Liver Transplantation Using Allografts With Recent Liver Blunt Trauma: A Nationwide Audit From the French CRISTAL Biomedicine Agency Registry.

BACKGROUND: In the current setting of organ shortage, brain-dead liver donors with recent liver trauma (RLT) represent a potential pool of donors. Yet, data on feasibility and safety of liver transplantation (LT) using grafts with RLT are lacking. METHODS: All liver grafts from brain-dead donors with RLT proposed for LT between 2010 and 2018 were identified from the nationwide CRISTAL registry of the Biomedicine Agency. The current study aimed at evaluating 1-y survival as the primary endpoint. RESULTS: Among 11 073 LTs, 142 LTs (1.3%) using grafts with RLT were performed. These 142 LTs, including 23 split LTs, were performed from 131 donors (46.1%) of 284 donors with RLT proposed for LT. Transplanted grafts were procured from donors with lower liver enzymes levels ( P < 0.001) and less advanced liver trauma according to the American Association for the Surgery of Trauma liver grading system ( P < 0.001) compared with not transplanted grafts. Before allocation procedures, 20 (7%) of 284 donors underwent damage control intervention. During transplantation, specific liver trauma management was needed in 19 patients (13%), consisting of local hemostatic control (n = 15), partial hepatic resection on back-table (n = 3), or perihepatic packing (n = 1). Ninety-day mortality and severe morbidity rates were 8.5% (n = 12) and 29.5% (n = 42), respectively. One-year overall and graft survival rates were 85% and 81%, and corresponding 5-y rates were 77% and 72%, respectively. CONCLUSIONS: Using liver grafts from donors with RLT seems safe with acceptable long-term outcomes. All brain-dead patients with multiorgan trauma, including liver injury, should be considered for organ allocation.

Optimal donation of kidney transplants after controlled circulatory death.

Controlled donation after circulatory death (cDCD) is used for "extended criteria" donors with poorer kidney transplant outcomes. The French cDCD program started in 2015 and is characterized by normothermic regional perfusion, hypothermic machine perfusion, and short cold ischemia time. We compared the outcomes of kidney transplantation from cDCD and brain-dead (DBD) donors, matching cDCD and DBD kidney transplants by propensity scoring for donor and recipient characteristics. The matching process retained 442 of 499 cDCD and 809 of 6185 DBD transplantations. The DGF rate was 20% in cDCD recipients compared with 28% in DBD recipients (adjusted relative risk [aRR], 1.43; 95% confidence interval [CI] 1.12-1.82). When DBD transplants were ranked by cold ischemia time and machine perfusion use and compared with cDCD transplants, the aRR of DGF was higher for DBD transplants without machine perfusion, regardless of the cold ischemia time (aRR with cold ischemia time <18 h, 1.57; 95% CI 1.20-2.03, vs aRR with cold ischemia time ≥18 h, 1.79; 95% CI 1.31-2.44). The 1-year graft survival rate was similar in both groups. Early outcome was better for kidney transplants from cDCD than from matched DBD transplants with this French protocol.

Effect of Lung-Protective Ventilation in Organ Donors on Lung Procurement and Recipient Survival.

Rationale: Previous studies have shown that a lung-protective strategy, which aims at minimizing ventilator-induced lung injury (with low Vt/high positive end-expiratory pressure as the main pillars), in selected potential organ donors after brain death increased lung eligibility and procurement.Objectives: This prospective nationwide cohort study aimed to evaluate the impact of lung-protective ventilation (PV) in nonselected donors on lung procurement and recipient survival after lung transplantation.Methods: We included all reported donors aged 18-70 years after brain death without a lung recovery contraindication and with at least one organ recovered between January 2016 and December 2017. PV was defined as Vt ≤8 ml/kg predicted body weight and positive end-expiratory pressure ≥8 cm H2O. The association between PV at the time of lung proposal (T1) and lung procurement was determined by multivariable logistic regression stratified by propensity score quintile to account for PV and non-PV group differences in baseline characteristics. We studied 1-year survival of recipients from donors with or without PV at T1.Measurements and Main Results: Of 1,626 included lung donors, 1,109 (68%) had at least one lung proposed; 678 (61%) of these had at least one lung recovered. At T1, only 25.6% of donors with at least one lung proposed for lung transplantation were ventilated with a protective strategy. For donors with a lung proposal, the probability of lung procurement was increased with PV at T1 (odds ratio, 1.43; 95% confidence interval [CI], 1.03-1.98; P = 0.03). One-year survival did not differ between recipients of lungs from donors with and without PV (82.7%, 95% CI 76.0-87.8% vs. 82.3%, 95% CI 78.5-85.4%; P = 0.94).Conclusions: The use of lung PV in nonselected donors may increase lung procurement. One-year survival did not differ between recipients of lungs from donors with PV or from those without PV.

