Performance of APACHE III models in an Australian ICU

STUDY OBJECTIVE: Evaluation of the performance of the APACHE (acute physiology and chronic health evaluation) III ICU and hospital mortality models at an Australian tertiary adult ICU. DESIGN: Noninterventional, observational study. SETTING: Metropolitan, Australian, tertiary referral medical/surgical ICU. PATIENTS: A total of 3,398 consecutive eligible admissions from January 1, 1995, to December 31, 1997. MEASUREMENTS: Prospective collection of demographic, diagnostic, physiologic, laboratory, admission, and discharge data. RESULTS: The patient sample was younger and more commonly male, with more comorbidities and a different operative and referral source mix, compared to the APACHE III development sample. Receiver operating characteristic curve areas for ICU (0.92) and hospital mortality (0.90) demonstrated excellent discrimination. Observed ICU mortality (9.9%) did not significantly differ from the prediction of the APACHE III model (8.9%) or the APACHE III model adjusted for hospital characteristics (10.5%). The hospital mortality (16.0%) was underestimated by the APACHE III model [13.6%; chi(2)(1) = 7.4; p = 0.01]. With proprietary adjustments for hospital characteristics (14.9%) or referenced to the US database (15.6%), agreement was closer. Good calibration was found with all models except the unadjusted hospital mortality model. CONCLUSION: In contrast to other non-American studies, this Australian study demonstrates that the APACHE III can perform well on independent assessment. As perfect discrimination and calibration cannot coexist in a probabilistic model with dichotomous outcomes, performance of APACHE III models with proprietary adjustment for hospital characteristic provide a good compromise for use in quality surveillance.     

Results of report cards for patients with congestive heart failure depend on the method used to adjust for severity

BACKGROUND: The validity of outcome report cards may depend on the ways in which they are adjusted for risk. OBJECTIVES: To compare the predictive ability of generic and disease-specific survival prediction models appropriate for use in patients with heart failure, to simulate outcome report cards by comparing survival across hospitals and adjusting for severity of illness using these models, and to assess the ways in which the results of these comparisons depend on the adjustment method. DESIGN: Analysis of data from a prospective cohort study. SETTING: A university hospital, a Veterans Affairs (VA) medical center, and a community hospital. PATIENTS: Sequential patients presenting in the emergency department with acute congestive heart failure. MEASUREMENTS: Unadjusted 30-day and 1-year mortality across hospitals and 30-day and 1-year mortality adjusted by using disease-specific survival prediction models (two sickness-at-admission models, the Cleveland Health Quality Choice model, the Congestive Heart Failure Mortality Time-Independent Predictive Instrument) and generic models (Acute Physiology and Chronic Health Evaluation [APACHE] II, APACHE III, the mortality prediction model, and the Chadson comorbidity index). RESULTS: The community hospital's unadjusted 30-day survival rate (85.0%) and the VA medical center's unadjusted 1-year survival rate (60.9%) were significantly lower than corresponding rates at the university hospital (92.7% and 67.5%, respectively). No severity model had excellent ability to discriminate patients by survival rates (all areas under the receiver-operating characteristic curve < 0.73). Whether the VA medical center, the community hospital, both, or neither had worse survival rates on simulated report cards than the university hospital depended on the prediction model used for adjustment. CONCLUSIONS: Results of simulated outcome report cards for survival in patients with congestive heart failure depend on the method used to adjust for severity.     

