In Reply

In Reply.—I would be pleased to address the important points raised by the authors of both replies to the editorial on the diagnostic and management autopsy (DMA). Let me address each one specifically.

Importantly, please be assured that I am extremely familiar with the autopsy service in a teaching hospital. I certainly agree with the statements by Dr Krywanczyk and Dr Mount that the principal autopsy objective is to determine why a patient has died and that improved quality assurance follows from effective autopsies. In fact, I was privileged to be a part of the panel sponsored by the National Academy of Medicine on diagnostic error, and I strongly supported the statement in the report¹  that the autopsy is critical to disclose diagnostic errors, precisely because of the published information cited by the authors of this letter.^2–4  What the members of the committee did not provide is a potential solution to the problem that the vast majority of next of kin are unwilling to provide consent for a traditional autopsy. Therefore, little information is gained for improved clinical performance because there is no postmortem review of any kind. It is clearly stated in the original editorial⁵  that the DMA is another option for postmortem analysis and not a replacement for the traditional autopsy. The fact that there are some advantages to the DMA over the traditional autopsy in no way implies that the traditional autopsy should not be made available for the next of kin. Having watched the clinical laboratory test menu become increasingly more genetic and costly from the early 1970s until the present, I am firmly convinced that a general pathologist has a much more difficult time understanding the significance of the different alleles in the pharmacogenetic analysis of clopidogrel (Plavix). On the other hand, experts who review the current literature involving hypercoagulability and vascular diseases are much more likely to effectively interpret a CYP2C19 *2/*2 result and understand its significance in deciding among the possible treatment options with clopidogrel, prasugrel, and ticagrelor. Although not stated in the original editorial, it is understood that genetic information may or may not be informative, because the coagulation experts are also well aware of the overwhelming number of variants of unknown significance. Finally, the statement “His wife was pleased to avoid an autopsy” ⁵  was simply a statement of fact. The decedent's wife said she was “relieved” that the medical examiner did not require an autopsy.

With regard to the second reply to the editorial by Dr Wolf, let me also address each point.

The use of the first person in the editorial was chosen because I was reporting on my own personal feelings about the tragic event in the paragraph cited. Of course, the DMA would only be performed with permission of the next of kin, similar to the traditional autopsy. This could have been stated more clearly in the original editorial. The DMA is a review of hospital performance and is not meant to substitute for a required forensic autopsy. The DMA has several characteristics similar to a mortality and morbidity conference. However, the DMA has some elements that are not typical for many or most mortality and morbidity conferences. The DMA is an interdepartmental activity involving primarily experts in the disease that led to the patient's death. The communication of results extends outside the department to all relevant quality-focused leaders within the institution. It is also done, to completion, within a few days of death.

Taken together, we need far more autopsies, and providing an alternative to the traditional autopsy, without removing it, was the intended message. Thank you to the authors of both replies to the editorial.