- Email: [email protected]
No difference in mortality between men and women after out-of-hospital cardiac arrest
Resuscitation 96 (2015) 78–84
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
No difference in mortality between men and women after out-of-hospital cardiac arrest夽 Matilde Winther-Jensen a,∗ , Jesper Kjaergaard a , Michael Wanscher a , Niklas Nielsen b , Jørn Wetterslev c , Tobias Cronberg d , David Erlinge e , Hans Friberg f , Yvan Gasche g , Janneke Horn h , Jan Hovdenes i , Michael Kuiper j , Tommaso Pellis k , Pascal Stammet l , Matthew P. Wise m , Anders Åneman n , Christian Hassager a a
The Heart Centre, Copenhagen University Hospital, Copenhagen, Denmark Department of Anesthesia and Intensive Care, Helsingborg Hospital, Helsingborg, Sweden c Copenhagen Trial Unit, Rigshospitalet, Copenhagen, Denmark d Department of Neurology, Skåne University Hospital, Lund, Sweden e Department of Cardiology, Skåne University Hospital, Lund, Sweden f Department of Anesthesia and Intensive Care, Skåne University Hospital, Lund, Sweden g Department of Intensive Care, Geneva University Hospital, Geneva, Switzerland h Department of Intensive Care, Academic Medical Centrum, Amsterdam, The Netherlands i Department of Anesthesia and Intensive Care, Oslo University Hospital, Rikshospitalet, Oslo, Norway j Department of Intensive Care, Leeuwarden Medical Centrum, Leeuwarden, The Netherlands k Department of Intensive Care, Santa Maria degli Angeli, Pordenone, Italy l Department of Anesthesia and Intensive Care, Centre Hospitalier de Luxembourg, Ernest Barblé, Luxembourg m Department of Intensive Care, University Hospital of Wales, Cardiff, United Kingdom n Department of Intensive Care, Liverpool hospital, Sydney, NSW, Australia b
a r t i c l e
i n f o
Article history: Received 23 March 2015 Received in revised form 23 June 2015 Accepted 29 June 2015 Keywords: Out of hospital cardiac arrest Therapeutic hypothermia Women Sex
a b s t r a c t Aim: Comparing the outcome after out-of-hospital cardiac arrest (OHCA) in men and women and to determine whether sex modifies the effect of targeted temperature management (TTM) at 33 or 36 ◦ C. Methods: The TTM trial randomized 950 patients to TTM at 33 or 36 ◦ C for 24 h. This predefined sub-study of the TTM trial assessed survival and neurological outcome defined as Cerebral Performance Category (CPC) and modified Rankin Scale (mRS) using female sex as main predictor of outcome, in relation to level of TTM and other confounding factors. Results: Compared to men, women more often had OHCA at home, p = 0.04 and less often had bystander defibrillation, p = 0.01. No other differences in arrest circumstances were found. Coronary angiography (CAG) and percutaneous coronary intervention (PCI) <24 h after ROSC was less often performed in women, both: p = 0.02. Female sex was associated with higher mortality in univariate analysis, hazard ratio (HR) = 1.29, CI = 1.04–1.61, p = 0.02 compared to men. Adjusting for demographic factors (age and comorbidity), arrest circumstances, pre-hospital findings, inclusion sites, treatments and status at admission reduced this: HR = 1.11, CI = 0. 87–1.41, p = 0.42, and sex was no longer an independent risk factor for death. The effect of sex did not modify the effect of TTM at 33 and 36 ◦ C, pinteraction = 0.73. Conclusion: Female sex seems associated with adverse outcome, but this association is largely explained by differences in arrest circumstances and in-hospital treatment. Our data shows no interaction between sex and the effect of targeting 33 vs. 36 ◦ C. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction 夽 A Spanish translated version of the summary of this article appears as Appendix in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2015.06.030. ∗ Corresponding author. E-mail address: [email protected] (M. Winther-Jensen). http://dx.doi.org/10.1016/j.resuscitation.2015.06.030 0300-9572/© 2015 Elsevier Ireland Ltd. All rights reserved.
Unexpected cardiac arrest is reported to be more common in men1 with an annual rate of cardiac arrest of 0.16% vs. 0.085% in women. Previous studies have reported worse2,3 outcome in women whereas others find a survival benefit of
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
women compared to men in out-of-hospital cardiac arrest (OHCA) victims.4–7 It should be kept in mind that the cardiac arrest populations studied are not always comparable. Some studies included the entire population of emergency service (EMS) attended cardiac arrests,1,3–10 whilst other focused on the population of successfully resuscitated patients.2,11 The observed differences in outcome may in part be explained by dissimilarity in demographic characteristics and in circumstances of the cardiac arrest. Women are reported to have a lower rate of primary rhythm of ventricular fibrillation (VF), to be older at the time of OHCA, less often suffer witnessed arrest, less frequently receive cardiopulmonary resuscitation (CPR) by bystanders and have OHCA from cardiac etiology.1,6,7 Disparity in utilization of coronary angiography (CAG) and percutaneous coronary intervention (PCI) between men and women have been found previously,12–14 which poses the question whether in-hospital treatment may in part contribute to the reported mortality differences along with biological differences, pre-hospital factors and OHCA characteristics. This study is a post hoc analysis of the Target Temperature Management (TTM) trial, which compared target temperatures of 33 ◦ C and 36 ◦ C, finding no overall differences in survival and neurological outcome as well as no interaction of sex and level of TTM.15 This pre-defined post-hoc sub-study reports an in-depth analysis of the differences in mortality as well as neurological outcome and interaction between age and sex in a contemporary cohort of comatose OHCA survivors consisting of men and women, and investigates whether the effect of target temperature management at 33 vs. 36 ◦ C is different in men and women.
