Myocardial perfusion and oxidative stress after 21% vs. 100% oxygen ventilation and uninterrupted chest compressions in severely asphyxiated piglets

Resuscitation 106 (2016) 7–13

Contents lists available at ScienceDirect

Resuscitation journal homepage: www.elsevier.com/locate/resuscitation

Experimental paper

Myocardial perfusion and oxidative stress after 21% vs. 100% oxygen ventilation and uninterrupted chest compressions in severely asphyxiated piglets夽 Anne Lee Solevåg a,b,c,∗ , Georg M. Schmölzer a,b , Megan O’Reilly a,b , Min Lu a,b , Tze-Fun Lee a,b , Lisa K. Hornberger b , Britt Nakstad c , Po-Yin Cheung a,b a

Centre for the Studies of Asphyxia and Resuscitation, Neonatal Research Unit, Royal Alexandra Hospital, Edmonton, Canada Department of Pediatrics, University of Alberta, Edmonton, Canada c Department of Pediatric and Adolescent Medicine, Akershus University Hospital, Lørenskog, Norway b

a r t i c l e

i n f o

Article history: Received 24 November 2015 Received in revised form 17 April 2016 Accepted 14 June 2016 Keywords: Newborn Animals Asphyxia neonatorum Cardiopulmonary resuscitation Heart massage Oxygen

a b s t r a c t Aim: Despite the minimal evidence, neonatal resuscitation guidelines recommend using 100% oxygen when chest compressions (CC) are needed. Uninterrupted CC in adult cardiopulmonary resuscitation (CPR) may improve CPR hemodynamics. We aimed to examine 21% oxygen (air) vs. 100% oxygen in 3:1 CC:ventilation (C:V) CPR or continuous CC with asynchronous ventilation (CCaV) in asphyxiated newborn piglets following cardiac arrest. Methods: Piglets (1–3 days old) were progressively asphyxiated until cardiac arrest and randomized to 4 experimental groups (n = 8 each): air and 3:1 C:V CPR, 100% oxygen and 3:1 C:V CPR, air and CCaV, or 100% oxygen and CCaV. Time to return of spontaneous circulation (ROSC), mortality, and clinical and biochemical parameters were compared between groups. We used echocardiography to measure left ventricular (LV) stroke volume at baseline, at 30 min and 4 h after ROSC. Left common carotid artery blood pressure was measured continuously. Results: Time to ROSC (heart rate ≥100 min−1 ) ranged from 75 to 592 s and mortality 50–75%, with no differences between groups. Resuscitation with air was associated with higher LV stroke volume after ROSC and less myocardial oxidative stress compared to 100% oxygen groups. CCaV was associated with lower mean arterial blood pressure after ROSC and higher myocardial lactate than those of 3:1 C:V CPR. Conclusion: In neonatal asphyxia-induced cardiac arrest, using air during CC may reduce myocardial oxidative stress and improve cardiac function compared to 100% oxygen. Although overall recovery may be similar, CCaV may impair tissue perfusion compared to 3:1 C:V CPR. © 2016 Elsevier Ireland Ltd. All rights reserved.

Introduction

Abbreviations: CC, chest compressions; CPR, cardiopulmonary resuscitation; CCaV, continuous CC and asynchronous ventilation; C:V, chest compression to ventilation; ILCOR, International Liaison Committee on Resuscitation; ROSC, return of spontaneous circulation; SpO2 , oxygen saturation; HR, heart rate; paCO2 , partial arterial CO2 ; ETCO2 , end-tidal CO2 ; VT , tidal volume; CO, cardiac output; LVOT, left ventricular outflow tract; PW, pulsed wave; SV, stroke volume; CPAP, continuous positive airway pressure; PEEP, positive end-expiratory pressure; PEA, pulseless electrical activity; PPV, positive pressure ventilation; MMP, matrix metalloproteinase; GSH, glutathione; GSSG, oxidized glutathione. 夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2016.06.014. ∗ Corresponding author at: Neonatal Research Unit, Royal Alexandra Hospital, 10240 Kingsway Avenue NW, T5H 3V9 Edmonton, Alberta, Canada. E-mail address: [email protected] (A.L. Solevåg). http://dx.doi.org/10.1016/j.resuscitation.2016.06.014 0300-9572/© 2016 Elsevier Ireland Ltd. All rights reserved.

