|Year : 2019 | Volume
| Issue : 3 | Page : 141-146
Estimating a reliable cutoff point of 1-propanol in postmortem blood as marker of microbial ethanol production
Vassiliki A Boumba1, Nikolaos Kourkoumelis2, Kallirroi Ziavrou1, Theodore Vougiouklakis1
1 Laboratory of Forensic Medicine and Toxicology, Faculty of Medicine, School of Health Sciences, University of Ioannina, Ioannina, Greece
2 Laboratory of Medical Physics, Faculty of Medicine, School of Health Sciences, University of Ioannina, Ioannina, Greece
|Date of Web Publication||18-Sep-2019|
Vassiliki A Boumba
Laboratory of Forensic Medicine and Toxicology, Faculty of Medicine, School of Health Sciences, University of Ioannina, 451 10 Ioannina
Source of Support: None, Conflict of Interest: None
The interpretation of the ethanol analysis results in postmortem cases may be challenging when the origin of postmortem ethanol (antemortem ingestion or microbial production) is under dispute. In this study, we investigated the suitability of blood 1-propanol cutoff concentration as a reliable marker for the discrimination between “positive” and “negative” for postmortem ethanol production (PEP) autopsy blood samples by performing receiver operating characteristic (ROC) analysis. The results indicated that a threshold 1-propanol concentration of 0.104 mg/dL had an area under the curve of 0.90 (standard deviation = 0.03), sensitivity of 79%, and specificity of 91% for predicting PEP with 95% confidence interval. This means that the positive for PEP autopsy blood samples yield significantly differentiated PEP than approximately 90% of the controls. The estimated concentration of 1-propanol is an applicable threshold (cutoff) value for autopsy blood 1-propanol to discriminate between “positive” and “negative” samples for PEP. We named this threshold 1-propanol concentration (of 0.104 mg/dL) as “1-propanol criterion.” In an effort to test its applicability to postmortem cases, we evaluated blood ethanol and 1-propanol from 222 postmortem cases records. The results showed that 10% of the tested samples were positive for PEP, and only a few of them were from bodies with signs of putrefaction at autopsy. This finding indicates that PEP in a corpse could take place before the appearance of obvious putrefaction. We suggest the ROC-based calculation of the 1-propanol cutoff concentration, at 0.104 mg/dL, as an effective method for “flagging” blood samples as positive for PEP irrespectively of the presence or not of putrefaction.
Keywords: 1-propanol, blood, ethanol, forensic, postmortem
|How to cite this article:|
Boumba VA, Kourkoumelis N, Ziavrou K, Vougiouklakis T. Estimating a reliable cutoff point of 1-propanol in postmortem blood as marker of microbial ethanol production. J Forensic Sci Med 2019;5:141-6
|How to cite this URL:|
Boumba VA, Kourkoumelis N, Ziavrou K, Vougiouklakis T. Estimating a reliable cutoff point of 1-propanol in postmortem blood as marker of microbial ethanol production. J Forensic Sci Med [serial online] 2019 [cited 2019 Oct 17];5:141-6. Available from: http://www.jfsmonline.com/text.asp?2019/5/3/141/267156
| Introduction|| |
The analysis of ethanol in postmortem cases constitutes a decisive and important part of the death investigation process, and although blood ethanol concentrations are easily and accurately determined; the interpretation of the results may be challenging when the origin of postmortem ethanol is under dispute (antemortem ingestion or microbial production). The presence of higher alcohols (1-propanol, isobutanol, 1-butanol, amyl and isoamyl alcohol) in an autopsy blood sample has been considered as markers of postmortem microbial ethanol production.,,,, Moreover, semi-quantitative relationships between the 1-propanol and ethanol concentrations (expressed as ratios) have been reported.,,, However, other studies have questioned the role of 1-propanol in the discrimination between antemortem ethanol consumption and postmortem ethanol formation., On the other hand, the determination of biomarkers of alcohol consumption in autopsy blood samples, such as ethyl glucuronide, ethyl sulfate, phosphatidylethanol, and fatty acid ethyl esters, has been suggested, although their presence has been considered unreliable in determining the source of ethanol postmortem, especially in highly putrefied cases.,,,, A variety of confounding factors and artifacts can obscure postmortem ethanol production (PEP). Therefore, the identification of reliable laboratory markers for the assessment and documentation of PEP samples remains a challenging field in postmortem toxicology.
