Severity Of Illness Scoring Systems Research Articles

Comparing ensemble learning algorithms and severity of illness scoring systems in cardiac intensive care units: a retrospective study.

Beatriz Nistal-Nuño designed a machine learning system type of ensemble learning for patients undergoing cardiac surgery and intensive care unit cardiology patients, based on sequences of cardiovascular physiological measurements and other intensive care unit physiological measurements in addition to static features, which generates a score for prediction of mortality of cardiac intensive care unit patients. ■ Gradient Boosting Machine and Random Forest models were built for prediction of mortality at cardiac intensive care units. ■ A total of 9,761 intensive care unit stays of patients admitted under a Cardiac Surgery and Cardiac Medical services were studied. ■ The AUROC and AUPRC values were significantly superior to seven conventional systems compared. ■ The machine learning models' calibration curves were substantially closer to the ideal line. Logistic Regression has been used traditionally for the development of most predictor tools of intensive care unit mortality. The purpose of this study is to combine shared risk factors between patients undergoing cardiac surgery and intensive care unit cardiology patients to develop a risk score for prediction of mortality in cardiac intensive care unit patients, using machine learning. Gradient Boosting Machine and Distributed Random Forest models were developed based on 9,761 intensive care unit-stays from the MIMIC-III database. Sequential and static features were collected. The primary endpoint was intensive care unit mortality prediction. Discrimination, calibration, and accuracy statistics were evaluated. The predictive performance of traditional scoring systems was compared. Machine learning models' AUROC and AUPRC were significantly superior to all conventional systems for the primary endpoint (p<0.05), with AUROC of 0.9413 for Gradient Boosting Machine and 0.9311 for Distributed Random Forest. Sensitivity was 0.6421 for Gradient Boosting Machine, 0.6 for Distributed Random Forest, and <0.3 for all conventional systems except for serial SOFA (0.6316). Precision was 0.574 for Gradient Boosting Machine, 0.566 for Distributed Random Forest, and <0.5 for all conventional systems. Diagnostic odds ratio was 58.8144 for Gradient Boosting Machine, 51.2926 for Distributed Random Forest and <34 for all conventional systems. Brier score was 0.025 for Gradient Boosting Machine and 0.028 for Distributed Random Forest, being worse for the traditional systems. Calibration curves of Gradient Boosting Machine and Distributed Random Forest were substantially closer to the ideal line. The machine learning models showed superiority over the traditional scoring systems compared, with Gradient Boosting Machine having the best performance. Discrimination and calibration were excellent for Gradient Boosting Machine, followed by Distributed Random Forest. The machine learning methods exhibited better capacity for most accuracy statistics.

Importance of clinical parameters for cultivation of critical care thinking by online teaching of critical care medicine

BackgroundThe teaching of critical care medicine is a very important task, especially during the COVID-19 pandemic. The understanding of critical care parameters is the foundation and core, which is conducive to the formation of clinical thinking. This study is to evaluate the training effect of teaching of critical care parameters based on an online platform, and explore the teaching methods of critical care medicine that can help to cultivate trainees’ clinical thinking and practical ability.MethodsQuestionnaires were released before and after the training through the official new media platform “Yisheng” application (APP) of China Medical Tribune involving 1109 participants. The trainees who filled in the questionnaire in APP and received training were randomly selected as the investigated population. Statistical description and analysis were carried out using SPSS 20.0 and Excel 2020.ResultsThe trainees were mainly attending physicians in tertiary hospitals and above. Among all critical care parameters, trainees paid more attention to critical hemodynamics, respiratory mechanics, severity of illness scoring systems, critical ultrasound, and critical hemofiltration. The degree of satisfaction with the courses was high, especially the course of critical hemodynamics was scored the highest. The trainees believed that the course contents were of great help to clinical work. However, no significant difference was found in the trainees’ understanding or cognition of the connotation of the parameters before and after the training.ConclusionTeaching of critical care parameters based on an online platform is conducive to improving and consolidating the clinical care ability of trainees. However, it is still necessary to strengthen the cultivation of clinical thinking in critical care. In the future, the integration of theory with practice must be strengthened in clinical practice, ultimately achieving the homogeneous diagnosis and treatment of patients with critical illness.