Resuscitation of asphyxiated infants with 100% oxygen is associated with increased oxidative stress,1 increased morbidity and mortality.2,3 A meta-analysis reported that using air instead of 100% oxygen reduces mortality by 30%, corresponding to the potential saving of more than 100,000 infant lives annually.2,3 However, available studies examined infants who were supported with assisted ventilation only, without the need of chest compression (CC). The use of air instead of 100% oxygen in neonatal cardiac arrest is debated and the 2015 International Liaison Committee on Resuscitation (ILCOR) guidelines advocate 100% oxygen when CC are needed.4 However, the ILCOR acknowledges the minimal evidence to support this recommendation. Indeed, newborn animal studies indicate that ventilation with air is as effective as 100% oxygen during CC, at least after brief asystole5,6 ILCOR

8

A.L. Solevåg et al. / Resuscitation 106 (2016) 7–13

may permit the inclusion of new data to inform practice if it becomes available before the next complete evidence evaluation is due.7,8 Restoration of myocardial function after hypoxia-ischemia depends on blood oxygenation and myocardial perfusion. In adult animals, performing uninterrupted CC increases coronary perfusion pressure (CPP).9 No data is available to suggest that this may also be the case in newborns. Schmölzer et al.10 found no difference in return of spontaneous circulation (ROSC) or mortality comparing 3:1 C:V CPR and continuous CC with asynchronous ventilation (CCaV) using 100% oxygen in asphyxiated newborn piglets with profound bradycardia. No study has investigated air vs. 100% oxygen during CCaV in neonatal CPR. The aim of this study was to compare air and 100% oxygen combined with 3:1 C:V CPR or CCaV in newborn piglets with severe asphyxia and cardiac arrest. We

hypothesized that CCaV with air would improve ROSC compared to the currently recommended 100% oxygen with 3:1 C:V CPR.

Methods Subjects Newborn mixed breed piglets (1–3 days, 1.7–2.4 kg) were obtained on the day of experimentation from the Swine Research Technology Center, University of Alberta. All experiments were conducted in accordance with the guidelines and approved by the Animal Care and Use Committee (Health Sciences), University of Alberta and presented according to the ARRIVE guidelines.11 The protocol is presented in Fig. 1.

Fig. 1. Experimental protocol.

A.L. Solevåg et al. / Resuscitation 106 (2016) 7–13

Animal preparation The piglets were anesthetized with Isoflurane 1–5%, tracheotomized and mechanically ventilated (Sechrist infant ventilator model IV-100; Sechrist Industries, Anaheim, CA) at a 25 min−1 rate, peak inspiratory pressure (PIP) 25 cm H2 O and positive endexpiratory pressure (PEEP) 5 cm H2 O. After vascular access was obtained, hydration was maintained with intravenous infusions of 5% dextrose 10 mL/kg/h and 0.9% NaCl 2 mL/kg/h, and anesthesia was changed to intravenous morphine 50–200 mcg/kg/h and propofol 0.1–0.2 mg/kg/h. An intravenous bolus of morphine (0.15 mg/kg) and pancuronium (0.1 mg/kg) was given before tracheotomy. From stabilization onwards, muscle relaxants were not used. If signs of pain or distress, the piglets received an intravenous bolus of morphine (0.15 mg/kg) and/or acepromazine (0.25 mg/kg). Piglets were allowed to recover from surgical instrumentation for 1 h during which the ventilator rate and airway pressure were adjusted to keep the partial arterial CO2 (paCO2 ) 35–45 mmHg as determined by end-tidal CO2 (ETCO2 ) and periodic arterial blood gases. Surgical procedures A 5-French ArgyleTM single-lumen catheter (Covidien, Dublin, Ireland) was inserted into the left common carotid artery (CA) for continuous blood pressure monitoring and blood sampling. A 5French ArgyleTM double-lumen catheter (Covidien) was inserted in the external jugular vein on the same side for fluid and medication infusion. The piglet was tracheotomized and a 3.5 uncuffed endotracheal tube was inserted and fixed to the trachea. A real-time ultrasonic flow probe (2SB; Transonic Systems Inc., Ithica, NY) was placed around the right CA. Systemic arterial pressure and heart rate (HR) were continuously measured with a Hewlett Packard 78833B monitor (Hewlett Packard Co., Palo Alto, CA). Respiratory parameters A respiratory function monitor (Respironics, Philips, Andover, MA) continuously measured tidal volume (VT ), ventilation rate and ETCO2 at a sample rate of 200 Hz. The flow sensor was placed between the endotracheal tube and the mechanical ventilator or T-piece. Echocardiography Echocardiography was performed with a Vivid 7 and a 5S probe (GE Healthcare, Buckinghamshire, UK). In the animals subjected to asphyxia and resuscitation, cardiac output (CO) was measured at baseline, and 30 min and 4 h after ROSC. Aortic valve (AoV) diameter was measured at baseline from the right parasternal window. This diameter was used to calculate the valve cross sectional area (CSA, r2 ), which was used for all calculations of CO. A subcostal five-chamber view was used to obtain left ventricular outflow tract (LVOT) flow measurements using pulsed wave (PW) Doppler with an angle of interrogation of <20◦ . CO (mL/kg/min) was calculated off-line on the Vivid 7 from the stroke volume (SV) (velocity time integral averaged from three consecutive PW Doppler signals across the aortic valve (LVOT flow in cm) × CSA (in cm2 )) × HR. Biochemical analyses Lactate was extracted from frozen left ventricular (LV) tissue with 6% perchloric acid/0.5 mM ethylene glycol tetraacetic acid. After adding potassium carbonate, the supernatant was collected for lactate determination by a nicotinamide adenine dinucleotide enzyme-coupled colorimetric assay with spectrophotometry at 340 nm12 (Spectramax 190; Molecular Devices, Sunnyvale, CA). The activity levels of matrix metalloproteinase (MMP)-2 (64 kDa, 72 kDa and 75 kDa) and MMP-9 in LV tissue homogenates were determined by gelatin zymographic analysis as previously

