Exploring the applicability of equine blood to bloodstain pattern analysis

Abstract

Bloodstain pattern analysis (BPA) is the forensic application of the interpretation of distinct patterns which blood exhibits during a bloodletting incident, providing key evidence with its ability to map the sequence of events. Here, we explore the use of equine blood and investigate its suitability within the field of BPA. Blood is a complex fluid, and finding a suitable safe substitute to human blood that encompasses all of its characteristics has been the focus of many investigations. Animal blood has been concluded as the closest and therefore the most suitable alternate. However, it seems that currently only porcine blood is most prominently utilised.

In this study, equine blood was investigated, using two different anti-clotting methods, where blood impacts were explored over a typical range of varying impact velocities upon a selection of commonly encountered surfaces. Key BPA parameters, such as the diameters of the resulting bloodstains, number of spines and area of origin were measured, which were subsequently applied into previously derived BPA equations.

We find that defibrinated equine blood is a suitable alternative and offers the same conclusive outcomes to human blood. This gives bloodstain pattern investigators and researchers an additional choice of blood which can be of benefit when certain bloods are difficult to attain or when the activity involves the usage of a large quantity of blood. Additionally we explore the effect on BPA of aged blood, which revealed a significant decrease in stain diameter of up to 12.78 % when blood has been left for 57 days. A shelf life of no more than 12 days is recommended when blood is refrigerated at 4℃.

Keywords

Equine blood bloodstain pattern analysis forensic science aged blood

Introduction

Bloodstain pattern analysis (BPA) is the study and interpretation of bloodstains found at crime scenes.¹ Observations made from these stains can ascertain particular information such as the sequence of events and movement of the victim and/or assailant during a bloodletting incident.^1,2 The importance of BPA has been widely recognised in the legal and forensic divisions.¹ Recent studies have involved a more quantitative approach to what is considered a very subjective field, establishing new ways of predicting the angle of impact,³ impact velocity⁴ and point of origin⁵ of an attack. It has been observed within these and other studies⁶ that porcine (pig) is the preferred choice of blood, showing reputable applicability to human blood. Safety aspects have to be considered when utilising human blood. The use of animal blood has been acknowledged as a suitable substitute to human blood, decreasing but not eliminating the risks of coming into contact with pathogens and diseases such as HIV, hepatitis and so on.⁶ Despite the dependable reputation of porcine blood, other animal bloods are still employed. A study carried out by Christman et al.⁷ compared various animal bloods (swine, bovine, equine and porcine) to determine suitability as a human blood substitute, where impacts and the general appearance of the blood were compared.⁷ Although this study gives an insight into the functionality of animal blood as a human blood substitute, it uses now outdated terminology (low impact velocity, etc.) and suffers from a lack of any real statistical analysis which is considered obligatory within the forensic field. It is therefore the purpose of this current study to determine the use and applicability of equine blood, a more commercially available and animal friendly blood,⁸ in the interpretation of bloodstain patterns with the use of these newer quantitative BPA approaches.^3–5

Pioneering work from Hulse-Smith et al.⁴ reported, using porcine blood and utilising several already well-defined equations, that using the bloodstain diameter and number of spines of a bloodstain, the impact velocity and blood-drop size can be determined.⁴ This paper will use these equations and directly compare results for equine blood obtained with those reported by Hulse-Smith et al.⁴

The creation of a bloodstain is dependent on a variety of parameters and constants: gravity, velocity, droplet diameter, objects used (weapon), impacted surface and blood characteristics (such as surface tension, density and viscosity). Some of these parameters (viscosity, surface tension, droplet diameter and density) can be accounted for using the Reynolds and Weber numbers, as explored throughout the work of Hulse-Smith et al.⁴ The Reynolds number (Re) is a dimensionless ratio relating to the ratio of fluid inertia to viscous forces, essentially quantifying these two forces for known flow conditions.⁴ The Reynolds number (Re) is expressed by:

Re = \frac{ρ D_{o} V_{o}}{μ}

(1)

where D_o is the drop diameter, V_o is the impact velocity, μ is the blood viscosity and ρ is the blood density. Equivalently, the Weber number (We) expresses the ratio of inertia to surface tension forces, where σ is surface tension:

We = \frac{ρ {DoV}_{0}^{2}}{σ}

(2)

The Reynolds number can be utilised further to find the maximum drop spread diameter to drop diameter ratio, as described by equation (3), where D_max corresponds to the maximum drop spread diameter.⁴ Equation (3) suggests that when inertia from the drop impact is high enough, surface tension can be neglected. Weber and Reynolds numbers equating to We >> Re^0.5 are necessary to satisfy the above statement.⁴

\frac{D_{\max}}{D_{0}} = \frac{{Re}^{0.25}}{2}

(3)

It has been widely recognised that D_s, the final stain diameter, is equivalent to D_max due to the unremarkable difference experienced when the drop rebounds.⁴ However, it is noted that this may alter when we consider different surfaces (i.e. plastic) where a larger wetting angle may be exhibited. An experimental correction value was added to equation (3) to account for experimental discrepancies encountered between measured and calculated values:

\frac{D_{s}}{D_{0}} = C_{d} \frac{{Re}^{0.25}}{2}

(4)

where the experimental constant, C_d equates to 1.11.⁴ Similarly, the Weber number can be further utilised in the calculation of number of spines N, expresses in equation (5):

N = 1.14 \sqrt{We}

(5)

When a blood drop impacts a surface, the drop expands outwards and creates deviations on the periphery of the stain, most commonly referred to as spines. Spines are defined as any rise and fall beyond an otherwise smooth rim.⁴ This includes waves, triangles, lines or other protrusions.⁴ The calculation of spines has proven to be somewhat subjective, explaining why another experimental constant, C_n, was introduced into equation (5)⁴ to account for these discrepancies; where C_n has been reported to be a value of 0.838:

N = 1.14 C_{n} \sqrt{We}

(6)

With the use of equine and human blood and drop tests, and through the applications of the above equations (1)–(6), comparisons of the observed diameter, number of spines and Weber and Reynolds numbers were calculated and contrasted with one another, and previous published values obtained for porcine blood.⁴

Consequently, in this paper, we explored the use of equine blood and determined its suitability and potential use in BPA. Two samples of equine blood were used, each with different anti-clotting methods employed. The first was defibrinated, where the fibrin (an essential component in clotting) is removed. Second, we used an anticoagulant, Alsever’s, which is considered to be the most stable and commonly used anticoagulant.⁹ Last, we turned our attention to the consideration of the age of blood and its potential effect upon BPA experimentation. It is well documented that the age of porcine blood alters bloodstains, decreasing the diameter of the bloodstain as the blood ages,⁶ and therefore considerations pertaining to shelf life have to be made when conducting experiments or crime-scene reconstruction to account for these changes. Presently, it is unknown if this is the case for equine blood, which we duly explore, since any such changes would need to be accounted for in future BPA activities to avoid possible misinterpretations of patterns.

Experimental

Resources

All equine blood (defibrinated and Alsever’s) was obtained from TCS-Biosciences Ltd. at a packed cell volume (PCV %) of 45%. Human blood (ethically approved as governed by ethics at Manchester Metropolitan University) was freshly drawn on the day of the experiment using a venepuncture. Blood was drawn into purple blood tubes containing the anticoagulant EDTA. Surfaces were new: paper (80 gsm; standard A4), plastic sheet (Medium Density Polyethylene), tile (ceramic gloss) and cold-rolled steel.

