American Journal of Emergency Medicine 38 (2020) 1778-1781

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

journal homepage:

Safety of push-dose phenylephrine in adult ICU patients

Brian A. Kurish, PharmD ^a, Cesar Alaniz, PharmD ^b, James T. Miller, PharmD ^c, Nicholas Farina, PharmD ^c,?

^a ProMedica Toledo Hospital, Department of Pharmacy, United States of America

^b University of Michigan College of Pharmacy, United States of America

^c Michigan Medicine, Department of Pharmacy, United States of America

Introduction

Vasopressor medications are used to increase blood pressure in hypotensive, intensive care unit (ICU) patients. Typically, vaso- pressors are administered as continuous intravenous (IV) infu- sions. Due to the quick onset and short half-life of vasopressor medications, these infusions are able to be quickly titrated to a tar- get blood pressure. However, the administration of continuous in- fusions can be delayed because of the time required to compound and deliver the medications. Recent evidence even suggests that infusion dead space can delay medication administration by several minutes [1]. Greater duration of hypotension has been as- sociated with increased risk of mortality, acute kidney injury, and myocardial injury in adult ICU patients [2].

The terms “push-dose phenylephrine (PDP)” and “bolus-dose phenylephrine” refer to administration of phenylephrine as an IV push. Phenylephrine vials or syringes can be stored close to pa- tients in medication dispensing cabinets and emergency kits, allowing for a more rapid delivery of the medication to the bedside. PDP has been used in the operating room for many decades [3]. Re- cently, PDP use has expanded to the management of hypotension in the emergency department and the ICU.

Administration of push-dose vasopressors carries potential safety risks. Hypertension may occur if blood pressure is over- corrected. Additionally, phenylephrine may lead to a reflex brady- cardia. Indeed, case reports have identified episodes of cardiac arrest which were attributed to systemic absorption of phenyleph- rine via topical administration [4-6]. Few studies have assessed the safety of PDP use outside of the operating room [7-9]. These safety analyses focused primarily on the aforementioned adverse effects of hypertension and reflex bradycardia in the setting of peri- procedural use of PDP and include few ICU patients. The purpose of this study was to identify the frequency of adverse events and describe changes in hemodynamics following PDP administration in patients in the ICU.

* Corresponding author at: Michigan Medicine, 1111 E. Catherine Street, Victor Vaughan Building Room 301, Ann Arbor, MI 48109, United States of America.

E-mail address: [email protected] (N. Farina).

Methods

Study design

This was a single center retrospective cohort study conducted in a large academic medical center. This institution includes 99 adult ICU beds spanning medical, surgical, trauma/burn, cardiac, cardiothoracic surgery, and neurosurgery ICUs. Pre-packaged phenylephrine (1000 ug/10 mL) syringes are stocked in automated dispensing cabinets in all ICUs. These syringes are able to be removed from automated dispens- ing cabinets via an override function if necessary. During the study pe- riod, there was no institutional protocol in place for administration of PDP. Although PDP may be administered by physicians or nurses at this institution, all administrations of PDP were under the direct super- vision of a physician at bedside. This study was approved by the institu- tional review board.

Patient selection

Data was collected using the electronic medical record to identify doses of PDP administered in an adult ICU between January and June 2017. Use of PDP was organized into discrete PDP episodes. A PDP epi- sode included both the Initial administration of PDP and any subsequent doses of PDP given within a sixty-minute time-frame. Thus, multiple PDP episodes could occur in a single patient. All PDP episodes adminis- tered in adult patients (age >= 18) were included. Episodes in which vital signs were not recorded within a 30-minute period both before and after initial PDP dose were excluded from analysis. Episodes in which other bolus dose vasopressors, such as epinephrine, were used prior to phenylephrine were also excluded.

