A Comparison between Conventional Methods and Proton- Magnetic Resonance Spectroscopy (H1-MRS) for Estimation of Postmortem Interval in Six Different Species: A Systematic Literature Review

Search Strategy

We conducted an electronic search in ten electronic search engines: PubMed, Google Scholar, Scopus, Web of Science, Virtual Health Library (VHL), Popline, International

Scientific Indexing (ISI), WHO Global Health Library, New York Academy of Medicine Grey Literature Report (NYAM), and System for Information on Grey Literature in Europe (SIGLE). They were searched for relevant studies published before January 2023, as shown in Figure 1. We used the search terms “postmortem interval”, “Livor mortis”, “Rigor mortis”, “Body cooling”, “Putrefaction Signs”, and “proton-magnetic resonance spectroscopy”, and then the search string was modified to be suitable for each database. Reference lists of some published papers were examined for more sources not found in the database searches.

Click to enlarge

Study Inclusion Criteria and Study Selection

Only original articles (published before January 2023) concerning the use of conventional methods or H1-MRS in the determination of PMI in humans and different animal species were included in this systematic review. There was no restriction on a certain language, publication year, or country, but we limited the included studies to those trials that were applied away from the effect of ambient temperature on brain metabolite concentrations. Letters, editors’ correspondences, news, and studies whose data could not be reliably retrieved or extracted were excluded. A preliminary assessment of the titles and abstracts was performed to identify relevant articles. After that, full texts were screened, and the eligible texts were used in the data synthesis.

Data Extraction

The following data were extracted from the included studies: the first author’s name, year of publication, year of carrying out of the experiment, study design, country and city of origin, characteristics of human and animal models (species, age, gender, weight), cause of death, number of included subjects in each experiment, name of metabolites produced in each species, duration of spectroscopy measurement, and time interval from death until the occurrence of brain decomposition in all species. Some graphs were plotted to be used in data synthesis. In addition, the identification and elimination of overlapping datasets were performed under extensive care, especially for multiple studies containing the same species.

Data Analysis

Meta-analysis was inapplicable due to a lack of studies on the same species with the same circumstantial evidence and the same environmental conditions.

Quality Assessment

A total of 11 studies were identified to be assessed by using the Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2)” tool, which is highly recommended by the Cochrane Collaboration for diagnostic test accuracy (DTA). QUADAS-2 addressed “risk of bias” and “applicability concerns” through 4 domains: patient selection, index test, reference standard, and flow and timing, as shown in Table 1. The quality of the included studies was evaluated as medium.

Notably, most studies did not compare the results of H1- MRS with the results of the conventional methods for PMI calculation, so “reference standard” and “flow and timing” were categorised as “unclear” risk of bias and applicability concerns.

Canine

For almost more than a century, pathologists have depended on many conventional methods, such as alterations in body temperature, lividity, rigor mortis, and signs of putrefaction, to identify the time elapsed after death in different animals [11]. Some people used to own dogs in their houses, which is why animal cadavers might be found beside their dead owners in crimes. Therefore, studying the postmortem changes in canine cadavers has been a point of interest to help pathologists determine the PMI of dogs and consequently the PMI of their owners [12]. Based on that, PMI was determined as follows: during the 1st day PM, the rectal temperature of the canine was above the ambient temperature, the cornea dried out, rigor mortis started to develop in most muscle groups, and blue to green skin discolouration developed in the caudal abdomen and flank [13].

On the 3rd day PM, the rectal temperature equaled the ambient temperature, the upper eyes were shrunken, the rigidity of some joints was lost and skin discolouration of the abdomen and flank became deeper. On the 7th day PM, the eyes collapsed, rigor mortis was still present in the jaw and hind limbs, and signs of putrefaction showed great development. Finally, by three weeks PM, the cornea and adjacent membranes were colonised by molds and fungi, green, blue, and brown skin discolouration was so marked, and the epidermal layer was easily detached [13].

On another note, in 1995, Choe et al. investigated the alterations in brain metabolite concentrations in dogs before and after death until two days PM [14]. Dogs were treated with anaesthetic agents and then killed by intracardiac injection of 20 ml 10% KCl. The results revealed the presence of four premortem metabolites that were considerably high and characterised for the premortem period (livability) in dogs. They were N-acetyl-aspartate (NAA), creatine (Cr), choline (Cho) and inositol (Ino). There was no resonance signal of lactate before death.

