- Email: [email protected]
Biventricular measurement of blood strontium in real cases of drowning
Forensic Science
Forensic Science International 69 (1994) 139-148
ELSEVIER
lnternihnal
Biventricular measurement of blood strontium in real cases of drowning J. Azparren* a, I. de la Rosab, M. Sanchoa aInstituto National de Toxicologia de Madrid. C/ Luis Cabrera, 9, 28002 Madrid, Spain b Veterinary University, Zaragoza. Spain
Received 17 May 1994; accepted 22 June 1994
Abstract
A simple method for blood strontium determination by spetrophotometry atomic absorption with Zeeman correction is proposed. This method is applied to real cases of death by drowning, where the difference in blood strontium concentration between the left and the right side of the heart could contribute, together with other data, to the investigation of causes of death in cases of drowning in salt water. The differences obtained in blood strontium concentrations between the left and the right side of the heart in cases of supposed ‘typical drowning’ in sea water were always > 75 pg SrA, compared to < 20 rg Sr/l found in two supposed cases of ‘atypical drowning’. Keywork
Drowning; Strontium; Hemodilution
1. Introduction Usually, in cases of drowning, the death mechanism is an irreversible cerebral anoxia, as a consequence of aspiration of water into the air passages [ 11.These cases are known as ‘typical drowning’ or ‘wet drowning’. When they occur in fresh water, a large amount of water passes from the alveoli into the pulmonary capillaries and a hemodilution takes place in the left heart. On the contrary, in sea water, not only does water leave the arterial blood and a hemoconcentration occur, but also ions of * Corresponding author. 0379-0738/94/$07.00 0 1994 Eisevier Science Ireland Ltd. All rights reserved SSDI 0379-0738(94)01581-O
140
J. A:parren
et al. /Forensic
Sci. ht.
69 (1994)
139-148
strontium from sea water appear to enter into the arterial blood [2,3]. This event produces a marked difference in the strontium concentration between arterial and venous blood. In certain cases of submersion, death occurs without or with only a small passage of strontium from water into the blood, for example, in cases of ‘dry drowning’ where death presumably is caused by fatal cerebral hypoxia due to a laryngeal spasm, in cases of immersion syndrome (vagal inhibition) and in cases of postmortem immersion of the victim. In these cases, the strontium concentration in the blood of the left and right ventricles should be of a similar amount. Therefore, the measurement of one of the substances whose concentration in blood is affected by the penetration of water or ions into the lung should be a means to investigate the type of death. To this end, different analyses have been carried out: blood chlorine [4,5]; blood iron [6]; blood magnesium [7]; atria1 natriuretic peptide (ANP) (81; diatoms [9,10] and strontium in serum [11,12], which was proposed a long time ago [13], because of the large difference between the concentration of strontium in human blood and that of sea water (16-43 &l and 8000 &l, respectively). Strontium is a common component in the marine salts and is found in oceans and seas, at concentrations of - 8000 &l [ 14,151, with variations depending on salinity. However, in cases of drowning, no matter what this variation would be, the difference of the strontium concentration between blood and sea water would always be big enough to allow strontium in blood to be considered as an indicator. The strontium concentration in rivers varies enormously. In some rivers, this concentration is found to be similar to those found in blood, but in other rivers, the concentration could be more than ten times higher. Therefore, in cases of drowning in rivers, it is very important to analyze samples of water taken where the corpse was found. 2. Material In the present work, we have analyzed samples of blood from the left and the right side of the heart, submitted by different pathologist forensics taken from 28 cases of submersion that occurred since the end of 1991. In most of the cases, we have also analyzed samples of water taken near the place where the corpse was found. A total of seven cases were excluded for different reasons: cases with attempts of resucitation, including intracardial injection, cases of ‘wet drowning’ that occurred in rivers where samples of water were not available, and cases whose blood samples had a hemoglobin concentration of c 50 g/l. We have also analyzed 20 different blood samples without anticoagulants from corpses where the cause of death was known to be other than drowning, in order to determine the ‘normal’ level of strontium. 3. Analytical method 3.1. Apparatus A Perkin-Elmer model Z-3030 atomic absorption spectrophotometer equipped with Zeeman effect, standard Perkin-Elmer hollow cathode strontium lamp and AS-
J. Azparren
et al. /Forensic
Sci. ht.
69 (1994)
139-148
141
40 autosampler were used. Eppendorf automatic pipettes were employed for all preparation. A Vibromixer (Heidolph, REAX 2000) was used for mixing the solutions. 3.2. Reagents and materials These were as follows: -
distilled de-ionized water was employed for the preparation of all solutions; Triton X-100 (Merck); 0.5% (v/v) solution in deionized water; nitric acid (suprapure, Merck); and plastic tips for Eppendorf pipettes were rinsed with ultrapure water immediately before use.
