[Reliability of the measurement of glucose at the bedside of critical patients]. / Fiabilidad de la determinación de la glucemia a la cabecera del paciente en pacientes críticos.

INTRODUCTION: The objective of this study is to measure the reliability of three measurement methods at the bedside of the patient, of glucose in the critical patient compared with the measurement of glucose in the central laboratory. MATERIAL AND METHODS: Observational, perspective study developed in a polyvalent unit of 18 beds for four months. Patients who had arterial catheter were included. Eight samples obtained at the patient's bedside were compared with the plasma glucose (gold Standard): three in capillary blood, four in arterial blood and one in arterial blood gases from a syringe. The measurements at bedside were conducted with reactive strips MediSense Optium Plus and glucometer MediSense Optium. A comparison was made of the means used in the Student's T test and Bland and Altman analysis. RESULTS: We obtained 630 samples in 70 patients. Mean glucose (SD) in mg/dl was: a) capillary samples: 149 (38), 149 (35), 147 (37); b) arterial samples: 140 (34), 142 (35), 143 (35), 142 (34); arterial gas sample syringe: 143 (33); c) plasma glucose: 138(33). There were significant differences (p < 0.001) between plasma glucose and capillary samples but not with arterial samples (p=0.2). In the arterial samples, the presence of some factors, such as vasoactive drugs, glycated solution perfusion, insulin perfusion and plasma concentration of hemoglobin, increase error and dispersion regarding the gold standard. CONCLUSIONS: The measurement of glucose at bedside in critical patients is more reliable in arterial samples than in capillary ones.

Fiabilidad de la determinación de la glucemia a la cabecera del paciente en pacientes críticos / Reliability of the measurement of glucose at the bedside of critical patients

Introducción. El objetivo de este estudio es determinar la fiabilidad de tres métodos de determinación, a pie de cama, de la glucemia en el paciente crítico comparados con la determinación de glucemia en el laboratorio central. Material y métodos. Estudio observacional prospectivo desarrollado en una Unidad polivalente de 18 camas durante 4 meses. Se incluyeron pacientes que portaban catéter arterial. Se compararon con la glucemia plasmática (patrón oro) 8 muestras obtenidas a la cabecera del paciente: tres en sangre capilar, 4 en sangre arterial y una de sangre arterial en jeringa de gases. Las determinaciones a la cabecera fueron realizadas con tiras reactivas MediSense® Optium™ Plus y glucómetro MediSense® Optium™. Se realizó una comparación de medias mediante la prueba de la «t» de Student y análisis de Bland y Altman. Resultados. Obtuvimos 630 muestras en 70 pacientes. La glucemia media (desviación estándar [DE]) en mg/dl fue: a) muestras capilares: 149 (38), 149 (35), 147 (37); b) muestras arteriales: 140(34), 142 (35), 143 (35), 142 (34); muestra arterial en jeringa de gases: 143 (33); c) glucemia plasmática: 138 (33). Hubo diferencias significativas (p < 0,001) entre la glucemia plasmática y las muestras capilares, pero no con las muestras arteriales (p = 0,2). En las muestras arteriales la presencia de algunos factores, como fármacos vasoactivos, perfusión de soluciones glucosadas, perfusión de insulina y concentración plasmática de hemoglobina, aumenta el error y la dispersión respecto al patrón oro. Conclusiones. En enfermos críticos la medida de la glucemia a pie de cama es más fiable en muestras arteriales que en muestras capilares

Introduction. The objective of this study is to measure the reliability of three measurement methods at the bedside of the patient, of glucose in the critical patient compared with the measurement of glucose in the central laboratory. Material and methods. Observational, perspective study developed in a polyvalent unit of 18 beds for four months. Patients who had arterial catheter were included. Eight samples obtained at the patient's bedside were compared with the plasma glucose (gold Standard): three in capillary blood, four in arterial blood and one in arterial blood gases from a syringe. The measurements at bedside were conducted with reactive strips MediSense® Optium™ Plus and glucometer MediSense® Optium™. A comparison was made of the means used in the Student's T test and Bland and Altman analysis. Results. We obtained 630 samples in 70 patients. Mean glucose (SD) in mg/dl was: a) capillary samples: 149 (38), 149 (35), 147 (37); b) arterial samples: 140 (34), 142 (35), 143 (35), 142 (34); arterial gas sample syringe: 143 (33); c) plasma glucose: 138(33). There were significant differences (p < 0.001) between plasma glucose and capillary samples but not with arterial samples (p=0.2). In the arterial samples, the presence of some factors, such as vasoactive drugs, glycated solution perfusion, insulin perfusion and plasma concentration of hemoglobin, increase error and dispersion regarding the gold standard. Conclusions. The measurement of glucose at bedside in critical patients is more reliable in arterial samples than in capillary ones

[Estimate of creatinine clearance with formulae: a valid option compared with conventional measurement?]. / Estimación del aclaramiento de creatinina mediante fórmulas: una opción válida frente a la medición convencional?

Determination of creatinine clearance is the most commonly used method to evaluate renal function, requiring collection of urine from a given period of time, usually 24 hours. Cockcroft and Gault developed a formula to estimate this parameter without urine samples, however some authors have shown the low reliability of this and similar nomograms. The present study is an attempt to evaluate the differences between the laboratory determination of this parameter and its evaluation using Cockcroft's equation. The study included 24 patients who had creatinine clearance determined through biochemical methods and with the formula proposed by Cockcroft. Mean clearance in the first instance was 81.4 mL/min and 65.5 mL/min in the second one. The estimated mean difference in creatinine clearance obtained by both methods was 16.4 mL/min. A significant difference between clearance determinations was found using the above methods and therefore it is concluded that the values obtained with Cockcroft's formula are hardly reliable since the method tends to subestimate creatinine clearance. Also, two new formulas were developed which correlate better with the results obtained in our series. To conclude, we recommend that creatinine clearance should be measured at the laboratory using serum and 24-hour creatinine determination.