Intensive Care Medicine

Published by Springer Verlag
Online ISSN: 1432-1238
Print ISSN: 0342-4642
To examine variables associated with postextubation respiratory distress in chronic obstructive pulmonary disease (COPD) patients. Prospective, clinical investigation. Intensive care unit of a university hospital. Forty COPD patients, considered ready for extubation. We recorded, from the digital display of a standard ventilator, breathing frequency (f), tidal volume (VT) and f/VT for the respiratory pattern, airway occlusion pressure at 0.1 s (P0.1) for the respiratory drive and measured blood gases: i) before extubation, following 30 min of a 6 cm H2O pressure support (PS) ventilation trial, ii) 1 h after extubation, at the 30th min of a face mask 4 cm H2O PS ventilation trial. According to the weaning outcome, the patients were divided into two groups: respiratory distress, and non-respiratory distress within 72 h of the discontinuation of mechanical ventilation. The respiratory distress was defined as the combination of f more than 25 breaths/min, an increase in PaCO2 of at least 20% compared with the value measured after extubation, and pH lower than 7.35. We determined whether those patients who developed respiratory distress after extubation differed from those who did not. Respiratory pattern data and arterial blood gases recorded, either before or after extubation, and P0.1 recorded before extubation, were inadequate to differentiate the two groups. Only P0.1 recorded 1 h after the discontinuation of mechanical ventilation differentiated the patients who developed respiratory distress from those who did not (4.2+/-0.9 vs 1.8+/-0.8, p < 0.01). P0.1 recorded after extubation may be a good indicator of postextubation respiratory distress. Measuring P0.1 and/or the analysis of the evolution of this parameter could facilitate decisions during the period following extubation.
Characteristics of included studies
Assessment of quality and risk of bias Author Year Randomisation Allocation concealment Blinding Intention to treat analysis No loss to follow-up 
Forest plot of pooled estimates for mortality. 6 % HES 130 = 6 % hydroxyethyl starch with a molecular weight of 130 kD and a molar substitution ratio of approximately 0.4. CI = confidence interval. Studies reporting at least one event in each group are arranged in ascending year of publication. Weights are from random effects analysis
Forest plot of pooled estimates for need for renal replacement therapy. 6 % HES 130 = 6 % hydroxyethyl starch with a molecular weight of 130 kD and a molar substitution ratio of approximately 0.4. CI = confidence interval. Studies reporting at least one event in each group are arranged in ascending year of publication. Weights are from random effects analysis  
Purpose: To determine whether fluid resuscitation of acutely ill adults with 6 % hydroxyethyl starch (6 % HES 130) with a molecular weight of 130 kD and a molar substitution ratio of approximately 0.4 (6 % HES 130) compared with other resuscitation fluids results in a difference in the relative risk of death or treatment with renal replacement therapy (RRT). Methods: Systematic review and meta-analysis of randomized controlled trials comparing intravascular fluids for resuscitation of hospitalised adults that reported mortality or treatment with RRT. The risk of bias was assessed independently by two reviewers and meta-analysis was performed using random effects. Results: Thirty-five trials enrolling 10,391 participants were included. The three largest trials had the lowest risk of bias, were published (or completed) in 2012, and together enrolled 77 % of all participants. Death occurred in 928 of 4,691 patients (19.8 %) in the 6 % HES 130 group versus 871 of 4,720 (18.5 %) in the control fluid groups relative risk (RR) in the 6 % HES 130 group 1.08, 95 % confidence interval (CI) 1.00 to 1.17, I (2) = 0 %). Treatment with RRT occurred in 378 of 4,236 patients (8.9 %) in the 6 % HES 130 group versus 306 of 4,260 (7.2 %) in the control fluid group (RR in the 6 % HES 130 group 1.25, 95 % CI 1.08 to 1.44, I (2) = 0 %). Conclusions: The quality and quantity of data evaluating 6 % hydroxyethyl starch (130/0.4 and 130/0.42) as a resuscitation fluid has increased in the last 12 months. Patients randomly assigned to resuscitation with 6 %HES 130 are at significantly increased risk of being treated with RRT.
To compare the colloids 5% albumin, 4% gelatin, and 6% HES 130/0.4 with one another and with normal saline regarding their plasma expanding effects at increased permeability and to compare the results with those from a previous study at normal permeability. Prospective controlled randomized laboratory study in a university research laboratory. 48 adult male Sprague-Dawley rats. Permeability was increased by an injection of 0.5 ml dextran 70 using the fact that dextran causes anaphylactic reaction in the rat. Plasma volume was determined ((125)I albumin tracer technique) after anesthesia, 1 h after dextran injection (before infusion for 10-15 min of 20 ml/kg bw of each of the colloids or 80 ml/kg saline), and 3 h later. Blood pressure, hematocrit, blood gases, and electrolytes were measured. CVP was measured in four rats. Plasma volume was 41.1+/-1.9 ml/kg at baseline (n=9), and 29.1+/-4.1 ml/kg (n=35) 1 h after the dextran injection. Three hours after infusion of the plasma expander plasma volume had increased by 17.1+/-3.4 ml/kg in the albumin group, 7.9+/-3.6 ml/kg in the gelatin group, 7.4+/-4.4 ml/kg in the HES group, and 12.2+/-3.1 ml/kg in the saline group. It was unchanged in a control group given no solution (n=7 for all groups). Albumin was a more effective plasma volume expander than gelatin or HES or saline (saline in 4 times larger volume). Gelatin and HES were equally effective. All solutions showed a smaller plasma expanding effect than observed in a previous study with normal permeability.
Characteristics of studies with (A) in vitro hemodilution, (B) in vivo hemodilution 
Outcomes of in vitro hemodilution: (A) TEG, (B) ROTEM, (C) Sonoclot 
Dose effects of hemodilution. Low dose defined as \40% hemodilution in vitro (range 10-33%) or \28 ml/kg (range 7-27 ml/kg) administered as in vivo hemodilution. High dose defined as C40% hemodilution in vitro (range 40-80%) and C28 ml/kg (range 28-63 ml/kg) 
Hydroxyethyl starch solutions (HES) are plasma volume expanders which affect hemostasis. Newer HES 130/0.4 is said to be safer. Reevaluation of published evidence is necessary after the recent retraction of studies. Systematic review of studies assessing HES 130/0.4 effects on hemostasis by thrombelastography (TEG, ROTEM) or Sonoclot (SCR) in comparison with crystalloid or albumin control fluids was performed. Only studies which provided statistical comparisons between study fluids were analyzed. Studies were divided into in vitro or in vivo hemodilution studies. We assessed study quality, HES effects which differed significantly from controls, values outside normal range, degree of hemodilution, and cumulative HES dose. Seventeen in vitro and seven in vivo hemodilution studies were analyzed. Four studies reported quality control measures. Nineteen studies (all 15 ROTEM studies, 3 of 5 in vitro TEG, and 1 of 2 SCR studies) showed a significant hypocoagulatory effect of HES 130/0.4 on clot formation, while clotting time was not uniformly affected. Three in vivo TEG studies with low HES doses or cancer patients found mixed or nonsignificant results. In studies which provided normal ranges (n = 9), more values were outside normal ranges in the HES than in the control groups (87/122 vs. 58/122, p < 0.001). Dose effects were apparent in the in vitro studies, which investigated higher dilutions up to 80%. In vivo studies were fewer and did not investigate doses >40 ml/kg. HES 130/0.4 administration results in a weaker and smaller clot. Until results from well-designed clinical trials are available, safer fluids should be chosen for patients with impaired coagulation.
Purpose: To assess the impact of 6% tetrastarch [hydroxyethyl starch (HES) 130/0.4 and 130/0.42] in severe sepsis patients. The primary outcome measure was 90-day mortality. Methods: A structured literature search was undertaken to identify prospective randomised controlled trials (RCTs) in adult patients with severe sepsis receiving 6% tetrastarch (of potato or waxy maize origin) as part of fluid resuscitation in comparison with other non-HES fluids after randomisation in the critical care setting. A systematic review and meta-analysis were performed. Results: Six RCTs were included (n = 3,033): three from 2012 (n = 2,913) had low risk of bias. Median tetrastarch exposure was 37.4 ml/kg (range 30-43 ml/kg). Ninety-day mortality was associated with tetrastarch exposure [relative risk (RR) 1.13; 95% confidence interval (CI) 1.02-1.25; p = 0.02] compared with crystalloid. The number needed to harm (NNH) was 28.8 (95 % CI 14.6-942.5). Publication bias and statistical heterogeneity (I(2) = 0%) were not present. Tetrastarch exposure was also associated with renal replacement therapy (p = 0.01; NNH 15.7) and allogeneic transfusion support (p = 0.001; NNH 9.9). No difference between groups was observed for 28-day mortality, for comparison with colloid as control, or for waxy maize-derived tetrastarch, but power was lacking. Overall mortality was associated with tetrastarch exposure (RR 1.13; 95% CI 1.02-1.25; p = 0.02). Conclusions: In our analysis, 6% tetrastarch as part of initial fluid resuscitation for severe sepsis was associated with harm and, as alternatives exist, in our view should be avoided.