3.3. Analytical procedure A 50 ~1 quantity of the well-mixed whole blood was diluted 20-fold with 950 ~1
of the Triton X-100 solution in the autosampler cup. A 10 ~1 quantity of blood solution or standard solution of various concentrations made in the same way was introduced into the graphite furnace, using the autosampler AS-40, and duplicate integrated absorbance measurements were made. The average calculated slope of a linear regression line was 16.0 x 10m3A.s. per ~g/l for the diluted blood and 17.1 x 10m3A.s. per j&l for the aqueous standard (1% (v/v) HN03) with r2 values calculated to be ~0.98. The similarity of these two slopes allows the determination of both diluted blood and aqueous material from the same aqueous calibration curve. See Tables 1 and 2. 3.4. Evaluations (a) Sensitivity. The sensitivity, defined as the mass (pg) of Sr required to produce
an average absorbance area of 0.0044 A.s., was 2.6 pg.
Table 1 Zeeman atomic
absorption
spectroscopy.
instrumental
parameters
Parameter
Value or comment
Current lamp Wavelength
20-25 mA 420.1 nm
L slit Signal mode Integrate time Inert gas
0.7 nm Peak area 2.5 s Argon Pyro-coated graphite IO PI On Spiked in diluted blood or aqueous (I% (v/v) HNO,)
Furnace type Sample volume Zeeman background Mode of calibration
correction
material
142
J. A:parren
Table 2 Temperature
et al. /Forensic
Sri. Int. 69 (1994)
139-148
program
Step
Dry
Char
Atomise
Burn out
Temperature (“C) Ramp time (s)
110 IO
1400 20
2400 0
2700
Hold time (s) Flow (mUmin)
30 300
30 300
5 0
I 5 300
(b) Detection limit. The detection limit calculated for blood and defined as three times the standard deviation of a matrix blank was 4.6 pg Sr/l blood, for n = 22. (cj Linear range. The linearity observed was up to at least 260 pg Sr/l blood. Linearity was evaluated by extending the calibration curve to -0.220 A.s. (d) Precision and accuracy. To evaluate the precision of the method, a series of blood samples containing strontium was analyzed (see Table 3). The blood used as diluent had been previously determined by the proposed method. No blood materials are commercially available with certified Sr concentration values. For this reason we have not evaluated the accuracy of the results.
4. Results As we can see in Tables 4-6, we have considered the value of 8000 &l as the ‘normal’ strontium concentration in water in cases of drowning in sea water, where samples of water were not available, when using the expression 1000 (LV-RV)/W - ratio of (left ventricle minus right ventricle) to water. In cases that occurred at the rivermouth, the strontium concentration in water could be quite variable depending on the tide type. For this reason, there could be an error in the expression 1000 (LV-
Table 3 Precision Additions
Blood concentration (ag
Mean measured”
WI blood)
(SD) (Irg Sri1 blood)
80 BgSrll blood 120 pg Sr/l blood 200 pg Sr/l blood 300 pg Sr/l blood
100.01 140.01 220.01 320.01
20.01 95.97 138.18 217.26 313.54
% = 6. bMeasured
addition.
-
by standard
(1.97)b (4.59) (4.71) (8.19) (5.17)
Recovery”
(%a)
(86-l 14) (86-99.7) (94.0- 102.0) (93.9-103.3) (95.6-99.7)
RSD” (‘%I)
9.85 4.78 3.41 3.77 1.65
4
7660 7120
522 * 35.4”
320.7 * 4.6”’
97.6 +z 8.47”
151.6 f 8.9”’
~~
348.3 * 12.6”’
880.5 * 5.7”
152.2 f 3.2”’
172.7 * 6.0”’
65 f 3.0”’
9
IO
II
“Number of replicates
101.4 * 5.9”’
97.1 + 4.8”’
I4
16.7”’
328.3 f
303.6 + 13.6’4’
15.7’+
I3
140.6 f
14.0”’
62.2 f
161.9 f
I2
l.314’
6300
199.4 * 12.1’“’
~
~
~
5800
~
6940
112.1 * 3.7’6’
l4+’
8
381.4 t
10.1’4’
??
145.6
5950
7
2475.6 f 57”’
1555 zt 12.6’*’
6
8000
4
239.3 + 37.3”’
21.9 f 4.214’
3
3465
273.6 * I IS@’
44.4 f 6.7’j’
2
~
110.8 + 0.3”’
Left ventricle (LV) concentration (pg Sr/l) f SD”
15.8 f 0.7”’
tration (rg WI) zt SD”
Right ventricle (RV) concen-
I
CkSe
Salt water
Table
0.5
3.1
9.8
16.7
88.5
31
34.0
154.7
23.7
55.4
27.2
66.1
Il.9
x loo0
<24
20
2
36-48
12-24
24-48
3-4
12-14
26
24-48
6
8.5
I6 f 3
<30
(h)
had been in water
20-6
29-S
23-12
26-10
29-9
6-10
s-7
25-3
27-10
27-8
6-5
l-3
7-1
7-12
death (day and month)
29
22
66
36
79
63
43
63
I7
25
42
37
43
41
Rwer’s mouth
Sea
Sea
Sea
Rwer’s mouth SCd
River’s mouth
River’s mouth River’s mouth
Sea
Sea
Sea
River’s mouth
SW
Increased
Normal
Increased
Lightly increased
Increased
Increased
Increased
Increased
Increased
Very increased
Increased
Yes
~~
~
-~--
Yes
Yes
~
-
Yes
Yes
Yes
edema
Small amount of frothy dark liquid Abundant frothy liquid
Small amount of frothy liquid Large amount of frothy liquid with bubbles Abundant frothy liquid Abundant frothy liquid Abundant frothy liquid Lightly frothy liquid
Abundant frothy liquid Large amount of colorless liquid Dark liquid with abundant bubbles Abundant frothy liquid Dark liquid with abundant bubbles
Yes
No
Yes
Yes
Yes
FOWa
Yes
-
Yes
Yes
Yes
~
Yes
in the surface of the lungs
I.
