ABSTRACT
Objectives: The aim of this study was to determine the predictors of mortality and describe laboratory trends among adults with confirmed COVID-19. Methods: The medical records of adult patients admitted to a referral hospital with COVID-19 were retrospectively reviewed. Demographic and clinical characteristics, and laboratory parameters, were compared between survivors and non-survivors. Predictors of mortality were determined by multivariate analysis. Mean laboratory values were plotted across illness duration. Results: Of 1215 patients, 203 (16.7%) had mild, 488 (40.2%) moderate, 183 (15.1%) severe, and 341 (28.1%) critical COVID-19 on admission. In-hospital mortality was 18.2% (0% mild, 6.1% moderate, 15.8% severe, 47.5% critical). Predictors of mortality were age ≥ 60 years, COPD, qSOFA score ≥ 2, WBC > 10 × 109/L, absolute lymphocyte count < 1000, neutrophil ≥ 70%, PaO2/FiO2 ratio ≤ 200, eGFR < 90 mL/min/1.73 m2, LDH > 600 U/L, and CRP > 12 mg/L. Non-survivors exhibited an increase in LDH and decreases in PaO2/FiO2 ratio and eGFR during the 2nd-3rd week of illness. Conclusion: The overall mortality rate was high. Predictors of mortality were similar to those of other reports globally. Marked inflammation and worsening pulmonary and renal function were evident among non-survivors by the 2nd-3rd week of illness.
ABSTRACT
Objective: This study aimed to describe community-acquired bacterial coinfection (CAI) and antimicrobial use among COVID-19 patients. Methods: Electronic records were retrospectively reviewed, and clinical data, laboratory data, antibiotic use, and outcomes of patients with and without CAI were compared. Results: Of 1116 patients, 55.1% received antibiotics within 48 hours, but only 66 (5.9%) had documented CAI, mainly respiratory (40/66, 60.6%). Patients with CAI were more likely to present with myalgia (p = 0.02), nausea/vomiting (p = 0.014), altered sensorium (p = 0.007), have a qSOFA ≥ 2 (p = 0.016), or require vasopressor support (p < 0.0001). Patients with CAI also had higher median WBC count (10 vs 7.6 cells/mm3), and higher levels of procalcitonin (0.55 vs 0.13, p = 0.0003) and ferritin (872 vs 550, p = 0.028). Blood cultures were drawn for almost half of the patients (519, 46.5%) but were positive in only a few cases (30/519, 5.8%). Prescribing frequency was highest at the start and declined only slightly over time. The mortality of those with CAI (48.5%) was higher compared with those without CAI (14.3%). Conclusion: Overall CAI rate was low (5.9%) and antimicrobial use disproportionately high (55.0%), varying little over time. The mortality rate of coinfected patients was high. Certain parameters can be used to better identify those with CAI and those who need blood cultures.
ABSTRACT
Objectives: To describe the clinical profile and outcomes of hospitalized patients with coronavirus disease 2019 (COVID-19) across the spectrum of disease severity. Methods: This retrospective study included adult patients with confirmed COVID-19 infection admitted to a referral hospital. Descriptive statistics, tests for trend, Kaplan-Meier curve and log-rank test were used to compare characteristics and outcomes across disease severity categories. Results: Of 1500 patients with COVID-19, 14.8% were asymptomatic, 13.5% had mild disease, 36.6% had moderate disease, 12.3% had severe disease and 22.7% had critical disease. Asymptomatic patients were admitted for a concurrent condition or for isolation. Patients aged >60 years, male gender and with co-morbidities had more severe disease. Fever, cough, shortness of breath, malaise, gastrointestinal symptoms and decreased sensorium were more common in patients with severe disease. Bilateral pulmonary infiltrates were common (51.1%), with sicker patients having more abnormal findings. The overall mortality rate was 15.1%. Adopting a symptom-based strategy reduced the length of hospitalization from a median of 13 [interquartile range (IQR) 7-21] days to 9 (IQR 5-14) days. Conclusion: The clinical profile and outcomes for this cohort of patients with COVID-19 was consistent with published reports. Asymptomatic infection was common, and universal testing may be a valuable strategy in the correct context, given the implications for infection control. A symptom-based strategy was found to reduce the length of hospitalization considerably.