- Email: [email protected]
The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study
G Model
ARTICLE IN PRESS
RESUS 7044 1–6
1
Resuscitation xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
2
The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study夽
3
4
5
Q1
6 7 8 9 10 11
Monica Parry a,∗ , Kyle Danielson a , Sarah Brennenstuhl a , Ian R. Drennan b,c , Laurie J. Morrison d a
Lawrence S. Bloomberg Faculty of Nursing, University of Toronto, Toronto, Canada Division of Emergency Medicine, Department of Medicine, University of Toronto, Toronto, Canada c Institute of Medical Science, Faculty of Medicine, University of Toronto, Toronto, Canada d Robert and Dorothy Pitts Chair in Acute Care and Emergency Medicine, Li Ka Shing Knowledge Institute, St. Michael’s Hospital Division of Emergency Medicine, Department of Medicine, University of Toronto, Toronto, Canada b
12
13 26
a r t i c l e
i n f o
a b s t r a c t
14 15 16 17 18 19
Article history: Received 9 September 2016 Received in revised form 24 December 2016 Accepted 13 January 2017
20
25
Keywords: Out-of-hospital cardiac arrest Diabetes Return of spontaneous circulation Survival
27
Introduction
21 22 23 24
28Q2 29 30 31 32 33 34 35
Background: Sudden cardiac arrest (SCA), confirmed absence of cardiac mechanical activity, is the leading cause of heart-related death in the US. Almost 85% of SCA occur out-of-hospital (OHCA), with very poor rates of return of spontaneous circulation (ROSC) and survival to hospital discharge. We sought to determine if diabetes status was associated with survival or ROSC following an OHCA. Methods: We completed a retrospective cohort study using data from the Toronto Regional RescuNet Epistry dataset, based upon data definitions defined by the Resuscitation Outcomes Consortium (ROC) Epistry-Cardiac Arrest and the Strategies for Post Arrest Resuscitation Care (SPARC) network datasets. Adults ≥18 years of age who experienced an OHCA, had data on diabetes status, and were treated by Emergency Medical Services (EMS) between 2012–2014 were included in the analysis (n = 10,097). We used bivariate analyses to examine relationships between diabetes status, Utstein elements and outcomes, and logistic regression to determine predictors of survival. Results: Diabetes prevalence was 27.8% (95% CI: 27.0–28.7). A larger proportion of those with diabetes had a non-shockable initial rhythm (28.8% vs. 25.1%; p < 0.01) and did not survive to hospital discharge (92.1% vs. 89.2%, p < 0.001). Diabetes status is associated with a decrease in survival, independent from a number of Utstein elements (adjusted OR = 0.76; 95% CI: 0.64–0.91, p = 0.003). Conclusions: This is the first Canadian study to examine the association between diabetes status and OHCA outcomes. Our findings suggest that diabetes status prior to arrest is associated with decreased survival. The growing prevalence of diabetes globally suggests a future burden related to OHCAs. © 2017 Elsevier B.V. All rights reserved.
Sudden cardiac arrest (SCA), confirmed absence of cardiac mechanical activity, is a significant public health issue and the leading cause of heart-related death in the United States (US).1 Approximately 85% of SCA occur out-of-hospital, leading to 326,200 deaths each year in the US1 and one death every 13 min in Canada.2 Of those out-of-hospital cardiac arrests (OHCAs) treated by Emergency Medical Services (EMS), only 10.6% survive to hospital discharge and 8.3% have favorable neurological outcomes.1
夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at DOI:10.1016/j.resuscitation.2017.01.011. ∗ Corresponding author at: Lawrence S. Bloomberg Faculty of Nursing, 155 College Street, Suite 130, Toronto, ON, Canada. Fax: +1 416 978 8222. E-mail address: [email protected] (M. Parry).
Several core Utstein elements have consistently predicted survival among EMS-treated arrests, including age, gender, witnessed arrest, arrest location, bystander cardiopulmonary resuscitation (CPR), first monitored rhythm, and EMS response times.3,4 These variables have been shown to predict approximately 72% of the variation in survival3,4 however, a portion of the variation remains unaccounted for, suggesting that there are unknown patient variables associated with survival after an OHCA. Identification of these variables would be an opportunity to reduce the public health burQ3 den of OHCA.4 The most common preventable comorbidities associated with SCA include coronary artery disease, myocardial infarction, and heart failure.5 Diabetes is a risk factor for all these comorbidities and has been shown to be associated with a 2–4 fold increase in the risk of SCA.6–8 Cardiac autonomic neuropathy, a common and serious complication of type 1 (T1) and type 2 (T2) diabetes,9–11 results from damage to autonomic fibers that innervate the heart and blood
http://dx.doi.org/10.1016/j.resuscitation.2017.01.011 0300-9572/© 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52
G Model RESUS 7044 1–6
ARTICLE IN PRESS M. Parry et al. / Resuscitation xxx (2017) xxx–xxx
2
88
vessels and increases the risk of SCA.12 Cardiac autonomic neuropathy has been linked to resting tachycardia, prolonged QT interval, increased silent myocardial ischemia and infarction, and increased mortality.13,14 Results from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial confirm that cardiac autonomic neuropathy is associated with a 1.55–2.95 times risk of all-cause and cardiovascular disease (CVD) mortality independent of multiple CVD risk factors.15 Others report that men with T2 diabetes have a 2.26-fold increase in SCA risk compared to men without diabetes, even after adjusting for age, body mass index (BMI), systolic blood pressure (SBP), smoking, cholesterol, known coronary artery disease (CAD), and family history of CAD.16 Globally, 8.3% of individuals aged 20–79 years of age17 and 25.9% of individuals over 65 years of age in the US have diabetes.18 In 2015, 9.3% of Canadians had diabetes, and this is projected to increase by 44% in the next decade.19 Diabetes is one of the largest public health emergencies of the 21st century. The World Health Organization (WHO) suggests diabetes is the third highest risk factor for premature mortality in the world20 ; it is estimated that slightly more men than women have diabetes, and one in two individuals remain undiagnosed due to misinterpretation or lack of symptoms early in the course of disease.21 Diabetes complications are common, and while there are no internationally agreed upon standards for diagnosing or measuring diabetes complications, it is estimated that up to 50% of individuals have complications at the time of diagnosis.22 Diabetes and its complications are a major cause of early mortality, and existing health statistics underestimate the number of these deaths.23 Better understanding the burden of diabetes and its complications would help direct public health actions, and improve care for all people with diabetes.24 The primary objective of this study was to determine the association between diabetes status (T1 and T2) and survival to hospital discharge from OHCA. The secondary objectives were to determine the association between diabetes status, and any ROSC and neurological outcome, at hospital discharge.
89
Methods
90
Study design and population
53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87
108
We completed a retrospective cohort study using data from the Toronto Regional RescuNet Epistry dataset and definitions developed by the Resuscitation Outcomes Consortium (ROC) Epistry-Cardiac Arrest and the Strategies for Post Arrest Resuscitation Care (SPARC) network datasets.25,26 This dataset is a population-based registry of consecutive OHCA attended by 911initiated EMS first responders in southern Ontario, serving a population of over 6.6 million residents in Durham, Halton, Muskoka, Peel, Simcoe, Toronto and York.27 Data are collected from pre-hospital call reports from seven land EMS agencies, fire departments, and one air ambulance service, and from in-hospital records from 44 destination hospitals and entered into a secure database. The Research Ethics Board at each participating hospital has given ethics approval for all retrospective studies using the Toronto RescuNet Epistry-Cardiac Arrest database. Adults ≥18 years of age who experienced an OHCA, had data on diabetes status, and were treated by EMS between 2012–2014 were included in the analysis (n = 10,097).
109
Measures
91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107
110 111 112
Exposure Diabetes status was ascertained using a variable based on the inhospital record (yes, no, not noted). If this information was missing
or status was not noted, data on diabetes status was obtained from the pre-hospital call report.
Covariates The Utstein elements are the most widely used set of variables closely associated with OHCA outcomes.3 The following Utstein elements were obtained and used in our analysis: age (years), gender, witness status (EMS/bystander witnessed, unwitnessed), location (public, private/residential, other),28 first response CPR (bystander, EMS, none) bystander application of automatic external defibrillator (AED), first monitored rhythm (ventricular fibrillation and ventricular tachycardia [shockable], or pulseless electrical activity and asystole [non-shockable]), EMS response time (call to 911 to arrival on scene time interval), and etiology of arrest. Etiology was defined as either obvious cause (e.g., drug overdose, asphyxiation, drowning) or presumed cardiac cause when there was no identifiable obvious cause of arrest.
