Artificial intelligence algorithms predict the efficacy of analgesic cocktails prescribed after orthopedic surgery.

BACKGROUND: Mixtures ('cocktails') of various analgesics are more effective in controlling post-operative pain because of potential synergetic effects. Few studies have investigated such effects in large combinations of analgesics and no studies have determined the probabilities of effectiveness. METHODS: We used one-hot encoding of the categorical variables reported pain levels and the administered cocktails (from a total of eight analgesics) and then applied an unsupervised neural network and then the unsupervised DBSCAN algorithm to detect clusters of cocktails. We used Bayesian statistics to classify the effectiveness of these cocktails. RESULTS: Of the 61 different cocktails administered to 750 patients, we found that four combinations of three to four analgesics were by far the most effective. All these cocktails contained Metamizole and Paracetamol; three contained Hydromorphone and two contained Diclofenac and one Diclofenac-Orphenadrine. The ML probability that these cocktails decreased pain levels ranged from 0.965 to 0.981. Choice of a most effective cocktail involves choosing the optimum in a 4-dimensional parameter space: maximum probability of efficacy, confidence interval about maximum probability, fraction of patients with increase in pain levels, relative number of patients with successful pain level decrease. CONCLUSIONS: We observed that administering one analgesic or at most two is not effective. We found no statistical indicators that interactions between analgesics in the most effective cocktails decreased their effectiveness. Pairs of most effective cocktails differed by the addition of only one analgesic (Diclofenac-Orphenadrine for one pair and Hydromorphone for the other). We conclude that the listed cocktails are to be recommended.

Monitoring of intracranial pressure in patients with severe traumatic brain injury: an Austrian prospective multicenter study.

OBJECTIVE: The goals of this study were to elucidate reasons why patients did or did not receive intracranial pressure (ICP) monitoring and to describe factors influencing hospital mortality after severe traumatic brain injury (TBI). DESIGN: Prospective multicenter cohort study. PATIENTS AND PARTICIPANTS: 88,274 patients consecutively admitted to 32 medical, surgical and mixed Austrian ICUs between 1998 and 2004. INTERVENTIONS: None. MEASUREMENTS AND RESULTS: 1,856 patients (2.1% of all ICU admissions) exhibited severe TBI (GCS < 9); of these, 1,031 (56%) had ICP monitoring. The "worst" and the "best" cases were both less likely to receive ICP monitoring. Younger patients, female patients, and patients with isolated TBI were more likely to receive ICP monitoring. Compared with large centers ICP was monitored more frequently [odds ratio (OR) 3.09, CI 2.42-3.94] in medium-sized centers. The 20% of patients with the highest likelihood to receive ICP monitoring were monitored in 91% of cases, and had the lowest hospital mortality (31%, OR 0.78, CI 0.37-1.64). Multivariate analysis revealed that severity of illness, TBI severity, isolated TBI, and the number of cases treated per year were associated with hospital outcome. Compared with the large centers, ORs for hospital mortality were 1.85 (CI 1.42-2.40) for patients from medium-sized centers and 1.91 (CI 1.24-2.93) for patients from small centers. CONCLUSIONS: ICP monitoring may possibly have some beneficial effects, but this needs further evaluation. Patients with severe TBI should be admitted to experienced centers with high patient volumes since this might improve hospital mortality rates.

Non-isotopic tyrosine kinetics using an alanyl-tyrosine dipeptide to assess graft function in liver transplant recipients - a pilot study.

BACKGROUND: Although accurate assessment of liver function in liver transplant recipients is of crucial importance for optimal timing of the procedure and for determining graft viability, none of the many available methods has proven reliable in the clinical routine. Thus, a novel non-isotopic assay of tyrosine kinetics using the tyrosine-containing dipeptide L-alanyl-L-tyrosine (Ala-Tyr) was tested for its clinical feasibility in patients undergoing orthotopic liver transplantation (OLT). METHODS: Plasma levels of tyrosine and clearance of tyrosine released after infusion of the dipetide Ala-Tyr were assessed before and one day after OLT in 10 liver transplant recipients with normal graft function, also in three organ donors and in three recipients showing poor graft function. Standard laboratory parameters (e.g. aminotransferases) and the plasma disappearance rate of indocyanine green were also measured. RESULTS: Following uneventful OLT, tyrosine plasma levels (before 127 +/- 15 micromol/vs. post-OLT 52 +/- 6 micromol/l, P < 0.05) and kinetics (tyrosine clearance: before 206 +/- 77 ml/min vs. post-OLT 371 +/- 109 ml/min, P < 0.05) were normalized. In cases of severe graft dysfunction, tyrosine kinetics (tyrosine clearance: 238 +/- 61 ml/min) resembled the situation in end-stage liver disease, whereas no such correlation was seen with conventional markers of liver function. Organ preservation had only a minor impact on tyrosine kinetics (n.s.). CONCLUSION: OLT rapidly normalizes both the plasma levels and the kinetics of tyrosine. Graft failure is associated with an immediate rise in plasma tyrosine levels and a delay in tyrosine elimination. Our results show that tyrosine clearance using the dipetide Ala-Tyr is a suitable non-isotopic, non-invasive indicator of graft viability in the early postoperative course following OLT.