Noticeably, just 30 minutes after death, a significant lactate signal appeared, giving a lactate/Cr ratio (w/w) from 2 to 2.86 mmol/kg, and within three hours PM, it increased to a range of 2.53 - 3.37 mmol/kg. After these early postmortem hours, the lactate/Cr ratio started to decrease gradually until reaching 2.02 - 2.94 mmol/kg at 24 hours PM (Table 2). In addition, the inositol/Cr ratio also increased to a range of 1.16 - 1.86 mmol/kg at three hours PM, but it kept increasing until reaching 1.43 - 3.11 mmol/kg at the end of the 1st day PM.

Considering the neuronal marker NAA, there was a gradual decrease in its concentration until it almost disappeared on the 3rd day PM. This might be attributed to its responsibility for the integrity of neuronal tissue, which was lost when ischemia occurred after death and continued to develop until the 3rd day PM. NAA and creatine concentrations were decreasing at the same time. This might explain why the NAA/Cr ratio was approximately constant and did not decrease substantially (Table 2). At 48 hours PM, autolysis occurred very rapidly in the neurons of the ventral thalamic nucleus and then in the small cells of the striatum, leading to the inability of H1-MRS to identify the signals of metabolites. Notably, slower decomposition occurred in the large pyramidal cells of the cortical layer of the parietal lobe, which were more durable [14]. Therefore, researchers turned to positioning the spectroscopy volume of interest in the brain cortex instead of the thalamus in their later studies [1, 5, 9, 10, 15, 16, 17].

(n =10). * (hPM) refers to hour/s postmortem. * (Cr) refers to creatine; (NAA) refers to N-acetyl aspartate. Table 2: Statistical Evaluation of Metabolite Ratios in the Dog Brain by Using H1-MRS (P<0.05) [14].

Bovine

There are no published data on the estimation of PMI by conventional methods in Bovine, as veterinary forensic studies on ruminants are lacking. However, it is known that pathologists rely on their experience to estimate it subjectively [13].

Therefore, we concluded the results of nuclear magnetic resonance (NMR) spectroscopy only. Bovine brain homogenate was used to represent each animal. Portions of approximately 4.0 g cortex were homogenised and stored in a refrigerator at 4°C until the end of proton NMR measurement at 195 hours PM [18]. Under refrigeration, the concentrations of lactate, creatine, myo-inositol, scyllo- inositol, glutamate, and glutamine remained constant from three hours PM until the end of the experiment, with ranges of 26.6 ± 2.2, 9.4 ± 0.6, 8.1 ± 0.7, 0.58 ± 0.04, 6.9 ± 0.4, and 2.4 ± 0.2 mmol/kg, respectively [18].

The high concentrations of lactate, creatine, and myo- inositol in the brain homogenate were indicative signs of death at three hours PM. In addition, the concentration of GABA, an inhibitory neurotransmitter in the brain tissue, at three hours PM ranged from 1.7 - 2.1 mmol/kg. At 8.5 hours PM, the concentration increased dramatically to 2.3 - 3.3 mmol/ kg, indicating that GABA is a characteristic brain metabolite for bovine brain homogenate at 8.5 hours PM. Additionally, there was a gradual increase in GABA concentration from the onset of death along the experimental time, reaching an average of 1.9 mmol/kg at three hours PM and 6.2 mmol/kg at 195 hours PM [18].

The decomposition of glycerophosphocholine (Gpc) into glycerol and choline after death resulted in an increase in the concentrations of glycerol and choline during the 195 hours of the experiment [18]. For example, at 22 hours PM, the concentration of glycerol ranged from 0.84 - 0.94 mmol/ kg, and at 31 hours PM, it increased to 0.95 - 1.15 mmol/ kg. Glycerol was a characteristic brain metabolite for bovine brain homogenate at that time range (22-31 hours PM).

At 147 hours PM– almost six days PM– the concentrations of choline and glycerol were identified as 1.35 - 1.87 mmol/ kg and 1.91 - 2.13 mmol/kg, respectively (Table 3).