Outcomes Survival to hospital discharge (yes/no) was our primary outcome. Secondary outcomes included any ROSC during the resuscitation attempt (yes/no) and neurological outcome at hospital discharge based on the modified Rankin Scale (mRS) of 0–2 (favourable) or 3–5 (unfavourable).29 The amount of missing data for most variables was negligible, ranging from 0% to about 2%, with the exception of EMS response time, which had 10.2% missing data (Table 1). The missing data occurred in cases where the first EMS responders were not from RescuNET participating services. Missing data was less than 1% for this variable in all cases where RescuNET participating EMS responders arrived first.
Statistical analysis Univariate statistics, including frequency counts, percentages and means, were used to describe the prevalence of diabetes and other characteristics of the study population. Bivariate associations between diabetes status and each of the Utstein data elements and outcomes were assessed using 2 tests and t-tests, as appropriate. Multivariable logistic regression was used to investigate the relationship between diabetes status and each of the outcomes (separately) while adjusting for the Utstein data elements. Preliminary steps included performing model diagnostics, such as assessing multicollinearity using the variance inflation factor (VIF) and testing for non-linearities between the predictors and outcome. There was no evidence of multicollinearity, but we did determine the need for a non-linear specification of age, which was represented using age and age2 in the modeling. Next, we tested for plausible interactions between diabetes status and age, gender, whether arrest was witnessed and location of arrest, but none were found. To build a final model, we employed a hierarchical approach, starting with modeling diabetes status alone, then cumulatively adding clusters of variables. The Utstein elements were added in three clusters: demographics (age, gender), circumstances of arrest (witnessed, location, first-response CPR, etiology), followed by first monitored rhythm. We excluded response times from the final model due to the large amount of missing data. Nested models were compared using the likelihood ratio test (LRT). Clusters were left in the model if they significantly contributed to model fit based on the LRT. Only 4.2% of cases were missing data for one or more of the covariates and could not be included in the modeling. Data analysis was undertaken using SPSS version 23. A two sided p-value of p < 0.05 was used to establish statistical significance.
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
113 114
115 116 117 118 119 120 121 122 123 124 125 126 127 128
129 130 131 132 133 134 135 136 137 138 139 140 141
142
143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
G Model
ARTICLE IN PRESS
RESUS 7044 1–6
M. Parry et al. / Resuscitation xxx (2017) xxx–xxx Table 1 Patient characteristics by diabetes status.
3
Q5
N% Event year 2012, n (%) 2013, n (%) 2014, n (%) Age Mean age, years Median age, years (IQR) Gender Male, n (%) Female, n (%) Witnessed status Witnessed, n (%) Not witnessed, n (%) Location Public, n (%) Private/Residential, n (%) Other (e.g., hotel & nursing home), n (%) Site Toronto, n (%) Durham, n (%) Simcoe, n (%) Muskoka, n (%) Peel, n (%) Halton, n (%) York, n (%) First-response CPR Bystander, n (%) EMS, n (%) None, n (%) Bystander AED applied, n (%) First monitored rhythm Ventricular fibrillation, n (%) PEA, n (%) Asystole, n (%) Shockable, n (%) Not Shockable, n (%) Other, n (%) Etiology No obvious cause, presumed cardiac, n (%) Obvious cause, presumed non-cardiac, n (%) Median EMS response time interval, mins (IQR)
Diabetes
No diabetes
2812 (27.8)
7285 (72.2)
p Value
Missing, n (%)
889 (31.6) 936 (33.3) 987 (35.1)
2343 (32.2) 2466 (33.9) 2476 (34.0)
ns
0 (0)
72 74 (18)
69 72 (26)
<0.001 <0.001
0 (0) 0 (0)
1761 (62.6) 1051 (37.4)
4574 (62.8) 2711 (37.2)
ns
0 (0)
1432 (50.9) 1380 (49.1)
3676 (50.5) 3608 (49.5)
ns
1 (0)
241 (8.7) 2282 (82.0) 259 (9.3)
673 (9.4) 5946 (83.0) 7.6 (7.6)
=0.01
154 (1.5)
1265 (45.0) 249 (8.9) 198 (7.0) 30 (1.1) 565 (20.1) 150 (5.3) 355 (12.6)
3219 (44.2) 759 (10.4) 649 (8.9) 75 (1.0) 1158 (15.9) 467 (6.4) 958 (13.2)
<0.001
0 (0)
1006 (35.8) 376 (14.1) 1409 (50.1) 66 (6.6)
2790 (38.3) 887 (12.2) 3607 (49.5) 181 (6.5)
=0.008 ns
27 (0.3)
416 (15.1) 673 (24.4) 1308 (47.4) 67 (2.4) 269 (9.7) 27 (1.0)
1207 (17.0) 1652 (23.3) 3172 (44.7) 239 (3.4) 740 (10.4) 80 (1.1)
=0.008
247 (2.4)
2666 (94.8) 146 (5.2) 6.0 (2.5)
6500 (89.2) 785 (10.8) 6.0 (2.4)
<0.001 ns
0 (0) 1071 (11.9)
0 (0)
ns, non-significant; IQR, interquartile range; CPR, cardiopulmonary resuscitation; EMS, emergency medical services; AED, automated external defibrillator; PEA, pulseless electrical activity; mins, minutes.
172
173 174
Results In total, there were 11,765 OHCA treated by EMS between 2012 and 2014. We excluded 209 cases less than 18 years of age. Another OHCA in database from 2012 – 2014 N = 20,371
1459 cases were excluded because of missing data on diabetes status (12.6% of otherwise eligible cases) (Fig. 1). Those with missing data on diabetes were younger (61 vs. 70 years), more often male (75.4% vs. 62.7%), had more cardiac arrests in public location (33.8%
Exclude Not treated by EMS N = 8,606
Cases Treated by EMS N = 11,765 Exclusion Criteria < 18 years of age (n = 209); excluded missing diabetes status (n=1,459) Eligible cases N = 10,097
Fig. 1. CONSORT diagram of included and excluded cases.
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
175 176 177 178
G Model
ARTICLE IN PRESS
RESUS 7044 1–6
M. Parry et al. / Resuscitation xxx (2017) xxx–xxx
4 Table 2 Survival by diabetes status.
Survival to hospital discharge Survived n (%) Died n (%) Any ROSC, n (%) Neurologic outcome, mRS score mRS 0–2, Favorable outcome n (%) mRS 3–5, Poor outcome n (%)
Diabetes
No diabetes
p Value
Missing, n (%)
221 (7.9) 2586 (92.1) 883 (31.4)
786 (10.8) 6493 (89.2) 2422 (33.2)
<0.001 ns
11 (0.1) 0 (0)
183 (88.0) 25 (12.0)
691 (91.3) 66 (8.7)
ns
42 (4.2)
ns, non-significant; ROSC, return of spontaneous circulation; mRS, modified Rankin Scale.
Table 3 Logistic regression of survival among adults with diabetes who had an out of hospital cardiac arrest treated by EMS in Ontario from 2012 to 2014 (n = 9677). Model 1 Unadjusted
OR
95% CI
Diabetes diagnosis 0.70 (0.60, 0.83) Age in years 2 Age in years Male gender Location (ref = Private/Residential) Public Other First-response CPR (ref = none) Bystander EMS Witnessed Obvious cause Shockable first monitored rhythm Chi-square 19.37 (df = 1) 19.37 (df = 1) Chi square 6127.34 −2 Loglikelihood
Model 2 Adjusted for demographics
Model 3 Adjusted for demographics and circumstances of arrest
Model 4 Adjusted for demographics, circumstances of arrest, and first monitored rhythm
OR
95% CI
OR
95% CI
OR
95% CI
0.78 0.95 0.99 1.62
(0.66, 0.93) (0.94, 0.95) (0.99, 0.99) (1.39, 1.90)
0.76 0.95 0.99 1.37
(0.64, 0.91) (0.94, 0.96) (0.99, 0.99) (1.16, 1.62)
0.88 0.96 1.00 0.95
(0.72, 1.06) (0.95, 0.97) (0.99, 1.00) (0.79, 1.14)
3.92 1.07
(3.27, 4.69) (0.74, 1.55)
2.59 1.16
(2.12, 3.15) (0.77,1.73)
1.20 2.29 4.38 1.02
(1.01, 1.43) (1.87, 2.82) (3.59, 5.34) (0.80, 1.31)
0.95 (0.79, 1.15) 2.99 (2.38, 3.75) 2.70 (2.19, 3.33) 2.17 (1.65, 2.85) 10.58 (8.80, 12.74) 1980.01 (df = 11) 707.98 (df = 1) 4166.71
484.77 (df = 4) 465.40 (df = 3) 5661.96
1272.02 (df = 10) 787.26 (df = 6) 4874.69
CPR, cardiopulmonary resuscitation; EMS, emergency medical services.
