Influence of blood thiosulfate produced by postmortem changes for the diagnosis of hydrogen sulfide poisoning in forensic autopsy

The data analyses were performed within the framework of routine medicolegal casework following the autopsy (2009) and ethical (2003) guidelines of the Japanese Society of Legal Medicine and approved by the ethics committee of Hokkaido University (No. 16-015). The need for informed consent was waived by our institutional ethics committee.

The average PMI was 1.4 (0.2–3) days in the non-decomposed group, 3.7 (2–14) days in the partially decomposed group, and 41.6 (3–150) days in the severely decomposed group. The average blood thiosulfate concentration of the severely decomposed group was 70.9 (10.5–266.6) μmol/L, while that of the partially decomposed group was 16.3 (0.1–52.7) μmol/L, with many cases exceeding 14 μmol/L (Tables 1 and 2). The mean blood thiosulfate concentration was 1.1 (0.1–3.6) μmol/L in the non-decomposed group and <4 μmol/L for all cases in this group (Tables 1 and 2). There was a statistically significant difference between each of the three groups (P < 0.01).

To investigate whether any specific relationship exists between blood thiosulfate concentrations and PMI in all cases, an approximate equation (regression line)—with thiosulfate concentration on the vertical axis (y) and PMI on the horizontal axis (x)—was used as follows: y = 1.3923x + 10.885; the correlation coefficient was 0.6735, indicating a trend of increasing thiosulfate with increasing PMI (Figure 1). In the severely decomposed group, a regression line was obtained, with y = 1.1599x + 31.239; the correlation coefficient was 0.5701 (Figure 2). In the partially decomposed (Figure 3) and non-decomposed groups, a correlation between PMI and thiosulfate concentration was not observed.

Figure 1.

Correlation between PMI and TS conc. in all samples. PMI, postmortem interval; TS conc., thiosulfate concentration.

Figure 2.

Correlation between PMI and TS conc. in the severely decomposed group. PMI, postmortem interval; TS conc., thiosulfate concentration.

Figure 3.

Correlation between PMI and TS conc. in the partially decomposed group. PMI, postmortem interval; TS conc., thiosulfate concentration.

The blood thiosulfate concentrations of healthy subjects have been reported to be extremely low. In a previous report by Kage et al. [1], the blood thiosulfate concentrations of 12 healthy subjects were below the detection limit (3 μmol/L) of the GC–MS method. Togawa et al. [7] reported an average of 0.61 (0.537–0.690) μmol/L thiosulfate in the serum of 5 healthy subjects, measured by high-performance liquid chromatography. Additionally, Kawanishi et al. [6] reported that the plasma thiosulfate concentrations of 5 healthy subjects were 2.41–3.04 (average: 2.68) μmol/L. These reports indicate that the blood thiosulfate concentration in healthy subjects was extremely low and did not exceed 4 μmol/L. The blood thiosulfate concentration of the non-decomposed group in this study was 0.1–3.6 (average: 1.1) μmol/L, and no cases had a concentration of >4 μmol/L. Therefore, cases with no evidence of decomposition would be expected to present low values, similar to those of healthy, living subjects. In the partially decomposed group, the values exceeded 4 μmol/L in 12/19 cases; in the severely decomposed group, all cases exceeded 4 μmol/L despite not being associated with hydrogen sulfide poisoning and therefore expected to present concentration of <4 μmol/L. From these results, it is clear that thiosulfate concentrations increase through postmortem decomposition, regardless of hydrogen sulfide poisoning. It is difficult to specify the timing of the postmortem increase in thiosulfate concentration; however, it is believed to rise when parts of the specimens decompose, with severely decomposed specimens undoubtedly generating thiosulfate postmortem.

