Elevación de creatinina sérica como indicador de necrosis pancreática / Elevation of serum creatinine as an indicator of pancreatic necrosis

La Pancreatitis Aguda (PA) es una patología que se autolimita en el 80% de los casos; estos casos en general evolucionan hacia la recuperación total. Su evolución puede ser de leve a severa. La forma grave varía desde un 10% a un 25% y se asocia con falla orgánica y/o complicaciones locales como necrosis pancreática. En Venezuela, la PA Severa es un importante problema de salud pública, encontrándose entre las primeras 25 causas de muerte. Este estudio plantea la utilización de una prueba de laboratorio ampliamente disponible, de fácil uso e interpretación, para pronosticar la aparición de complicaciones como necrosis pancreática. Objetivo: Determinar la utilidad de la creatinina sérica como factor predictivo de necrosis pancreática en pancreatitis aguda. Materiales y Metodos: Estudio de tipo analítico, transversal y retrospectivo. Se revisaron las historias clínicas de los pacientes que ingresaron al hospital “Dr. Miguel Pérez Carreño”, con diagnóstico de pancreatitis aguda entre 2008 y 2009. Se registró creatinina sérica y se relacionó con la clasificación de severidad tomográfica según Balthazar. Resultados: La población estuvo conformada por 50 casos, de éstos se excluyeron 4, por embarazo o enfermedad renal crónica. Treinta de sexo femenino (65%) y 16 masculino (35%). Edades comprendidas entre 18 a 77 años, con media de 40,2. La estancia hospitalaria media fue de 8,74 días. Del total de 46 pacientes, presentó Balthazar A 63% (n=29), B 17,39% (n=8), C 14,04% (n= 6) y D 6,5% (n=3). No se obtuvo ningún E. Al aplicar un análisis de varianza se observó relación estadística significativa directamente proporcional de la creatinina sérica de ingreso (p=0,001) y de las 48 horas (p=0,001) con el Balthazar y el hematocrito. Conclusiones: La evaluación de los niveles y variaciones de creatinina sérica son de utilidad para predecir la aparición de necrosis pancreática en pacientes con pancreatitis aguda.

Acute Pancreatitis (AP) is a self-limited pathology in 80% of the cases; these cases generally evolve towards total recovery. Its evolution can be mild or severe. The severe form varies from a 10% to a 25%, and is associated with organ failure and/or local complications as pancreatic necrosis. In Venezuela severe AP is an important public health problem, being in the first 25 causes of death. The present study proposes the use of a widely available laboratory test, of easy use and interpretation, to predict the appearance of complications as pancreatic necrosis. Objective: To determine the usefulness of the serum creatinine as predictive factor of pancreatic necrosis in acute pancreatitis. Materials and Methods: An analytic, transversal and retrospective study. Clinical histories of patients admitted to the “Dr. Miguel Perez Carreño” hospital with a diagnosis of acute pancreatitis between years 2008 and 2009 were reviewed. Serum creatinine was registered and compared according to the Balthazar classification of tomographic severity. Results: The sample was comprised by 50 cases, from which 4 were excluded, due to pregnancy or chronic renal disease. Thirty were female (65%) and 16 male (35%). Between the ages of 18 and 77, with mean age of 40,2. The median hospital stay was 8,74 days. From total of 46 patients, 63% had Balthazar A (n=29); B 17.39% (n=8); C 14,04% (n= 6) and D 6.5% (n=3). No Balthazar E was obtained. When applying a variance analysis, a significant statistical relation was observed, directly proportional to the serum creatinine upon admission (p=0,001) and the 48 hours (p=0,001) with the Balthazar and hematocrit. Conclusions: The evaluation of the levels and variations of serum creatinine is a useful tool for predicting the appearance of pancreatic necrosis in patients with acute pancreatitis.