179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201
vs. 9.2%), and more often had cardiac arrests of obvious cause compared to those included in the analysis. Cases with missing diabetes status also survived to hospital discharge less often (3.9% vs. 10%). Among the 10,097 OHCAs included in the analysis, 2812 cases had a diagnosis of diabetes, resulting in a prevalence of 27.8% (95% CI: 27.0–28.6). Cases were distributed across seven sites including Toronto (44.4%), Peel (17.1%), York (13.0%), Durham (10.0%), Simcoe (8.4%), Halton (6.1%), and Muskoka (1.0%). Of these, most occurred in residential locations (82.8%) and just over half (50.6%) were witnessed. Bystander CPR was attempted in over a third of cases (37.6%) and a bystander applied an AED infrequently (6.5%). In just over one in ten, first-response CPR was preformed by EMS (12.7%). Approximately 16% of cases had an initial cardiac rhythm of ventricular fibrillation (shockable), and 44.4% of cases had asystole and 23.0% had pulseless electrical activity (non-shockable). Those with diabetes were slightly older (72 vs. 69, p < 0.001), had a presumed cardiac cause for arrest (i.e., no obvious cause) (94.8% vs. 89.2%, p < 0.001), more often had an EMS-witnessed arrest (14.1% vs. 12.2%, p < 0.01), and presented with a non-shockable rhythm (28.8% vs. 25.1%, p < 0.01) compared to those without diabetes. There were no differences in diabetes status by gender, whether arrest was witnessed by bystanders, or in EMS response times (Table 1).
30% lower odds of survival (OR = 0.70, 95% CI: 0.60–0.83). When the relationship is adjusted for patient demographics (model 2) or together with circumstances of the arrest (model 3), the lower odds of survival among those with diabetes decreases slightly, but the association remains significant (model 2 OR = 0.78, 95% CI: 0.66–0.93; model 3 OR = 0.76, 95% CI: 0.64–0.91). When the first monitored rhythm is also included, the odds ratio remains below one, but becomes non-significant (model 4 OR = 0.88, 95% CI: 0.72–1.06). While response time was not included in the final modeling due to a large amount of missing data in cases where the first EMS responders were not from RescuNET, a supplementary analysis showed that, when included in a similar series of models, the pattern of findings remained unchanged (data available upon request). Any ROSC during resuscitation Approximately 31% (n = 883) of individuals with diabetes and 33% (n = 2422) of individuals without diabetes had any ROSC, with no group differences by diabetes status (Table 2). Multivariable modeling confirmed that there was no relationship between diabetes status and ROSC (data available on request). Neurological outcome at hospital discharge
202
203 204 205 206 207
208 209 210 211 212 213 214 215 216 217 218 219 220 221
222
223 224 225 226 227
228
Survival to hospital discharge Differences in outcomes between those with and without diabetes are presented in Table 2. A smaller proportion of those with diabetes survived to hospital discharge (7.9% vs. 10.8%, p < 0.001). Further investigation of this relationship using logistic regression (Table 3), shows that when unadjusted, diabetes is associated with
Exactly 88% (n = 183) of individuals with diabetes and 91.3% (n = 691) of individuals without diabetes had favorable neurological outcomes, with no group differences by diabetes status (Table 2). Multivariable modeling confirmed that there was no relationship between diabetes status and neurological outcome (data available on request).
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
229 230 231 232 233 234
G Model RESUS 7044 1–6
ARTICLE IN PRESS M. Parry et al. / Resuscitation xxx (2017) xxx–xxx
235
236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299
Discussion In this study, 27.8% of individuals had a diagnosis of diabetes. These results are similar to Ro30 who reported that 26.7% (n = 370) of individuals who had an OHCA with a presumed cardiac cause had diabetes. In our cohort, individuals with diabetes had significantly lower odds of survival to hospital discharge, even when adjusting for demographics (age, gender) and circumstances of the arrest (location, resuscitation attempted, witnessed, and obvious cause). However, when initial rhythm was included in the model, the association between diabetes status and survival lost significance. This may reflect, in part, a greater number of individuals with diabetes presenting with a non-shockable first monitored rhythm; a finding also demonstrated in other research.31 More clearly, however, it underlines the large effect that initial rhythm has on survival from OHCA, even when adjusting for diabetes status and other Utstein elements (AOR, 10.58, 95% CI: 8.80–12.74). The importance of first monitored rhythm for predicting survival has been shown in other studies.32,33 Our study adds to the literature by comparing survival from an OHCA in individuals with and without diabetes in Canada, and it is the first study to describe these outcomes in relation to existing Utstein elements. The finding that diabetes status is related to survival with an unadjusted analysis is consistent with previous research.34,35 Nehme et al.34 reported that individuals with diabetes had reduced rates of survival to hospital discharge compared to individuals without diabetes (6.8% vs. 13.4%, p < 0.001) and individuals with diabetes had reduced odds of good functional recovery at one year (OR, 0.57, 95% CI: 0.35–0.95). Jang et al.35 found diabetes status was negatively associated with survival (AOR, 0.84, 95% CI: 0.75–0.95) and neurological recovery (AOR, 0.81, 0.67–0.97), and diabetes modified the effect on OHCA survival depending on the presence or absence of pre-existing cardiac disease. Cardiac disease alone had a non-significant effect on survival and neurological recovery, but when combined with diabetes, odds of survival (AOR, 0.58, 0.45–0.74) and neurological recovery (AOR, 0.52, 95% CI: 0.36–0.75) were significantly lower. The results of this study demonstrate that OHCA in individuals with diabetes are more likely to present with non-shockable rhythms, which have the lowest survival rate.32 However, others report that diabetes also reduced survival to hospital discharge for individuals presenting with a shockable rhythm (AOR, 0.57, 95% CI: 0.38–0.86).34 A convergence of recent evidence helps to explain the link between diabetes and survival from OHCA. While the interrelationship between glucose metabolism, autonomic function and cardiac disease in diabetes are not well understood, they are thought to be related to a prolonged QT interval. The QT interval on an electrocardiogram (ECG) measures the total time for ventricular depolarization and repolarization. A prolonged QT interval corrected for heart rate (QTc) is predictive of mortality in in healthy adults.36 It has been positively associated with HbA1c concentration ( = 4, 95% CI: 2–6), and recent evidence suggests a lower threshold (QTc ≥ 409 milliseconds) is associated with autonomic dysfunction in individuals with diabetes.37 Cardiac autonomic neuropathy should be considered in models of CVD and mortality risk in diabetes, and although clinical symptoms occur late in diabetes, research suggests subclinical cardiac autonomic neuropathy can be detected within two years of diagnosis of T1 diabetes and one year of diagnosis of T2 diabetes.38 Results from the ACCORD trial suggests a combination of heart rate variability (HRV) and QT interval measurements derived from a standard ECG could identify a subset of individuals with diabetes at increased mortality risk independent of traditional CVD risk factors.15 We found no group differences in neurological outcomes by diabetes status. This finding is inconsistent with others34,35 who found diabetes was negatively associated with neurological recov-
5
ery. Ro et al.39 also reported that diabetes modified the positive effect of targeted temperature management (TTM) to 32–34 ◦ C on survival and neurological outcomes after OHCA in individuals with diabetes. That is, TTM was associated with good neurological recovery in individuals without diabetes following an OHCA, but not in individuals with diabetes.39 Limitations Approximately, 12.6% (n = 1459) of otherwise eligible cases had to be excluded from the analysis because of missing data on diabetes status. Excluded cases were younger, more frequently male, more often had an arrest in a public location with an obvious non-cardiac cause and less frequently survived. Based on these characteristics, it appears that some selection bias might have occurred. Because T2 diabetes is so often undiagnosed (almost 50% of cases, globally),40 a portion of those excluded likely had diabetes. Excluding those without data on diabetes may have made our analysis more conservative by removing cases with lower survival. Diabetes status was measured dichotomously; type and duration of diabetes, treatment, glycemic control, and presence of complications could not be considered in the analyses. Data on past CVD history was initially considered, but because of a large amount of missing data this was not controlled for in the analyses, which may indicate some residual confounding. Conclusion This is the first Canadian study that has examined the association between diabetes status and OHCA outcomes. Our findings indicate that diabetes is associated with decreased survival and the growing prevalence of diabetes globally suggests future burden related to OHCAs. While information about initial rhythm may trump that of diabetes status when predicting survival, knowing more about diabetes risk (e.g., pre-diabetes), diabetes control (e.g., HbA1c) and diabetes complications (e.g., renal function and proteinuria) may help to target interventions to prevent a SCA in these subgroups of individuals. Diabetes is associated with autonomic dysfunction and cardiac microvascular complications that increase the risk of SCA. Electrocardiogram screening for long QTc and complication risk of diabetes may be of value in routine complication assessments, but further research is needed in this area. Conflict of interest statement None. Acknowledgments This study was possible through grants from the Canadian Insti- Q4 tutes of Health Research (CIHR), the Laerdal Medical Foundation and the Heart and Stroke Foundation of Canada. Kyle Danielson was supported by a Canada Graduate Scholarship Master’s Award from CIHR and Ian Drennan was supported by a CIHR Banting and Best Doctoral Research Award. The RescuNet Epistry dataset was funded by the Laerdal Foundation, CIHR, the Heart and Stroke Foundation of Canada, Defence Research and Development Canada and the US National Institutes of Health funding for the Resuscitation Outcomes Consortium. The authors would like to thank Ms. Chantelle Nielson who contributed to the initial Rescu Writing Group Proposal while a student in the Undergraduate Student Summer Research Program at the Lawrence S. Bloomberg Faculty of Nursing, University of Toronto, Toronto, Ontario, Canada. The authors would also like to thank the Rescu Epistry investigators and emergency medical service oper-
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
300 301 302 303 304 305
306
307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
323
324 325 326 327 328 329 330 331 332 333 334 335 336 337
338
339
340
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356
G Model RESUS 7044 1–6
ARTICLE IN PRESS M. Parry et al. / Resuscitation xxx (2017) xxx–xxx
6
362
ators, providers and medical directors, as well as in-hospital staff in the SPARC network hospitals working together in the front line of emergency patient care for their continued commitment and contributions to high quality care and primary data collection in resuscitation research at Rescu, Li Ka Shing Knowledge Institute, St. Michael’s Hospital, Toronto, Ontario, Canada.
