Validated transport conditions maintain the quality of washed red blood cells.

BACKGROUND: Washing red blood cell (RBC) units prior to transfusion is indicated for certain patients. In the United States, units stored at 1°C-6°C or transported at 1°C-10°C are available for issue up to 24 h, if not used immediately. The washing procedure commonly utilizes room temperature saline resulting in units starting out above the allowed temperature range. This leads to wastage if units are issued and returned too quickly before having a chance to equilibrate in a transport cooler. STUDY DESIGN AND METHODS: Here we performed an experimental study of washed RBC quality comparing "ideal" storage conditions in a blood bank refrigerator to a "real-world" simulation of unit transport, including holding in a transport cooler. Twelve RBC units were washed and allocated evenly into either condition. RESULTS: Measurements at 0, 1, 3, 6, 12, and 24 h post-washing revealed that placement in a transport cooler was associated with higher unit temperature prior to 12 h (p = .013) with a maximum difference of 9.3°C. Despite this difference, several measures of unit quality including extracellular potassium, pH, lactate, and free hemoglobin were indistinguishable between conditions (p = .382, .224, .286, .691, respectively). We selected half of the tested units from our irradiated inventory and confirmed increased potassium leak (p < .001) and accumulation of free hemoglobin (p = .012) in irradiated units. DISCUSSION: Washed units stored under approved transport conditions are acceptable to return to inventory up to 24 h after washing and we provide a prediction interval-based temperature threshold for rejecting these units, permitting reduced waste.

Invasive mold infection of the gastrointestinal tract: A case series of 22 immunocompromised patients from a single academic center.

Invasive mold infection (IMI) of the gastrointestinal (GI) tract is a rare complication in immunocompromised patients that carries a high mortality rate. It is most often described in the setting of disseminated disease. Early diagnosis and treatment are critical in its management, but this is rarely obtained, leading to delayed therapy. To describe the clinical characteristics, treatment and outcomes of this infection, we reviewed all the cases of adult patients with histopathological findings from autopsy or surgical specimens that demonstrated fungal invasion into the GI tract at Stanford Hospital & Clinics from January 1997 to August 2020. Twenty-two patients that met criteria were identified and they were all immunocompromised, either due to their underlying medical conditions or the treatments that they received. The most common underlying disease was hematological malignancies (63.6%) and the most common symptoms were abdominal pain, GI bleeding and diarrhea. A majority of patients (72.7%) had disseminated invasive mold infection, while the rest had isolated GI tract involvement. In 2/3 of our cases, the fungal genus or species was confirmed based on culture or PCR results. Given the very high mortality associated with GI mold infection, this diagnosis should be considered when evaluating immunocompromised patients with concerning GI signs and symptoms. A timely recognition of the infection, prompt initiation of appropriate antifungal therapy as well as surgical intervention if feasible, are key to improve survival from this devastating infection. LAY SUMMARY: Patients with a weakened immune system can suffer from mold infections in the bowel, which are difficult to diagnose and have very high death rate. We examined such cases in our institution in order to learn about their clinical and microbiological features. This study can further improve our understanding of these infections in order to improve patient outcome.

Impact of a deep learning assistant on the histopathologic classification of liver cancer.

Artificial intelligence (AI) algorithms continue to rival human performance on a variety of clinical tasks, while their actual impact on human diagnosticians, when incorporated into clinical workflows, remains relatively unexplored. In this study, we developed a deep learning-based assistant to help pathologists differentiate between two subtypes of primary liver cancer, hepatocellular carcinoma and cholangiocarcinoma, on hematoxylin and eosin-stained whole-slide images (WSI), and evaluated its effect on the diagnostic performance of 11 pathologists with varying levels of expertise. Our model achieved accuracies of 0.885 on a validation set of 26 WSI, and 0.842 on an independent test set of 80 WSI. Although use of the assistant did not change the mean accuracy of the 11 pathologists (p = 0.184, OR = 1.281), it significantly improved the accuracy (p = 0.045, OR = 1.499) of a subset of nine pathologists who fell within well-defined experience levels (GI subspecialists, non-GI subspecialists, and trainees). In the assisted state, model accuracy significantly impacted the diagnostic decisions of all 11 pathologists. As expected, when the model's prediction was correct, assistance significantly improved accuracy (p = 0.000, OR = 4.289), whereas when the model's prediction was incorrect, assistance significantly decreased accuracy (p = 0.000, OR = 0.253), with both effects holding across all pathologist experience levels and case difficulty levels. Our results highlight the challenges of translating AI models into the clinical setting, and emphasize the importance of taking into account potential unintended negative consequences of model assistance when designing and testing medical AI-assistance tools.

Missed diagnosis and misdiagnosis of infectious diseases in hematopoietic cell transplant recipients: an autopsy study.

Hematopoietic cell transplantation (HCT) is potentially curative for patients with hematologic disorders, but carries significant risks of infection-related morbidity and mortality. Infectious diseases are the second most common cause of death in HCT recipients, surpassed only by progression of underlying disease. Many infectious diseases are difficult to diagnose and treat, and may only be first identified by autopsy. However, autopsy rates are decreasing despite their value. The clinical and autopsy records of adult HCT recipients at our center who underwent autopsy between 1 January 2000 and 31 December 2017 were reviewed. Discrepancies between premortem clinical diagnoses and postmortem autopsy diagnoses were evaluated. Of 185 patients who underwent autopsy, 35 patients (18.8%) had a total of 41 missed infections. Five patients (2.7%) had >1 missed infection. Of the 41 missed infections, 18 (43.9%) were viral, 16 (39.0%) were fungal, 5 (12.2%) were bacterial, and 2 (4.9%) were parasitic. According to the Goldman criteria, 31 discrepancies (75.6%) were class I, 5 (12.2%) were class II, 1 (2.4%) was class III, and 4 (9.8%) were class IV. Autopsies of HCT recipients frequently identify clinically significant infectious diseases that were not suspected premortem. Had these infections been suspected, a change in management might have improved patient survival in many of these cases. Autopsy is underutilized and should be performed regularly to help improve infection-related morbidity and mortality. Illustrative cases are presented and the lessons learned from them are also discussed.