Variation in ICU risk-adjusted mortality: impact of methods of assessment and potential confounders

BACKGROUND: Federal and state agencies are considering ICU performance assessment and public reporting; however, an accurate method for measuring performance must be selected. In this study, we determine whether a substantial variation in ICU mortality performance still exists in modern ICUs, and compare the predictive accuracy, reliability, and data burden of existing ICU risk-adjustment models. METHODS: A retrospective chart review of 11,300 ICU patients from 35 California hospitals from 2001 to 2004 was performed. We calculated standardized mortality ratios (SMRs) for each hospital using the mortality probability model III (MPM(0) III), the simplified acute physiology score (SAPS) II, and the acute physiology and chronic health evaluation (APACHE) IV risk-adjustment models. We compared discrimination, calibration, data reliability, and abstraction time for the models. RESULTS: Regardless of the model used, there was a large variation in SMRs among the ICUs studied. The discrimination and calibration were adequate for all risk-adjustment models. APACHE IV had the best discrimination (area under the receiver operating characteristic curve [AUC], 0.892) compared to MPM(0) III (AUC, 0.809), and SAPS II (AUC, 0.873; p < 0.001). The models differed substantially in data abstraction times, as follows: MPM(0)III, 11.1 min (95% confidence interval [CI], 8.7 to 13.4); SAPS II, 19.6 min (95% CI, 17.0 to 22.2); and APACHE IV, 37.3 min (95% CI, 28.0 to 46.6). CONCLUSIONS: We found substantial variation in the ICU risk-adjusted mortality rates that persisted regardless of the risk-adjustment model. With unlimited resources, the APACHE IV model offers the best predictive accuracy. If constrained by cost and manual data collection, the MPM(0) III model offers a viable alternative without a substantial loss in accuracy.     

Validation of risk adjustment models for in-hospital percutaneous transluminal coronary angioplasty mortality on an independent data set.

OBJECTIVES: We sought to validate recently proposed risk adjustment models for in-hospital percutaneous transluminal coronary angioplasty (PTCA) mortality on an independent data set of high risk patients undergoing PTCA. BACKGROUND: Risk adjustment models for PTCA mortality have recently been reported, but external validation on independent data sets and on high risk patient groups is lacking. METHODS: Between July 1, 1994 and June 1, 1996, 1,476 consecutive procedures were performed on a high risk patient group characterized by a high incidence of cardiogenic shock (3.3%) and acute myocardial infarction (14.3%). Predictors of in-hospital mortality were identified using multivariate logistic regression analysis. Two external models of in-hospital mortality, one developed by the Northern New England Cardiovascular Disease Study Group (model NNE) and the other by the Cleveland Clinic (model CC), were compared using receiver operating characteristic (ROC) curve analysis. RESULTS: In this patient group, an overall in-hospital mortality rate of 3.4% was observed. Multivariate regression analysis identified risk factors for death in the hospital that were similar to the risk factors identified by the two external models. When fitted to the data set, both external models had an area under the ROC curve >0.85, indicating overall excellent model discrimination, and both models were accurate in predicting mortality in different patient subgroups. There was a trend toward a greater ability to predict mortality for model NNE as compared with model CC, but the difference was not significant. CONCLUSIONS: Predictive models for PTCA mortality yield comparable results when applied to patient groups other than the one on which the original model was developed. The accuracy of the two models tested in adjusting for the relatively high mortality rate observed in this patient group supports their application in quality assessment or quality improvement efforts.     

Minimal early mortality in CABG--simply a question of surgical quality?

BACKGROUND: The increasing number of risk scores and models for the evaluation of the early risk after cardiac surgery reflects the interest in 'calculating' the risk of adverse events. Different time intervals, but also different 'types' of death are generally accepted in the evaluation of early mortality. The aim of this study was to focus on the differences in the calculation of early mortality and to focus on their potentially misleading impact on risk stratification. METHODS: We investigated 7,436 patients who underwent coronary artery bypass grafting from June 30, 1988 through June 30, 2001. A follow-up was performed 180 days after operation (98.7 % complete). RESULTS: According to the definition of 30-day mortality to represent the total time interval between an intervention and the 30th postoperative day, the 30-day mortality was 5.92 % (n = 440 patients). Hospital mortality reflects the number of deaths from the day of intervention through the patient's individual discharge, independent of any fixed time interval. Hospital mortality was 5.86 % (n = 436 patients) in our patient group. 30-day hospital mortality requires the investigation of hospital mortality until the 30th postoperative day; in-hospital and general mortality after the 30th postoperative day remained excluded from the analysis; 30-day hospital mortality was 5.19 % (n = 386 patients). Assuming a maximum hospital stay of 5 days, hospital mortality would decrease to 2.64 % (n = 196 patients). CONCLUSIONS: 30-day mortality, hospital mortality and 30-day hospital mortality are used to determine early outcome. The present data indicate the vulnerability of non-standardized time intervals to discharge policy. However, both hospital mortality and 30-day hospital mortality are predominantly used in current risk scores and models. In view of the comparability and meaning of data, the methodology for the evaluation of early risk should be reconsidered.     