363
References
357 358 359 360 361
364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404
1. Mozaffarian D, Benjamin E, Go A, et al. Heart disease and stroke statistics—2015 update. Circulation 2015;131:e1–294. 2. Robertson R. Sudden death from cardiac arrest—improving the odds. N Engl J Med 2000;343:1259–60. 3. Perkins G, Jacobs I, Nadkarni V, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update of the Utstein resuscitation registry templates for out-of-hospital cardiac arrest. Circulation 2015;131:1286–300. 4. Rea T, Cook A, Stiell I, et al. Predicting survival after out-of-hospital cardiac arrest: Role of the Utstein data elements. Ann Emerg Med 2010;55:249–57. 5. Bergner DW, Goldberger JJ. Diabetes mellitus and sudden cardiac death: what are the data? Cardiol J 2010;17:117–29. 6. Jouven X, Lemaitre RN, Rea TD, Sotoodehnia N, Empana JP, Siscovick DS. Diabetes, glucose level, and risk of sudden cardiac death. Eur Heart J 2005;26:2142–7. 7. Laukkanen J, Makikallio T, Ronkainen K, Karppi H, Kuri S. Impaired fasting plasma glucose and type 2 diabetes are related to the risk of out-of-hospital sudden cardiac death and all-cause mortality. Diabetes Care 2013;36:1166–71. 8. Ohira T, Maruyama M, Imano H, et al. Risk factors fo sudden cardiac death among Japanese: the circulatory risk in communities study. J Hypertens 2012;30:1137–43. 9. Kuehl M, Stevens MJ. Cardiovascular autonomic neuropathies as complications of diabetes mellitus. Nat Rev Endocrinol 2012;8:405–16. 10. Tesfaye S, Boulton AJ, Dyck PJ, et al. Diabetic neuropathies: update on definitions, diagnostic criteria, estimation of severity, and treatments. Diabetes Care 2010;33:2285–93. 11. Dimitropoulos G, Tahrani AA, Stevens MJ. Cardiac autonomic neuropathy in patients with diabetes mellitus. World J Diabetes 2014;5:17–39. 12. Singh JP, Larson MG, O’Donnell CJ, et al. Association of hyperglycemia with reduced heart rate variability. Am J Cardiol 2000;86:309–12. 13. Young L, Wackers F, Chyun D, et al. Cardiac outcomes after screening for asymptomatic coronary artery disease in patients with type 2 diabetes: the DIAD study: a randomized controlled trial. JAMA 2009;301:1547–55. 14. Maser R, Mitchell B, Vinik A, Freeman R. The association between cardiovascular autonomic neuropathy and mortality in individuals with diabetes: a metaanalysis. Diabetes Care 2003;26:1895–901. 15. Pop-Busui R, Evans G, Gerstein H, et al. Effects of cardiac autonomic dysfunction on mortality risk in the action to control cardiovascular risk in diabetes (ACCORD) trial. Diabetes Care 2010;33:1578–84. 16. Laukkanen JA, Makikallio TH, Ronkainen K, Karppi J, Kurl S. Impaired fasting plasma glucose and type 2 diabetes are related to the risk of out-of-hospital sudden cardiac death and all-cause mortality. Diabetes Care 2013;36:1166–71. 17. IDF. International Diabetes Federation Annual Report 2014. Belgium2014. 18. ADA. Statistics about diabetes. 2016. Accessed June 1, 2016, 2016.
19. CDA. Diabetes in Canada. 2015; 2016. https://www.diabetes.ca/getmedia/ 513a0f6c-b1c9-4e56-a77c-6a492bf7350f/diabetes-charter-backgroundernational-english.pdf.aspx. Accessed June 29, 2016. 20. IDF. IDF Diabetes Atlas. 7th ed; 2015. http://www.idf.org/diabetesatlas/keymessages.html. Accessed January 2nd, 2016. 21. CDA. Diabetes in Ontario. 2015. 22. UN. World Population Prospects. Geneva, Switzerland2015. 23. Yang WS, Zhao W, Xiao J, et al. Medical care and payment for diabetes in china: enormous threat and great opportunity. PLoS One 2012;7:1–13. 24. WHO. Per Capita Total Expenditure on Health. Geneva, Switzerland2012. 25. Morrison L, Nichol G, Rea T, et al. Rationale, development and implementation of the resuscitation outcomes consortium epistry-cardiac arrest. Resuscitation 2008;78:161–9. 26. Lin S, Morrison L, Brooks S. Development of a data dictionary for the Strategies for Post Arrest Resuscitation Care (SPARC) network for post cardiac arrest research. Resuscitation 2011;82:419–22. 27. Drennan I, Strum R, Byers A, et al. Out-of-hospital cardiac arrest in high-rise buildings: delays to patient care and effect on survival. CMAJ 2016;188:413–9. 28. Jacobs I, Berg R, Billi J, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries. Circulation 2004;110:3385–97. 29. Edgren E, Hedstrand U, Kelsey S, Sutton-Tyrrell K, Safar P. Assessment of neurological prognosis in comatose survivors of cardiac arrest. Lancet 1994;343:1055–9. 30. Ro Y, Shin S, Song K, et al. Risk of diabetes mellitus on incidence of out-of-hospital cardiac arrests: a case-control study. PLoS One 2016;11:1–11. 31. Terman S, Hume B, Meurer W, Silbergleit R. Impact of presenting rhythm on short- and long-term neurological outcome in comatose survivors of cardiac arrest treated with therapeutic hypothermia. Crit Care Med 2014;42:2225–34. 32. Andrew E, Nehme Z, Lijovic M, Bernard SA, Smith K. Outcomes following outof-hospital cardiac arrest with initial cardiac rhythm of asystole or pulseless electrical activity in Victoria, Australia. Resuscitation 2014;85:1633–9. 33. Wibrandt I, Norsted K, Schmidt H, Schierbeck J. Predictors for outcome among cardiac arrest patients: the importance of initial cardiac arrest rhythm versus time to return of spontaneous circulation, a retrospective cohort study. BMC Emerg Med 2015;15:1–7. 34. Nehme Z, Nair R, Andrew E, et al. Effect of diabetes and pre-hospital blood glucose level on survival and recovery after out-of-hospital cardiac arrest. Paper presented at: Crit Care Resusc 2016. 35. Jang D, Shin S, Ro Y, et al. Interaction of the diabetes mellitus and cardiac diseases on survival outcomes in out-of-hospital cardiac arrest. Am J Emerg Med 2016;34:702–7. 36. Schouten E, Dekker J, Meppelink P, Kok F, Vandenbroucke J, Pool J. QT interval prolongation predicts cardiovascular mortality in an apparently health population. Circulation 1991;84:1516–23. 37. Stern K, Cho Y, Benitez-Aguirre P, et al. QT interval, corrected for heart rate, is associated with HbA1c concentration and autonomic function in diabetes. Diabet Med 2016:1–7. 38. Pfeifer M, Weinberg C, Cook D, et al. Autonomic neural dysfunction in recently diagnosed diabetic subjects. Diabetes Care 1984;7:447–53. 39. Ro Y, Shin S, Song K, et al. Interaction effects between hypothermia and diabetes mellitus on survival outcomes after out-of-hospital cardiac arrest. Resuscitation 2015;90:35–41. 40. Beagley J, Guariguata L, Weil C, Motala A. Global estimates of undiagnosed diabetes in adults. Diabetes Res Clin Pract 2014;103:150–60.