2. Methods 2.1. Study design This study is a pre-defined sub-study of the TTM-trial which included 950 OHCA patients admitted to 36 intensive care units (ICU’s) in Europe and Australia from November 2010 to January 2013.15 Inclusion criteria included age ≥18 years, OHCA of presumed cardiac cause, sustained ROSC and Glasgow Coma Scale <8 (patient not able to obey verbal commands). The primary outcome was all-cause mortality at the end of the trial in July 2013 and neither the primary nor secondary or tertiary outcomes of neurological outcome were different in the 33 ◦ C and 36 ◦ C groups.15,16 A prespecified sub group analysis did not show any overall difference of the effect of TTM at 33 vs. 36 ◦ C men and women.15 Randomization was performed through a web-based system, assigning patients in a 1:1 ratio stratified for site to either 33 ◦ C or 36 ◦ C target temperature treatment.17 Active treatment was continued in all patients until blinded neurological evaluation had been performed 108 h after arrest or later.18 Four patients withdrew consent and 7 patients were excluded from the intention-to-treat population and thus 939 patients were analyzed for mortality and neurological outcome.18 Pre-hospital data such as initial rhythm, presence of bystander, bystander performing CPR and time to ROSC were collected according to Utstein guidelines.19 EEG at 12-36 h and SSEP at 48–72 h were strongly recommended by protocol. All other treatment and clinical investigations were left at the discretion of the treating physician. We assessed severity of comorbidities by creating a comorbidity index based on the Charlson comorbidity index. The Charlson index is a weighted and validated index based on 22 conditions and is used to predict short-term mortality.20
79
2.2. Ethical approval The TTM-trial is registered at ClinicalTrials.gov (Identifier: NCT01020916) and the protocol was approved by ethical committees in each participating country. Informed consent was waived or obtained from all participants or their next of kin according to national legislation. 2.3. Outcome Mortality was assessed using the survival status in the TTMdataset, obtained from hospital registry, public registry or contact with patient, relatives or physician. Survival status and neurological function was assessed 6 months after randomization with 100% of patients successfully assessed for mortality and 99% for neurological outcome. An assessor, blinded for intervention, evaluated neurological outcome of the surviving patients by Cerebral Performance Category (CPC) and modified Rankin Scale (mRS) during a follow-up visit at 180 days after OHCA. 2.4. Statistical analyses Data are presented as total number of patients and proportion (%) of patients in men and women. Continuous data are reported as mean and standard deviation (SD), median and interquartile range (IQR) or median with 25–75% quartiles for time intervals and differences were assessed by unpaired two-sample t-test for differences in means, Pearson’s x2 test was used to test for distribution differences between two categories and Monte Carlo simulation was applied upon encountering counts ≤5. Kruskal–Wallis rank sum test was used to test the distribution of variables among more than two categories. Logistic regression was used to assess factors influencing favorable vs. unfavorable neurological outcome and performance of acute CAG and PCI, defined as CAG and PCI performed ≤24 h post ROSC. Mortality was assessed by Kaplan–Meier plots and Cox regression and compared using log rank tests. Multivariate proportional hazard Cox regression was applied to assess the influence on other variables, including inclusion site (largest two sites vs. other sites as defined in the TTM main study18 ). Hazard ratio (HR) and 95% confidence interval (CI) are reported. Survival curves were adjusted to the mean values of the population and plotted for men and women in order to assess survival differences if mean values in all variables are assumed. Differences between men and women in neurological outcome were tested with x2 test, also for favorable vs. non-favorable outcome and Monte Carlo simulation was applied upon encountering counts ≤5. All analyses were performed in R.3.1.2.21 Missing values were present at levels <10% and thus we did not perform imputation. 3. Results Randomization was reasonably balanced between the two sexes. Men and women did not differ with regards to age, type of cooling device presence of bystander, bystander CPR, time from cardiac arrest to basic life support (BLS), advanced life support (ALS) and ROSC (Table 1). We also found no difference in lactate at admission, primary rhythm (shockable vs. non-shockable) and ECG at admission (Table 1). Women more often had OHCA at home than men (61% vs. 52%, p = 0.04), had a lower comorbidity index (57% of women had CI 0 vs. 49% men, p = 0.04) and bystander defibrillation was less frequent in women (4% vs. 11%, p = 0.01, Table 1). Acute CAG within 24 h after ROSC was performed less frequently in women than in men (56% vs. 66%, p = 0.02, Table 1) but delay to CAG was not different. The same was true for PCI
80
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
Table 1 Population and OHCA characteristics in men and women. OHCA: out-of-hospital cardiac arrest, CPR: cardiopulmonary resuscitation, BLS: basic life support, ALS: advanced life support, ROSC: return of spontaneous circulation, VF: ventricular fibrillation, VT: ventricular tachycardia, PEA: pulseless electrical activity, CA: cardiac arrest, ECG: electrocardiogram, LBBB: left bundle branch block, CCI: Charlson comorbidity index.
