G Model

ARTICLE IN PRESS

RESUS-7347; No. of Pages 9

Resuscitation xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

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

Clinical paper

Conventional versus chest-compression-only cardiopulmonary resuscitation by bystanders for children with out-of-hospital cardiac arrest夽 Yoshikazu Goto a,∗ , Akira Funada a , Yumiko Goto b a b

Department of Emergency and Critical Care Medicine, Kanazawa University Hospital, Kanazawa, Japan Department of Cardiology, Yawata Medical Centre, Komatsu, Japan

a r t i c l e

i n f o

Article history: Received 19 July 2017 Received in revised form 11 October 2017 Accepted 18 October 2017 Keywords: Out-of-hospital cardiac arrest Cardiopulmonary resuscitation Children Infant Epidemiology

a b s t r a c t Aim: It is unclear whether chest-compression-only cardiopulmonary resuscitation (CC-CPR) is therapeutically equivalent to conventional CPR for children with out-of-hospital cardiac arrest (OHCA). We aimed to determine the association of CC-CPR and conventional CPR with outcomes in discrete child patient populations with OHCA. Methods: We analysed 6810 children (aged <18 years) using Japanese registry from 2007 to 2014. Main outcomes measure was 30-day neurologically intact survival after OHCA, defined as Glasgow-Pittsburgh cerebral performance categories 1 or 2. Results: In propensity score-matched children aged 1–17 years (n = 2682), overall neurologically intact survival rate was significantly higher after conventional CPR than after CC-CPR (9.4% vs. 6.0%, P = 0.001). However, there was no significant difference between the two CPR modalities in patients with cardiac aetiology (14.2% vs. 11.8%, P = 0.32), initial shockable rhythm (35.3% vs. 31.7%, P = 0.59), or age ≥8 years (12.4% vs. 9.8%, P = 0.13). For matched infants (n = 1994), no significant differences were observed in overall neurological intact survival between conventional CPR and CC-CPR (2.2% vs. 1.3%, P = 0.17). In infant subgroup analyses, neurologically intact survival was similar between the CPR modalities for cardiac aetiology (0.3% vs. 1.0%; P = 0.37) and for witnessed arrest (6.2% vs. 6.0%; P = 0.98). Conclusions: In the majority of the paediatric subgroups, conventional CPR was associated with improved outcomes compared to CC-CPR. CC-CPR was associated with 30-day neurologically intact survival similar to conventional CPR for children with OHCA aged ≥8 years, for children aged 1–17 years with cardiac aetiology or initial shockable rhythm, and for infants with cardiac aetiology or witnessed arrest. © 2017 Elsevier B.V. All rights reserved.

1 Introduction Early bystander cardiopulmonary resuscitation (CPR) for outof-hospital cardiac arrest (OHCA) is crucial in the chain of survival [1–5]. Fortunately, the bystander CPR rate in children has increased recently from ∼30% [6,7] to ∼50% [8–12]. In order to increase bystander CPR, in 2008, the American Heart Association (AHA) recommended chest-compression-only CPR (CC-CPR) for adults with OHCA [13]. However, CC-CPR for cardiac arrest does not apply to patients with non-cardiac origin, unwitnessed arrest, or children.

夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at http://10.1016/j.resuscitation.2017.10.015. ∗ Corresponding author at: Kanazawa University Hospital, Department of Emergency and Critical Care Medicine, Takaramachi 13-1, Kanazawa 920-8640, Japan, E-mail address: [email protected] (Y. Goto).

Specifically, two investigations of children with OHCA demonstrated that receiving CC-CPR was associated with inferior 30-day intact neurological survival rates compared with conventional CPR (chest compressions with rescue breaths) [14,15]. However, in children aged 1–17 years with presumed cardiac aetiology, CC-CPR was equally associated with 30-day neurologically intact survival compared with conventional CPR [14]. Based on these findings, in 2015, the International Liaison Committee On Resuscitation (ILCOR) [5] recommended that rescuers provide conventional CPR for infants and children in cardiac arrest; if rescuers could not provide rescue breaths, they should perform CC-CPR. Recent evidence from Japan [16] shows that CC-CPR for children is associated with improved 30-day neurologically intact survival compared with no bystander CPR; no statistically significant differences were observed for CCCPR compared with conventional CPR regardless of arrest aetiology, witness status, or age subgroups. We hypothesized that CC-CPR by bystanders produces neurologically intact survival equivalent

https://doi.org/10.1016/j.resuscitation.2017.10.015 0300-9572/© 2017 Elsevier B.V. All rights reserved.

Please cite this article in press as: Goto Y, et al. Conventional versus chest-compression-only cardiopulmonary resuscitation by bystanders for children with out-of-hospital cardiac arrest. Resuscitation (2017), https://doi.org/10.1016/j.resuscitation.2017.10.015

G Model RESUS-7347; No. of Pages 9

ARTICLE IN PRESS Y. Goto et al. / Resuscitation xxx (2017) xxx–xxx

2

to conventional CPR for children with OHCA aged ≥8 years and for children aged <8 years with presumed cardiac origin or initial shockable rhythm. Using propensity-score matching analyses, the present study investigated a large cohort of paediatric OHCAs to assess differences in post-OHCA outcomes between CC-CPR and conventional CPR according to age.

2 Methods 2.1 Study design This was a nationwide, population-based, observational study of all children (aged <18 years) receiving resuscitation performed by emergency medical services (EMS) personnel after OHCA in Japan between January 1, 2007, and December 31, 2014. Cessation of cardiac mechanical activity was confirmed by the absence of signs of circulation, indicating cardiac arrest [17]. As determined by the attending physicians and EMS personnel, the cause of arrest was presumed to be cardiac unless there was evidence to suggest trauma, hanging, drowning, drug overdose, asphyxia, respiratory disease, cerebrovascular disease, malignant tumours, or any other non-cardiac aetiology. This study was conducted with the approval of the ethics committee at Kanazawa University.