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459
ARTICLE IN PRESS
RESUS 7044 1–6
1
Resuscitation xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
2
The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study夽
3
4
5
Q1
6 7 8 9 10 11
Monica Parry a,∗ , Kyle Danielson a , Sarah Brennenstuhl a , Ian R. Drennan b,c , Laurie J. Morrison d a
Lawrence S. Bloomberg Faculty of Nursing, University of Toronto, Toronto, Canada Division of Emergency Medicine, Department of Medicine, University of Toronto, Toronto, Canada c Institute of Medical Science, Faculty of Medicine, University of Toronto, Toronto, Canada d Robert and Dorothy Pitts Chair in Acute Care and Emergency Medicine, Li Ka Shing Knowledge Institute, St. Michael’s Hospital Division of Emergency Medicine, Department of Medicine, University of Toronto, Toronto, Canada b
12
13 26
a r t i c l e
i n f o
a b s t r a c t
14 15 16 17 18 19
Article history: Received 9 September 2016 Received in revised form 24 December 2016 Accepted 13 January 2017
20
25
Keywords: Out-of-hospital cardiac arrest Diabetes Return of spontaneous circulation Survival
27
Introduction
21 22 23 24
28Q2 29 30 31 32 33 34 35
Background: Sudden cardiac arrest (SCA), confirmed absence of cardiac mechanical activity, is the leading cause of heart-related death in the US. Almost 85% of SCA occur out-of-hospital (OHCA), with very poor rates of return of spontaneous circulation (ROSC) and survival to hospital discharge. We sought to determine if diabetes status was associated with survival or ROSC following an OHCA. Methods: We completed a retrospective cohort study using data from the Toronto Regional RescuNet Epistry dataset, based upon data definitions defined by the Resuscitation Outcomes Consortium (ROC) Epistry-Cardiac Arrest and the Strategies for Post Arrest Resuscitation Care (SPARC) network datasets. Adults ≥18 years of age who experienced an OHCA, had data on diabetes status, and were treated by Emergency Medical Services (EMS) between 2012–2014 were included in the analysis (n = 10,097). We used bivariate analyses to examine relationships between diabetes status, Utstein elements and outcomes, and logistic regression to determine predictors of survival. Results: Diabetes prevalence was 27.8% (95% CI: 27.0–28.7). A larger proportion of those with diabetes had a non-shockable initial rhythm (28.8% vs. 25.1%; p < 0.01) and did not survive to hospital discharge (92.1% vs. 89.2%, p < 0.001). Diabetes status is associated with a decrease in survival, independent from a number of Utstein elements (adjusted OR = 0.76; 95% CI: 0.64–0.91, p = 0.003). Conclusions: This is the first Canadian study to examine the association between diabetes status and OHCA outcomes. Our findings suggest that diabetes status prior to arrest is associated with decreased survival. The growing prevalence of diabetes globally suggests a future burden related to OHCAs. © 2017 Elsevier B.V. All rights reserved.
Sudden cardiac arrest (SCA), confirmed absence of cardiac mechanical activity, is a significant public health issue and the leading cause of heart-related death in the United States (US).1 Approximately 85% of SCA occur out-of-hospital, leading to 326,200 deaths each year in the US1 and one death every 13 min in Canada.2 Of those out-of-hospital cardiac arrests (OHCAs) treated by Emergency Medical Services (EMS), only 10.6% survive to hospital discharge and 8.3% have favorable neurological outcomes.1
夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at DOI:10.1016/j.resuscitation.2017.01.011. ∗ Corresponding author at: Lawrence S. Bloomberg Faculty of Nursing, 155 College Street, Suite 130, Toronto, ON, Canada. Fax: +1 416 978 8222. E-mail address: [email protected] (M. Parry).
Several core Utstein elements have consistently predicted survival among EMS-treated arrests, including age, gender, witnessed arrest, arrest location, bystander cardiopulmonary resuscitation (CPR), first monitored rhythm, and EMS response times.3,4 These variables have been shown to predict approximately 72% of the variation in survival3,4 however, a portion of the variation remains unaccounted for, suggesting that there are unknown patient variables associated with survival after an OHCA. Identification of these variables would be an opportunity to reduce the public health burQ3 den of OHCA.4 The most common preventable comorbidities associated with SCA include coronary artery disease, myocardial infarction, and heart failure.5 Diabetes is a risk factor for all these comorbidities and has been shown to be associated with a 2–4 fold increase in the risk of SCA.6–8 Cardiac autonomic neuropathy, a common and serious complication of type 1 (T1) and type 2 (T2) diabetes,9–11 results from damage to autonomic fibers that innervate the heart and blood
http://dx.doi.org/10.1016/j.resuscitation.2017.01.011 0300-9572/© 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52
G Model RESUS 7044 1–6
ARTICLE IN PRESS M. Parry et al. / Resuscitation xxx (2017) xxx–xxx
2
88
vessels and increases the risk of SCA.12 Cardiac autonomic neuropathy has been linked to resting tachycardia, prolonged QT interval, increased silent myocardial ischemia and infarction, and increased mortality.13,14 Results from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial confirm that cardiac autonomic neuropathy is associated with a 1.55–2.95 times risk of all-cause and cardiovascular disease (CVD) mortality independent of multiple CVD risk factors.15 Others report that men with T2 diabetes have a 2.26-fold increase in SCA risk compared to men without diabetes, even after adjusting for age, body mass index (BMI), systolic blood pressure (SBP), smoking, cholesterol, known coronary artery disease (CAD), and family history of CAD.16 Globally, 8.3% of individuals aged 20–79 years of age17 and 25.9% of individuals over 65 years of age in the US have diabetes.18 In 2015, 9.3% of Canadians had diabetes, and this is projected to increase by 44% in the next decade.19 Diabetes is one of the largest public health emergencies of the 21st century. The World Health Organization (WHO) suggests diabetes is the third highest risk factor for premature mortality in the world20 ; it is estimated that slightly more men than women have diabetes, and one in two individuals remain undiagnosed due to misinterpretation or lack of symptoms early in the course of disease.21 Diabetes complications are common, and while there are no internationally agreed upon standards for diagnosing or measuring diabetes complications, it is estimated that up to 50% of individuals have complications at the time of diagnosis.22 Diabetes and its complications are a major cause of early mortality, and existing health statistics underestimate the number of these deaths.23 Better understanding the burden of diabetes and its complications would help direct public health actions, and improve care for all people with diabetes.24 The primary objective of this study was to determine the association between diabetes status (T1 and T2) and survival to hospital discharge from OHCA. The secondary objectives were to determine the association between diabetes status, and any ROSC and neurological outcome, at hospital discharge.
89
Methods
90
Study design and population
53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87
108
We completed a retrospective cohort study using data from the Toronto Regional RescuNet Epistry dataset and definitions developed by the Resuscitation Outcomes Consortium (ROC) Epistry-Cardiac Arrest and the Strategies for Post Arrest Resuscitation Care (SPARC) network datasets.25,26 This dataset is a population-based registry of consecutive OHCA attended by 911initiated EMS first responders in southern Ontario, serving a population of over 6.6 million residents in Durham, Halton, Muskoka, Peel, Simcoe, Toronto and York.27 Data are collected from pre-hospital call reports from seven land EMS agencies, fire departments, and one air ambulance service, and from in-hospital records from 44 destination hospitals and entered into a secure database. The Research Ethics Board at each participating hospital has given ethics approval for all retrospective studies using the Toronto RescuNet Epistry-Cardiac Arrest database. Adults ≥18 years of age who experienced an OHCA, had data on diabetes status, and were treated by EMS between 2012–2014 were included in the analysis (n = 10,097).
109
Measures
91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107
110 111 112
Exposure Diabetes status was ascertained using a variable based on the inhospital record (yes, no, not noted). If this information was missing
or status was not noted, data on diabetes status was obtained from the pre-hospital call report.
Covariates The Utstein elements are the most widely used set of variables closely associated with OHCA outcomes.3 The following Utstein elements were obtained and used in our analysis: age (years), gender, witness status (EMS/bystander witnessed, unwitnessed), location (public, private/residential, other),28 first response CPR (bystander, EMS, none) bystander application of automatic external defibrillator (AED), first monitored rhythm (ventricular fibrillation and ventricular tachycardia [shockable], or pulseless electrical activity and asystole [non-shockable]), EMS response time (call to 911 to arrival on scene time interval), and etiology of arrest. Etiology was defined as either obvious cause (e.g., drug overdose, asphyxiation, drowning) or presumed cardiac cause when there was no identifiable obvious cause of arrest.
Outcomes Survival to hospital discharge (yes/no) was our primary outcome. Secondary outcomes included any ROSC during the resuscitation attempt (yes/no) and neurological outcome at hospital discharge based on the modified Rankin Scale (mRS) of 0–2 (favourable) or 3–5 (unfavourable).29 The amount of missing data for most variables was negligible, ranging from 0% to about 2%, with the exception of EMS response time, which had 10.2% missing data (Table 1). The missing data occurred in cases where the first EMS responders were not from RescuNET participating services. Missing data was less than 1% for this variable in all cases where RescuNET participating EMS responders arrived first.