Development and validation of the Ontario acute myocardial infarction mortality prediction rules

OBJECTIVES: To develop and validate simple statistical models that can be used with hospital discharge administrative databases to predict 30-day and one-year mortality after an acute myocardial infarction (AMI). BACKGROUND: There is increasing interest in developing AMI "report cards" using population-based hospital discharge databases. However, there is a lack of simple statistical models that can be used to adjust for regional and interinstitutional differences in patient case-mix. METHODS: We used linked administrative databases on 52,616 patients having an AMI in Ontario, Canada, between 1994 and 1997 to develop logistic regression statistical models to predict 30-day and one-year mortality after an AMI. These models were subsequently validated in two external cohorts of AMI patients derived from administrative datasets from Manitoba, Canada, and California, U.S. RESULTS: The 11-variable Ontario AMI mortality prediction rules accurately predicted mortality with an area under the receiver operating characteristic (ROC) curve of 0.78 for 30-day mortality and 0.79 for one-year mortality in the Ontario dataset from which they were derived. In an independent validation dataset of 4,836 AMI patients from Manitoba, the ROC areas were 0.77 and 0.78, respectively. In a second validation dataset of 112,234 AMI patients from California, the ROC areas were 0.77 and 0.78 respectively. CONCLUSIONS: The Ontario AMI mortality prediction rules predict quite accurately 30-day and one-year mortality after an AMI in linked hospital discharge databases of AMI patients from Ontario, Manitoba and California. These models may also be useful to outcomes and quality measurement researchers in other jurisdictions.     

Validation of the hospital outcome prediction equation (HOPE) model for monitoring clinical performance

Background: The aim of this study was to validate a risk-adjusted hospital outcome prediction equation (HOPE) using a statewide administrative dataset.
Methods: Retrospective observational study using multivariate logistic regression modelling. Calibration and discrimination were assessed by standardized mortality ratio (SMR), area under the receiver operating characteristic plot (ROC AUC), Hosmer–Lemeshow contingency tables and goodness-of-fit statistic in an independent dataset, and in all 23 important tertiary, metropolitan and regional hospitals. The dependent variable was in-hospital death. All consecutive adult hospital separations between 1 July 2004 and 30 June 2006, excluding obstetric and day-case only admissions, from all acute health services within the State of Victoria, Australia were included.
Results: A total of 379 676 consecutive records (1 July 2004 to 30 June 2005) was used to derive the HOPE model. Six variables (age, male sex, admission diagnosis, emergency admission, aged-care resident and inter-hospital transfer) were selected for inclusion in the final model. It was validated in the 384 489 consecutive records from the following year (1 July 2005 to 30 June 2006). The 95% confidence interval for the SMR was 0.98–1.02, and for the ROC AUC, 0.87–0.88. Discrimination and (one or more) calibration criteria were achieved in 22 (96%) of the 23 hospitals.
Conclusion: The HOPE model is a simple risk-adjusted outcome prediction tool, based on six variables from data that are routinely collected for administrative purposes and appears to be a reliable predictor of hospital outcome.     

Intensivmedizin bei älteren Patienten: Wie nützlich sind die Scoresysteme APACHE II und III?