2.2 Study setting Japan has nearly 127 million residents in an area of 378,000 km2 . The Fire and Disaster Management Agency (FDMA) of Japan supervises the nationwide EMS system, while the local fire stations operate the local EMS systems. EMS personnel are trained and permitted to use several resuscitative methods, including automated external defibrillators, insertion of an airway adjunct, insertion of a peripheral intravenous line, and administration of Ringers lactate solution. Further, certain emergency personnel are permitted to insert a tracheal tube and administer intravenous epinephrine. Importantly, Japanese law prohibits EMS personnel from terminating resuscitation in the field. Accordingly, most patients with OHCA undergo CPR by EMS providers and are subsequently transported to hospitals. Emergency telephone dispatchers in Japan are required to provide CPR instructions for CC-CPR if it is difficult for them to administer rescue breathing since 2006 [18].

2.3 Data collection and quality control Since 2005, The FDMA in Japan has launched an on-going, prospective, population-based, observational study involving all OHCA patients receiving EMS treatment [17]. Specifically, EMS personnel at each treatment centre recorded patient data using an Utstein-style template in cooperation with the physician in charge. The recorded data were then transferred to individual local fire stations and subsequently integrated into the data registry system on the FDMA database. Ultimately, all data were stored in the nationwide database developed by the FDMA for public use. With permission from the FDMA, we analysed de-identified patient data contained within this database for the present investigation. Neurological outcomes were stratified utilizing the Cerebral Performance Category (CPC) scale (category 1: good cerebral performance; category 2: moderate cerebral disability; category 3: severe cerebral disability; category 4: coma or vegetative state; and category 5: death) [19]. For all patients, CPC categorization was determined by the attending physician.

2.4 Study endpoints Primary endpoints included 30-day neurologically intact survival, defined as a CPC of 1 or 2 and 30-day survival. 2.5 Statistical analysis We compared 30-day outcomes between conventional CPR and CC-CPR, categorizing patients into two age groups: <1 year (infants) or 1–17 years. To perform rigorous adjustments for differences in the baseline characteristics of patients, we utilized both logistic regression analyses for unmatched patients as well as propensity-score matching analyses to adjust for selection bias when comparing outcomes between conventional CPR and CC-CPR. In analyses of unmatched cohorts, both univariate and multivariate logistic regression analyses were performed in order to estimate the association between patient outcome and the type of bystander CPR performed. In propensity-score matching analyses, we estimated two propensity scores by fitting a logisticregression model that includes 17 variables for infants and 18 variables for patients aged 1–17 years as described in Table 1. We performed one-to-one nearest-neighbour matching between patients with conventional CPR and CC-CPR without replacement, using a caliper width equal to 0.20 of the standard deviation of the logit of the propensity score [20]. Before analysing outcomes, we assessed the success of the propensity-matching procedure by comparing the distribution of patient characteristics in the matched sample by calculating an absolute standardized difference [21]. An absolute standardized difference greater than or equal to 0.1 was considered indicative of a significant imbalance between the two cohorts [22]. To compare the 30-day outcomes between bystander CPR modality, we utilized either chi-squared or Fishers exact tests and further analysed subgroups according to aetiology (cardiac/non-cardiac), initial rhythm (shockable/non-shockable), witnessed status (yes/no), age (1–7 years/8–17 years). Continuous variables are expressed as the mean ± standard deviation, while categorical variables are expressed as percentages. As an estimate of effect size and variability, we reported odds ratio (OR) with 95% confidence intervals (CIs). All statistical analyses were performed using the JMP statistical package (Version 13, SAS Institute Inc., Cary, NC, USA). All reported tests were two-tailed with a P-value <0.05 considered statistically significant. 3 Results Over the 8 years, data for 967,683 patients were compiled in the database. Of these cases, 1.3% of patients (n = 12,708) were treated by EMS personnel and were eligible for study enrolment (Fig. 1). Excluding patients with public access defibrillation-only CPR, rescue breathing-only CPR, and no bystander CPR, we analysed 2842 infants and 3968 children aged 1–17 years with OHCA. In infants, 1994 of the 2842 patients with bystander CPR (70.2%) were matched as described. Further, in children aged 1–17 years, patient matching was achieved for 2682 of 3968 patients with bystander CPR (67.6%). Absolute standardized differences in the matched cohorts were considerably improved in each age cohort (Tables 1 and 2). Regardless of age, the proportions of patients receiving bystander CPR and CC-CPR significantly increased during the study (all Ptrend < 0.001, Supplementary Tables S1 and S2). Further, 30-day survival for infants was significantly increased over the study period (Ptrend < 0.01, Supplementary Table S1). However, the proportions of 30-day CPC 1–2 for both age groups did not significantly increase (Supplementary Tables S1 and S2). Table 3 describes the results of the logistic regression analyses in the unmatched cohorts by age. Between age groups, the

Please cite this article in press as: Goto Y, et al. Conventional versus chest-compression-only cardiopulmonary resuscitation by bystanders for children with out-of-hospital cardiac arrest. Resuscitation (2017), https://doi.org/10.1016/j.resuscitation.2017.10.015

Age 1–17 years (n = 3968)

Conventional CPR

Compression-only CPR

(n = 1191)

(n = 1651)

1

ASD

Conventional CPR

Compression-only CPR

(n = 1496)

