Modeling in-Hospital Patient Survival During the First 28 Days After Intensive Care Unit Admission: a Prognostic Model for Clinical Trials in General Critically Ill Patients

OBJECTIVE: The objective of the study was to develop a model for estimating patient 28-day in-hospital mortality using 2 different statistical approaches. DESIGN: The study was designed to develop an outcome prediction model for 28-day in-hospital mortality using (a) logistic regression with random effects and (b) a multilevel Cox proportional hazards model. SETTING: The study involved 305 intensive care units (ICUs) from the basic Simplified Acute Physiology Score (SAPS) 3 cohort. PATIENTS AND PARTICIPANTS: Patients (n = 17138) were from the SAPS 3 database with follow-up data pertaining to the first 28 days in hospital after ICU admission. INTERVENTIONS: None. MEASUREMENTS AND RESULTS: The database was divided randomly into 5 roughly equal-sized parts (at the ICU level). It was thus possible to run the model-building procedure 5 times, each time taking four fifths of the sample as a development set and the remaining fifth as the validation set. At 28 days after ICU admission, 19.98% of the patients were still in the hospital. Because of the different sampling space and outcome variables, both models presented a better fit in this sample than did the SAPS 3 admission score calibrated to vital status at hospital discharge, both on the general population and in major subgroups. CONCLUSIONS: Both statistical methods can be used to model the 28-day in-hospital mortality better than the SAPS 3 admission model. However, because the logistic regression approach is specifically designed to forecast 28-day mortality, and given the high uncertainty associated with the assumption of the proportionality of risks in the Cox model, the logistic regression approach proved to be superior.

Mortality after Discharge from Intensive Care: the Impact of Organ System Failure and Nursing Workload Use at Discharge

OBJECTIVES: Mortality after ICU discharge accounts for approx. 20-30% of deaths. We examined whether post-ICU discharge mortality is associated with the presence and severity of organ dysfunction/failure just before ICU discharge. PATIENTS AND METHODS: The study used the database of the EURICUS-II study, with a total of 4,621 patients, including 2,958 discharged alive to the general wards (post-ICU mortality 8.6%). Over a 4-month period we collected clinical and demographic characteristics, including the Simplified Acute Physiology Score (SAPS II), Nine Equivalents of Nursing Manpower Use Score, and Sequential Organ Failure Assessment (SOFA) score. RESULTS: Those who died in the hospital after ICU discharge had a higher SAPS II score, were more frequently nonoperative, admitted from the ward, and had stayed longer in the ICU. Their degree of organ dysfunction/failure was higher (admission, maximum, and delta SOFA scores). They required more nursing workload resources while in the ICU. Both the amount of organ dysfunction/failure (especially cardiovascular, neurological, renal, and respiratory) and the amount of nursing workload that they required on the day before discharge were higher. The presence of residual CNS and renal dysfunction/failure were especially prognostic factors at ICU discharge. Multivariate analysis showed only predischarge organ dysfunction/failure to be important; thus the increased use of nursing workload resources before discharge probably reflects only the underlying organ dysfunction/failure. CONCLUSIONS: It is better to delay the discharge of a patient with organ dysfunction/failure from the ICU, unless adequate monitoring and therapeutic resources are available in the ward.

A Comparative Study of Cardiovascular Tolerability with Slow Extended Dialysis Versus Continuous Haemodiafiltration in the Critical Patient

Background: In the haemodynamically unstable patient the method of treatment of acute renal failure is still largely controversial. The purpose of our study was to compare slow extended dialysis with continuous haemodiafiltration in the critical patient with indication for renal replacement therapy and haemodynamic instability. Patients and Methods: This is a cohort study comparing in 63 ventilated critical patients a 12 month period when only continuous haemodiafiltration was used (n=25) with an equal period of slow extended dialysis (n=38). Our primary objective was to evaluate the impact of the dialytic procedure on cardiovascular stability in those patients. As secondary aims we considered system coagulation/thrombosis and predictors of mortality. In the two groups we analysed the first session performed, the second session performed and the average of all the sessions performed in each patient. Results: In these patients, mortality in the intensive care unit was high (68% in the continuous haemodiafiltration group and 63% in the slow extended dialysis group). We did not find any association between the dialytic technique used and death; only the APACHE score was a predictor of death. Slow extended dialysis was a predictor of haemodynamic stability, a negative predictor of sessions that had to be interrupted for haemodynamic instability, and a predictor of achieving the volume removal initially sought. Slow extended dialysis was also associated with less coagulation of the system. Conclusions: Our data suggested that slow extended dialysis use was not inferior to continuous haemodiafiltration use in terms of cardiovascular tolerability.