Large amount of clear liquid NO
300 cc
7s cc
l-1.1
NO
Small amount of dark liquid NO
II.
Large amount of clear liquid
Large
and
Yes
Yes
Abundant
Yes
mouth nose
Finethefroth at
s
5
5
SD”
%/I)
Left
5.5”’
64.4 *
I9
hEslimdted
“Number
2.7”’
23.3 *
I8
time between
of replicates.
(pg
+ 3.8 “’
?? 1.7”’
zt SD”
death
109.6
20.4
*
1.2’”
7.4”’
and autopsy.
f
116.7 + 4.7”’
20.5
I7
12.5”
36.4 ?? 1.9”
*
34.1
41.8
+ 2.2”’
20.8
%/I)
concen-
ventricle
tration
(LV)
I6
*
(pg
tration
ventricle
concen-
Right
(> 50 gg G/l)
(RV)
water
15
Case
Brackish
Table
580
92
575
342
800
71.9
<0
153.6
15.8
16.6
x
1000
48
S-12
4Ob
3
<24
and
(h)
30-9
15-3
2-11
31-7
l-l
month)
(day
death
in water
had been
55
24
75
66
29
River
River
place
Watering
Pool
River
Increased
~
creased
Lightly
Acute
~
Very
in-
Yes
icreased
edema
amount
liquid
~
liquid
amount of frothy
Large
of liquid
Large
frothy
Abundant
Yes
~
500 ml
> 500 ml
> 500 ml
Yes
“OX
the lungs
mouth of
in the surface
and
5
5
??
z
%
5
?
; % 3 2.
* %
5
J. Azpurren et al. /Forensic
Sci. ht.
69 (1994)
139-148
145
146
J. A:parren et al. /Forensic
Sri. ht. 69 (1994)
139-148
RV)/W, since the water samples were not taken at the same moment as death took place. In order to determine the reference levels of strontium in the blood of corpses, we have analyzed 20 different blood samples without anticoagulants taken from corpses where the cause of death was other than drowning. We have found an average concentration of 27.97 pg/l and a range of 14.64-47.46 &I. ‘A few hours’ interval between the death and the collection of the sample could mean a great variation in the concentration of the strontium in water in those two different moments, so that the sample would have no value at all. 5. Discussion 5.1. Cases that occurred in sea water or at the river mouth
According to the autopsy findings, only in case 13 was a ‘dry lung’ found, and in cases 14 and 7 the appearance of the lung was considered doubtful. In our study, as we can see in Table 4, there are two cases, 13 and 14, whose values for 1000 (LVRV)/W (3.1 and 0.5) were quite different from those obtained in the rest of the cases (>9). Cases 13 and 14 seem to show no entrance of strontium from sea water to the arterial blood during the time the victims were alive, which could indicate cases of laryngeal spasm with no water introduction into the lung; cases of vagal inhibition where death took place immediately after a cardiac arrest; or cases of postmortem immersion of the victim. In case 7, according to autopsy findings, the lungs have not the typical appearances found in ‘wet drowning’ cases. The forensic pathologist suggested that the cause of death could be a cardiac arrest. In our laboratory we found a diatom in the femur, together with a high alcohol concentration in blood and vitreous humor - 4.6 and 3.5 g/l, respectively - as well as marked differences in the concentrations of chlorides in blood at the left and the right side of the heart - 154 and 128 mmol Cl/l, respectively - and of hemoglobin - 100 and 69 g/l, respectively. Therefore, it seemed that the victim had swallowed a certain quantity of water before dying. The high concentrations of strontium found in venous blood in cases 13 and 14 - 3 12 and 97 &l, respectively - were above those considered as ‘reference-values’ (14.6-47.5 &l) and could be explained by a postmortem diffusion that increased the strontium concentration in blood of both ventricles by the same amount. The differences found in the strontium concentration in blood from the left and the right sides of the heart in the rest of the cases, which are supposed to be cases of ‘wet drowning’, is in the range 75-900 &l. This variation could be partially explained by the different amount of strontium that had been passing from sea water to the arterial blood during the time the victim remained alive with water in the lung. 5.2. Cases that occurred in water with >_50 pg S-/l
At this time, we are considering those cases of drowning that occurred in rivers or natural pools, with strontium concentration greater than the ‘reference values’ of strontium in blood. The behaviour of strontium concentration in arterial blood in
J. Azparren et al. /Forensic Sci. hr. 69 (1994) 139-148
141
