Can we really use the Stewart-Fencl method to analyze acid-base derangements in clinical practice?