Influence of ICU-bed availability on ICU admission decisions.

BACKGROUND: The potential influence of bed availability on triage to intensive care unit (ICU) admission is among the factors that may influence the ideal ratio of ICU beds to population: thus, high bed availability (HBA) may result in the admission of patients too well or too sick to benefit, whereas bed scarcity may result in refusal of patients likely to benefit from ICU admission. METHODS: Characteristics and outcomes of patient admitted in four ICUs with usual HBA, defined by admission refusal rate less than 11 % because of bed unavailability, were compared to patients admitted in six ICUs with usual low bed availability (LBA), i.e., an admission refusal rate higher than 10 % during a 90-day period. RESULTS: Over the 90 days, the mean number of days with no bed available was 30 ± 16 in HBA units versus 48 ± 21 in LBA units (p < 0.01). The proportion of admitted patients was significantly higher in the HBA (80.1 %; n = 659/823) than in the LBA units [61.6 %: n = 480/779; (p < 0.0001)]. The proportion of patients deemed too sick to benefit from admission was higher in LBA (9.0 %; n = 70) than in the HBA (6.3 %; n = 52) units (p < 0.05). The HBA group had a significantly greater proportion of patients younger than 40 years of age (22.5 %; n = 148 versus 14 %; n = 67 in LBA group; p < 0.001) and higher proportions of patients with either high or low simplified acute physiologic score II values. CONCLUSIONS: Bed availability affected triage decisions. Units with HBA trend to admit patients too sick or too well to benefit.

Therapeutic hypothermia after nonshockable cardiac arrest: the HYPERION multicenter, randomized, controlled, assessor-blinded, superiority trial.

BACKGROUND: Meta-analyses of nonrandomized studies have provided conflicting data on therapeutic hypothermia, or targeted temperature management (TTM), at 33°C in patients successfully resuscitated after nonshockable cardiac arrest. Nevertheless, the latest recommendations issued by the International Liaison Committee on Resuscitation and by the European Resuscitation Council recommend therapeutic hypothermia. New data are available on the adverse effects of therapeutic hypothermia, notably infectious complications. The risk/benefit ratio of therapeutic hypothermia after nonshockable cardiac arrest is unclear. METHODS: HYPERION is a multicenter (22 French ICUs) trial with blinded outcome assessment in which 584 patients with successfully resuscitated nonshockable cardiac arrest are allocated at random to either TTM between 32.5 and 33.5°C (therapeutic hypothermia) or TTM between 36.5 and 37.5°C (therapeutic normothermia) for 24 hours. Both groups are managed with therapeutic normothermia for the next 24 hours. TTM is achieved using locally available equipment. The primary outcome is day-90 neurological status assessed by the Cerebral Performance Categories (CPC) Scale with dichotomization of the results (1 + 2 versus 3 + 4 + 5). The primary outcome is assessed by a blinded psychologist during a semi-structured telephone interview of the patient or next of kin. Secondary outcomes are day-90 mortality, hospital mortality, severe adverse events, infections, and neurocognitive performance. The planned sample size of 584 patients will enable us to detect a 9% absolute difference in day-90 neurological status with 80% power, assuming a 14% event rate in the control group and a two-sided Type 1 error rate of 4.9%. Two interim analyses will be performed, after inclusion of 200 and 400 patients, respectively. DISCUSSION: The HYPERION trial is a multicenter, randomized, controlled, assessor-blinded, superiority trial that may provide an answer to an issue of everyday relevance, namely, whether TTM is beneficial in comatose patients resuscitated after nonshockable cardiac arrest. Furthermore, it will provide new data on the tolerance and adverse events (especially infectious complications) of TTM at 32.5-33.5°C. TRIAL REGISTRATION: ClinicalTrials.gov: NCT01994772 .

Acute respiratory distress syndrome (ARDS)-associated acute cor pulmonale and patent foramen ovale: a multicenter noninvasive hemodynamic study.

