Artificial intelligence-supported total parenteral nutrition management in neonatal intensive care units: A systematic review of clinical efficacy, safety, and system integration.
Mustafa Alpkan Çiçek, Ercan Tutak, Mesut Dursun et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review evaluated the efficacy, safety, and integration of AI-supported total parenteral nutrition (TPN) management in neonatal intensive care units (NICUs). Thirteen quantitative studies were analyzed, revealing that computerized physician order entry (CPOE) significantly reduced medication errors from 10.8% to 3.2%. The findings suggest favorable outcomes in clinical efficacy and patient safety, although the evidence quality is predominantly moderate to low.
Key findings
- CPOE implementation reduced the PN medication error rate from 10.8% to 3.2%.
- The TPN2.0 transformer model achieved a Pearson R = 0.94 correlation with expert decisions.
- CPOE and rule-based CDSS improved macronutrient target attainment and glycemic control.
Why it matters
The integration of AI in TPN management could enhance clinical outcomes and patient safety in NICUs, addressing critical nutritional needs in vulnerable populations. Understanding these advancements is essential for improving care protocols.
What to keep in mind
The review is based on thirteen studies of predominantly moderate-to-low certainty, which may limit the generalizability of the findings.