Method

Blood-drop tests for all bloods were dispensed manually using two different sized pipettes (1 mm and 1.77 mm, inner tip diameter), from varying heights (30.5, 60.9, 91.4 and 121.9 cm) and onto different surfaces (paper, plastic, tile and steel). A minimum amount of pressure was applied to the pipette each time, and the drops were performed by the same analyst to maintain continuity. The surface roughness of each experimental surface (Table 1) was deduced using a TESA-rugosurf 10 at five different points. The roughness is known to affect the observed number of spines⁴ and gives greater applicability to real-life scenarios where surfaces are not all the same. Blood drops were filmed in slow motion at 1200 frames per second (fps) using the Casio Ex-F1 digital camera such that a drop diameter (D_o) could be readily established.⁴ A diverse range of heights was used to create varying impact velocities (V_o) as expressed in Table 2. Velocities were calculated using the equation:

V_{o} = \sqrt{2 gh}

(7)

where g is the acceleration of gravity and h is the release height of the blood which is measured from the tip of the pipette to the impact surface. Here, it was assumed that the path of the blood drop can be defined by the equations for the movement of a rigid object, therefore disregarding parameters such as drag.³ These particular heights were chosen, as they can be directly compared to the results obtained by Hulse-Smith⁴ who also used animal blood (porcine) as a human blood substitute. It is noted that there is a possibility of oscillation during the initial phase of the falling drop. However, studies have shown that this oscillation is less dramatic than that produced in a raindrop and is therefore likely not to have a marked effect.¹⁰ It has also been demonstrated that the drop oscillation will not be a significant factor, as oscillation does not occur after a 40 cm release height, which is below most of the heights used in this experiment.¹¹

Table 1.

Average surface roughness of each experimental surface.

Surfaces	Roughness (µm) (N = 5)
Paper	2.62 ± 0.20
Plastic	0.72 ± 0.15
Tile	0.07 ± 0.01
Cold-rolled steel	0.54 ± 0.20

Table 2.

Release heights of blood drops calculated from the tip of pipette to the impacting surface and converted into impact velocity with the use of equation (7).

Release heights (cm)	Impact velocity (m/s)
30.5	2.45
60.9	3.46
91.4	4.24
121.9	4.89

Temperature is an important factor requiring careful consideration and control, since it affects the viscosity of the blood, which can lead to stain distortion (i.e. stain elongation, etc.).¹² Therefore, all blood was maintained at room temperature during all experimentation, although it is noted that body temperature is approaching 37℃ and could potentially give differing real-life results. It was found that mimicking body temperature would be problematic to establish and keep at a constant, especially when air exposure and temperature of the equipment are considered. Viscosity measurements for each blood (defibrinated, etc.) were performed using a Kinexus Pro Rheometer where blood was tested at room temperature.

The remaining blood (300 mL) was refrigerated at 4℃ in the original 500 mL container which was sealed with Para film for storage in order to conduct the aging experiment. The blood was agitated throughout the experiment using a magnetic stirrer. The stirrer was set low (15 rpm) to prevent cell lysis.

The room in which the experiments were conducted was temperature controlled (at 22℃). The temperature was monitored using a steel ibutton temperature data logger.

Age experiment

The same experimental procedure was adhered to for the aged equine blood, though D_o was not deemed necessary to calculate, as diameter and number of spines were the focal interests of the aged blood. Drops were again manually dispensed from heights of 30.5, 60.9, 91.4 and 121.9 cm onto a paper surface (80 mg, A4 standard). Drops were repeated five times, and stain diameters were measured using a magnifying loupe. Viscosity measurements were performed using a Kinexus Pro Rheometer.

Results and discussion

Blood-type properties

The physical properties of blood are of high importance, since they determine the resultant pattern of blood and are utilised in many of the aforementioned equations (see Introduction). Taking this into consideration, the physical properties of equine, porcine and human blood are compared in Table 3 which show no significant difference between the physical properties of equine blood^13,14 and those of porcine and human blood,⁴ suggesting the potential use of equine blood in BPA.

Table 3.

A comparison of published values obtained for the physical properties of equine, porcine and human blood (all unadulterated).