OBJECTIVE: Scores like APACHE (Acute Physiology And Chronic Health Evaluation) were evaluated for unselected intensive care unit (ICU) admissions. Can they also be used for risk stratification and quality assurance in selected subgroups like elderly patients? METHODS: Over a 3-year period data of all admissions of a 12 bed interdisciplinary ICU were collected. APACHE II and III scores and probabilities of hospital deaths were compared with observed outcomes. The discriminatory power was evaluated by calculating the areas under the receiver operating characteristic (ROC) curves. Calibration was analyzed with standardized mortality ratios (SMR) and the Hosmer-Lemeshow goodness-of-fit statistic. RESULTS: Of 3382 admissions due to exclusion criteria, 2795 patients were analyzed, 1396 (49.9%) of these were > or = 65 years, mean age 75 (65-99) years. 62.5% were non-operative, 37.5% postoperative admissions, 35% after emergency operations. ICU mortality was 11.7%, hospital mortality 25.1%. The areas under the ROC curves were 0.77 for APACHE II and 0.79 for APACHE III (whole collective 0.83 and 0.85, respectively). The SMR was 1.17 for APACHE II and 1.23 for APACHE III compared with 1.06 and 1.22 for all patients, respectively. Calibration for elderly patients was insufficient for APACHE II (Hosmer-Lemeshow chi-square = 19, p < 0.025) as well as for APACHE III (chi-square = 41, p < 0.001), while it was good for all patients for APACHE II (chi-square = 12, p > 0.1) but not so for APACHE III (chi-square = 48, p < 0.001). CONCLUSIONS: APACHE II and III both show good discrimination for elderly patients although a little inferior than for all patients. Both scores can be used for risk stratification of elderly ICU patients. Mortality prognosis is not sufficient for geriatric patients although APACHE II calibrates well for all. Application of these scores for quality assurance in selected subgroups like elderly patients cannot be recommended based on these data.     

Development and Validation of an Acute Heart Failure-Specific Mortality Predictive Model Based on Administrative Data

Background

Acute heart failure (AHF) with its high in-hospital mortality is an increasing burden on healthcare systems worldwide, and comparing hospital performance is required for improving hospital management efficiency. However, it is difficult to distinguish patient severity from individual hospital care effects. The aim of this study was to develop a risk adjustment model to predict in-hospital mortality for AHF using routinely available administrative data.

Methods

Administrative data were extracted from 86 acute care hospitals in Japan. We identified 8620 hospitalized patients with AHF from April 2010 to March 2011. Multivariable logistic regression analyses were conducted to analyze various patient factors that might affect mortality. Two predictive models (models 1 and 2; without and with New York Heart Association functional class, respectively) were developed and bootstrapping was used for internal validation. Expected mortality rates were then calculated for each hospital by applying model 2.

Results

The overall in-hospital mortality rate was 7.1%. Factors independently associated with higher in-hospital mortality included advanced age, New York Heart Association class, and severe respiratory failure. In contrast, comorbid hypertension, ischemic heart disease, and atrial fibrillation/flutter were found to be associated with lower in-hospital mortality. Both model 1 and model 2 demonstrated good discrimination with c-statistics of 0.76 (95% confidence interval, 0.74-0.78) and 0.80 (95% confidence interval, 0.78-0.82), respectively, and good calibration after bootstrap correction, with better results in model 2.

Conclusions

Factors identifiable from administrative data were able to accurately predict in-hospital mortality. Application of our model might facilitate risk adjustment for AHF and can contribute to hospital evaluations.     

Community-acquired bacteremia at a teaching versus a nonteaching hospital: impact of acute severity of illness on 30-day mortality