Time to death and relative risks for 1-year mortality in the hydroxyethyl starch (HES) 130/0.42 and Ringer's acetate groups, respectively. a Survival curves 1 year after randomisation for the two intervention groups in the modified intention-to-treat population. Kaplan-Meier analysis censored at 1 year showed that the survival time did not significantly differ between the two groups [intervention = HES: median survival 82 days, 95 % confidence interval (CI) 0-200 days; intervention = Ringer's: median survival 328 days (no standard error, so 95 % CI cannot be calculated]. The P of the log rank test was 0.20; the reasoning for using log rank test is given in the ESM. b Relative risks with 95 % CI for 1-year mortality in the HES group compared with the Ringer's acetate group in all patients and in the two pre-defined subgroups (shock defined as mean arterial pressure of \70 mmHg, ongoing treatment with vasopressor/inotropic agents or plasma lactate of [4.0 mmol/l in the hour prior to randomisation and acute kidney injury defined as renal Sepsis-related (=Sequential) Organ Failure Assessment score (SOFA) score of[2 [plasma creatinine of [170 lmol/l (1.92 mg/dl) or urinary output of \500 ml] in the 24 h prior to randomisation 
Adjustment for risk factors at baseline for death at 1 year as assessed by uni-and multivariate logistic regression analyses on the complete case dataset 
We assessed long-term mortality and hospitalisation in patients with severe sepsis resuscitated with hydroxyethyl starch (HES) or Ringer's acetate. This was an investigator-initiated, parallel-grouped, blinded randomised trial using computer-generated allocation sequence and centralised allocation data that included 804 patients with severe sepsis needing fluid resuscitation in 26 general intensive care units (ICUs) in Scandinavia. Patients were allocated to fluid resuscitation using either 6 % HES 130/0.42 or Ringer's acetate during ICU admission. We assessed mortality rates at 6 months, 1 year and at the time of longest follow-up and days alive and out of hospital at 1 year. The vital status of all patients was obtained at a median of 22 (range 13-36) months after randomisation. Mortality rates in the HES versus Ringer's groups at 6 months were 53.3 (212/398 patients) versus 47.5 % (190/400) [relative risk 1.12; 95 % confidence interval (CI) 0.98-1.29; P = 0.10], respectively; at 1 year, 56.0 (223/398) versus 51.5 % (206/400) (1.09; 95 % CI 0.96-1.24; P = 0.20), respectively; at the time of longest follow-up, 59.8 (238/398) versus 56.3 % (225/400) (1.06; 95 % CI 0.94-1.20; P = 0.31), respectively. Percentage of days alive and out of hospital at 1 year in the HES versus Ringer's groups was 24 (0-87 days) versus 63 % (0-90) (P = 0.07). The long-term mortality rates did not differ in patients with severe sepsis assigned to HES 130/0.42 versus Ringer's acetate, but we could not reject a 24 % relative increased or a 4 % relative decreased mortality at 1 year with HES at the 95 % confidence level.
Impairment of haemostasis has been described with slowly degradable medium molecular weight hydroxyethyl starch (MMW-HES), whereas rapidly degradable MMW-HES is generally considered to have no important effects on blood coagulation. This study was undertaken to investigate the effects of a rapidly degradable MMW-HES plasma substitute on primary haemostasis and blood coagulation in human subjects. Randomised, cross-over study. Research unit of a university hospital. Nine healthy, adult male volunteers. A 60-min intravenous infusion of 1 l HES 200/0.5/6 (HAES-steril 6%) or 4% albumin (control). The infusion of HES resulted in decreased circulating levels of von Willebrand factor antigen (from 85+/-8% to 59+/-6% after HES vs from 80+/-7% to 69+/-8% after albumin, p<0.05) and ristocetin cofactor activity (from 93+/-4 to 67+/-4% after HES vs from 79+/-5 to 75+/-5% after albumin, p<0.01). This was associated with an impairment of in vitro platelet function as determined with the PFA-100 platelet function analyser (closure time with collagen/epinephrine from 120+/-7 to 159+/-14 s after HES vs from 121+/-7 to 137+/-10 s after albumin, p<0.05; with collagen/ADP from 88+/-3 to 116+/-9 s and from 103+/-4 to 114+/-7 s after HES and albumin, respectively, p=0.01). The infusion of 1 l of HES 200/0.5/6 in healthy human subjects results in moderately decreased plasma levels of von Willebrand factor associated with impairment of platelet function.
Study flow chart 
Westley croup score system 
Baseline characteristics of study participants 
Treatment effect on hemodynamic and respiratory parameters 
Nebulized L-epinephrine has been recommended for the treatment of viral croup. However, the few studies assessing its effect on post-extubation stridor (PES) have shown conflicting results. We compared the efficacy and safety of nebulized L-epinephrine at three different doses for the treatment of PES. We conducted a prospective, randomized, double-blind trial including all consecutive children with a PES score of ≥4 (Westley score). The primary efficacy outcome was change in PES score at 40 min. A reduction of ≥2 points in stridor score was defined as clinically significant. A total of 96 patients were randomly assigned to receive one of three doses of nebulized L-epinephrine upon achieving a PES score of 4 or more following extubation. Stridor score and vital signs were recorded before treatment, and at 20, 40, 60 and 180 min after nebulization. Baseline characteristics were similar among all study groups. No significant difference was detected among the treatments based on change in Westley score by intent-to-treat analysis. In addition, the difference in the number of patients who clinically improved among the treatment groups was not significant (p = 0.54). Patients receiving 5 ml nebulized epinephrine had a significant increase of systolic and diastolic blood pressure at 40 and 180 min. Nebulized L-epinephrine at doses of 0.5, 2.5 and 5 ml demonstrated a lack of dose response in effect on PES and a modestly clinically significant increase in undesired side effects (heart rate and blood pressure) at higher doses.
The use of synthetic colloids for resuscitation and volume replacement is common in the intensive care unit. Although adverse reactions have been reported to colloid solutions, the incidence of severe reactions to the starch derivatives is low. We report a case of an anaphylactoid reaction to pentastarch (200/0.5) in a young asthmatic who received it as a fluid challenge in the intensive care unit. The pathogenesis and implications of such a reaction in an asthmatic are discussed.
Patients characteristics at enrolment (n=10). SAPS II Simplified Acute Physiology Score II [15]. Data are expressed as mean€SD or % 
To determine the steady-state plasma and epithelial lining fluid (ELF) concentrations of piperacillin/tazobactam (P/T) administered to critically ill patients with severe bacterial pneumonia. Prospective, open-label study. An intensive care unit and research ward in a university hospital. Ten adult patients with severe nosocomial bacterial pneumonia on mechanical ventilation. All subjects received a 30-min intravenous infusion of P/T 4 g/0.5 g every 8 h. The steady-state plasma and ELF concentrations of P/T were determined by high-performance liquid chromatography. The mean+/-SD steady-state plasma trough, peak, and intermediate concentrations were 8.5+/-4.6 microg/ml, 55.9+/-21.6 microg/ml, and 24.0+/-13.8 microg/ml for piperacillin, and 2.1+/-1.0 microg/ml, 4.8+/-2.1 microg/ml, and 2.4+/-1.2 microg/ml for tazobactam, respectively. The mean+/-SD steady-state intermediate ELF concentrations were 13.6+/-9.4 microg/ml for piperacillin and 2.1+/-1.1 microg/ml for tazobactam, respectively, showing a mean percentage penetration of piperacillin and tazobactam into ELF of 56.8% and 91.3 %, respectively, with a P/T ratio of 6.5:1. Our results show that during the treatment of severe nosocomial pneumonia, a regimen of P/T 4 g/0.5 g every 8 h might provide insufficient concentrations into lung tissue to exceed the MIC of many causative pathogens. This suggests that higher doses of P/T should be administered in order to maximize the antibiotic concentration at the site of infection, or that a second antimicrobial agent should be used in association.