Forecasting ICU Census by Combining Time Series and Survival Models.

Retrospective observational study. We developed the algorithm using data from the Medical-Surgical ICU (MSICU) at University Hospital, London, Canada and validated using data from the Critical Care Trauma Centre (CCTC) at Victoria Hospital, London, Canada. Adult patient admissions (7,434) to the MSICU and (9,075) to the CCTC from 2015 to 2021. None. We developed an Autoregressive integrated moving average time series model that forecasts patients arriving in the ICU and a survival model using MODS, NEMS, and other factors to estimate patient LOS. The models were combined to create an algorithm that forecasts ICU census for planning horizons ranging from 1 to 7 days. We evaluated the algorithm quality using several fit metrics. The root mean squared error ranged from 2.055 to 2.890 beds/d and the mean absolute percentage error from 9.4% to 13.2%. We show that this forecasting algorithm provides a better fit when compared with a moving average or a time series model that directly forecasts ICU census. Additionally, we evaluated the performance of the algorithm using data during the global COVID-19 pandemic and found that the error of the forecasts increased proportionally with the number of COVID-19 patients in the ICU. It is possible to develop accurate tools to forecast ICU census. This type of algorithm may be important to clinicians and managers when planning ICU capacity as well as staffing and surgical demand planning over a short time horizon.

Cytokine Levels and Severity of Illness Scoring Systems to Predict Mortality in COVID-19 Infection.

Various scoring systems and cytokines have been cited as predicting disease severity in COVID-19 infection. This study analyzed the link between mortality rate, levels of cytokines, and scoring systems such as the Glasgow Coma Scale (GCS), Acute Physiologic Assessment and Chronic Health Evaluation II (APACHE II), Sequential Organ Failure Assessment (SOFA), and Charlson Comorbidity Index in patients infected with COVID-19. Adult patients infected with COVID-19 were followed up in the intensive care unit (ICU) and analyzed prospectively. We measured serum cytokine levels (Interleukin-10 (IL-10), Interleukin-8 (IL-8), Interleukin-6 (IL-6), Interleukin-1β (IL-1β), tumor necrosis factor-alpha (TNF-α) and High mobility group box 1 (HMGB-1)) and recorded GCS, APACHE II, SOFA, and Charlson comorbidity index scores on admission to the ICU. Receiver operating curve (ROC) analysis was performed to predict mortality from IL-1β, IL-6 IL-10, IL-8, TNF-α, and HMGB-1 values. Study participants were grouped as follows: Group A, survivors, and Group B, deceased, during the 28-day follow-up. The mean age was 65.69 (±13.56) in Group A (n = 36) and 70.85 (±10.06) in Group B (n = 27). The female/male ratio was 23/40. Age, sex, body mass index (BMI), comorbid illnesses, GCS, APACHE II, SOFA, and Charlson scores, duration of hospitalization or ICU admission, therapeutic choices, and lymphocyte, PMNL, NLR, platelet, D-dimer, fibrinogen, GGT, CRP, procalcitonin, and lactate levels were similar between the groups. The frequency of acute kidney injury (AKI) was higher in Group B (p = 0.005). Serum IL-10, IL-8, IL-6, IL-1β, TNF-α, HMGB-1, ferritin, and LDH values were higher, and PaO2/FiO2 was lower in Group B than in Group A. ROC analysis showed that there was an association between serum IL-1β (>1015.7), serum IL-6 (>116.7), serum IL-8 (>258.4), serum IL-10 (>247.5), serum TNF-α (>280.7), and serum HMGB-1 (>23.5) and mortality. AKI gave rise to a greater risk of mortality (odds ratio: 7.081, p = 0.014). Mortality was associated with serum IL-10, IL-8, IL-6, IL-1β, TNF-α, and HMGB-1 but not with GCS, APACHE II, SOFA, or Charlson comorbidity index scores. AKI increased the risk of mortality by seven times. Our findings suggest that cytokine levels (serum IL-10, IL-8, IL-6, IL-1β, TNF-α, and HMGB-1) were predictors of mortality in COVID-19 infection. In addition, our results might give an opinion about the course of COVID-19 infection.

