Cardiac Agents during Neonatal Cardiopulmonary Resuscitation

At birth, 0.1% of term infants and up to 15% of preterm infants receive extensive cardiopulmonary resuscitation (CPR) defined as chest compressions (CCs) with or without epinephrine (adrenaline) [1]. Neonatal CPR is associated with high mortality and severe short- and long-term neurologic sequelae [2, 3]. In a retrospective cohort study, 83% of infants who received delivery room CPR achieved return of spontaneous circulation (ROSC), while 64% survived hospital discharge, of which 25% received 10 or more minutes of CPR [4].

The Neonatal Consensus of Science and Treatment Recommendation (CoSTR) recommends epinephrine, preferably given intravenously (IV), at a dose of 0.01–0.03 mg/kg [5]. Alternatively, it can be given via an endotracheal tube (ET) at a dose of 0.05–0.1 mg/kg; however, this is not the preferred route and is only used if other routes are not feasible. The epinephrine dose should be repeated every 3–5 min during CCs. The 2020 Neonatal CoSTR has added intraosseous (IO) route as an alternative to administer epinephrine and fluids [6]. The 8th edition of Neonatal Resuscitation Program (NRP) from the American Academy of Pediatrics recommends IO epinephrine be administered at a dose of 0.01–0.03 mg/kg [7].

The 8th edition of the NRP simplified the CoSTR epinephrine dose recommendation (0.01–0.03 mg/kg) to an initial IV epinephrine dose of 0.02 mg/kg for educational efficiency [7]. Subsequent epinephrine doses could be increased, but they should not exceed 0.03 mg/kg/dose [7]. These recommendations are extrapolated from adult and pediatric human data and neonatal animal studies, as neonatal human data about the optimal dose and route of epinephrine administration are lacking.

Vasopressin, an antidiuretic hormone, has vasoconstrictive action through V₁-receptor stimulation resulting in pulmonary vasodilation and systemic vasoconstriction. Three neonatal animal studies using transitional and post-transitional models report conflicting results [8‒10]. In pediatric and adult studies [11, 12], vasopressin improved survival when asystole was the cardiac arrest rhythm. Therefore, vasopressin might be an alternative to epinephrine during neonatal resuscitation. The aim of this review was to summarize the current available literature on cardiac agents used during neonatal resuscitation.

PubMed and EMBASE were searched from their inception to March 3, 2023. Reference lists of included studies and review papers were checked to identify further potential papers for review.

Epinephrine is an endogenous catecholamine with a high affinity for α₁, α_2, β₁, and β₂-receptors in vascular and cardiac smooth muscle (Fig. 1) [13]. It causes vasoconstriction via stimulation of α₁-receptors in vascular smooth muscle, while α₂-receptors cause coronary vasoconstriction [14]. Through β₁-receptor stimulation and activation of Gs-adenyl cyclase-cAMP-protein kinase, signaling cascade occurs, increasing heart rate (chronotropy), conduction velocity (dromotropy), contractility (inotropy), and rate of myocardial relaxation (lusitropy) [14]. β₂-receptor stimulation follows the same pathway as β₁-receptors and leads to smooth muscle relaxation and increased contractility in the myocardium. The in vivo effects of epinephrine depend on dose, number of receptors available on target tissues, affinity of receptors, and local target tissue environments.