(n = 2472)

ASD1

242 245 156 144 114 126 76 88

(20.3) (20.6) (13.1) (12.1) (9.6) (10.6) (6.4) (7.4)

192 251 169 178 201 199 228 233

(11.6) (15.2) (10.2) (10.8) (12.2) (12.1) (13.8) (14.1)

0.241 0.141 0.09 0.04 0.08 0.05 0.251 0.221

241 251 203 216 165 158 118 144

(16.1) (16.8) (13.6) (14.4) (11.0) (10.6) (7.9) (9.6)

221 221 272 311 351 336 389 371

(8.9) (8.9) (11.0) (12.6) (14.2) (13.6) (15.7) (15.0)

0.221 0.241 0.08 0.05 0.09 0.09 0.241 0.161

36 77 397 144 234 93 210 NA 671

(3.0) (6.5) (33.3) (12.1) (19.7) (7.8) (17.6)

(4.4) (5.6) (40.2) (17.1) (11.4) (7.9) (13.5)

(56.3)

73 92 663 282 188 130 223 NA 971

(58.8)

0.07 0.04 0.141 0.141 0.231 <0.01 0.111 NA 0.05

51 121 447 206 278 149 244 8 924

(3.4) (8.3) (29.9) (13.8) (18.6) (10.0) (16.3) (5.9) (61.8)

92 204 843 409 390 206 328 9.2 1540

(3.7) (8.3) (34.1) (16.6) (15.8) (8.4) (13.3) (5.9) (62.3)

0.02 0.01 0.09 0.08 0.07 0.06 0.09 0.211 0.01

474 717

(39.8) (60.2)

710 941

(43.0) (57.0)

0.07 0.07

500 996

(33.4) (66.6)

733 1739

(29.7) (70.3)

0.08 0.08

32 1159

(2.7) (97.3)

40 1611

(2.4) (97.6)

0.02 0.02

148 1348

(9.9) (90.1)

168 2304

(6.8) (93.2)

0.111 0.111

976 148 67

(81.9) (12.4) (5.6)

1388 227 36

(84.1) (13.7) (2.2)

0.06 0.04 0.181

950 300 246

(63.5) (20.1) (16.4)

1697 464 311

(68.6) (18.8) (12.6)

0.111 0.03 0.111

910 1

(76.4) (0.1)

1389 0

(84.1) (0)

0.201 NA

1006 90

(67.3) (2.2)

1853 55

(75.0) (6.0)

0.171 0.191

1160 209 181 12 117 29 11 7.4 26.0

(97.4) (17.6) (15.2) (1.0) (9.8) (2.4) (0.9) (3.9) (9.9)

1602 269 242 30 153 48 18 7.4 26.8

(97.0) (16.3) (14.7) (1.8) (9.3) (2.9) (1.1) (3.1) (9.3)

0.02 0.03 0.02 0.07 0.02 0.03 0.02 0.01 0.09

1448 301 275 182 286 172 72 7.8 28.8

(96.8) (20.1) (18.4) (12.2) (19.1) (11.5) (4.8) (3.8) (10.1)

2393 407 371 215 546 272 112 7.5 29.4

(96.8) (16.5) (15.0) (8.7) (22.1) (11.0) (4.5) (3.3) (10.4)

<0.01 0.09 0.09 0.111 0.07 0.02 0.01 0.08 0.05

Values are reported as n (%) unless indicated otherwise. AED, automated external defibrillator; ASD, absolute standardized difference; CPR, cardiopulmonary resuscitation; NA, not available; SD, standard deviation. 1 An ASD of equal or more than 0.1 was considered to indicate a substantial imbalance between the two cohorts. † Numbers of patients with missing data were 6 (0.21%) in the aged <1 year cohort and 6 (0.15%) in the aged 1–17 years cohort. ‡ Numbers of patients with missing data were 8 (0.28%) in the aged <1 year cohort and 10 (0.25%) in the aged 1–17 years cohort.

ARTICLE IN PRESS

Year 2007 2008 2009 2010 2011 2012 2013 2014 Geographic Japanese regions North (Hokkaido) Northeast (Tohoku) East (Kanto-Koshinetsu) Central (Chubu-Hokuriku) Midwest (Kinki) West (Chugoku-Shikoku) South (Kyushu-Okinawa) Age, y, mean (SD) Male sex Aetiology of cardiac arrest Presumed cardiac Non-cardiac Initial cardiac rhythm Ventricular fibrillation or tachycardia Pulseless electrical activity/asystole Bystander witnessed status No witness Family member Nonfamily member Dispatcher CPR instruction Offered Use of public access AED by bystander CPR by emergency responder Emergency lifesaving technician present in ambulance Physician present in ambulance Prehospital advanced medication by attended physician Defibrillation by emergency responder Use of advanced airway management Insertion of intravenous line Epinephrine administration Call-to-response time, min, mean (SD)† Call-to-hospital arrival time, min, mean (SD)‡

Age <1 year (n = 2842)

G Model

RESUS-7347; No. of Pages 9

Characteristic

Y. Goto et al. / Resuscitation xxx (2017) xxx–xxx

Please cite this article in press as: Goto Y, et al. Conventional versus chest-compression-only cardiopulmonary resuscitation by bystanders for children with out-of-hospital cardiac arrest. Resuscitation (2017), https://doi.org/10.1016/j.resuscitation.2017.10.015

Table 1 Baseline Characteristics of Unmatched Patients according to Age.

