A minimally invasive autopsy guided by imaging techniques can accurately identify ischemic heart disease, according to a study presented at RSNA 2017 in Chicago.
Autopsies are useful in healthcare quality control to educate students and provide mortality statistics for research funding. But conventional autopsies have declined in recent decades, said Ivo Wagensveld, MSc, a researcher with Erasmus Medical Center in Rotterdam, Netherlands.
“One of the reasons that these autopsy rates are declining is the invasiveness of the procedure and the fear of mutilation of the deceased’s body,” Wagensveld said during RSNA. “Therefore, less invasive autopsy methods might provide a solution and maybe can increase the number of postmortem diagnostics.”
Wagensveld and colleagues developed the following process for a minimally invasive autopsy: whole-body postmortem CT; MRI of the head, thorax and abdomen; and CT-guided biopsies of normal parenchyma and imaging abnormalities. They tested the diagnostic accuracy of this method against conventional, invasive autopsy in 99 adult cadavers—61 men and 38 women with a mean age of 62.5.
Conventional autopsy detected 34 cases of acute myocardial ischemia and 40 cases of chronic ischemia. Postmortem MRI combined with biopsy demonstrated 97 percent sensitivity and 95 percent specificity for detecting acute ischemia and 90 percent sensitivity and 75 percent specificity for identifying chronic ischemia.
“Imaging-based autopsies can potentially serve as an alternative to conventional autopsies, although other pathologies still need to be thoroughly validated,” Wagensveld said. “CT-guided biopsies increased diagnostic value greatly when added to postmortem MRI.”
In addition, the researchers calculated a calcium score and the area under the Receiver Operating Curve (ROC) curve for each case. In this scenario, an area under the curve (AUC) of 1 represents a perfect predictive test, while an AUC of 0.5 is no better than a coin flip.
Wagensveld said calcium score was a “fair diagnostic test” for chronic infarction (AUC of 0.73) but didn’t provide much value in detecting acute infarction (0.61).