Experimental estimation of postmortem interval using multivariate analysis of proton NMR metabolomic data

Nuclear magnetic resonance (NMR) spectroscopy has recently been applied to metabolic studies. In particular, metabolic profiles of tissues or of the whole body can easily be acquired through multivariate analysis of NMR spectra. The present study investigates metabolic changes after death in rat femoral muscles using pattern recognition of proton NMR spectra. Rats were killed by suffocation, cocaine overdose and induced respiratory failure, and then low molecular weight metabolites extracted using perchlorate from excised tissues were measured using proton NMR. All spectral data were processed and assessed by multivariate analysis to obtain metabolic profiles of the tissues. The results of principal component analysis (PCA) score plots soon after death showed that the metabolic profiles of the tissues differed according to the mode of death. The principal component (PC) scores of the data varied hourly and correlated with postmortem interval. The present results showed that NMR-based metabolic profiling could provide useful information with which to estimate postmortem intervals and causes of death.