To define the mechanisms of the stable and prolonged post-operative plasma volume expansion observed with Hydroxyethyl Starches (HES) and to determine whether a partial intravascular hydrolysis of large molecules contribute to reinforce the colloid-osmotic effect. Prospective, pharmacologic study using single dose of drug. University-based, post-anesthesia care unit. The protocol was performed during the post-operative period, in 10 patients after stable recovery from general anesthesia for carotid endarterectomy. HES 200/0.62 (500 ml) was infused over 30 min. Standard hemodynamic and biological variables, HES concentration and colloid osmotic pressure were obtained at each measurement. Plasma volume was calculated using 51Cr-labelled RBCs. Patterns of changes in number average molecular weight (MWn) and weight average MW (MWw) were measured using gel permeation chromatography. Measurements were obtained at control, end of infusion, 1 h, 3 h, 6 h and 24 h after infusion. Plasma volume increased by 693 ml (+21%) after the infusion of HES and remained constant over 24 h. HES concentration progressively decreased to reach a value of 35% of the peak at 24 h. MWn and MWw, initially decreased when compared with the dose solution and changed little in the 24 h study period. Diuresis significantly decreased at 3 h up to 24 h. Plasma albumin decreased after infusion and then progressively increased to reach a significantly higher value at 24 h than after infusion. Initial plasma volume expansion and decrease in HES concentration agree with previously-published data. Maintenance of plasma volume expansion over 24 h was not related to a partial intravascular hydrolysis. Low elimination rate of HES, extravascular mobilization of albumin and post-operative renal adaptations were possibly the 3 main mechanisms to explain a prolonged plasma volume expansion with HES 200/0.62, 6%.
To investigate the effect of ketamine on endotoxin modulation of inositol 1,4,5-triphosphate (IP3) formation in cardiomyocytes. A prospective observational cell culture study. A research laboratory in the University of Hirosaki School of Medicine. Neonatal rat cardiomyocytes. We investigated bradykinin-induced IP3 production in the presence of lipopolysaccharide (LPS) and the effect of ketamine on the LPS modulation of IP3 formation. The LPS modulation of IP3 formation was measured in the presence of BM13177 (a thromboxane A2 (TXA2) receptor inhibitor) or GDPbetaS (a GTP-binding protein inhibitor). U46619 (a TXA2 agonist)-induced IP3 production was measured in the presence of ketamine, and the ketamine modulation of U46619-induced IP3 production was measured in the presence of W7 (a Ca2+ releasing agent) and verapamil (a Ca2+ channel blocker). One micromole ketamine significantly attenuated the LPS-induced IP3 production from 763.8+/-34.6 to 461.6+/-65.1 pmol mg protein(-1). Ten micromoles of BM13177 or 1 mM GDPbetaS significantly blocked LPS modulation of bradykinin-induced IP3 production from 786.0+/-33.8 to 218.6+/-21.6 and 226.8+/-25.4 pmol mg protein(-1). One micromole of ketamine significantly decreased U46619-induced IP3 production from 857.3+/-45.0 to 632.9+/-64.5 pmol mg(-1) protein. The ketamine inhibition of U46619-induced IP3 production was enhanced by W7 and inhibited by verapamil. Ketamine decreased LPS-induced IP3 formation and the ketamine inhibition was associated with inhibition of the TXA2-IP3 sequence. Inhibition of TXA2 by ketamine was associated with a decrease in intracellular Ca2+.
Characteristics of the 72 patients included in the coloni- zation study 
Patients colonized by P. aeruginosa during the period of the study (*Two patients colonized by different strains during ICU stay) 
To identify routes and patterns of colonization with Pseudomonas aeruginosa in intubated patients to design strategies of prevention for respiratory infection. Prospective and observational study in the 16-bed intensive care unit of a teaching hospital. Ninety-eight intubated patients were investigated over a 3-year period. Those ventilated less than 72 h were excluded. Samples from the tap water from each patient's room, stomach, oropharynx, subglottic secretions, trachea, and rectum were collected when the patient was intubated, and then three times per week. Pulsed-field gel electrophoresis was performed to type the strains. We identified 1,607 isolates pertaining to 35 different pulsotypes. Overall 54.2% of patients presented colonization, and tracheal colonization was present in 30.5%. Ten patients had colonization at intubation, and four of these developed ventilator-associated pneumonia (VAP) after a mean of 4+/-2 days. ICU-acquired colonization occurred in 31 patients, and 4 of these developed VAP after a median of 10+/-5 days. P. aeruginosa was isolated from the room's tap water in 62.4% of samples. More than 90% of tap water samples had pulsotypes 1 and 2, which were frequently isolated in the stomach (59%) but were only rarely associated with VAP. Although colonization/infection with P. aeruginosa in intubated patients tends to be endogenous, exogenous sources should not be ruled out. A combination of early identification (and eradication) of airways colonization by P. aeruginosa plus infection control measures targeted to reduce cross-contamination should be the basis to prevent pulmonary infection.
S-100B protein is a promising marker of injury severity and outcome after head injury. We examined the relationship between serum S-100B concentrations and injury severity, clinical course, survival, and treatment efficacy after severe traumatic brain injury (TBI). Prospective observational study in a neurosurgical intensive care unit. 102 adult patients with severe TBI, admitted between June 2001 and November 2003 (30 months). Serum S-100B levels were measured by immunoluminometric technique on admission and every 24 h thereafter for a maximum of 7 days. Initial S-100B levels were significantly related to pupillary status, computed tomography severity 1, and 1-month survival. Cox's proportional hazard regression analysis showed that initial S-100B was an independent predictor of 1-month survival, in the presence of dilated pupils, and with increased age. Subjects with initial levels above 1 microg/l had a nearly threefold increased probability of death within 1 month. Serum S-100B alteration indicated neurological improvement or deterioration. Finally, surgical treatment reduced S-100B levels. Serum S-100B protein reflects injury severity and improves prediction of outcome after severe TBI. S-100B may also have a role in assessing the efficacy of treatment after severe TBI.
To assess thyroid function abnormalities in survivors of severe head trauma and to examine their relationship with indices of brain injury severity. Prospective study. General intensive care unit (ICU) in a university hospital. Twenty-two (21 men) head-injured patients, with a median age of 25.5 years at the time of injury, were investigated. Severity of brain trauma was assessed by Glasgow Coma Scale (GCS) score, Marshall Computerized Tomographic Classification, intracranial pressure levels and serum S-100b concentrations measured over a 6-day period. Thyroid function testing was performed 1 year after ICU discharge and included the measurement of free thyroxine, triiodothyronine and thyrotropin. On admission to the ICU, GCS ranged from 3 to 8. Peak S-100b was 1.49 microg/l (range: 0.37-5.26 microg/l). Median triiodothyronine and thyrotropin were 123 ng/dl and 1.60 microIU/ml, respectively. Free thyroxine was 1.08+/-0.22 ng/dl (range: 0.7-1.5 ng/dl). Overall, 7 of the 22 patients (32%) had thyroid dysfunction. Four patients had central hypothyroidism and three patients had subclinical hypothyroidism. Peak S-100b correlated negatively with free thyroxine (r=-0.47, p=0.02). There were no correlations between other brain injury severity indices and thyroid hormone levels. A significant subset of brain injury patients presents with changes in thyroid function 1 year after ICU discharge; these depend upon biochemical serum markers of head trauma severity.
To determine possible changes in outcome from acute respiratory distress syndrome (ARDS) and to compare severity of lung injury and methods of treatment from 1967 to 1994. Computerized (Medline, Current Contents) and manual (Cumulated Index Medicus) literature search using the key word and/or title ARDS. Only clinical studies published as full papers reporting data on both patient mortality (survival) and oxygenation index (PaO2/FIO2) were included. Single case reports, abstracts, reviews and editorials were excluded from evaluation. Relevant data were extracted in duplicate, followed by quality checks on approximately 80% of data extracted. 101 papers reporting on 3264 patients were included: 48 studies (2207 patients) were performed in the USA, 43 studies (742 patients) in Europe and 10 studies (315 patients) elsewhere. Mortality reported in these studies was 53 +/- 22% (mean +/- SD), with no apparent trend towards a higher survival (1994: 22 studies, mortality 51 +/- 19%). The mean PaO2/FIO2 ratio remained unchanged throughout the observation period (118 +/- 47 mmHg). No correlation could be established between outcome and PaO2/FIO2 or lung injury score. Patients who underwent pressure-limited ventilation had a significantly lower mortality (35 +/- 20%) than patients on volume-cycled ventilation (54 +/- 22%) or patients for whom there was no precise information on ventilatory support (59 +/- 19%). Significantly lower PaO2/FIO2 ratios (61 +/- 17 mmHg) were observed in patients prior to extracorporeal lung assist, together with mortality rates in the range of those for conventionally treated patients (55 +/- 22%). The mortality of ARDS patients remained constant throughout the period studied. Therefore, the standard for outcome in ARDS should be a mortality in the 50% range. Neither PaO2/FIO2 ratio nor lung injury score was a reliable predictor for outcome in ARDS. Patients might benefit from pressure-limited ventilatory support, as well as extracorporeal lung assist. Since crucial data were missing in most clinical studies, thus preventing direct comparison, we emphasize the importance of using standardized definitions and study entry criteria.