Kidney Transplant From Uncontrolled Donation After Circulatory Death: Contribution of Normothermic Regional Perfusion.

BACKGROUND: The French uncontrolled donors after circulatory death (DCD) protocol restricts donor age to <55 years, no-flow time to <30 minutes, and functional warm ischemia time to <150 minutes. In situ kidney perfusion can be performed at either 4°C (in situ cooling [ISC]) or 33-36°C (normothermic regional perfusion [NRP]). Hypothermic machine perfusion is systematically used. Only nonimmunized first transplant recipients were eligible. To improve the management of uncontrolled DCD, we tried to identify factors predictive of outcome. METHODS: We identified all kidney transplants from uncontrolled DCD between 2007 and 2014 from the French Transplant Registry. Risk factors for primary nonfunction (PNF; n = 37) and poor renal function (estimated glomerular filtration rate < 30 mL/min or graft loss at 1 y, n = 66) were analyzed by using a multivariate logistic model. RESULTS: This study analyzed 499 kidney transplantations, 50% of which were performed with NRP. Mean functional warm ischemia time was 135 minutes. Mean cold ischemia time was 14 hours. The principal PNF risk factor was young donor age (odds ratio [OR] = 0.95; P = 0.002). A sensitivity analysis showed a higher risk of PNF with ISC than with NRP (OR = 4.5; P = 0.015). Risk factors for poor renal function were donor body mass index (OR = 1.2; P < 0.001) and ISC versus NRP. Univariate analysis of uncontrolled DCD-specific risk factors showed no-flow time, functional warm time, and cold ischemia time did not affect the risk of PNF or poor renal function. CONCLUSIONS: Uncontrolled DCD kidneys are an additional source of valuable transplants. NRP appears to decrease graft failure by restoring oxygenated blood as the first step of preconditioning.

Being Convinced and Taking Responsibility: A Qualitative Study of Family Members' Experience of Organ Donation Decision and Bereavement After Brain Death.

OBJECTIVES: Family members of brain dead patients experience an unprecedented situation in which not only they are told that their loved one is dead but are also asked to consider organ donation. The objective of this qualitative study was to determine 1) what it means for family members to make the decision and to take responsibility, 2) how they interact with the deceased patient in the ICU, 3) how family members describe the impact of the process and of the decision on their bereavement process. DESIGN: Qualitative study using interviews with bereaved family members who were approached for organ donation after the death of their relative in the ICU (brain death). SETTING: Family members from 13 ICUs in France. SUBJECTS: Bereaved family members who were approached for organ donation after the death of their relative in the ICU (brain death). INTERVENTION: None. MEASUREMENTS AND RESULTS: Twenty-four interviews were conducted with 16 relatives of organ donor patients and with eight relatives of nonorgan donor patients. Three themes emerged: 1) taking responsibility-relatives explain how they endorse decisional responsibility but do not experience it as a burden, on the contrary; 2) ambiguous perceptions of death-two groups of relatives emerge: those for whom ambiguity hinders their acceptance of the patient's death; those for whom ambiguity is an opportunity to accept the death and say goodbye; and 3) donation as a comfort during bereavement. CONCLUSIONS: In spite of caregivers' efforts to focus organ donation discussions and decision on the patient, family members feel a strong decisional responsibility that is not experienced as a burden but a proof of their strong connection to the patient. Brain death however creates ambivalent experiences that some family members endure whereas others use as an opportunity to perform separation rituals. Last, organ donation can be experienced as a form of comfort during bereavement provided family members remain convinced their decision was right.