Through β-adrenergic receptor stimulation, epinephrine increases myocardial oxygen demands, respiratory and metabolic acidosis (common during neonatal asphyxia), and inhibits hemodynamic responses (e.g., aggravated hypertension or tachycardia after ROSC) [15]. Increased myocardial oxygen demand could delay ROSC and potentially lead to brain damage if hypoxia persists [16, 17]. Epinephrine administration has been associated with intracranial hemorrhage in premature infants [16], most likely related to rebound hypertension after ROSC [18]. At high doses, α-adrenergic vasoconstrictive effect can reduce blood flow to the intestinal tract or kidneys, potentially resulting in necrotizing enterocolitis or renal failure [16, 19]. Observational studies reported that preterm infants receiving CCs and epinephrine have significantly lower survival and higher neurodevelopmental impairments, cystic periventricular leukomalacia, bronchopulmonary dysplasia, and grade 3–4 intraventricular hemorrhages, in comparison with infants not receiving CPR [3, 20, 21]. This may likely be due to the overall outcome of requiring CPR rather than the effects of epinephrine itself; however, this cannot be explained through observational studies alone. While epinephrine might be needed during CPR, healthcare providers must be aware of the potential adverse effects with epinephrine administration.

IV epinephrine provides higher bioavailability compared to ET epinephrine and could be administered via the umbilical or peripheral vein. A neonatal piglet study comparing administration of 0.01–0.03 mg/kg IV or ET epinephrine during CPR reported IV-administered epinephrine to be more efficacious than the ET route, suggesting that a higher dose may be required for ET administration [22]. Burchfield et al. [23] randomized asphyxiated neonatal lambs to various IV-epinephrine dosages and determined that, while heart rate was highest following 0.1 mg/kg high-dose epinephrine (p < 0.05), cardiac output and stroke volume were significantly reduced compared to the other doses (p < 0.05). These decreases appear to be directly related to the dose-dependent rise in systemic vascular resistance [23].

The NRP recommends a normal saline flush volume of 3 mL, which was changed from previous recommendations of 0.5–1 mL [7, 24]. This was based on a randomized animal study demonstrating that a larger flush volume of 3 mL/kg following epinephrine administration increases the incidence of ROSC in term lambs [25]. ROSC was achieved in 5/7 lambs treated with 0.03 mg/kg epinephrine administered through a low umbilical venous catheter (UVC) followed with a 1 mL flush, and 9/9 in lambs with a 3 mL/kg flush (∼10–12 mL; p = 0.02) [25]. For a given flush volume (1 mL or 3 mL/kg), 0.03 mg/kg epinephrine resulted in 11-fold higher odds of ROSC compared to 0.01 mg/kg epinephrine (adjusted OR [95% CI] 11.3 [1.1–112.25], p = 0.04) [25]. The use of a high-flush volume (3 mL/kg) for either 0.01 or 0.03 mg/kg epinephrine did not significantly affect odds of ROSC compared to the low-flush volume (1 mL; adjusted OR 4.35 [0.45–42.32]) [25]. Although 0.03 mg/kg epinephrine followed with a 3 mL/kg flush resulted in 100% ROSC, the authors speculated that this appeared to be mostly due to the higher epinephrine dose, rather than the larger flush volume.

A further study using the same near-term perinatal lamb model of asphyxia-induced cardiac arrest compared 1 and 2.5 mL absolute normal saline flush after UVC epinephrine administration [26]. Achievement of ROSC following the first dose of epinephrine was non-significantly higher (p = 0.08) with 2.5 mL flushes (12/15) compared to a 1 mL flush (3/7) [26]. While this study suggests that a higher flush volume may increase the likelihood of ROSC, the control and experimental groups have varying sample sizes; as such, the results should be interpreted with caution. The cumulative epinephrine dose and median time to ROSC were not different between 1 mL and 2.5 mL flushes (p = 0.27 and p = 0.71, respectively) [26]. While a 3 mL flush might achieve faster ROSC, it is concerning that clinical recommendations are based entirely on the findings of a single preclinical animal trial that did not use an absolute flush volume of 3 mL but rather a relative flush volume of 3 mL/kg [25].

