Bacterial infections in patients with COVID-19: the impact of procalcitonin testing on antibiotics prescription in the real world.

BACKGROUND: Bacterial infections are not prevalent among patients hospitalized with COVID-19, while unnecessary prescription of antibiotics was commonly observed. This study aimed to determine the impact of procalcitonin testing on antibiotics prescription in the real-world setting. METHODS: We performed a territory-wide retrospective cohort study involving all laboratory-confirmed patients hospitalized in public hospitals in Hong Kong in 2020 with COVID-19. We determined the prevalence of bacterial co-infections (documented infections within 72 h of admission) and secondary bacterial infections (infections after 72 h of admission) and antibiotics consumption, and the correlation between procalcitonin testing and antibiotics prescription. RESULTS: The cohort included 8666 patients, with mean age 45.3 ± 19.9 years, 48.5% male, and comorbidities in 26.9%. Among 2688 patients with bacterial cultures performed, 147 (5.5%) had bacterial co-infections, and 222 (8.3%) had secondary bacterial infections. Antibiotics were prescribed for 2773 (32.0%) patients during the hospital admission. Procalcitonin tests were performed for 2543 (29.3%) patients. More patients with procalcitonin testing received antibiotics (65.9% vs. 17.9%, p < 0.001). Procalcitonin testing was associated with 5-fold increased risk of antibiotics prescription after adjusting for confounding variables. At hospital level, procalcitonin testing correlated with antibiotics prescription. Patients with procalcitonin level < 0.5 ng/mL had a lower probability of antibiotics initiation and shorter duration of antibiotics therapy. CONCLUSIONS: Procalcitonin testing was not associated with lower prescription of antibiotics. Patients with low procalcitonin level had lower antibiotics exposure, supporting the use of procalcitonin to exclude bacterial infections aiding early stopping of antibiotics among patients hospitalized with COVID-19.

Risks of post-colonoscopic polypectomy bleeding and thromboembolism with warfarin and direct oral anticoagulants: a population-based analysis.

BACKGROUND: There were limited data on the risk of post-polypectomy bleeding (PPB) in patients on direct oral anticoagulants (DOAC). We aimed to evaluate the PPB and thromboembolic risks among DOAC and warfarin users in a population-based cohort. METHODS: We performed a territory-wide retrospective cohort study involving patients in Hong Kong from 2012 to 2020. Patients who received an oral anticoagulant and had undergone colonoscopy with polypectomy were identified. Propensity-score models with inverse probability of treatment weighting were developed for the warfarin-DOAC and between-DOAC comparisons. The primary outcome was clinically significant delayed PPB, defined as repeat colonoscopy requiring haemostasis within 30 days. The secondary outcomes were 30-day blood transfusion requirement and new thromboembolic event. RESULTS: Apixaban was associated with lower PPB risk than warfarin (adjusted HR (aHR) 0.39, 95% CI 0.24 to 0.63, p<0.001). Dabigatran (aHR 2.23, 95% CI 1.04 to 4.77, adjusted p (ap)=0.035) and rivaroxaban (aHR 2.72, 95% CI 1.35 to 5.48, ap=0.002) were associated with higher PPB risk than apixaban. In subgroup analysis, apixaban was associated with lower PPB risk in patients aged ≥70 years and patients with right-sided colonic polyps.For thromboembolic events, apixaban was associated with lower risk than warfarin (aHR 0.22, 95% CI 0.11 to 0.45, p<0.001). Dabigatran (aHR 2.60, 95% CI 1.06 to 6.41, ap=0.033) and rivaroxaban (aHR 2.96, 95% CI 1.19 to 7.37, ap =0.013) were associated with higher thromboembolic risk than apixaban. CONCLUSIONS: Apixaban was associated with a significantly lower risk of PPB and thromboembolism than warfarin, dabigatran and rivaroxaban, particularly in older patients with right-sided polyps.