Perioperative management of adult diabetic patients. Preoperative period.

In diabetic patients undergoing surgery, we recommend assessing glycaemic control preoperatively by assessing glycated haemoglobin (HbA1c) levels and recent capillary blood sugar (glucose) levels, and to adjust any treatments accordingly before surgery, paying particular attention to specific complications of diabetes. Gastroparesis creates a risk of stasis and aspiration of gastric content at induction of anaesthesia requiring the use of a rapid sequence induction technique. Cardiac involvement can be divided into several types. Coronary disease is characterised by silent myocardial ischaemia, present in 30-50% of T2D patients. Diabetic cardiomyopathy is a real cause of heart failure. Finally, cardiac autonomic neuropathy (CAN), although rarely symptomatic, should be investigated because it causes an increased risk of cardiovascular events and a risk of sudden death. Several signs are suggestive of CAN, and confirmation calls for close perioperative surveillance. Chronic diabetic kidney disease (diabetic nephropathy) aggravates the risk of perioperative acute renal failure, and we recommend measurement of the glomerular filtration rate preoperatively. The final step of the consultation concerns the management of antidiabetic therapy. Preoperative glucose infusion is not necessary if the patient is not receiving insulin. Non-insulin drugs are not administered on the morning of the intervention except for metformin, which is not administered from the evening before. The insulins are injected at the usual dose the evening before. The insulin pump is maintained until the patient arrives in the surgical unit. It should be remembered that insulin deficiency in a T1D patient leads to ketoacidosis within a few hours.

Perioperative management of adult diabetic patients. Review of hyperglycaemia: definitions and pathophysiology.

Diabetes mellitus is defined by chronic elevation of blood glucose linked to insulin resistance and/or insulinopaenia. Its diagnosis is based on a fasting blood-glucose level of ≥1.26g/L or, in some countries, a blood glycated haemoglobin (HbA1c) level of >6.5%. Of the several forms of diabetes, type-2 diabetes (T2D) is the most common and is found in patients with other risk factors. In contrast, type-1 diabetes (T1D) is linked to the autoimmune destruction of ß-pancreatic cells, leading to insulinopaenia. Insulin deficiency results in diabetic ketoacidosis within a few hours. 'Pancreatic' diabetes develops from certain pancreatic diseases and may culminate in insulinopaenia. Treatments for T2D include non-insulin based therapies and insulin when other therapies are no longer able to control glycaemic levels. For T1D, treatment depends on long (slow)-acting insulin and ultra-rapid analogues of insulin administered according to a 'basal-bolus' scheme or by continuous subcutaneous delivery of insulin using a pump. For patients presenting with previously undiagnosed dysglycaemia, investigations should determine whether the condition corresponds to pre-existing dysglycaemia or to stress hyperglycaemia. The latter is defined as transient hyperglycaemia in a previously non-diabetic patient that presents with an acute illness or undergoes an invasive procedure. Its severity depends on the type of surgery, the aggressiveness of the procedure and its duration. Stress hyperglycaemia may lead to peripheral insulin resistance and is an independent prognostic factor for morbidity and mortality.

Grief Symptoms in Relatives Who Experienced Organ Donation Requests in the ICU.