The 2020 CoSTR added IO epinephrine as an alternative for IV administration if UVC access is not feasible or unsuccessful [6]. The initial IO dose is the same as the IV dose at 0.01–0.03 mg/kg and repeated every 3–5 min during CCs [6, 7]. The bioavailability of IO epinephrine is not different from that of IV epinephrine [27]. Roberts et al. [28] reported that 7/9 (78%) near-term lambs who received IO epinephrine and 10/12 (83%) receiving IV epinephrine achieved ROSC. Furthermore, plasma epinephrine levels and the time and number of epinephrine doses needed to achieve ROSC were not different between the groups [28]. A systematic review by Scrivens et al. [29] summarized the literature on IO access in the newborn population until 2019 and determined IO equipment should be available on neonatal units and appropriate training should be implemented. Following subgroup analysis of two randomized clinical trials, another systematic review reported no statistically significant interactions between IV or IO route of drug administration and clinical outcomes [30].

In newborns, the recommended IO site is the proximal tibia, which has a mean medullary diameter of 7 mm, providing a narrow margin of error to correctly place an IO needle [31]. There are no comparisons between speed of UVC or IO needle placement during neonatal resuscitation available (Table 1). However during simulation, the average time for IO placement was 46 s faster compared to UVC placement (59 vs. 105 s; p < 0.001) [32]. Furthermore, a prospective surveillance study reported successful IO placement in 75% of neonates on the first attempt, with overall successful IO access achieved in 91% of neonates [33].

Case reports have reported extravasation, limb ischemia, compartment syndrome, fractures, infection (local or osteomyelitis), air or fat embolism, and amputation [31, 34‒37]. However, these complications following IO placement are rare, suggesting that IO is a safe and effective alternative if UVC access is unsuccessful or not feasible.

Epinephrine administered by the ET route at a dose of 0.1 mg/kg (range 0.05–0.1 mg/kg), with a maximum dose of 0.3 mg, is recommended as an alternative when IV access has not yet been established or is not possible [6]. ET epinephrine administration should be followed by several positive-pressure inflations to allow epinephrine to be distributed within the lungs [6]. Once vascular access has been established, epinephrine should be immediately given using the IV or IO dose if heart rate remains <60/min [7].

During neonatal simulation, time to intubate was less than 2 min compared to ∼6 min for UVC placement (Table 2) [43]. In a neonatal lamb model of asphyxia-induced cardiac arrest, Vali et al. [38] reported the median time to ROSC was 4.5 min with ET epinephrine compared to 2 min with IV epinephrine (p = 0.02). The authors speculated that a higher ET epinephrine dose of 0.1 mg/kg might be needed at birth as epinephrine might be diluted by the lung liquid [38]. The main disadvantage of ET epinephrine is lower bioavailability and resultant lower rates of ROSC compared to IV or IO routes.

In 1999, the International Liaison Committee on Resuscitation recommended a lower ET dose between 0.01 and 0.03 mg/kg [44]. However, observational studies reported that newborns receiving this dose were less likely to achieve ROSC and more likely to require subsequent IV epinephrine, resulting in changing the recommendation to the current dose [40, 41]. ET epinephrine appears to be absorbed into the systemic circulation following deposition into the lung mucosa [45]. A cohort study reported a mortality rate of 25% in preterm infants who received ET epinephrine (3 of 12 infants with a gestational age of 25–33 weeks) [39]. Another cohort study reported rates of ROSC of 23% (3/13) and 18% (3/17) in newborns receiving 0.03 mg/kg and 0.05 mg/kg of ET epinephrine in the delivery room alone, respectively (p = 0.71) [40]. The 24 newborns who did not respond to ET epinephrine subsequently received IV epinephrine, of which 71% (17/23) achieved ROSC after one or more IV epinephrine doses [40]. ROSC was achieved in 75% (15/20) of neonates receiving only IV epinephrine (initial dose of 0.01 mg/kg, increasing dosage with subsequent doses) [40]. Barber and Wyckoff analyzed neonatal resuscitative data from the delivery room and determined that only 32% (14 of 44) of neonates receiving only an initial ET epinephrine dose between 0.01 and 0.03 mg/kg achieved ROSC, while the remaining neonates required subsequent rescue IV epinephrine [41]. A rescue dose of 0.01 mg/kg IV epinephrine resulted in successful ROSC in 77% (23 of 30) of newborns after failing to respond to ET epinephrine [41]. A systematic review reported similar survival with ET and IV administration of epinephrine [42].

