Review of literature

Müller and Hundt3 in 1976 gave the first alert about the possible interference from using alcohol antiseptics for cleansing the venipuncture site. In their elegant study, the authors collected three sequential evacuated blood tubes from each of 10 healthy volunteers, for measuring ethanol concentrations with GC. The skin was cleansed with an antiseptic containing 10% chlorhexidine and 70% ethanol. Before alcohol evaporation had occurred, the needle was inserted into the vein and the first two blood tubes were completely filled, whereas the needle was withdrawn from the vein while the third blood tube was still aspirating. Importantly, the presence of ethanol on the skin did not produce a measurable concentration of blood alcohol in the first two tubes, whereas a measurable concentration of blood alcohol was found in the third tube in eight out of the ten subjects, with blood alcohol values between 0.4 - 743.9 mmol/L.

In a study by Goldfinger TM et al.4, they simultaneously obtained blood samples in emergency patients for ethanol content from both antecubital fossae, using an alcohol pad on one arm and a non-alcohol-containing germicidal solution on the other. Fifty patients with ethanol concentrations greater than zero were statistically analyzed. Twenty patients surveyed had no measurable alcohol level by either technique. There was no significant difference in the blood alcohol concentration obtained by either method of skin preparation in both groups (P<.01). Blood alcohol concentration incidentally obtained in the emergency department by routine isopropyl alcohol skin preparation is an accurate laboratory parameter.

Miller et al.5 reported a retrospective analysis of prospective data collected from a study of blood ethanol levels after the use of the alcohol-based hand sanitizer (ABHS). A total of 5 male volunteers were enrolled. Eight of the 10 total blood ethanol level measurements were drawn after skin preparation with Kendall WEBCOL® Alcohol Preps (APP) containing 70% isopropyl alcohol. All had an initial and post-ABHS application blood alcohol level (BAL) drawn, for a total of 10 BAL measurements. Measurements upon completion of the study were <5 mg/dL in all 5 study participants and in each of the 10 blood draws regardless of skin preparation technique. This study demonstrates that the use of isopropyl skin prep pads is unlikely to cause significant false-positive blood ethanol levels.

In a case reported by Yigit O et al.6, they encountered an obviously high blood alcohol level in a 20-year-old worker brought to Emergency Department after accidentally having his head crushed under a tree trunk. Head and maxillofacial computed tomography was ordered and blood samples were taken. When the results arrived, a very high blood alcohol level as 453 mg/dL was seen. The patient was questioned again for alcohol consumption; however, he denied having ingested any alcohol. The laboratory was questioned about whether there was any problem with the test measuring method and devices; the technicians denied any such problem. When the nurse who collected the blood sample was asked about swabbing the skin, it was learned that she used an alcohol swab first and then a povidone-iodine swab before blood sampling. A new sample was collected again after povidone-iodine swabbing and the blood alcohol level was measured as 0.3 mg/dL, which was within normal limits.

In a study reported by Tucker A et al.7, twenty eight Volunteers aged >18 years had paired venous blood tests, which were drawn within 2 min of each other. One arm was swabbed with a 70% isopropyl alcohol swab and allowed to dry before venepuncture. The other was swabbed with saline, and these concurrent samples were used as controls. BAL was tested using the enzymatic method. Pathologists analysing the samples were blinded to the swabbing technique used. The mean differences and standard deviations of each of the paired samples were analysed using Student's t‐test. Fifty‐six paired venous blood samples were obtained from volunteers. Mean BAL in the isopropyl alcohol‐swabbed group was 3.27 mg/dL with a standard deviation of 1.14 mg/dL. Mean BAL in the saline‐swabbed group was 3.41 mg/dL with a standard deviation of 1.11 mg/dL. The mean difference was 0.14 mg/dL, with a standard error of 0.157. There was no statistically significant difference between the groups. Hence concluded that use of 70% isopropyl alcohol swabs does not significantly affect BAL when used before venepuncture.

In a study done by Lippi G et al.8, venous blood was collected from 23 healthy volunteers, with two sequential procedures. In the first blood collection, 2 mL of alcoholic antiseptic (0.5% chlorhexidine, 70% ethanol) were place on a gauge pad, the venipuncture site of right arm was cleaned but the antiseptic was not let to dry before phlebotomy. In the second blood collection, 2 mL of the same alcoholic antiseptic were placed on another gauge pad, the venipuncture site of left harm was cleaned and the antiseptic was accurately cleansed before phlebotomy. Ethanol was measured with a reference GC technique in whole blood and EDTA plasma, and a commercial enzymatic assay in EDTA plasma. The concentration of alcohol in all EDTA plasma samples was always lower than the limit of detection of the enzymatic assay (i.e., 2.2 mmol/L; 0.1 g/L). Similarly, alcohol concentration was also undetectable using a reference GC technique (i.e., < 0.22 mmol/L; 0.01 g/L) in EDTA plasma and whole blood. Hence conclusion was drawn that using ethanol-containing antiseptics before venipuncture may not be causes of spurious or false positive results of alcohol measurement at least when ideal venipunctures are performed.