these cases is difficult to predict; on the one hand it is expected that strontium passes from the water to the arterial blood, but on the other hand, a hemodilution in the arterial blood could be produced due to the possible hypotonic condition of this water. According to the autopsy findings from five cases, four of them were considered as ‘typical drowning’ cases. In the other case (19), the forensic pathologist thought it could be a death caused by vagal inhibition. In our laboratory, only in two cases (17 and 19) did we find an appreciable difference in strontium concentration between the blood of the left and the right chambers. In the other three cases, this difference was found to be < 15 &l. In case 19, we found similar values in the concentration in blood of the left and the right ventricles in hemoglobin - 173 and 172 g/l - and chloride - 98 and 110 rnmol/l. This was in agreement with the cause of death proposed by the forensic pathologist. The concentration of strontium here was in contradiction with the rest of the data, and perhaps could be affected by factors such as contamination of strontium in the container or by the place where the blood sample was taken from. A homogenization of the blood in the ventricles before the sampling should be necessary, to avoid a concentration of strontium towards the lower levels of the heart, as was observed to have occurred in the blood sample containers received. 5.3. Cases that occurred in fresh water (cSOt.~g Sr/l) No appreciable differences were found in the concentration of strontium in blood from the left and the right side of the heart, due to the similar concentration of strontium found in water and blood. Case number 24 was a case of death due to inhalation of CO in the bathtub, confirmed by the high concentration of carboxyhemoglobin found in blood. Case number 23 was a ‘typical drowning’ case confirmed by the elevated number of diatoms found in several organs and blood. 6. Conclusion The combined measurement of blood strontium in two ventricles seems to help in the diagnosis of drowning in salt water. In fact, the differences in the concentration of strontium in two ventricles, obtained in 12 cases of supposed ‘typical drowning’ that occurred in salt water, were 175 pgSr/l, against ~20 PgSrll obtained in two cases where death could have been produced by causes that did not imply a transfer of strontium from water to blood, according to anatomopathological observations during the autopsy. Where drowning occurred in rivers with strontium concentration in the range 90-800 &l, only in certain cases could the strontium determination be useful in the diagnosis of death by drowning. Acknowledgements We wish to thank Lourdes Saez for proof reading the English in this manuscript.
J. Azparren et al. /Forensic
148
Sri. Int. 69 (1994) 139-148
References H.G. Swann and N.R. Spafford, Body salt and water changes during fresh and sea water drowning. Texas Rep. Biol. Med., 9 (1951) 356-382. PI H.G. Swann, M. Bruer, C. Moore and B.L. Vezien, Fresh water and sea water drowning a study of the terminal cardiac and biochemical events. Texas Rpts. Biol. Med., 5 (1947) 423-437. 131 J. Peam, Pathophysiology of drowning. Med. J. Australia, 142 (1985) 586-588. [41 O.A. Gettler, A method for determination of death by drowning. J. Am. Med. Assoc., 66 (1921)
Ill
1650.
PI H.D. Stanley, CF. Henry and E.S. Henry, Blood changes in man following death due to drowning. Arch. Pathol., 56 (1963) 454-461.
161 G. Canepa, Fer hematique et submersion. An. Med. Leg., 1 (1963) 27-33.
171 A.R. Moritz, Chemical methods for the determination of death by drowning. Physiol. Rev., 24 (1944) 70-88.
PI J.A. Llorente, E. Villanueva, C. Hemindez-Cueto
and J.D. Luna, Plasmatic Levels of atria1 natriuretic peptide (ANP) in drowning. A pilot study. Forensic Sci. Int., 44 (1990) 69-75. 191 N. Foged, Diatoms and drowning - once more. Forensic Sci. Int., 21 (1983) 153-l 59. 1101 A.J. Peabody, Diatoms and drowning. A review. Med. Sci. Law, 20(4) (1980) 254-261. [Ill M. Piette, J. Timperman and N. Parisis, Serum strontium as a medico-legal diagnostic indicator of drowning. Med. Sci. Law., 29(2) (1989) 162-171 WI M.A. Amin, A.H. Samia, M.A. Kabil et al. Serum strontium estimation as a diagnostic criterion of the type of drowning water. Forensic Sci. Int., 28 (1985) 47-52. [I31 S. Icard, La prense de la mort par submersion suivant qu’elle a en lieu dans une riviere ou dans la mer. Rev. Pathol. comparee d’Hygiene gen., 32 (1932) 559-571. 1141 J.P. Riley and G. Skirrow, Chemical Oceanography, 2nd. edn., Vol. 2, Academic Press, New York, 1975, pp. 52-60. 1151 E.D. Goldberg, In the sea. Vol. 2. In M.N. Hill (ed.), Wiley Interscience, New York, 1963, pp. 3-25.