The NRP does not currently recommend epinephrine administration through a supraglottic airway device, which is an alternative airway if intubation is not feasible or possible [46]. Theoretically, it provides a route for epinephrine administration into the trachea and lungs; however, a potential downside is that as it is only inserted as far as the pharynx, its mechanism of absorption and bioavailability may differ from ET.

There is limited evidence about the efficacy of epinephrine administration via a supraglottic airway device. In a pediatric pig model of asphyxiated cardiac arrest, Chen et al. [47] compared epinephrine administration via ET, injection into the upper end of the supraglottic airway, or with a catheter inserted into the distal supraglottic airway. The highest peak plasma epinephrine levels were measured with either ET epinephrine or distal supraglottic airway administration, while the lowest levels were observed in the group that received an epinephrine injection into the upper end of the supraglottic airway [47]. No study has examined epinephrine administration via supraglottic airway device in a newborn animal model or in newborn infants.

Nasal epinephrine is currently not recommended during neonatal resuscitation, and limited data are available. A study using an adult canine model reported no difference in rates of ROSC between nasal and IV epinephrine, while another reported highest rates of ROSC and coronary perfusion pressure (CPP) with high-dose epinephrine in conjunction with low-dose phentolamine [48, 49].

Only one neonatal study assessed nasal epinephrine during neonatal resuscitation. Songstad et al. [50] randomized 22 asphyxiated near-term lambs to epinephrine given by IV (0.01 mg/kg), ET (0.1 mg/kg), or intranasally (0.1 mg/kg total or 0.05 mg/kg/nostril) compared to 5 mL IV saline during CCs. No significant differences in mean (SD) time to ROSC between IV saline (11.2 [1.2] min), nasal (9.2 [2.2] min), and ET epinephrine (10.3 [2.4] min) (p > 0.05) were reported [50]. Time to ROSC was significantly shorter with IV epinephrine (2.4 [0.4] min; p > 0.05) compared to all other groups [43]. Following one treatment dose, 5/6 lambs administered IV epinephrine achieved ROSC, and 1/5 and 1/6 in the ET and nasal epinephrine groups achieved ROSC, respectively [50]. No lambs achieved ROSC with only IV saline [50]. The rate of increase in heart rate, mean arterial blood pressure, carotid blood flow, and pulmonary blood flow was significantly greater (p < 0.05) with IV epinephrine compared to ET or nasal epinephrine [50]. Lambs who received IV epinephrine also had significantly higher overall plasma troponin 1 levels compared to all other groups, indicating a higher degree of cardiac muscle injury compared to other groups [43]. While nasal epinephrine might be an alternative to recommended routes of administration, no human data are available to suggest this possibility.

Intramuscular (IM) epinephrine injection is the first-line treatment for systemic anaphylaxis, but it is not recommended during neonatal CPR [51]. Berkelhamer et al. [52] administered IM epinephrine (0.1 mg/kg) into the deltoid muscle of four term fetal lambs following 30 s of CCs and reported no significant increase in plasma epinephrine concentrations. Bioavailability of IM epinephrine may be further limited in neonates with severe acidosis or during asystole as CCs may be insufficient in achieving adequate muscle perfusion for circulation of IM-deposited epinephrine [13]. A case report described significant tissue damage at the IM epinephrine injection site [53]. Until further neonatal studies are conducted, the IM route for epinephrine administration during resuscitation should only be used during research studies.