To evaluate the impact of noninvasive positive pressure mechanical ventilation (NPPV) on ventilator-associated pneumonia (VAP). Prospective observational study. Medical intensive care unit (ICU) of a university teaching hospital. Cohort of 320 consecutive patients staying in the ICU more than 2 days and mechanically ventilated for > or = 1 day. VAP was diagnosed when, satisfying classical clinical and radiological criteria, fiberoptic bronchoalveolar lavage and/or protected specimen brush grew > or = 10(4) and > or = 10(3) CFU/ml, respectively, of at least one microorganism. Patients were classified into four subgroups according to the way in which mechanical ventilation was delivered: NPPV then tracheal intubation (TI) (n = 38), TI then NPPV (n = 23), TI only (n = 199), and NPPV only (n = 60). Occurrence of VAP was estimated by incidence rate and density of incidence. Risk factors for VAP were assessed by logistic regression analysis. Twenty-seven patients had 28 episodes of VAP. The incidence rates for patients with VAP were 18% in NPPV-TI, 22% in TI-NPPV, 8% in TI, and 0% in NPPV (p < 0.0001). The density of incidence of VAP was 0.85 per 100 days of TI and 0.16 per 100 days of NPPV (p = 0.04). Logistic regression showed that length of ICU stay and ventilatory support were associated with VAP. There is a significantly lower incidence of VAP associated with NPPV compared to tracheal intubation. This is mainly explained by differences in patient severity and risk exposure.
Flow chart describing the patient selection process from the initial to the final sample  
Distribution of observed and expected deaths across risk classes
Forest plots describing standardized mortality ratios (SMRs) with their 95 % confidence intervals for some important subsets. The numbers of patients for each subset are given in parentheses. The dotted line indicates the overall SMR based on each of the two scores  
More recent severity scores should be more reliable than older ones because they account for the improvement in medical care over time. To provide more insight into this issue, we compared the predictive ability of the Simplified Acute Physiology Score (SAPS) II and SAPS 3 (originally developed from data collected in 1991-1992 and 2002, respectively) on a sample of critically ill patients. This was a prospective observational study on 3,661 patients from 103 Italian intensive care units. Standardized mortality ratios (SMRs) were calculated. Assessment of calibration across risk classes was performed using the GiViTI calibration belt. Discrimination was evaluated by means of the area under the receiver operating characteristic analysis. Both scores were shown to discriminate fairly. SAPS 3 largely overpredicted mortality, more than SAPS II (SMR 0.63, 95 % CI 0.60-0.66 vs. 0.87, 95 % CI 0.83-0.91). This result was consistent and statistically significant across all risk classes for SAPS 3. SAPS II did not show relevant deviations from ideal calibration in the first two deciles of risk, whereas in higher-risk classes it overpredicted mortality. Both scores provided unreliable predictions, but unexpectedly the newer SAPS 3 turned out to overpredict mortality more than the older SAPS II.
The respiratory and hemodynamic effects of halothane in patients with status asthmaticus who required mechanical ventilation was evaluated. Halothane was administered in 12 patients in a concentration of 1% for thirty minutes. Standard drug treatments and ventilator settings were not modified during halothane administration. The following data were collected before and after halothane administration: arterial blood gases, peak inspiratory pressure, VD/VT, pulmonary arterial pressure, right heart pressures and cardiac index (by means of the thermodilution method). After halothane treatment PaCO2 significantly decreased, arterial pH increased, peak inspiratory pressure decreased and VD/VT decreased significantly. Mean pulmonary arterial pressure and right heart pressures decreased and the cardiac index was unchanged. The heart rate significantly decreased and arrhythmias did not occur during halothane administration. The administration of halothane in patients with status asthmaticus requiring mechanical ventilation produces a rapid reduction in bronchospasm and barotraumatic injury and a rapid improvement in arterial blood gases, without any adverse hemodynamic effects.
Study flowchart. CXR Chest radiograph, ICU Intensive care unit. See text for definitions of ''on-demand'' and ''daily routine'' individual CXRs  
Therapeutic impact of chest radiographs. Depicts the changes in patient's care that would not have occurred without the chest radiograph (CXR). These changes may involve indwelling devices (a), diagnostic interventions (b) or therapeutic interventions (c). The 144 CXRs performed for no other reason than ICU admission and/or insertion of an indwelling device were not analyzed (because these two indications are often shared by both daily routine and on-demand prescription policies). For antimicrobial therapy, hydric depletion, physiotherapy and inotropic agents, the therapeutic intervention consisted of the initiation, change, discontinuation or continuation (whereas the discontinuation was scheduled). See the text for definitions of ''on-demand'' and ''daily routine'' individual CXRs. Results are expressed as the percentage of CXRs for each modality of prescription: on-demand or daily routine  
Clinical value of chest radiographs with respect to the PaO2/FiO2 ratio. CXR Chest radiograph, MV mechanical ventilation. The PaO 2 /FiO 2 ratio has been changed into a categorical variable according to the respiratory component of the Sequential Organ Failure Assessment score (SOFA) score. *When analyzing either all the CXRs or only ''ondemand'' CXRs, there was a significant association between PaO 2 /FiO 2 and the incidence of expected findings not impacting patient care (p \ 0.001 and p = 0.001, respectively, with the v 2 test for independence). This was not the case for ''daily routine'' CXRs (p = 0.47). Therefore, the lower the PaO 2 / FiO 2 ratio is, the higher the clinical value of ''on-demand'' CXRs, whereas ''daily routine'' CXRs were associated with a low clinical value whatever the PaO 2 /FiO 2 ratio  
Timetable of the CXRs performed on RadioDay: during (scheduled CXRs) or outside the morning round (unscheduled CXRs). CXR Chest radiograph, ICU intensive care unit. MV patients Mechanically ventilated patients. As compared with ICUs that have adopted a ''daily routine'' policy of CXR prescription (in patients under MV), fewer CXRs were performed during the morning round (scheduled CXRs) in ''on-demand'' ICUs and, importantly, this did not increase the rate of CXRs performed outside the morning round (unscheduled CXRs)  
To analyze the current practices of chest radiograph (CXR) prescription and their clinical impact. Prospective snapshot observational study (on RadioDay) combined with a survey. Patients who were given a CXR on RadioDay. One hundred four French intensive care units (ICUs). On RadioDay, 854 CXRs (in 804 patients) were ordered. For the "CXRs morning round," the prescription policy was declared to be "on-demand" (in 63 % of the ICUs), "daily routine only in mechanically ventilated patients (MV)" (30 %) or, less frequently, "daily routine in all patients" (7 %). When analyzing the two main local policies, as compared with "daily routine only in MV" ICUs, in "on-demand" ICUs: (1) fewer CXRs were ordered (0.6 ± 0.3 vs. 0.9 ± 0.2 CXRs/patient, p < 0.001) with no increase in the rate of unscheduled CXRs (i.e., CXRs performed outside the morning round), and (2) individual CXRs were more often followed by a therapeutic intervention (which would not have occurred without the CXR): 34 vs. 25 % of the CXRs (p < 0.05). Last, in case of severe respiratory disease (low PaO(2)/FiO(2) ratio), it is noteworthy that the clinical value of "on-demand" individual CXRs was still markedly higher than that of "daily routine" CXRs. Nearly two-thirds of the participating ICUs adopted the "on-demand" strategy of prescription, which was associated with a lower rate of CXRs with no increase in unscheduled CXRs and was of higher clinical value than a "daily routine in MV" strategy. Importantly, the study design did not allow assessing if the "on-demand" strategy had missed or delayed some diagnoses.
