Potassium Reference Values in Neonates: Impact of Sampling Method and Clinical Condition

Background and ObjectivesCurrent potassium reference intervals for neonates fail to account for sampling method differences and prematurity-related factors, leading to unnecessary resampling and interventions. We aimed to establish sampling-specific potassium reference intervals for term and preterm neonates using comprehensive electronic health record data. MethodsWe analyzed 195 606 blood gas measurements from 10 290 neonates (2007-2024) at a tertiary neonatal intensive care unit. After ex-cluding values during severely impaired clinical conditions using integrated clinical metadata, we derived reference intervals for venous, arterial, and cap-illary samples. Multivariate analysis identified factors affecting potassium homeostasis. ResultsFrom 55 664 included values, capillary samples showed signif-icantly higher potassium levels than arterial or venous samples. Reference intervals (2.5th-97.5th percentiles) for neonates >7 days: venous [2.6, 5.5] mM, arterial [2.6, 5.9] mM, capillary [3.0, 6.3] mM. Time-matched analysis of 5403 paired samples showed capillary-specific intervals achieved 88% sensitivity and 94% specificity for detecting true hyperkalemia compared to arterial or venous controls. ConclusionCapillary blood gas potassium levels require distinct, higher reference intervals than venous or arterial samples in neonates. Implementa-tion of sampling-specific reference ranges may reduce false-positive results and unnecessary interventions in this vulnerable population. Article SummaryLarge EHR-based study defines sampling-specific neonatal potassium reference intervals; higher capillary ranges reduce false positives and maybe unnecessary interventions. Whats Known on This SubjectExisting neonatal potassium reference intervals often ignore sampling modality and prematurity, contributing to clinical uncertainty and unnecessary repeat testing in neonates. What This Study AddsProvides sampling-specific potassium reference intervals for neonates; capillary samples require distinct, higher ranges, improving discrimination of true hyperkalemia and reducing false positives. Contributors Statement PageRudolf G. Ascherl: Conceptualized and designed the study, coordinated data extraction, perfomed analysis and visualization, drafted and revised the initial manuscript. Benjamin W. Ackermann: Contributed to study design, assisted in revising the manuscript. Matthias Knupfer: Provided clinical oversight, interpreted findings, critically reviewed the manuscript. Equal contribution: Rudolf G. Ascherl and Benjamin W. Ackermann. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.