BACKGROUND: Few studies have focused recently on the epidemiology of community-acquired bacteremia (CAB) and there have been few comparisons of CAB in teaching versus nonteaching hospitals. OBJECTIVES: To compare the clinical characteristics, acute severity of illness, and 30-day mortality of patients with CAB admitted to a teaching and a nonteaching hospital and to define predictors of 30-day mortality among patients with CAB that would be identifiable at the time of admission to the hospital. METHODS: This was a retrospective study of CAB at a teaching hospital (n = 174 episodes) compared to a community nonteaching hospital (n = 74 episodes) during 1995. Data collected included demographic characteristics, underlying diseases, sources of CAB, and antimicrobial therapy. Acute severity of illness on admission was measured by using the acute physiology score component of the Acute Physiology and Chronic Health Evaluation III system (APS APACHE III).Main Outcome Measure: Status, dead or alive, 30 days after admission for CAB. RESULTS: At the nonteaching hospital, patients were older but, on average, significantly less acutely ill (as measured by the admission APS APACHE III score) than were those at the teaching hospital. In contrast, patients with HIV infection, posttransplantation, or on hemodialysis were identified only at the teaching hospital. Overall, organisms causing CAB at both hospitals were similar except that Staphylococcus aureus CAB occurred significantly more often at the teaching hospital and Escherichia coli CAB occurred more often at the nonteaching hospital. There was no significant difference in 30-day mortality in patients with CAB between the teaching hospital (19.3%) and the nonteaching hospital (16.7%; P =.63). APS APACHE III score on admission identified episodes of CAB with a low- and a high-risk for 30-day mortality at both hospitals. Independent predictors of 30-day mortality were APACHE III score on admission (P <.001) and pneumonia as a source of CAB (P =.012). CONCLUSIONS: Among patients with CAB, acute severity of illness on admission was the most important predictor of 30-day mortality at both hospitals. Even though patients with CAB were, on average, more severely ill at the time of admission to the teaching hospital, 30-day mortality rates were not significantly different between the two hospitals because deaths correlated with high APS APACHE III scores at both facilities. The APS APACHE III score on admission provides important prognostic information among patients with CAB.     

Comparison of outcome from intensive care admission after adjustment for case mix by the APACHE III prognostic system

STUDY OBJECTIVES: To evaluate the acute physiology, age, chronic health evaluation III (APACHE III) scoring system in the context of general adult ICUs in the United Kingdom. DESIGN: Prospective, noninterventional, cohort study. SETTING: Seventeen general adult ICUs in a discrete area of southwest England. PATIENTS: 12,793 patients admitted between April 1, 1993 and December 31, 1995. MEASUREMENTS: Sociodemographic and severity-of-illness data were collected for all patients admitted to the study units. Formal goodness-of-fit tests were applied and observed mortality was compared with that predicted by using the APACHE III system. RESULTS: For the group of ICUs as a whole, the risk-adjusted standardized mortality ratio (SMR) was 1.23 (95% confidence intervals, 1.12-1.25). For 11 out of 17 ICUs, the SMR was significantly greater than unity (p < 0.05). Calibration, as tested by Hosmer-Lemeshow statistics, was poor (H2 = 312.54; C2 = 332.85; df = 8; p < 0.01); however, model discrimination was good with a total correct classification rate of 82.9% and an area under the receiver operating characteristic curve of 0.89. CONCLUSIONS: The excess mortality observed after case-mix adjustment using the APACHE III system in this study may be the result of either poor intensive care performance as compared with the United States or a failure of the APACHE III equation to fit the UK data.     

两种评价急性肾衰竭患者预后及肾脏转归积分模型的比较

目的　比较急性生理和平素健康评估Ⅱ (APACHEⅡ )与急性肾小管坏死 个体严重程度指数 (ATN ISI)两种积分模型对急性肾衰竭 (ARF)患者的预后和肾脏转归的预示效果。方法　回顾性分析了近 1 0年的 42 2例ARF患者资料 ,比较两种积分模型对患者病死率及肾脏转归的预测效果 ,并采用两种积分评定方式对ARF发生 30、45、60d后的肾脏转归进行了判别分析。结果　随着两种模型积分值的增加 ,患者的病死率升高 ,当ATN ISI积分≥ 0 85、APACHEⅡ积分≥ 35时病死率为 1 0 0 % ;APACHEⅡ和ATN ISI模型的ROC曲线下的面积分别为 0 81 7± 0 0 2 1和 0 880± 0 0 1 8,表明两种模型对ARF患者病死率的判别均有意义。对肾脏转归的判别 ,ATN ISI在各评定时间的判别符合率均高于APACHEⅡ ;ATN ISI积分≥ 0 75时 ,均需依赖透析治疗 ;<0 75但≥ 0 58时 ,肾功能未恢复正常 ;肾功能完全恢复者积分值均在 0 58以内。APACHEⅡ积分≥ 2 6时 ,均需依赖透析治疗 ;<2 6时 ,肾功能完全恢复和肾功能不全病人之间无明显积分界限 ;但≤ 2 2时 ,上述二者所占比例分别为 80 4%和1 9 6 %。结论　两种积分模型对ARF患者的病死率及肾脏转归均有较好的预示效果 ,但ATN ISI积分模型对肾脏转归的预示价值更优于APACHEⅡ。     