Use of CytoSorb in Traumatic Amputation of the Forearm and Severe Septic Shock.

Severe trauma associated with later disability and mortality still constitutes a major health and socioeconomic problem throughout the world. While primary morbidity and mortality are mostly related to initial injuries and early complications, secondary lethality is strongly linked to the development of systemic inflammatory response syndrome, sepsis, and ultimately multiple organ dysfunction syndrome. We herein report on a 49-year-old male patient who was admitted to the hospital after a traumatic amputation of his right forearm that was cut off while working on a landfill. After initial treatment for shock, he received immediate replantation and was transferred to the ICU. Due to the anticipated risk of a complex infection, continuous renal replacement therapy in combination with CytoSorb was initiated. During the course of the combined treatment, a rapid improvement in hemodynamics was noticed, as well as a significant reduction of IL-6 and lactate levels. Despite a recurring septic episode and the necessity for amputation, the patient clinically stabilized and underwent complete recovery. The early treatment with a combination of CVVHDF and CytoSorb was accompanied by an attenuation of the systemic inflammatory reaction, which subsided without major or permanent organ damage, despite the impressive pathogen spectrum and the pronounced local damage.

Creating Clinical Fuzzy Automata with Fuzzy Arden Syntax.

Formal constructs for fuzzy sets and fuzzy logic are incorporated into Arden Syntax version 2.9 (Fuzzy Arden Syntax). With fuzzy sets, the relationships between measured or observed data and linguistic terms are expressed as degrees of compatibility that model the unsharpness of the boundaries of linguistic terms. Propositional uncertainty due to incomplete knowledge of relationships between clinical linguistic concepts is modeled with fuzzy logic. Fuzzy Arden Syntax also supports the construction of fuzzy state monitors. The latter are defined as monitors that employ fuzzy automata to observe gradual transitions between different stages of disease. As a use case, we re-implemented FuzzyARDS, a previously published clinical monitoring system for patients suffering from acute respiratory distress syndrome (ARDS). Using the re-implementation as an example, we show how key concepts of fuzzy automata, i.e., fuzzy states and parallel fuzzy state transitions, can be implemented in Fuzzy Arden Syntax. The results showed that fuzzy state monitors can be implemented in a straightforward manner.

Molecular adsorbent recirculating system and hemostasis in patients at high risk of bleeding: an observational study.

INTRODUCTION: Liver failure is associated with reduced synthesis of clotting factors, consumptive coagulopathy, and platelet dysfunction. The aim of the study was to evaluate the effects of liver support using a molecular adsorbent recirculating system (MARS) on the coagulation system in patients at high risk of bleeding. METHODS: We studied 61 MARS treatments in 33 patients with acute liver failure (n = 15), acute-on-chronic liver failure (n = 8), sepsis (n = 5), liver graft dysfunction (n = 3), and cholestasis (n = 2). Standard coagulation tests, standard thromboelastography (TEG), and heparinase-modified and abciximab-fab-modified TEG were performed immediately before and 30 minutes after commencement of MARS, and after the end of MARS treatment. Prostaglandin I2 was administered extracorporeally to all patients; 17 patients additionally received unfractioned heparin. RESULTS: Three moderate bleeding complications in three patients, requiring three to four units of packed red blood cells, were observed. All were sufficiently managed without interrupting MARS treatment. Although there was a significant decrease in platelet counts (median, 9 G/l; range, -40 to 145 G/l) and fibrinogen concentration (median, 15 mg/dl; range, -119 to 185 mg/dl) with a consecutive increase in thrombin time, the platelet function, as assessed by abciximab-fab-modified TEG, remained stable. MARS did not enhance fibrinolysis. CONCLUSION: MARS treatment appears to be well tolerated during marked coagulopathy due to liver failure. Although MARS leads to a further decrease in platelet count and fibrinogen concentration, platelet function, measured as the contribution of the platelets to the clot firmness in TEG, remains stable. According to TEG-based results, MARS does not enhance fibrinolysis.

