Impact of antibiotic timing on mortality from Gram-negative bacteraemia in an English district general hospital: the importance of getting it right every time.

OBJECTIVES: There is limited evidence that empirical antimicrobials affect patient-oriented outcomes in Gram-negative bacteraemia. We aimed to establish the impact of effective antibiotics at four consecutive timepoints on 30 day all-cause mortality and length of stay in hospital. METHODS: We performed a multivariable survival analysis on 789 patients with Escherichia coli, Klebsiella spp. and Pseudomonas aeruginosa bacteraemias. Antibiotic choices at the time of the blood culture (BC), the time of medical clerking and 24 and 48 h post-BC were reviewed. RESULTS: Patients that received ineffective empirical antibiotics at the time of the BC had higher risk of mortality before 30 days (HR = 1.68, 95% CI = 1.19-2.38, P = 0.004). Mortality was higher if an ineffective antimicrobial was continued by the clerking doctor (HR = 2.73, 95% CI = 1.58-4.73, P < 0.001) or at 24 h from the BC (HR = 1.83, 95% CI = 1.05-3.20, P = 0.033) when compared with patients who received effective therapy throughout. Hospital-onset infections, 'high inoculum' infections and elevated C-reactive protein, lactate and Charlson comorbidity index were independent predictors of mortality. Effective initial antibiotics did not statistically significantly reduce length of stay in hospital (-2.98 days, 95% CI = -6.08-0.11, P = 0.058). The primary reasons for incorrect treatment were in vitro antimicrobial resistance (48.6%), initial misdiagnosis of infection source (22.7%) and non-adherence to hospital guidelines (15.7%). CONCLUSIONS: Consecutive prescribing decisions affect mortality from Gram-negative bacteraemia.

Severe acute respiratory coronavirus virus 2 (SARS-CoV-2) nosocomial transmission dynamics, a retrospective cohort study of two healthcare-associated coronavirus disease 2019 (COVID-19) clusters in a district hospital in England during March and April 2020.

OBJECTIVE: To understand the transmission dynamics of severe acute respiratory coronavirus virus 2 (SARS-CoV-2) in a hospital outbreak to inform infection control actions. DESIGN: Retrospective cohort study. SETTING: General medical and elderly inpatient wards in a hospital in England. METHODS: Coronavirus disease 2019 (COVID-19) patients were classified as community or healthcare associated by time from admission to onset or positivity using European Centre for Disease Prevention and Control definitions. COVID-19 symptoms were classified as asymptomatic, nonrespiratory, or respiratory. Infectiousness was calculated from 2 days prior to 14 days after symptom onset or positive test. Cases were defined as healthcare-associated COVID-19 when infection was acquired from the wards under investigation. COVID-19 exposures were calculated based on symptoms and bed proximity to an infectious patient. Risk ratios and adjusted odds ratios (aORs) were calculated from univariable and multivariable logistic regression. RESULTS: Of 153 patients, 65 were COVID-19 patients and 45 of these were healthcare-associated cases. Exposure to a COVID-19 patient with respiratory symptoms was associated with healthcare-associated infection irrespective of proximity (aOR, 3.81; 95% CI, 1.6.3-8.87). Nonrespiratory exposure was only significant within 2.5 m (aOR, 5.21; 95% CI, 1.15-23.48). A small increase in risk ratio was observed for exposure to a respiratory patient for >1 day compared to 1 day from 2.04 (95% CI, 0.99-4.22) to 2.36 (95% CI, 1.44-3.88). CONCLUSIONS: Respiratory exposure anywhere within a 4-bed bay was a risk, whereas nonrespiratory exposure required bed distance ≤2.5 m. Standard infection control measures required beds to be >2 m apart. Our findings suggest that this may be insufficient to stop SARS-CoV-2 transmission. We recommend improving cohorting and further studies into bed distance and transmission factors.

Long-term outcome and risk factors for late mortality in Gram-negative bacteraemia: a retrospective cohort study.

OBJECTIVES: The long-term outcomes of patients following Gram-negative bacteraemia (GNB) are poorly understood. Here we describe a cohort of patients with GNB over a 2-year period and determine factors associated with late mortality (death between Days 31 and 365 after detection of bacteraemia). METHODS: This was a single-centre, retrospective, observational cohort study of 789 patients with confirmed Escherichia coli, Klebsiella spp. or Pseudomonas aeruginosa bacteraemia with a follow-up of 1 year. Multivariable survival analysis was used to determine risk factors for late mortality in patients who survived the initial 30-day period of infection. RESULTS: Overall, 1-year all-cause mortality was 36.2%, with 18.1% of patients dying within 30 days and 18.1% of patients suffering late mortality. An adverse antimicrobial resistance profile [hazard ratio (HR) = 1.095 per any additional antimicrobial category, 95% confidence interval (CI) 1.018-1.178; P = 0.014] and infection with P. aeruginosa (HR = 2.08, 95% CI 1.11-3.88; P = 0.022) were independent predictors of late mortality. Other significant factors included Charlson comorbidity index and length of hospitalisation after the index blood culture. CONCLUSION: Patients with GNB have a poor long-term prognosis. Risk factors for greater mortality at 1 year include co-morbidity, length of hospitalisation, and infecting organism and its resistance profile.

Lessons of the month 1: Polytrauma in a geriatric patient resulting in reactivation of cytomegalovirus infection and secondary cold agglutinin disease-induced haemolytic anaemia.

Haemolysis is a rare but serious complication of cytomegalovirus (CMV) infection, described mostly in immunocompromised patients, the pathogenesis of which is yet to be fully elucidated. We describe a case of Coombs-positive haemolytic anaemia in an older Caucasian immunocompetent woman after polytrauma, resulting from suspected reactivation of CMV infection and secondary cold agglutinin disease. Clinicians should consider CMV infection in the differential diagnosis of haemolytic anaemia in immunocompetent older adults who are admitted with significant musculoskeletal trauma. Early treatment with B-cell depletion therapy to halt cold agglutinin production can be life-saving.