Reference values of lithium in postmortem femoral blood

The main recipients of lithium, people diagnosed with bipolar disorder, show an increased mortality in both natural and unnatural causes of death. Based on international data persons diagnosed with bipolar disorder comprise 2.3-9.6% of all suicidal deaths. In cases of suicide among those suffering from bipolar disorder, 17-53% are due to fatal intoxications. Diagnosing fatal intoxications is often challenging, particularly when the reference information needed to interpret the concentration of a drug is lacking or scarce.

The aim of this study was to establish postmortem femoral blood reference concentrations of lithium, providing both fatal and "normal" postmortem concentrations, as well as to investigate the impact of the mode of intoxication and to study the co-detection of lithium and antidepressant drugs in intoxications and controls.

In Sweden, forensic autopsies are performed in unnatural and obscure deaths. This study included all autopsies in which lithium was found during the study period (1992-2010). Lithium was not included in the regular drug screen, but analysed upon request using flame photometry, ion-selective electrode detection or atomic absorption spectrophotometry. Each case was evaluated according to an established strategy, with strict inclusion and exclusion criteria followed by a multi-observer manual review (Fig. 1, Table 1). The cases included were classified as single intoxications (group A), multi-drug intoxications (group B) or controls (group C). The control group only included cases where death by intoxication and antemortem incapacitation by drugs could be ruled out.

During the study period, lithium was found in 124 cases. After application of inclusion and exclusion criteria and the subsequent manual review, 21 cases were classified as group A (n=4), group B, (n=7) and group C (n=10). The femoral blood lithium concentrations in group A (median 2.69mmol/l) and group B (median 2.10mmol/l) were significantly different (p=0.01) compared to group C (median 0.2mmol/l). There were however no statistically significant difference between the concentrations in groups A and B. The most common mode of death in intoxications was acute-on-chronic (n=10), but the impact of chronic use on the fatal blood concentrations could not be evaluated since there was just one case without previous use. There was no difference in the proportion of co-detections of lithium and antidepressants between intoxication cases and controls.