The measurement of plasma sodium and potassium concentrations is needed in critical care settings when it must be measured quickly and accurately. The purpose of this study was to compare the results obtained in dogs with a validated laboratory system (Vitros 250) and with the ECS 2000, a recently commercialized analyser. Seventy-one canine Li-heparin plasmas were analyzed with the Vitros 250 and the ECS 2000. The comparison between plasma and whole blood was performed with the ECS 2000 on 36 samples. The correlation was excellent for potassium (KECS = 0,90 × KVitros + 0,29; r = 0,96) and good for sodium (NaECS = 1,19 NaVitros - 24,38; r = 0,79). Between-series imprecision (n = 22) was 12% and 1,6% for sodium and potassium respectively. The ECS 2000 sodium results were significantly higher and the bias was proportional, ranging from 0,6 to 4,9 mmol/L. Potassium results were moderately lower with the ECS 2000 and the bias was proportional, ranging from 0,11 to 0,44 mmol/L. The clinical discrimination "normal" vs. "high" or "low" according to the Vitros 250 reference intervals, showed that 94% and 90% of the results felt into the same categories for sodium and potassium respectively. The comparison of the results obtained in whole blood and plasma using the ECS 2000 showed an excellent agreement; the difference between the two series was not significant for potassium, but sodium concentrations were significantly lower than those obtained from plasma. Centrifuging the sample does not improve the analytic quality. Consequently, the ECS 2000, quick and easy to use, gives very good results and is a good alternative to more expensive analysers for veterinary clinics.