Performance of standard severity scoring systems for outcome prediction in patients admitted to a respiratory intensive care unit in North India

OBJECTIVE: There are little data on the value of using severity scoring systems developed in western countries to assess critically ill patients in India. The authors evaluated the performance of Acute Physiology and Chronic Health Evaluation version II (APACHE II), Simplified Acute Physiology Score version II (SAPS II) and Mortality Probability Models version II at admission and at 24 h (MPM(0) and MPM(24), respectively) in predicting patient outcomes in their Respiratory Intensive Care Unit. METHODS: Data from 459 consecutive adult admissions were collected prospectively. Standardized mortality ratios were computed as an index of the overall model performance. Model calibration was assessed using Lemeshow-Hosmer goodness-of-fit tests and through calibration curves. Model discrimination was assessed through receiver operating curve analysis and by drawing 2 x 2 classification matrices. RESULTS: Overall standardized mortality ratio exceeded 1.5 for all models. All models had modest discrimination (area under receiver-operating-characteristic curves 0.66-0.78) and poor calibration (high Lemeshow-Hosmer C and H statistic values). All models had a tendency to underpredict hospital death in patients with lower mortality probability estimates. There were no major differences between the models with regard to either discrimination or calibration performance. CONCLUSIONS: Standard severity scoring systems developed in western countries are poor at predicting patient outcome in critically ill patients admitted to a respiratory intensive care unit in Northern India. Caution must be exercised in using such models in their present form on Indian patients until either they are customized for local use or fresh models are developed from Indian cohorts.     

Validation of the LOD score compared with APACHE II score in prediction of the hospital outcome in critically ill patients

The Logistic Organ Dysfunction score (LOD) is an organ dysfunction score that can predict hospital mortality. The aim of this study was to validate the performance of the LOD score compared with the Acute Physiology and Chronic Health Evaluation II (APACHE II) score in a mixed intensive care unit (ICU) at a tertiary referral university hospital in Thailand. The data were collected prospectively on consecutive ICU admissions over a 24 month period from July1, 2004 until June 30, 2006. Discrimination was evaluated by the area under the receiver operating characteristic curve (AUROC). The calibration was assessed by the Hosmer-Lemeshow goodness-of-fit H statistic. The overall fit of the model was evaluated by the Brier's score. Overall, 1,429 patients were enrolled during the study period. The mortality in the ICU was 20.9% and in the hospital was 27.9%. The median ICU and hospital lengths of stay were 3 and 18 days, respectively, for all patients. Both models showed excellent discrimination. The AUROC for the LOD and APACHE II were 0.860 [95% confidence interval (CI) = 0.838-0.882] and 0.898 (95% Cl = 0.879-0.917), respectively. The LOD score had perfect calibration with the Hosmer-Lemeshow goodness-of-fit H chi-2 = 10 (p = 0.44). However, the APACHE II had poor calibration with the Hosmer-Lemeshow goodness-of-fit H chi-2 = 75.69 (p < 0.001). Brier's score showed the overall fit for both models were 0.123 (95%Cl = 0.107-0.141) and 0.114 (0.098-0.132) for the LOD and APACHE II, respectively. Thus, the LOD score was found to be accurate for predicting hospital mortality for general critically ill patients in Thailand.     