Receiver operating characteristic (ROC) curve and the area under the curve (AUC) reported as ROC analysis are extensively used in clinical epidemiology for the assessment of the discrimination accuracy of diagnostic biomarkers (e.g., serum markers) and imaging tests in classification modeling of the “diseased” from “nondiseased” (healthy controls). The conventional evaluation of a diagnostic test with dichotomous outcome (positive/negative test results) uses sensitivity and specificity as metrics of accuracy of the test, so as, sensitivity and specificity vary across the different thresholds, and the sensitivity is inversely related to specificity. The ROC curve (the plot of sensitivity vs. specificity) produces a “separator” scale, on which results of the diseased and nondiseased form a pair of overlapping distributions, with the derived AUC being an effective index of accuracy (the inherent ability of the test to discriminate between diseased and healthy controls). The complete separation of the two underlying distributions implies a perfectly discriminating test, while complete overlap implies no discrimination.
The aim of this study was to investigate, by performing ROC analysis, the suitability of the postmortem blood 1-propanol “cutoff” concentration, as a reliable marker for the discrimination between “positive” and “negative” for PEP autopsy blood samples.
| Materials and Methods|| |
Determination of ethanol and 1-propanol
Autopsy blood ethanol and 1-propanol concentrations were determined by headspace-gas chromatography-flame ionization detector (HS-GC-FID) as described previously. GC analyses were performed on a Shimadzu GC-17A gas chromatograph equipped with a SUPELCOWAX™−10 fused silica capillary column (30 m × 0.25 mm, film thickness 0.25 μm) and with a FID. The GC was fitted with a Shimadzu AOC-5000 headspace-GC automated sample pretreatment and injection system. The temperature of the injection port, the column, and the FID was 115°C, 60°C, and 260°C, respectively. The carrier gas was helium with a flow rate of 0.7 mL/min with a constant pressure of 65 kPa. The injection inlet was set in a split mode with a split ratio of 10:1. The samples were incubated at 50°C for 8.0 min prior to injection with an agitation speed of 500 rpm. A 2.5 mL gas-tight syringe was used heated at 105°C. A HS aliquot of 500 μL was sampled for analysis with a feel speed of 250 μL/s and a pull-up delay of 500 ms. The injection speed was 1000 μL/s at a needle penetration depth of 20 mm. After injection syringe was flushed with helium for 1.5 min. Acquisition time was 10 min. In 10 mL HS vials containing 0.50 g ammonium sulfate were added 500 μL of the calibration solution or the culture medium or blood culture and 500 μL of the internal standard solution. The vials were then sealed with metal crimp caps fitted with silicone septa and were put to the HS autosampler for analysis. Ammonium sulfate was added to increase the ionic strength of the solution. Calibration curves were constructed for each analyte in six concentration levels within the relevant working concentration range. Stock aqueous standard solutions were prepared in concentration 4.00% w/v for ethanol and 0.25% w/v and 0.40% w/v for 1-propanol. Acetonitrile was used as internal standard for the determination of volatiles in aqueous solution of 100 mg/dL. The above solutions were stored at 4°C for up to 6 months. Working solutions were prepared on a daily basis by mixing the appropriate volumes of the corresponding stock solutions of each analyte and double distilled (DD) water. Blood quality control samples were prepared from stock solutions in three different concentrations for each analyte. Limits of detection (LOD) and quantitation were 0.01 mg/dL, 0.03 mg/dL, and 0.02 mg/dL, 0.04 mg/dL for ethanol and 1-propanol, respectively.
Sample definition and selection
Autopsy blood samples included in the study were from cases routinely autopsied in the Laboratory of Forensic Medicine and Toxicology of the Faculty of Medicine, University of Ioannina, on request by the Police. According to the Greek legislation, all criminal, suspicious, accidental, suicidal, sudden, and unexpected, or in any way unexplained death cases should be subjected to medicolegal investigation by the forensic pathologist who is authorized to decide on the sampling site and the types of postmortem specimen that would be collected and subjected to toxicological analysis.