Age (years) Temperature 33 36 Cooling by surface device OHCA at home Witnessed OHCA Bystander CPR Bystander defibrillation Time from OHCA to BLS (min) Time from OHCA to ALS (min) Time from OHCA to ROSC (min) Lactate at admission (mmol/L) First monitored rhythm VF/VT PEA Asystole ROSC—bystander CPR/defibrillation Acute coronary angiography (<24 h) Time to CAG (h) Acute PCI (<24 h) Time to PCI (h) ECG findings at admission Acute ST-elevation Acute LBBB Normal or unchanged Other abnormality CAG > 24 h post ROSC Time to CAG (days) PCI > 24 h post ROSC Time to PCI (days) CT of the brain Time to CT (h) EEG Time to EEG (h) MRI of the brain Time to MRI (h) SSEP Time to SSEP (h) Pacemaker Time to Pacemaker (days) ICD Time to ICD (days) Cause of death Cerebral Cardiovascular Multi organ failure Other/unknown Charlson comorbidity index CI 0 CI 1 CI 2 CI ≥ 3
Men (n = 761)
Women (n = 178)
65 (56–72)
66 (57.25–73)
0.66
393 (52%) 368 (48%) 567 (75%) 392 (52%) 685 (90%) 555 (73%) 86 (11%) 1 (0–2) 9 (6–13) 25 (17–39) 6 (3.0–9.4)
80 (45%) 98 (55%) 127 (71%) 108 (61%) 153 (86%) 128 (72%) 7 (4%) 1 (0–3) 10 (6–14) 24 (17–40) 6.15 (3.1–9.93)
0.13
600 (79%) 49 (6%) 86 (11%) 12 (2%) 500 (66%) 2 (1–3) 330 (43%) 2 (2–3)
129 (72%) 16 (9%) 27 (15%) 1 (1%) 99 (56%) 2 (1–3) 59 (33%) 2 (2–3)
0.08
321 (42%) 40 (5%) 179 (24%) 212 (28%) 81 (11%) 8 (5–17) 44 (6%) 9 (4.75–17.25) 263 (35%) 14 (3–88.75) 325 (43%) 72 (51–99.5) 30 (4%) 213 (140.5–307.2) 164 (22%) 93 (70–117) 14 (2%) 19 (15–22.75) 114 (15%) 16 (11–24)
63 (35%) 15 (8%) 39 (22%) 59 (33%) 21 (12%) 2 (1–7) 12 (7%) 6 (3–9) 79 (44%) 32 (3–97) 64 (36%) 84.5 (57.25–125.5) 5 (3%) 309 (184–411) 34 (19%) 99 (71–109) 6 (3%) 14.5 (7.25–20.25) 25 (14%) 16 (12–20)
0.42
213 (59%) 81 (23%) 41 (11%) 23 (6%)
53 (52%) 30 (29%) 15(15%) 4 (4%)
374 (49%) 226 (30%) 121 (16%) 40 (5%)
102 (57%) 34 (19%) 32 (18%) 10 (6%)
within 24 h after ROSC (33% vs. 43%, p = 0.02, Table 1) with similar delay. However, female sex was not associated with odds of having CAG or PCI within 24 h post ROSC in logistic regression when adjusting for confounding factors (CAG: OR = 0.49, CI = 0.14–1.97, p = 0.28, PCI: OR = 0.61, CI = 0.23–1.68 p = 0.32) in the total patient population. There was no interaction between female sex and STEMI, p = 0.57. However, when stratifying by STEMI and No-STEMI subgroups female sex was associated with lower odds of having CAG performed in the STEMI group (ORcagstemi = 0.46, CIcagstemi = 0.21–1.00, pcagstemi = 0.045), but not with statistically significant lower odds of PCI in the STEMI group (ORpcistemi = 0.56, CIpcistemi = 0.30–1.09, ppcistemi = 0.08) nor CAG/PCI in the No-STEMI group, (ORcagnostemi = 0.71, CIcagnostemi = 0.44–1.13, pcagnostemi = 0.15, ORpcinostemi = 0.78, CIpcinostemi = 0.42–1.37, ppcinostemi = 0.40).