Statistical analysis Univariate statistics, including frequency counts, percentages and means, were used to describe the prevalence of diabetes and other characteristics of the study population. Bivariate associations between diabetes status and each of the Utstein data elements and outcomes were assessed using 2 tests and t-tests, as appropriate. Multivariable logistic regression was used to investigate the relationship between diabetes status and each of the outcomes (separately) while adjusting for the Utstein data elements. Preliminary steps included performing model diagnostics, such as assessing multicollinearity using the variance inflation factor (VIF) and testing for non-linearities between the predictors and outcome. There was no evidence of multicollinearity, but we did determine the need for a non-linear specification of age, which was represented using age and age2 in the modeling. Next, we tested for plausible interactions between diabetes status and age, gender, whether arrest was witnessed and location of arrest, but none were found. To build a final model, we employed a hierarchical approach, starting with modeling diabetes status alone, then cumulatively adding clusters of variables. The Utstein elements were added in three clusters: demographics (age, gender), circumstances of arrest (witnessed, location, first-response CPR, etiology), followed by first monitored rhythm. We excluded response times from the final model due to the large amount of missing data. Nested models were compared using the likelihood ratio test (LRT). Clusters were left in the model if they significantly contributed to model fit based on the LRT. Only 4.2% of cases were missing data for one or more of the covariates and could not be included in the modeling. Data analysis was undertaken using SPSS version 23. A two sided p-value of p < 0.05 was used to establish statistical significance.
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
113 114
115 116 117 118 119 120 121 122 123 124 125 126 127 128
129 130 131 132 133 134 135 136 137 138 139 140 141
142
143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
G Model
ARTICLE IN PRESS
RESUS 7044 1–6
M. Parry et al. / Resuscitation xxx (2017) xxx–xxx Table 1 Patient characteristics by diabetes status.
3
Q5
N% Event year 2012, n (%) 2013, n (%) 2014, n (%) Age Mean age, years Median age, years (IQR) Gender Male, n (%) Female, n (%) Witnessed status Witnessed, n (%) Not witnessed, n (%) Location Public, n (%) Private/Residential, n (%) Other (e.g., hotel & nursing home), n (%) Site Toronto, n (%) Durham, n (%) Simcoe, n (%) Muskoka, n (%) Peel, n (%) Halton, n (%) York, n (%) First-response CPR Bystander, n (%) EMS, n (%) None, n (%) Bystander AED applied, n (%) First monitored rhythm Ventricular fibrillation, n (%) PEA, n (%) Asystole, n (%) Shockable, n (%) Not Shockable, n (%) Other, n (%) Etiology No obvious cause, presumed cardiac, n (%) Obvious cause, presumed non-cardiac, n (%) Median EMS response time interval, mins (IQR)
Diabetes
No diabetes
2812 (27.8)
7285 (72.2)
p Value
Missing, n (%)
889 (31.6) 936 (33.3) 987 (35.1)
2343 (32.2) 2466 (33.9) 2476 (34.0)
ns
0 (0)
72 74 (18)
69 72 (26)
<0.001 <0.001
0 (0) 0 (0)
1761 (62.6) 1051 (37.4)
4574 (62.8) 2711 (37.2)
ns
0 (0)
1432 (50.9) 1380 (49.1)
3676 (50.5) 3608 (49.5)
ns
1 (0)
241 (8.7) 2282 (82.0) 259 (9.3)
673 (9.4) 5946 (83.0) 7.6 (7.6)
=0.01
154 (1.5)
1265 (45.0) 249 (8.9) 198 (7.0) 30 (1.1) 565 (20.1) 150 (5.3) 355 (12.6)
3219 (44.2) 759 (10.4) 649 (8.9) 75 (1.0) 1158 (15.9) 467 (6.4) 958 (13.2)
<0.001
0 (0)
1006 (35.8) 376 (14.1) 1409 (50.1) 66 (6.6)
2790 (38.3) 887 (12.2) 3607 (49.5) 181 (6.5)
=0.008 ns
27 (0.3)
416 (15.1) 673 (24.4) 1308 (47.4) 67 (2.4) 269 (9.7) 27 (1.0)
1207 (17.0) 1652 (23.3) 3172 (44.7) 239 (3.4) 740 (10.4) 80 (1.1)
=0.008
247 (2.4)
2666 (94.8) 146 (5.2) 6.0 (2.5)
6500 (89.2) 785 (10.8) 6.0 (2.4)
<0.001 ns
0 (0) 1071 (11.9)
0 (0)
ns, non-significant; IQR, interquartile range; CPR, cardiopulmonary resuscitation; EMS, emergency medical services; AED, automated external defibrillator; PEA, pulseless electrical activity; mins, minutes.
172
173 174
Results In total, there were 11,765 OHCA treated by EMS between 2012 and 2014. We excluded 209 cases less than 18 years of age. Another OHCA in database from 2012 – 2014 N = 20,371
1459 cases were excluded because of missing data on diabetes status (12.6% of otherwise eligible cases) (Fig. 1). Those with missing data on diabetes were younger (61 vs. 70 years), more often male (75.4% vs. 62.7%), had more cardiac arrests in public location (33.8%
Exclude Not treated by EMS N = 8,606
Cases Treated by EMS N = 11,765 Exclusion Criteria < 18 years of age (n = 209); excluded missing diabetes status (n=1,459) Eligible cases N = 10,097
Fig. 1. CONSORT diagram of included and excluded cases.
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
175 176 177 178
G Model
ARTICLE IN PRESS
RESUS 7044 1–6
M. Parry et al. / Resuscitation xxx (2017) xxx–xxx
4 Table 2 Survival by diabetes status.
Survival to hospital discharge Survived n (%) Died n (%) Any ROSC, n (%) Neurologic outcome, mRS score mRS 0–2, Favorable outcome n (%) mRS 3–5, Poor outcome n (%)
Diabetes
No diabetes
p Value
Missing, n (%)
221 (7.9) 2586 (92.1) 883 (31.4)
786 (10.8) 6493 (89.2) 2422 (33.2)
<0.001 ns
11 (0.1) 0 (0)
183 (88.0) 25 (12.0)
691 (91.3) 66 (8.7)
ns
42 (4.2)
ns, non-significant; ROSC, return of spontaneous circulation; mRS, modified Rankin Scale.
Table 3 Logistic regression of survival among adults with diabetes who had an out of hospital cardiac arrest treated by EMS in Ontario from 2012 to 2014 (n = 9677). Model 1 Unadjusted
OR
95% CI
Diabetes diagnosis 0.70 (0.60, 0.83) Age in years 2 Age in years Male gender Location (ref = Private/Residential) Public Other First-response CPR (ref = none) Bystander EMS Witnessed Obvious cause Shockable first monitored rhythm Chi-square 19.37 (df = 1) 19.37 (df = 1) Chi square 6127.34 −2 Loglikelihood
Model 2 Adjusted for demographics
Model 3 Adjusted for demographics and circumstances of arrest
Model 4 Adjusted for demographics, circumstances of arrest, and first monitored rhythm
OR
95% CI
OR
95% CI
OR
95% CI
0.78 0.95 0.99 1.62
(0.66, 0.93) (0.94, 0.95) (0.99, 0.99) (1.39, 1.90)
0.76 0.95 0.99 1.37
(0.64, 0.91) (0.94, 0.96) (0.99, 0.99) (1.16, 1.62)
0.88 0.96 1.00 0.95
(0.72, 1.06) (0.95, 0.97) (0.99, 1.00) (0.79, 1.14)
3.92 1.07
(3.27, 4.69) (0.74, 1.55)
2.59 1.16
(2.12, 3.15) (0.77,1.73)
1.20 2.29 4.38 1.02
(1.01, 1.43) (1.87, 2.82) (3.59, 5.34) (0.80, 1.31)
0.95 (0.79, 1.15) 2.99 (2.38, 3.75) 2.70 (2.19, 3.33) 2.17 (1.65, 2.85) 10.58 (8.80, 12.74) 1980.01 (df = 11) 707.98 (df = 1) 4166.71
484.77 (df = 4) 465.40 (df = 3) 5661.96
1272.02 (df = 10) 787.26 (df = 6) 4874.69
CPR, cardiopulmonary resuscitation; EMS, emergency medical services.