Vasopressin is an antidiuretic hormone with vasoconstrictive action through V₁-receptor stimulation (Fig. 2) [54]. It is postulated to be beneficial through its effects of combined pulmonary vasodilation and systemic vasoconstriction and is not affected by respiratory or metabolic acidosis [55]. The V₁-receptor associates with the stimulatory G_q/11 protein, with an end result of muscular contraction [54]. Vasopressin does not possess β-adrenergic activity, and, as such, does not increase myocardial oxygen consumption [55].

Vasopressin may be an alternative; however, no studies have been conducted in newborns, so data are largely extrapolated from adult and pediatric studies. Asystolic adult cardiac arrest patients randomized to vasopressin rather than epinephrine had significantly higher rates of survival to hospital admission (29% vs. 20%, p = 0.02) and discharge (5% vs. 2%, p = 0.04) [11]. Similarly, Mann et al. [12] reviewed pediatric episodes of CPR over a 3-year period (1997–2000) at a tertiary hospital and identified four children who received 0.4 international units (IU)/kg/dose of IV vasopressin as rescue therapy following unsuccessful conventional CPR techniques. ROSC occurred in 3/4 children following vasopressin administration; 2/4 survived >24 h, and one survived to hospital discharge [12]. A further case series of 8 episodes of refractory cardiac arrest in 2-month- to 5-year-old patients given 15–20 μgram/kg/dose terlipressin (synthetic analog of vasopressin) reported ROSC in 6/8 episodes, and 4/7 patients survived hospital discharge without neurologic injury (one patient died 12 h after ROSC) [56]. These adult and pediatric data suggest that vasopressin might be an alternative during neonatal CPR, which most commonly occurs due to hypoxia and presents with asystole or bradycardia as the cardiac arrest rhythms. However, the optimal dose and route of vasopressin administration for neonatal resuscitation are unknown.

The IO/IV dose of vasopressin for adults is 40 IU, and studies report a comparable bioavailability between IO and IV administration [57‒60]. There is a lack of studies examining ET vasopressin administration, with ET vasopressin and endobronchial (EB) vasopressin each assessed in a single study [61, 62]. In an adult canine model ET vasopressin increased diastolic blood pressure higher and faster compared with ET and EB epinephrine (p < 0.05) and kept it elevated for a longer period of time [61]. Adolescent pigs who received IV vasopressin had higher CPP compared with EB vasopressin, while the EB group had a CPP plateau effect that appeared to last longer than with IV vasopressin [62].

A study using a supraglottic airway device to deliver aerosolized vasopressin (1 IU/kg, diluted to 5 mL) in pediatric Yorkshire pigs reported no apparent changes in hemodynamics following administration [63]. Vasopressin delivered using a supraglottic airway with a catheter inserted into the trachea did report significant increases in mean and diastolic arterial pressure [63]. No studies were found on the use of nasal vasopressin in neonatal resuscitation.

A neonatal piglet model of asphyxia cardiac arrest randomized 65 piglets to low- (0.01 mg/kg) or high-dose (0.03 mg/kg) epinephrine, low- (0.2 IU/kg) or high-dose (0.4 IU/kg) vasopressin, or saline (control) during CPR [8]. High-dose vasopressin had a higher survival rate (n = 9/10 [90%]) compared with control (n = 5/12 [43%]; p = 0.03) and low-dose epinephrine (n = 5/13 [36%]; p = 0.006) [8]. Comparisons of survival between low-dose vasopressin, high-dose epinephrine, and all other groups were not significantly different [8]. Post-resuscitation, there was a significant increase in superior vena cava flow with high-dose vasopressin (p < 0.001) compared to control, high-, and low-dose epinephrine [8]. The epinephrine-resuscitated groups had increased post-resuscitation troponin compared to the vasopressin group [8]. A post-translational piglet model of asphyxia-induced asystolic cardiac arrest randomized 16 piglets to IV vasopressin (0.4 IU/kg) or epinephrine (0.02 mg/kg) [10]. Vasopressin significantly improved post-ROSC survival times compared to the epinephrine group (240 [240–240] versus 65 [30–240] min, respectively, p = 0.02), as well as arterial pressure, cerebral blood oxygen saturation, and heart rate 4 h after ROSC [10].