3

Conventional CPR

Compression-only CPR

(n = 997)

(n = 997)

ASD1

Conventional CPR

Compression-only CPR

(n = 1341)

(n = 1341)

ASD1

(17.0) (19.7) (12.0) (12.3) (10.8) (11.6) (7.6) (8.7)

172 190 120 121 100 106 96 92

(17.2) (19.0) (12.0) (12.1) (10.0) (10.6) (9.6) (9.2)

0.01 0.02 <0.01 0.01 0.03 0.03 0.07 0.02

194 194 180 199 163 153 117 141

(14.5) (14.5) (13.4) (14.8) (12.2) (11.4) (8.7) (10.5)

186 199 186 196 174 149 119 132

(13.9) (14.8) (13.9) (14.6) (13.0) (11.1) (8.9) (9.8)

0.02 0.01 0.01 0.01 0.02 0.01 0.01 0.02

32 64 363 136 155 80 167 NA 564

(3.2) (6.4) (36.3) (13.6) (15.5) (8.0) (16.7)

(2.4) (6.9) (36.5) (12.8) (15.8) (7.8) (17.6)

(56.5)

24 69 364 128 158 78 176 NA 564

(56.5)

0.05 0.02 <0.01 0.02 0.01 0.01 0.02 NA 0

48 115 411 193 244 124 206 8.1 829

(3.6) (8.6) (30.7) (14.4) (18.2) (9.3) (15.4) (5.9) (61.8)

51 106 422 199 229 118 216 7.9 828

(3.8) (7.9) (31.5) (14.8) (17.1) (8.8) (16.1) (5.8) (61.7)

0.01 0.02 0.02 0.01 0.03 0.02 0.02 0.02 <0.01

399 598

(40.0) (60.0)

416 581

(41.7) (58.3)

0.03 0.03

437 904

(32.6) (67.4)

450 891

(33.6) (66.4)

0.02 0.02

28 969

(2.8) (97.2)

25 972

(2.5) (97.5)

0.02 0.02

119 1222

(8.9) (91.1)

120 1221

(9.0) (91.0)

<0.01 <0.01

835 127 35

(83.8) (12.7) (3.5)

833 130 34

(83.6) (13.0) (3.4)

0.01 0.01 0.01

871 270 200

(65.0) (20.1) (14.9)

870 278 193

(64.9) (20.7) (14.4)

<0.01 0.01 0.01

792 0

(79.4) (0)

783 0

(78.5) (0)

0.02 NA

937 55

(69.9) (4.1)

943 49

(70.0) (3.7)

0.01 0.02

975 172 151 12 98 25 8 7.3 26.1

(97.8) (17.2) (15.1) (1.2) (9.8) (2.5) (0.8) (3.2) (9.6)

966 163 150 12 96 24 9 7.3 26.4

(96.9) (16.3) (15.0) (1.2) (9.6) (2.4) (0.9) (3.2) (9.7)

0.06 0.02 <0.01 0 0.01 0.01 0.01 0.01 0.03

1297 260 236 147 263 155 60 7.70 28.9

(96.7) (19.4) (17.6) (11.0) (19.6) (11.6) (4.5) (3.7) (10.2)

1291 248 232 143 259 123 56 7.70 28.9

(96.3) (18.5) (17.3) (10.7) (19.3) (9.2) (4.2) (3.6) (10.5)

0.02 0.02 0.01 0.01 0.01 0.08 0.01 0.01 <0.01

Values are reported as n (%) unless indicated otherwise. AED, automated external defibrillator; ASD, absolute standardized difference; CPR, cardiopulmonary resuscitation; NA, not available; SD, standard deviation. 1 An ASD of equal or more than 0.1 was considered to indicate a substantial imbalance between the two cohorts.

ARTICLE IN PRESS

170 197 120 123 108 116 76 87

Y. Goto et al. / Resuscitation xxx (2017) xxx–xxx

Year 2007 2008 2009 2010 2011 2012 2013 2014 Geographic Japanese regions North (Hokkaido) Northeast (Tohoku) East (Kanto-Koshinetsu) Central (Chubu-Hokuriku) Midwest (Kinki) West (Chugoku-Shikoku) South (Kyushu-Okinawa) Age, y, mean (SD) Male sex Aetiology of cardiac arrest Presumed cardiac Non-cardiac Initial cardiac rhythm Ventricular fibrillation or tachycardia Pulseless electrical activity/asystole Bystander witnessed status No witness Family member Nonfamily member Dispatcher CPR instruction Offered Use of public access AED by bystander CPR by emergency responder Emergency lifesaving technician present in ambulance Physician present in ambulance Prehospital advanced medication by attended physician Defibrillation by emergency responder Use of advanced airway management Insertion of intravenous line Epinephrine administration Call-to-response time, min, mean (SD) Call-to-hospital arrival time, min, mean (SD)

Age 1–17 years (n = 2682)

Age <1 year (n = 1994)

G Model

Characteristic

RESUS-7347; No. of Pages 9

4

Please cite this article in press as: Goto Y, et al. Conventional versus chest-compression-only cardiopulmonary resuscitation by bystanders for children with out-of-hospital cardiac arrest. Resuscitation (2017), https://doi.org/10.1016/j.resuscitation.2017.10.015

Table 2 Baseline Characteristics of Matched Patients according to Age.

G Model RESUS-7347; No. of Pages 9

ARTICLE IN PRESS Y. Goto et al. / Resuscitation xxx (2017) xxx–xxx

Fig. 1. Flowchart of Patient Inclusion Criteria.