179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201
vs. 9.2%), and more often had cardiac arrests of obvious cause compared to those included in the analysis. Cases with missing diabetes status also survived to hospital discharge less often (3.9% vs. 10%). Among the 10,097 OHCAs included in the analysis, 2812 cases had a diagnosis of diabetes, resulting in a prevalence of 27.8% (95% CI: 27.0–28.6). Cases were distributed across seven sites including Toronto (44.4%), Peel (17.1%), York (13.0%), Durham (10.0%), Simcoe (8.4%), Halton (6.1%), and Muskoka (1.0%). Of these, most occurred in residential locations (82.8%) and just over half (50.6%) were witnessed. Bystander CPR was attempted in over a third of cases (37.6%) and a bystander applied an AED infrequently (6.5%). In just over one in ten, first-response CPR was preformed by EMS (12.7%). Approximately 16% of cases had an initial cardiac rhythm of ventricular fibrillation (shockable), and 44.4% of cases had asystole and 23.0% had pulseless electrical activity (non-shockable). Those with diabetes were slightly older (72 vs. 69, p < 0.001), had a presumed cardiac cause for arrest (i.e., no obvious cause) (94.8% vs. 89.2%, p < 0.001), more often had an EMS-witnessed arrest (14.1% vs. 12.2%, p < 0.01), and presented with a non-shockable rhythm (28.8% vs. 25.1%, p < 0.01) compared to those without diabetes. There were no differences in diabetes status by gender, whether arrest was witnessed by bystanders, or in EMS response times (Table 1).
30% lower odds of survival (OR = 0.70, 95% CI: 0.60–0.83). When the relationship is adjusted for patient demographics (model 2) or together with circumstances of the arrest (model 3), the lower odds of survival among those with diabetes decreases slightly, but the association remains significant (model 2 OR = 0.78, 95% CI: 0.66–0.93; model 3 OR = 0.76, 95% CI: 0.64–0.91). When the first monitored rhythm is also included, the odds ratio remains below one, but becomes non-significant (model 4 OR = 0.88, 95% CI: 0.72–1.06). While response time was not included in the final modeling due to a large amount of missing data in cases where the first EMS responders were not from RescuNET, a supplementary analysis showed that, when included in a similar series of models, the pattern of findings remained unchanged (data available upon request). Any ROSC during resuscitation Approximately 31% (n = 883) of individuals with diabetes and 33% (n = 2422) of individuals without diabetes had any ROSC, with no group differences by diabetes status (Table 2). Multivariable modeling confirmed that there was no relationship between diabetes status and ROSC (data available on request). Neurological outcome at hospital discharge
202
203 204 205 206 207
208 209 210 211 212 213 214 215 216 217 218 219 220 221
222
223 224 225 226 227
228
Survival to hospital discharge Differences in outcomes between those with and without diabetes are presented in Table 2. A smaller proportion of those with diabetes survived to hospital discharge (7.9% vs. 10.8%, p < 0.001). Further investigation of this relationship using logistic regression (Table 3), shows that when unadjusted, diabetes is associated with
Exactly 88% (n = 183) of individuals with diabetes and 91.3% (n = 691) of individuals without diabetes had favorable neurological outcomes, with no group differences by diabetes status (Table 2). Multivariable modeling confirmed that there was no relationship between diabetes status and neurological outcome (data available on request).
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
229 230 231 232 233 234
G Model RESUS 7044 1–6
ARTICLE IN PRESS M. Parry et al. / Resuscitation xxx (2017) xxx–xxx
235
236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299
Discussion In this study, 27.8% of individuals had a diagnosis of diabetes. These results are similar to Ro30 who reported that 26.7% (n = 370) of individuals who had an OHCA with a presumed cardiac cause had diabetes. In our cohort, individuals with diabetes had significantly lower odds of survival to hospital discharge, even when adjusting for demographics (age, gender) and circumstances of the arrest (location, resuscitation attempted, witnessed, and obvious cause). However, when initial rhythm was included in the model, the association between diabetes status and survival lost significance. This may reflect, in part, a greater number of individuals with diabetes presenting with a non-shockable first monitored rhythm; a finding also demonstrated in other research.31 More clearly, however, it underlines the large effect that initial rhythm has on survival from OHCA, even when adjusting for diabetes status and other Utstein elements (AOR, 10.58, 95% CI: 8.80–12.74). The importance of first monitored rhythm for predicting survival has been shown in other studies.32,33 Our study adds to the literature by comparing survival from an OHCA in individuals with and without diabetes in Canada, and it is the first study to describe these outcomes in relation to existing Utstein elements. The finding that diabetes status is related to survival with an unadjusted analysis is consistent with previous research.34,35 Nehme et al.34 reported that individuals with diabetes had reduced rates of survival to hospital discharge compared to individuals without diabetes (6.8% vs. 13.4%, p < 0.001) and individuals with diabetes had reduced odds of good functional recovery at one year (OR, 0.57, 95% CI: 0.35–0.95). Jang et al.35 found diabetes status was negatively associated with survival (AOR, 0.84, 95% CI: 0.75–0.95) and neurological recovery (AOR, 0.81, 0.67–0.97), and diabetes modified the effect on OHCA survival depending on the presence or absence of pre-existing cardiac disease. Cardiac disease alone had a non-significant effect on survival and neurological recovery, but when combined with diabetes, odds of survival (AOR, 0.58, 0.45–0.74) and neurological recovery (AOR, 0.52, 95% CI: 0.36–0.75) were significantly lower. The results of this study demonstrate that OHCA in individuals with diabetes are more likely to present with non-shockable rhythms, which have the lowest survival rate.32 However, others report that diabetes also reduced survival to hospital discharge for individuals presenting with a shockable rhythm (AOR, 0.57, 95% CI: 0.38–0.86).34 A convergence of recent evidence helps to explain the link between diabetes and survival from OHCA. While the interrelationship between glucose metabolism, autonomic function and cardiac disease in diabetes are not well understood, they are thought to be related to a prolonged QT interval. The QT interval on an electrocardiogram (ECG) measures the total time for ventricular depolarization and repolarization. A prolonged QT interval corrected for heart rate (QTc) is predictive of mortality in in healthy adults.36 It has been positively associated with HbA1c concentration ( = 4, 95% CI: 2–6), and recent evidence suggests a lower threshold (QTc ≥ 409 milliseconds) is associated with autonomic dysfunction in individuals with diabetes.37 Cardiac autonomic neuropathy should be considered in models of CVD and mortality risk in diabetes, and although clinical symptoms occur late in diabetes, research suggests subclinical cardiac autonomic neuropathy can be detected within two years of diagnosis of T1 diabetes and one year of diagnosis of T2 diabetes.38 Results from the ACCORD trial suggests a combination of heart rate variability (HRV) and QT interval measurements derived from a standard ECG could identify a subset of individuals with diabetes at increased mortality risk independent of traditional CVD risk factors.15 We found no group differences in neurological outcomes by diabetes status. This finding is inconsistent with others34,35 who found diabetes was negatively associated with neurological recov-
5
ery. Ro et al.39 also reported that diabetes modified the positive effect of targeted temperature management (TTM) to 32–34 ◦ C on survival and neurological outcomes after OHCA in individuals with diabetes. That is, TTM was associated with good neurological recovery in individuals without diabetes following an OHCA, but not in individuals with diabetes.39 Limitations Approximately, 12.6% (n = 1459) of otherwise eligible cases had to be excluded from the analysis because of missing data on diabetes status. Excluded cases were younger, more frequently male, more often had an arrest in a public location with an obvious non-cardiac cause and less frequently survived. Based on these characteristics, it appears that some selection bias might have occurred. Because T2 diabetes is so often undiagnosed (almost 50% of cases, globally),40 a portion of those excluded likely had diabetes. Excluding those without data on diabetes may have made our analysis more conservative by removing cases with lower survival. Diabetes status was measured dichotomously; type and duration of diabetes, treatment, glycemic control, and presence of complications could not be considered in the analyses. Data on past CVD history was initially considered, but because of a large amount of missing data this was not controlled for in the analyses, which may indicate some residual confounding. Conclusion This is the first Canadian study that has examined the association between diabetes status and OHCA outcomes. Our findings indicate that diabetes is associated with decreased survival and the growing prevalence of diabetes globally suggests future burden related to OHCAs. While information about initial rhythm may trump that of diabetes status when predicting survival, knowing more about diabetes risk (e.g., pre-diabetes), diabetes control (e.g., HbA1c) and diabetes complications (e.g., renal function and proteinuria) may help to target interventions to prevent a SCA in these subgroups of individuals. Diabetes is associated with autonomic dysfunction and cardiac microvascular complications that increase the risk of SCA. Electrocardiogram screening for long QTc and complication risk of diabetes may be of value in routine complication assessments, but further research is needed in this area. Conflict of interest statement None. Acknowledgments This study was possible through grants from the Canadian Insti- Q4 tutes of Health Research (CIHR), the Laerdal Medical Foundation and the Heart and Stroke Foundation of Canada. Kyle Danielson was supported by a Canada Graduate Scholarship Master’s Award from CIHR and Ian Drennan was supported by a CIHR Banting and Best Doctoral Research Award. The RescuNet Epistry dataset was funded by the Laerdal Foundation, CIHR, the Heart and Stroke Foundation of Canada, Defence Research and Development Canada and the US National Institutes of Health funding for the Resuscitation Outcomes Consortium. The authors would like to thank Ms. Chantelle Nielson who contributed to the initial Rescu Writing Group Proposal while a student in the Undergraduate Student Summer Research Program at the Lawrence S. Bloomberg Faculty of Nursing, University of Toronto, Toronto, Ontario, Canada. The authors would also like to thank the Rescu Epistry investigators and emergency medical service oper-
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
300 301 302 303 304 305
306
307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
323
324 325 326 327 328 329 330 331 332 333 334 335 336 337
338
339
340
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356
G Model RESUS 7044 1–6
ARTICLE IN PRESS M. Parry et al. / Resuscitation xxx (2017) xxx–xxx
6
362
ators, providers and medical directors, as well as in-hospital staff in the SPARC network hospitals working together in the front line of emergency patient care for their continued commitment and contributions to high quality care and primary data collection in resuscitation research at Rescu, Li Ka Shing Knowledge Institute, St. Michael’s Hospital, Toronto, Ontario, Canada.
