Unacceptable variability of hemoglobin estimation on samples obtained from ear punctures.

Sampling techniques were studied for the determination of the hemoglobin in volunteer blood donors. First, finger, ear, and venous samples were collected from 50 volunteers. Then five volunteers were monitored for a three-day period and finger, ear, and venous samples were collected daily. Thirdly, microhematocrits were done on the postdonation EDTA samples of 500 donors who had been screened using ear puncture copper sulfate hemoglobin estimations. Fourth, 25,437 donations were followed up to see whether capillary sampling techniques affected the rate of donor deferral. And finally, the ear sampling technique was varied using vigorous rubbing and manipulation to see if there was abnormal circulation that would affect the hematocrit results from capillary blood from the ear. We found that the microhematocrits averaged 7 per cent higher from blood obtained from ear puncture than from either finger puncture or venous puncture. Microhematocrits obtained from blood obtained by ear puncture were less accurate and less precise than blood from finger punctures. Finger puncture samples exclude approximately 6 per cent more donors than does blood from ear punctures.