Javier Trujillano

Pleural fluid interleukin‐8 and C‐reactive protein for discriminating complicated non‐purulent from uncomplicated parapneumonic effusions

Background and objective:  This study was designed to test the hypothesis that measurement of IL‐8 and CRP in pleural fluid could improve the identification of patients with non‐purulent parapneumonic effusions that ultimately require chest tube drainage.

Stratification of the severity of critically ill patients with classification trees

BackgroundDevelopment of three classification trees (CT) based on the CART (Classification and Regression Trees), CHAID (Chi-Square Automatic Interaction Detection) and C4.5 methodologies for the calculation of probability of hospital mortality; the comparison of the results with the APACHE II, SAPS II and MPM II-24 scores, and with a model based on multiple logistic regression (LR).MethodsRetrospective study of 2864 patients. Random partition (70:30) into a Development Set (DS) n = 1808 and Validation Set (VS) n = 808. Their properties of discrimination are compared with the ROC curve (AUC CI 95%), Percent of correct classification (PCC CI 95%); and the calibration with the Calibration Curve and the Standardized Mortality Ratio (SMR CI 95%).ResultsCTs are produced with a different selection of variables and decision rules: CART (5 variables and 8 decision rules), CHAID (7 variables and 15 rules) and C4.5 (6 variables and 10 rules). The common variables were: inotropic therapy, Glasgow, age, (A-a)O2 gradient and antecedent of chronic illness. In VS: all the models achieved acceptable discrimination with AUC above 0.7. CT: CART (0.75(0.71-0.81)), CHAID (0.76(0.72-0.79)) and C4.5 (0.76(0.73-0.80)). PCC: CART (72(69-75)), CHAID (72(69-75)) and C4.5 (76(73-79)). Calibration (SMR) better in the CT: CART (1.04(0.95-1.31)), CHAID (1.06(0.97-1.15) and C4.5 (1.08(0.98-1.16)).ConclusionWith different methodologies of CTs, trees are generated with different selection of variables and decision rules. The CTs are easy to interpret, and they stratify the risk of hospital mortality. The CTs should be taken into account for the classification of the prognosis of critically ill patients.

Time spent in the emergency department and mortality rates in severely injured patients admitted to the intensive care unit: An observational study

PURPOSE The aim of this study was to identify the determinants of a shorter emergency department time (EDt) in patients with severe trauma (STPs) admitted to the intensive care unit and determine whether EDt influences mortality. PATIENTS AND METHODS A prospective observational study of STPs (2005-2007) was conducted. With the variables available from the ED, 2 multiple logistic regression models (MLRM) were created: one for the factors associated with EDt less than or equal to median and the other with mortality. RESULTS A total of 243 patients were included. The mean age was 43 years; 76% were male. The overall mortality rate was 20%. The median EDt was 120 minutes. The independent factors that were associated with the MLRM for an EDt of 120 minutes or less included age less than 60 years, mechanical ventilation, severe traumatic brain injury, and a trauma and injury severity score of 20 or higher. The MLRM for mortality was age greater than 60 years, mechanical ventilation, traumatic brain injury and shock. An EDt of 120 minutes or less was associated with an increased risk of death in the univariate analysis but not in the MLRM. CONCLUSIONS Patients in the ED with indicators of high trauma severity have a reduced EDt but a higher mortality rate. Advanced age increases both mortality and EDt. With the factors included in the model, EDt was not an independent factor for mortality in STPs.

Aproximación a la metodología basada en árboles de decisión (CART). Mortalidad hospitalaria del infarto agudo de miocardio

Objective: To provide an overview of decision trees based on CART (Classification and Regression Trees) methodology. As an example, we developed a CART model intended to estimate the probability of intrahospital death from acute myocardial infarction (AMI). Method: We employed the minimum data set (MDS) of Andalusia, Catalonia, Madrid and the Basque Country (20012002), which included 33,203 patients with a diagnosis of AMI. The 33,203 patients were randomly divided (70% and 30%) into the development (DS; n = 23,277) and the validation (VS; n = 9,926) sets. The CART inductive model was based on Breiman’s algorithm, with a sensitivity analysis based on the Gini index and cross-validation. We compared the results with those obtained by using both logistic regression (LR) and artificial neural network (ANN) (multilayer perceptron) models. The developed models were contrasted with the VS and their properties were evaluated with the area under the ROC curve (AUC) (95% confidence interval [CI]). Results: In the DS, the CART showed an AUC = 0.85 (0.860.88), LR 0.87 (0.86-0.88) and ANN 0.85 (0.85-0.86). In the VS, the CART showed an AUC = 0.85 (0.85-0.88), LR 0.86 (0.85-0.88) and ANN 0.84 (0.83-0.86). Conclusions: None of the methods tested outperformed the others in terms of discriminative ability. We found that the CART model was much easier to use and interpret, because the decision rules generated could be applied without the need for mathematical calculations.