INTRODUCTION: Acute cor pulmonale (ACP) and patent foramen ovale (PFO) remain common in patients under protective ventilation for acute respiratory distress syndrome (ARDS). We sought to describe the hemodynamic profile associated with either ACP or PFO, or both, during the early course of moderate-to-severe ARDS using echocardiography. METHODS: In this 32-month prospective multicenter study, 195 patients with moderate-to-severe ARDS were assessed using echocardiography during the first 48 h of admission (age: 56 (SD: 15) years; Simplified Acute Physiology Score: 46 (17); PaO2/FiO2: 115 (39); VT: 6.5 (1.7) mL/kg; PEEP: 11 (3) cmH2O; driving pressure: 15 (5) cmH2O). ACP was defined by the association of right ventricular (RV) dilatation and systolic paradoxical ventricular septal motion. PFO was detected during a contrast study using agitated saline in the transesophageal bicaval view. RESULTS: ACP was present in 36 patients, PFO in 21 patients, both PFO and ACP in 8 patients and the 130 remaining patients had neither PFO nor ACP. Patients with ACP exhibited a restricted left ventricle (LV) secondary to RV dilatation and had concomitant RV dysfunction, irrespective of associated PFO, but preserved LV systolic function. Despite elevated systolic pulmonary artery pressure (sPAP), patients with isolated PFO had a normal RV systolic function. sPAP and PaCO2 levels were significantly correlated. CONCLUSIONS: In patients under protective mechanical ventilation with moderate-to-severe ARDS, ACP was associated with LV restriction and RV failure, whether PFO was present or not. Despite elevated sPAP, PFO shunting was associated with preserved RV systolic function.

Prevalence and prognostic value of acute cor pulmonale and patent foramen ovale in ventilated patients with early acute respiratory distress syndrome: a multicenter study.

PURPOSE: We sought to determine the prevalence of and factors associated with acute cor pulmonale (ACP) and patent foramen ovale (PFO) at the early phase of acute respiratory distress syndrome (ARDS), and to assess their relation with mortality. METHODS: In this prospective multicenter study, 200 patients submitted to protective ventilation for early moderate to severe ARDS [PaO2/F(I)O2: 115 ± 39 with F(I)O2: 1; positive end-expiratory pressure (PEEP): 10.6 ± 3.1 cmH2O] underwent transthoracic (TTE) and transesophageal echocardiography (TEE) <48 h after admission. Echocardiograms were independently interpreted by two experts. Factors associated with ACP, PFO, and 28-day mortality were identified using multivariate regression analysis. RESULTS: TEE depicted ACP in 45/200 patients [22.5%; 95% confidence interval (CI) 16.9-28.9%], PFO in 31 patients (15.5%; 95% CI 10.8-21.3%), and both ACP and PFO in 9 patients (4.5%; 95% CI 2.1-8.4%). PFO shunting was small and intermittent in 27 patients, moderate and consistent in 4 patients, and large or extensive in no instances. PaCO2 >60 mmHg was strongly associated with ACP [odds ratio (OR) 3.70; 95% CI 1.32-10.38; p = 0.01]. No factor was independently associated with PFO, with only a trend for age (OR 2.07; 95% CI 0.91-4.72; p = 0.08). Twenty-eight-day mortality was 23%. Plateau pressure (OR 1.15; 95% CI 1.05-1.26; p < 0.01) and air leaks (OR 5.48; 95% CI 1.30-22.99; p = 0.02), but neither ACP nor PFO, were independently associated with outcome. CONCLUSIONS: TEE screening allowed identification of ACP in one-fourth of patients submitted to protective ventilation for early moderate to severe ARDS. PFO shunting was less frequent and never large or extensive. ACP and PFO were not related to outcome.

Refusal of intensive care unit admission due to a full unit: impact on mortality.

RATIONALE: Intensive care unit (ICU) beds are a scarce resource, and patients denied intensive care only because the unit is full may be at increased risk of death. OBJECTIVE: To compare mortality after first ICU referral in admitted patients and in patients denied admission because the unit was full. METHODS: Prospective observational multicenter cohort study of consecutive patients referred for ICU admission during two 45-day periods, conducted in 10 ICUs. MEASUREMENTS AND MAIN RESULTS: Of 1,762 patients, 430 were excluded from the study, 116 with previously denied admission to another ICU and 270 because they were deemed too sick or too well to benefit from ICU admission. Of the remaining 1,332 patients, 1,139 were admitted, and 193 were denied admission because the unit was full (65 were never admitted, 39 were admitted after bumping of another patient, and 89 were admitted on subsequent referral). Crude Day 28 and Day 60 mortality rates in the nonadmitted and admitted groups were 30.1 versus 24.3% (P = 0.07) and 33.3 versus 27.2% (P = 0.06), respectively. Day 28 mortality adjusted on age, previous disease, Glasgow scale score less than or equal to 8, shock, creatinine level greater than or equal to 250 µmol/L, and prothrombin time greater than or equal to 30 seconds was nonsignificantly higher in patients refused ICU admission only because of a full unit compared with patients admitted immediately. Patients admitted after subsequent referral had higher mortality rates on Day 28 (P = 0.05) and Day 60 (P = 0.04) compared with directly admitted patients. CONCLUSIONS: Delayed ICU admission due to a full unit at first referral is associated with increased mortality.