ABSTRACT
OBJECTIVE: Shorter courses of intravenous antibiotics for young infants with urinary tract infection (UTI) have myriad advantages. As practice shifts toward shorter intravenous treatment courses, this study aimed to determine the safety of early intravenous-to-oral antibiotic switch and identify risk factors for bacteraemia with UTI. METHODS: Retrospective audit of infants aged ≤90 days with a positive urine culture at a quaternary paediatric hospital over 4 years (2016-2020). Data were collected from the hospital electronic medical record and laboratory information system. Short-course intravenous antibiotic duration was defined as <48 hours for non-bacteraemic UTI and <7 days for bacteraemic UTI. Multivariate analysis was used to determine patient factors predicting bacteraemia. RESULTS: Among 427 infants with non-bacteraemic UTI, 257 (60.2%) were treated for <48 hours. Clinicians prescribed shorter intravenous courses to infants who were female, aged >30 days, afebrile and those without bacteraemia or cerebrospinal fluid pleocytosis. Treatment failure (30-day UTI recurrence) occurred in 6/451 (1.3%) infants. All had non-bacteraemic UTI and one received <48 hours of intravenous antibiotics. None had serious complications (bacteraemia, meningitis, death). Follow-up audiology occurred in 21/31 (68%) infants with cerebrospinal fluid pleocytosis, and one had sensorineural hearing loss. Bacteraemia occurred in 24/451 (5.3%) infants, with 10 receiving <7 days intravenous antibiotics with no treatment failure. Fever and pyelonephritis were independent predictors of bacteraemia. CONCLUSION: Short-course intravenous antibiotics for <48 hours for young infants with non-bacteraemic UTI should be considered, provided meningitis has been excluded. Treatment failure and serious complications were rare in young infants with UTI.
ABSTRACT
BACKGROUND: Candida species are the most common cause of systemic fungal infections in children. Risk factors for candidemia vary in different patient populations, posing challenges for clinical prediction of infection. We describe the epidemiology and clinical disease of candidemia in children admitted to a tertiary pediatric hospital. METHODS: Retrospective audit of children ≤18 years of age with candidemia at a tertiary pediatric hospital over a 16-year period. RESULTS: There were 139 episodes of candidemia in 124 children. A central venous catheter was present in 94% of episodes, prior antibiotic exposure in 86% and parenteral nutrition in 43%. During the study period, the proportion of candidemia due to non-albicans Candida spp. increased primarily due to a rise in C. krusei. Colonization with Candida spp. in the 30 days before developing candidemia was identified in 40% of episodes and the species was concordant in 60%. Infection at other sites was rare, including pulmonary dissemination (9/38, 24%), renal fungal disease (9/114, 8%), fungal endophthalmitis (8/102, 8%) and hepatosplenic nodules (5/92, 5%). Overall, 8/127 (6%) isolates were fluconazole-resistant (7 C. krusei and 1 C. glabrata) and 7/127 (6%) had intermediate susceptibility to fluconazole. The overall 30-day mortality was 12% and significant risk factors for mortality on multivariate analysis were male sex, liver disease and mucositis. CONCLUSIONS: Our study outlines low rates of disseminated candidiasis and low mortality associated with candidemia at our institution. Additionally, it suggests that prior colonization may be an important risk factor, however, this should be validated in large prospective controlled studies.