Data collection

Patient demographic data collected included age, sex, weight, ICU type, and history of heart failure with reduced (b30%) left ventricular ejection fraction (LVEF). Variables collected for each PDP episode in- cluded causes of hypotension, initial PDP dose, administration time, and use of concomitant vasopressor infusion. For each PDP episode, pre- and post-dose blood pressures and heart rates were recorded. Ad- ditional data collected included requirement of vasopressor infusion within 24 h of initial PDP, fluid administration in the 60 min prior to

https://doi.org/10.1016/j.ajem.2020.05.087

0735-6757/(C) 2020

B.A. Kurish et al. / American Journal of Emergency Medicine 38 (2020) 1778-1781 1779

initial PDP, and use of PDP in patients without central IV access. Data was abstracted by two of the study investigators (BK, NF) using a stan- dardized data collection form. Data abstractors were not blinded to

Table 1

Baseline characteristics of patients receiving PDP

N = 155

study hypotheses. An Interrater agreement test was not performed.

Outcomes and statistics

The primary outcome was a composite safety endpoint of new onset hypertension (systolic blood pressure N 180 mmHg or diastolic blood pressure N 110 mmHg), or bradycardia (heart rate b 50 bpm) at first vital signs recorded following initial PDP administration, a new extrav- asation event within 48 h following administration of PDP, and cardiac arrest within 10 min of initial PDP administration. Extravasation events were recorded based on chart review. Secondary outcomes included in- dividual components of the primary composite endpoint and changes in blood pressure. A post-hoc analysis was conducted comparing hemody- namic response in patients receiving phenylephrine at greater and lower doses than the median.

An unpaired Student t-test was used to compare parametric contin- uous variables, and a Mann-Whitney U test was used for nonparametric continuous variables. Categorical variables were compared using either a Chi-Square or Fisher’s Exact test. The association between PDP use and cardiac arrest was assessed using the Naranjo algorithm [10].

Results

Baseline demographics and characteristics

There were 188 PDP episodes occurring in 155 patients (Fig. 1). Pa- tient demographics are displayed in Table 1. Mean patient age was

62.1 (+- 16.6) years and male patients comprised 66.5% of the popula- tion. At least one vasopressor infusion was already being administered at the time of PDP administration in 47 (25%) PDP episodes. In these pa- tients, median vasopressor dose was 0.3 ug/kg/min of Norepinephrine equivalents, although doses ranged from 0.02-1.49 ug/kg/min (2.1-111.9 ug/min). ICU mortality occurred in 45.5% (70/154) of patients.

Table 2 describes characteristics of PDP administration episodes. The most common initial PDP dose administered was 100 ug, which was given in 92 (48.7%) episodes. PDP administrations occurred primarily in the surgical (37%) and medical (32%) ICUs, although all adult ICUs were represented in the cohort. Median time from PDP administration to first documented repeat blood pressure was 5 min (IQR 5 min). Fol- lowing PDP administration, a new vasopressor infusion was initiated in the subsequent 24 h following 80 (42.5%) of episodes. Among these new vasopressor infusion initiations, 41 (51.2%) were started within

Age, mean (STD) 62.1(16.6)

Male, n (%) 125 (66.5)

Weight (kg), median (IQR) 79 (28)

History of Heart Failure with Reduced LVEF, n (%) 18 (9.6)

an hour of PDP administration. Median volume of fluid administered in the hour following PDP administration was 122 mL (IQR 493 mL). Fluid volumes of 500 mL or greater were administered within an hour of PDP administration following 52 episodes (27.6%).

Outcomes

The primary composite safety outcome occurred in 8 (4.3%) PDP ep- isodes (Table 2). Separated into specific adverse events, PDP administra- tion was associated with 3 (1.6%) episodes of hypertension, 2 (1.1%)

episodes of bradycardia, 1 (0.5%) extravasation event, and 2 (1.1%) cases of cardiac arrest. None of the cases of hypertension required inter- vention. One case of bradycardia was treated with 0.5 mg of IV atropine and the other did not require intervention. The case of extravasation re- solved without intervention or lasting harm

The first patient who suffered cardiac arrest following PDP adminis- tration was admitted to the neurosurgery ICU for management of a mas- sive intracranial hemorrhage with Midline shift. The patient became hypotensive and two PDP doses of 100 ug were administered. Despite this intervention, blood pressure continued to fall and the patient went into asystole. The Naranjo algorithm scored this event as 2, indi- cating this was a possible Adverse drug reaction.