	Equine blood^12,14	Porcine blood⁴	Human blood⁴
Density (kg/m³)	1050	1062	1052–1063
Viscosity (mPa.s)	3.3–5.6	3.4–6.1	3.8–5.1
Surface tension (×10⁻²N/m)	5.1	5.3–5.8	5.1–5.7
PCV (%)	32–46	38.9–46.3	40.0–45.0

Drop diameters

Slow-motion filming, as detailed in the experimental section, was performed, enabling the capture of scaled still images of blood drops as they depart the pipette tip to allow the drop diameters, D_o, to be deduced as depicted in Figure 1. Drops were discharged as close to the scale as possible to avoid any size distortion. Such drop diameters result in average drop diameter values of 4 mm and 5 mm when defibrinated blood was utilised, 3.5 mm and 4 mm when Alsever’s blood was dropped and 4 mm and 4.5 mm when human blood was utilised (1 mm and 1.77 mm inner tip diameter pipettes respectively). These results are in excellent agreement with previous blood-drop diameters reported by Hulse-Smith et al.⁴ and Willis et al.¹⁵ where human¹⁵ and porcine blood were used,⁴ showing no difference in size (between 3.0 mm and 4.3 mm) or general shape of the droplet exhibited.

Figure 1.

Blood drops captured using the Casio Ex-F1 digital camera slow motion filming at 1200 fps. Scaled stills were produced and measured. (a) A still of defibrinated equine blood drop using a 1 mL pipette (1 mm inner tip diameter). (b) Defibrinated equine blood drop using a 1 mL pipette (1.77 mm inner tip diameter).

Blood type

First, we considered the size of bloodstains (D_s) produced when all blood types were released from heights of 30.5, 60.9, 91.4 and 121.9 cm. These particular heights were chosen so that direct comparisons could be made later with those values obtained for porcine blood,⁴ which have also been implemented at these height values. Blood was dropped onto paper with a 1 mL pipette (1.77 mm inner tip diameter). Stains were left to dry completely and were then measured using a magnifying loupe. Figure 2 explores the magnitude of the bloodstain diameter (D_s) as a function of release height, revealing that the anti-coagulated (Alsever’s) equine blood has a marked effect upon the size of the bloodstain produced. Bloodstains are evidently larger in diameter with the use of Alsever’s equine blood. This is an unexpected outcome as we have previously discussed, this form of anticoagulant is the most stable and commonly used.⁹ When investigations into the physical properties of the blood types were conducted, we found that the Alsever’s equine blood has a much lower viscosity of 2.5 mPa.s compared to that of defibrinated equine blood and human blood, which gave viscosity values of 4.7 mPa.s and 3.7 mPa.s respectively. Dynamic viscosity measurements were carried out on the blood using a Kinexus Pro Rheometer where viscosity measurements were obtained at room temperature (25^°C). Viscosity versus shear rate peaks were formulated, since viscosity is affected by shear rate. However, at shear rates of 100 s⁻¹ and above (which should be experienced throughout this study), the peak peters out to a constant. This constant was therefore taken as the viscosity of the blood. The diverse range of viscosity values are likely to be the explanation behind the much larger bloodstain diameters exhibited for the Alsever’s equine blood. As viscosity decreases, fluid – in this case blood – flows more freely and therefore travels further, resulting in a larger bloodstain diameter. Unpaired t-tests were performed in order to establish the statistical significance of the results. Each equine blood was statistically compared to human blood, taking the largest stain sizes exhibited, where the largest difference was observed. It was found that Alsever’s equine blood is statistically significant to human blood with a p-value of .0001, signifying that the use of Alsever’s does have a significant effect on the bloodstain diameter displayed and therefore is not an appropriate human blood substitute. Conversely, when defibrinated equine blood was statistically compared to human blood, the results were found not to be statistically different from human blood with a p-value of .1411. This implies that defibrinated equine blood is the more viable human blood alternative, as it exhibits bloodstains that are of a similar size and, importantly, not statistically significantly different from that of human blood.

Figure 2.

Comparisons of blood stain diameters (D_s) for defibrinated equine blood (squares, ▪), human blood (triangles, ▴) and anti-coagulated equine blood (circles,•) dropped upon a paper surface, identifying that defibrinated equine blood gives the greatest comparability to human blood; N = 5.