Forensic Science International 69 (1994) 139-148
ELSEVIER
lnternihnal
Biventricular measurement of blood strontium in real cases of drowning J. Azparren* a, I. de la Rosab, M. Sanchoa aInstituto National de Toxicologia de Madrid. C/ Luis Cabrera, 9, 28002 Madrid, Spain b Veterinary University, Zaragoza. Spain
Received 17 May 1994; accepted 22 June 1994
Abstract
A simple method for blood strontium determination by spetrophotometry atomic absorption with Zeeman correction is proposed. This method is applied to real cases of death by drowning, where the difference in blood strontium concentration between the left and the right side of the heart could contribute, together with other data, to the investigation of causes of death in cases of drowning in salt water. The differences obtained in blood strontium concentrations between the left and the right side of the heart in cases of supposed ‘typical drowning’ in sea water were always > 75 pg SrA, compared to < 20 rg Sr/l found in two supposed cases of ‘atypical drowning’. Keywork
Drowning; Strontium; Hemodilution
1. Introduction Usually, in cases of drowning, the death mechanism is an irreversible cerebral anoxia, as a consequence of aspiration of water into the air passages [ 11.These cases are known as ‘typical drowning’ or ‘wet drowning’. When they occur in fresh water, a large amount of water passes from the alveoli into the pulmonary capillaries and a hemodilution takes place in the left heart. On the contrary, in sea water, not only does water leave the arterial blood and a hemoconcentration occur, but also ions of * Corresponding author. 0379-0738/94/$07.00 0 1994 Eisevier Science Ireland Ltd. All rights reserved SSDI 0379-0738(94)01581-O
140
J. A:parren
et al. /Forensic
Sci. ht.
69 (1994)
139-148
strontium from sea water appear to enter into the arterial blood [2,3]. This event produces a marked difference in the strontium concentration between arterial and venous blood. In certain cases of submersion, death occurs without or with only a small passage of strontium from water into the blood, for example, in cases of ‘dry drowning’ where death presumably is caused by fatal cerebral hypoxia due to a laryngeal spasm, in cases of immersion syndrome (vagal inhibition) and in cases of postmortem immersion of the victim. In these cases, the strontium concentration in the blood of the left and right ventricles should be of a similar amount. Therefore, the measurement of one of the substances whose concentration in blood is affected by the penetration of water or ions into the lung should be a means to investigate the type of death. To this end, different analyses have been carried out: blood chlorine [4,5]; blood iron [6]; blood magnesium [7]; atria1 natriuretic peptide (ANP) (81; diatoms [9,10] and strontium in serum [11,12], which was proposed a long time ago [13], because of the large difference between the concentration of strontium in human blood and that of sea water (16-43 &l and 8000 &l, respectively). Strontium is a common component in the marine salts and is found in oceans and seas, at concentrations of - 8000 &l [ 14,151, with variations depending on salinity. However, in cases of drowning, no matter what this variation would be, the difference of the strontium concentration between blood and sea water would always be big enough to allow strontium in blood to be considered as an indicator. The strontium concentration in rivers varies enormously. In some rivers, this concentration is found to be similar to those found in blood, but in other rivers, the concentration could be more than ten times higher. Therefore, in cases of drowning in rivers, it is very important to analyze samples of water taken where the corpse was found. 2. Material In the present work, we have analyzed samples of blood from the left and the right side of the heart, submitted by different pathologist forensics taken from 28 cases of submersion that occurred since the end of 1991. In most of the cases, we have also analyzed samples of water taken near the place where the corpse was found. A total of seven cases were excluded for different reasons: cases with attempts of resucitation, including intracardial injection, cases of ‘wet drowning’ that occurred in rivers where samples of water were not available, and cases whose blood samples had a hemoglobin concentration of c 50 g/l. We have also analyzed 20 different blood samples without anticoagulants from corpses where the cause of death was known to be other than drowning, in order to determine the ‘normal’ level of strontium. 3. Analytical method 3.1. Apparatus A Perkin-Elmer model Z-3030 atomic absorption spectrophotometer equipped with Zeeman effect, standard Perkin-Elmer hollow cathode strontium lamp and AS-
J. Azparren
et al. /Forensic
Sci. ht.
69 (1994)
139-148
141
40 autosampler were used. Eppendorf automatic pipettes were employed for all preparation. A Vibromixer (Heidolph, REAX 2000) was used for mixing the solutions. 3.2. Reagents and materials These were as follows: -
distilled de-ionized water was employed for the preparation of all solutions; Triton X-100 (Merck); 0.5% (v/v) solution in deionized water; nitric acid (suprapure, Merck); and plastic tips for Eppendorf pipettes were rinsed with ultrapure water immediately before use.
