983: PERIPHERAL VASOPRESSOR INFUSION FOR SHOCK: A QUALITY ASSESSMENT PROJECT IN A COMMUNITY ICU

Abstract

Introduction: Vasopressor medications in shock are generally administered via central line but placement requires technical proficiency and in community ICU/ED settings, provider availability is not guaranteed around the clock. Evaluating the feasibility of peripheral vasopressor use in a real-world community hospital-based ICU allows us to explore alternative options. The primary objective was to assess the incidence of infiltration, skin necrosis, or limb loss resulting from peripheral vasopressor administration. A secondary objective was to assess the size and location of peripheral intravenous lines and the need for secondary central IV accesses. Institutional IRB approval was obtained. Methods: EMR records for all patients admitted with shock (any type) to a 12-bed community medical ICU receiving vasopressor support during 2018 were individually chart reviewed by a staff intensivist. The local sentinel event reporting system, and physician and nursing documentation were reviewed for any incidence of IV extravasation, or limb damage/loss. Pharmacy records of phentolamine/nitroglycerin paste orders ensured all cases of extravasation were captured. Results: There were 136 unique admissions requiring vasopressor support in 2018. 37.5 % of patients had central lines placed already while 44.9% of patients had vasopressors infused via peripheral IV. Of the 61 patients with peripheral IVs, 30 (49.2%) required placement of a central line. Vasopressors used included phenylephrine (47.1%), norepinephrine (41.9%), vasopressin (7.4%), dobutamine (2.9%) and epinephrine (0.7%). The average duration of vasopressor infusion through peripheral access was 592 mins. There were 8 episodes of infiltration amongst the peripheral IV group, with no episodes of skin damage or necrosis. The median IV size was 20 gauge, and the incidence of infiltration showed no significant association with gauge size (95%CI: -1.4, 0.61; P = 0.43). Conclusions: This dataset suggests that cutaneous infiltration of vasoactive medications is infrequent. Within this patient sample topical nitroglycerine or IV phentolamine, an antidote for cutaneous, drug-induced vasoconstriction, was rarely required. Gauge size in peripheral vasopressor administration appeared to make no difference.