At 195 hours PM, the three characteristic metabolites, namely, alanine, choline, and glycerol, were determined at concentrations of 3.2 - 3.4, 2.02 - 2.30, and 2.52 - 2.56 mmol/ kg, respectively. These specific concentrations appeared to confirm the beginning of the 9th day PM for the bovine bodies. Although the decrease in NAA concentration was insignificant during the first two days PM, it started to decrease gradually, reaching an average ranging from 3.1 - 3.7 mmol/kg on the 9th day PM. Notably, NAA is decomposed into its primary components acetate and aspartate, whose concentrations accordingly increased along the 195 hours PM [18].

(n =24). * (hPM) refers to hour/s postmortem. Table 3: Metabolite Concentrations ((mmol/kg)) in Bovine Brain Homogenate (P<0.05) [18].

Porcine

Porcine is characterised by light rosy skin and less hair than other animals, so postmortem changes appear obviously on their skin after death. During the first 12 hours PM, lividity was prominent in the pig carcass with hypostatic pink skin discolouration, in addition to a beginning of rigor mortis in the muscle group. At the end of the 1st day PM, lividity was darker colored over the head, neck, and trunk, and rigor mortis extended to include the whole body muscles. Purple to black discolouration was observed in the lower abdominal region on the 2nd day PM. The marbling of the trunk region and early onset of bloating were marked signs of the beginning of the decomposition stage on the 3rd day PM. Then, blisters filled with maggots appeared on the trunk region, and the skin started to slip. Towards the end of this stage, bloating developed in the neck, hair was lost, and brown discolouration developed around the edges. These changes were observed in spring/summer in South Africa, and there is no doubt that the rate of corpse decomposition is faster in summer and spring than in winter and autumn [19].

Magnetic resonance spectroscopy was also applied to porcine in a study by Banaschak, et al [13].

Heads of young pigs were obtained from an abattoir, processed, and stored at 21 ± 1°C for three weeks.

The analysis of resonance spectra of pig brain metabolites revealed the presence of alanine, acetate, and choline signals throughout the experiment. In the first four days PM, prominent concentrations of NAA, creatine, and lactate were determined, while succinate and free trimethylammonium (f TMA) began to appear at the beginning of the 5th day PM. The disappearance of NAA at approximately five days and a half PM (~130 hours PM) and creatine at seven days PM (~170 hours PM) marked their postmortem times. Interestingly, lactate was the most durable metabolite, which disappeared almost at the first third of the 9th day PM (~200 hours PM) [15].

Rabbits

The decomposition rate and insect succession of rabbit carcasses occur simply and more rapidly than those of large- weighted carcasses [20]. Corpse decomposition can be divided into five stages (fresh, bloat, active decay, advanced decay, and dry) [21, 22]. The fresh stage lasted approximately 12 hours for outdoor-placed carcasses and 36 hours for indoor-placed carcasses. Then, the bloat stage followed the fresh stage until the 1st and 2nd day PM. High decomposition of outdoor and indoor-placed carcasses marked the active decay stage on the 3rd day PM, and skeletonisation of both carcasses marked the advanced decay stage on the 5th day PM. Finally, the dry stage developed on the 19th day PM for outdoor-placed carcasses and on the 16th day PM for indoor- placed carcasses [22]. Notably, calliphorid flies were the first forensic insects to arrive and breed on rabbit carcasses in all seasons of the year [23].

In addition, regarding the application of the analytical technique H1-MRS on the brains of rabbits in the postmortem period, it showed a substantial decrease in the relative concentrations of NAA, choline, and creatine within 24 hours PM [24, 25]. PMI was estimated through two independent equations using the NAA/Cr and Ch/Cr ratios as independent variables, and the average deviations of the two estimated PMIs were 2.1 - 37.9 min and 1.69 - 40.87 min, respectively [24]. This indicated that NAA/Cr and Ch/Cr were significant metabolic ratios for the estimation of PMI in rabbits.

Sheep

As mentioned before, forensic studies on ruminants are rare. Therefore, we concluded the available published data on sheep as follows: Lucilia cuprina was the first forensic fly that arrived and breaded on sheep cadavers upon death. Minutes- hours PM, it was attracted to sheep cadavers and started a colonisation cycle (egg- maggot- pupa- adult). It is known as a “sheep green bottle” and is only found on corpses located in a drier climate, as it is a very specific insect. Pathologists have used it as a tool for the estimation of PMI by classifying the age of its developmental stages [26].