Computed tomography DICOM images analysis allows a quantitative measurement of organ weight, volume and specific gravity in humans. The brain weight, volume and specific gravity of 15 traumatic brain-injury patients (3+/-2 days after trauma) were computed using a specially designed software (BrainView). Data were compared with those obtained from 15 healthy subjects paired for age and overall intracranial volume. Hemisphere weight were 91 g higher in patients than in controls (1167+/-101 vs 1076+/-112 g; p<0.05). Specific gravity of hemispheres (1.0367+/-0.0017 vs 1.0335+/-0.0012 g/ml; p<0.001), brainstem (1.0302+/-0.0016 vs 1.0277+/-0.0015 g/ml; p<0.001) and cerebellum (1.0396+/-0.0020 vs 1.0375+/-0.0015 g/ml; p<0.05) was significantly higher in traumatic brain injury (TBI) patients than in controls (all p<0.0001 without interaction). This increase in specific gravity was evenly distributed between the hemispheres, the brainstem and the cerebellum, and the grey and white matter. It was more pronounced in the rostral than in the caudal areas of the hemispheres. It was independent of the volume of brain contusion, of the mechanism of head injury, of natremia and of initial Glasgow coma score. Human TBI patients present a diffuse increase in specific gravity. This observation is in sharp opposition with the data derived from the experimental literature.
(a) To assess whether differences in lung morphology observed in patients with adult respiratory distress syndrome (ARDS) are associated with differences in cardiorespiratory parameters, lung mechanics, and outcome. (b) To propose a new ARDS Severity Score to identify patients with a high mortality risk. Prospective study over a 53-month period. Fourteen-bed surgical intensive care unit of a university hospital. Seventy-one consecutive patients with early ARDS. Cardiorespiratory parameters were measured using a Swan-Ganz catheter, the pressure-volume (PV) curve was measured using the gross syringe method, and fast spiral computed tomography (CT) was performed. Patients with diffuse attenuations (n = 16) differed from patients with lobar attenuations (n = 26) regarding: (a) mortality rate (75% vs. 42%, p = 0.05), (b) incidence of primary ARDS (82% vs. 50%, p = 0.03), (c) respiratory compliance (47 +/- 12 vs. 64 +/- 16 ml per cmH2O(-1) p = 0.04), and (d) lower inflexion point (8.4 +/- 2.0 vs. 4.6 +/- 2.0 cmH2O, p = 0.001). A third group of patients with patchy attenuations (n = 29) had a mortality rate of 41 %, a respiratory compliance of 56 +/- 18 ml per cmH2O(-1) and a lower inflexion point of 6.3 +/- 2.7 cmH2O. The bedside chest radiograph accurately assessed lung morphology in only 42% of the patients. In contrast to the scores based on the bedside chest radiograph, a new ARDS Severity Score based on CT lung morphology and cardiorespiratory parameters identified a subgroup of patients with a high mortality rate (> or = 60%). In patients with ARDS, differences in lung morphology are associated with differences in outcome and lung mechanics. A new ARDS Severity Score based on CT lung morphology and cardiorespiratory parameters accurately identified patients with the most severe forms of ARDS and a mortality rate above 60%.
Interleukin (IL)-10 levels produced by cultured peripheral blood mononuclear cells ( PBMC ) in healthy donors and sepsis patients in relation to -1082 promoter polymorphism. Isolated PBMC from healthy donors ( n = 14 for AA and n = 14 for AG/GG) and sepsis patients ( n = 12 for AA and n = 12 for AG/GG) were cultured for 24 h at 37 ◦ C, with 
Interleukin (IL)-10 levels produced by cultured peripheral blood mononuclear cells ( PBMC ) in survivors and non-survivors in relation to -1082 promoter polymorphism. Isolated PBMC from survivors ( n = 7 for AA and n = 4 for AG/GG) and non-survivors ( n = 5 for AA and n = 8 for AG/GG) were cultured for 24 h at 37 ◦ C with either C3-binding glycoprotein ( C3bgp ), lipopolysaccharide ( LPS ), phytohemagglutinin ( PHA ), pokeweed mitogen ( PWM ) or non-stimulated ( N ). The quantity of IL-10 was determined in culture supernatants by ELISA test. The results are presented as mean ± SE. ( ∗ p < 0 . 05, non-survivors vs survivors with the same genotype; ◦ p < 0 . 05, 
To investigate the -1082 (A/G) polymorphism in the promoter of the IL-10 gene in terms of IL-10 production from stimulated peripheral blood mononuclear cells (PBMC) and to evaluate the relationship of this polymorphism with susceptibility to severe sepsis and the outcome of the disease. Case-control study. Research laboratory of Molecular Biology and Immunology and University Hospital ICU, Faculty of Medicine, Trakia University. A total of 53 healthy volunteers and 33 patients in ICU meeting the criteria for severe sepsis were included. The amplification refractory mutation system PCR was used for IL-10-1082 polymorphism detection. Isolated PBMC were stimulated with either C3-binding glycoprotein (C3bgp), lipopolysaccharide (LPS), phytohemagglutinin (PHA),or pokeweed mitogen (PWM). IL-10 production was measured in culture supernatants. The AA genotype was associated with lower IL-10 production in LPS-, PHA- or PWM-stimulated healthy PBMC. Patients with severe sepsis had significant elevation of A allele, compared with healthy controls (74.2% vs 52.8%; p=0.0062). Carriage of at least one copy of IL-10-1082 G allele in sepsis patients and in healthy controls resulted in a statistically significant increase in IL-10 production from stimulated PBMC. Surviving sepsis patients had a significant decrease of IL-10-1082 allele G frequency, compared with controls (17% vs 47.2%; p=0.012). An association between increased IL-10 production and poor outcome from sepsis was observed. The A allele of the -1082 polymorphism in the interleukin-10 gene promoter is associated with sepsis susceptibility, whereas G allele is associated with higher stimulated interleukin-10 production and increased mortality in severe sepsis.
Blood cultures were obtained from 39% of all 574 admissions to our Medical Intensive Care Unit. (ICU); in 109 (19%) a pathogenic organism was demonstrated. 45% of the septicaemias were detected within the first 48 h of ICU stay have been considered as "non ICU-acquired". Septicaemic patients were significantly older, had longer ICU stays and a higher mortality rate (62%) than non septicaemic patients (28%) (p less than 0.05). Gram negative organisms (69%) predominated over gram positive (29%) and Serratia marcescens and coagulase positive Staphylococcus were the most frequently isolated. Shock appeared in 32% and had an extremely high mortality (91%) and was associated with the presence of "multiple species septicaemia". Prior to the septicaemia the survivors differed from the fatalities only in the level of serum albumin; this was significantly lower in patients with gram negative in comparison with gram positive septicaemias and in patients who developed shock. Arterial, pulmonary artery and urinary catheters, and endotracheal devices were used frequently in these patients and were statistically associated with the presence of septicaemia. The airway was the most frequent possible source for the septicaemia.
To determine the incidence of end-of-life decisions in intensive care unit (ICU) patients. Prospective data collection and questionnaire in a 31-bed medicosurgical ICU in a university hospital. All 109 ICU patients who died during a 3-month period (April-June 2001). Members of the ICU team were also invited to complete a questionnaire regarding the circumstances of each patient's death. Cardiopulmonary resuscitation was performed in 21 of the patients; other mechanisms leading to death were brain death (n=19), refractory shock (n=17), and refractory hypoxemia (n=2). The decision was taken in the remaining 50 patients to withdraw (n=43) or withhold (n=7) therapy. Questionnaires were completed for 68 patients, by physician and nurse in 40 cases, physician only in 20 cases, and nurse only in 8 cases. Questionnaires were obtained for 34 of 50 patients for whom a decision was made to limit therapy. Respondents generally felt that the decision was timely (n=28, 82%), 5 (15%) felt the decision was too late, and one (3%) that the decision was made too soon, before the family could be informed. Therapeutic limitations are frequent in patients dying in the ICU, with withdrawing more common than withholding life support. Generally members of the ICU staff were satisfied with the end-of-life decisions made.
A full-term, male neonate developed persistent pulmonary hypertension, and responded to high-frequency oscillatory ventilation and inhaled nitric oxide (INO). Discontinuation of INO was attempted three times and was followed by severe desaturations due to right-to-left shunt through the patent ductus arteriosus and patent foramen ovale. As a result of this rebound pulmonary hypertension, the neonate was maintained on INO therapy for 6 days. Successful discontinuation was achieved by using the phosphodiesterase inhibitor, dipyridamole. We speculate that during exogenous INO therapy, endogenous nitric oxide was inhibited, thus cyclic guanosine 3',5'-monophophate, the smooth muscle relaxant, was rapidly hydrolyzed. By inhibiting phosphodiesterase, smooth muscle relaxation occurred, and consequently weaning from INO was achieved.