RATIONALE: Studies show that the quality of end-of-life communication and care have a significant impact on the living long after the death of a relative and have been implicated in the burden of psychological symptoms after the ICU experience. In the case of organ donation, the patient's relatives are centrally involved in the decision-making process; yet, few studies have examined the impact of the quality of communication on the burden of psychological symptoms after death. OBJECTIVES: To assess the experience of the organ donation process and grief symptoms in relatives of brain-dead patients who discussed organ donation in the ICU. METHODS: We conducted a multicenter longitudinal study in 28 ICUs in France. Participants were the relatives of brain-dead patients who were approached to discuss organ donation. Relatives were followed-up by phone at three time points: at 1 month, to complete a questionnaire describing their experience of the organ donation process; at 3 months, to complete the Hospital Anxiety and Depression Scale and the Impact of Event Scale-Revised; and at 9 months, to complete the Impact of Event Scale-Revised and the Inventory of Complicated Grief. MEASUREMENTS AND MAIN RESULTS: In total, 202 relatives of 202 patients were included, of whom 158 consented to and 44 refused organ donation. Interviews were conducted at 1, 3, and 9 months with 78%, 68%, and 58% of relatives, respectively. The overall experience of the organ donation process was significantly more burdensome for relatives of nondonors. They were more dissatisfied with communication (27% vs. 10%; P = 0.021), more often shocked by the request (65% vs. 19%; P < 0.0001), and more often found the decision difficult (53% vs. 27%; P = 0.017). However, there were no significant differences in grief symptoms measured at 3 and 9 months between the two groups. Understanding of brain death was associated with grief symptoms; our results show a higher prevalence of complicated grief symptoms among relatives who did not understand the brain death process than among those who did (75% vs. 46.1%; P = 0.026). CONCLUSIONS: Experience of the organ donation process varied between relatives of donor versus nondonor patients, with relatives of nondonors experiencing lower-quality communication, but the decision was not associated with subsequent grief symptoms. Importantly, understanding of brain death is a key element of the organ donation process for relatives.

Perioperative management of adult diabetic patients. Intraoperative period.

Perioperative hyperglycaemia (>1.80g/L or 10mmol/L) increases morbidity (particularly due to infection) and mortality. Hypoglycaemia can be managed in the perioperative period by decreasing blood sugar levels with insulin between 0.90 and 1.80g/L but it may occur more frequently when the goal is strict normoglycaemia. We propose continuous administration of insulin therapy via an electronic syringe (IVES) in type-1 diabetes (T1D) and type-2 diabetes (T2D) patients if required or in cases of stress hyperglycaemia. Stopping a personal insulin pump requires immediate follow on with IVES insulin. We recommend 4mg dexamethasone for the prophylaxis of nausea and vomiting, rather than 8mg, combined with another antiemetic drug. The use of regional anaesthesia (RA), when possible, allows for better control of postoperative pain and should be prioritised. Analgesic requirements are higher in patients with poorly controlled blood sugar levels than in those with HbA1c<6.5%. The struggle to prevent hypothermia, the use of RA and multimodal analgesia (which allow for a more rapid recovery of bowel movements), limitation of blood loss, early ambulation and minimally invasive surgery are the preferred measures to regulate perioperative insulin resistance. Finally, diabetes does not change the usual rules of fasting or of antibiotic prophylaxis.

Perioperative management of adult diabetic patients. Specific situations.

Ambulatory surgery can be carried out in diabetic patients. By using a strict organisational and technical approach, the risk of glycaemic imbalance is minimised, allowing the patients to return to their previous way of life more quickly. Taking into account the context of ambulatory surgery, with a same day discharge, the aims are to minimise the changes to antidiabetic treatment, to maintain adequate blood sugar control and to resume oral feeding as quickly as possible. The preoperative evaluation is the same as for a hospitalised patient and recent glycaemic control (HbA1c) is necessary. Perioperative management and the administration of treatment depend on the number of meals missed. The patient can return home after taking up usual feeding and treatment again. Hospitalisation is necessary if significant glycaemic imbalance occurs. In pregnancy, it is necessary to distinguish between known pre-existing diabetes (T1D or T2D) and gestational diabetes, defined as glucose intolerance discovered during pregnancy. During labour, blood sugar levels should be maintained between 0.8 and 1.4g/L (4.4-8.25mmol/L). Control of blood sugar levels is obtained by using a continuous administration of insulin using an electronic syringe (IVES) together with a glucose infusion. Post-partum, management depends on the type of diabetes: in T1D and T2D patients a basal-bolus scheme is restarted with decreased doses while in gestational diabetes insulin therapy is stopped after delivery. Antidiabetic treatment is again necessary if blood sugar levels remain>1.26g/L (7mmol/L).