3.3. Analytical procedure A 50 ~1 quantity of the well-mixed whole blood was diluted 20-fold with 950 ~1
of the Triton X-100 solution in the autosampler cup. A 10 ~1 quantity of blood solution or standard solution of various concentrations made in the same way was introduced into the graphite furnace, using the autosampler AS-40, and duplicate integrated absorbance measurements were made. The average calculated slope of a linear regression line was 16.0 x 10m3A.s. per ~g/l for the diluted blood and 17.1 x 10m3A.s. per j&l for the aqueous standard (1% (v/v) HN03) with r2 values calculated to be ~0.98. The similarity of these two slopes allows the determination of both diluted blood and aqueous material from the same aqueous calibration curve. See Tables 1 and 2. 3.4. Evaluations (a) Sensitivity. The sensitivity, defined as the mass (pg) of Sr required to produce
an average absorbance area of 0.0044 A.s., was 2.6 pg.
Table 1 Zeeman atomic
absorption
spectroscopy.
instrumental
parameters
Parameter
Value or comment
Current lamp Wavelength
20-25 mA 420.1 nm
L slit Signal mode Integrate time Inert gas
0.7 nm Peak area 2.5 s Argon Pyro-coated graphite IO PI On Spiked in diluted blood or aqueous (I% (v/v) HNO,)
Furnace type Sample volume Zeeman background Mode of calibration
correction
material
142
J. A:parren
Table 2 Temperature
et al. /Forensic
Sri. Int. 69 (1994)
139-148
program
Step
Dry
Char
Atomise
Burn out
Temperature (“C) Ramp time (s)
110 IO
1400 20
2400 0
2700
Hold time (s) Flow (mUmin)
30 300
30 300
5 0
I 5 300
(b) Detection limit. The detection limit calculated for blood and defined as three times the standard deviation of a matrix blank was 4.6 pg Sr/l blood, for n = 22. (cj Linear range. The linearity observed was up to at least 260 pg Sr/l blood. Linearity was evaluated by extending the calibration curve to -0.220 A.s. (d) Precision and accuracy. To evaluate the precision of the method, a series of blood samples containing strontium was analyzed (see Table 3). The blood used as diluent had been previously determined by the proposed method. No blood materials are commercially available with certified Sr concentration values. For this reason we have not evaluated the accuracy of the results.
4. Results As we can see in Tables 4-6, we have considered the value of 8000 &l as the ‘normal’ strontium concentration in water in cases of drowning in sea water, where samples of water were not available, when using the expression 1000 (LV-RV)/W - ratio of (left ventricle minus right ventricle) to water. In cases that occurred at the rivermouth, the strontium concentration in water could be quite variable depending on the tide type. For this reason, there could be an error in the expression 1000 (LV-
Table 3 Precision Additions
Blood concentration (ag
Mean measured”
WI blood)
(SD) (Irg Sri1 blood)
80 BgSrll blood 120 pg Sr/l blood 200 pg Sr/l blood 300 pg Sr/l blood
100.01 140.01 220.01 320.01
20.01 95.97 138.18 217.26 313.54
% = 6. bMeasured
addition.
-
by standard
(1.97)b (4.59) (4.71) (8.19) (5.17)
Recovery”
(%a)
(86-l 14) (86-99.7) (94.0- 102.0) (93.9-103.3) (95.6-99.7)
RSD” (‘%I)
9.85 4.78 3.41 3.77 1.65
4
7660 7120
522 * 35.4”
320.7 * 4.6”’
97.6 +z 8.47”
151.6 f 8.9”’
~~
348.3 * 12.6”’
880.5 * 5.7”
152.2 f 3.2”’
172.7 * 6.0”’
65 f 3.0”’
9
IO
II
“Number of replicates
101.4 * 5.9”’
97.1 + 4.8”’
I4
16.7”’
328.3 f
303.6 + 13.6’4’
15.7’+
I3
140.6 f
14.0”’
62.2 f
161.9 f
I2
l.314’
6300
199.4 * 12.1’“’
~
~
~
5800
~
6940
112.1 * 3.7’6’
l4+’
8
381.4 t
10.1’4’
??
145.6
5950
7
2475.6 f 57”’
1555 zt 12.6’*’
6
8000
4
239.3 + 37.3”’
21.9 f 4.214’
3
3465
273.6 * I IS@’
44.4 f 6.7’j’
2
~
110.8 + 0.3”’
Left ventricle (LV) concentration (pg Sr/l) f SD”
15.8 f 0.7”’
tration (rg WI) zt SD”
Right ventricle (RV) concen-
I
CkSe
Salt water
Table
0.5
3.1
9.8
16.7
88.5
31
34.0
154.7
23.7
55.4
27.2
66.1
Il.9
x loo0
<24
20
2
36-48
12-24
24-48
3-4
12-14
26
24-48
6
8.5
I6 f 3
<30
(h)
had been in water
20-6
29-S
23-12
26-10
29-9
6-10
s-7
25-3
27-10
27-8
6-5
l-3
7-1
7-12
death (day and month)
29
22
66
36
79
63
43
63
I7
25
42
37
43
41
Rwer’s mouth
Sea
Sea
Sea
Rwer’s mouth SCd
River’s mouth
River’s mouth River’s mouth
Sea
Sea
Sea
River’s mouth
SW
Increased
Normal
Increased
Lightly increased
Increased
Increased
Increased
Increased
Increased
Very increased
Increased
Yes
~~
~
-~--
Yes
Yes
~
-
Yes
Yes
Yes
edema
Small amount of frothy dark liquid Abundant frothy liquid
Small amount of frothy liquid Large amount of frothy liquid with bubbles Abundant frothy liquid Abundant frothy liquid Abundant frothy liquid Lightly frothy liquid
Abundant frothy liquid Large amount of colorless liquid Dark liquid with abundant bubbles Abundant frothy liquid Dark liquid with abundant bubbles
Yes
No
Yes
Yes
Yes
FOWa
Yes
-
Yes
Yes
Yes
~
Yes
in the surface of the lungs
I.
