Slower Drip, Safer Lines? Study Tests Minimum Infusion Rates in Preterm Infants

In the fragile world of neonatal intensive care, some of the most consequential decisions involve numbers so small they barely register on everyday intuition. Two milliliters per hour is roughly the pace at which a standard intravenous drip would deliver a teaspoon of fluid over forty minutes, yet for the tiniest premature infants that seemingly trivial figure sits at the center of a genuine clinical dilemma. A new exploratory study published in BMC Pediatrics suggests that lowering the minimum infusion rate through peripherally inserted central catheters, the slender lifelines that carry nutrition into the smallest babies, may be safer than many clinicians have assumed, and could help preterm infants hold on to precious intravenous nutrition for longer as they transition to feeding by mouth.

Peripherally inserted central catheters, universally abbreviated as PICCs, are thin, flexible tubes threaded from a peripheral vein until the tip rests near the heart. In extremely low birth weight and very low birth weight infants, babies born weighing under 1,500 grams and often under 1,000 grams, these catheters are the primary vehicle for parenteral nutrition, the glucose, amino acids, lipids, and electrolytes delivered directly into the bloodstream when the immature gut cannot yet absorb enough nourishment. Because these infants are so small, their total daily fluid allowance is astonishingly restricted, sometimes only a few tens of milliliters per kilogram of body weight, which means every milliliter of the infusion must be carefully budgeted.

Herein lies the problem that motivated the research team led by Jian Cao of Shanghai Children’s Hospital and colleagues at the Second People’s Hospital of Kashi in Xinjiang, China. Many neonatal units enforce a conventional minimum infusion rate of 3 milliliters per hour through PICCs, based on the empirical belief that slower flows allow blood, lipid deposits, or precipitated minerals to stagnate inside the catheter lumen and trigger occlusion. But as a preterm infant matures and enteral feeding through the gut increases, the amount of parenteral nutrition the baby actually needs shrinks. If the pump cannot be set below 3 milliliters per hour, the infusion may deliver more fluid than the baby can tolerate, forcing clinicians to stop parenteral nutrition altogether and run the catheter on maintenance fluid instead, a practice known as functional idling. During such idling, the baby loses the nutritional stream that the catheter was placed to provide.

The study, published on 9 October 2026, took the form of a retrospective before-after cohort evaluation at a single center. The researchers compared 64 extremely or very low birth weight infants admitted during 2025, when the unit adopted a minimum infusion rate of 2 milliliters per hour, with a historical control group of 54 comparable infants treated during 2024 under the traditional 3 milliliter per hour threshold. To keep the comparison as clean as possible, the team applied a strict inclusion criterion requiring a confirmed central tip location for every catheter, and both groups followed identical, unified catheter maintenance protocols. Baseline characteristics of the two cohorts were statistically comparable, with no significant differences reported across the measured variables.

The primary safety question was catheter occlusion, the feared complication in which the lumen becomes blocked and the catheter must be removed prematurely. Using Fisher’s exact test, the investigators found no statistically significant difference between the groups: occlusion occurred in 6.2 percent of catheters in the lower-rate group versus 7.4 percent in the conventional group, a risk difference of minus 1.2 percentage points with a 95 percent confidence interval stretching from minus 12.0 to plus 8.7 percentage points. As a sensitivity analysis, the team applied Firth’s penalized logistic regression, a statistical technique designed for small samples and rare events that produces more stable estimates when ordinary logistic regression would struggle, and again found no signal of harm.

Secondary outcomes pointed in the same reassuring direction. Unplanned catheter removal, an indicator of complications severe enough to force the line out early, occurred in 4 of 64 catheters, or 6.2 percent, in the intervention group compared with 8 of 54, or 14.8 percent, in the control group, a difference that did not reach statistical significance with a P value of 0.14. Catheter-related bloodstream infection, the other major safety concern with any central line, was observed in none of the 64 intervention catheters versus 2 of 54, or 3.7 percent, of control catheters, again not statistically significant. The direction of these differences, while not conclusive, offered no evidence that the slower infusion rate endangered the infants.