Aproximación metodológica al uso de redes neuronales artificiales para la predicción de resultados en medicina

: In clinical practice, there is an increasing interest in obtaining adequate models of prediction. Within the possible available alternatives, the artificial neural networks (ANN) are progressively more used. In this review we first introduce the ANN methodology, describing the most common type of ANN, the Multilayer Perceptron trained with backpropagation algorithm (MLP). Then we compare the MLP with the Logistic Regression (LR). Finally, we show a practical scheme to make an application based on ANN by means of an example with actual data. The main advantage of the RN is its capacity to incorporate nonlinear effects and interactions between the variables of the model without need to include them a priori. As greater disadvantages, they show a difficult interpretation of their parameters and large empiricism in their process of construction and training. ANN are useful for the computation of probabilities of a given outcome based on a set of predicting variables. Furthermore, in some cases, they obtain better results than LR. Both methodologies, ANN and LR, are complementary and they help us to obtain more valid models.

Left carotid adventitial vasa vasorum signal correlates directly with age and with left carotid intima-media thickness in individuals without atheromatous risk factors

ObjectiveThe early identification of the onset of subclinical atheromatosis is essential in reducing the high mortality risk from cardiovascular disease (CVD) worldwide. Although carotid intima-media thickness (cIMT) is the most commonly used early predictor of ongoing atherosclerosis, an experimental model of atherosclerosis, demonstrated that increases in adventitial microvessels (vasa vasorum (VV)) precede endothelial dysfunction. Using the reported accuracy of contrast-enhanced ultrasound (CEU) to measure carotid adventitial VV, this study assessed whether measurements of carotid adventitial VV serve as a marker of subclinical atherosclerotic lesions in a control population with none of the classical risk factors for CVD.Methods and resultsMeasurements of cIMT (B-mode ultrasound) and adventitial VV (CEU) were conducted in 65 subjects, 30–70 years old, 48% men, with none of the classical risk factors for CVD. Adventitial VV strongly correlated with its own cIMT only in the left carotid artery. Importantly, the left carotid adventitial VV directly correlated with age.ConclusionsThe increases with age in left carotid adventitial VV in individuals with zero risk for atheromatosis suggest that the measurement of carotid adventitial VV could be an accurate and sensitive marker for the diagnosis of subclinical atheromatosis and therefore a prominent tool for monitoring the efficacy of anti-atheromatous therapies.

Atención del paciente crítico pediátrico en una UCI de adultos: utilidad del índice PIM

OBJECTIVES An analysis is made of the characteristics of patients younger than 14 years treated in an adult ICU (AICU), to determine the procedures and techniques required by such patients, and to evaluate the use of the Pediatric Index of Mortality (PIM) in stratifying severity. DESIGN A retrospective observational study was carried out. SETTING An AICU of a second level hospital. PATIENTS We studied 130 patients aged from 1 month to 14 years (average age 6.1±4 years) treated in the AICU from January 1997 to December 2010. VARIABLES OF INTEREST Clinical-demographic parameters, diagnosis, clinical procedures, PIM score, length of stay, transfer to pediatric ICU (PICU), and mortality. Classification by destination (AICU, PICU) and outcome (alive, dead). PIM and assessment of the diagnostic performance curve (ROC) for mortality. RESULTS The average age of the patients was 6.1±4 years. Most common diagnoses: trauma (26.9%) and sepsis (22.3%). Main procedures: mechanical ventilation (58.5%), central venous line (74.6%) and vasoactive drugs (20%). A total of 64.6% were transferred to PICU, and the overall mortality was 13%. Patients who stayed in the AICU were older (8.2±4 vs 5.5±4 years, p<0.001), had low morbidity, and their stay was short (44.5±38 hours). The PIM score was significantly higher in the patients who died (60±20 AICU, 38±30 PICU) than in those who survived (4±1 AICU, 9±1 PICU) (p<0.001). ROC curve with AUC=0.91 (95%CI: 0.85 to 0.98). CONCLUSIONS The PIM score can stratify severity and identify patients at an increased risk of death. Critical child care in the AICU requires the presence of adequate materials and the continuous learning of procedures adapted to pediatric patients in order to ensure adequate care.