Data, Anomalies; a Call to Cease and Desist.

Data, Anomalies; a Call to Cease and Desist.

The predictive value of phase angle on long-term outcome after ICU admission

The use of severity of illness scoring systems, including the Acute Physiology and Chronic Health Evaluation (APACHE) III score, has made it possible to compare groups of patients and evaluate treatment strategies. Phase angle, derived from bio-impedance analysis, reflects tissue quality and quantity in which cell mass, membrane integrity and hydration state are represented. We hypothesized that phase angle on ICU admission may serve as a proxy for physical frailty and as such can be used as an additional predictor of long-term mortality after ICU admission. A single-center prospective observational cohort study with consecutive patients, admitted to the ICU between June 2018 and June 2019. Demographic data, APACHE III, comorbidity, and phase angle in the first 6h after ICU admission were collected and the ICU, hospital, and 1-year survival were registered. Of all 1023 patients, 115 (11%) died within a year after ICU admission. Nonsurvivors had higher APACHE III scores than survivors [86 (65-119) vs. 55 (46-67), p<0.001]. Phase angle was significantly higher in survivors than in nonsurvivors [5.4 (4.7-6.4) vs. 4.7 (3.9-6.0), p<0.001]. Univariate analysis showed an association between mortality and admission type, sepsis, presence of malignancy, APACHE III, and PhA. Multivariate logistic regression analysis using these variables confirmed low PhA to be an independent predictor of 1-year mortality (OR: 1.81; CI: 1.09-2.97; p=0.02), in addition to presence of malignancy (OR: 2.30; CI: 1.31-4.02; p=0.004) and APACHE III score (OR: 1.03; CI: 1.02-1.04; p<0.001). In this single center study, low phase angle was independently associated with 1-year all-cause mortality after ICU admission. gov number: NCT0444976.

How Are Emerging Data Translated Into Clinical Practice? A Mixed Methods Investigation of Coronavirus Disease 2019 Institutional Treatment Protocols.

BackgroundEarly in the coronavirus disease 2019 (COVID-19) pandemic, there was minimal data to guide treatment, and we lacked understanding of how clinicians translated this limited evidence base for potential therapeutics to bedside care. Our objective was to systematically determine how emerging data about COVID-19 treatments was implemented by analyzing institutional treatment protocols.MethodsTreatment protocols from North American healthcare facilities and recommendations from guideline-issuing bodies were collected. Qualitative data on treatment regimens and their applications were extracted using an adapted National Institutes of Health/US Food and Drug Administration experimental therapeutics framework. Structured data on risk factor and severity of illness scoring systems were extracted and analyzed using descriptive statistics.ResultsWe extracted data from 105 independent protocols. Guideline-issuing organizations published recommendations after the initial peak of the pandemic in many regions and generally recommended clinical trial referral, with limited additional guidance. Facility-specific protocols favored offering some treatment (96.8%, N = 92 of 95), most commonly, hydroxychloroquine (90.5%), followed by remdesivir and interleukin-6 inhibitors. Recommendation for clinical trial enrollment was limited largely to academic medical centers (19 of 52 vs 9 of 43 community/Veterans Affairs [VA]), which were more likely to have access to research studies. Other themes identified included urgent protocol development, plans for rapid updates, contradictory statements, and entirely missing sections, with section headings but no content other than “in process.”ConclusionsIn the COVID-19 pandemic, emerging information was rapidly implemented by institutions into clinical practice and, unlike recommendations from guideline-issuing bodies, heavily favored administering some form of therapy. Understanding how and why evidence is translated into clinical care is critical to improve processes for other emerging diseases.

Multicenter International Cohort Validation of a Modified Sequential Organ Failure Assessment Score Using the Richmond Agitation-sedation Scale.