Machine Learning Prediction Models for In-Hospital Mortality After Transcatheter Aortic Valve Replacement

ObjectivesThis study sought to develop and compare an array of machine learning methods to predict in-hospital mortality after transcatheter aortic valve replacement (TAVR) in the United States.BackgroundExisting risk prediction tools for in-hospital complications in patients undergoing TAVR have been designed using statistical modeling approaches and have certain limitations.MethodsPatient data were obtained from the National Inpatient Sample database from 2012 to 2015. The data were randomly divided into a development cohort (n = 7,615) and a validation cohort (n = 3,268). Logistic regression, artificial neural network, naive Bayes, and random forest machine learning algorithms were applied to obtain in-hospital mortality prediction models.ResultsA total of 10,883 TAVRs were analyzed in our study. The overall in-hospital mortality was 3.6%. Overall, prediction models’ performance measured by area under the curve were good (>0.80). The best model was obtained by logistic regression (area under the curve: 0.92; 95% confidence interval: 0.89 to 0.95). Most obtained models plateaued after introducing 10 variables. Acute kidney injury was the main predictor of in-hospital mortality ranked with the highest mean importance in all the models. The National Inpatient Sample TAVR score showed the best discrimination among available TAVR prediction scores.ConclusionsMachine learning methods can generate robust models to predict in-hospital mortality for TAVR. The National Inpatient Sample TAVR score should be considered for prognosis and shared decision making in TAVR patients.     

Prevalence, predictors, and outcomes of conservative medical management in non-ST-segment elevation acute coronary syndromes in Gulf RACE-2

We assessed the prevalence, predictors, and in-hospital and long-term outcomes of conservative medical management for patients with non-ST-segment elevation acute coronary syndrome (NSTEACS) compared with percutaneous coronary intervention (PCI) and coronary artery bypass graft surgery (CABG). This prospective study conducted from October 2008 to June 2009 in 65 hospitals from 6 Arabian Gulf countries included 30-day and 1-year mortality follow-up for 3661 patients. Compared with conservative management group (2859 patients; 78.1%), the PCI group (638; 17.4%) had significantly better unadjusted and adjusted in-hospital (odds ratio [OR]: 0.40, 95% confidence interval [CI]: 0.17-0.97), 30-day (OR: 0.44, 95% CI: 0.24-0.76) and 1-year (OR: 0.58, 95% CI: 0.40-0.87) mortality rates. Comparison with the CABG group (164; 4.5%) yielded similar results with inclusion of patients scheduled for CABG after hospital discharge. Independent predictors of conservative medical management were mainly country of residence and history of prior CABG.     

Prospective evaluation of 3 risk stratification scores in cardiac surgery

BACKGROUND: The aim of the study was to evaluate 3 different risk stratification scores in cardiac surgery, based on the hospital results of 1,299 patients. METHODS: From June 1995 to December 1997, all patients (n = 1,299) undergoing coronary artery bypass grafting (CABG) and/or heart valve surgery were prospectively enrolled. The postoperative in-hospital outcome (mortality, morbidity and length of hospital stay) was analysed in relation to three different risk stratification scores (Parsonnet, Higgins and French score). RESULTS: The results of 1,299 patients (mean age 62.8 +/- 10.2 years) were analysed. 10 patients died, accounting for a total mortality of 0.8%. 13 patients (1%) underwent cardiopulmonary resuscitation. In 25 patients (1.9%), perioperative myocardial infarction occurred. Performance of the 3 systems was assessed by evaluating discrimination with receiver operating characteristic (ROC) curves. The area under the ROC curve was 0.761 for Parsonnet, 0.786 for Higgins and 0.798 for French score. The French and the Higgins score showed an increase of in-hospital mortality, morbidity and length of stay in relation to increasing risk classes. CONCLUSION: For objective evaluation of the outcome in cardiac surgery, case-mix severity needs to be considered, which is reflected by preoperative risk stratification scores. In our study, all the 3 scores showed a high discrimination and are appropriate tools to assess mortality in cardiac surgery. Especially the French and the Higgins score (restricted to 5 groups), due to their simplicity, were useful to predict postoperative outcome in clinical routine.     