The severe traumatic brain injury in Austria: early rehabilitative treatment and outcome.

BACKGROUND: Severe traumatic brain injury (TBI) is a great economical and logistic problem in the health care system which reduces the quality of life and productivity of the patient. The purpose of this study is to evaluate the outcome of patients after severe brain trauma according to the course of their rehabilitation. METHODS: Patients with TBI were divided into three groups. Group A; after early rehabilitation (n = 16), B; following a standard rehabilitation procedure after work accidents (n = 34) and C; undergone standard rehabilitation procedure after accidents at home (n = 12). Glasgow Coma Scale (GCS), Post traumatic amnesia (PTA) during acute care, Glasgow Outcome Scale Extended (GOSE) and Functional Independence Measurement (FIM) were measured before and after rehabilitation. Long-term outcomes (12 months post injury) were measured with the Community Integration Questionnaire (CIQ). RESULTS: Group A showed a significantly shorter time span from hospital admission until rehabilitation center admission than B and C (p < 0.001). PTA was significantly lower in group B than in group A (p = 0.038). GOSE of patients within group C was significantly lower (p = 0.004) at hospital discharge. FIM was significantly higher in B (p = 0.005) at the time of admission to rehabilitation center. At the time of discharge FIM showed no significant differences between the groups. CIQ showed a trend to improving scores in group A. CONCLUSION: Despite the similar level of severity of TBI and outcome prognosis group A showed the best rehabilitation effect and long-term outcome.

Iced versus room temperature injectate for assessment of cardiac output, intrathoracic blood volume, and extravascular lung water by single transpulmonary thermodilution.

PURPOSE: To assess the accuracy of iced versus room (RT) temperature single transpulmonary thermodilution (STPD) measurements for cardiac output, intra-thoracic blood, volume and extravascular lung water. MATERIALS AND METHODS: We studied 15 critically ill patients in a surgical intensive care unit with sepsis/septic shock (n = 8), pancreatitis (n = 2), acute liver failure (n = 2), orthotopic liver transplantation (n = 2) and lung resection (n = 1). All patients were sedated and mechanically ventilated. A 4-French femoral arterial catheter was inserted into each patient and connected to the pulse contour computer system (PiCCO). The pulse contour computer was then consecutively calibrated by triplicate STPD with 20 mL of RT and iced saline solution. The measurements with RT injectate were performed with a special in-line sensor adapted for measurement with RT injectate. All measurements were completed in less than 10 min. RESULTS: A total of 144 measurements were carried out. Linear regression analysis revealed good correlation between the two methods [r = 0.95; r = 0.91 and r = 0.97 for iced v RT cardiac index (CI), intrathoracic blood volume index (ITBVI) and extravascular lung water index (ELWI) respectively]. The bias +/- 2 * standard deviation of difference was -0.2 +/- 0.7 L/min/m2 for CIIT v CIRT; -4,9 +/- 194 mL/m2 for ITBVIIT v ITBVIRT and -0.535 +/- 1,5 mL/kg for ELWIIT v ELWIRT. CIRT and ELWIRT were measured slightly higher compared to IT injectate (P <.05). CONCLUSIONS: CI, ITBVI, and ELWI assessed by STPD with RT injectate are well correlated with measurements by iced injectates. According to our results room temperature injectates can be used in critically ill patients for assessment of CI, ITBVI and ELWI, which is more convenient for both the patients and medical staff and is also less expensive.

Bilateral lung transplantation with intra- and postoperatively prolonged ECMO support in patients with pulmonary hypertension.