p Value
0.44 0.04 0.27 0.53 0.01 0.27 0.87 0.97 0.90
0.02 0.66 0.02 0.99
0.71 0.049 0.73 0.21 0.02 0.95 0.12 0.19 0.62 0.82 0.54 0.13 0.39 0.26 0.93 0.14
0.01
0.04
For CAG and PCI performed after 24 h post ROSC, there was no difference between men and women (p = 0.71, Table 1) but delay to CAG was shorter in women (2 vs. 8 days, p < 0.05, Table 1). For neurological prognostication, CT of the brain was less often performed in men (35% vs. 44%, p = 0.02, Table 1) but delay was not different (Table 1). EEG, MRI and SSEP frequency did not differ. There was also no difference between pacemaker and implantable cardioverter defibrillator (ICD) implantation and delay to implantation. More women died from cardiovascular causes (29% vs. 23%, p < 0.01, Table 1) and more men died from cerebral causes (59% vs. 52%, p = 0.01, Table 1). More women died from multi organ failure (15% vs. 11%, p = 0.01). Death from unknown causes was 6% in men and 4% in women Hazard ratios (HR) for death in women was significantly higher in univariate analysis at HR: 1.29, CI: 1.04–1.61, p = 0.02, Table 2,
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
81
Table 2 Hazard ratios for female sex from Cox regression. Univariate
Female sex
Model 1
Model 2
Model 3
HR (95% CI)
p
HR (95% CI)
p
HR (95% CI)
p
HR (95% CI)
p
1.29 (1.04–1.61)
0.02
1.26 (1.01–1.57)
0.04
1.13 (0.90–1.42)
0.29
1.11 (0.87–1.41)
0.42
Model 1, basic demographic factors: Female sex, age at arrest, Comorbidity index. Model 2, risk factors associated with OHCA and resuscitation: female sex, arrest at home, age at arrest, inclusion at 2 largest sites, bystander CPR, bystander defibrillation, time to ROSC (min), Comorbidity index, shockable primary rhythm. Model 3, Demographic, Pre-hospital–and in-hospital risk factors: female sex, TTM 36 C, age at arrest, public arrest, inclusion at 2 largest sites, witnessed arrest, bystander CPR, bystander defibrillation, time to ROSC (min), lactate at admission, CAG first 24 h post ROSC, Comorbidity index, cooling device, STEMI at admission ECG, shockable primary rhythm.
Fig. 1. Survival plots: (A) Observed mortality in men and women, full study period. Differences tested by log-rank test, p < 0.02. (B) Survival curves adjusted to the mean values of the total population.
Fig. 1A, and this was still significant when adjusting for demographic variables (age and comorbidity) (HR: 1.26, CI: 1.01–1.57, p = 0.04, Table 2). When adding OHCA circumstances to the model the risk was still increased; but was not significant (HR; 1.13, CI: 0.90–1.42, p = 0.29, Table 2). This did not change when adjusting for in-hospital treatment (HR: 1.11, CI: 0.87–1.41, p = 0.42, Table 2). When adjusting men and women to the mean values of the population, there was no difference in mortality, Fig. 1B. No significant interaction between sex and TTM group was found (p = 0.73) (Fig. 2), or between sex and age (p = 0.35). Neither was there any interaction between sex and primary rhythm (p = 0.61). Specific sub-analysis did not reveal any interaction between sex and age <50, p = 0.63. Neurological outcome including death was significantly different between men and women for both CPCmen = 3 (1–5), CPCwomen = 5 (1–5), p = 0.047) and mRSmen = 4 (1–6), mRSwomen = 6 (2–6), p = 0.01), Fig. 3 implying that both are worse in women. When limiting analysis to survivors; CPC did not differ between men and women CPCmen = 1 (1–1), CPCwomen = 1 (1–2), p = 0.19, Fig. 3) but mRS remained significantly different mRSmen = 1 (0–2), mRSwomen = 1 (1–1), p = 0.01, Fig. 3). In linear regression, female sex was not associated with unfavorable outcome in survivors when adjusting for other confounders, neither assessed by CPC score (p = 0.27) or mRS (p = 0.83). Sex and target temperature did not interact neither for CPC (p = 0.77) or mRS (p = 0.65).
4. Discussion Female sex was associated with higher overall mortality. This difference may be explained by differences in OHCA circumstances and maybe also in in-hospital use of CAG and PCI, despite female sex only being associated with use of CAG in the STEMI group. Thus, when adjusting for other confounders, there was no longer a statistically significant mortality difference, even though the point estimate continued to suggest an increased risk in women. In this
data set, female sex did not modify the effect of 33 or 36 ◦ C and thus the data neither suggest a benefit of TTM at 33 ◦ C compared to 36 ◦ C in women nor in men, nor in any age groups in the two sexes. 4.1. Pre-hospital factors and acute in-hospital management OHCA in women occurred less often in public places, but in the present study this did not affect the rate of CPR administered by bystanders in the cohort, contrary to other findings.1,7 However, the higher proportion of men that received defibrillation may in part be explained by the higher proportion of cardiac arrest in public places, since AEDs are predominantly available in public spaces.22 Women are often found to be older than men at arrest,1,6,7 but we did not find age differences in our study. Contrary to the investigations previously discussed,1,4–10 the current study only deals with successfully resuscitated patients who survived long enough for ICU admission and enrollment in the TTM trial. In case older age of women in studies that encompass the entire cardiac arrest population is driven by unsuccessfully resuscitated patients, this would explain the lack of difference in the current study. We did, as others, find significant differences between the use of acute CAG12,13 and PCI,14 and in the group of patients presenting with STEMI, female sex was associated with lower chance of CAG being performed in multivariate linear regression. The reason for not performing CAG/PCI was not recorded in our dataset, so the difference in utilization and to which extent it was justified cannot be further explored. It should be noted that the difference did not seem to be rooted in a different proportion of STEMI, age or severity of comorbidities, since these factors did not differ between men and women. 4.2. Outcome When comparing studies regarding survival in cardiac arrest patients it is crucial to keep in mind whether the population studied
82
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
Fig. 2. Bar plot of % mortality in women and men stratified by TTM allocation group and age quartiles with 95% confidence intervals. Differences in quartiles were tested by log-rank test. There is no significant interaction between sex and TTM allocation group, p = 0.35.