In a multicenter, international cohort, we aimed to validate a modified Sequential Organ Failure Assessment (mSOFA) using the Richmond Agitation-Sedation Scale, hypothesized as comparable to the Glasgow Coma Scale (GCS)-based Sequential Organ Failure Assessment (SOFA). The SOFA score, whose neurologic component is based on the GCS, can predict intensive care unit (ICU) mortality. But, GCS is often missing in lieu of other assessments, such as the also reliable and validated Richmond Agitation Sedation Scale (RASS). Single-center data suggested an RASS-based SOFA (mSOFA) predicted ICU mortality. Our nested cohort within the prospective 2016 Fourth International Study of Mechanical Ventilation contains 4120 ventilated patients with daily RASS and GCS assessments (20,023 patient-days, 32 countries). We estimated GCS from RASS via a proportional odds model without adjustment. ICU mortality logistic regression models and c-statistics were constructed using SOFA (measured GCS) and mSOFA (measured RASS-estimated GCS), adjusted for age, sex, body-mass index, region (Europe, USA-Canada, Latin America, Africa, Asia, Australia-New Zealand), and postoperative status (medical/surgical). Cohort-wide, the mean SOFA=9.4+/-2.8 and mean mSOFA = 10.0+/-2.3, with ICU mortality = 31%. Mean SOFA and mSOFA similarly predicted ICU mortality (SOFA: AUC = 0.784, 95% CI = 0.769-0.799; mSOFA: AUC = 0.778, 95% CI = 0.763-0.793, P = 0.139). Across models, other predictors of mortality included higher age, female sex, medical patient, and African region (all P < 0.001). We present the first SOFA modification with RASS in a "real-world" international cohort. Estimating GCS from RASS preserves predictive validity of SOFA to predict ICU mortality. Alternative neurologic measurements like RASS can be viably integrated into severity of illness scoring systems like SOFA.

Utilization of an ICU Severity of Illness Scoring System to Triage Patients With ST-Elevation Myocardial Infarction.

To subjectively identify low-risk ST-elevation myocardial infarction (STEMI) patients and triage this low-risk population to an intermediate level of care. Many patients with STEMI are admitted to the intensive care unit (ICU), however, a large portion do not merit ICU admission. We sought to examine whether, among post-STEMI patients admitted to the ICU, if an easily obtainable subjective scoring system could predict low-risk patients and safely triage them to an intermediate level of care. Retrospective observational study at Christiana Hospital, a 900-bed regional referral center. Data were defined by the ACTION Registry and CathPCI Registry. Acute Physiology and Chronic Health Evaluation (APACHE) predictions were retrieved for all patients with STEMI and were analyzed for complications, length of stay, and inhospital mortality. We then examined subjective criteria to triage patients with STEMI out of the ICU. Among 253 patients with STEMI, 179 (70.75%) were classified as low risk (intermediate level care appropriate) and 74 (29.25%) were classified as high risk (ICU appropriate). The mean age was 64.95 years. The APACHE III score was right skewed with a mean of 36.97 and a median of 31. There was a significant difference between the APACHE III score of low-risk patients and the APACHE III score of high-risk patients (P < .001). In conclusion, patients characterized as low risk, as defined by our criteria, had low APACHE III scores and a low likelihood of complications post-STEMI. This low-risk population could potentially be admitted to an intermediate level of care, avoiding the ICU altogether.

The value of missing information in severity of illness score development.

We aim to investigate the hypothesis that using information about which variables are missing along with appropriate imputation improves the performance of severity of illness scoring systems used to predict critical patient outcomes. We quantify the impact of missing and imputed variables on the performance of prediction models used in the development of a sepsis-related severity of illness scoring system. Electronic health records (EHR) data were compiled from Christiana Care Health System (CCHS) on 119,968 adult patients hospitalized between July 2013 and December 2015. Two outcomes of interest were considered for prediction: (1) first transfer to intensive care unit (ICU) and (2) in-hospital mortality. Five different prediction models were employed. Indicators were utilized in these prediction models to identify when variables were missing and imputed. We observed statistically significant gains in prediction performance when moving from models that did not indicate missing information to those that did. Moreover, this increase was higher in models that use summary variables as predictors compared to those that use all variables. When developing prediction models using longitudinal EHR data, researchers should explore the incorporation of indicators for missing variables along with appropriate imputation.

Validation and Modification of a Severity of Illness Score for Children With Acute Hematogenous Osteomyelitis.