5

CPR, cardiopulmonary resuscitation; EMS, emergency medical services.

proportions of 30-day survival and 30-day CPC 1 or 2 were significantly reduced after CC-CPR than after conventional CPR, described as follows (CC-CPR vs. conventional CPR): for infants, survival: 6.8% (112/1651) vs. 9.7% (115/1191) [adjusted OR, 0.66; 95% CI, 0.49–0.89]; CPC 1 or 2: 1.3% (21/1651) vs. 2.4% (29/1191) [adjusted OR, 0.51; 95% CI, 0.27–0.95]; for children aged 1–17 years, survival: 12.1% (300/2472) vs. 19.4% (290/1496) [adjusted OR, 0.61; 95% CI, 0.50–0.74]; CPC 1 or 2: 5.0% (123/2472) vs. 10.0% (149/1496) [adjusted OR, 0.50; 95% CI, 0.37–0.67]. Fig. 2 shows the matched patient numbers and 30-day outcomes by year. In infants, no significant differences were found in the two types of bystander CPR for every 2-year period. In children aged 1–17 years, 30-day CPC 1–2 after conventional CPR was significantly higher than that after CC-CPR in the 2007–2008 and 2009–2010 periods, but not in the 2011–2012 and 2013–2014 periods. The P-value for trend in 30-day CPC 1–2 was found to be significant only in CC-CPR for children aged 1–17 years (P = 0.02). Fig. 3 shows the results of outcome comparisons between conventional CPR and CC-CPR following propensity-score matching by age. Among infants, no significant differences were observed

between the two types of bystander CPR with respect to overall 30-day outcomes. When stratified into subgroups, the proportion of 30-day CPC 1–2 was significantly higher after conventional CPR than after CC-CPR in infants with non-cardiac origin (3.5% [21/598] vs. 1.5% [9/581]; P = 0.04) and unwitnessed (1.4% [12/835] vs. 0.4% [3/833]; P = 0.03). However, no significant difference was observed in 30-day CPC 1–2 in patients with cardiac origin and with witnessed arrest. When stratified by initial rhythm, no significant differences were identified between either bystander CPR modality for infants with respect to 30-day outcomes. Of the patients aged 1–17 years, the proportions of overall 30-day survival and 30-day CPC 1 or 2 were significantly higher after conventional CPR than after CC-CPR (survival: 18.6% [250/1341] vs. 12.7% [170/1341]; P < 0.001; CPC 1–2: 9.4% [126/1341] vs. 6.0% [80/1341]; P = 0.001). When stratified into subgroups, proportions of 30-day favourable outcomes were similar for children with cardiac origin and initial shockable rhythm. When stratified by age subgroup, the proportion of 30-day survival was significantly higher after conventional CPR than after CC-CPR for patients aged 8–17 years (20.2% [132/652] vs. 15.8% [102/646], P = 0.04). However, the observed differences were

Please cite this article in press as: Goto Y, et al. Conventional versus chest-compression-only cardiopulmonary resuscitation by bystanders for children with out-of-hospital cardiac arrest. Resuscitation (2017), https://doi.org/10.1016/j.resuscitation.2017.10.015

G Model RESUS-7347; No. of Pages 9

ARTICLE IN PRESS Y. Goto et al. / Resuscitation xxx (2017) xxx–xxx

6 Table 3 Comparison of Outcomes in the Unmatched Cohorts by Age.

Age <1 year Total no. of unmatched patients 30-day survival, n/total n (%) Unadjusted, OR (95% CI) Adjusted§ , OR (95% CI) 30-day CPC 1 or 2, n/total n (%) Unadjusted, OR (95% CI) Adjusted§, OR (95% CI) Age 1–17 years Total no. of unmatched patients 30-day survival, n/total n (%) Unadjusted, OR (95% CI) Adjusted¶ , OR (95% CI) 30-day CPC 1 or 2, n/total n (%) Unadjusted, OR (95% CI) Adjusted¶ , OR (95% CI) Patients aged 1–7 years, n 30-day survival, n/total n (%) Unadjusted, OR (95% CI) Adjusted¶ , OR (95% CI) 30-day CPC 1 or 2, n/total n (%) Unadjusted, OR (95% CI) Adjusted¶ , OR (95% CI) Patients aged 8–17 years, n 30-day survival, n/total n (%) Unadjusted, OR (95% CI) Adjusted¶ , OR (95% CI) 30-day CPC 1 or 2, n/total n (%) Unadjusted, OR (95% CI) Adjusted¶ , OR (95% CI)

Overall unmatched patients

Conventional CPR

2842 227/2842

(8.0)

50/2842

(1.8)

1191 115/1191 Reference Reference 29/1191 Reference Reference

3968 590/3968

(14.9)

272/3968

(6.9)

1816 236/1816

(13.0)

71/1816

(3.9)

2152 354/2152

(16.5)

201/2152

(9.3)

1496 290/1496 Reference Reference 149/1496 Reference Reference 781 133/781 Reference Reference 49/781 Reference Reference 715 157/715 Reference Reference 100/715 Reference Reference

Compression-only CPR

(9.7)

(2.4)

(19.4)

(10.0)

(17.0)

(6.3)

(22.0)

(14.0)

1651 112/1651 0.68 0.66 21/1651 0.52 0.51 2472 300/2472 0.57 0.61 123/2472 0.47 0.50 1035 103/1035 0.54 0.52 22/1035 0.32 0.27 1437 197/1437 0.56 0.72 101/1437 0.46 0.68

(6.8) (0.52–0.89)† (0.49–0.89)† (1.3) (0.29–0.91)* (0.27–0.95)*

(12.1) (0.48–0.68)‡ (0.50–0.74)‡ (5.0) (0.37–0.61)‡ (0.37–0.67)‡ (10.0) (0.41–0.71)‡ (0.38–0.69)‡ (2.1) (0.19–0.54)‡ (0.16–0.46)‡ (13.7) (0.45–0.71)‡ (0.55–0.94)* (7.0) (0.35–0.62)‡ (0.47–0.97)‡