9

described.13 The LV oxidized glutathione (GSSG) and total glutathione (GSH) were analyzed and the ratio of GSSG/GSH was calculated as described previously.14 Experimental protocol (Fig. 1) Experimental animals were randomized to four groups (n = 8 each): Group 1: 3:1 C:V CPR consisting of repeated series of 3 CC (CC rate 90 min−1 ) with an interruption to deliver one inflation with air (ventilation rate 30 min−1 ). Group 2: 3:1 C:V CPR consisting of 3 CC (CC rate 90 min−1 ) with an interruption to deliver one inflation with 100% oxygen (ventilation rate 30 min−1 ). Group 3: CCaV (CC rate 90 min−1 ) with air (ventilation rate 30 min−1 ). Group 4: CCaV (CC rate 90 min−1 ) with 100% oxygen (ventilation rate 30 min−1 ). A sham-operated group underwent the same surgical protocol, stabilization and equivalent experimental periods with neither asphyxia nor resuscitation (n = 4). Asphyxia was induced by reducing FiO2 to 0.08 and reducing the ventilator rate by 10 min−1 every 10 min until a rate of 0 min−1 . After 10 min with PEEP of 5 cm H2 O, the ventilator was disconnected and the endotracheal tube clamped. Pulseless electrical activity (PEA) was defined as a peak flow velocity across the aortic valve of <0.5 m/s (echocardiography), CA flow <5 mL min−1 and no audible heartbeat (auscultation). Once PEA was diagnosed, we waited for 30 s before giving positive pressure ventilation (PPV) with air (PIP 25 cm H2 O/PEEP 5 cm H2 O) for 30 s with a Neopuff T-Piece (Fisher & Paykel, Auckland, NZ). When CC was started after 30 s of PPV, FiO2 was kept at 21% or increased to 100% according to randomization. Manual CC was performed by the same investigator (ML) each time and a metronome was set at 120 min−1 in the 3:1 C:V groups and 90 min−1 in the CCaV groups. We aimed to give PPV with a rate of 30 min−1 , and thus achieve a total of 120 events min−1 in all intervention groups. If there was no ROSC after 30 s of CC, epinephrine (0.02 mg/kg) was given intravenously and repeated every 3 min as needed to a maximum of 4 doses. CPR was discontinued if ROSC was not achieved after 15 min. As previously described,6 ROSC was defined as an unassisted HR ≥ 100 min−1 . After ROSC, FiO2 was adjusted according to SpO2 and paO2 ; and piglets were observed for 4 h before being euthanized with i.v. phenobarbital (100 mg/kg). Randomization and blinding Piglets were randomly allocated to sham-operated or experimental groups (http://www.randomizer.org). To reduce randomization bias a two-step randomization was used (Fig. 1). After surgical procedures and stabilization a sequentially numbered, sealed envelope containing another envelope with “sham” or “intervention” written on the outside was opened (step one). “Intervention” piglets were then subjected to hypoxia and asphyxia. After 15 min of hypoxia, the “intervention” envelope containing one of the four resuscitation allocations was opened by one of the threemember resuscitation team (TFL). The investigator who determined PEA and ROSC (ALS) was blinded to the FiO2 . The 3:1 C:V or CCaV allocation was not revealed to the whole resuscitation team until after PEA had been diagnosed. Off-line measurements of CO were performed in a blinded fashion. Sample size estimates Our primary outcome measure was time to ROSC. Previous observational data showed a mean (standard deviation (SD)) time to ROSC of 180 (25) s in piglets resuscitated with 100% oxygen and 3:1 C:V CPR. We hypothesized that CCaV with air would reduce time to ROSC. A sample size of 32 piglets (8 per group) would then be sufficient to detect a clinically important (33%) reduction in time to ROSC with 80% power and a 2-tailed alpha error of 0.05.

10

A.L. Solevåg et al. / Resuscitation 106 (2016) 7–13

Table 1 Baseline characteristics.