Children with osteomyelitis demonstrate a wide spectrum of illness. Objective measurement of severity is important to guide resource allocation and treatment decisions, particularly for children with advanced illness. The purpose of this study is to validate and improve a previously published severity of illness scoring system for children with acute hematogenous osteomyelitis (AHO). Children with AHO were prospectively studied during evaluation and treatment by a multidisciplinary team who provided care according to evidence-based guidelines to reduce variation. A severity of illness score was calculated for each child and correlated with surrogate measures of severity. Univariate analysis was used to assess the significance of each parameter within the scoring model along with new parameters, which were evaluated to improve the model. The scoring system was then modified by the addition of band count to replace respiratory rate. The modified score was calculated and applied to the prospective cohort followed by correlation with the surrogate measures of severity. One hundred forty-eight children with AHO were consecutively studied. The original severity of illness score correlated well with length of stay and other established measures of severity. Band percent of the white blood cell differential ≥1.5% was found to be significantly associated with severity and chosen to replace respiratory rate in the model. The modified calculated severity scores correlated well with the chosen surrogate measures and significantly differentiated children with osteomyelitis on the basis of causative organism, length of stay, intensive care, surgeries, bacteremia, and disseminated or multifocal disease. The findings of this study validate the previously published severity of illness scoring tool in large cohort of children who were prospectively evaluated. The replacement of respiratory rate with band count improved the scoring system.

An Evaluation of the Influence of Body Mass Index on Severity Scoring.

Although one third or more of critically ill patients in the United States are obese, obesity is not incorporated as a contributing factor in any of the commonly used severity of illness scores. We hypothesize that selected severity of illness scores would perform differently if body mass index categorization was incorporated and that the performance of these score models would improve after consideration of body mass index as an additional model feature. Retrospective cohort analysis from a multicenter ICU database which contains deidentified data for more than 200,000 ICU admissions from 208 distinct ICUs across the United States between 2014 and 2015. First ICU admission of patients with documented height and weight. One-hundred eight-thousand four-hundred two patients from 189 different ICUs across United States were included in the analyses, of whom 4,661 (4%) were classified as underweight, 32,134 (30%) as normal weight, 32,278 (30%) as overweight, 30,259 (28%) as obese, and 9,070 (8%) as morbidly obese. None. To assess the effect of adding body mass index as a risk adjustment element to the Acute Physiology and Chronic Health Evaluation IV and Oxford Acute Severity of Illness scoring systems, we examined the impact of this addition on both discrimination and calibration. We performed three assessments based upon 1) the original scoring systems, 2) a recalibrated version of the systems, and 3) a recalibrated version incorporating body mass index as a covariate. We also performed a subgroup analysis in groups defined using World Health Organization guidelines for obesity. Incorporating body mass index into the models provided a minor improvement in both discrimination and calibration. In a subgroup analysis, model discrimination was higher in groups with higher body mass index, but calibration worsened. The performance of ICU prognostic models utilizing body mass index category as a scoring element was inconsistent across body mass index categories. Overall, adding body mass index as a risk adjustment variable led only to a minor improvement in scoring system performance.

EResearch in acute kidney injury: a primer for electronic health record research.

Acute kidney injury (AKI) has a significant impact on patient morbidity and mortality as well as overall health care costs. eResearch, which integrates information technology and information management to optimize research strategies, provides a perfect platform for necessary ongoing AKI research. With the recent adoption of a widely accepted definition of AKI and near-universal use of electronic health records, eResearch is becoming an important tool in AKI research. Conducting eResearch in AKI should ideally be based on a relatively uniform methodology. This article is the first of its kind to describe a methodology for pursuing eResearch specific to AKI and includes an illustrative database example for critically ill patients. We discuss strategies for using serum creatinine and urine output in large databases to identify and stage AKI and ways to interpolate missing values and validate data. Issues specific to the pediatric population include variation in serum creatinine with growth, varied severity of illness scoring systems and medication dosage based on weight. Many of these same strategies used to optimize AKI eResearch can be applicable to real-time AKI alerts with potential integration of additional clinical variables.

Using Illness Rating Systems to Predict Discharge Location Following Total Knee Arthroplasty.