Effects of surface roughness

Next, we considered blood impacting upon different experimental surfaces, as a variety of surfaces should provide us with various surface roughnesses (texture of a surface). Figure 3(a) and (b) demonstrates the observed average bloodstain diameters (D_s) obtained when defibrinated equine blood was dropped onto paper, plastic, tile and cold-rolled steel. When comparing surface roughness, in accordance with the final stain diameters (D_s) ascertained during the drop tests, it is apparent that there is an effect. However, due to the inconsistency of the tested surfaces – that is, the surface roughness is heterogeneous in nature – there are anomalies within the results. This is demonstrated in Figure 3(a) where it appears that the smaller stains were produced on the cold-rolled steel instead of the paper where the surface was much rougher. It is nevertheless evident from the observed bloodstains that there is a decrease in stain diameter in accordance with surface roughness, such that the lowest roughness value gives larger bloodstain diameters (D_s). This could further be related to the original drop diameter, where the 5 mm droplets (D_o) produced significantly larger bloodstains than those of the 4 mm droplet. Similar results were exhibited for both Alsever’s equine blood (Figure 4(a) and (b)) and human blood (Figure 5(a) and (b)), where bloodstains decreased when the surface roughness increased. This is further demonstrated by performing t-tests where all p-values for the three bloods were found to be statistically significant (p < .0015). When we compare all three bloods, it is evident that Alsever’s equine blood on the whole generates much larger stains than that of human and defibrinated equine blood. This is consistent with the results reported earlier (see Blood Type section).

Figure 3.

(a) Bloodstain diameters for defibrinated equine blood on different surfaces – paper (upside-down triangles, ▾), plastic (triangles, ▴), tile (circles, •) and cold-rolled steel (squares, ▪) – using a 1 mL pipette (inner tip diameter 1 mm), depicting the rougher the surface the smaller the bloodstains and vice versa; N = 5. (b) Bloodstain diameters for defibrinated equine blood on different surfaces – paper (upside-down triangles,▾), plastic (triangles, ▴), tile (circles,•) and cold-rolled steel (squares, ▪) – using a 1 mL pipette (inner tip diameter 1.77 mm). Similarly, the smaller tipped pipette bloodstains appear larger as surface roughness decreases; N = 5.

Figure 4.

(a) Bloodstain diameters (D_s) for Alsever’s equine blood released upon different surfaces from a range of release heights – paper (squares, ▪), plastic (upside-down triangles,▾), tile (circles, •) and cold-rolled steel (triangles, ▴) – using a 1 mL pipette (inner tip diameter 1 mm). Direct comparisons with defibrinated equine blood reveal significant changes in bloodstain diameters (D_s); N = 5. (b) Bloodstain diameters for Alsever’s equine blood on different surfaces – paper (squares, ▪), plastic (triangles, ▴), tile (circles, •) and cold-rolled steel (upside-down triangles,▾) – using a 1 mL pipette (inner tip diameter 1.77 mm). Similarly, the smaller tipped pipette bloodstains appear to be larger in size than ones experienced using defibrinated equine blood; N = 5.

Figure 5.

(a) Bloodstain diameters (D_s) for human blood released upon different surfaces from a range of release heights – paper (squares, ▪), plastic (triangles, ▴), tile (circles, •) and cold-rolled steel (upside-down triangles,▾) – using a 1 mL pipette (inner tip diameter 1 mm). Direct comparisons with both equine bloods reveal defibrinated equine blood to be the best match; N = 5. (b) Bloodstain diameters for human blood on different surfaces – paper (squares, ▪), plastic (triangles,▴), tile (circles, •) and cold-rolled steel (upside-down triangles,▾) – using a 1 mL pipette (inner tip diameter 1.77 mm). Again, the smaller tipped pipette bloodstains further demonstrated the applicability of defibrinated equine blood as a human blood substitute; N = 5.