Large amount of clear liquid NO
300 cc
7s cc
l-1.1
NO
Small amount of dark liquid NO
II.
Large amount of clear liquid
Large
and
Yes
Yes
Abundant
Yes
mouth nose
Finethefroth at
s
5
5
SD”
%/I)
Left
5.5”’
64.4 *
I9
hEslimdted
“Number
2.7”’
23.3 *
I8
time between
of replicates.
(pg
+ 3.8 “’
?? 1.7”’
zt SD”
death
109.6
20.4
*
1.2’”
7.4”’
and autopsy.
f
116.7 + 4.7”’
20.5
I7
12.5”
36.4 ?? 1.9”
*
34.1
41.8
+ 2.2”’
20.8
%/I)
concen-
ventricle
tration
(LV)
I6
*
(pg
tration
ventricle
concen-
Right
(> 50 gg G/l)
(RV)
water
15
Case
Brackish
Table
580
92
575
342
800
71.9
<0
153.6
15.8
16.6
x
1000
48
S-12
4Ob
3
<24
and
(h)
30-9
15-3
2-11
31-7
l-l
month)
(day
death
in water
had been
55
24
75
66
29
River
River
place
Watering
Pool
River
Increased
~
creased
Lightly
Acute
~
Very
in-
Yes
icreased
edema
amount
liquid
~
liquid
amount of frothy
Large
of liquid
Large
frothy
Abundant
Yes
~
500 ml
> 500 ml
> 500 ml
Yes
“OX
the lungs
mouth of
in the surface
and
5
5
??
z
%
5
?
; % 3 2.
* %
5
J. Azpurren et al. /Forensic
Sci. ht.
69 (1994)
139-148
145
146
J. A:parren et al. /Forensic
Sri. ht. 69 (1994)
139-148
RV)/W, since the water samples were not taken at the same moment as death took place. In order to determine the reference levels of strontium in the blood of corpses, we have analyzed 20 different blood samples without anticoagulants taken from corpses where the cause of death was other than drowning. We have found an average concentration of 27.97 pg/l and a range of 14.64-47.46 &I. ‘A few hours’ interval between the death and the collection of the sample could mean a great variation in the concentration of the strontium in water in those two different moments, so that the sample would have no value at all. 5. Discussion 5.1. Cases that occurred in sea water or at the river mouth
According to the autopsy findings, only in case 13 was a ‘dry lung’ found, and in cases 14 and 7 the appearance of the lung was considered doubtful. In our study, as we can see in Table 4, there are two cases, 13 and 14, whose values for 1000 (LVRV)/W (3.1 and 0.5) were quite different from those obtained in the rest of the cases (>9). Cases 13 and 14 seem to show no entrance of strontium from sea water to the arterial blood during the time the victims were alive, which could indicate cases of laryngeal spasm with no water introduction into the lung; cases of vagal inhibition where death took place immediately after a cardiac arrest; or cases of postmortem immersion of the victim. In case 7, according to autopsy findings, the lungs have not the typical appearances found in ‘wet drowning’ cases. The forensic pathologist suggested that the cause of death could be a cardiac arrest. In our laboratory we found a diatom in the femur, together with a high alcohol concentration in blood and vitreous humor - 4.6 and 3.5 g/l, respectively - as well as marked differences in the concentrations of chlorides in blood at the left and the right side of the heart - 154 and 128 mmol Cl/l, respectively - and of hemoglobin - 100 and 69 g/l, respectively. Therefore, it seemed that the victim had swallowed a certain quantity of water before dying. The high concentrations of strontium found in venous blood in cases 13 and 14 - 3 12 and 97 &l, respectively - were above those considered as ‘reference-values’ (14.6-47.5 &l) and could be explained by a postmortem diffusion that increased the strontium concentration in blood of both ventricles by the same amount. The differences found in the strontium concentration in blood from the left and the right sides of the heart in the rest of the cases, which are supposed to be cases of ‘wet drowning’, is in the range 75-900 &l. This variation could be partially explained by the different amount of strontium that had been passing from sea water to the arterial blood during the time the victim remained alive with water in the lung. 5.2. Cases that occurred in water with >_50 pg S-/l
At this time, we are considering those cases of drowning that occurred in rivers or natural pools, with strontium concentration greater than the ‘reference values’ of strontium in blood. The behaviour of strontium concentration in arterial blood in
J. Azparren et al. /Forensic Sci. hr. 69 (1994) 139-148
141