Autopsies were routinely performed 12–48 h after reporting of the death (or finding of the body); corpses were kept refrigerated at 4°C till autopsy. The collection site was primarily peripheral, followed by the heart cavity or any other available site in cases, with severe trauma or advanced putrefaction. Autopsy blood samples were collected in sterilized blood collection tubes of 4.0 mL (containing ethylenediaminetetraacetic acid and KF) (VACUETTE ®, Greiner, Labortechnik, Austria) and stored at 4°C till analysis. Analyses for ethanol and 1-propanol were performed 2–24 h post autopsy. All samples were measured in duplicates, and mean values of the two measurements were reported. Blood alcohol content (BAC) should have shown an agreement of ± 5%; otherwise, a third measurement was performed. Concentrations of all volatiles higher than the respective LODs were considered positive for the evaluation presented herein. The review of the archives of the relative autopsy reports revealed the manner of death (natural, violent, or undetermined) and the putrefaction state of the body at autopsy. Autopsy reports were written and signed by the forensic pathologist after evaluating the autopsy findings, toxicological and/or histological results for each case. Common obvious signs of putrefaction at autopsy were considered the greenish discoloration of the abdomen and genitals, possibly spread up to breast and limbs, the abdomen swelling, and the bloated body. Otherwise, in the absence of any obvious sign of putrefaction at autopsy, cases were characterized as without putrefaction.
The procedure followed for the microbiological testing of the autopsy blood samples was reported previously. Briefly: blood samples were collected during autopsy into sterilized collection tubes containing anticoagulant. A loopful of blood was plated on selective agar: Gram negatives were identified by plating a loopful of buffered peptone water on Violet Red Bile Agar and Chromocult Coliform Agar and incubation aerobically at 37°C for 24 h. For the isolation of Gram-positive strains, Mannitol salt agar and D-Coccosel Bile Esculin agar were used, respectively. The plates were incubated at 37°C aerobically. For the isolation of clostridium species, a loopful was streaked onto blood agar plates, the inoculated plates were incubated anaerobically using the GENbag anaer incubation systems for 48 h at 37°C and were examined for b-hemolysis.
ROC curve analyses were used to determine the optimal reliable cutoff value for the postmortem 1-propanol concentration as a marker for microbial ethanol production, in an effort to differentiate between blood samples with PEP (positives) and without PEP (negatives). An applicable cutoff value for blood 1-propanol to distinguish between cases with possible PEP and not possible PEP was estimated.
| Results|| |
The case-specific parameters evaluated for the ROC analysis were the putrefaction state of the body at autopsy (with or without obvious sign of putrefaction) and the manner of death (natural, violent, and unspecified). The blood selected as “true negatives” for PEP (131 samples) were sampled during autopsy from respective number of corpses (131 cases) without obvious signs of putrefaction. The “true negatives” were all sampled from natural death cases, and those with ethanol concentration higher than the LOD (0.01 mg/dL) were found negative for microbial burden. Therefore, the ethanol detected in this group of blood samples was attributed to antemortem consumption.
The blood selected as “true positives” for PEP (40 samples) were sampled during autopsy from the respective number of corpses (40 cases) with obvious (early to severe) signs of putrefaction; the “true positives” were sampled either from unspecified cause of death cases (29 cases) or from violent death cases with extended open traumas (11 cases). All the true positive samples carried microbial burden (belonging either to Gram positives or Gram negatives or clostridia species). Consequently, ethanol and/or 1-propanol in these blood samples had originated (partly or totally) by postmortem microbial production. Considering the described characteristics and the concentrations of ethanol and 1-propanol of the autopsy blood samples, the sensitivity (true positive rate) and specificity (true negative rate) were calculated. The concentrations of ethanol and 1-propanol as well as other characteristics of the true positives and true negatives samples/cases are shown in [Table 1].
|Table 1: Evaluated cases for the receiver operating characteristic analysis (n=171)|
Click here to view
We generated ROC curve to determine the ability of the autopsy blood 1-propanol concentration to predict PEP. The optimal reliable cutoff value for the postmortem blood 1-propanol concentration as a marker for microbial ethanol production was determined. The performance of 1-propanol as a diagnostic variable was quantified by the area under the ROC curve (AUC) [Figure 1]. The results indicated that a threshold 1-propanol concentration of 0.104 mg/dL had an AUC of 0.90 (standard deviation = 0.03), sensitivity of 79%, and specificity of 91% for predicting PEP within 95% confidence interval. This means that the positive for PEP autopsy blood samples yield significantly differentiated PEP than approximately 90% of the controls. The estimated concentration of 1-propanol is an applicable threshold (cutoff) value for autopsy blood 1-propanol to distinguish between “positive” and “negative” samples for PEP. We named this threshold 1-propanol concentration (of 0.104 mg/dL) as “1-propanol criterion.”