OBJECTIVE: Lung transplantation for pulmonary hypertension (PH) is usually performed on cardiopulmonary bypass, with the disadvantage of full systemic anticoagulation, uncontrolled allograft reperfusion and aggressive ventilation. These factors can be avoided with intra- and postoperatively prolonged extracorporeal membrane oxygenator (ECMO) support. PATIENTS AND METHODS: Between February 1999 and March 2001, 17 consecutive patients with PH (systolic pulmonary artery pressure >70 mmHg) of different etiologies underwent bilateral lung transplantation (BLTX). There were 11 females and six males in the age range from 7 to 50 years (mean age, 28.4+/-12.9 years). Six patients were preoperatively hospitalized, four in the intensive care unit (ICU), one was on ECMO for 3 weeks pretransplantation, and one was resuscitated and bridged with ECMO for 1 week until transplantation. Femoral venoarterial ECMO support with heparin-coated circuits was set up after induction of anesthesia and discontinued at the end of surgery (n=3) or extended for 12 h median into the postoperative period (n=14). Postoperative ventilation pressure was kept below 25 mmHg. Allograft function at 2 h after discontinuation of ECMO, outcome and adverse events were monitored in all patients. Mean follow up time was 18+/-11.4 months. RESULTS: The perioperative mortality was 5.9% (n=1). Arterial oxygen pressure measured 2 h after weaning from ECMO, and under standard mechanical ventilation with a peak pressure of 25 mmHg and inspired oxygen fraction of 0.4, was 157+/-28 mmHg. The mean pulmonary artery pressures were reduced to 29+/-3,4 from 66+/-15 mmHg before transplantation. Postoperative complications included rethoracotomy due to bleeding (n=4) and temporary left ventricular failure (n=4). Median ICU stay was 12 days. Incidence of rejection within the first 100 days was 0.4 per patient. CONCLUSION: BLTX with intraoperative and postoperatively prolonged ECMO support provides excellent initial organ function due to optimal controlled reperfusion and non-aggressive ventilation. This results in improved outcome even in advanced forms of PH.

Liver support in fulminant liver failure after hemorrhagic shock.

Acute liver failure (ALF) is a rare clinical syndrome associated with a mortality of up to 80% and its management remains an interdisciplinary challenge. Despite recent improvements in intensive care management, the mortality of patients with ALF remains high and is related to complications such as cerebral edema, sepsis and multiple organ failure. Emergency orthotopic liver transplantation (OLT) is currently the only effective treatment for those patients who are unlikely to recover spontaneously. Nevertheless, OLT is not always possible because of the shortage of the organs and/or complications related to ALF. Newly introduced liver-assist devices can temporarily support the patient's liver until native liver recovers or can serve as a bridging device until a liver graft is available. The support devices use both cell-based and non-cell-based techniques. One of the latest non-cell-based extracorporeal hepatic support devices, the molecular adsorbent recycling system (MARS), is based on the concept of albumin dialysis. MARS utilises selective hemodiafiltration with countercurrent albumin dialysis aiming to selectively remove both water-soluble and albumin-bound toxins of the low and middle molecular-weight range. We report on a young patient who presented with clinical symptoms of ischemic hepatitis and multi-organ failure (APACHE II score 38-->predicted postoperative mortality 87%) due to prolonged hemorrhagic shock. OLT was contraindicated because of history of pancreas cancer with metastases. It was necessary to use aggressive conservative therapy and an extracorporeal liver-assist device until liver regeneration began and hemodynamic conditions were stable. The patient underwent five treatments with MARS. During the treatment, there were improvements of hemodynamics, respiratory function, acid-base disturbances and laboratory parameters. The plasma disappearance rate of indocyanine green, a parameter of dynamic liver function, improved during MARS treatment. Although repeated neurological examination predicted diffuse brain damage (brain oedema, decreased cerebral blood flow), the patient recovered without any neurological deficits. The patient survived and was discharged from the hospital in good condition. In this case MARS treatment was successful in supporting the patient through the most critical period of ALF.

Septic shock secondary to ß-hemolytic streptococcus-induced necrotizing fasciitis treated with a novel cytokine adsorption therapy.