Further investigations were performed with the use of equations (3)–(6). Figures 6–8 depict the ratio of drop spread (D_s/D_o) in relation to the Reynolds number, where lines of ‘best fit’ where undertaken. Equations (3) and (4) are pre-existing equations and were found not to satisfy the data obtained completely with any of the three bloods tested. Therefore, new constants were established that gave a ‘best fit’ to the scatter data presented. New constants (C_d) were found to equal 0.9, 1.09 and 0.88 for defibrinated equine blood, Alsever’s equine blood and human blood respectively. Comparing the three constants, it is even clearer that Alsever’s equine blood is inconsistent with human blood, and it would therefore be advised that defibrinated blood is used when conducting BPA reconstructions or general experimentation. Similarly, when the number of spines was considered, for each blood, in relation to the Weber number, new constants were again developed which ‘best fit’ the given data more effectively than the original equations (5) and (6). This found the new constants (C_n) to be 0.46, 0.72 and 0.45 for defibrinated equine blood, Alsever’s equine blood and human blood, shown in Figures 9–11. All calculations were determined using the physical properties expressed in Table 3 and the viscosity values determined earlier for each blood type. Clearly, the range of scatter observed gives a large error on C_n, due to the number of spines being greatly affected by the impacting surface, and the tile surface is clearly pulling down the C_n value. However, note the deviation from that predicted previously by Hulse-Smith⁴ suggesting that counting the number of spines is subjective.

Figure 6.

A new line of ‘best fit’(solid line) was established when considering the spread factor versus the Reynolds number utilising defibrinated equine blood on different surfaces: paper, plastic tile and cold-rolled steel. This is comparing to the original line of ‘best fit’ (dotted line) using equation (3)⁴ and the line of ‘best fit’ (dashed line) developed by Hulse-Smith et al.⁴ using equation (4); N = 5.

Figure 7.

A representation of the effects of Alsever’s treated equine blood dropped on different surfaces – paper, plastic, tile and cold-rolled steel – and the effects this has on the plot of spread factor versus Reynolds number, subsequently leading to an alteration of the constant value, established using a new line of ‘best fit’ (solid line). This more accurately fit the scatter compared with the original line of ‘best fit’ (dotted line) found using equation (3)⁴ and the line of ‘best fit’ (dashed line) developed by Hulse-Smith et al.⁴ using equation (4); N = 5.

Figure 8.

A new line of ‘best fit’ (solid line) was established when considering the spread factor versus the Reynolds number utilising human blood on different surfaces: paper, plastic tile and cold-rolled steel. This was compared to the original line of ‘best fit’ (dotted line) using equation (3)⁴ and the line of best fit (dashed line) developed by Hulse-Smith et al.⁴ using equation (4); N = 5.

Figure 9.

The number of spines, N, as a function of the Weber number exhibited when utilising defibrinated equine blood on different surfaces – paper, plastic, tile and cold-rolled steel – versus the Weber number. The number of spines is highly influenced by the surface roughness, consequently leading to a new constant being developed, with the use of a line of ‘best fit’. The new line of ‘best fit’ (solid line) fitted the scatter spread more accurately compared to the original line of ‘best fit’ (dotted line) using equation (5)⁴ and the line of ‘best fit’ (dashed line) incorporated by Hulse Smith et al.⁴ using equation (6); N = 5.

Figure 10.

A new constant was developed when the number of spines, N, as a function of the Weber number was considered after the utilisation of Alsever’s equine blood on different surfaces – paper, plastic, tile and cold-rolled steel – versus the Weber number. The new line of ‘best fit’ (solid line) fitted the scatter spread more accurately compared to the original line of ‘best fit’ (dotted line) using equation (5)⁴ and the line of ‘best fit’ (dashed line) incorporated by Hulse Smith et al.⁴ using equation (6); N = 5.

Figure 11.

Human blood was tested to determine number of spines, N, as a function of the Weber number when it was deposited onto different surfaces – paper, plastic, tile and cold-rolled steel – as surface roughness is highly influential in number of spines. A new constant was developed, with the use of a line of ‘best fit’. The new line of ‘best fit’ (solid line) fitted the scatter spread more accurately compared to the original line of ‘best fit’ (dotted line) using equation (5)⁴ and the line of best fit (dashed line) incorporated by Hulse Smith et al.⁴ using equation (6); N = 5.