On another note, several studies were conducted on sheep models for PMI calculation through alterations in their brain metabolite concentrations. Sheep were slaughtered, and heads with attached spinal cords were sealed by plasticine immediately after decapitation to prevent any bacterial invasion or loss of CSF, placed in plastic containers, and stored at room temperature (21 ± 3°C) for investigation [1, 5, 9, 10, 16].

In the premortem period, the analysis of results revealed the presence of propionate and butyrate signals at concentrations of 1.7 mmol/kg and 3.1 mmol/kg, respectively, but no signals of isobutyrate and f TMA [9]. At seven hours PM, the propionate signal disappeared, the butyrate concentration decreased to almost 2.7 mmol/kg, and isobutyrate and f TMA signals remained absent (Table 3). Nevertheless, there were clear signals of creatine and choline compounds in that time range [9, 16]. NAA had a predominant signal with an average concentration of 5-6 mmol/kg a few hours PM. However, it decreased gradually to

3 - 5 mmol/kg at the end of the 1st day PM and to 1 - 3 mmol/ kg on the 2nd day PM [3, 4, 10].

On the 3rd day PM, the propionate concentration reached approximately 3.6 mmol/kg and continued to increase significantly (p < 0.05) until the end of the experiment. fTMA had a prominent signal with a concentration ranging from 6 - 40 mmol/kg from the 4th day to the 7th day PM [9]. In parallel, when the concentrations of alanine, acetate, and lactate increased, the concentrations of NAA, total creatine, and choline decreased over time [16].

On the 9th day PM, there were substantial signals of alanine, acetate, and succinate, while NAA, total creatine, and lactate almost disappeared. On the 15th day PM, there was a noticeable increase in the concentrations of propionate, butyrate, fTMA, and isobutyrate, in addition to acetate, which broke a record, reaching a concentration greater than 200 mmol/kg [16].

*(Total n = 9). * (hPM) refers to hour/s postmortem. Table 4: Metabolite concentrations (mmol/kg) estimated by NMR in the sheep brain [9].

Human

Using conventional methods for the identification of PMI is very common and objective in humans. As livor mortis determined PMI from half an hour to approximately eight hours after death according to its degree [4]. Additionally, algor mortis developed from half to an hour postmortem and continued until approximately 10-12 hPM when the body temperature equaled the environmental temperature. In a parallel way, rigor mortis started within almost two hPM and developed from head to feet in 12 hours, stunned for an additional 12 hours, and then disappeared from feet to head from 24 hPM to 36 hPM [4].

From a forensic entomologic perspective, blowflies are deposited on body orifices at highly variable times depending on the conditions of the cadaver and the environment. Overall, the life cycle of blowflies to become adults is approximately 18 to 24 days, divided into life stages referring to different periods of death [23].

In addition, considering the application of H1-MRS on human subjects, it had some limitations, including number, measurement, and ethical considerations. In spite of the lack of trials and cadavers, researchers could drive promising results for the calculation of PMI using H1-MRS. Human subjects used in these studies were those reported dead in closed areas at room temperature, transported to the Institute of Legal Medicine, and stored at 4°C until scanned [1, 9].

The spectroscopic examination and analysis took almost a couple of hours, which was much shorter than any other conventional criminological investigation [1]. In the early stage, there was a decrease in NAA and NAAG, an obvious increase in lactate and acetate [17], and detectable alterations in the metabolites of Cat. II [9]. Additionally, it was concluded that the overall absence of isobutyrate, fTMA, and propionate signals referred to death from five to six hPM (table 4) [10]. Isobutyrate and fTMA signals appeared after death, as they were not found in the live human brain [9]. The isobutyrate signal started to appear two days PM, and the fTMA signal appeared on the 3rd day PM [10]. After the first three days, creatine showed a decrease followed by disappearance, while acetate and fTMA showed a dramatic increase [17].

(n =4). *(p < 0.05; Z-test, one-tail; calculated from group deviation when n > 1). Table 5: Metabolite Concentrations (mmol/kg) Estimated by H1-MRS in the Human Brain [9].