Pediatrics · Medical Literature
Randomized controlled, double blind trial of use of vancomycin catheter-lock to prevent Gram-positive bloodstream infection in neonates.
Abstract
Central lines are more frequently used in infant newborns; they offer advantages but are related to traumatic complications, obstruction and infection. Antibiotics in catheter locks may decrease infections. The aim of this work was to evaluate the efficacy of vancomycin catheter lock to prevent Gram-positive bloodstream infection. Randomized controlled doubled blind trial in newborn with central catheter in first 48 hours. Parents informed consent was obtained. Two groups were formed: experimental group with vancomycin catheter lock twice a day for 30 minutes and control group with normal saline solution catheter lock. Primary outcome was Gram-positive bacterial bloodstream infection. Trial was approved by the project review committees of the National Institute of Pediatrics. Sixty-one patients were included, with mean gestational age of 34.9 weeks of gestation and 2,044 g of weight. Mean stay of catheter was 14.6 days and parenteral nutrition 12 days. It was lower suspicion of sepsis in experimental group (25 vs. 34.5%), fewer bloodstream infections (3.1 vs. 13.8%) and lower infection density per 1,000 catheter/days (2.3 vs. 4.6), but none of these were statistically significative. Vancomycin catheter-lock diminished bloodstream infections at 25% respect control group and 50% density infection per 1000 catheter/days, without being statistically significant.
Clinical relevance
Does twice-daily vancomycin catheter-lock therapy (30-minute dwell time) reduce Gram-positive bloodstream infection in neonates with central venous catheters compared with normal saline catheter-lock? __SNAPSHOT__ {"studyDesign":"Randomized controlled, double-blind trial","setting":"National Institute of Pediatrics, Mexico (single-center tertiary pediatric institution); study approved by institutional project review committees","sampleSize":"n=61 total; experimental group n=32 (vancomycin lock), control group n=29 (normal saline lock) — group sizes inferred from reported percentages (calculated)","population":"Neonates with a newly placed central venous catheter, enrolled within the first 48 hours of catheter insertion; mean gestational age 34.9 weeks; mean birth weight 2,044 g; mean catheter dwell time 14.6 days; mean parenteral nutrition duration 12 days","intervention":"Vancomycin catheter-lock solution instilled twice daily for 30 minutes per dwell","comparator":"Normal saline catheter-lock solution (same frequency and dwell duration)","primaryOutcome":"Gram-positive bacterial bloodstream infection (confirmed infection; suspected sepsis and infection density per 1,000 catheter-days reported as secondary or co-primary measures)","followUp":"Duration of catheter placement (mean 14.6 days); longer-term follow-up not reported","fundingSource":"Not reported"}
Key findings
Among 61 neonates, confirmed bloodstream infection occurred in 3.1% of the vancomycin-lock group vs. 13.8% of the saline-lock group — an absolute risk reduction of 10.7 percentage points (calculated) and a relative risk reduction of approximately 78% (calculated). Suspected sepsis was also lower in the experimental group (25% vs. 34.5%), and infection density per 1,000 catheter-days was halved (2.3 vs. 4.6). However, none of these differences reached statistical significance. Confidence intervals, p-values, and NNT are not reported in the abstract; an NNT of approximately 9 can be calculated from the absolute risk reduction (calculated), but must be interpreted cautiously given non-significant results. No adverse event data are reported in the abstract.
Practice implications
This trial demonstrated numerically consistent trends favoring vancomycin catheter-lock across all infection-related endpoints, but none achieved statistical significance — most likely because the study was substantially underpowered. With only 61 participants, the trial was too small to reliably detect even a clinically meaningful difference in a relatively rare binary outcome. The findings are directionally consistent with prior, larger RCTs in neonates — notably the Garland et al. (2005) trial, which demonstrated a statistically significant reduction in catheter-related bloodstream infection with vancomycin-heparin lock in critically ill neonates — but the current study cannot independently confirm or refute efficacy. The population (mean gestational age 34.9 weeks, mean weight 2,044 g) represents predominantly moderate preterm and late preterm neonates; results may not generalize to extremely preterm or very low birth weight neonates who bear the greatest infection burden. Critically, safety data — including potential for asymptomatic hypoglycemia, vancomycin resistance emergence, and systemic absorption — are not addressed in the abstract. Only an abstract was available for this summary; full-text review is necessary before any clinical application. Practice should not change based on this study alone. __LIMITATIONS__ Critically underpowered: n=61 is insufficient to detect statistically significant differences for a low-frequency binary outcome such as confirmed bloodstream infection; type II error is the most likely explanation for null results,All reported differences were non-significant; no p-values or confidence intervals are provided in the abstract, preventing formal assessment of precision,Single-center study conducted at one Mexican pediatric referral institution, limiting generalizability to other NICUs, healthcare systems, and patient populations,Relatively moderate preterm population (mean GA 34.9 weeks); extremely preterm and VLBW neonates — those at highest CRBSI risk — may not be adequately represented,Absence of adverse event reporting in the abstract: vancomycin resistance surveillance, systemic absorption, and hypoglycemia (a known complication of antibiotic-lock dwell) are not described,Blinding integrity is not verifiable from the abstract alone,Vancomycin concentration, formulation, and whether heparin was co-administered are not specified — limiting comparability with prior trials,Infection density per 1,000 catheter-days is reported without denominators, making independent verification impossible,Funding source not reported; potential conflicts of interest cannot be assessed,Only an abstract is available; full-text methodology, randomization sequence, allocation concealment, and statistical analysis plan cannot be evaluated __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ This small double-blind RCT (n=61) found numerically lower rates of Gram-positive bloodstream infection, suspected sepsis, and infection density with twice-daily vancomycin catheter-lock versus saline in neonates, but none of the differences reached statistical significance — most likely due to critical underpowering. The directional findings are biologically plausible and consistent with prior positive trials in this population, but this study alone provides insufficient evidence to support or modify current practice. A larger, adequately powered trial with pre-specified safety endpoints is required before routine adoption of this strategy.