363
References
357 358 359 360 361
364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404
1. Mozaffarian D, Benjamin E, Go A, et al. Heart disease and stroke statistics—2015 update. Circulation 2015;131:e1–294. 2. Robertson R. Sudden death from cardiac arrest—improving the odds. N Engl J Med 2000;343:1259–60. 3. Perkins G, Jacobs I, Nadkarni V, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update of the Utstein resuscitation registry templates for out-of-hospital cardiac arrest. Circulation 2015;131:1286–300. 4. Rea T, Cook A, Stiell I, et al. Predicting survival after out-of-hospital cardiac arrest: Role of the Utstein data elements. Ann Emerg Med 2010;55:249–57. 5. Bergner DW, Goldberger JJ. Diabetes mellitus and sudden cardiac death: what are the data? Cardiol J 2010;17:117–29. 6. Jouven X, Lemaitre RN, Rea TD, Sotoodehnia N, Empana JP, Siscovick DS. Diabetes, glucose level, and risk of sudden cardiac death. Eur Heart J 2005;26:2142–7. 7. Laukkanen J, Makikallio T, Ronkainen K, Karppi H, Kuri S. Impaired fasting plasma glucose and type 2 diabetes are related to the risk of out-of-hospital sudden cardiac death and all-cause mortality. Diabetes Care 2013;36:1166–71. 8. Ohira T, Maruyama M, Imano H, et al. Risk factors fo sudden cardiac death among Japanese: the circulatory risk in communities study. J Hypertens 2012;30:1137–43. 9. Kuehl M, Stevens MJ. Cardiovascular autonomic neuropathies as complications of diabetes mellitus. Nat Rev Endocrinol 2012;8:405–16. 10. Tesfaye S, Boulton AJ, Dyck PJ, et al. Diabetic neuropathies: update on definitions, diagnostic criteria, estimation of severity, and treatments. Diabetes Care 2010;33:2285–93. 11. Dimitropoulos G, Tahrani AA, Stevens MJ. Cardiac autonomic neuropathy in patients with diabetes mellitus. World J Diabetes 2014;5:17–39. 12. Singh JP, Larson MG, O’Donnell CJ, et al. Association of hyperglycemia with reduced heart rate variability. Am J Cardiol 2000;86:309–12. 13. Young L, Wackers F, Chyun D, et al. Cardiac outcomes after screening for asymptomatic coronary artery disease in patients with type 2 diabetes: the DIAD study: a randomized controlled trial. JAMA 2009;301:1547–55. 14. Maser R, Mitchell B, Vinik A, Freeman R. The association between cardiovascular autonomic neuropathy and mortality in individuals with diabetes: a metaanalysis. Diabetes Care 2003;26:1895–901. 15. Pop-Busui R, Evans G, Gerstein H, et al. Effects of cardiac autonomic dysfunction on mortality risk in the action to control cardiovascular risk in diabetes (ACCORD) trial. Diabetes Care 2010;33:1578–84. 16. Laukkanen JA, Makikallio TH, Ronkainen K, Karppi J, Kurl S. Impaired fasting plasma glucose and type 2 diabetes are related to the risk of out-of-hospital sudden cardiac death and all-cause mortality. Diabetes Care 2013;36:1166–71. 17. IDF. International Diabetes Federation Annual Report 2014. Belgium2014. 18. ADA. Statistics about diabetes. 2016. Accessed June 1, 2016, 2016.
19. CDA. Diabetes in Canada. 2015; 2016. https://www.diabetes.ca/getmedia/ 513a0f6c-b1c9-4e56-a77c-6a492bf7350f/diabetes-charter-backgroundernational-english.pdf.aspx. Accessed June 29, 2016. 20. IDF. IDF Diabetes Atlas. 7th ed; 2015. http://www.idf.org/diabetesatlas/keymessages.html. Accessed January 2nd, 2016. 21. CDA. Diabetes in Ontario. 2015. 22. UN. World Population Prospects. Geneva, Switzerland2015. 23. Yang WS, Zhao W, Xiao J, et al. Medical care and payment for diabetes in china: enormous threat and great opportunity. PLoS One 2012;7:1–13. 24. WHO. Per Capita Total Expenditure on Health. Geneva, Switzerland2012. 25. Morrison L, Nichol G, Rea T, et al. Rationale, development and implementation of the resuscitation outcomes consortium epistry-cardiac arrest. Resuscitation 2008;78:161–9. 26. Lin S, Morrison L, Brooks S. Development of a data dictionary for the Strategies for Post Arrest Resuscitation Care (SPARC) network for post cardiac arrest research. Resuscitation 2011;82:419–22. 27. Drennan I, Strum R, Byers A, et al. Out-of-hospital cardiac arrest in high-rise buildings: delays to patient care and effect on survival. CMAJ 2016;188:413–9. 28. Jacobs I, Berg R, Billi J, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries. Circulation 2004;110:3385–97. 29. Edgren E, Hedstrand U, Kelsey S, Sutton-Tyrrell K, Safar P. Assessment of neurological prognosis in comatose survivors of cardiac arrest. Lancet 1994;343:1055–9. 30. Ro Y, Shin S, Song K, et al. Risk of diabetes mellitus on incidence of out-of-hospital cardiac arrests: a case-control study. PLoS One 2016;11:1–11. 31. Terman S, Hume B, Meurer W, Silbergleit R. Impact of presenting rhythm on short- and long-term neurological outcome in comatose survivors of cardiac arrest treated with therapeutic hypothermia. Crit Care Med 2014;42:2225–34. 32. Andrew E, Nehme Z, Lijovic M, Bernard SA, Smith K. Outcomes following outof-hospital cardiac arrest with initial cardiac rhythm of asystole or pulseless electrical activity in Victoria, Australia. Resuscitation 2014;85:1633–9. 33. Wibrandt I, Norsted K, Schmidt H, Schierbeck J. Predictors for outcome among cardiac arrest patients: the importance of initial cardiac arrest rhythm versus time to return of spontaneous circulation, a retrospective cohort study. BMC Emerg Med 2015;15:1–7. 34. Nehme Z, Nair R, Andrew E, et al. Effect of diabetes and pre-hospital blood glucose level on survival and recovery after out-of-hospital cardiac arrest. Paper presented at: Crit Care Resusc 2016. 35. Jang D, Shin S, Ro Y, et al. Interaction of the diabetes mellitus and cardiac diseases on survival outcomes in out-of-hospital cardiac arrest. Am J Emerg Med 2016;34:702–7. 36. Schouten E, Dekker J, Meppelink P, Kok F, Vandenbroucke J, Pool J. QT interval prolongation predicts cardiovascular mortality in an apparently health population. Circulation 1991;84:1516–23. 37. Stern K, Cho Y, Benitez-Aguirre P, et al. QT interval, corrected for heart rate, is associated with HbA1c concentration and autonomic function in diabetes. Diabet Med 2016:1–7. 38. Pfeifer M, Weinberg C, Cook D, et al. Autonomic neural dysfunction in recently diagnosed diabetic subjects. Diabetes Care 1984;7:447–53. 39. Ro Y, Shin S, Song K, et al. Interaction effects between hypothermia and diabetes mellitus on survival outcomes after out-of-hospital cardiac arrest. Resuscitation 2015;90:35–41. 40. Beagley J, Guariguata L, Weil C, Motala A. Global estimates of undiagnosed diabetes in adults. Diabetes Res Clin Pract 2014;103:150–60.
Please cite this article in press as: Parry M, et al. The association between diabetes status and survival following an out-of-hospital cardiac arrest: A retrospective cohort study. Resuscitation (2017), http://dx.doi.org/10.1016/j.resuscitation.2017.01.011
405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459