Comparison of three methods of evaluation of acid-base disorders in pediatric patients requiring intensive care

Objetivo: comparar el rendimiento diagnóstico de los métodos tradicional, brecha aniónica corregida y de Stewart, para el diagnóstico de los trastornos ácido-base e hiperlactatemia en niños críticos. Método: se recolectaron datos clínicos y de laboratorio de niños al ingreso a la unidad de cuidado intensivo mediante gases, ionograma, albumina y lactato tomados simultaneamente. Se compararon los diagnósticos acido-base según los tres métodos. Resultados: 60 pacientes fueron analizados, 51 (85%) presentaron hipoalbuminemia, 18 (30%) presentaron hiperlactatemia (lactato 2 mEq/L) y 8 pacientes (13,3%) hiperlactatemia severa (lactato 5 mEq/L). 41(68,3%) tenían acidosis metabólica, 60% de ellos tenian acumulación de aniones no medibles según Stewart. Hubo 7 pacientes (11.7%) con acidosis por aniones no medibles con pH y HCO3 normal. La sensibilidad y especificidad para el diagnostico de hiperlactatemia fueron 68% y 40% método de Stewart, 89% y 40% Exceso de bases y 77 y 59% Brecha aniónica. Conclusiones: los métodos de brecha aniónica, exceso de bases y Stewart muestran regular desempeño para diagnosticar hierlactatemia. La frecuente alcalosis por hipoalbuminemia, enmascara trastornos acidificantes por aniones no medibles que deben ser buscados. / Abstract. Objective: to compare the diagnostic performance of the Stewart’s physicochemical method, the anion gap method and the traditional method for the diagnosis of acid-base disorders and lactic acidosis in critical ill children. Methods: We collected blood gases, complete electrolytes, albumin and serum lactate data from children on admission to the intensive care unit. Results: 60 patients were included in the trial, 51 (85%) had hypoalbuminemia, 18 (30%) had hyperlactatemia (lactate 2 mEq / L) and 8 patients (13.3%) severe hyperlactatemia (lactate 5 mEq / L), 41 (68.3%) had metabolic acidosis and 60% had an accumulation of unmeasured anions according to Stewart’s method. We identified seven patients (11.7%) with unmeasured anions acidosis with normal pH and HCO3. The sensitivity and specificity for the diagnosis of hyperlactatemia were 68% and 40% for Stewart’s method, 89% and 40% base excess and 77 and 59% for the anion gap method. Conclusions: for this population all the studied methods showed a regular performance in the diagnosis of lactic acidosis. Hypoalbuminemic alkalosis frequently masks acidifying disorders by unmeasured anions, and it should be sought.Otr