[Influence of sampling techniques on blood sodium, potassium and hemoglobin concentration measured by a blood gas...

A study was carried out to assess the influence of sampling technique on sodium, potassium and haemoglobin blood concentrations measured simultaneously with a bedside blood gas analyser (Ciba Corning 288) and in the central laboratory. Blood samples from forty intensive care unit patients were used to test cight techniques: capillary tubes (Radiometer & Corning), preheparinized syringes (Terumo 2.5 ml, Sherwood 1 and 3 ml, Corning 1 ml) and Radiometer 2 ml) and a conventional syringe (Plastimed) which was rinsed with heparin just before use. Each result was compared with that obtained by the laboratory. For haemoglobin concentrations, the results were underestimated by the bedside analyser with Terumo, Corning, Radiometer and Plastimed syringes. The differences between results varied with the haemoglobin concentration, and precision was low: 38 to 55% of results showed a difference of more than 1 g.dl-1, and 13 to 25% a difference of more than 2 g.dl-1. The results obtained with the capillary tubes and the preheparinized syringes (Sherwood) showed differences independent of the value, as well as higher precision: 13 to 20% of values differed more than 1 g.dl-1, and 0 to 3% more than 2 g.dl-1. The capillary tubes moderately overestimated the haemoglobin concentration (+0.02 to +0.65 g.dl-1) while Sherwood's syringes underestimated it (-0.48 to -0.94 g.dl-1). For the sodium and potassium concentrations, the eight methods gave results similar to those obtained by the laboratory. It is concluded that these modules for bedside biological analysis, when used in optimal conditions, rapidly provide reliable results for some parameters.