The second patient who suffered cardiac arrest following PDP administration had experienced a Prolonged hospitalization for intrahepatic cholangiocarcinoma which was complicated by persistent atrial flutter. The patient acutely decompensated and was transferred to the surgical ICU with concern for septic shock. He received a single PDP dose of 200 ug. Less than 5 min later, the patient developed pulseless electrical activity which was treated with chest compressions and one dose of epinephrine prior to obtaining return of spontaneous circulation. A transthoracic echocardiogram performed 5 days prior to the arrest was notable for a left ventricular ejection fraction of 20%. The Naranjo algorithm scored this event 4, indicating this was a possible adverse drug reaction.

Since the median initial PDP dose administered was 100 ug, patients receiving 100 ug or less of phenylephrine were compared to those who received N100 ug. Results of this comparison are shown in Table 3.

Fig. 1. Patient enrollment criteria.

1780 B.A. Kurish et al. / American Journal of Emergency Medicine 38 (2020) 1778-1781

Table 2

Characteristics of PDP Administration Episodes

Cause of hypotension, n (%)

N = 188

This study also captured potentially serious adverse events associ- ated with PDP administration, particularly cardiac arrest. Current litera- ture linking phenylephrine to cardiac arrest is scarce. Three publications cite a total of five episodes of cardiac arrest associated with topical ad-

Sepsis 25

(13.3)

Cardiac 30

(16.0)

Hypovolemia/Bleeding 29

(15.4)

Procedure 44

(23.4)

Multiple causes 27

(14.4)

Unknown 33

(17.6)

MAP prior to administration (mmHg), median (IQR) 56 (16)

Concurrent vasopressor infusion, n (%) 47 (25)

ministration of phenylephrine in the operating room [4-6]. With the ex- ception of a single case in a 26 year-old male, these cases occurred in pediatric patients. Consequently, it is difficult to quantify the risk associ- ated with PDP use in the ICU based on this literature. In this study, two cases of cardiac arrest occurred within 10 min of a PDP dose. Both epi- sodes were rated as possible Adverse drug reactions after the Naranjo al- gorithm was applied. Unfortunately, no adverse drug reaction scoring tool is equipped to assess a situation such as cardiac arrest as many criteria in the algorithm are unable to be assessed. In the authors’ opin- ion, the first case of cardiac arrest is very unlikely to have been caused by PDP. It is more likely that PDP contributed to the arrest in the second case given the temporal relationship between PDP administration and

Total fluid volume administered in one hour prior to PDP

administration (mL), median (IQR)

Received at least 500 mL of fluid in one hour prior to PDP administration, n(%)

42.5

(149)

26

(13.8)

cardiac arrest. Of note, the patient in this case likely had a low LVEF at the time of cardiac arrest. Phenylephrine may reduce cardiac output in patients with low left ventricular ejection fraction via multiple mecha-

Receipt of blood products in one hour prior to PDP administration, n(%) 7 (3.7) Initial PDP dose (mcg), median (IQR) 100

(100)

Requirement of repeat PDP administration within 10 min, n (%) 54

(28.7%)

Administration without central access, n (%) 78

(41.5%)

ICU at time of PDP Administration, n (%)

Surgical ICU 69

(36.7%)

Medical ICU 60

(31.7%)

Cardiothoracic ICU 17

(9.0%)

Cardiac Care Unit 16

(8.5%)

Neurosurgical ICU 14

(7.4%)

Trauma/Burn ICU 12

(6.4%)

Baseline characteristics between the two groups were similar. There was no difference in the incidence of adverse events (3.0% vs 5.6%, p

= 0.31). The median change in MAP was 4.8 mmHg in the lower dose group compared to 12.6 mmHg in the higher dose group. However, this difference failed to reach statistical significance (p = 0.07).