Age (days) Weight (kg) Male/female Heart rate (bpm) MAP (mmHg) CO (mL/kg/min) pH Base excess (mmol/L) PaCO2 (mmHg) SpO2 (%) Lactate (mmol/L) Arterial hemoglobin (g/L)

3:1 C:V + 21% oxygen (n = 8)

3:1 C:V + 100% oxygen (n = 8)

CCaV + 21% oxygen (n = 8)

CCaV + 100% oxygen (n = 8)

Sham (n = 4)

p-Value

1.5 (1.0–2.8) 1.8 (1.7–2.2) 4/4 207 (177–235) 73 (66–80) 317 (228–421) 7.4 (7.3–7.5) 3 (−2 to 6) 42 (38–45) 93 (92–95) 2.7 (2.2–3.3) 73 (67–87)

2.0 (2.0–3.0) 2.0 (1.8–2.4) 6/2 223 (171–230) 74 (68–79) 345 (286–414) 7.5 (7.4–7.5) 4 (−2 to 6) 36 (34–39) 93 (90–98) 2.5 (1.8–2.8) 79 (72–86)

2.0 (2.0–3.0) 2.2 (2.0–2.3) 5/3 225 (200–239) 74 (72–75) 342 (310–401) 7.4 (7.4–7.4) 2 (0–3) 40 (38–43) 88 (84–93) 1.9 (1.8–3.5) 79 (60–87)

1.5 (1.0–2.0) 1.9 (1.8–2.1) 6/2 202 (174–255) 69 (60–75) 364 (336–542) 7.4 (7.4–7.5) 2 (0–7) 41 (38–46) 92 (86–95) 2.5 (1.8–4.0) 72 (70–81)

1.5 (1.0–2.75) 2.0 (1.8–2.2) 3/1 199 (166–233) 67 (64–67) – 7.5 (7.3–7.5) 4 (−5 to 8) 37 (36–39) 93 (92–98) 2.4 (1.9–3.7) 78 (61–92)

0.34 0.32 0.79 0.89 0.10 0.76 0.75 0.97 0.15 0.46 0.81 0.99

Data are presented as median (IQR); MAP – mean arterial blood pressure, CO – cardiac output.

Data collection and analysis Piglet demographics including age, weight and gender were recorded. Transonic flow probe, HR and pressure transducer outputs were digitized and recorded with PowerLab® LabChart® software (ADInstruments, Dunedin, NZ). VT , ventilation rate and ETCO2 were analyzed using the Flow Tool Physiologic Waveform Viewer (Philips Healthcare, Wallingford, CT). Continuous variables are presented as median (interquartile range (IQR)). For the respiratory parameters, the mean value for each piglet was calculated first, then the median of the means calculated for each intervention group. Kaplan–Meier survival graphs were used and the proportions of surviving piglets until 4 h after ROSC in the oxygen and CC groups were compared by z test. All other data were compared between groups using Friedman’s ANOVA and the Mann–Whitney U test for continuous variables, and 2 for categorical variables. p-Values were 2-sided and p < 0.05 was considered statistically significant. Statistical analyses were performed with IBM SPSS 22 for Mac (IBM Corporation, Armonk, NY). Results Piglet baseline characteristics are presented in Table 1. Characteristics of resuscitation, immediate recovery and survival are presented in Table 2. ROSC and mortality Of the 32 animals subjected to asphyxia and PEA, 7 (22%) did not achieve ROSC despite CPR. Time to ROSC in these animals was set to 900 s (15 min) for comparisons between groups using non-parametric tests. The piglets that achieved ROSC did so after 75–592 s of CPR. There was no difference in the number of animals achieving ROSC, time to ROSC, and mortality between the groups

(Table 2), including when stratifying the results according to FiO2 (p = 0.90 for time to ROSC and 0.36 for mortality) or CC method (p = 0.84 for time to ROSC and 0.64 for mortality). Survival graphs for FiO2 and CC groups are presented in Fig. 2. Respiratory parameters Respiratory parameters during CPR are presented in Table 3. ETCO2 during CPR was not different between all groups nor when stratified according to FiO2 and CC method. The VT during CPR was different between all groups (p = 0.05) with significantly higher VT in 3:1 C:V CPR than CCaV (p = 0.01), but no difference when stratifying according to FiO2 (Table 3). Minute ventilation (MV) during CPR was not different between all groups nor when stratified according to FiO2 and CC method (Table 3). The fraction of simultaneous ventilations and CC was not different between all groups (p = 0.08), but were higher in CCaV (median (IQR) 22 (5–39)%) compared to 3:1 C:V CPR (5 (1–13)%), p = 0.031. Hemodynamic parameters CPR characteristics are presented in Table 3. Differences in CC frequency were observed between all groups, with a higher CC frequency in CCaV than 3:1 C:V CPR. Despite a higher number of CC/min in CCaV, DBP during CPR did not increase, and the number of epinephrine doses used during CPR was not different between the four groups. No differences in the temporal changes in HR (p = 0.86) or mean arterial blood pressure (MAP) (p = 0.44) were observed between the four groups. The DBP and arterial lactate at 4 h after ROSC were not different between all groups, or stratified according to FiO2 and CC method (Table 4). Piglets resuscitated with air had a higher median LV stroke volume compared to 100% oxygen at 30 min (1.4 vs. 1.0 mL/kg, p = 0.01) and 4 h (0.8 vs. 0.5 mL/kg, p = 0.06) after ROSC. However, due to a

Table 2 Characteristics of hypoxia, resuscitation and survival of 4 groups of asphyxiated piglets.