Approach to the methodology of classification and regression trees

OBJECTIVE To provide an overview of decision trees based on CART (Classification and Regression Trees) methodology. As an example, we developed a CART model intended to estimate the probability of intrahospital death from acute myocardial infarction (AMI). METHOD We employed the minimum data set (MDS) of Andalusia, Catalonia, Madrid and the Basque Country (2001-2002), which included 33,203 patients with a diagnosis of AMI. The 33,203 patients were randomly divided (70% and 30%) into the development (DS; n = 23,277) and the validation (VS; n = 9,926) sets. The CART inductive model was based on Breimans algorithm, with a sensitivity analysis based on the Gini index and cross-validation. We compared the results with those obtained by using both logistic regression (LR) and artificial neural network (ANN) (multilayer perceptron) models. The developed models were contrasted with the VS and their properties were evaluated with the area under the ROC curve (AUC) (95% confidence interval [CI]). RESULTS In the DS, the CART showed an AUC = 0.85 (0.86-0.88), LR 0.87 (0.86-0.88) and ANN 0.85 (0.85-0.86). In the VS, the CART showed an AUC = 0.85 (0.85-0.88), LR 0.86 (0.85-0.88) and ANN 0.84 (0.83-0.86). CONCLUSIONS None of the methods tested outperformed the others in terms of discriminative ability. We found that the CART model was much easier to use and interpret, because the decision rules generated could be applied without the need for mathematical calculations.

Factores de riesgo de la disfunción hepática asociada a la nutrición parenteral

BACKGROUND AND OBJECTIVE The objective of this study is to describe the incidence of hepatic dysfunction (HD) in our hospital and evaluate the possible risk factors associated with HD development as an improvement of the caring process received by patients treated with parenteral nutrition (PN). PATIENTS AND METHOD A prospective study of patients (n=994) who required PN during the period 2000-2004. HD is the identification of an increase above 1,5 of the top reference value of alkaline phosphatase (40-450U/l) and gamma glutamyl transpeptidase (11-49U/l) associated with an increase of transaminases (5-32U/l) and a total bilirrubin higher than 1,2mg/dl. RESULTS The incidence of HD was 4,9% (n=49). Days with PN were significantly higher in the HD group: median (interquartile range): 30 (20-59) vs 15 (8-25) days (p<0.001). In the univariated HD analysis, the variables that reached significant odds ratio were: the critical patient condition, the PN duration, the total calorie contribution higher than 25kcal/kg, to exceed 3g of carbohydrates/kg, to administer more than 0.8g/kg of lipids and to exceed 0.16g of nitrogen/kg. In the multivariated analysis, the variables selected as independent risk factors were: to exceed 3 weeks of PN, to be a critical patient and a contribution over 0.16g of nitrogen/kg. CONCLUSIONS The present profiles of the patients who will develop HD are those with prolonged PN. These patients undergo processes and critical therapy, where the specialists must monitor, not only calorie contribution, carbohydrates or lipids, but proteins as well.

Classification tree analysis for the discrimination of pleural exudates and transudates

Abstract Background: Classification and regression tree (CART) analysis is a non-parametric technique suitable for the generation of clinical decision rules. We have studied the performance of CART analysis in the separation of pleural exudates and transudates. Methods: Basic demographic, radiologic and laboratory data were retrospectively evaluated in 1257 pleural effusions (204 transudates and 1053 exudates, according to standard clinical criteria) and submitted for CART analysis. The models discriminative ability was compared with that of Lights criteria, in both the original formulation and an abbreviated version, i.e., deleting the pleural fluid (PF)/serum lactate dehydrogenase (LDH) ratio from the triad. Results: A first CART model built starting from all available data identified PF/serum protein ratio and PF LDH ratios as the two best discriminatory parameters. This algorithm achieved a sensitivity of 96.8%, slightly lower than that of classical Lights criteria (98.5%) and comparable to that of the abbreviated Lights criteria (97.0%), and significantly better specificity (85.3%) compared to both classical (74.0%) and abbreviated (79.4%) Lights criteria. A second CART model developed after excluding serum measurements selected PF protein and PF LDH as the most discriminatory variables, and correctly classified 97.2% of exudates and 77.0% of transudates. Conclusions: CART-based algorithms can efficiently discriminate between pleural exudates and transudates. Clin Chem Lab Med 2007;45:82–7.