Perioperative management of adult diabetic patients. Postoperative period.

Follow on from continuous intravenous administration of insulin with an electronic syringe (IVES) is an important element in the postoperative management of a diabetic patient. The basal-bolus scheme is the most suitable taking into account the nutritional supply and variable needs for insulin, reproducing the physiology of a normal pancreas: (i) slow (long-acting) insulin (=basal) which should immediately take over from IVES insulin simulating basal secretion; (ii) ultra-rapid insulin to simulate prandial secretion (=bolus for the meal); and (iii) correction of possible hyperglycaemia with an additional ultra-rapid insulin bolus dose. A number of schemes are proposed to help calculate the dosages for the change from IV insulin to subcutaneous insulin and for the basal-bolus scheme. Postoperative resumption of an insulin pump requires the patient to be autonomous. If this is not the case, then it is mandatory to establish a basal-bolus scheme immediately after stopping IV insulin. Monitoring of blood sugar levels should be continued postoperatively. Hypoglycaemia and severe hyperglycaemia should be investigated. Faced with hypoglycaemia <3.3mmol/L (0.6g/L), glucose should be administered immediately. Faced with hyperglycaemia >16.5mmol/L (3g/L) in a T1D or T2D patient treated with insulin, investigations for ketosis should be undertaken systematically. In T2D patients, unequivocal hyperglycaemia should also call to mind the possibility of diabetic hyperosmolarity (hyperosmolar coma). Finally, the modalities of recommencing previous treatments are described according to the type of hyperglycaemia, renal function and diabetic control preoperatively and during hospitalisation.

Perioperative management of adult diabetic patients. The role of the diabetologist.

A patient should be referred to a diabetologist perioperatively in several circumstances: preoperative recognition of a previously unknown diabetes or detection of glycaemic imbalance (HbA1c <5% or >8%); during hospitalisation, recognition of a previously unknown diabetes, persisting glycaemic imbalance despite treatment or difficulty resuming previously used chronic treatment; postoperatively and after discharge from hospital, for all diabetic patients in whom HbA1c is >8%.

Physiological and self-assessed psychological stress induced by a high fidelity simulation course among third year anesthesia and critical care residents: An observational study.

INTRODUCTION: The use of high fidelity simulators in Medicine can improve knowledge, behaviour and practice but may be associated with significant stress. Our objective was to measure physiological and psychological self-assessed intensity of stress before and after a planned simulation training session among third year anaesthesia and critical care residents. METHODS: A convenience sample of 27 residents participating in a simulation training course was studied. Stress was evaluated by self-assessment using a numerical scale and by salivary amylase concentration before and after the session. Technical and non-technical (using the Aberdeen Anaesthetists' Non Technical Skills scale) performances were assessed through videotapes analysis. RESULTS: The median stress score was 5 (2-8) before and 7 (2-10) after the simulation session (P<0.001). For 48% of residents studied, the stress score after the session was superior or equal to 8/10. Salivary amylase concentration increased significantly after the session compared to before the session, respectively (1,250,440±1,216,667 vs. 727,260±603,787IU/L, P=0.008). There was no significant correlation between stress parameters and non-technical performance. DISCUSSION: Simulation-induced stress, as measured by self-assessment and biological parameter, is high before the session and increases significantly during the course. While this stress did not seem to impact performance negatively, it should be taken into account.

Uncontrolled donation after circulatory death: European practices and recommendations for the development and optimization of an effective programme.

The shortage of organs remains one of the biggest challenges in transplantation. To address this, we are increasingly turning to donation after circulatory death (DCD) donors and now in some countries to uncontrolled DCD donors. We consolidate the knowledge on uncontrolled DCD in Europe and provide recommendations and guidance for the development and optimization of effective uncontrolled DCD programmes.