To compare the results of cardiac output measurements obtained by lithium dilution and transpulmonary thermodilution in paediatric patients. Design: A prospective study. Paediatric intensive care unit in a university teaching hospital. Twenty patients (age 5 days-9 years; weight 2.6-28.2 kg) were studied. Between two and four comparisons of lithium dilution cardiac output (LiDCO) and transpulmonary thermodilution (TPCO) were made in each patient. Results from three patients were excluded: in one patient there was an unsuspected right-to-left shunt, in two patients there was a problem with blood sampling through the lithium sensor. There were 48 comparisons of LiDCO and TPCO in the remaining 17 patients over a range of 0.4-6 l/min. The mean of the differences (LiDCO-TPCO) was -0.1 +/- 0.3 (SD) l/min. Linear regression analysis gave LiDCO = 0.11 + 0.90 x TPCO l/min (r2 = 0.96). There were no adverse effects in any patient. These results suggest that the LiDCO method can be used to provide safe and accurate measurement of cardiac output in paediatric patients. The method is simple and quick to perform, requiring only arterial and venous catheters, which will already have been inserted for other reasons in these patients.
To analyze the epidemiology and factors influencing mortality of ICU-acquired bacteremia. Prospective clinical study. A medical-surgical ICU in an university hospital. We recorded variables from 111 consecutive ICU-acquired episodes for a 3-year period. The attack rate was 1.9 episodes per 100 patient-days. The commonest isolates were coagulase-negative staphylococci, Staphylococcus aureus, Pseudomonas aeruginosa, and Escherichia coli. Intravascular catheters were the most frequent source of infection. Overall mortality was 31.5%, and 65.7% of all deaths were directly attributable to infection. Bacteremia from intra-abdominal, lower respiratory tract or unknown origin were associated with a poor prognosis. A logistic regression analysis defined intraabdominal origin (p = 0.01, OR = 15.7) and presence of shock (p = 0.04, OR = 3.3) as independently influencing the risk of death. No significant differences were found for the remaining variables studied. Epidemiology and etiology of ICU-acquired bacteremia does not differ seriously in respect to nosocomial bacteremia among unselected populations, although it is associated with a greater incidence and overall mortality. Presence of shock is the most important modificable variable affecting the outcome.
To study current strategies in nutritional management of pediatric intensive care units (PICUs) in Europe, focusing on energy requirements. Survey by a 35-item questionnaire sent to representatives of 242 PICUs in 28 countries. Addresses were obtained from national PICU associations and the members' list of the European Society of Pediatric and Neonatal Intensive Care. Staff members of 111 European PICUs (46%) from 24 countries. Predominantly physicians were reported to be responsible for nutritional support. In 73% of PICUs a multidisciplinary nutritional team was available. In most PICUs daily energy requirements were estimated using weight, age, predictive equations and correction factors. In 17% of PICUs energy expenditure was regularly measured by indirect calorimetry. Nutritional status was mostly assessed by weight, physical examination, and a wide range of biochemical blood parameters. Approximately 70% of PICUs used dedicated software for nutritional support. A similar percentage of PICUs regarded "nutrition" as a research topic and part of the residents' training program. Most European PICUs regard nutritional support as an important aspect of patient care, as shown by the presence of nutritional teams, software, research, and education. However, energy requirements of pediatric intensive care patient were based predominantly on estimations rather than on measurements.
A retrospective study of the medical records of 115 consecutive cases of meningococcal disease with 11 fatalities was carried out in order to evaluate the power of 15 clinical and laboratory variables available on admission in predicting a fatal outcome. On linear discriminant analysis, six variables showed a significant discriminating power in predicting death: low systolic blood pressure, low platelet count, extensive petechiae, high body temperature, low CSF polynuclear cell count and absence of meningism. From a stepwise linear discriminant analysis, two alternative procedures for prognostic evaluation were derived. If a large high risk group is accepted which will include practically all patients at risk of death, a prognostic evaluation based only on systolic blood pressure on admission is sufficient. Alternatively, if unconventional, potentially hazardous therapy is considered for high risk patients, a small high risk group may be defined. Patients with systolic blood pressure less than 100 mm Hg, platelet count less than or equal to 125 X 10(9)/l and body temperature greater than 39 degrees on admission constitute a small group with a very high mortality. A similar risk group was defined if platelet count less than or equal to 125 X 10(9)/l was substituted for extensive petechiae. The advantage of the latter procedure is that only a simple bedside examination is required for the prognostic evaluation.
To assess the usefulness of venous oxygen saturation in the jugular bulb (SjO(2)) as a complementary test for the diagnosis of brain death. Prospective observational study. Polytrauma intensive care unit (ICU) of an acute-care teaching hospital in Santander, Spain. We studied 118 (44%) out of 270 patients with severe head injury and intracranial hemorrhage meeting criteria of brain death (lack of cardiac response to atropine, unresponsive apnea, and iso-electric EEG in the absence of shock, hypotension and treatment with muscle relaxants and/or central nervous system (CNS) depressant drugs). At the moment at which clinical diagnosis of brain death was made and an iso-electric EEG was obtained, simultaneous oxygen saturation in central venous blood (right atrium) (SvO(2)) and jugular venous bulb (SjO(2)) samples was measured. The ratio between SvO(2) and SjO(2), expressed as CvjO(2) (the so-called central venous-jugular bulb oxygen saturation rate; CvjO(2) = SvO(2)/SjO(2)) was calculated. CvjO(2) less than 1 was obtained in 114 patients [mean (SD): 0.89 (0.02)], whereas CvjO(2) greater than 1 was obtained in only 4 (3.38%). In the group of 152 survivors, a single patient was discharged from the ICU in a vegetative state in which CvjO(2) was below 1. CvjO(2)as a complementary test for the diagnosis of brain death showed 96.6% sensitivity, 99.3% specificity, and 99.1% and 97.4% positive and negative predictive values, respectively. Central venous-jugular bulb oxygen saturation rate below 1 together with accepted clinical criteria (unresponsive coma with brainstem areflexia) provides non-invasive assessment of cerebral circulatory arrest that can help to suspect brain death.
Distribution of the studied patients after randomization . SBT Spontaneous breathing trial  
To investigate the possibility of successful extubation performing a spontaneous breathing trial (SBT) in pressure support ventilation (PSV) with target durations of 30 and 120 min. Prospective and randomized study in two medical-surgical adult intensive care units. 98 adult patients supported by mechanical ventilation for at least 48 h and considered ready for a weaning trial. An SBT conducted in PSV with 7 cmH(2)O and patients randomly assigned to two groups with target durations of 30- and 120-min. In the 30-min group 43 patients (93%) tolerated the SBT and were extubated while 4 (9%) needed reintubation within 48 h; in the 120-min group 46 patients (88%) successfully completed the trial and were extubated while 2 (4%) were reintubated. ICU mortality in the groups with short and long periods was 6% and 4%, and in-hospital mortality 20% and 17%, respectively. Those successful in the 30- and 120-min groups had similar length of ICU stay (6 and 7 days, respectively) and in-hospital length of stay (20 and 25 days, respectively). Compared to the successfully extubated, the reintubated patients had significantly higher length of ICU stay and mortality (17 vs. 6 days and 33 vs. 3.6%, respectively). An SBT with PSV of 7 cmH(2)O lasting 30 min is equally effective in recognizing the successfully extubated patients as a 120-min trial.
We investigated whether a treatment according to a clinical algorithm could improve the low survival rates in acute respiratory distress syndrome (ARDS). Uncontrolled prospective trial. One university hospital intensive care department. 122 patients with ARDS, consecutively admitted to the ICU. ARDS was treated according to a criteria-defined clinical algorithm. The algorithm distinguished two main treatment groups: The AT-sine-ECMO (advanced treatment without extracorporeal membrane oxygenation) groups (n = 73) received a treatment consisting of a set of advanced non-invasive treatment options, the ECMO treatment group (n = 49) received additional extracorporeal membrane oxygenation (ECMO) using heparin-coated systems. The groups differed in both APACHE II (16 +/- 5 vs 18 +/- 5 points, p = 0.01) and Murray scores (3.2 +/- 0.3 vs 3.4 +/- 0.3 points, p = 0.0001), the duration of mechanical ventilation prior to admission (10 +/- 9 vs 13 +/- 9 days, p = 0.0151), and length of ICU stay in Berlin (31 +/- 17 vs 50 +/- 36 days, p = 0.0016). Initial PaO2/FIO2 was 86 +/- 27 mm Hg in AT-sine-ECMO patients that improved to 165 +/- 107 mm Hg on ICU day 1, while ECMO patients showed an initial PaO2/FIO2 of 67 +/- 28 mm Hg and improvement to 160 +/- 102 mm Hg was not reached until ICU day 13. QS/QT was significantly higher in the ECMO-treated group and exceeded 50% during the first 14 ICU days. The overall survival rate in our 122 ARDS patients was 75%. Survival rates were 89% in the AT-sine ECMO group and 55% in the ECMO treatment group (p = 0.0000). We conclude that patients with ARDS can be successfully treated with the clinical algorithm and high survival rates can be achieved.