Discussion

In this study, the incidence of adverse events associated with PDP administration in critically ill adults was b5%. This is similar to previous studies assessing the safety of PDP used in the emergency department [7,8]. The only other PDP study conducted in the ICU identified a much higher frequency of adverse events at 11.6% [9]. However, this study likely overestimated the Adverse event rate of PDP by including an increase in heart rate of 30% as an adverse event [11]. Phenylephrine is unlikely to cause tachycardia due to lack of ? receptor activity. Rates of non-tachycardia related adverse events are similar between the pre- vious study in ICU patients and this current study. Unfortunately, there is little data from the operating room to inform adverse events associ- ated with use of PDP despite frequent use.

Unlike the prior studies of PDP, this analysis assessed for the adverse event of extravasation following PDP administration. Previous literature has demonstrated that risk of extravasation with continuous infusions of phenylephrine is low [12,13]. This study demonstrated similar results with PDP. Despite over 40% of PDP administrations occurring via a pe- ripheral line, only 1 case of extravasation, which resolved without inter- vention or lasting harm, was noted. Peripheral administration of PDP appears to be safe.

nisms, including a reduction in heart rate and increase in afterload [14-16]. In the current study, patients had a history of heart failure with reduced ejection fraction in 18 episodes of PDP use. However, data regarding acute changes in LVEF measured by echocardiography was not routinely collected for all patients. More research is necessary to describe the safety of PDP in patients with reduced LVEF.

Study results indicate that patients who receive higher doses of PDP have greater responses in blood pressure. Acutely hypotensive critically ill patients may benefit from a more aggressive initial PDP dose or from a quick repeat PDP dose should response be inadequate. These results are similar to those described by Swenson and colleagues, who found that PDP doses of N200 ug are more effective at increasing MAP than smaller doses [7]. Although incidence of adverse events were greater when an initial PDP dose N100 ug was given, the low overall adverse event rate in this study limits the ability to drawn strong conclusions from this finding.

Limitations to this study include its retrospective nature and the ex- clusion of a large number of patients due to inadequate documentation. Extravasation events were captured retrospectively via chart review; it is possible that not all extravasation events were reported. Additionally, 25% of patients in this study were receiving a vasopressor infusion at the time of PDP administration. Increases in vasopressor infusion rate at the same time of PDP administration may have confounded the effects of PDP on MAP. Moreover, other institutions may have different practices regarding phenylephrine products and their use, which may limit the external validity of these results. At this institution, phenylephrine was supplied in 10 mL syringes (concentration 100 ug/mL). Although the premixed syringes include precise medication volumes, the error as- sociated with human administration of small aliquots may be substan- tial. One study found that each 1 mL aliquot could deliver anywhere from 0.864 to 1.053 mL of fluid [17]. It is likely that even when a nurse or a physician intended to administer a discrete dose of phenylephrine,

Table 3

PDP dose-response analysis.

	Dose <= 100 ug	Dose N 100 ug	P value
N	98	90
Dose (mcg), median (IQR)	100 (0)	200 (200)	b 0.01
MAP prior to PDP, median (IQR)	55.9 (15.2)	55.7 (16.0)	1.00
Change in MAP (mm Hg), median (IQR)	4.8 (18.6)	12.6 (26.7)	0.07
Repeat PDP administration within 10 min, n (%)	30 (30.6)	24 (26.7)	0.63
Change in HR (bpm), median (IQR)	-1 (8)	-1 (16)	0.91
Hypertension, n (%)	0 (0)	3 (3.3)	0.11
Bradycardia, n (%)	1 (1.0)	1 (1.1)	1.00
Extravasation, n (%)	1 (1.0)	0 (0)	1.00
Cardiac Arrest, n (%)	1 (1.0)	1 (1.1)	1.00
Any Adverse Event, n (%)	3 (3.1)	5 (6.7)	0.48

B.A. Kurish et al. / American Journal of Emergency Medicine 38 (2020) 1778-1781 1781

the dose actually provided to the patient differed. Finally, preload ex- pansion with a fluid challenged was rare prior to use of PDP in this study. Although results of this study indicate that PDP can effectively raise blood pressure without a prior fluid challenge, it is possible that PDP efficacy would have been impacted by a more aggressive fluid chal- lenge prior to PDP administration. Given the retrospective nature of this study, study investigators were unable to determine if patients were assessed for fluid responsiveness prior to PDP administration.