The limits of succinylcholine for critically ill patients.

BACKGROUND: Urgent tracheal intubations are common in intensive care units (ICU), and succinylcholine is one of the first-line neuromuscular blocking drugs used in these situations. Critically ill patients could be at high risk of hyperkalemia after receiving succinylcholine because one or more etiologic factors of nicotinic receptor upregulation can be present, but there are few data on its real risk. Our objectives in this study were to determine the factors associated with arterial potassium increase (ΔK) and to assess the occurrence of acute hyperkalemia ≥6.5 mmol/L after succinylcholine injection for intubation in the ICU. METHODS: In a prospective, observational study, all critically ill patients intubated with succinylcholine in an ICU were screened. Only intubations with arterial blood gases and potassium measurements before and after (K(after)) a succinylcholine injection were studied. RESULTS: During 18 months, 131 critically ill patients were intubated after receiving succinylcholine with arterial potassium before and after intubation (K(after)) for a total of 153 intubations. After multivariate analysis, the only factor associated with ΔK was the length of ICU stay before intubation (ρ = 0.561, P < 0.001). The factors associated with K(after) ≥6.5 mmol/L (n = 11) were the length of ICU stay (P < 0.001) and the presence of acute cerebral pathology (P = 0.047). The threshold of 16 days was found highly predictive of acute hyperkalemia ≥6.5 with 37% (95% confidence interval: 19%-58%) of K(after) ≥6.5 after the 16th day compared with only 1% (95% confidence interval: 0%-4%) of K(after) ≥6.5 when succinylcholine was injected during the first 16 days. CONCLUSIONS: This study shows that the risk of ΔK after succinylcholine injection is strongly associated with the length of ICU stay. The risk of acute hyperkalemia ≥6.5 mmol/L is highly significant after 16 days.

Renal arterial resistance in septic shock: effects of increasing mean arterial pressure with norepinephrine on the renal resistive index assessed with Doppler ultrasonography.

OBJECTIVE: To determine the effects of increasing mean arterial pressure (MAP) on renal resistances assessed by Doppler ultrasonography in septic shock. DESIGN AND SETTING: Prospective, single-center, nonrandomized, open-label trial in the surgical intensive care unit in a university teaching hospital. PATIENTS AND PARTICIPANTS: 11 patients with septic shock who required fluid resuscitation and norepinephrine to increase and maintain MAP at or above 65 mmHg. INTERVENTIONS: Norepinephrine was titrated in 11 patients in septic shock during three consecutive not randomized periods of 2 h to achieve a MAP at successively 65, 75, and 85 mmHg. MEASUREMENTS AND RESULTS: At the end of each period hemodynamic parameters and renal function variables (urinary output, creatinine, clearance) were measured, and Doppler ultrasonography was performed on interlobar arteries to assess the renal resistive index. When increasing MAP from 65 to 75 mmHg, urinary output increased significantly from 76 +/- 64 to 93 +/- 68 ml/h and the resistive index significantly decreased from 0.75 +/- 0.07 to 0.71 +/- 0.06. No difference was found between 75 and 85 mmHg. CONCLUSIONS: Doppler ultrasonography and resistive index measurements may help determine in each patient the optimal MAP for renal blood flow and may be a relevant end-point to titrate the hemodynamic treatment in septic shock.

Clinical review: initial management of blunt pelvic trauma patients with haemodynamic instability.

Pelvic trauma can lead to severe, uncontrollable haemorrhage and death related to prolonged shock and multiple organ failure. Massive retroperitoneal haematoma should be assumed to be present in cases of post-traumatic haemodynamic instability associated with pelvic fracture in the absence of extrapelvic haemorrhagic lesions. This review describes the pathophysiology of retroperitoneal haematoma in trauma patient with blunt pelvic fracture, considering the roles of venous and arterial bleeding. Efficacy and safety of haemostatic procedures are also discussed, and particular attention is given to the efficacy of pelvic angiographic embolization and external pelvic fixation. A decision making algorithm is proposed for the treatment of trauma patients with pelvic fracture that takes haemodynamic status and associated lesions into account.

