[Safety of administration of norepinephrine through peripheral vein line in patients with septic shock]

To analysis the risk factors and safety of administration of norepinephrine (NE) via peripheral vein line (PVL) in patients with septic shock.

A single-center retrospective study was conducted. According to the Lanzhou University Second Hospital information system (HIS) and nursing adverse events report cards, patients with septic shock administrated with NE via PVL to correct the hypotension from January 1st 2015 to December 31st, 2019 were enrolled. The patients' general information, placement location of peripheral venous catheter and venousneedle type, characteristics of NE usage and patient general condition when extravasation occurred were collected. The univariate analysis and Logistic regression were used to analyze risk factors associated with extravasation. Also, the receiver operator characteristic curve (ROC curve) was drawn, and the predictive value of risk factors for extravasation was analyzed.

A total of 1 022 cases with NE were enrolled. After a preliminary screening, a total of 910 cases with NE were used to correct low blood pressure, including 116 cases of peripheral venous infusion. The average age was (52.91±18.69) years old, with majority of female (77 cases, 66.4%). Basic diseases were mainly chronic obstructive pulmonary disease (COPD, 100 cases, 86.2%), followed by hypertension (91 cases, 78.4%), coronary heart disease (87 cases, 75.0%), type 2 diabetes (74 cases, 63.8%) respectively, the primary disease was septic shock in 109 cases (94.0%). A total of 147 peripheral venous catheters were inserted, and the most common site of puncture was the forearm [78.9% (116/147)], followed by the hand [12.2% (18/147)] and the median cubital vein [8.8% (13/147)]. 89.9% of the needles were 20 G in diameter, and 75 cases (64.7%) were converted to central venous catheters (CVC) during subsequent treatment due to continuous infusion of NE. Six patients (5.2%) had extravasation, the median time of extravasation was 29 (23-39) hours, and the median time of NE was 23 (11-53) hours, including 2 patients with an infusion concentration of 60 mg/L and 4 patients with an infusion concentration of 120 mg/L.The infusion speed was 0.5-1.0 μg×kg^-1×min^-1, and the average speed of infusion was (0.75±0.04) μg×kg^-1×min^-1 when extravasation. Univariate and binary Logistic regression analysis showed that the risk factors related to the occurrence of extravasation included: (1) patient factors: the presence of basic diseases, hypertension [odds ratio (OR) = 3.11, 95% confidence interval (95%CI) was 3.09-3.12, P = 0.001] and edema (OR = 1.79, 95%CI was 1.32-2.99, P = 0.032). (2) Factors of infusion fluid itself: long-term (> 24 hours) infusion (OR = 2.91, 95%CI was 1.04-5.96, P = 0.040), infusion concentration > 60 mg/L (OR = 1.88, 95%CI was 1.32-3.99, P = 0.024), infusion speed > 0.3 μg×kg^-1×min^-1 (OR = 2.43, 95%CI was 2.38-2.51, P = 0.029) and diameter of needles < 20 G (OR = 3.11, 95%CI was 3.09-3.22, P = 0.033). (3) Medical personnel factors: lack of observation and assessment (OR = 1.09, 95%CI was 1.03-6.77, P = 0.043). The ROC curve analysis showed that: edema, long-term infusion (> 24 hours), infusion rate > 0.3 μg×kg^-1×min^-1 and diameter of needles < 20 G had a certain predictive value for extravasation of NE through peripheral venous infusion in patients with septic shock, the area under ROC curve (AUC) was 0.610, 0.762, 0.672, 0.629, 95%CI was 0.508-0.713, 0.675-0.849, 0.571-0.772, 0.525-0.732, and P values were 0.044, 0.000, 0.002, 0.019, respectively.

Hypertension, edema, long-term infusion (> 24 hours), infusion concentration > 60 mg/L, infusion speed > 0.3 μg×kg^-1×min^-1, diameter of needle < 20 G, and lack of observation and evaluation by medical staff regularly were risk factors affecting the safety of peripheral intravenous NE in patients with septic shock. Peripheral NE should be avoided in the presence of the above risk factors.