Values are reported as n/total n (%) unless indicated otherwise. CI, confidence interval; CPC, cerebral performance category; CPR, cardiopulmonary resuscitation; OR, odds ratio. * p < 0.05. † p < 0.01. ‡ p < 0.001. § Adjusted for a predefined set of potential 11 confounders: aetiology of cardiac arrest, initial cardiac rhythm, bystander witness status, sex, geographic Japanese regions, year, prehospital advanced medication by attended physician, defibrillation by emergency responder, use of advanced airway management, call-to-response time, and call-to-hospital arrival time. ¶ Adjusted for a predefined set of potential 13 confounders: above mentioned 11 confounders plus age and epinephrine administration.

no longer significant for 30-day CPC 1–2 (12.4% [81/652] vs. 9.8% [63/646], P = 0.13).

4 Discussion Using propensity-score matching analyses, this nationwide population-based observational study analysed the data of a large cohort of paediatric OHCAs from the All-Japan Utstein Registry for 8 years. In unmatched cohorts, CC-CPR was associated with decreased odds of 30-day favourable outcomes compared with conventional CPR. However, after propensity-score matching, CC-CPR had similar neurologically intact survival to conventional CPR in children aged 1–17 years with cardiac aetiology, initial shockable rhythm, or ≥8 years compared with conventional CPR. Further, we demonstrate that CC-CPR was equivalent to conventional CPR for infants with cardiac aetiology or witnessed arrest. Therefore, we propose that CC-CPR by bystanders may provide a reasonable alternative to conventional CPR in certain paediatric OHCAs, and may help to increase the rate of bystander CPR by reducing both procedural complexity as well as barriers to bystander action [13]. Although differences in cardiac arrest aetiology between children and adults necessitate procedural differences in resuscitation technique, no evidence exists identifying a precise age to initiate adult CPR techniques [2]. According to the Guidelines 2000 for CPR and emergency cardiovascular care (ECC) published by AHA with ILCOR [1], an “adult” is defined as any individual ≥8 years of age based largely on practical criteria and ease of teaching. Further, the 2005 AHA Guidelines for CPR and ECC recommended that adult

guidelines for the lay rescuer apply to victims approximately 8 years of age and older; for healthcare providers, adult guidelines apply to post-pubescent victims (approximately 12–14 years of age) [2]. In the 2010 AHA guidelines for CPR and ECC (as well as in the 2015 AHA update), adult BLS guidelines apply during and past puberty [3,4]. The findings in this paper suggest that the current adult BLS guidelines for the lay rescuer may extend to children aged ≥8 years, and children aged 1–17 years with cardiac aetiology of arrest or a shockable rhythm. In the present investigation, conventional CPR was found to be better in infants with cardiac arrest of non-cardiac origin (the proportion among matched patients was 59% [1179/1994]) and unwitnessed status (84% [1668/1994]), in children aged 1–17 years with cardiac arrest of non-cardiac origin (67% [1795/2682]) and non-shockable status (91% [2443/2682]), and in children aged 1–7 years (52% [1384/2682]), compared to CC-CPR in terms of 30-day CPC 1–2 rate. Based on these results, the patients who should receive conventional CPR are infants with unwitnessed status and children aged 1–17 years with non-shockable status, regardless of the aetiology of cardiac arrest. Emergency dispatch centres in Japan have increasingly become more active in relaying CPR instructions to citizens performing CPR [15]. Interestingly, dispatcher-assisted instruction of CPR in Japan was converted from conventional CPR to CC-CPR in 200618 before the AHA recommendation of CC-CPR in 2008 [13]. Critically owing to these EMS efforts, the proportion of CC-CPR has increased significantly, accounting for over 70% of all bystander CPR performed during the study period in both age groups (Supplementary

Please cite this article in press as: Goto Y, et al. Conventional versus chest-compression-only cardiopulmonary resuscitation by bystanders for children with out-of-hospital cardiac arrest. Resuscitation (2017), https://doi.org/10.1016/j.resuscitation.2017.10.015

G Model RESUS-7347; No. of Pages 9

ARTICLE IN PRESS Y. Goto et al. / Resuscitation xxx (2017) xxx–xxx

7

Fig. 2. 30-day Outcomes by Survival and Good Neurological Outcome Following Out-of-Hospital Cardiac Arrest among 4676 Matched Children. A, Patient numbers. B, Infants. C, Children aged 1–17 years. CPC, cerebral performance category; CPR, cardiopulmonary resuscitation. *Cochran-Armitage trend test was performed.

Fig. 3. 30-day Outcomes in Matched Cohorts by Age and Subgroup. A, Survival rate in infants. B, CPC 1–2 rate in infants. C, Survival rate in children aged 1–17 years. D, CPC 1–2 rate in children aged 1–17 years. CPC, cerebral performance category; CPR, cardiopulmonary resuscitation.

