Cardiac arrest is a dreadful event in pediatric intensive care unit (PICU). Cardiopulmonary resuscitation (CPR) involves chest compression and manual ventilation at appropriate intervals. Return of spontaneous circulation (ROSC) is the initial therapeutic goal in cardiac arrest and is a measure of initial success. Vasopressor medications are often used during CPR. These medications increase aortic diastolic pressure, thereby improving coronary perfusion pressure, which facilitates ROSC [1]. Epinephrine is the most widely studied and the first line vasoactive drug as per the pediatric advanced life support (PALS) guidelines [2]. Vasopressin, a potent vasoconstrictor, is well studied in adult cardiac arrest [3]. The recent advanced cardio-vascular life support guideline recommends that Vaso-pressin, combined with epinephrine, may be considered in adult cardiac arrest resuscitation [3]. Though the pediatric in-hospital cardiac arrest (IHCA) outcome has improved from 39% to 77% in high-income countries, data from low-and middle-income countries are lacking or are under-reported [2]. Animal studies, case series, and feasibility pilot studies have shown encouraging results for the use of vasopressin in pediatric cardiac arrest [4-6]. This study hypothesized that epinephrine plus vasopressin would be associated with a higher rate of ROSC as compared to epinephrine plus placebo in the pediatric intensive care unit cardio-pulmonary resuscitation.

A computer-generated, unstratified, block randomi-zation with variable block sizes of four, six, and eight was used with an allocation ratio of 1:1 by a person not involved in the study. Individual assignments were kept in serially numbered boxes. Each box contained ten identically looking one mL ampoules of either vasopressin or placebo (normal saline). The original label in each ampoule was removed and replaced by an opaque paper. Each box was serially numbered and allocated to the patient according to the random sequence. The serially-numbered trial drug boxes were kept in a separate place in PICU to avoid the wrong allocation in the stressful environment. Only one trial drug box was kept in the crash cart, which contained all the emergency drugs and equipment required for CPR. The nurses were instructed to open the trial drug box, which was kept in the crash cart during CPR. The investigator ensured the replacement of the trial drug box in the crash cart according to the serial number once the trial drug was used. Multiple simulation sessions were carried out and discussed before the start of the study. Injection normal saline (sodium chloride 0.9%, 1 mL, Serum Institute of India Pvt Ltd), injection epinephrine (Bioaderna, 1 mg per 1 mL, Health Biotech Ltd) and injection vasopressin (Vascel 20, 20 Unit per mL, CELON laboratories Pvt Ltd) were used in this study. The institute’s central pharmacy supplied the trial drugs. The participants, treating team and nurses administering the medications, and the investigators, were unaware of the treatment assignments. The person who collected and entered the data into the datasheet and the study statistician were unaware of the treatment assign-ment throughout the analyses. The treatment assignment was disclosed, after the first draft of the result was finalized.

All patients received CPR in accordance with the PALS-2015 guidelines established by the American Heart Association (AHA) [2]. This includes the support of airway, breathing, including supplemental oxygen, evaluation of cardiac rhythm, high-quality CPR with minimally interrupted chest compressions, electrical defibrillation if appropriate, and medications except for the trial drugs. The resuscitation team members were trained to provide CPR as per the PALS 2015 guidelines [2]. The facility for standby extracorporeal membrane oxygenation (ECMO) is not available in the study site (PICU). Our hospital has no approved guidelines for ‘Do not resuscitate’ instructions. Epinephrine plus vasopressin group received intravenous epinephrine (1:10000) 0.1 mL per kg and vasopressin (1:1.5 dilution in normal saline) 0.1 mL per kg (=0.8 unit per kg; maximum dose of 5 mL, 40 unit). Epinephrine plus placebo group received intravenous epinephrine (1:10000) 0.1 mL per kg and placebo (1:1.5 dilution in normal saline) 0.1 mL per kg. The trial drugs were given as bolus doses, concurrently if two vascular accesses were available or within 10 seconds gap if one vascular access was available. The trial drugs were given at an interval of every three minutes until ROSC or a maximum of five doses, whichever was earlier. Three mL normal saline flush was given after adminis-tration of each dose of the trial drug. Subsequently, if needed, epinephrine was continued as per protocol. Post-resuscitation care was provided to the patients who achieved ROSC as per the unit protocol (from PALS-2015 guidelines) [2]. All patients were followed up until death or 90 day post-discharge. The functional status of the survivor was assessed by using the pediatric cerebral performance category (PCPC) scale and pediatric overall performance category (POPC) scale (lower the score, better the neurological outcome) [7]. Data regarding the cardiac arrest events and their outcomes were collected as per the Utstein style template and in the predesigned proforma [8-10].

The primary outcome was the proportion of patients who achieved ROSC. The secondary outcomes were (i) survival rate (at 24 hours, PICU, hospital, and 90-day of discharge), (ii) functional status (at PICU, hospital, and 90-day of discharge), (iii) need for organ support(s), (iv) length of stay in PICU and hospital, and (v) adverse effect(s) of the study drugs if any. ROSC was defined as the restoration of a spontaneous perfusing rhythm that results in more than an occasional gasp, fleeting palpable pulse, or arterial waveform [2,3,10]. Sustained ROSC was defined as not requiring chest compressions for 20 consecutive minutes after obtaining ROSC and signs of perfusion [2,3,10]. The probability of adverse trial drug reaction was assessed by Naranjo algorithm [11].

