EVALUATION OF COMPLETE BLOOD-COUNT RESULTS FROM A NEW, ON-SITE HEMOCYTOMETER COMPARED WITH A LABORATORY-BASED HEMOCYTOMETER

Objective: To compare point of care results obtained from an on site h emocytometer with values provided by an institutional laboratory instr ument. Design: A prospective laboratory evaluation. Setting: The centr al laboratory and cardiac surgical intensive care unit of a university -affiliated tertiary care center. Patients: Normal range comparison wa s performed using blood specimens routinely obtained from 48 hospitali zed patients for complete blood count analysis. The second evaluation was performed on blood specimens routinely obtained (in the intensive care unit) after cardiac surgery involving extracorporeal circulation in a series of 187 consecutive patients. Measurements and Main Results : Hemoglobin concentration, platelet count, mean corpuscular volume, m ean platelet volume, and red and white blood cell counts were measured with both onsite (MD 16, Coulter Electronics, Hialeah, FL) and labora tory (STKS, Coulter Electronics) instruments. Hematocrit and red cell distribution width were calculated using measured variables. Blood spe cimens were obtained from two distinct patients series. To evaluate me asurement values within the normal range, a series of 48 routinely obt ained blood specimens for complete blood count analysis sis in our ins titutional laboratory were utilized for concurrent analysis with the o n-site hemocytometer. To evaluate measure ment values out of the norma l range, a second comparison involved measurements performed on blood specimens obtained in the cardiac surgical intensive care unit for com plete blood count analysis. Linear regression demonstrated good correl ations between on-site and laboratory hemoglobin concentration (r(2) = .97), hematocrit (r(2) = .95), platelet count (r(2) = .97), mean corp uscular volume (r(2) = .91), red cell distribution width (r(2) = .80), and red (r(2) = .95) and white (r(2) = .96) blood cell count results. A marginal correlation was observed between mean platelet volume valu es (r(2) = .47). Bias analysis (mean +/- 2 so) demonstrated similar me asurements between on-site and laboratory hemoglobin concentration, he matocrit, platelet count, red blood cell count, white blood cell count , mean platelet volume, mean corpuscular volume, and red cell distribu tion width. Conclusions: On site hemoglobin concentration, hematocrit, white blood cell count, red blood cell count, red cell distribution w idth, and platelet count values compare well with those results obtain ed from the laboratory. The Mn 16 hemocytometer (Coulter Electronics) provides on-site hematologic results that can provide an accurate and rapid quantitative assessment of platelets, and red and white blood ce lls. Rapid access to information obtained from this type of system may be clinically useful, especially in critically ill patients.