Very early steroid withdrawal in simultaneous pancreas-kidney transplants.

BACKGROUND: Simultaneous pancreas-kidney (SPK) transplantation is an effective treatment for patients suffering from type 1 diabetes mellitus. Conventional immunosuppressive treatments include steroids that may induce insulin resistance and are responsible for many side effects. In de novo SPK, early withdrawal of corticosteroids may be an important issue. METHODS: A total of 24 consecutive patients with type 1 diabetes mellitus had been treated by SPK transplantation. All of them had a short induction therapy with anti-thymoglobulin (ATG) and steroids for only 4 days, association with CellCept and tacrolimus. The rate of acute rejection, graft and patient survival and side effects have been analysed. RESULTS: Patient and kidney survival was 100% and the pancreas survival was 95.6% at 1 year. The rate of acute rejection of kidney and pancreas was 4.2% and 8.3% at 6 months, respectively. The mean serum creatinine was 98.9+/-19.6 micromol/l and the mean HbA1c concentration was 5.1%+/-0.5% at 6 months. Only four patients developed a cytomegalovirus primo-infection, associated in one case with pneumonia, whereas 75% of patients developed a bacterial infection. Because of the occurrence of leucopenia and/or diarrhoea, CellCept has been dramatically decreased in 33% of cases and required the re-introduction of steroids. CONCLUSION: A short induction with ATG and steroids associated with a chronic therapy with CellCept and tacrolimus is safe and efficient in preventing acute renal rejection in SPK.

Beta2-adrenergic agonist protects human endothelial cells from hypoxia/reoxygenation injury in vitro.

OBJECTIVE: Circulatory shock results in hypoxia/reoxygenation processes that lead to the release of reactive oxygen species, endothelial injury, and multiple organ failure. Previous data suggest that beta2-adrenergic agonists prevent endothelial dysfunction. The study aimed at determining whether the beta2-adrenergic agonist formoterol protects endothelial cells against hypoxia/reoxygenation injury in vitro. DESIGN: Prospective controlled trial. SETTING: University hospital research laboratory. SUBJECTS: Cultured human umbilical vein endothelial cells (HUVECs). INTERVENTIONS: Confluent HUVECs were sealed in a flow-through chamber mounted on an inverted microscope and perfused with a constant flow of Krebs medium. After 1 hr of equilibration, HUVECs underwent 2 hrs of hypoxia and 1 hr of reoxygenation. Cell death at the end of reoxygenation and reactive oxygen species formation were assessed with fluorescent probes propidium iodide and 2',7'-dichlorodihydrofluorescein diacetate, respectively. The effects of the beta2-adrenergic agonist formoterol, the beta2-adrenergic antagonist ICI 118,551 and the nitric oxide synthase inhibitor L-NNA were investigated. Statistical analysis was performed with analysis of variance followed by post hoc Fisher's test. MEASUREMENTS AND MAIN RESULTS: Hypoxia/reoxygenation increased cell death (hypoxia/reoxygenation 29 +/- 4% vs. control 1 +/- 5%, p < .05) and endothelial reactive oxygen species production (hypoxia/reoxygenation 126 +/- 4% vs. control 108 +/- 4%, p < .05). Formoterol reduced cell death in a concentration-dependent manner (EC95 = 10 mol/L) and reduced endothelial reactive oxygen species production (hypoxia/reoxygenation + formoterol EC95 109 +/- 4% vs. hypoxia/reoxygenation 126 +/- 4%, p < .05). When added to formoterol EC95, ICI 118,551 and L-NNA abolished the formoterol-induced cell protection and reduced reactive oxygen species production. CONCLUSIONS: These results indicate that formoterol reduces endothelial cell death and reactive oxygen species production in this in vitro hypoxia/reoxygenation model. These effects are beta2-adrenergic specific and are partially mediated by nitric oxide synthase.