The ROSC rate varies between 47% and 64.6%, as reported by previous studies [12,13]. We assumed that the primary outcome of interest in the control group was 50%. We calculated the sample size based upon the assumption of 30% improvement in the primary outcome by the intervention with 80% power at the 5% significance (two-sided) and 1:1 allocation. Thirty-nine patients were required in each group by calculation. With a 10% attrition rate, the final sample size was estimated as 86 [12-14]. The sample size was calculated using the software nQuery version 4.0.

Statistical analysis: Data were analyzed according to their assigned groups (intention to treat analysis). The distribution of data was checked with the Kolmogorov-Smirnov Z test. Continuous variables were compared between the two groups by Student’s t-test for normally distributed or by the Mann-Whitney U test for skewed data. Proportions were compared by the Chi-square test (or Fisher’s exact test if expected cell frequencies were less than five). Kaplan-Meier curve and log-rank test were used to analyze ‘time to event’ data followed by Cox proportional hazard regression analysis to adjust for the prespecified baseline factors (age, sex, and PRISM-III score). The relative risk and hazard ratio, with a 95% confidence interval, was calculated as appropriate. All tests were two-tailed, and a P value of less than 0.05 was considered statistically significant. IBM SPSS software 20.0 (IBM Corp) and Epi Info 7 (7.0.9.7, CDC) were used for data analysis.

The study flow is depicted in Fig. 1. Ninety patients were enrolled (epinephrine plus vasopressin, n=45, and epinephrine plus placebo n=45) after the screening of 118 patients. The baseline characteristics and clinical variables are described in Table I. The median (IQR) time to first cardiac arrest since admission was similar between groups [2 (1-7) vs 2 (1-5) day; P=0.75]. The most common (80%) arrest rhythm was pulseless electrical activity (PEA). Hemodynamic abnormality (67.8%) was the most common event that led to arrest, followed by respiratory events (23.3%). Respiratory failure was an underlying illness in 76 (84.4%) patients and sepsis in 60 (66.7%) patients. The median (IQR) duration of CPR was similar between groups [18 (10-30) vs 15 (6–30) minutes; P=0.96].

Fig. 1 Study flow chart.

 

This randomized controlled trial enrolled 90 patients who underwent CPR in PICU. We found no significant difference in the proportion of patients who achieved ROSC and survival rate between the Epinephrine plus Vasopressin group and Epinephrine plus Placebo group. In our study, the overall rate of achieving ROSC was 54.4%, and survival to hospital discharge was 2.2%. This observation contrasts with the data from high-income countries, where the rates were more than 80% and 40%, respectively [15]. The potential reasons could be the uniform reporting registries, universal healthcare prog-rams, training of health care workers, and accessibility to extracorporeal membrane oxygenation (ECMO).

The outcome of a pediatric cardiac arrest depends upon many factors, including the initial presenting rhythm, the place of cardiac arrest, early recognition of the arrest, and the underlying conditions. In the previous studies done in high-income countries, asystole (55%) was the initial arrest rhythm, and respiratory failure was the common precipitating factor [1,9]. Nevertheless, they enrolled patients not only from PICU but also from the emergency department and general ward [1,9]. In contrast, this study enrolled patients only from PICU, where stringent monitoring helped identify the arrest much earlier, before progressing to asystole. Similar to our study setting, Rathore, et al. [12] reported bradycardia (52.2%) and sepsis (71%) as the initial arrest rhythm and underlying diagnosis, respectively. They reported a higher ROSC rate (64.6%) and survival to hospital discharge (14%). However, only 21% of CPR occurred in PICU in that study. In our study, patients were enrolled only from PICU. So, the study population was different. Generally, PICU patients are sicker and the majority of them have multiple organ dysfunction requiring organ support. Also, the initial rhythm is an important factor in predicting the outcome; bradycardia rhythm with a pulse is more likely to recover than pulseless non-shockable rhythms [12].

Acknowledgments: Mrs. S. Raja Deepa B.Com, MCA (JIPMER Campus, Puducherry, India) for blinded data handling, review and editing of the manuscript; Mr. Rakesh Mohindra (Punjab University, Chandigarh, India) and Mrs. Thenmozhi M (M.Sc, Ph.D., Senior Demonstrator, CMC, Vellore, India) for helping with statistical analysis and Mrs. Harpreet Kaur (Punjab University, Chandigarh, India), and Mrs. Neelima Chadha (Tulsi Das Library, PGIMER, Chandigarh, India) for helping with the medical literature search.

Note: The preliminary study data was presented in the 21st National Conference of IAP Intensive Care Chapter (NCPIC 2019), from 5th to 8th December, 2019, Bengaluru.

Contributors: RR: had full access to all the data in the study and took responsibility for the integrity of the data and the accuracy of the data analysis; RR: Study concept and design; AS, MC, KM, RSK, AJ, RB: acquisition, analysis, or interpretation of data and drafting of the first manuscript; MC, RB, NB, SM: protocol development and revision of the manuscript; RR, SM: critical revision of the manuscript for important intellectual content; RR, NB: study supervision. RR: is the guarantor of the paper. All authors approved the final version of the manuscript.

Funding: None; Competing interest: None stated.