Fig. 3. Neurological outcome: (A) Neurological outcome assessed by CPC in men and women. (B) Neurological outcome assessed by mRS in men and women. Outcome was significantly different, CPC: p = 0.047, mRS: p = 0.01. In survivors, CPC was not different between men and women (p = 0.19) but mRS was with a higher rate of favorable outcome in men (p = 0.01).
consists of all EMS-attended OHCA patients, or those successfully resuscitated. Since female OHCA patients are often older,1,6,7 the risk of being widowed and live alone is higher, and it is likely that assessing survival in women compared to men in these two different populations represent completely different scenarios, because severe cases with longer time to ALS/BLS may not be equally distributed among the two sexes. In addition, different end points in studies can make comparisons challenging. Studies on EMS-attended cardiac arrest assess different combinations of endpoints such as ROSC,1,4–9 survival to discharge,1,6,8–10 neurological outcome3,9 or 30-day survival.3,4,7,9 Results in these studies point in different directions, with women more frequently being resuscitated1,5,6,8 but with no difference in survival,1,6,8 worse3 as well as better survival compared to male OHCA patients.4,7 The population of successfully resuscitated patients has been studied with survival to admission11 and 30-day survival2 but still results differ with men having better survival in univariate analysis2 and the excess risk found in women being smaller with increasing age.11 Our study contributes with data on a minimum of 6-month survival as well as neurological outcome in a resuscitated population of OHCA patients, and we find that the higher mortality found in women is not statistically significant when adjusting for confounders. When adjusting men and women to the background population the risk difference is also not significant. Survival is increased in adjusted curves (Fig. 1B) compared to univariate curves (Fig. 1A), as the adjustment uses mean values in analysis and thus does not take into account differences in OHCA circumstances between men and women. As such, the
mortality risk found in women may to some extent be explained by the lower frequency of public arrest and lower rate of utilization of PCI. Neurological outcome was distributed similarly between men and women as measured by CPC or mRS, but when limiting the analysis to survivors, a higher rate of unfavorable outcome in women when assessed by mRS was found. However, female sex was not associated with outcome when adjusting for other factors. Other studies have suggested that women have poorer outcome after stroke,23,24 which may, to some extent be explained by women being less often independent and more often widowed.23,24 Civil status was not recorded in our study, but patients who were CPC 3 or 4 pre-arrest were excluded. The CPC score has previously been criticized because it is not a reflection of the patient’s own assessment of function.25 It is weighted towards mental function and may overestimate the number of patients with favorable outcome.26,27 The TTM trial also performed mRS to detect differences in physical ability and increase differentiation among the good outcomes group.26 Both scales have previously been applied in men and women. Neurological outcome including death was different among men and women, but in survivors female sex is not associated with lower odds of favorable outcome in multivariate linear regression. This does not rule out that further testing would reveal functional differences between men and women. We did not detect differences related to age between men and women for mortality or neurological outcome. We also could not replicate a survival advantage in women of reproductive age (<50) that others report for OHCA9,28 as well as for in-hospital cardiac
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
arrest.29 Others again have not been able to detect survival benefit in women of reproductive age,8 and as only 29 women and 89 men in our data were <50, the power to detect interactions is likely limited. 4.3. Effect of target temperature There was no difference in the effect of target temperature in men and women overall, even though Fig. 2a appears to show a different effect in women in the two oldest age quartiles. However, as these subgroups are small this is likely due to random error. Further analysis showed no influence of age in mortality, but again, the sample size at younger and older ages is small and thus the power to detect differences in subgroups is limited. This should be further explored in a larger sample, designed to detect differences in age groups in the two sexes. 