Hypoxia time (min) At cardiac arrest

Resuscitation Achieving ROSC (n (%)) Immediately after ROSC Survival 4 h after ROSC (n (%))

pH paCO2 (mmHg) Lactate (mmol/L) CC/min (n) Epinephrine doses (n) Heart rate (bpm) MAP (mmHg)

3:1 C:V + 21% O2 (n = 8)

3:1 C:V + 100% O2 (n = 8)

CCaV + 21% O2 (n = 8)

CCaV + 100% O2 (n = 8)

p-Value

31 (32–40) 6.6 (6.5–6.7) 96 (87–109) 18.3 (14.1–18.9) 92 (83–94) 1 (0–3) 6 (75) 218 (149–241) 77 (66–87) 2 (25)

31 (26–34) 6.5 (6.5–6.7) 82 (60–100) 17.6 (15.0–19.9) 85 (77–90) 1 (0–4) 6 (75) 207 (186–226) 89 (63–105) 4 (50)

33 (29–42) 6.6 (6.5–6.8) 53 (44–92) 17.8 (16.6–19.8) 92 (90–96) 1 (1–1) 8 (100) 208 (171–221) 88 (82–99) 3 (38)

33 (31–37) 6.7 (6.5–7.0) 76 (48–107) 17.9 (17.4–18.2) 90 (89–94) 1 (1–4) 5 (63) 213 (178–237) 92 (80–103) 3 (38)

0.79 0.44 0.06 0.29 0.05 0.80 0.32 0.83 0.14 0.79

Data are presented as median (IQR); MAP – mean arterial blood pressure, ROSC – return of spontaneous circulation, CC – chest compressions.

Table 3 Respiratory and hemodynamic parameters during cardiopulmonary resuscitation.

VT (mL/kg) Ventilation rate (per minute) MV (mL/kg/min) ETCO2 (mmHg) CC/min (n) CA flow (mL min−1 ) DBP (mmHg) Epinephrine doses (n)

3:1 C:V + 21% oxygen (n = 8)

3:1 C:V + 100% oxygen (n = 8)

17 (11–22) 36 (33–39)

21 (15–22) 39 (36–40)

390 (373–528) 15 (10–20) 92 (83–94) 9 (5–12)

578 (420–691) 14 (9–17) 85 (77–90) 4 (2–12)

32 (29–53) 1 (0–3)

CCaV + 21% oxygen (n = 8)

CCaV + 100% oxygen (n = 8)

13 (10–16) 37 (32–53) 386 (274–505) 14 (10–19) 92 (90–96) 8 (5–12)

19 (5–) 1 (0–4)

14 (13–15) 39 (36–47) 525 (355–565) 11 (9–13) 90 (89–94) 7 (3–10)

15 (9–30) 1 (1–1)

44 (25–) 1 (1–4)

p-Value

0.05 0.70 0.19 0.29 0.05 0.68 0.30 0.80

21% oxygen (n = 16) 14 (10–17) 37 (33–41) 517 (389–666) 15 (10–19) 92 (90–94) 8 (5–12) 29 (12–43) 1 (1–1)

100% oxygen (n = 16) 15 (13–21) 39 (36–40) 593 (542–710) 12 (9–15) 90 (87–92) 5 (3–8) 29 (10–55) 1 (0–4)

p-Value

0.21 0.43 0.13 0.09 0.09 0.31 0.86 0.59

3:1 C:V (n = 16) 18 (14–2) 38 (34–40) 604 (466–805) 14 (10–18) 89 (78–92) 5 (3–12) 32 (28–47) 1 (0–4)

CCaV (n = 16) 14 (13–15) 38 (36–49) 539 (431–620) 12 (10–18) 91 (90–94) 7 (4–10) 16 (10–46) 1 (1–2)

p-Value

0.01 0.74 0.16 0.44 0.04 0.48 0.35 0.72

Table 4 Markers of oxidative stress and perfusion 4 h after return of spontaneous circulation. 3:1 C:V + 21% oxygen (n = 2) Arterial lactate 4 h after ROSC (mmol/L) DBP (mmHg) Left ventricular GSSG/GSH ratio Left ventricular lactate (␮mol/mg protein) Left ventricular MMP-9 Left ventricular MMP-2

4.6 (3.4–) 34 (21–) 0.10 (0.09–0.12)

3:1 C:V + 100% oxygen (n = 4) 5.1 (3.2–) 44 (35–52) 0.13 (0.11–0.20)

CCaV + 21% oxygen CCaV + 100% (n = 3) oxygen (n = 3)

p-Value

17.9 (1.2–)

8.2 (4.2–)

0.74

9.7 (2.9–18.0)

31 (31–38) 0.17 (0.12–)

0.50 0.04

32 (24–45) 0.10 (0.08–0.11)

32 (27–) 0.08 (0.07–0.08)