PurposeTotal knee arthroplasty (TKA) is increasing in frequency and cost. Optimization of discharge location may reduce total expenditure while maximizing patient outcomes. Although preoperative illness rating systems—including the American Society for Anesthesiologists Physical Classification System (ASA), severity of illness scoring system (SOI), and Mallampati rating scale (MP)—are associated with patient morbidity and mortality, their predictive value for discharge location, length of stay (LOS), and total costs remains unclear.Materials and MethodsWe conducted a retrospective analysis of 677 TKA patients (550 primary and 127 revision) treated at a single institution. The influence of ASA, SOI, and MP scores on discharge locations, LOS, and total costs was assessed using multivariable regression analyses.ResultsNone of the systems were significant predictors of discharge location following TKA. SOI scores of major or higher (β=2.08 days, p<0.001) and minor (β=−0.25 days, p=0.009) significantly predicted LOS relative to moderate scores. Total costs were also significantly predicted by SOI scores of major or higher (β=$6,155, p=0.022) and minor (β=−$1,163, p=0.007).ConclusionsSOI scores may be harnessed as a predictive tool for LOS and total costs following TKA, but other mechanisms are necessary to predict discharge location.

Early hospital mortality prediction of intensive care unit patients using an ensemble learning approach

BackgroundMortality prediction of hospitalized patients is an important problem. Over the past few decades, several severity scoring systems and machine learning mortality prediction models have been developed for predicting hospital mortality. By contrast, early mortality prediction for intensive care unit patients remains an open challenge. Most research has focused on severity of illness scoring systems or data mining (DM) models designed for risk estimation at least 24 or 48h after ICU admission. ObjectivesThis study highlights the main data challenges in early mortality prediction in ICU patients and introduces a new machine learning based framework for Early Mortality Prediction for Intensive Care Unit patients (EMPICU). Materials and methodsThe proposed method is evaluated on the Multiparameter Intelligent Monitoring in Intensive Care II (MIMIC-II) database. Mortality prediction models are developed for patients at the age of 16 or above in Medical ICU (MICU), Surgical ICU (SICU) or Cardiac Surgery Recovery Unit (CSRU). We employ the ensemble learning Random Forest (RF), the predictive Decision Trees (DT), the probabilistic Naive Bayes (NB) and the rule-based Projective Adaptive Resonance Theory (PART) models. The primary outcome was hospital mortality. The explanatory variables included demographic, physiological, vital signs and laboratory test variables. Performance measures were calculated using cross-validated area under the receiver operating characteristic curve (AUROC) to minimize bias. 11,722 patients with single ICU stays are considered. Only patients at the age of 16 years old and above in Medical ICU (MICU), Surgical ICU (SICU) or Cardiac Surgery Recovery Unit (CSRU) are considered in this study. ResultsThe proposed EMPICU framework outperformed standard scoring systems (SOFA, SAPS-I, APACHE-II, NEWS and qSOFA) in terms of AUROC and time (i.e. at 6h compared to 48h or more after admission). Discussion and conclusionThe results show that although there are many values missing in the first few hour of ICU admission, there is enough signal to effectively predict mortality during the first 6h of admission. The proposed framework, in particular the one that uses the ensemble learning approach – EMPICU Random Forest (EMPICU-RF) offers a base to construct an effective and novel mortality prediction model in the early hours of an ICU patient admission, with an improved performance profile.

Performance of Simplified Acute Physiology Score 3 In Predicting Hospital Mortality In Emergency Intensive Care Unit.