Univariate results and ROC results of eight prognostic scoring systems (s.e. standard error) 
To investigate the accuracy of eight different prognostic scores (Stiehm, Niklasson, Leclerc, Garlund, the MOC score, Tesero, the Glasgow Meningococcal Septicaemia Prognostic Score (GMSPS) and Tüyzüs) in the prediction of fatal outcome in meningococcal disease. Combined prospective and retrospective study. A 175-bed pediatric department of a university hospital providing secondary care to +/- 180,000 inhabitants and serving as a referral center. The Pediatric Intensive Care (14 beds) is one of the six PICUs in the Netherlands and provides tertiary care for children under 18 years. During an 8-year period (1986-1994) 125 children (mean age 4 years, 10 months) with culture-proven meningococcal disease were studied: 34 patients presenting with meningitis, 33 patients with septic shock and 58 patients with meningitis and septic shock. All eight scores discriminated above average between survivors and non-survivors, as expressed by the corresponding Receiver Operator Characteristic (ROC) curves. The area under the ROC curve (AUC) ranged from 0.74 for the Garlund score to 0.93 for the GMSPS. The GMSPS performed significantly better than its competitors, even after exclusion of the base deficit as one of the score components (AUC = 0.92). It showed above average calibration when logistically transformed into a probability of mortality, and accurately identified a subgroup of patients with no mortality. None of the scores correctly identified non-survivors. The GMSPS is a simple score that can be reliably used for risk classification and the identification of low-risk patients.
We analyzed the haemodynamic effects of high doses of r-Met Hu IL-2 [ala-125] in four patients with advanced cancer. Haemodynamic parameters were measured daily from days 1 to 6 of treatment. Mean arterial pressure decreased significantly (98 +/- 11 mmHg on day 1 versus 84 +/- 7 mmHg on day 5; p = 0.0435) as did systemic vascular resistance (2042 +/- 296 dynes s cm-5 m-2 on day 1 versus 1166 +/- 87 dynes s cm-5 m-2 on day 5; p = 0.003). There was a significant increase in mean pulmonary artery pressure (13.25 +/- 3.30 mmHg on day 1 versus 20.75 +/- 7.41 mmHg on day 5; p = 0.03), systemic oxygen consumption (173.5 +/- 37.8 ml min-1 m-2 on day 1 versus 257.8 +/- 20.5 ml min-1 m-2 on day 5; p = 0.02) and cardiac index (3.86 +/- 0.58 l min-1 m-2 on day 1 versus 5.77 +/- 0.21 l min-1 m-2 on day 5; p = 0.008). There was no significant decrease in the arteriovenous oxygen content difference (4.5 +/- 0.8 ml dl-1 on day 1 versus 4.46 +/- 0.22 ml dl-1 on day 5). Increases in oxygen delivery (570 +/- 163 ml min-1 m-2 on day 1 versus 750 +/- 109 ml min-1 m-2 on day 5 and oxygen extraction ratio (29.95% +/- 6.37% on day 1 versus 34.60% +/- 4.35% on day 5) were not statistically significant. We concluded that the haemodynamic effect induced by high doses of r Hu-IL-2 is similar to that seen in septic shock.
The identification of risk factors contributing to the development of pulmonary oedema, pneumonia and late mortality in submersion victims. A retrospective study of 125 submersion victims. The medical intensive care unit in a university hospital. Baseline examination on admission consisted of history, physical examination, arterial blood gas analysis and a chest radiograph. Patients were then classified into four groups: class I, baseline examination negative; class II, baseline examination positive, but mechanical ventilation not needed on admission; class III, mechanical ventilation required on admission; class IV, patients suffering from cardiopulmonary arrest. All patients who were not successfully resuscitated or who had expired within 24 h after admission were excluded for determination of the risk of pulmonary oedema and pneumonia. Class I patients did not develop pulmonary complications; neither pulmonary oedema nor pneumonia occurred in this group. In the remaining classes the incidence of pulmonary oedema was 72% and that of pneumonia, 14.7%. Stepwise logistic regression showed that pulmonary oedema was related to the type of water (seawater, ditch water, swimming pool) victims were submerged in and to the neurological state both at the time of rescue and on admission. The development of pneumonia was related to the use of mechanical ventilation (the risk was 52%). Pneumonia was not related to neurological state at the time of rescue or on admission, to body temperature on admission, to the prophylactic administration of antibiotics or to the use of corticosteroids. Mortality was high in class IV patients, but low in all other patients. Early mortality was 18.4% while late mortality was 5.6%. There is no need to hospitalise submersion victims when there are no signs or symptoms of aspiration upon arrival in the emergency room. All other patients should be admitted to an intensive care unit. The risk of pneumonia is high when mechanical ventilation is necessary. Mortality is high in patients with circulatory arrest on admission, but low in all other patients.
Capillary leakage syndrome (CLS) is a frequent complication in sepsis, characterized by loss of intravasal fluids leading to generalized edema and hemodynamic instability despite massive fluid therapy. In spite of its importance no standardized diagnostic criteria are available for CLS. Prospective clinical study. 1,800-bed university hospital Six septic shock patients with CLS were compared to six control patients. CLS was clinically determined by generalized edema, positive fluid balance, and weight gain. Plasma volume was measured by indocyanine green, red blood cell volume by chromium-51 labeled erythrocytes, and colloid osmotic pressure before and 90 min after the administration of 300 ml 20% albumin. Extracellular water (ECW) was measured using the inulin distribution volume and bioelectrical impedance analysis. Red blood cells averaged 20.2 +/- 1.0 ml/ kg body weight in CLS patients and 23.3 +/- 4.1 in controls. ECW was higher in CLS patients than in controls (40.0 +/- 6.9 vs. 21.7 +/- 3.71; p< 0.05). ECW of inulin was correlated with that measured by bioelectrical impedance analysis (r = 0.74, p< 0.01). The increase in colloid osmotic pressure over the 90 min was less in CLS patients than in controls (1.1 +/- 0.3 vs. 2.8 +/- 1.3 mmHg;p< 0.05). These results suggest that measurements of an increased ECW using bioelectrical impedance analysis combined with a different response of colloid osmotic pressure to administration of albumin can discriminate noninvasively between patients with and those without CLS.
To compare the efficacy, safety, and cost of midazolam and propofol in prolonged sedation of critically ill patients. Randomized, prospective study. General intensive care unit (ICU) in a 1100-bed teaching hospital. 67 critically ill, mechanically ventilated patients. Patients were invasively monitored and mechanically ventilated. A loading dose [midazolam 0.11 +/- 0.02 (SEM), propofol 1.3 +/- 0.2] was administered, followed by continuous infusion, titrated to achieve a predetermined sedation score. Sedation was continued as long as clinically indicated. Mean duration of sedation was 141 and 99 h (NS) for midazolam and propofol, respectively, at mean hourly doses of 0.070 +/- 0.003 midazolam and 1.80 +/- 0.08 propofol. Overall, 68% of propofol patients versus 31% of midazolam (p < 0.001) patients had a > 20% decrease in systolic blood pressure after the loading dose, and 26 versus 45% (p < 0.01) showed a 25% decrease in spontaneous minute volume. Propofol required more daily dose adjustments (2.1 +/- 0.1 vs 1.4 +/- 0.1, p < 0.001). Nurse-rated quality of sedation with midazolam was higher (8.2 +/- 0.1 vs 7.3 +/- 0.1 on a 10-cm visual analog scale, p < 0.001). Resumption of spontaneous respiration was equally rapid. Recovery was faster after propofol (p < 0.02), albeit with a higher degree of agitation. Amnesia was evident in all midazolam patients but in only a third of propofol patients. The cost of propofol was 4-5 times higher. Both drugs afforded reliable, safe, and controllable long-term sedation in ICU patients and rapid weaning from mechanical ventilation. Midazolam depressed respiration, allowed better maintenance of sedation, and yielded complete amnesia at a lower cost, while propofol caused more cardiovascular depression during induction.