Calibração da creatinina sérica com uso de equação de regressão: influência na performance da fórmula do estudo MDRD / Serum creatinine calibration with regression equation: influence on MDRD formula performance

Diferenças nos métodos de medida da creatinina sérica podem determinar amplas variações na taxa de filtração glomerular (TFG) estimada com fórmulas. Para a padronização da medida da creatinina, deve ser usado método calibrado rastreável para medida de referência com ID-MS (isotope dilution mass spectrometry). Objetivo: Avaliar a TFG estimada com a equação MDRD (Modification of Diet in Renal Disease) original (MDRDo, creatinina método não calibrado) e a equação MDRD re-expressa (MDRDr, método calibrado por ID-MS), comparando-as com a TFG medida pelo 51Cr-EDTA (método padrão) em indivíduos normais. Métodos: Foram avaliados 101 indivíduos, com idade média de 38±12 anos, sendo 45 homens. A TFG foi medida pela técnica de injeção única do 51Cr-EDTA (TFG 51Cr-EDTA) e estimada pelas equações MDRDo: 186 x creatinina sérica-1,154 x idade-0,203 x 0,742 (se mulher) x 1,210 (se negro) e MDRDr, substituindo-se o valor 186 por 175 na equação. A creatinina sérica foi medida pelo método de Jaffe não calibrado e transformada em calibrado com a fórmula: y=1,07x-0,249, obtida previamente por regressão. A concordância entre os métodos foi avaliada através da análise de Bland&Altman. Resultados: Os valores médios para as TFG 51Cr-EDTA, MDRDr e MDRDo foram de 105±18, 102±21 e 84±13 ml/min/1,73 m², respectivamente. Acurácia (percentual de casos de TFG estimada que não desviam em mais de 15% do valor medido) foi maior com o uso da MDRDr em relação à MDRDo (57% vs. 35%, P=0,002). O viés (diferença entre TFG medida e estimada) para 51Cr-EDTA e MDRDr foi de 3±23 ml/min/1,73 m². Para 51Cr-EDTA e MDRDo o viés foi significativamente maior, sendo de 21±18 ml/min/1,73 m². No entanto, a precisão, avaliada como desvio padrão do viés indicou elevada dispersão nos dois casos. Conclusão: O uso da equação re-expressa do MDRD, empregando a creatinina calibrada, produz uma estimativa mais acurada da TFG do que a equação original do MDRD.

Differences in methods of measurement of serum creatinine may provide wide variations in glomerular filtration rate (GFR) estimated with formulas. To standardize the measurement of creatinine, calibrated methods should be used, traceable to the reference ID-MS (isotope dilution mass spectrometry) method. Aim: To evaluate the performance of GFRs estimated with the original Modification of Diet in Renal Disease study equation (MDRDo; non-calibrated creatinine method) and with the re-expressed MDRD equation (MDRDr; ID-MS creatinine), comparing them with the GFR measured by 51Cr-EDTA (standard method ) in normal adults. Methods: 101 subjects, aged 38±12 years, 45 (45%) men were evaluated. GFR was measured by single-injection 51Cr-EDTA (GFR 51Cr-EDTA) technique and estimated by the following equations - MDRDo: 186 x serum creatinine-1. 154 x age-0.203 x 0.742 (if female) x 1.210 (if black), and MDRDr, replacing the value 186 by 175 in the equation. Serum creatinine was measured by a non-calibrated Jaffe’s method and transformed into calibrated with the formula: y=1.07x-0.249, previously obtained by regression. The agreement between methods was assessed by the Bland&Altman analyses. Results: The mean GFR 51Cr-EDTA, MDRDr and MDRDo were 105±18, 102±21 and 84±13 ml/min/1, 73 m2, respectively. There was no agreement between 51Cr-EDTA and MDRDo GFR (P <0.001), but it was present between 51Cr-EDTA and MDRDr GFR (P=0.149). Accuracy (percentage of cases of estimated GFR within 15% of measured value) was higher with the use of MDRDr in comparison to MDRDo (57% vs. 35%, P=0.002). Bias (diference between measured and estimated GFR) for 51Cr-EDTA and MDRDr was 3±23 ml/min/1.73 m². For 51Cr-EDTA and MDRDo the bias was significantly higher, 21±18 ml/min/1.73 m². However, the precision, evaluated as standard deviation of bias indicated a huge variation in both cases. Conclusion: The re-expressed MDRD equation, using calibrated creatinine, is a more accurate estimation of GFR than.