Objectives Published discrepancy rates between emergency department (ED) and hospital discharge (HD) diagnoses vary widely (from 6.5 to 75.6%). The goal of this study was to determine the extent of diagnostic discrepancy and its impact on length of hospital stay (LOS), up-triage to the intensive care unit (ICU) and in-hospital mortality. Methods A retrospective chart review of adult patients admitted from the ED to a hospitalist service at a tertiary hospital was performed. The ED and HD diagnoses were compared and classified as concordant, discordant, or symptom diagnoses according to predefined criteria. Logistic regression analysis was conducted to examine the associations of diagnostic discordance with in-hospital mortality and up-triage to the ICU. A linear regression model was used for the length of stay. Results Of the 636 patients whose records were reviewed, 418 (217 [51.9%] women, with a mean age of 64.1 years) were included. Overall, 318 patients (76%) had concordant diagnoses, while 91 (21.77%) had discordant diagnoses. Only 9 patients (2.15%) had symptom diagnoses. A discordant diagnosis was associated with increased mortality (OR: 3.64; 95% CI: 1.026–12.91; p=0.045) and up-triage to the ICU (OR: 5.51; 95% CI: 2.43–12.5; p<0.001). The median LOS was significantly greater for patients with discordant diagnoses (7 days) than for those with concordant diagnoses (4.7 days) (p=0.004). Symptom diagnosis did not affect the mortality or ICU up-triage. Conclusions One in five hospitalized patients had discordant HD and admission diagnoses. This diagnostic discrepancy was associated with significant impacts on patient morbidity and mortality.

Objective Bleeding and thrombosis are common complications during Extracorporeal Membrane Oxygenation (ECMO) support for COVID-19 patients. We sought to examine the relationship between inflammatory status, coagulation effects, and observed bleeding and thrombosis in patients receiving venovenous (VV) ECMO for COVID-19 respiratory failure. Study Design Cross-sectional cohort study Settings Quaternary care institution. Patients The study period from April 1, 2020, to January 1, 2021, we included all patients with confirmed COVID-19 who received VV ECMO support. Intervention None. Measurements and Main Results Thirty-two patients were supported with VV ECMO during the study period, and 17 patients (53%) survived to hospital discharge. The ECMO nonsurvivors mean lactate dehydrogenase (LDH) levels were markedly elevated in comparison to survivors (1046 u/L [IQR = 509, 1305] vs 489 u/L [385 658], p = 0.003). Platelet/fibrinogen dysfunction, as reflected by the low Maximum Amplitude (MA) on viscoelastic testing, was worse in nonsurvivors (65.25 mm [60.68, 67.67] vs 74.80 mm [73.10, 78.40], p = 0.01). Time-group interaction for the first seven days of ECMO support, showed significantly lower platelet count in the nonsurvivors (140 k/ul [103, 170] vs 189.5 k/ul [ 146, 315], p < 0.001) and higher D-dimer in (21 μg/mL [13, 21] vs 14 μg/mL [3, 21], p < 0.001) in comparison to the survivors. Finally, we found profound statistically significant correlations between the clinical markers of inflammation and markers of coagulation in the nonsurvivors group. The ECMO nonsurvivors experienced higher rate of bleeding (73.3% vs 35.3%, p = 0.03), digital ischemia (46.7% vs 11.8%, p = 0.02), acute renal failure (60% vs 11.8%, p = 0.01) and bloodstream infection (60% vs 23.5%, p = 0.03). Conclusion The correlation between inflammation and coagulation in the nonsurvivors supported with VV ECMO could indicate dysregulated inflammatory response and worse clinical outcomes.

A 61-year-old male presented with a history of ischemic cardiomyopathy requiring left ventricular assist device (LVAD) implantation 9 months prior to presentation. The patient was on aspirin and warfarin as part of his LVAD management. The patient had chest pain and was found to have non-ST elevation myocardial infarction. Despite being on warfarin, a subtherapeutic international normalized ratio of 1.6 was measured on admission. Lactate dehydrogenase was 694 U/L (12–146) and high-sensitivity troponin peaked at 47,093 ng/L. Left heart catheterization revealed an extensive aortic root thrombus (AT). AT is an uncommon cause of thromboembolic events in long-term LVAD patients. Thrombotic events in LVAD patients are pump thrombosis, cerebrovascular accidents, heparin-induced thrombocytopenia, and rarely, AT. There is no well-described management of such a rare complication. In our report, we suggest a multidisciplinary approach and consideration of conservative treatment of these patients.

No abstract