Risk factors affecting infiltration of peripheral venous lines in infants.

The influence of 11 variables on the infiltration of peripheral venous lines (PVLs) was evaluated during 151 infusions in patients younger than 1 year of age. Infusions were followed from the time of cannula placement to discontinuation. Fifty-eight percent of PVLs were infiltrated by 36.30 +/- 33.53 hours (mean +/- SD; median 40 hours, range 10 to 187 hours). No difference between infiltrated and noninfiltrated PVLs was noted with regard to patient age, gender, weight, cannula type, cannula gauge, cannula site, infusion device, potassium or dextrose concentration, medications, or rate of solution administration (P greater than 0.05). Infiltration was observed more often in black than in white infants (P = 0.03) and in patients with lower controller solution head heights (P = 0.01). The time to infiltration was decreased significantly for steel verus Teflon cannulas (P = 0.02), for administration of intravenous medication versus no administration of medication (P = 0.03), for peripheral parenteral nutrition solutions compared with 5% or 10% dextrose solutions (P = 0.014), and with increasing cannula gauge (P = 0.05). The time to infiltration did not differ significantly for gravity-controlled versus positive-pressure infusion device delivery (P = 0.51) or for potassium concentrations less than or equal to 20 mEq/L versus greater than 20 mEq/L (P = 0.13). Infusion device occlusion alarms were associated with only 19% of infiltrations. No sloughing of skin or necrosis of tissue occurred related to infiltration.