Background:Since the 1980s, severity of illness scoring systems has gained increasing popularity in Intensive Care Units (ICUs). Physicians used them for predicting mortality and assessing illness severity in clinical trials. The objective of this study was to assess the performance of Simplified Acute Physiology Score 3 (SAPS 3) and its customized equation for Australasia (Australasia SAPS 3, SAPS 3 [AUS]) in predicting clinical prognosis and hospital mortality in emergency ICU (EICU).Methods:A retrospective analysis of the EICU including 463 patients was conducted between January 2013 and December 2015 in the EICU of Peking University Third Hospital. The worst physiological data of enrolled patients were collected within 24 h after admission to calculate SAPS 3 score and predicted mortality by regression equation. Discrimination between survivals and deaths was assessed by the area under the receiver operator characteristic curve (AUC). Calibration was evaluated by Hosmer-Lemeshow goodness-of-fit test through calculating the ratio of observed-to-expected numbers of deaths which is known as the standardized mortality ratio (SMR).Results:A total of 463 patients were enrolled in the study, and the observed hospital mortality was 26.1% (121/463). The patients enrolled were divided into survivors and nonsurvivors. Age, SAPS 3 score, Acute Physiology and Chronic Health Evaluation Score II (APACHE II), and predicted mortality were significantly higher in nonsurvivors than survivors (P < 0.05 or P < 0.01). The AUC (95% confidence intervals [CIs]) for SAPS 3 score was 0.836 (0.796–0.876). The maximum of Youden's index, cutoff, sensitivity, and specificity of SAPS 3 score were 0.526%, 70.5 points, 66.9%, and 85.7%, respectively. The Hosmer-Lemeshow goodness-of-fit test for SAPS 3 demonstrated a Chi-square test score of 10.25, P = 0.33, SMR (95% CI) = 0.63 (0.52–0.76). The Hosmer-Lemeshow goodness-of-fit test for SAPS 3 (AUS) demonstrated a Chi-square test score of 9.55, P = 0.38, SMR (95% CI) = 0.68 (0.57–0.81). Univariate and multivariate analyses were conducted for biochemical variables that were probably correlated to prognosis. Eventually, blood urea nitrogen (BUN), albumin, lactate and free triiodothyronine (FT3) were selected as independent risk factors for predicting prognosis.Conclusions:The SAPS 3 score system exhibited satisfactory performance even superior to APACHE II in discrimination. In predicting hospital mortality, SAPS 3 did not exhibit good calibration and overestimated hospital mortality, which demonstrated that SAPS 3 needs improvement in the future.

Assessment and comparison of 3 mortality prediction models SAPS II, APACHE II and SOFA for prediction of mortality in patients of sepsis

Background: Little is known about outcomes of patients admitted to the ICU with severe sepsis and septic shock, despite the seriousness of sepsis as a public health problem in developing countries. Understanding sepsis outcome studies is hampered by lack of an agreed severity of illness scoring system for sepsis patients. The objective of the present study is to assess and compare the validity of 3 mortality prediction models SAPS 2, APACHE II and SOFA for prediction of mortality in patients of sepsis.Methods: One hundred patients of Sepsis were selected after applying the inclusion and exclusion criteria. Informed consent was taken from the patients or their relatives A careful and detailed history was recorded to assess the onset and duration of clinical events and the probable risk factors for the same; a detailed general physical examination was performed. Blood sampling for CBC, RFT, LFT and arterial blood gas analysis was done. SAPS 2, APACHE II and SOFA scores were calculated on the day of admission.Results: The ROC analysis shows that the best discrimination was provided by SAPS 2 score (AUROC=0.981), followed by APACHE II (AUROC=0.978) and SOFA (AUROC=0.911).Conclusions: SAPS 2 score was superior to the APACHE II and SOFA scores for predicting survival in patients with septic shock but a combination of factors must be taken in consideration to estimate the prognosis in the ICU.

Validation of SICK score - A severity of illness scoring system at a tertiary care centre in Puducherry

Objectives: Validation of SICK score for its ability to predict out come. Validation based on predominant system involved. Methods: Prospective study done at tertiary care centre in puducherry. Inclusion criteria: All children admitted in PICU between 1 month & 12 year in 6 month period. Exclusion criteria: Patients with surgical illness, congenital anomalies, non availability of consent. Physical variables of Temperature, Heart Rate, Respiratory Rate, Blood pressure, Capiallary Refill time, Oxygen saturation & sensoriurn (using AVPU scale) were recorded at the time of admission. SICK score was calculated and outcome was recorded as survived or expired. Association of patients outcome with total score was done using logistic regression analysis. The receiver operating characteristic (ROC) curve analysis was performed to decide the optimum cut off point in terms of total score that provides maximum sensitivity & specificity. Area under ROC curve (AUC) was used to compare the predictive power of total score. All analysis was carried out using SPPS software version 13.0. Results: Total of 354 Children were analyzed. For SICK score of <1 mortality was 0%, which gradually increased to 100% with a score of 7 or more. The area under the ROC curve was 93% (95%: CI: 89.8% - 97.3%) indicating good predictive ability of the score. Conclusion: SICK score has good predictive ability and uses only physical criteria. Further SICK score is not influenced by predominant system involved.