21% oxygen (n = 5)

100% oxygen (n = 7) 7.0 (3.7–10.6) 40 (33–52) 0.14 (0.11–0.22)

p-Value

0.66 0.20 0.005

3:1 C:V (n = 6) 4.6 (3.3–10.9) 43 (30–49) 0.11 (0.10–0.14)

CCaV (n = 6)

p-Value

10.6 (3.5–18.0)

0.43

35 (30–46) 0.12 (0.08–0.22)

0.49 0.83

1.1 (0.8–1.5)

1.1 (0.8–1.2)

1.9 (1.8–)

2.2 (0.9–)

0.23

1.4 (0.9–1.9)

1.1 (0.9–2.2)

0.95

1.1 (0.8–1.2)

2.1 (1.3–2.3)

0.05

1.0 (0.6–1.4) 0.9 (0.7–1.0)

1.8 (1.0–2.5) 0.8 (0.6–1.1)

1.3 (1.0–) 0.8 (0.7–)

1.8 (0.7–) 0.9 (0.6–)

0.53 0.92

1.2 (0.7–1.4) 0.9 (0.7–1.0)

1.8 (1.0–2.5) 0.9 (0.6–0.9)

0.23 0.73

1.2 (0.8–2.2) 0.8 (0.7–1.0)

1.5 (0.9–1.9) 0.9 (0.6–1.0)

0.84 0.62

A.L. Solevåg et al. / Resuscitation 106 (2016) 7–13

Data are presented as median (IQR); C:V – chest compression:ventilation, CCaV – continuous chest compression and asynchronous ventilation, VT – tidal volume, MV – minute ventilation, ETCO2 – end-tidal CO2 , CC – chest compressions, CA – carotid artery, DBP – diastolic blood pressure. The bold p-values are the ones that are significant.

Data are presented as median (IQR); C:V – chest compression:ventilation, CCaV – continuous chest compression and asynchronous ventilation, CC – chest compressions, MMP – matrix metalloproteinase, GSSG – oxidized glutathione, GSH – glutathione, DBP – diastolic blood pressure. The bold p-values are the ones that are significant.

11

12

A.L. Solevåg et al. / Resuscitation 106 (2016) 7–13

performed in a synchronized or asynchronized fashion. While respiratory and hemodynamic parameters during CPR were not influenced by FiO2 or CC method, post-ROSC hemodynamics were modestly compromised after 100% oxygen resuscitation and CCaV. Elevated markers of oxidative stress in LV tissue supported the finding of lower SV after 100% oxygen resuscitation. A lower post-ROSC MAP in CCaV was accompanied by a higher LV lactate, lower VT and more CC min−1 compared to 3:1 C:V CPR piglets. Oxygen

Fig. 2. Kaplan–Meier survival graph for the oxygen and chest compression groups. 21% vs. 100% oxygen, p = 0.998 z-test; p = 0.485 1-way on Rank. 3:1 vs. CCaV, p = 0.998 z-test; p = 0.937 1-way on Rank.

higher post-ROSC HR in 100% oxygen piglets, the median CO at 30 min and 4 h after ROSC was not significantly different between air and 100% oxygen (277 vs. 227 mL/kg and 170 vs. 117 mL/kg, respectively). MAP was lower after CCaV compared to 3:1 C:V CPR at 15 min (p = 0.05), but not at 4 h (p = 0.20) after ROSC. Biochemical analyses Biochemical results from the LV myocardium are presented in Table 4. LV lactate was higher after CCaV than 3:1 C:V CPR, but not different between oxygen groups. The LV GSSG/GSH ratio, a tissue marker of oxidative stress, was different between the four groups (p = 0.04), higher in piglets resuscitated with 100% oxygen compared to air (p = 0.005) with no difference between groups stratified by CC method (Table 4). There was no difference in LV MMP-2 and MMP-9 between groups (Table 4). Discussion We designed this study to examine air vs. 100% oxygen combined with 3:1 C:V or CCaV during neonatal CPR. We found that recovery from cardiac arrest was similar irrespective of air or 100% oxygen ventilation and whether CC and ventilations were