these cases is difficult to predict; on the one hand it is expected that strontium passes from the water to the arterial blood, but on the other hand, a hemodilution in the arterial blood could be produced due to the possible hypotonic condition of this water. According to the autopsy findings from five cases, four of them were considered as ‘typical drowning’ cases. In the other case (19), the forensic pathologist thought it could be a death caused by vagal inhibition. In our laboratory, only in two cases (17 and 19) did we find an appreciable difference in strontium concentration between the blood of the left and the right chambers. In the other three cases, this difference was found to be < 15 &l. In case 19, we found similar values in the concentration in blood of the left and the right ventricles in hemoglobin - 173 and 172 g/l - and chloride - 98 and 110 rnmol/l. This was in agreement with the cause of death proposed by the forensic pathologist. The concentration of strontium here was in contradiction with the rest of the data, and perhaps could be affected by factors such as contamination of strontium in the container or by the place where the blood sample was taken from. A homogenization of the blood in the ventricles before the sampling should be necessary, to avoid a concentration of strontium towards the lower levels of the heart, as was observed to have occurred in the blood sample containers received. 5.3. Cases that occurred in fresh water (cSOt.~g Sr/l) No appreciable differences were found in the concentration of strontium in blood from the left and the right side of the heart, due to the similar concentration of strontium found in water and blood. Case number 24 was a case of death due to inhalation of CO in the bathtub, confirmed by the high concentration of carboxyhemoglobin found in blood. Case number 23 was a ‘typical drowning’ case confirmed by the elevated number of diatoms found in several organs and blood. 6. Conclusion The combined measurement of blood strontium in two ventricles seems to help in the diagnosis of drowning in salt water. In fact, the differences in the concentration of strontium in two ventricles, obtained in 12 cases of supposed ‘typical drowning’ that occurred in salt water, were 175 pgSr/l, against ~20 PgSrll obtained in two cases where death could have been produced by causes that did not imply a transfer of strontium from water to blood, according to anatomopathological observations during the autopsy. Where drowning occurred in rivers with strontium concentration in the range 90-800 &l, only in certain cases could the strontium determination be useful in the diagnosis of death by drowning. Acknowledgements We wish to thank Lourdes Saez for proof reading the English in this manuscript.
J. Azparren et al. /Forensic
148
Sri. Int. 69 (1994) 139-148
References H.G. Swann and N.R. Spafford, Body salt and water changes during fresh and sea water drowning. Texas Rep. Biol. Med., 9 (1951) 356-382. PI H.G. Swann, M. Bruer, C. Moore and B.L. Vezien, Fresh water and sea water drowning a study of the terminal cardiac and biochemical events. Texas Rpts. Biol. Med., 5 (1947) 423-437. 131 J. Peam, Pathophysiology of drowning. Med. J. Australia, 142 (1985) 586-588. [41 O.A. Gettler, A method for determination of death by drowning. J. Am. Med. Assoc., 66 (1921)
Ill
1650.
PI H.D. Stanley, CF. Henry and E.S. Henry, Blood changes in man following death due to drowning. Arch. Pathol., 56 (1963) 454-461.
161 G. Canepa, Fer hematique et submersion. An. Med. Leg., 1 (1963) 27-33.
171 A.R. Moritz, Chemical methods for the determination of death by drowning. Physiol. Rev., 24 (1944) 70-88.
PI J.A. Llorente, E. Villanueva, C. Hemindez-Cueto
and J.D. Luna, Plasmatic Levels of atria1 natriuretic peptide (ANP) in drowning. A pilot study. Forensic Sci. Int., 44 (1990) 69-75. 191 N. Foged, Diatoms and drowning - once more. Forensic Sci. Int., 21 (1983) 153-l 59. 1101 A.J. Peabody, Diatoms and drowning. A review. Med. Sci. Law, 20(4) (1980) 254-261. [Ill M. Piette, J. Timperman and N. Parisis, Serum strontium as a medico-legal diagnostic indicator of drowning. Med. Sci. Law., 29(2) (1989) 162-171 WI M.A. Amin, A.H. Samia, M.A. Kabil et al. Serum strontium estimation as a diagnostic criterion of the type of drowning water. Forensic Sci. Int., 28 (1985) 47-52. [I31 S. Icard, La prense de la mort par submersion suivant qu’elle a en lieu dans une riviere ou dans la mer. Rev. Pathol. comparee d’Hygiene gen., 32 (1932) 559-571. 1141 J.P. Riley and G. Skirrow, Chemical Oceanography, 2nd. edn., Vol. 2, Academic Press, New York, 1975, pp. 52-60. 1151 E.D. Goldberg, In the sea. Vol. 2. In M.N. Hill (ed.), Wiley Interscience, New York, 1963, pp. 3-25.