|Figure 1: Receiver operating characteristic curve to determine the ability of the autopsy blood 1-propanol concentration to predict postmortem ethanol production and its characteristics|
Click here to view
The “1-propanol criterion” was used to evaluate the origin of ethanol in autopsy blood samples from 222 autopsy cases, in an effort to test its applicability to postmortem cases. Tested cases were selected from our chromatogram archives according to ethanol or/and 1-propanol concentrations: both concentrations should have been equal or higher than the respective limits of quantification.
It was found that 24 samples (24/222, 10.8%) were “flagged” as positive for PEP. The retrospective review of the relevant autopsy reports revealed that the majority of positive for PEP samples (22/24) were violent death cases without putrefaction at autopsy. The two positives for PEP samples were collected from natural death cases, and they had relatively low amounts of blood ethanol concentration and signs of putrefaction at autopsy. Negative for PEP cases (198/222) were all from violent or natural deaths. A portion of 53/198 blood samples has originated from corpses that had open wounds at autopsy. Furthermore, a minority of the respective corpses (6/53) had signs of putrefaction at autopsy. Characteristics of the tested samples/cases are presented in [Table 2].
|Table 2: Application of the “1-propanol criterion” in autopsy blood sample|
Click here to view
Characteristics of the cases which were flagged as positives for PEP, are presented in [Table 3]. Most positive cases (22/24) were violent deaths, and the respective corpses did not have any signs of putrefaction at autopsy. Interestingly, 14/22 (64%) positive cases were violent deaths, and corpses had open wounds at autopsy (V1, V4–7, V9–10, V13–14, and V18–22). Moreover, three blood samples had been sampled from drowned persons (V3, V11, and V17).
|Table 3: Case-specific characteristics of the autopsy bloods characterized as positives for postmortem ethanol production|
Click here to view
| Discussion|| |
PEP is a phenomenon difficult to recognize with certainty. The putrefaction state of the cadaver, the clinical history, the determination of glucose levels, the identification of microbes in the analyzed sample, discrepancies between ethanol concentration from various sampling sites and different specimens, and the presence of other volatiles have been suggested as indicators of PEP in the effort to achieve feasible accuracy in interpreting the ethanol analysis results.,
Among volatiles possible present in an autopsy blood sample, 1-propanol is the volatile mostly correlated to putrefaction and subsequent microbial ethanol production. Furthermore, some studies suggest determination of the ratio of postmortem ethanol and 1-propanol concentrations to verify the existence and the extent of PEP.,,, However, the reliability of the ethanol to 1-propanol concentration ratio, as a biomarker to establish whether the ethanol found in an autopsy blood sample has resulted from postmortem microbial production or from antemortem ingestion, is still under debate since it is disproved by the study of Liang et al. Despite this, the current literature agrees that 1-propanol might provide valuable information to assist with the interpretation of ethanol findings in decomposed bodies and drive the toxicology to further and more specific analyses, such as the determination of ethyl glucuronide.
Postmortem microbial activity is the main source of 1-propanol in a postmortem specimen.,, Other possible sources of 1-propanol in a postmortem blood sample are the antemortem consumption of alcoholic beverages containing 1-propanol (among other higher alcohols, known also as congener alcohols),, which result in relatively low blood 1-propanol concentrations (0.042 mg/dL on average when BAC is up to 1.22 g/L); and the antemortem endogenous production of congener alcohols, which result in negligible concentrations of 1-propanol (up to 48 μg/L or 0.0048 mg/dL in serum). Another possible source of postmortem 1-propanol is the antemortem consumption of products containing 1-propanol, which is expecting to result in high postmortem blood 1-propanol concentrations (such cases are rare and were not included in this study). Therefore, the presence of 1-propanol in postmortem blood is not per se a marker of postmortem ethanol neoformation, and criteria should be established to differentiate between positive and negative for PEP samples. This study was conducted in an effort to contribute to this goal.
