Sudden unexplained death in schizophrenia (SUD-SCZ) is not uncommon and its incidence is approximately three times higher than that in the general population. However, diagnosis of SUD-SCZ remains a great challenge in forensic pathology. This study designed a two-phase study to investigate whether three proteins, namely two potassium ion channel proteins (KCNJ3 and KCNAB1) and one spliceosome protein (SF3B3) that were identified in our previous work, could be applied in the postmortem diagnosis of SUD-SCZ. Immunohistochemical staining of the three biomarkers, followed by a rigorous quantitative analysis, was performed on heart specimens from both SUD-SCZ and control groups. A diagnostic software based on the logistic regression formula derived from the test phase data was then constructed. In the test phase, we found that the staining intensities of KCNJ3, KCNAB1, and SF3B3 were all significantly lower in the SUD-SCZ group (n = 20) as compared with the control group that died from non-natural causes (n = 25), with fold-changes being 14.85 (p < 0.001), 4.13 (p = 0.028) and 2.12 (p = 0.048), respectively. Receiver operating characteristic analysis further illustrated that combination of the three biomarkers achieved the optimal diagnostic specificity (92%) and area under the curve (0.886). In the validation phase, the diagnostic software was confirmed to be a promising tool for predicting the risk of SUD-SCZ in authentic cases. Our study provided a valid strategy towards the practical diagnosis of SUD-SCZ by using KCNJ3, KCNAB1, and SF3B3 proteins as diagnostic biomarkers.
Read full abstract