Previous experimental studies have failed to demonstrate a difference in hemodynamic parameters following resuscitation with air or 100% oxygen.15 In our study, 100% oxygen resulted in a lower post-ROSC SV compared to air. Reduced CO and poor systemic perfusion may be a possible mechanism underlying the increased mortality in asphyxiated infants exposed to hyperoxia.2,3 The use of air in neonatal asphyxia-induced cardiac arrest and CC has been subjected to major debate and rejected because of the opposing effects of oxidative stress and anaerobic metabolism. There is a lack of data about the effectiveness of air in severely compromised animals needing prolonged resuscitation efforts, with no data available in the respective clinical population.4 The animals in the current study were severely compromised with a fairly low ROSC rate (78%), long ROSC times (up to 10 min), and high mortality (63%). The results strengthen our previous findings in asphyxiated piglets that air is as effective as 100% oxygen during neonatal CPR.6 We also observed that air attenuated myocardial oxidative stress without compromising the replenishment of oxygen debt as evident by a similar LV lactate compared to 100% oxygen. Indeed, 100% oxygen exposure of asphyxiated infants may increase reoxygenation injury and cause death or neurological impairment.2,3 In asphyxiated piglets, Borke et al.16 demonstrated that myocardial MMP-2 was elevated after hypoxic-ischemia, and higher in piglets resuscitated with 100% oxygen compared with air. We did not find a difference in MMP-2 and -9 between experimental groups, possibly due to characteristics of the animal model and a lack of power, i.e. few animals surviving 4 h after ROSC for tissue collection. Nonetheless, in agreement with Cheung et al.,14 we found that a lower SV in the 100% oxygen piglets was associated with a higher LV GSSG/GSH ratio compared with air resuscitated piglets. Our results indicate that air may be less harmful than 100% oxygen during neonatal CC. Chest compressions Our results are in agreement with previous piglet studies suggesting that ROSC does not change when attempts are made to increase CPP by performing uninterrupted CC.10,17,18 Berkowitz et al.19 reported no difference in hemodynamic parameters including CPP when infant piglets received CCaV compared to 5:1 C:V CPR, concluding that sufficient intrathoracic pressures could be achieved during conventional CPR in small animals (and human infants) due to the shape and compliance of their chests. Alternatively, severe acidosis with systemic vasodilation may contribute to a blunted effect of CCaV in profound asphyxia. As almost all the piglets in our study received one or more doses of epinephrine before achieving ROSC, CC alone, even when uninterrupted, may not generate sufficient CPP to achieve ROSC in these animals. Results from this and other studies indicate that epinephrine may be necessary to obtain ROSC in severe asphyxia.20 This is in contrast to adult animal data9,21,22 and adult observational studies23 where an advantage of uninterrupted CC has been demonstrated. A frequently raised concern, confirmed by our study, is the potential for CC to interfere with assisted ventilations during CCaV in newborn infants. In agreement with our previous manikin study,24 CCaV resulted in a lower VT. Interestingly, we found higher LV lactate 4 h after CCaV

A.L. Solevåg et al. / Resuscitation 106 (2016) 7–13

compared to 3:1 C:V CPR, even though the difference in arterial lactate was not significant. One might speculate that exposure of the myocardium to a higher number of CC during CCaV may have contributed to tissue lactate accumulation. Limitations Newborn piglets have similar anatomy and pathophysiology to human neonates at near-term gestation. However, in addition to anesthetic and surgical confounding factors, all piglets had already undergone fetal to neonatal transition, making respiratory and hemodynamic responses to CPR not entirely comparable to delivery room resuscitation. Nevertheless, the translational value remains high and is relevant given the poor outcome of newborns who have severe combined acidosis with pH < 6.6, high lactate and receive aggressive resuscitation for >10 min.25 Continuous CC in advanced pediatric CPR is performed at a rate of 100 min−1 , and one might argue that both a higher CC and ventilation rate could be beneficial in neonatal CCaV. As a higher number of CC and/or ventilations could be a potential confounder, we decided to aim for the same number of CC (90) and ventilations (30) min−1 in all intervention groups in order to single out the effect of performing uninterrupted CC. Despite adequate randomization of animals, there were minor differences between groups including paCO2 , BE and CC rate, this may limit the translation of our findings. Conclusion In conclusion, in severely asphyxiated piglet with prolonged resuscitation efforts and high mortality, air ventilation was as effective as 100% oxygen during CC. Uninterrupted CC did not improve ROSC. CPR with a pause for one ventilation with air after every third CC has proven to be effective in animal models,6,5 and should be investigated in human infants. Conflict of interest statement The authors declare no conflict of interest. Acknowledgements ALS is supported by the Canadian Institutes of Health Research (operating grant MOP-CIA-299111 to PYC and travel award to ALS) and the South-Eastern Norway Regional Health Authority. MOR is supported by a Fellowship of Molly Towell Perinatal Foundation. GMS is supported by a Heart and Stroke Foundation/University of Alberta Professorship for Neonatal Resuscitation and by a Heart and Stroke Foundation of Canada Research Scholarship. The sponsor of the study had no role in study design, data collection, data analysis, data interpretation or writing of the report. References 1. Vento M, Asensi M, Sastre J, Garcia-Sala F, Pallardo FV, Vina J. Resuscitation with room air instead of 100% oxygen prevents oxidative stress in moderately asphyxiated term neonates. Pediatrics 2001;107:642–7.