ROC analysis is used in clinical epidemiology to quantify how accurately medical diagnostic tests can discriminate between two patient states, typically referred to as “diseased” and “not diseased.” In the current study, ROC curves were used to evaluate the predictive ability of blood 1-propanol as a biomarker of PM microbial ethanol production. ROC analysis was applied in a significant number of autopsy cases with known blood ethanol and 1-propanol concentrations and putrefaction state at autopsy. The putrefaction state at autopsy (and secondary the manner of death) was used as case-specific characteristic to discriminate “true positive” cases from “true negatives.” The absence of microbes was an additional characteristic of true negative cases. Autopsy blood ethanol and 1-propanol concentrations were used to evaluate specificity and sensitivity, as measures of accuracy, across different thresholds of 1-propanol concentrations. The result showed that 1-propanol concentration, at 0.104 mg/dL, is an effective threshold concentration (“cut-off”) for “flagging” an autopsy blood sample as positive for PEP. We named this threshold 1-propanol concentration as “1-propanol criterion.”
The “1-propanol criterion” was applied to 222 autopsy blood samples to characterize them as “positive” or “negative” for PEP. The results showed that 10% (24/222) of the tested samples were positives for PEP. Interestingly, only two samples were from corpses that had obvious signs of putrefaction at autopsy. Blood ethanol concentrations for these two samples (2/24 cases) were relatively low and within the range of ethanol concentrations that usually is considered of microbial origin in putrefied corpses., The majority of them identified as positive samples (20/22 cases, 91%) including those with external or internal traumatic lesions were from corpses without obvious signs of putrefaction at autopsy. It is worth mentioning that the presence of wounds to the corpse increases the possibility of a microbial invasion from the environment to the dead body, and hence, it promotes the expansion of internal microflora. Microbial activity and as a consequence PEP could have happened in a corpse before the appearance of obvious signs of putrefaction, and in such cases, an early marker of PEP is essential. This contribution suggests an objective and measurable laboratory indicator of microbial ethanol production, beyond and irrespectively of the decomposition state of the corpse (putrefaction), or the presence of traumatic lesions, or any other case-specific characteristic that has been previously related to PEP.,,,,, Our results show that the “1-propanol criterion” could be per se a valuable identifier of “positive” samples for PEP.
Another advantage of the “1-propanol criterion” is that it provides a means of quantitative estimation of PEP: as 1-propanol concentration increases, the PEP is expected to extend. The typically reported postmortem microbial ethanol concentrations in decomposed bodies are up to 70 mg/dL (0.7 g/L), although concentrations as high as 220 mg/dL (2.20 g/L) have been reported in some documented cases as well as, an extreme case of 300 mg/dL (3.00 g/L). Therefore, blood ethanol levels higher than 0.7 mg/dL in decomposed bodies have apparently resulted from antemortem ethanol consumption. If ethanol is detected in autopsy blood with no obvious putrefaction, the “1-propanol criterion” suggests that PEP is due to ethanol neoformation. Furthermore, ethanol concentration due to postmortem production is expected to increase when the concentration of 1-propanol increases beyond the cutoff concentration of 0.104 mg/dL. In such cases, the determination of biomarkers of ethanol consumption ,,,, could assist the better interpretation of the ethanol analysis results; although the extent of the microbial produced ethanol would still have remained obscure. Therefore since the phenomenon of microbial ethanol production is highly complex and depends on many variables, we expect a deviation in such cases which need more data to become clarified.
| Conclusion|| |
We suggest the ROC-based calculation of the 1-propanol cutoff concentration, at 0.104 mg/dL, as an effective method for “flagging” autopsy blood samples as positive or negative for PEP.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Leikin JB, Watson WA. Post-mortem toxicology: What the dead can and cannot tell us. J Toxicol Clin Toxicol 2003;41:47-56.
Flanagan RJ, Connally G. Interpretation of analytical toxicology results in life and at postmortem. Toxicol Rev 2005;24:51-62.
Boumba VA, Ziavrou KS, Vougiouklakis T. Biochemical pathways generating post-mortem volatile compounds co-detected during forensic ethanol analyses. Forensic Sci Int 2008;174:133-51.
Boumba VA, Economou V, Kourkoumelis N, Gousia P, Papadopoulou C, Vougiouklakis T. Microbial ethanol production: Experimental study and multivariate evaluation. Forensic Sci Int 2012;215:189-98.
Boumba VA, Kourkoumelis N, Gousia P, Economou V, Papadopoulou C, Vougiouklakis T. Modeling microbial ethanol production by E. Coli under aerobic/anaerobic conditions: Applicability to real postmortem cases and to postmortem blood derived microbial cultures. Forensic Sci Int 2013;232:191-8.