A perinatal lamb model of asystole cardiac arrest randomized 27 lambs to receive either 0.03 mg/kg epinephrine or 0.4 IU/kg vasopressin administered into a low UVC [9]. ROSC was achieved in seven out of 10 lambs in the epinephrine group with a mean (SD) time to ROSC of 8 (2) min, while three of nine lambs in the vasopressin group achieved ROSC in 13 (6) min [9]. After one dose of epinephrine, five of 10 lambs achieved ROSC, while only one of nine lambs who received vasopressin achieved ROSC following the first dose [9]. p values on incidence and time to ROSC were not reported, so comparisons cannot be made.

Adolescent pigs in cardiac arrest following hemorrhagic shock were administered high-dose epinephrine (0.2 mg/kg epinephrine), vasopressin (0.8 IU/kg), or saline, and reported ROSC in 7/7, 6/7, and 0/4 pigs receiving vasopressin, epinephrine, and placebo, respectively [64]. After ROSC, all pigs who received epinephrine developed metabolic acidosis and died within 60 min, while vasopressin-treated pigs had less acidosis and survived >60 min (p < 0.01) [64]. In an adult pig model of cardiac arrest, repeated administration of vasopressin during prolonged CPR maintained the CPP >20–30 mm Hg threshold needed for successful defibrillation [65]. Repeated vasopressin doses had a prolonged increase in CPP compared to epinephrine and are speculated to support ROSC as significantly more pigs given vasopressin achieved ROSC compared with those given epinephrine (6/6 vs. 0/6, respectively; p < 0.05) [65].

Adult and pediatric data suggest that vasopressin improves outcomes if asystole is the cardiac arrest rhythm. While newborn infants are much more likely to present with asystole than any other pulseless rhythm, there is no evidence about vasopressin in neonatal patients. While neonatal animal data have had conflicting results, there is a pilot randomized cluster trial ongoing comparing vasopressin and epinephrine during neonatal CPR (NCT05738148).

The current neonatal resuscitation guidelines recommend epinephrine be primarily given via the IV or IO route, with the ET route as an alternative if these routes are not feasible or unsuccessful. The IV and IO dose ranges between 0.01 and 0.03 mg/kg, which should be repeated every 3–5 min during CCs. However, the optimal dosing and route of administration of epinephrine remain unknown. There is evidence from adult and pediatric studies that vasopressin might be an alternative to epinephrine; however, the neonatal data are scarce. Neonatal animal studies examining vasopressin during neonatal resuscitation report conflicting results; however, these differences may be a result of differing animal models and methodologies. Improved hemodynamic parameters and post-ROSC rates of survival with vasopressin compared to epinephrine are promising. A human pilot randomized cluster-trial is currently ongoing, examining epinephrine and vasopressin during neonatal resuscitation (NCT05738148).

The authors have no conflicts of interest to declare.

We would like to thank the public for donating money to our funding agencies: MR is a recipient of the University of Alberta Graduate Recruitment Scholarship, Canadian Institutes of Health Research Canada Graduate Scholarships-Master’s program, and Walter H. Johns Graduate Fellowship. GMS is a recipient of Heart and Stroke Foundation Canada Grant-in-Aid G-22-0031980 and a Project Grant from the Canadian Institute of Health Research (479386) to examine vasopressors in the neonatal period. GP and CTR are supported by National Health and Medical Research Council of Australia (NHMRC) Investigator Grants (CR: 1175634 and GP:1173731).

Conception and design: G.M.S., M.R., M.O.R., and P.-Y.C. Drafting of the first draft: M.R. Collection, assembly, analysis, and interpretation of the data; critical revision of the article for important intellectual content; and final approval of the article: G.M.S., P.-Y.C., M.R., M.O.R., G.R.P., and C.T.R.