The Impact of Hemolysis on the Accuracy of Ethanol Determinations

Recently, the accuracy of ethanol determinations by headspace (HS) gas chromatography (GC) has been questioned when performed on hemolyzed blood specimens (1, 2). While there are various studies concerning this subject, they have not yet been consolidated to provide an easy reference for review. The purpose of this letter is to provide a brief review of recent challenges on the accuracy of ethanol determinations of hemolyzed blood specimens and references which address them. The challenges generally focus on assumptions that the matrix of hemolyzed blood is different than whole blood, will affect an accurate ethanol quantitation and is no longer representative of the biological components within the blood circulating in the human body. One of the challenges focuses on the use of a small gauge needle when drawing blood specimens (1). Hemolysis is the process of the breakdown of red blood cells in blood. Hemolyzed blood specimens occur somewhat frequently in clinical laboratory practice and can affect the measurement of many clinical assays performed in clinical laboratories with clinical analyzers (3, 4) although most hemolyzed specimens do not exceed the clinically significant threshold of free hemoglobin of ≥0.3 g/L (3, 5). Hemolysis can be caused by a pathological condition in vivo or by some sample collection, transport and storage conditions in vitro (3, 4). Although small needle size (>21 gauge) is frequently mentioned as a potential cause of in vitro hemolysis, there is conflicting evidence in the literature on the actual biological impact of this parameter (6) and the use of butterfly needle systems with 21, 23 or 25 gauge needles is commonly described for venipuncture (7). Hematocrit is a clinical test that measures the percentage of the volume of whole blood that is composed of red blood cells (8). The scientific literature does not support the assertion that a small needle size will cause a change in the composition of the whole blood sample to be more consistent with that of a serum sample, i.e., have significantly fewer red blood cells (9, 10). Human red blood cells have a diameter of 6–8 μm and a thickness of 2 μm (3). The internal diameter of a 25-gauge needle is 260 μm (11). For comparison, it would be similar to asserting that ping pong balls (40 mm or 1.57 inches in diameter), if they were able to be deformed and shaped like a donut, would have difficulty going through a 51 inch (4.25 feet) diameter hula hoop which normally has a maximum diameter of 42 inches. What may be affected if the sample does happen to be hemolyzed is the analytical results of clinical tests for blood composition such as hematocrit (12) not the actual composition of the drawn sample of whole blood. Lippi et al. (12) drew whole blood with a 20-gauge needle, divided it into three separate portions and examined the effect of hemolysis on clinical parameters including hematocrit. The first portion was analyzed without modification and was verified to not be hemolyzed. The second and third portions were treated prior to analysis by passing the blood through a 30-gauge needle 5 times to cause moderate hemolysis and 10 times to cause gross hemolysis. The sample volume was identical. All of the same cellular material was present in all samples with exception of differing portions of the erythrocytes which had been lysed. As expected the measured hematocrit values were significantly lower after moderate and gross hemolysis of the samples even though the overall composition of the blood was identical. Clinical assays for ethanol utilizing the enzymatic method of analysis on serum/plasma have been reported to be affected by hemolysis resulting in a decrease in the measured ethanol concentration (13–15). Gadsden (13) demonstrated that the interference was spectrophotometric in nature (ultraviolet) and a highly hemolyzed sample would cause a decrease in the measured ethanol concentration due to an elevation of the instrument blank reading performed prior to the addition of reagents to the sample. Forensic assays for ethanol utilize the GC method, usually with HS sampling of the specimen (16–21) and are not affected by hemolyzed specimens. Matrix effects have been reported to be insignificant for HS–GC analysis when using at least a 5:1 dilution of the specimen (17, 21). Therefore, water-based calibrators have been employed to test urine and whole blood specimens whether the specimens are hemolyzed or not (16, 17, 19). Kristoffersen et al. (19) observed no significant difference in ethanol concentrations when analyzing hemolyzed and non-hemolyzed blood by HS–GC. The authors did report, however, that the decrease in ethanol concentration after room temperature storage was greater in hemolyzed specimens and theorized that the decrease was due to greater availability of oxyhemoglobin in hemolyzed blood for oxyhemoglobin-mediated oxidation of ethanol (22). When evaluating the ability to accurately determine ethanol concentration on clotted specimens, Senkowski and Thompson (23) reported that there was no significant difference between ethanol analysis on whole blood and clotted specimens that were first homogenized with a tissue grinder. The mechanical grinding in a tissue grinder used in that study would not only break up the clots, but would also most certainly lyse intact erythrocytes. In other studies, HS–GC has been used as the reference method to evaluate the effect of hemolysis on clinical enzymatic assays because it is unaffected by sample hemolysis (13, 24). Assertions that hemolysis affects the accuracy of the quantitation of ethanol in forensic blood specimens measured by HS–GC is not supported by the literature.