Identifying potentially ineffective care in the sickest critically ill patients on the third ICU day

OBJECTIVE: To determine if an increase in the third-ICU-day acute physiology score (APS) of the APACHE (acute physiology and chronic health evaluation) III prognostic system can identify potentially ineffective care. DESIGN: Retrospective cohort study. SETTING: Academic medical center. PATIENTS: Adult patients with first-ICU-day predicted mortality rate > or = 80%. MEASUREMENTS: Demographics, ICU admission source, admission type, ICU admission diagnosis, first- and third-ICU-day APSs, APACHE III score, APACHE III-predicted hospital mortality, hospital discharge status, 100-day survival, and ICU and hospital length of stay. RESULTS: A total of 302 patients (age [mean +/- SD], 64.7 +/- 15.8 years; 54.3% male gender) were included in the study. Respiratory failure was the most common reason for ICU admission. Nonoperative admissions accounted for 94.7%. The first- and third-ICU-day APSs were 106.8 +/- 19.8 and 70.5 +/- 29.9, respectively. The first- and third-ICU-day predicted hospital mortality rates were 87.8 +/- 5.3% and 86.5 +/- 14.8%, respectively. The hospital mortality rate was 61.3%, and the 100-day survival rate 28.5%. The third-ICU-day APS was higher than the first-ICU-day APS in 34 patients (11.3%). Only 2 of these 34 patients (6%) survived to hospital discharge, compared to 115 of 268 patients (43%) without an increase in APS (p < 0.0001). Of the two hospital survivors with increased APS, only one patient survived 100 days after hospital discharge. In predicting 100-day mortality, the sensitivity of an increase in the third-ICU-day APS was 15.3% (95% confidence interval, 11.1 to 20.7%), specificity was 98.8% (95% confidence interval, 93.7 to 99.8%), positive predictive value was 97.1% (95% confidence interval, 85.1 to 99.5%), and negative predictive value was 31.7% (95% confidence interval, 26.4 to 37.5%). CONCLUSIONS: A higher APS on the third ICU day, compared to the first ICU day, identifies potentially ineffective care in patients with the first-day predicted hospital mortality rate > or = 80%.     

Model to predict mortality in critically ill adults with acute kidney injury

Summary

Background and objectives

Acute kidney injury (AKI) requiring dialysis is associated with high mortality. Most prognostic tools used to describe case complexity and to project patient outcome lack predictive accuracy when applied in patients with AKI. In this study, we developed an AKI-specific predictive model for 60-day mortality and compared the model to the performance of two generic (Sequential Organ Failure Assessment [SOFA] and Acute Physiology and Chronic Health Evaluation II [APACHE II]) scores, and a disease specific (Cleveland Clinic [CCF]) score.

Design, setting, participants, & measurements

Data from 1122 subjects enrolled in the Veterans Affairs/National Institutes of Health Acute Renal Failure Trial Network study; a multicenter randomized trial of intensive versus less intensive renal support in critically ill patients with AKI conducted between November 2003 and July 2007 at 27 VA- and university-affiliated centers.

Results

The 60-day mortality was 53%. Twenty-one independent predictors of 60-day mortality were identified. The logistic regression model exhibited good discrimination, with an area under the receiver operating characteristic (ROC) curve of 0.85 (0.83 to 0.88), and a derived integer risk score yielded a value of 0.80 (0.77 to 0.83). Existing scoring systems, including APACHE II, SOFA, and CCF, when applied to our cohort, showed relatively poor discrimination, reflected by areas under the ROC curve of 0.68 (0.64 to 0.71), 0.69 (0.66 to 0.73), and 0.65 (0.62 to 0.69), respectively.

Conclusions

Our new risk model outperformed existing generic and disease-specific scoring systems in predicting 60-day mortality in critically ill patients with AKI. The current model requires external validation before it can be applied to other patient populations.