McIvor et al.9 conducted a study on 20 volunteers, in whom venepuncture through the skin soaked in either ethyl or isopropyl alcohol was performed. A non-alcoholic cetrimide/chlorhexidine swab was used as a control. All subjects were initially alcohol free. Ethyl alcohol was detected in only one blood sample and the level was found to be only 0.4 mg% (roughly the limit of detection of the assay). A slightly higher level of isopropyl alcohol (3 mg%) was found in one of the blood samples. Alcohol estimation was carried out by head-space gas chromatography using a pair of instruments each fitted with a column exhibiting different retention characteristics. The system was similar to that used by other UK Forensic Science Laboratories where accurate, definitive results are a necessity. It was concluded that under very testing conditions only minute ethanol interference is produced by using alcohol-based skin cleansing swabs. This minimal interference is unlikely to affect clinical sample results, and even in a forensic situation the inadvertent use of alcohol-based swabs is unlikely to lead to a miscarriage of justice.

In a study by Ryder KW et al.10, the effect of various pre-packaged skin cleansers on the results of serum ethanol analyses performed with the Du Pont automatic clinical analyzer was reported. When added directly to serum in concentrations of either 0.625 or 1.56% (v/v), neither polyvinyl pyrrolidone iodine nor benzalkonium chloride affected the ethanol results. The cross-reactivity of isopropanol with the automatic clinical analyser ethanol procedure was 3.9%. The greatest interference in the measured ethanol concentration was from the addition of green soap tincture, which contained 30% ethanol. The effect of improper phlebotomy technique on ethanol measurements was also investigated by performing venepunctures through a pool of 100% ethanol on the skin. No ethanol was detected in these samples unless an ethanol-soaked sponge was pressed over the venepuncture site while the needle was withdrawn from the skin. It was thus concluded that when correct phlebotomy technique is used, skin cleansing agents should not affect the results of ethanol measurements determined with the Du Pont automatic clinical analyser.

Peek et al.11 reported opposite results. The authors collected blood from 10 healthy volunteers during heavy drinking by either cleansing the venepuncture site with absolute ethanol on one arm or leaving the venepuncture site un-swabbed on the opposite arm. Blood ethanol concentration was found to be significantly increased (up to 3.9 mmol/L) in samples collected from the ethanol-cleansed arm compared to the un-swabbed arm. Unlike these findings, the concentration of blood ethanol was unaltered when ethanol was replaced with isopropanol.

McIvor and Cosbey12 measured BAC by means of head-space gas chromatography in 20 subjects, whose blood had been drawn after soaking the venepuncture site with ethanol, isopropyl alcohol or non-alcoholic cetrimide/chlorhexidine swab, and concluded that a modest ethanol interference could be appreciated by using alcohol-based skin cleansing swabs.

Malingré et al.13 studied the possible impact of cleansing the venepuncture site with an alcohol swab, by measuring blood ethanol concentration using both gas chromatography and an enzymatic assay in 50 patients aged ≤ 12 years and in 20 healthy volunteers. In no sample the concentration of blood ethanol was found to be measurable (i.e., < 4.3 mmol/L in all cases).

Higuchi et al.14 in his study collected blood from 40 healthy subjects after cleansing the venepuncture site with either ethanol or saline. Although no blood uptake of ethanol or physiological saline could be recorded by the reference gas chromatography technique when both fluids were allowed to dry for 1 min, a minimal alcohol intake (i.e., around 0.2 mmol/L) was however noticed in 40% percent of subjects when the venepuncture was performed immediately after cleaning the skin with ethanol (i.e., within 5 seconds). The rate of contamination considerably increased to 70% when the needle was voluntarily touched or swabbed by an ethanol-soaked cotton pad.

In a study conducted by Tatiya HS15, ten healthy adult volunteers were selected randomly for blood collection. Samples collected were preserved in commercially available grey topped bulbs with sodium fluoride as preservative. Samples were processed on the same day in a NABL accredited lab where lab technicians were kept blind about study. Irrespective of alcohol content of swab applied for cleaning, insignificant values of serum alcohol concentration as <10mg/dl i.e. lower than the limit of detection of enzymatic assay were detected, concluding that there is no significant effect of alcohol swabbing prior to venpuncture site on serum alcohol concentration.