The exploratory process and outcome measures, however, told a striking story. Functional idling, the proportion of catheters forced into a nutrition-free holding pattern, plummeted from 57.4 percent in the conventional-rate period to 18.8 percent in the lower-rate period, a difference the authors reported as highly significant. The median duration over which parenteral nutrition could be delivered through the PICC lengthened from 11.5 days, with an interquartile range of 7.0 to 16.0 days, to 14.0 days, with an interquartile range of 12.0 to 16.2 days, a difference with a P value of 0.002. Most compellingly for anyone who has watched a premature infant struggle to grow, the daily weight gain velocity rose from 11.99 plus or minus 1.82 grams per kilogram per day to 13.27 plus or minus 1.69 grams per kilogram per day, a statistically significant improvement. The authors are careful to note that these three variables are exploratory and partly determined by the infusion-rate policy itself, meaning the design cannot prove that the rate change directly caused the nutritional gains, but the physiological logic is coherent: allowing the pump to run slower lets clinicians keep delivering nutrition within the baby’s tight fluid budget for more days.

The technical reasoning behind why slower flows might block catheters deserves scrutiny, because the study’s null finding on occlusion challenges a widely held assumption. Catheter occlusion in neonates is typically attributed to fibrin sheaths forming around the catheter tip, blood refluxing into the lumen between flushes, precipitation of calcium and phosphate from the nutrition solution, or lipid deposits from fat emulsions. The traditional thinking held that a minimum flow of 3 milliliters per hour keeps the lumen sufficiently flushed to prevent stagnation. The new data suggest that at 2 milliliters per hour, with the same maintenance and flushing protocols in place, this stagnation effect either does not materialize or is too small to detect in a cohort of this size. It is worth emphasizing that the confidence interval around the occlusion risk difference was wide, spanning roughly 12 percentage points in either direction, so a modest true increase in occlusion risk cannot be excluded by this study alone.

Indeed, the authors are unusually candid about the limits of their evidence. The study was explicitly not powered to establish safety, equivalence, or non-inferiority; it was designed as a hypothesis-generating, exploratory evaluation of a practice change already adopted in many units worldwide, where rates of 1 to 2 milliliters per hour are used despite the absence of rigorous supporting trials. The before-after design introduces its own vulnerabilities, since temporal changes in staffing, patient mix, or co-interventions between 2024 and 2025 could theoretically have influenced outcomes, although the comparable baseline characteristics and unified maintenance protocols mitigate some of these concerns. The single-center setting in Kashi, Xinjiang, also raises the standard question of generalizability to units with different patient populations, formulations, and nursing practices.

What makes this study resonate beyond its modest sample size is the way it illuminates a quiet gap between clinical practice and clinical evidence. Neonatal teams around the world have been lowering minimum infusion rates on physiological intuition, trading a theoretical occlusion risk for the tangible benefit of sustained nutrition during the most growth-critical weeks of life. This exploratory cohort now offers the first structured comparison suggesting that the trade may be favorable, with lower idling rates, longer nutrition delivery, and faster weight gain at no detectable cost in catheter complications. The authors call for an adequately powered prospective non-inferiority trial to confirm the finding, and such a trial would need to enroll hundreds of infants to narrow the confidence intervals enough to rule out a clinically meaningful occlusion increase. Until then, the study stands as a careful, honest data point in a debate that directly touches the growth and survival of the smallest patients in medicine, and a reminder that in neonatal care, even a single milliliter per hour can carry the weight of a scientific question.

Subject of Research: Minimum infusion rate safety for peripherally inserted central catheters in extremely and very low birth weight preterm infants receiving parenteral nutrition

Article Title: Safety of a 2 mL/h minimum infusion rate for peripherally inserted central catheters in preterm infants: an exploratory single-center before-after cohort study

Article References: Cao, J., Liu, H., Zhang, W., Zhang, C., Ge, A., & Li, J. (2026). Safety of a 2 mL/h minimum infusion rate for peripherally inserted central catheters in preterm infants: an exploratory single-center before-after cohort study. BMC Pediatrics. https://doi.org/10.1186/s12887-026-07784-5

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07784-5

Keywords: peripherally inserted central catheter, preterm infants, parenteral nutrition, minimum infusion rate, catheter occlusion, extremely low birth weight, very low birth weight, neonatal intensive care, catheter-related bloodstream infection, weight gain velocity, before-after cohort study, BMC Pediatrics