INTRODUCTION: Numerous animal studies and preliminary data from a clinical trial in septic patients demonstrated that a decrease in blood cytokine levels using an extracorporeal cytokine filter (CytoSorb) can effectively attenuate the inflammatory response during sepsis and possibly improve outcomes. METHODS: A 60-year-old female was admitted to hospital due to a forearm fracture. After surgical wound care by osteosynthesis the patient developed surgical wound infection which progressed to necrotizing fasciitis. All diagnostic criteria for SIRS were evident with additional proven infection from ß-hemolytic streptococcus. On admission to the ICU, the patient presented a full picture of multiple organ dysfunction syndrome due to septic shock including kidney failure, lung failure as well as thrombocytopenia, metabolic acidosis, and arterial hypotension. RESULTS: After one day on mechanical ventilation and an IL-6 level of 70,000 pg/ml the patient was treated with CytoSorb therapy over a period of four days, resulting in a significant reduction of IL-6 to 66 pg/ml and an overall improvement of the patient's condition. Despite the necessity of enucleation, the patient was successfully stabilized until control of the surgical infectious source was achieved. Importantly, treatment was safe and well-tolerated, without any adverse events. CONCLUSIONS: This is the first report of the clinical application of CytoSorb hemoadsorption in combination with a CRRT in a patient with septic shock. CytoSorb as described was able to significantly reduce IL-6 plasma levels and decrease vasopressor need while no adverse and device-related events occurred. CytoSorb seems to be an interesting and safe extracorporeal therapy to stabilize and bridge septic patients to surgery or recovery.

Molecular adsorbent recirculating system in patients with early allograft dysfunction after liver transplantation: a pilot study.

Early allograft dysfunction (EAD) after orthotopic liver transplantation (OLT) causes marked morbidity and mortality. We conducted a prospective pilot study to assess the safety and efficacy of molecular adsorbent recirculating system (MARS) in treatment of EAD after OLT. Twelve consecutive adult liver allograft recipients with a median age of 48 years, 9 of whom were male, were prospectively included and supported with MARS. EAD was defined as the presence of at least 2 of the following: serum bilirubin >10 mg/dL, prothrombin time <40%, aspartate aminotransferase or alanine transferase >1,000 U/L, and plasma disappearance rate of indocyanine green (PDR(ICG)) <10% per minute within 72 hours after reperfusion. One-year patient and graft survival was 66%. There was a significant decrease in serum bilirubin (P = 0.002), serum creatinine (P = 0.006), and aspartate aminotransferase (P = 0.005) and a significant increase in PDR(ICG) (P = 0.007) after MARS treatment. Prothrombin time, albumin level, and platelet count remained stable. Sustained improvement of renal and neurological function and of mean arterial pressure were observed. No MARS-related adverse effects occurred. MARS treatment provides a safe approach to the treatment of EAD after OLT. On the basis of this pilot study, a multicenter randomized clinical trial that uses MARS treatment in EAD after OLT has been initiated.

Comparison of invasive and noninvasive measurement of plasma disappearance rate of indocyanine green in patients undergoing liver transplantation: a prospective investigator-blinded study.

Plasma disappearance rate of indocyanine green (PDRICG) has been proposed for assessment of liver function in liver transplants donors and recipients, in patients with chronic liver failure, and as a prognostic factor in critically ill patients. The assessment of PDRICG using a newly developed noninvasive digital pulse densitometry method was simultaneously compared to invasive aortic fiber-optic method in patients undergoing orthotopic liver transplantation (OLT). Fourteen consecutive liver transplant candidates (11 male, 3 female) were prospectively enrolled into the study. A 4F aortic catheter with an integrated fiber-optic device and a thermistor was inserted via a femoral artery sheath for invasive aortic (INV) PDRICG assessment in all patients. The fiber-optic device was connected to a computer system (COLD-Z021, PULSION Medical Systems, Munich, Germany). A finger-piece sensor was used for non-invasive (NINV) pulse-densitometric PDRICG assessment. For the PDRICG assessment.5 mg/kg of ICG in cooled saline (10-15 mL) was injected through a central venous catheter. The assessments of PDRICG were performed after induction of anesthesia, after clamping of the hepatic artery, after clamping of the inferior vena cava, after reperfusion of the graft, and on the first postoperative day. During the PDRICG measurements, the investigators were blinded for the results of the noninvasive monitoring. Seventy-one pairs of measurements were performed successfully. PDRICG ranged from 0%/min to 43.8 %/min (11.6%/min +/- 9.6 %/min, mean +/- SD) for invasive and from 2.6%/min to 36.1 %/min (10%/min +/- 7.6 %/min, mean +/- SD) for noninvasive assessment method. The linear regression analysis yielded the equation: PDRICG(NINV) = 1.493 +/- 0.735 x PDRICG(INV), with a correlation coefficient of r = 0.93 (P <.0001). The analysis according to Bland and Altman showed a good agreement between the PDRICG(NINV) and PDRICG(INV) with a mean bias 1.5 +/- 3.8 for all measurements. In conclusion, according to these results, the noninvasive transcutaneous pulse-densitometric method correlates well with the invasive aortic fiber-optic method and thus can be used in patients undergoing liver transplantation.

