Objectives: To describe the predictive utility of the d-dimer assay among patients with the coronavirus disease 2019 syndrome for unprovoked lower extremity deep venous thrombosis. Design: Prospective observational study with retrospective data analysis. Setting: Academic medical center surgical ICU. Patients: Seventy-two intubated patients with critical illness from coronavirus disease 2019. Interventions: Therapeutic anticoagulation after imaging diagnosis of the first three deep venous thrombosis cases was confirmed; therapeutic anticoagulation as prophylaxis thereafter to all subsequent ICU admissions. Measurements and Main Results: Seventy-two patients with severe coronavirus disease 2019 were screened for deep venous thrombosis after ICU admission with 102 duplex ultrasound examinations, with 12 cases (16.7%) of lower extremity deep venous thrombosis identified. There were no differences between groups with respect to age, renal function, or biomarkers except for d-dimer (median, 12,858 ng/mL [interquartile range, 3,176–30,770 ng/mL] for lower extremity deep venous thrombosis vs 2,087 ng/mL [interquartile range, 638–3,735 ng/mL] for no evidence of deep venous thrombosis; p < 0.0001). Clinical screening tools (Wells score and Dutch Primary Care Rule) had no utility. The C-statistic for d-dimer concentration was 0.874 ± 0.065. At the model-predicted cutoff value of 3,000 ng/mL, sensitivity was 100%, specificity was 51.1%, positive predictive value was 21.8%, and negative predictive value was 100%. Conclusions: Lower extremity deep venous thrombosis is prevalent in coronavirus disease 2019 disease and can be present on ICU admission. Screening has been recommended in the context of the pro-inflammatory, hypercoagulable background milieu. d-dimer concentrations are elevated in nearly all coronavirus disease 2019 patients, and the test appears reliable for screening for lower extremity deep venous thrombosis at or above a concentration of 3,000 ng/mL (more than 13-fold above the normal range). Full anticoagulation is indicated if the diagnosis is confirmed, and therapeutic anticoagulation should be considered for prophylaxis, as all coronavirus disease 2019 patients are at increased risk.

We sought to examine the relationship between liver transplant-related total cost, patient outcome, and hospital resource utilization at freestanding children's hospitals. Using the PHIS database, a retrospective study of 374 patients that underwent liver transplantation at 15 freestanding children's hospitals from July 2010 to December 2012 was performed. One-year graft failure and patient mortality rates from July 2010 to December 2012 for each center were also obtained from the SRTR. There was a 5.1-fold difference in median cost (median $146 444, range $59 487-302 058, P<.001) between all centers. A 2.4-fold difference existed in median LOS (median 15 days, range 9-22 days, P<.001) across centers. Median postoperative ICU stay varied from 0 to 7 days (median 4 days, P<.001). Overall, 30-day readmission rate was 55% (31.3%-100%, P<.001). One-year graft failure varied from 0% to 19.1%, with an overall rate of 5.5% (P=.279). One-year patient mortality for all centers was 2.3% (range 0%-11.1%, P=.016). Higher total cost did not correlate with lower readmission rates, patient mortality, graft failure, or any other variable. These data suggest that identifying practice patterns at low-cost centers and implementing them at higher-cost centers may decrease the cost of pediatric liver transplantation without compromising outcomes.

No abstract

No abstract