5. Limitations The power calculation for the TTM trial was based on the assumption that there are no mortality differences between men and women. Thus power is not sufficient in a stratified analysis on mortality differences between men and women at younger or older ages, where sample size is smaller in the strata. The lack of registration of reasons for not performing CAG/PCI hinders detection of patterns behind these decisions other that the crude demographic data reported. 6. Conclusion Comatose female and male OHCA survivors differ from each other with regards to arrest circumstances and in-hospital treatment. Female sex is associated with worse outcome, but this seems to be explained to large extent by differences in arrest circumstances and perhaps in-hospital management. Women received fewer CAG and PCI than men, but female sex was not associated with lower utilization of CAG/PCI in multivariate analysis. It is not clear whether the fewer CAG and PCI are justified by a lower likelihood of indications for catheter related interventions. Target temperature management at 33 ◦ C does not seem to be associated with benefit compared to 36 ◦ C in neither men nor women. Conflict of interest statement MSc. Winther–Jensen, Drs. Åneman, Cronberg, Pellis, Hovdenes, Horn,Kuiper, Wanscher, Gasche, Stammet, Koopmans, Hassager and Erlinge report no conflicts of interest. Dr. Friberg reports lecture fees from BARD Medical and NatusInc., outside the submitted work. Dr. Kjaergaard reports grants from The EU Interreg IV A pro-gramme funding for establishing ‘Centre for Resuscitation Sciencein the Oresund Region’ 2013–2014, grants from Danish HeartFoundation: co-funding for the ‘Centre for Resuscitation Science in the Oresund Region’ 2013–2014, during the conduct of the study. Dr. Nielsen reports Speakers Honorarium for BARD Medical. Dr. Wise reports personal fees from BARD medical, outside the submitted work. Funding This work was supported by the European regional development fund through the Interreg IV A OKS programme [NYPS ID: 167157] and the Danish Heart Foundation [grant no: 13-04-R94A4516-22755] The TTM main study was supported by independent research grants from the Swedish Heart–Lung Foundation,
83
Arbetsmarknadens Försäkringsaktiebolag Insurance Foundation, Swedish Research Council, RegionSkåne (Sweden), National Health Service (Sweden), Thelmaoega Foundation, Krapperup Foundation, Thure Carlsson Foundation, Hans-Gabriel and Alice Trolle-Wachtmeister Foundation for Medical Research, Skåne University Hospital, TrygFonden (Denmark), and European Clinical Research Infrastructures Network.
References 1. Kim C, Fahrenbruch CE, Cobb La, Eisenberg MS. Out-of-hospital cardiac arrest in men and women. Circulation 2001;104:2699–703, http://dx.doi.org/ 10.1161/hc4701.099784. 2. Søholm H, Møller JE, Pedersen F, et al. Tertiary centres have improved survival compared to other hospitals in the Copenhagen area after outof-hospital cardiac arrest. Resuscitation 2012;84:162–7, http://dx.doi.org/ 10.1016/j.resuscitation.2012.06.029. 3. Akahane M, Ogawa T, Koike S, et al. The effects of sex on out-of-hospital cardiac arrest outcomes. Am J Med 2011;124:325–33, http://dx.doi.org/ 10.1016/j.amjmed.2010.10.020. 4. Adielsson A, Hollenberg J, Karlsson T, et al. Increase in survival and bystander CPR in out-of-hospital shockable arrhythmia: bystander CPR and female gender are predictors of improved outcome. Experiences from Sweden in an 18-year perspective. Heart 2011;97:1391–6, http://dx.doi.org/10.1136/hrt.2011.222711. 5. Wigginton JG, Pepe PE, Bedolla JP, DeTamble L a, Atkins JM. Sex-related differences in the presentation and outcome of out-of-hospital cardiopulmonary arrest: a multiyear, prospective, population-based study. Crit Care Med 2002;30:S131–6. 6. Perers E, Abrahamsson P, Bång a, Engdahl J, Lindqvist J, Karlson BW. There is a difference in characteristics and outcome between women and men who suffer out of hospital cardiac arrest. Resuscitation 1999;40:133–40. 7. Herlitz J, Engdahl J, Svensson L, Young M, Angquist K-A, Holmberg S. Is female sex associated with increased survival after out-of-hospital cardiac arrest? Resuscitation 2004;60:197–203, http://dx.doi.org/10.1016/ j.resuscitation.2003.09.012. 8. Bray JE, Stub D, Bernard S, Smith K. Exploring gender differences and the oestrogen effect in an Australian out-of-hospital cardiac arrest population. Resuscitation 2013;84:957–63, http://dx.doi.org/10.1016/j.resuscitation.2012.12.004. 