia associated with saline bronchoscopic lavage and thus may be considered safer for patients with acute lung injury.