Between 1 January 1999 and 30 September 2005, 658 (20.2%) of the 3,257 patients admitted to the ICU died. Of the 658 deaths, 86 (13%) had an autopsy. During the study period, our autopsy rates averaged 13% per year (range, 7.7% to 21.2% per year).
Of the 86 patients who underwent an autopsy, 38 (44%) were women and 48 (56%) were men. The mean age was 54 ± 16 years. The mean length of stay (LOS) in the ICU was 9 ± 8 days, and the mean LOS in the hospital was 19 ± 18 days. Twenty-four patients (28%) were surgical patients and 62 (72%) were medical. Of the 24 surgical patients, 10 (42%) underwent thoracotomy for lung or esophageal cancer, 10 (42%) gastrointestinal/hepatobiliary surgery for hepatic or pancreatic cancer, 2 (8%) orthopedic surgery for sarcoma, 1 head and neck cancer surgery, and 1 gynecologic cancer surgery. Of the 62 medical patients, 25 (40%) had undergone hematopoietic stem cell transplantation (HSCT), 18 (29%) had hematologic malignancies (leukemias or lymphomas), and 19 (31%) had solid tumors.
Major missed diagnoses (discordant cases) were noted in 22 patients (26%) (group 1): 12 (54%) patients had class I discrepancies, 7 (32%) had class II discrepancies, and 3 (14%) had both class I and class II discrepancies. Among the 22 discordant cases, 6 had undergone surgery, 6 had hematologic malignancies, 6 had solid tumors, and 4 underwent HSCT.
Opportunistic infections were the most common class I discrepancies, followed by cardiac complications (thrombotic endocarditis, myocardial infarction, and heart failure) (Table ). The opportunistic infections were due to one of several pathogens (viral, fungal, bacterial, and parasitic). The lung was the most commonly infected site, with pneumonia and empyema present in seven patients, followed by central nervous system infections (two patients), gastrointestinal infections (two patients), and widely disseminated disease (two patients). The majority of class II discrepancies were accounted for by cardiopulmonary complications (n = 7) attributed to pulmonary emboli and thrombotic endocarditis (Table ).
Class I and class II discrepancies
Clinical diagnoses were confirmed by autopsy in 49 patients (57%) (group 2). Most of the confirmed diagnoses were due to bacterial or fungal infections. Autopsy was inconclusive in 15 patients (17%) (group 3). Of the 15 patients, 12 (80%) were medical patients and 3 (20%) were surgical. The majority of group 3 patients died of multiple organ failure and systemic inflammatory response of unknown etiology, and no specific cause of death could be discerned on autopsy. The autopsies of these patients showed diffuse alveolar damage in the lung and diffuse non-specific inflammatory response with scaring and fibrosis in other organs, and positive cultures were not obtained.
There were no statistically significant differences in age or gender between the patients who had missed major diagnoses (group 1) and those with autopsy confirmation of premortem clinical diagnoses (group 2) (Table ). However, the patients with no pathologic diagnosis made on autopsy (group 3) had a significantly longer ICU LOS compared to those with autopsy confirmation of premortem clinical diagnoses (p = 0.05). Overall, patients with autopsy confirmation of premortem clinical diagnoses were not significantly different from those with missed diagnoses (p = 0.11).
Characteristics of critically ill cancer patients who underwent autopsy