We can therefore conclude that a decrease in drop diameter and an increase in surface roughness ultimately produce smaller bloodstains. This is concurrent with previous research in which porcine blood was tested.⁴ We can also conclude that defibrinated equine blood is overall more suitable as a human blood substitute, giving consistently comparable results throughout experimentation and on any surface type. Results obtained for number of spines observed when using all bloods are also in excellent agreement with previous research,⁴ showing that the rougher the surface, the greater preponderance of spines. It is noted that there is a lower overall number of spines acquired within this study compared to that of Hulse-Smith.⁴ However, this may be attributed to the physical property deviations of the tested bloods or the subjective technique of counting spines.

Area of origin

The application of the area of origin is a vital piece of information which pinpoints the area in which a bloodletting incident took place. The area of origin provides the analyst with a 3D perspective, helping them to ‘picture’ the incident and therefore giving greater insight into the series if events.¹⁶ Here, we used the tangent method to calculate the area of origin, where bloodstains and the area of convergence were measured manually. We compared only human blood and defibrinated equine blood for this section of analysis, as the Alsever’s equine blood proved to provide less comparable results to that of human blood throughout the previous experiments and further experimentation was deemed frivolous. To create the area of origin, a series of impacts were generated using a Proctor Little Nipper Rat Trap®, where 1–1.5 mL of blood was loaded onto the edge of the trap. The rat trap was set onto a large sheet of paper at ground level and at 12 cm height to create differing area of origins. When the rat-trap device snapped, the impact spatter collected onto the paper (lining) where it could be measured and visually compared. Figure 12(a) and (b) show the impacts of human and defibrinated equine blood respectively. Visually, both blood impacts are very similar and are hard to distinguish from one another. When the area of origin was calculated for all impacts, it was found that both could accurately calculate the area of origin and that only a small variance between the two existed of around ±2 cm, which is relatively insignificant in terms of determining an area of origin.

Figure 12.

Blood impacts used to calculate the area of origin for both (a) human blood and (b) defibrinated equine blood.

New versus aged blood

Finally, we explored the effects of age on equine blood. New equine blood and aged equine blood (over the time periods of 12 days, 14 days and 57 days) were all compared, defining the differences by alteration in diameter size and number of spines. Figure 13 illustrates the differences experienced. Diameters were found to decrease significantly after 14 days by up to 6.38% and decrease further after 57 days by up to 12.78%. A Student’s t-test was performed to determine the significance of these results. A p-value of .0003 was calculated using the values for 57 days and fresh blood at a height of 200 cm, and a p-value of .0020 when 14 days was compared to fresh blood at the same height. These results are considered to be statistically significant, and therefore it can be stated that age of blood has a significant effect on the bloodstain diameter. The aging process of blood increased the viscosity to 5 mPA.s and 5.3 mPa.s for 14 and 57 days respectively, ultimately resulting in smaller stains.⁶ However, it was found that the PCV% was unaffected.¹⁷ It is unknown why this affect occurred, but is thought to be accounted for by the effects of aging in red blood cells (RBCs).¹⁷ We can discount plasma viscosity as being the factor, as previous studies¹⁷ have found that plasma viscosity increases due to fibrinogen production, and this was removed in our blood. Age was not found to have any effect on the number of spines observed. Results concur with previous reports, where aged porcine blood was investigated and resulted in a decrease in diameter size.⁶ Consideration should be made when undertaking future experiments to the time period in which equine blood can be utilised, as it is clear that age has a significant effect on the bloodstain patterns produced.

Figure 13.

Aging defibrinated equine blood released (using a 1 mL pipette; 1.77 mm inner tip diameter) from a range of heights (30.5, 60.9, 91.4 and 121.9 cm). It is clear that a decrease in the bloodstain diameter is observed as the blood gets older. The age of the equine blood ranged from 57 days old (stars), 14 days old (circles), 12 days old (diamonds) to new blood (square); N = 5.

Discussion

In this paper, we have explored the application of equine blood upon BPA, demonstrating the effects of its usage on the interpretation of bloodstain patterns, with the use of blood spatter experiments and the employment of predefined equations. It has been established that defibrinated equine blood can be used as a BPA human blood substitute.

