Elsevier

Cardiovascular Pathology

Volume 41, July–August 2019, Pages 24-28
Cardiovascular Pathology

Original Article
Autopsy and clinical discrepancies in patients undergoing extracorporeal membrane oxygenation: a case series,☆☆

https://doi.org/10.1016/j.carpath.2019.03.001Get rights and content

Highlights

  • Extracorporeal life support is used as a salvage procedure to treat refractory cardiopulmonary failure.

  • There are limited data addressing discrepancies between pre- and postmortem findings in patients undergoing extracorporeal membrane oxygenation (ECMO).

  • Major discrepancies were found between premortem and postmortem diagnoses in all patients who underwent ECMO in our analysis.

  • Central ECMO was associated with significantly fewer discrepancies compared to peripheral and conversion ECMO runs.

  • We found significant overall discrepancies between pre- and postmortem findings in patients undergoing ECMO from 2004 through 2015 in our institution

Abstract

Background

Extracorporeal life support is used as a salvage procedure to treat refractory cardiopulmonary failure. There are limited data addressing discrepancies between pre- and postmortem findings in patients undergoing extracorporeal membrane oxygenation (ECMO). We investigated discrepancies between clinical and autopsy findings in patients placed on ECMO to assess in what proportion of patients were there significant cardiovascular or other pathologies present that were not clinically apparent prior to death.

Methodology

After institutional review board approval, a list of deceased ECMO patients who underwent autopsy examination from 2004 through 2015 was obtained from our institutional database. Retrospective analyses of findings on clinical investigations done while patients were on ECMO and findings on autopsy examination were compared and stratified according to modified Goldman Criteria, which classify discrepancies into four grades depending on their impact on patient's management and mortality.

Results

Of 19 patients, 18 patients had venoarterial ECMO (9 central + 5 peripheral + 4 conversions of ECMO type) and 1 patient received venovenous ECMO. Clinically unrecognized findings were found on autopsy in all patients. 56.6% of total discrepancies found were major [class I/II; e.g., myocardial infarction (MI), intracranial bleeding]. All patients had major discrepancies (class I/II) with an average of 4.21 class I discrepancies per patient. Class I discrepancies are findings which could have altered the course of treatment and survival of the patient if recognized premortem. The most common discrepancies were cardiovascular (MI 63.2%, marked cardiac remodeling 42.1%, severe coronary disease 31.6%) in nature across four classes of discrepancies.

Conclusions

We found major discrepancies between premortem and postmortem diagnoses in patients who underwent ECMO. Our findings underscore difficulties in clinically diagnosing events on ECMO as well as the need for enhanced surveillance and better diagnostic techniques in ECMO patients. Further prospective studies are necessary to understand effects of ECMO on major organs.

Introduction

Extracorporeal membrane oxygenation (ECMO, also known as extracorporeal life support) is a form of advanced life support in which venous blood is taken from the patient and pumped through an external artificial circuit to a membrane lung (or oxygenator), where carbon dioxide is removed and oxygen is added to the blood. The blood is then returned to the patient's venous or arterial circulation. Venovenous (VV) ECMO is used to provide gas exchange support for patients with severe respiratory failure refractory to conventional ventilatory support. Venoarterial (VA) ECMO is used to provide both circulatory and gas exchange support for patients with severe cardiac failure refractory to inotropic support. Venous blood from the patient is accessed from the large central veins and returned to the arterial system in the ascending aorta (central VA ECMO) or a peripheral artery (femoral, carotid, or axillary artery, referred to as peripheral VA ECMO). Patients who need salvage ECMO therapy are often very unstable and are generally managed under heavy sedation, at least initially. Developments in cardiopulmonary bypass machinery with newer pump technologies, better cannulation techniques, and improvised management guidelines have resulted in better survival of ECMO patients in recent years [1]. However, there are limited data addressing discrepancies between pre- and postmortem findings in patients undergoing ECMO.

Autopsy remains an important tool for assuring and improving the quality of medical care by monitoring the diagnostic accuracy and treatment of patients. Although the possibility that a given autopsy will reveal important unsuspected diagnoses has decreased over time, it remains sufficiently high that some of the reversible clinical conditions remain undiagnosed when the patient is critically ill and heavily sedated [2]. Early detection and active treatment for these previously missed diagnoses and complications may have a significant impact on the patients' mortality and morbidity. Despite the presence of publications reporting the discrepancies between clinical findings and autopsy reports [2], [3], autopsy rates have been declining worldwide over the past few decades [4], [5].

Previous studies reporting the discrepancy between clinical findings and autopsy reports in pediatric ECMO patients have been published in the literature, while analytical studies in adult ECMO patients were rarely found. The aims of this study were to investigate discrepancies between clinical and autopsy findings in patients placed on ECMO at our institution and to assess the value of autopsy in these patients.

Section snippets

Study design and patient population

This retrospective single-center observational study was conducted on deceased ECMO patients who underwent autopsy examination from 2004 through 2015 in our institution. After institutional review board approval, clinical and autopsy details of these patients were obtained from our institutional ECMO database and the hospital medical records. All adult patients who were admitted on ECMO to the intensive care unit were included in the study. In accordance with local laws, we referred all

Results

Of the 122 patients who received ECMO between 2004 and 2015, 53 (43.4%) adult patients died on ECMO; all of them were referred for autopsy. Nineteen adult patients had autopsy performed. There were 19 patients included in our analysis; 13 of them were males. The mean age of the patients was 47.1 years, and the mean duration of ECMO was 12.5 days. The demographics of our patients are shown in Table 2. Eighteen patients received VA ECMO for a primary cardiac cause, while one received VV ECMO for

Discussion

Our results demonstrate that autopsies are valuable in patients who underwent ECMO as all the patients in our analysis had at least one major discrepancy between the clinical findings and autopsy reports. The major clinical discrepancies were significant, implying that fatal near misses are common and recognition of clinical signs could be challenging in patients who are deeply sedated during ECMO. Patients also had an average of more than seven major critical events (class I or class II

Conclusions

In conclusion, we found major discrepancies between premortem and postmortem diagnoses in patients who underwent ECMO. Our findings highlight the possible difficulties in clinically diagnosing events on ECMO. Low threshold for investigations coupled with improved scrutiny and better diagnostic techniques in ECMO patients might enhance lesser discrepancies and better outcomes in ECMO patients. The use of newer advanced beside investigations and monitoring modalities may help to detect the

References (12)

There are more references available in the full text version of this article.

Ethical approval reference: DSRB 2016/0026.

☆☆

Conflicts of interest: none.

View full text