Drug compatibility in neonatal intensive care units: gaps in knowledge and discordances.

In this work, we reviewed the compatibility data of drug combinations frequently administrated in nine Spanish neonatal intensive care units (NICUs) and analyzed the degree of agreement among three highly used databases (Micromedex, King Guide to Parenteral Admixtures, and Stabilis) through Cohen's kappa coefficient statistical analysis. Among 1945 drug combinations analyzed, 283 were compatible, 421 were potentially compatible, 216 were incompatible, 139 were controversial, and there was no data for 886 combinations. In general, the three databases showed a strong degree of agreement: Micromedex vs. King Guide (κ = 0.746; p < 0.001), King Guide vs. Stabilis (κ = 0.743; p < 0.001), and Micromedex vs. Stabilis (κ = 0.691; p < 0.001). However, in 6 of 648 (Micromedex vs. King Guide), 3 of 357 (King Guide vs. Stabilis), and 32 of 606 (Micromedex vs. Stabilis) comparisons, drug pairs were compatible according to the first database and incompatible according to the second, indicating discordances among databases.Conclusion: There is a gap in knowledge about physical compatibility of a great number of drug combinations commonly used in NICUs. Although the three databases showed strong concordance, for some drug combinations, important discrepancies were found. Thus, there is a need for further studies on drug compatibility to increase safety of intravenous administration. What is Known: • Y site-administration in NICUs is very common and some administration errors are related to the lack of information on the compatibility of intravenous drugs. • Physical compatibility data of drugs frequently used in NICUs is still very limited. What is New: • Physical compatibility data of drug combinations commonly used in Spanish NICUs was reviewed in three highly used admixture databases: Micromedex, King Guide to Parenteral Admixtures and Stabilis, and our results showed a strong degree of agreement between them, however for some drug combinations, important discrepancies were found. • Our results indicated that there is still a large gap in knowledge about physical compatibility of a great number of drug combinations commonly used in NICUs..