Equine blood follows the same trend as both the human and porcine blood⁶ in that when height increases, so does the stain diameter. When utilising the pre-existing equations, the need for new constants to ‘best fit’ the given data was recognised. New values for the Reynolds numbers are smaller than those obtained previously for porcine blood but do not deviate so much from the original equation as to be questionable. When the human blood constant and defibrinated equine blood constant were compared, they deviated little with one another. Conversely, Alsever’s equine blood was found to be inconsistent with human blood and therefore not a viable substitute. Constant values for the Weber number suggest a much lower progression of number of spines than values found for porcine blood. This is accounted for by the subjective nature of spine counting.

Aged, defibrinated equine blood presented large changes in diameter size as the blood got older, with deviations amounting to 12.78% when blood was 57 days old. This is thought to be related to the change in the viscosity of the blood over time and the effect of the aging process upon the RBCs. Similar deviations are experienced when porcine blood is utilised.⁶

Overall, it can be concluded that defibrinated equine blood is a reliable substitute for human blood. Any deviations experienced from the previously reported results obtained from the utilisation of porcine blood are due to inconsistent physical properties used for the comparisons or the subjective nature of BPA. Further analysis using a greater variation in physical properties and impacted surfaces needs to be undertaken. This will provide a greater understanding of the true usefulness of these more quantitative approaches. It would also be useful to conduct experiments with fresh blood that does not contain anticoagulants, as this may lead to more significant results.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

References

Wonder AY. Bloodstain pattern evidence: objective approaches and case applications. 1st ed. California: Academic Press, 2007.

Peschel

Kunz

Rothschild

. Blood stain pattern analysis. Forensic Sci Med Pathol 2011; 7: 257–257.

Knock

Davison

. Predicting the position of the source of blood stains for angled impacts. J Forensic Sci 2007; 52: 1044–1044.

Hulse-Smith

Mehdizadeh

Chandra

. Deducing drop size and impact velocity from circular bloodstains. J Forensic Sci 2005; 50: 54–54.

de Bruin

Stoel

Limborgh

JCM

. Improving the point of origin determination in bloodstain pattern analysis. J Forensic Sci 2011; 56: 1476–1476.

Raymond

Smith

Liesegang

. The physical properties of blood – forensic considerations. Sci Justice 1996; 36: 153–153.

Christman DV. A study to compare and contrast animal blood to human blood product, Virginia Dep. Forensic Sci., Bloodstain Pattern Analysts Training manual, Section 4.2.3.1, Oct 1997.

www.tcsbiosciences.co.uk/animal_blood_products.php (accessed June 2012).

www.pmlmicro.com/assets/TDS/133.pdf (accessed June 2012).

10.

Bevel

Gardner

. Bloodstain pattern analysis: with an introduction to crime scene reconstruction, Boca Raton, FL: CRC Press, 1997.

11.

Raymond

Smith

Liesegang

. Oscillating blood droplets – implications for crime scene reconstruction. Sci Justice 1996; 36: 161–161.

12.

Brownson

DAC

Banks

. Crime scene investigation: the effect of drug contaminated bloodstains on bloodstain pattern analysis. Anal Methods 2010; 2: 1885–1885.

13.

Hamazaki T, and Shishido H. Thromb Res 1983; 30: 587.

14.

Palenske

Saunders

. Comparisons of blood viscosity between amphibians and mammals at 3 C and 38 C. J Thermal Bio 2002; 27: 479–479.

15.

Willis

Piranian

Donaggio

. Errors in the estimation of the distance of fall and angles of impact blood drops. Forensic Sci Int 2001; 123: 1–1.

16.

James SH, Kish PE, and Sutton TP. Principles of bloodstain pattern analysis: theory and practice. Baton Rouge, LA: CRC Press, Taylor and Francis Group, 2005.

17.

Carallo

Irace

De Franceschi

. The effect of aging on blood and plasma viscosity. An 11.6 years follow-up study. Clin Hemorheol Microcirc 2011; 47: 67–67.