Extracorporeal albumin dialysis in patients with Amanita phalloides poisoning.

BACKGROUND: Ingestion of Amanita phalloides is the most common cause of lethal mushroom poisoning. The relative late onset of symptoms is a distinct diagnostic feature of Amanita intoxication and also the main reason of failure for extracorporeal removal of Amanita-specific toxins from the gut and circulation. PATIENTS AND METHODS: Extracorporeal albumin dialysis (ECAD) has been used on six consecutive patients admitted after A. phalloides poisoning with acute liver failure (ALF). RESULTS: Six patients, with mean age of 46 years (range: 9-70 years), underwent one to three ECAD treatments. The mean time from mushroom ingestion until the first ECAD treatment was 76 h. Two patients regenerated spontaneously under ECAD treatment and orthotopic liver transplantation (OLT) could be avoided. Two patients were successfully bridged to OLT and one patient died because of cerebral herniation. One patient was treated with ECAD immediately after OLT because of the graft dysfunction and survived without re-transplantation. CONCLUSION: ECAD appeared to be a successful treatment perspective in supporting liver regeneration or in sufficient bridging to OLT and also in treatment of graft dysfunction after OLT in patients with A. phalloides poisoning.

Assessment of ventricular contractile function during orthotopic liver transplantation.

Hemodynamic alterations are a well-known phenomenon that influence the outcome of orthotopic liver transplantation (OLT). Whether or not myocardial dysfunction, which has various causes, contributes to this instability is still debated. Previous transesophageal echocardiography (TEE) studies have presented controversial data, not leading to final clarification. This is mainly because the impact of other contributing factors (inotropic support, alternating preload conditions and temperature) remained unaccounted for. We therefore measured the left ventricular shortening fraction (LVSF), a parameter reflecting myocardial contractility, in 10 consecutive patients undergoing OLT without veno-venous bypass. We measured during preparation (PP), during the anhepatic (AP) phase and the immediate reperfusion phase (RP). During the AP we observed a significant decrease of LVSF which never fell to subnormal levels in the majority of our patients, whereas during the RP, LVSF returned to PP values. These findings support the assumption that myocardial function is influenced by OLT, but that it plays only a minor role in the occurrence of hemodynamic instability, which could mainly be attributed to volume fluctuations.

Effect of acute renal failure requiring renal replacement therapy on outcome in critically ill patients.

OBJECTIVES: Acute renal failure is a complication in critically ill patients that has been associated with an excess risk of hospital mortality. Whether this reflects the severity of the disease or whether acute renal failure is an independent risk factor is unknown. The aim of this study was to analyze severity of illness and mortality in a group of critically ill patients with acute renal failure requiring renal replacement therapy in a number of Austrian intensive care units. DESIGN: Prospective, multicenter cohort study. PATIENTS AND SETTING: A total of 17,126 patients admitted consecutively to 30 medical, surgical, and mixed intensive care units in Austria over a period of 2 yrs. MEASUREMENTS AND MAIN RESULTS: Analyzed data included admission data, Simplified Acute Physiology Score, Logistic Organ Dysfunction system, Simplified Therapeutic Intervention Scoring System, length of intensive care unit stay, intensive care unit mortality, and hospital mortality. Of the admitted patients, 4.9% (n = 839) underwent renal replacement therapy because of acute renal failure (renal replacement therapy patients). These patients had a significantly higher hospital mortality (62.8% vs. 15.6%, p<.001), which remained significantly higher even when renal replacement therapy patients were matched with control subjects for age, severity of illness, and treatment center. Since univariate analysis demonstrated further intensity of treatment to be an additional predictor for outcome, a multivariate model including therapeutic interventions was developed. Five interventions were associated with nonsurvival (mechanical ventilation, single vasoactive medication, multiple vasoactive medication, cardiopulmonary resuscitation, and treatment of complicated metabolic acidosis/alkalosis). In contrast, the use of enteral nutrition predicted a favorable outcome. CONCLUSIONS: The results of our study suggest that acute renal failure in patients undergoing renal replacement therapy presents an excess risk of in-hospital death. This increased risk cannot be explained solely by a more pronounced severity of illness. Our results provide strong evidence that acute renal failure presents a specific and independent risk factor for poor prognosis.