Develop a procedure that would allow the estimation of drug concentrations in forensic bloodstains of unknown volume.

Blood is the primary material for quantitative toxicological analysis and interpretation of the results. However, in some cases it may be collected after a long time after the incident that is of interest to investigative bodies. This time interval may result in the elimination or a significant decrease in the concentrations of possibly present drugs, which may have a key influence on the interpretation. An example of such events may be hit-and-run cases that relate to causing a car accident and fleeing the scene. In such accidents, injuries are common and there is a chance that bloodstains will remain in the car or on the driver's clothes. Therefore bloodstains can become the best or sometimes the only material to be used to prove driving under the influence of drugs.

Quantitative analysis of forensic bloodstains samples is impossible without determining the volume of blood from which the stain was formed. The ability to quantify the blood volume is therefore necessary. In this work, endogenous amino acids: valine and leucine were used to determine the volume of blood. In order to isolate the analytes from bloodstains on different surfaces (plastic and cotton) the elution mixture consisted of an acetonitrile:water (60:40) and sodium chloride (0.9%) was used. Analyses were carried out using LC-MS/MS technique. The developed protocol was tested on 32 authentic forensic samples (mostly cases of people driving under the influence of amphetamine). Different volume bloodstains were prepared for each forensic case.

The results of bloodstains analyses were compared with the results of whole-blood analyses. The accuracy was in the range from −87.7 to +471.0%, however, for most of the cases more accurate results were obtained. The mean and median agreement between the concentration estimated from bloodstain and concentration determined in whole-blood were 48.8% and 13.4% for plastic, and 37.7% and 33.9% for cotton, respectively. The variables concentrations estimated from bloodstains and concentrations calculated from whole-blood were found to be strongly correlated. The Pearson correlation coefficients were: for plastic r(29)=0.85 and for cotton r(25)=0.91 (the P-value for both<0.00001). The best correlation coefficients were observed for the smallest bloodstains. The slope of the line for all results was 1.0784 (R2=0.7588). The relative standard deviation between concentrations estimated using valine and leucine were in the range 0.5–28.3% (n=40).

In the hit-and-run incident, the suspected person may be arrested after such a long time that toxicological analysis of possibly disclosed bloodstains may be the only option to prove driving under the influence of drugs. In the procedure presented here, the authors used endogenous blood components to determine its volume. Critical factor in the calculations is the estimation of the volume of the blood on the basis of amino acids, as the content of these compounds may differ due to individual variation. Taking into account this possible diversity, significant correlations between concentrations in authentic whole-blood and bloodstains were observed.

This is the first work in which quantitative analysis of drugs in bloodstains have been attempted. A special protocol was developed for this purpose. The research indicates that estimation of the concentration of drugs in bloodstains from the blood of an unknown volume is possible. The best accuracy was obtained by analysing 3 stains of different sizes and calculating the average. The protocol has been extensively tested on real forensic samples and the obtained results were successfully compared with drug concentrations determined in whole-blood.