9. Kitamura T, Iwami T, Nichol G, et al. Reduction in incidence and fatality of out-of-hospital cardiac arrest in females of the reproductive age. Eur Heart J 2010;31:1365–72, http://dx.doi.org/10.1093/eurheartj/ehq059. 10. Teodorescu C, Reinier K, Uy-Evanado A, Ayala J, Mariani R, Wittwer L, et al. Survival advantage from ventricular fibrillation and pulseless electrical activity in women compared to men: the oregon sudden unexpected death study. J Interv Card Electrophysiol 2012;34:219–25. 11. Arrich J, Sterz F, Fleischhackl R, et al. Gender modifies the influence of age on outcome after successfully resuscitated cardiac arrest: a retrospective cohort study. Medicine (Baltimore) 2006;85:288–94, http://dx.doi.org/ 10.1097/01.md.0000236954.72342.20. 12. Waldo SW, Armstrong EJ, Kulkarni A, et al. Comparison of clinical characteristics and outcomes of cardiac arrest survivors having versus not having coronary angiography. Am J Cardiol 2013;111:1253–8, http://dx.doi.org/ 10.1016/j.amjcard.2013.01.267. 13. Alfredsson J, Stenestrand U, Wallentin L, Swahn E. Gender differences in management and outcome in non-ST-elevation acute coronary syndrome. Heart 2007;93:1357–62, http://dx.doi.org/10.1136/hrt.2006.102012. 14. Radovanovic D, Erne P, Urban P, Bertel O, Rickli H, Gaspoz J-M. Gender differences in management and outcomes in patients with acute coronary syndromes: results on 20,290 patients from the AMIS Plus Registry. Heart 2007;93:1369–75, http://dx.doi.org/10.1136/hrt.2006.106781. 15. Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature management at 33 ◦ C versus 36 ◦ C after cardiac arrest. N Engl J Med 2013;369:2197–206. 16. Lilja G, Nielsen N, Friberg H, et al. Cognitive function in survivors of outof-hospital cardiac arrest after target temperature management at 33 ◦ C versus 36 ◦ C. Circulation 2015, http://dx.doi.org/10.1161/CIRCULATIONAHA. 114.014414. 17. Nielsen N, Hovdenes J, Bishop G, et al. Target temperature management after out-of-hospital cardiac arrest—a randomized, parallel-group, assessor-blinded clinical trial—rationale and design. Am Heart J 2012;163:541–8. 18. Nielsen N, Wetterslev J, Cronberg T, et al. Supplement to: targeted temperature management at 33 ◦ C versus 36 ◦ C after cardiac arrest. N Engl J Med 2013;369:2197–206, http://dx.doi.org/10.1056/NEJMoa1310519. 19. Langhelle A, Nolan J, Herlitz J, et al. Recommended guidelines for reviewing, reporting, and conducting research on post-resuscitation care: the Utstein style. Resuscitation 2005;66:271–83. 20. Charlson ME, Pompei P, Ales KLMC. A new method of classifying prognostic in longitudinal studies: development. J Chronic Dis 1987;40:373–83. 21. R Core Team. R foundation for statistical computing. R.3.01. Vienna, Austria: R Core Team; 2013 http://www.R–project.org/
84
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
22. Folke F, Gislason GH, Lippert FK, et al. Differences between out-of-hospital cardiac arrest in residential and public locations and implications for public-access defibrillation. Circulation 2010;122:623–30, http://dx.doi.org/ 10.1161/CIRCULATIONAHA.109.924423. 23. Kapral MK, Fang J, Hill MD, et al. Sex differences in stroke care and outcomes: results from the registry of the Canadian Stroke Network. Stroke 2005;36:809–14, http://dx.doi.org/10.1161/01.STR.0000157662.09551.e5. 24. Reeves MJ, Bushnell CD, Howard G, et al. Sex differences in stroke: epidemiology, clinical presentation, medical care, and outcomes. Lancet Neurol 2008;7:915–26, http://dx.doi.org/10.1016/S1474-4422(08)70193-5. 25. Hsu JW, Madsen CD, Callaham ML. Quality-of-life and formal functional testing of survivors of out-of-hospital cardiac arrest correlates poorly with traditional neurologic outcome scales. Ann Emerg Med 1996;28:597–605.
26. Rittenberger JC, Raina K, Holm MB, Kim YJ, Callaway CW. Association between cerebral performance category, modified Rankin Scale, and discharge disposition after cardiac arrest. Resuscitation 2011;82:1036–40, http://dx.doi.org/10.1016/j.resuscitation.2011.03.034. 27. Raina KD, Callaway C, Rittenberger JC, Holm MB. Neurological and functional status following cardiac arrest: method and tool utility. Resuscitation 2008;79:249–56. 28. Johnson MA, Haukoos JS, Larabee TM, et al. Females of childbearing age have a survival benefit after out-of-hospital cardiac arrest. Resuscitation 2012;84:639–44, http://dx.doi.org/10.1016/j.resuscitation.2012.09.011. 29. Topjian Aa, Localio AR, Berg Ra, et al. Women of child-bearing age have better in hospital cardiac arrest survival outcomes than do equal-aged men. Crit Care Med 2010;38:1254–60, http://dx.doi.org/10.1097/CCM.0b013e3181d8ca43.
