A RANDOMIZED CONTROLLED TRIAL: SUPRAPUBIC ASPIRATION VERSUS URINARY CATHETERIZATION IN THE NEONATAL INTENSIVE CARE UNIT

Abstract

Abstract BACKGROUND Urinary tract infection (UTI) is common in infants (10%). Urinary catheterization (UC) and supra-pubic aspiration (SPA) allow for sterile collection of urine. Some studies suggest SPA has a lower contamination rate than UC, however the optimal method of urine collection for culture remains unclear. OBJECTIVES To determine if there is: 1) a lower contamination rate in urine obtained by SPA assisted by bladder ultrasound versus UC in infants admitted to a neonatal intensive care unit (NICU); 2) a difference between procedures in: success rates for obtaining urine, short-term complication rates, and number of attempts required. DESIGN/METHODS A multicentre, unblinded, randomized controlled trial from 04/2013 to 05/2016. All gestational age (GA) infants greater than 72 hours of age investigated for UTI were eligible for randomization. Crossover to the other procedure could occur after 2 hours or 2 failed attempts. Target sample size was 165. Contamination was defined as growth of ≥2 microorganisms (SPA, UC) or growth <104 CFU/ml (UC). Primary analysis was by intention-to-treat. RESULTS Enrolment was stopped for futility. 906 families were approached with 151 providing consent. 50 infants were eligible for randomization with 47 randomized (SPA n=23, UC n=24). Table 1 shows participant demographics. UTI incidence was 13% for SPA, 8% for UC (p=0.67). Crossover rates were high: 56% for SPA; 21% for UC. No urine sample was obtained in 2 participants per group despite attempts. There was no statistically significant difference between the 2 groups in contamination rates (14% SPA, 18% UC, p=1.00). No short-term complications were reported in either group. Prior to crossover, there was no difference in the number of attempts (i.e. up to 2 attempts pre-crossover or to successfully obtain urine) per procedure (p=0.08); however, there was a difference in the success rates per procedure (44% SPA, 75% UC, p=0.04). CONCLUSION This trial found no difference in contamination rates between SPA and UC, while noting the low enrolment and high crossover rate. More procedural training (particularly for SPA) may ensure greater procedural confidence and success. Post-procedural consent could improve enrolment given that equipoise on the optimal urine collection method still exists.