Tables S1 and S2). Particularly, the prevalence of CC-CPR for infants with OHCA may contribute to the increase of 30-day survival in

OHCA infants (Supplementary Tables S1). Although the proportion of infants with witnessed arrest was minor (16.8% [478/2842]), the

Please cite this article in press as: Goto Y, et al. Conventional versus chest-compression-only cardiopulmonary resuscitation by bystanders for children with out-of-hospital cardiac arrest. Resuscitation (2017), https://doi.org/10.1016/j.resuscitation.2017.10.015

G Model RESUS-7347; No. of Pages 9 8

ARTICLE IN PRESS Y. Goto et al. / Resuscitation xxx (2017) xxx–xxx

present results indicate that CC-CPR was equivalent to conventional CPR for infants with witnessed arrest; this finding demonstrates that the advantages of hemodynamic maintenance by continuous chest compression far surpass any disadvantages due to insufficient blood oxygen saturation. Based on the adjusted OR of CC-CPR for 30-day outcomes in matched cohorts, we calculated estimated 30-day outcomes (Supplementary Table S3). If we only performed CC-CPR in all children with OHCA, the rates of 30-day outcomes would be significantly decreased in children aged 1–7 years (survival rate: from 13.4% to 8.4%, CPC 1–2 rate: from 4.5% to 2.0%, all P < 0.001); however, the rates would not change in infants and children aged 8–17 years. The results of the present investigation are consistent with those from a study by Kitamura [14], which analysed an identical Japanese database from 2005 to 2007. Fukuda [16] evaluated Japanese children aged 1–17 years with OHCA between 2011 and 2012, and demonstrated that CC-CPR had effects on 30-day CPC 1–2 similar to those obtained with conventional CPR, regardless of arrest aetiology, witness status, or age subgroup. Interestingly, these results are somewhat inconsistent with the present findings. Unlike Fukuda’s study, we excluded children who were not treated by EMS personnel because the accuracy of OHCA is not fully verified before EMS arrival. Moreover, we accounted for several prehospital cofounding variables including geographic region, because regional disparities in prehospital care and in-hospital post-resuscitation care are prominent in Japan [23,24]. As shown in Fig. 2-C, no significant differences in 30-day CPC 1–2 were found between two types of bystander CPR in patients from 2011 to 2012, which was consistent with Fukuda’s findings. Furthermore, the rate of 30-day CPC 1–2 gradually improved in children receiving CC-CPR from the 2007–2008 period to the 2013–2014 period. Possible explanations for these results include the nationwide dissemination of the following recommendations based on the 2010 international CPR guidelines update: (1) CC-CPR with high-quality assist by a dispatcher on the phone, (2) change from A-B-C to C-A-B sequence for CPR, and (3) improvement of post-resuscitation care (e.g. targeted temperature management). Naim [8] demonstrated that conventional CPR was superior to CC-CPR in infants with respect to neurologically intact survival. In the present investigation, conventional CPR in infants was associated with an increased likelihood of favourable outcomes compared with CC-CPR in a regression analytical model; this treatment superiority was only observed in cases with non-cardiac aetiology or unwitnessed arrest after propensity-score matching. This finding may be attributable to inherent differences between EMS systems or post-cardiac arrest care. This observational study has several potential limitations. First, the actual aetiology of cardiac arrest was not fully verified. Some infants may have had sudden infant death syndrome, a common aetiology for arrest followed by trauma and respiratory disease [25]. A nationwide school-based ECG screening program for cardiovascular diseases has been developed for all first, seventh, and tenth graders since 1994 in Japan [26]. However, combined analyses of data for sudden cardiac death and/or OHCA and data from the school-based screening program have not been performed. Second, although the duration of bystander CPR prior to EMS arrival may have influenced the patient outcomes [27], our analysis could not account for this issue. Third, owing to the retrospective nature of the study, the data lacked sufficient detail required to perform further risk adjustment for outcomes (e.g., comorbid diseases, location of arrest, CPR quality, and in-hospital medication). Finally, caution must be exercised when generalizing these results to additional EMS systems, as a relatively infrequent use of epinephrine (<5% in the present study) was observed, compared with prominent epinephrine use in the United States (65%) [10]. Therefore, an adequately powered randomized controlled trial will be required

to determine the role of CC-CPR by bystanders for children with OHCA. 5 Conclusions Conventional CPR was associated with improved outcomes compared to CC-CPR in the majority of the paediatric subgroups following OHCA. CC-CPR by bystander was associated with similar neurologically intact survival to conventional CPR for children with OHCA aged ≥8 years, for children aged 1–17 years with cardiac aetiology or initial shockable rhythm, and for infants with cardiac aetiology or witnessed arrest. Funding sources This work was supported by the Japan Society for the Promotion of Science (KAKENHI Grant Number 15K08543), which had no role in the design and implementation of the study, analysis and interpretation of the data, or approval of the manuscript. Conflicts of interest None. Acknowledgement We thank all the EMS personnel and participating physicians in Japan and FDMA for their generous cooperation in establishing and maintain the database. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.resuscitation.2017. 10.015. References [1]. The American Heart Association in collaboration with the International Liaison Committee on Resuscitation (ILCOR). Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Part 3: adult basic life support. The American Heart Association in collaboration with the International Liaison Committee on Resuscitation. Circulation 2000;102:I22–59. [2]. American Heart Association. American Heart Association Guideline for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2005;112:IV1–203. [3]. Berg MD, Schexnayder SM, Chameides L, et al. Part 13: paediatric basic life support: 2010 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2010;122:S862–75. [4]. Travers AH, Perkins GD, Berg RA, et al. Part 3: Adult basic life support and automated external defibrillation: 2015 International consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Circulation 2015;132:S51–83. [5]. de Caen AR, Maconochie IK, Aickin R, et al. Pediatric basic life support and pediatric advanced life support chapter collaborators. part 6: pediatric basic life support and pediatric advanced life support: 2015 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Circulation 2015;132:S177–203. [6]. Donoghue AJ, Nadkarni V, Berg RA, et al. Out-of-hospital paediatric cardiac arrest: an epidemiological review and assessment of current knowledge. Ann Emerg Med 2005;46:512–22. [7]. Atkins DL, Everson-Stewart S, Sears GK, et al. Epidemiology and outcomes from out-of-hospital cardiac arrest in children. The resuscitation outcomes consortium epistry–cardiac arrest. Circulation 2009;119:1484–91. [8]. Naim MY, Burke RV, McNally BF, et al. Association of bystander cardiopulmonary resuscitation with overall and neurologically favourable survival after paediatric out-of-hospital cardiac arrest in the United States: a report from the cardiac arrest registry to enhance survival surveillance registry. JAMA Pediatr 2017;171:133–41. [9]. Goto Y, Funada A, Goto Y. Duration of prehospital cardiopulmonary resuscitation and favourable neurological outcomes for paediatric out-of-hospital cardiac arrests: a nationwide, population-based cohort study. Circulation 2016;134:2046–59.