Nanikawa R, Ameno K, Hashimoto Y, Hamada K. Medicolegal studies on alcohol detected in dead bodies – Alcohol levels in skeletal muscle. Forensic Sci Int 1982;20:133-40.
Felby S, Nielsen E. Congener production in blood samples during preparation and storage. Blutalkohol 1995;32:50-8.
Moriya F, Hashimoto Y. Postmortem production of ethanol and n-propanol in the brain of drowned persons. Am J Forensic Med Pathol 2004;25:131-3.
Petković SM, Simić MA, Vujić DN. Postmortem production of ethanol in different tissues under controlled experimental conditions. J Forensic Sci 2005;50:204-8.
Chang J, Kollman SE. The effect of temperature on the formation of ethanol by Candida albicans
in blood. J Forensic Sci 1989;34:105-9.
Liang H, Kuang S, Guo L, Yu T, Rao Y. Assessment of the role played by N-propanol found in postmortem blood in the discrimination between antemortem consumption and postmortem formation of ethanol using rats. J Forensic Sci 2016;61:122-6.
Høiseth G, Karinen R, Christophersen AS, Olsen L, Normann PT, Mørland J. A study of ethyl glucuronide in post-mortem blood as a marker of ante-mortem ingestion of alcohol. Forensic Sci Int 2007;165:41-5.
Thierauf A, Kempf J, Perdekamp MG, Auwärter V, Gnann H, Wohlfarth A, et al.
Ethyl sulphate and ethyl glucuronide in vitreous humor as postmortem evidence marker for ethanol consumption prior to death. Forensic Sci Int 2011;210:63-8.
Thompson PM, Hill-Kapturczak N, Lopez-Cruzan M, Alvarado LA, Dwivedi AK, Javors MA. Phosphatidylethanol in postmortem brain and serum ethanol at time of death. Alcohol Clin Exp Res 2016;40:2557-62.
Liu Y, Zhang X, Li J, Huang Z, Lin Z, Wang J, et al.
Stability of ethyl glucuronide, ethyl sulfate, phosphatidylethanols and fatty acid ethyl esters in postmortem human blood. J Anal Toxicol 2018;42:346-52.
Santunione AL, Verri P, Marchesi F, Rustichelli C, Palazzoli F, Vandelli D, et al.
The role of ethyl glucuronide in supporting medico-legal investigations: Analysis of this biomarker in different postmortem specimens from 21 selected autopsy cases. J Forensic Leg Med 2018;53:25-30.
Hajian-Tilaki K. Receiver operator characteristic analysis of biomarkers evaluation in diagnostic research. J Clin Diagn Res 2018;12:LE01-8.
Ziavrou K, Boumba VA, Vougiouklakis TG. Insights into the origin of postmortem ethanol. Int J Toxicol 2005;24:69-77.
Kugelberg FC, Jones AW. Interpreting results of ethanol analysis in postmortem specimens: A review of the literature. Forensic Sci Int 2007;165:10-29.
Krause D, Wehner HD. Blood alcohol/congeners of alcoholic beverages. Forensic Sci Int 2004;144:177-83.
Rodda LN, Beyer J, Gerostamoulos D, Drummer OH. Alcohol congener analysis and the source of alcohol: A review. Forensic Sci Med Pathol 2013;9:194-207.
Wunder C, Hain S, Koelzer SC, Paulke A, Verhoff MA, Toennes SW. Lack of effects of a “sobering” product, “Eezup!”, on the blood ethanol and congener alcohol concentration. Forensic Sci Int 2017;278:101-5.
Liebich HM, Buelow HJ, Kallmayer R. Quantification of endogenous aliphatic alcohols in serum and urine. J Chromatogr 1982;239:343-9.
Saukko P, Knight B. Knights Forensic Pathology: The Pathophysiology of Death. 3rd
ed. New York: Oxford University Press; 2004.
Gilliland MG, Bost RO. Alcohol in decomposed bodies: Postmortem synthesis and distribution. J Forensic Sci 1993;38:1266-74.
Zumwalt RE, Bost RO, Sunshine I. Evaluation of ethanol concentrations in decomposed bodies. J Forensic Sci 1982;27:549-54.
[Table 1], [Table 2], [Table 3]