RESUMO
AIM: To determine agreement between neonatal capillary blood gases taken with plastic and glass tubes. METHODS: An agreement study was carried out in a regional tertiary neonatal unit. Inpatient babies ≥1 kg were recruited to the study when parents gave consent. After taking the routine glass capillary tube sample, a plastic tube sample was taken and run if the heel continued to bleed. Successful sample pairs were recorded and analysed against pre-defined acceptable differences. Assessment was also made of differences in failure rates between tube types for each parameter. RESULTS: Twenty-eight babies provided 135 blood gas pairs, of which five pairs were excluded. Successful pairing of results was achieved for pH in 105 valid samples. There were more failed plastic samples than glass, reaching significance for almost all parameters. pH, pO2 and pCO2 showed poor agreement (<80%) between glass and plastic tubes. On limited analysis of one successful blood gas pair per neonate to minimise bias, results remained non-reassuring. CONCLUSION: The findings of this study do not advocate switching from glass to plastic capillary tubes in our Newborn Intensive Care Unit. Further studies are required to assess agreement of glass and plastic capillary tubes for neonatal blood gas sampling.