Please cite this article in press as: Goto Y, et al. Conventional versus chest-compression-only cardiopulmonary resuscitation by bystanders for children with out-of-hospital cardiac arrest. Resuscitation (2017), https://doi.org/10.1016/j.resuscitation.2017.10.015

G Model RESUS-7347; No. of Pages 9

ARTICLE IN PRESS Y. Goto et al. / Resuscitation xxx (2017) xxx–xxx

[10]. Fink EL, Prince DK, Kaltman JR, et al. Unchanged paediatric out-of-hospital cardiac arrest incidence and survival rates with regional variation in North America. Resuscitation 2016;107:121–8. [11]. Inoue M, Tohira H, Williams T, et al. Incidence, characteristics and survival outcomes of out-of-hospital cardiac arrest in children and adolescents between 1997 and 2014 in Perth, Western Australia. Emerg Med Australas 2017;29:69–76. [12]. Rajan S, Wissenberg M, Folke F, et al. Out-of-hospital cardiac arrests in children and adolescents: incidences, outcomes, and household socioeconomic status. Resuscitation 2015;88:12–9. [13]. Sayre MR, Berg RA, Cave DM, et al. Hands-only (compression-only) cardiopulmonary resuscitation: a call to action for bystander response to adults who experience out-of-hospital sudden cardiac arrest: a science advisory for the public from the American Heart Association Emergency Cardiovascular Care Committee. Circulation 2008;117:2162–7. [14]. Kitamura T, Iwami T, Kawamura T, et al. Conventional and chest-compressiononly cardiopulmonary resuscitation by bystanders for children who have outof-hospital cardiac arrests: a prospective, nationwide, population-based cohort study. Lancet 2010;375:1347–54. [15]. Goto Y, Maeda T, Goto Y. Impact of dispatcher-assisted bystander cardiopulmonary resuscitation on neurological outcomes in children with out-of-hospital cardiac arrests: a prospective, nationwide, population-based cohort study. J Am Heart Assoc 2014;3:e000499. [16]. Fukuda T, Ohashi-Fukuda N, Kobayashi H, et al. Conventional versus compression-only versus no-bystander cardiopulmonary resuscitation for paediatric out-of-hospital cardiac arrest. Circulation 2016;134:2060–70. [17]. Kitamura T, Iwami T, Kawamura T, Nagao K, Tanaka H, Hiraide A. For the Implementation Working Group for All-Japan Utstein Registry of the Fire and Disaster Management Agency. Nationwide public access defibrillation in Japan. N Engl J Med 2010;362:994–1004. [18]. Fire and Disaster Management Agency of Japan, http://www.n-bouka.or.jp/ netnews/images/06 09/003 02.pdf. Accessed July 4, 2017 A report on advancements of emergency medical service systems [in Japanese]; 2006.

9

[19]. Jacobs I, Nadkami V, Bahr J, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries: a statement for healthcare professionals from a task force of the International Liaison Committee on Resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, Inter American Heart Foundation, Resuscitation Councils of Southern Africa). Circulation 2004;110:3385–97. [20]. Rosenbaum PR, Donald BR. Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. Am Stat 1985;39:33–8. [21]. Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res 2011;46:399–424. [22]. Normand ST, Landrum MB, Guadagnoli E, et al. Validating recommendations for coronary angiography following acute myocardial infarction in the elderly: a matched analysis using propensity scores. J Clin Epidemiol 2001;54:387–98. [23]. Okamoto Y, Iwami T, Kitamura T, et al. Regional variation in survival following paediatric out-of-hospital cardiac arrest. Circ J 2013;77:2596–603. [24]. Hasegawa K, Tsugawa Y, Camargo Jr CA, Hiraide A, Brown DF. Regional variability in survival outcomes of out-of-hospital cardiac arrest: the All-Japan Utstein Registry. Resuscitation 2013;84:1099–107. [25]. Gerein RB, Osmond MH, Stiell IG, Nesbitt LP, Burns S, OPALS Study Group. What are the aetiology and epidemiology of out-of-hospital paediatric cardiopulmonary arrest in Ontario, Canada? Acad Emerg Med 2006;13:653–8. [26]. Yoshinaga M. Prevalence of sudden death and out-of-hospital cardiac arrest in infants, children, and adolescents; what does it imply? Circ J 2013;77:2475–6. [27]. Iwami T, Kawamura T, Hiraide A, et al. Effectiveness of bystander-initiated cardiac-only resuscitation for patients with out-of-hospital cardiac arrest. Circulation 2007;116:2900–7.

Please cite this article in press as: Goto Y, et al. Conventional versus chest-compression-only cardiopulmonary resuscitation by bystanders for children with out-of-hospital cardiac arrest. Resuscitation (2017), https://doi.org/10.1016/j.resuscitation.2017.10.015