There are many reports regarding thiosulfate levels in blood samples collected from cadavers with hydrogen sulfide poisoning. Maebashi et al. [14] reported thiosulfate concentrations ranging from 14 μmol/L to 648 μmol/L in the blood of 17 cadavers with hydrogen sulfide poisoning-related deaths. Kage et al. [15,16,17] reported 25–220 μmol/L thiosulfate concentrations in the blood of cadavers who died of hydrogen sulfide poisoning in factory effluent, underground wastewater tanks, and volcanic gases backflowing into a geothermal power plant. Moretti et al. [18] reported that the concentration of thiosulfate in the cardiac blood of four people who died of hydrogen sulfide poisoning in the same tank ranged from 18.4 μmol/L to 35.1 μmol/L. The results of these studies suggest that most cases of death from hydrogen sulfide poisoning involve blood thiosulfate concentrations of ≥14 μmol/L.

Miyazato et al. [19] measured thiosulfate concentrations in the organs and bodily fluids of four cadavers with hydrogen sulfide poisoning-related deaths, two decomposed cases with deaths from causes other than hydrogen sulfide poisoning, and two cases with no evidence of decomposition. In the two obviously decomposed cases, thiosulfate concentrations in the left and right heart blood were 87.7 μmol/L and 81.8 μmol/L at 7 days postmortem and 38.0 μmol/L and 29.0 μmol/L at 12 days postmortem, respectively. This study was the first to report that thiosulfate concentrations increased to the same level as observed in cases of hydrogen sulfide poisoning [19].

In this study, the blood thiosulfate concentration in the severely decomposed group (n = 19) without hydrogen sulfide poisoning ranged from 10.5 μmol/L to 266.6 μmol/L (Table 1), which was 3–66 times higher than the blood thiosulfate concentration in the non-decomposed group. In 17/19 cases in the severely decomposed group, the blood thiosulfate concentrations were similar to those observed for death due to hydrogen sulfide poisoning (>14 μmol/L; Table 2). Although the level of thiosulfate in blood is considered a useful diagnostic indicator of death by hydrogen sulfide poisoning, the significant elevation of thiosulfate due to severe decomposition in this study indicates that it is difficult to diagnose hydrogen sulfide poisoning based on blood thiosulfate concentration in cases with advanced stages of decomposition.

In 7/19 cases in the partially decomposed group had normal values, and 8 cases had values as high as those observed for hydrogen sulfide poisoning-related deaths; in particular, 2 cases exceeded the threshold (14 μmol/L) for death by poisoning within about 2 days after death. There are no clear objective findings to determine whether level of thiosulfate increases due to postmortem decomposition. Therefore, if a case exhibits even partially decomposed findings, it is necessary to consider the possibility that the blood thiosulfate level may have increased due to postmortem, and make an informed judgment.

Thiosulfate levels were <4 μmol/L in all the cases in the non-decomposed group (Table 2). However, Miyazato et al. [19] reported that among two non-decomposed controls (not hydrogen sulfide poisoning), thiosulfate concentrations in the left and right heart blood were 9.0 μmol/L and 11.0 μmol/L with a 24-h PMI and 1.6 μmol/L and 0.8 μmol/L with a 29-h PMI, respectively. Despite there being a control case in the non-decomposed group, one case exceeded the normal value of 4 μmol/L for healthy subjects. Although the thiosulfate concentration did not exceed 14 μmol/L—a criterion for the diagnosis of hydrogen sulfide poisoning—it should be considered that mild elevation of thiosulfate may be observed in some postmortem environments with short PMIs (24 h).

A positive correlation between PMI and thiosulfate concentration was observed on a plot of all samples (Figure 1). In the severely decomposed group, a correlation was observed (Figure 2); however, no correlation was observed in the partially decomposed group (Figure 3). In the severely decomposed group, the degree of decomposition increased due to PMI, and the thiosulfate concentration was also considered to have increased. In the partially decomposed group, the lack of correlation may be due to the length of the PMI being shorter than that of the severely decomposed group. In addition, the individual differences due to the cause of death and the environment of the cadaver place were large. Further study is needed to clarify this issue.

Thiosulfate was found to be produced in the blood by postmortem decomposition. The elevation of blood thiosulfate concentration caused by postmortem decomposition is often as high as observed for death by hydrogen sulfide poisoning, which may have a significant impact on its diagnosis. Therefore, when measuring thiosulfate as an indicator of hydrogen sulfide poisoning, the effect of decomposition should be carefully considered.