In conclusion, the adverse event rate associated with PDP use is low. The association between use of PDP and cardiac arrest is unclear. How- ever, until more safety data exists use of PDP in patients with a reduced LVEF should be carefully considered.

CRediT authorship contribution statement

Brian A. Kurish: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing - original draft, Writing - review & editing. Cesar Alaniz: Conceptualization, Methodology, Project administration, Writing - review & editing. James T. Miller: Conceptualization, Methodology, Project administra- tion, Writing - review & editing. Nicholas Farina: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project ad- ministration, Writing - original draft, Writing - review & editing.

References

1. Gregerson BG, Larsen AL, Kelly SM, et al. Effect of intravenous infusion dead space on time to drug delivery in infants. Baylor Univ Med Cent Proc. 2018;31(2):168-70 Apr 3.
2. Maheshwari K, Nathanson BH, Munson SH, et al. The relationship between ICU hy- potension and in-hospital mortality and morbidity in septic patients. Intensive Care Med. 2018;44(6):857-67 Jun 1.
3. De Angelis J. Hemodynamic monitoring in the operating room. Resuscitation. 1984; 11(3-4):165-73.
4. Fisher DM, Groudine SB, Hollinger I, Jones J. VI Medical intelligence article New York state guidelines on the topical use of phenylephrine in the operating room. New York. 2000(3):859-64.
5. Kalyanaraman M, Carpenter RL, McGlew MJ, Guertin SR. Cardiopulmonary compro- mise after use of topical and submucosal ?-agonists: possible added complication by the use of ?-blocker therapy. Otolaryngol - Head Neck Surg. 1997;117(1):56-61.
6. Kouzegaran S, Khazaii T, Tanha AS. A case of cardiac arrest after topical phenyleph- rine administration in adenoidectomy surgery. Int J Pediatr. 2015;3(2.1):81-4.
7. Swenson K, Rankin S, Daconti L, et al. Safety of bolus-dose phenylephrine for hypo- tensive emergency department patients. Am J Emerg Med. 2018;36(10):1802-6.
8. Schwartz MB, Ferreira JA, Aaronson PM. The impact of push-dose phenylephrine use on subsequent preload expansion in the ED setting. Am J Emerg Med. 2016;34(12): 2419-22.
9. Rotando A, Picard L, Delibert S, et al. push dose pressors: experience in critically ill patients outside of the operating room. Am J Emerg Med. 2018;37(3):494-8.
10. Naranjo CA, Busto U, Sellers EM, et al. A method for estimating the probability of ad- verse drug reactions. Clin Pharmacol Ther. 1981;30(2):239-45.
11. Farina N. Asking the correct questions of Push dose vasopressors. Am J Emerg Med. 2019;37(8):1599.
12. Cardenas-Garcia J, Schaub KF, Belchikov YG, et al. Safety of peripheral intravenous administration of Vasoactive medication. J Hosp Med. 2015;10(9):581-5.
13. Datar S, Gutierrez E, Schertz A, Vachharajani V. Safety of phenylephrine infusion through peripheral intravenous catheter in the neurological intensive care unit. J In- tensive Care Med. 2018;33(10):589-92.
14. Yamazaki T, Shimada Y, Taenaka N, et al. Circulatory responses to afterloading with phenylephrine in hyperdynamic sepsis. Crit Care Med. 1982;10(7):432-5.
15. Soeding PF, Hoy S, Hoy G, et al. Effect of phenylephrine on the haemodynamic state and cerebral oxygen saturation during anaesthesia in the upright position. Br J Anaesth. 2013;111(2):229-34.
16. Morimatsu H, Ishikawa K, May CN, et al. The systemic and regional hemodynamic effects of phenylephrine in sheep under normal conditions and during early hyperdynamic sepsis. Anesth Analg. 2012;115(2):330-42 Aug.
17. Thobani SUSD. The accuracy and variability of Bolus injections with different sized syringes. Can J Anaesth. 1992;39(2):198-201.