Comparison of acute physiology and chronic health evaluation II (APACHE II) and acute physiology and chronic health evaluation IV (APACHE IV) severity of illness scoring systems, in a multidisciplinary ICU.

Background and Aims:Outcome prediction of critically ill patients is an integral part of care in an Intensive Care Unit (ICU). Acute Physiology and Chronic Health Evaluation (APACHE) scoring systems provide an objective means of mortality prediction in ICU. The aim of this study was to compare the performance of APACHE II and IV scoring system in our ICU.Material and Methods:All patients admitted to the ICU between January and June 2014 and who met the inclusion criteria were evaluated. APACHE II and IV score were calculated during the first 24 h of ICU stay based on the worst values. All patients were followed up till discharge from the hospital or death. Statistical analysis was performed using SPSS version 19.0. Discrimination of the model for mortality was assessed using receiver operating characteristic curve and calibration was assessed using the Hosmer-Lemeshow goodness-of-fit test.Results:Of a total 1268, 1003 patients were included in this study. The mean (±standard deviation) admission APACHE II score was 19.4 ± 8.9, and APACHE IV score was 59.1 ± 27.2. The APACHE scores were significantly higher among nonsurvivors than survivors (P < 0.001). The overall crude hospital mortality rate was 17.6%. APACHE IV had better discriminative power area under the ROC curve ([AUC] –0.82) than APACHE II (AUC-0.75). Both APACHE II and APACHE IV had poor calibration.Conclusions:APACHE IV showed better discrimination compared to APACHE II in our ICU population. Both APACHE II and APACHE IV had poor calibration. However, APACHE II calibrated better compared to APACHE IV.

Multiple Organ Dysfunction Score Is Superior to the Obstetric-Specific Sepsis in Obstetrics Score in Predicting Mortality in Septic Obstetric Patients.

Objectives:Mortality prediction scores have been used for a long time in ICUs; however, numerous studies have shown that they over-predict mortality in the obstetric population. With sepsis remaining a major cause of obstetric mortality, we aimed to look at five mortality prediction scores (one obstetric-based and four general) in the septic obstetric population and compare them to a nonobstetric septic control group.Subject and Design:Women in the age group of 16–50 years with an admission diagnosis or suspicion of sepsis were included. In a multicenter obstetric population (n = 797), these included all pregnant and postpartum patients up to 6 weeks postpartum. An age- and gender-matched control nonobstetric population was drawn from a single-center general critical care population (n = 2,461). Sepsis in Obstetric Score, Acute Physiology and Chronic Health Evaluation II, Simplified Acute Physiology Score II, Sequential Organ Failure Assessment, and Multiple Organ Dysfunction Scores were all applied to patients meeting inclusion criteria in both cohorts, and their area under the receiver-operator characteristic curves was calculated to find the most accurate predictor.Measurements and Main Results:A total of 146 septic patients were found for the obstetric cohort and 299 patients for the nonobstetric control cohort. The Sepsis in Obstetric Score, Acute Physiology and Chronic Health Evaluation II, Simplified Acute Physiology Score II, Sequential Organ Failure Assessment, and Multiple Organ Dysfunction Scores gave area under the receiver-operator characteristic curves of 0.67, 0.68, 0.72, 0.79, and 0.84 in the obstetric cohort, respectively, and 0.64, 0.72, 0.61, 0.78, and 0.74 in the nonobstetric cohort, respectively. The Sepsis in Obstetric Score performed similarly to all the other scores with the exception of the Multiple Organ Dysfunction Score, which was significantly better (p < 0.05).Conclusion:The Sepsis in Obstetric Score, designed specifically for sepsis in obstetric populations, was not better than general severity of illness scoring systems. Furthermore, the Sepsis in Obstetric Score performance was no different in an obstetric sepsis population compared to a nonobstetric sepsis population. The Multiple Organ Dysfunction Score is a simple organ-based score, and this result supports the use of organ-based outcome predictors in ICU even in an obstetric sepsis population.