To investigate the effects of tyrphostin AG 126, a tyrosine kinase inhibitor, on the multiple organ failure (MOF) caused by zymosan in the rat. Zymosan (500 mg/kg, suspended in saline solution, i.p.) causes an enhanced formation of reactive oxygen species, which contribute to the pathophysiology of MOF. After zymosan or saline administration, animals were monitored for 12 days. Treatment of rats with tyrphostin AG 126 (10 mg/kg, 3 mg/kg or 1 mg/kg intraperitoneally, 1 h and 6 h after zymosan) attenuated the peritoneal exudation and the migration of polymorphonuclear cells caused by zymosan in a dose-dependent fashion. Tyrphostin AG 126 also attenuated the lung, liver, and intestinal injury (histology) as well as the increase in the levels of myeloperoxidase and malondialdehyde caused by zymosan in the lung, liver, and intestine. Immunohistochemical analysis for nitrotyrosine, poly (ADP-ribose) polymerase (PAR), iNOS, and COX-2 revealed a positive staining in lung, liver and intestine from zymosan-treated rats. The degree of staining for nitrotyrosine, PAR, iNOS, and COX-2 were markedly reduced in tissue sections obtained from zymosan-treated rats which had received tyrphostin AG 126. Furthermore, treatment of rats with tyrphostin AG 126 significantly reduced the production of peroxynitrite and of pro-inflammatory cytokines TNF-alpha and IL-1beta. This study provides the first evidence that the protein kinase inhibitor tyrphostin AG 126 attenuates the degree of MOF associated with zymosan-induced peritonitis in the rat.
To assess the efficacy of gastric intramucosal pH for the evaluation of tissue perfusion and prediction of hemodynamic complications in critically ill children. Open prospective study without controls. Pediatric intensive care unit (ICU) of a tertiary care university pediatric hospital. Thirty critically ill children (16 boys and 14 girls), age range: 3 months-12 years. A tonometry catheter was placed in the stomach of all patients on admission to the pediatric ICU. Simultaneous tonometry and arterial gas measurements were made on admittance and every 6-12 h throughout the study; a total of 202 measurements were made. The catheter was removed after extubation and/or when the patient was hemodynamically stable. Intramucosal pH was calculated using the Henderson-Hasselbalch equation based on the pCO2 of the tonometer and arterial bicarbonate. Intramucosal pH values between 7.30 and 7.45 were considered to be normal. The patient's condition was analyzed using the Pediatric Risk Mortality Score (PRISM). The relations between intramucosal pH and the presence of major hemodynamic complications (cardiopulmonary arrest, shock), minor hemodynamic complications (hypotension, hypovolemia or arrhythmia), death, PRISM score and the duration of the stay in the pediatric ICU were analyzed. Intramucosal pH on admission was 7.48 +/- 0.15 on average (range 7.04-7.68). Five patients (16%) had an intramucosal pH lower than 7.30 on admission; these patients did not have a higher incidence of hemodynamic complications. The 16 patients (53%) who had an intramucosal pH of less than 7.30 at some time during the course of their disease had more hemodynamic complications than the patients who did not have pH lower than 7.30 (p < 0.0001). Every case of cardiopulmonary arrest and shock was related to intramucosal pH of less than 7.30. Patients with major complications (cardiopulmonary arrest and shock) had lower intramucosal pHs than those with minor hemodynamic complications (p = 0.03); similarly, they had low intramucosal pH readings more often than those with minor complications (p = 0.0032). Intramucosal pH values less than 7.30 had a sensitivity of 90% and a specificity of 98% as a predictor of hemodynamic complications. There was no relation between intramucosal pH lower than 7.30 and either PRISM or the duration of the stay in the pediatric ICU. Patients with intramucosal pH less than 7.20 had a higher PRISM than the patients who did not have pH lower than 7.20 (p < 0.05). A patient who died during the study due to cardiopulmonary arrest had prior intramucosal pH measurements of 7.23 and 7.10, and three patients died of late complications after the end of the study. Hemodynamic complications were not detected with arterial pH. Gap pH (arterial pH-intramucosal pH) and standard pH measurements yielded the same results as gastric intramucosal pH. Intramucosal pH could provide a useful early indication of hemodynamic complications in critically ill children.
Objective: To determine whether a) pre-operative measurement of gastric intramucosal pHi is predictive for mortality and morbidity in high-risk surgical patients and b) peri-operative improvement of global oxygen delivery (DO2) with fluids and dopexamine leads to increased gastric pHi and c) either improved global perfusion or improved splanchnic perfusion is related to the prevention of multiple organ failure (MOF). Design: Retrospective analysis of a double-blind, placebo-controlled, randomised study. Setting: General intensive care units from 14 hospitals. Patients: Two hundred eighty-six high-risk surgical patients. Interventions: Swan-Ganz and tonometer catheter placement; patients were stabilised pre-operatively using fluids, blood and/or oxygen to preset goals before receiving placebo or two doses of dopexamine (0.5 or 2.0 peri-operatively. Measurements and results: Haemodynamic assessment (including DO2 and oxygen consumption (VO2)) was performed together with measurement of gastric mucosal pHi pre-operatively and directly, 2, 6, 12, 24 and 36 h post-operatively. Retrospectively, patients were divided pre-operatively into two sub-groups based on the optimal cut-off value for mortality of the first pHi measurement after induction of anaesthesia as calculated by a receiver operator characteristic (ROC) curve analysis --low pHi group (< 7.35) and normal pHi (> or =7.35). Mortality in the low pHi, was higher than in the normal pHi, group (16.8 vs 2.3%; p = 0.0001). In the normal pHi group dopexamine, which was given prior to the first pHi measurement, had no effect on pHi, while DO2 increased significantly. In this group MOF score and number of patients with MOF remained similar for the treatment sub-groups. In the low pHi group gastric pHi increased significantly during dopexamine infusion (p = 0.008), despite the lack of an increase in DO2 and VO2. In this group the MOF score and the number of patients developing MOF decreased significantly with the use of dopexamine (p = 0.04). In both groups bicarbonate levels remained similar for the treatment subgroups. Conclusions: In high-risk surgical patients pre-operative measurement of pHi was predictive for mortality. The peri-operative response of pHi to dopexamine seemed to be dependent on pre-operative gastric pHi.
(1) To establish risk factors for the development of delirium in an intensive care unit (ICU) and (2) to determine the effect of delirium on morbidity, mortality and length of stay. Prospective study. Sixteen-bed medical/surgical ICU in a university hospital. Two hundred and sixteen consecutive patients admitted to the ICU for more than 24 h during 5 months were included in the study. Medical history, selected laboratory values, drugs received and factors that may influence patient psychological and emotional well-being were noted. All patients were screened with a delirium scale. A psychiatrist confirmed the diagnosis of delirium. Major complications such as self-extubation and removal of catheters, as well as mortality and length of stay were recorded. Forty patients (19%) developed delirium; of these, one-third were not agitated. In the multivariate analysis hypertension, smoking history, abnormal bilirubin level, epidural use and morphine were statistically significantly associated with delirium. Traditional factors associated with the development of delirium on general ward patients were not significant in our study. Morbidity (self-extubation and removal of catheters), but not mortality, was clearly increased. Predictive risk factors for the development of delirium in studies outside the ICU may not be applicable to critically ill patients. Delirium is associated with increased morbidity. Awareness of patients at risk may lead to better recognition and earlier intervention.
Paediatric admissions to a general ITU are briefly reviewed. Although it may not be an ideal arrangement it is suggested that children can be safely cared for in such a unit. The problems created by the child with acute upper airway obstruction are discussed to demonstrate the way in which the unit is organized to receive acutely ill children.
To examine plasma levels and circadian rhythm of interleukin 13 (IL-13), tumour necrosis factor (TNF) alpha and total serum cortisol in the systemic inflammatory response syndrome (SIRS). Prospective observational study in a 12-bed medical-surgical ICU of a 500-bed university hospital. Ten patients with SIRS and eight controls. Arterial blood was sampled hourly for 24 h for measurement of plasma IL-13, TNF- alpha, cortisol and white blood cell count (WCC) differential within 24 h of development of SIRS. There were significantly higher plasma IL-13 levels in SIRS patients than in controls (1282 vs. 713 pg/ml). IL-13 was significantly higher in patients with a diagnosis of sepsis than in those with non-infectious causes of SIRS (2080 vs. 515 pg/ml). In SIRS the elevation in IL-13 was associated with higher TNF-alpha and reduced WCC. A circadian rhythm was observed in plasma IL-13 secretion. No distinct circadian rhythm was noted for TNF- alpha, and the normal circadian rhythm of serum cortisol was lost. The anti-inflammatory cytokine IL-13 is elevated in early SIRS. Its plasma level exhibits a circadian rhythm and may be modulated in part by TNF-alpha. SIRS patients have disruption of the normal circadian rhythm of serum cortisol.
Top-cited authors
Richard Beale
  • Guy's and St Thomas' NHS Foundation Trust
Rui Moreno
  • Centro Hospitalar Universitário de Lisboa Central
Konrad Reinhart
  • Universitätsklinikum Jena
Jean-Louis Vincent
  • Université Libre de Bruxelles
Richard dellinger
  • Cooper University Hospital