13

2. Saugstad OD. Resuscitation of newborn infants: from oxygen to room air. Lancet 2010;376:1970–1. 3. Saugstad OD, Ramji S, Soll RF, Vento M. Resuscitation of newborn infants with 21% or 100% oxygen: an updated systematic review and meta-analysis. Neonatology 2008;94:176–82. 4. Perlman JM, Wyllie J, Kattwinkel J, et al. Part 7: Neonatal resuscitation: 2015 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Circulation 2015;132:S204–41. 5. Linner R, Werner O, Perez-de-Sa V, Cunha-Goncalves D. Circulatory recovery is as fast with air ventilation as with 100% oxygen after asphyxia-induced cardiac arrest in piglets. Pediatr Res 2009;66:391–4. 6. Solevag AL, Dannevig I, Nakstad B, Saugstad OD. Resuscitation of severely asphyctic newborn pigs with cardiac arrest by using 21% or 100% oxygen. Neonatology 2010;98:64–72. 7. Morley PT, Lang E, Aickin R, et al. Part 2: Evidence evaluation and management of conflicts of interest: 2015 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Circulation 2015;132:S40–50. 8. Wyllie J. Reply to: ‘What initial oxygen is best for preterm infants in the delivery room? A response to the 2015 neonatal resuscitation guidelines.’. Resuscitation 2016;101:e9. 9. Kern KB, Hilwig RW, Berg RA, Sanders AB, Ewy GA. Importance of continuous chest compressions during cardiopulmonary resuscitation: improved outcome during a simulated single lay-rescuer scenario. Circulation 2002;105:645–9. 10. Schmolzer GM, O’Reilly M, Labossiere J, et al. 3:1 compression to ventilation ratio versus continuous chest compression with asynchronous ventilation in a porcine model of neonatal resuscitation. Resuscitation 2014;85:270–5. 11. Kilkenny C, Browne WJ, Cuthill IC, Emerson M, Altman DG. Improving bioscience research reporting: the arrive guidelines for reporting animal research. PLoS Biol 2010;8:e1000412. 12. Bárány M, Glonek T. In: Frederiksen DW, Cunningham LM, editors. Methods in enzymology. New York: Academic Press; 1982. p. 624–76. 13. Emara M, Cheung PY. Inhibition of sulfur compounds and antioxidants on mmp2 and -9 at the activity level found during neonatal hypoxia-reoxygenation. Eur J Pharmacol 2006;544:168–73. 14. Cheung PY, Miedzyblocki M, Lee TF, Bigam DL. Effects of post-resuscitation administration with sodium hydrosulfide on cardiac recovery in hypoxiareoxygenated newborn piglets. Eur J Pharmacol 2013;718:74–80. 15. Fugelseth D, Borke WB, Lenes K, Matthews I, Saugstad OD, Thaulow E. Restoration of cardiopulmonary function with 21% versus 100% oxygen after hypoxaemia in newborn pigs. Arch Dis Child Fetal Neonatal Ed 2005;90:F229–34. 16. Borke WB, Munkeby BH, Halvorsen B, et al. Increased myocardial matrix metalloproteinases in hypoxic newborn pigs during resuscitation: effects of oxygen and carbon dioxide. Eur J Clin Invest 2004;34:459–66. 17. Solevag AL, Dannevig I, Wyckoff M, Saugstad OD, Nakstad B. Extended series of cardiac compressions during CPR in a swine model of perinatal asphyxia. Resuscitation 2010;81:1571–6. 18. Solevag AL, Dannevig I, Wyckoff M, Saugstad OD, Nakstad B. Return of spontaneous circulation with a compression:ventilation ratio of 15:2 versus 3:1 in newborn pigs with cardiac arrest due to asphyxia. Arch Dis Child Fetal Neonatal Ed 2011;96:F417–21. 19. Berkowitz ID, Chantarojanasiri T, Koehler RC, et al. Blood flow during cardiopulmonary resuscitation with simultaneous compression and ventilation in infant pigs. Pediatr Res 1989;26:558–64. 20. Solevag AL, Cheung PY, Lie H, et al. Chest compressions in newborn animal models: a review. Resuscitation 2015;96:151–5. 21. Berg RA, Sanders AB, Kern KB, et al. Adverse hemodynamic effects of interrupting chest compressions for rescue breathing during cardiopulmonary resuscitation for ventricular fibrillation cardiac arrest. Circulation 2001;104:2465–70. 22. Berg RA, Hilwig RW, Kern KB, Ewy GA. Bystander chest compressions and assisted ventilation independently improve outcome from piglet asphyxial pulseless cardiac arrest. Circulation 2000;101:1743–8. 23. Bobrow BJ, Spaite DW, Berg RA, et al. Chest compression-only CPR by lay rescuers and survival from out-of-hospital cardiac arrest. JAMA 2010;304:1447–54. 24. Solevag AL, Madland JM, Gjaerum E, Nakstad B. Minute ventilation at different compression to ventilation ratios, different ventilation rates, and continuous chest compressions with asynchronous ventilation in a newborn manikin. Scand J Trauma Resusc Emerg Med 2012;20:73. 25. Cheung PY, Robertson CM. Predicting the outcome of term neonates with intrapartum asphyxia. Acta Paediatr 2000;89:262–4.