Background Arginine deficiency is well recognised in very preterm infants (VPI) receiving parenteral nutrition (PN) and is associated with major morbidity. Current UK PN amino acid (AA) formulations are AA-P (arginine 8.4 g/100 gAA) and AA-V (arginine 6.2 g/100 gAA). Human milk contains 4 g/100 gAA arginine. We hypothesised that hyperalimentation in PN-dependent VPI would prevent arginine deficiency irrespective of AA formulation.

Aim To compare the plasma arginine levels in VPI (reference range; RR:54–78 micromol/l) randomised to receive control PN (CPN) or hyperalimentation PN (HPN) regimens in two randomised controlled trials (RCT).

Methods Both studies were single centre RCT with HPN containing 30% more protein/energy than CPN. RCT1 (AA-P) and RCT2 (AA-V) recruited infants <29weeks gestation. Actual daily protein/arginine intake was calculated using the nutritional intake data from the second week of life. Plasma AA levels were measured (week 2) if receiving >35% nutrients from PN.

Results Plasma AA levels were obtained at median (IQR) postnatal age 9 (8–10) days in both RCT1 and RCT2. Both studies achieved significantly higher actual daily protein and arginine intakes (Table 1) in HPN compared to CPN infants (p < 0.01). All median essential AA levels exceeded RR. While there was a trend towards higher median plasma arginine levels with HPN regimens (not significant), these remained below RR. Lower PN arginine content (AA-V) in RCT2 resulted in lower arginine intake to RCT1 despite higher protein intakes.

Conclusion These study data demonstrate hyperalimentation does not prevent low plasma arginine levels in VPI. The proportion of arginine in current UK neonatal PN AA formulations is too low.