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
No difference in mortality between men and women after out-of-hospital cardiac arrest夽 Matilde Winther-Jensen a,∗ , Jesper Kjaergaard a , Michael Wanscher a , Niklas Nielsen b , Jørn Wetterslev c , Tobias Cronberg d , David Erlinge e , Hans Friberg f , Yvan Gasche g , Janneke Horn h , Jan Hovdenes i , Michael Kuiper j , Tommaso Pellis k , Pascal Stammet l , Matthew P. Wise m , Anders Åneman n , Christian Hassager a a
The Heart Centre, Copenhagen University Hospital, Copenhagen, Denmark Department of Anesthesia and Intensive Care, Helsingborg Hospital, Helsingborg, Sweden c Copenhagen Trial Unit, Rigshospitalet, Copenhagen, Denmark d Department of Neurology, Skåne University Hospital, Lund, Sweden e Department of Cardiology, Skåne University Hospital, Lund, Sweden f Department of Anesthesia and Intensive Care, Skåne University Hospital, Lund, Sweden g Department of Intensive Care, Geneva University Hospital, Geneva, Switzerland h Department of Intensive Care, Academic Medical Centrum, Amsterdam, The Netherlands i Department of Anesthesia and Intensive Care, Oslo University Hospital, Rikshospitalet, Oslo, Norway j Department of Intensive Care, Leeuwarden Medical Centrum, Leeuwarden, The Netherlands k Department of Intensive Care, Santa Maria degli Angeli, Pordenone, Italy l Department of Anesthesia and Intensive Care, Centre Hospitalier de Luxembourg, Ernest Barblé, Luxembourg m Department of Intensive Care, University Hospital of Wales, Cardiff, United Kingdom n Department of Intensive Care, Liverpool hospital, Sydney, NSW, Australia b
a r t i c l e
i n f o
Article history: Received 23 March 2015 Received in revised form 23 June 2015 Accepted 29 June 2015 Keywords: Out of hospital cardiac arrest Therapeutic hypothermia Women Sex
a b s t r a c t Aim: Comparing the outcome after out-of-hospital cardiac arrest (OHCA) in men and women and to determine whether sex modifies the effect of targeted temperature management (TTM) at 33 or 36 ◦ C. Methods: The TTM trial randomized 950 patients to TTM at 33 or 36 ◦ C for 24 h. This predefined sub-study of the TTM trial assessed survival and neurological outcome defined as Cerebral Performance Category (CPC) and modified Rankin Scale (mRS) using female sex as main predictor of outcome, in relation to level of TTM and other confounding factors. Results: Compared to men, women more often had OHCA at home, p = 0.04 and less often had bystander defibrillation, p = 0.01. No other differences in arrest circumstances were found. Coronary angiography (CAG) and percutaneous coronary intervention (PCI) <24 h after ROSC was less often performed in women, both: p = 0.02. Female sex was associated with higher mortality in univariate analysis, hazard ratio (HR) = 1.29, CI = 1.04–1.61, p = 0.02 compared to men. Adjusting for demographic factors (age and comorbidity), arrest circumstances, pre-hospital findings, inclusion sites, treatments and status at admission reduced this: HR = 1.11, CI = 0. 87–1.41, p = 0.42, and sex was no longer an independent risk factor for death. The effect of sex did not modify the effect of TTM at 33 and 36 ◦ C, pinteraction = 0.73. Conclusion: Female sex seems associated with adverse outcome, but this association is largely explained by differences in arrest circumstances and in-hospital treatment. Our data shows no interaction between sex and the effect of targeting 33 vs. 36 ◦ C. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction 夽 A Spanish translated version of the summary of this article appears as Appendix in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2015.06.030. ∗ Corresponding author. E-mail address: [email protected] (M. Winther-Jensen). http://dx.doi.org/10.1016/j.resuscitation.2015.06.030 0300-9572/© 2015 Elsevier Ireland Ltd. All rights reserved.
Unexpected cardiac arrest is reported to be more common in men1 with an annual rate of cardiac arrest of 0.16% vs. 0.085% in women. Previous studies have reported worse2,3 outcome in women whereas others find a survival benefit of
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
women compared to men in out-of-hospital cardiac arrest (OHCA) victims.4–7 It should be kept in mind that the cardiac arrest populations studied are not always comparable. Some studies included the entire population of emergency service (EMS) attended cardiac arrests,1,3–10 whilst other focused on the population of successfully resuscitated patients.2,11 The observed differences in outcome may in part be explained by dissimilarity in demographic characteristics and in circumstances of the cardiac arrest. Women are reported to have a lower rate of primary rhythm of ventricular fibrillation (VF), to be older at the time of OHCA, less often suffer witnessed arrest, less frequently receive cardiopulmonary resuscitation (CPR) by bystanders and have OHCA from cardiac etiology.1,6,7 Disparity in utilization of coronary angiography (CAG) and percutaneous coronary intervention (PCI) between men and women have been found previously,12–14 which poses the question whether in-hospital treatment may in part contribute to the reported mortality differences along with biological differences, pre-hospital factors and OHCA characteristics. This study is a post hoc analysis of the Target Temperature Management (TTM) trial, which compared target temperatures of 33 ◦ C and 36 ◦ C, finding no overall differences in survival and neurological outcome as well as no interaction of sex and level of TTM.15 This pre-defined post-hoc sub-study reports an in-depth analysis of the differences in mortality as well as neurological outcome and interaction between age and sex in a contemporary cohort of comatose OHCA survivors consisting of men and women, and investigates whether the effect of target temperature management at 33 vs. 36 ◦ C is different in men and women.
2. Methods 2.1. Study design This study is a pre-defined sub-study of the TTM-trial which included 950 OHCA patients admitted to 36 intensive care units (ICU’s) in Europe and Australia from November 2010 to January 2013.15 Inclusion criteria included age ≥18 years, OHCA of presumed cardiac cause, sustained ROSC and Glasgow Coma Scale <8 (patient not able to obey verbal commands). The primary outcome was all-cause mortality at the end of the trial in July 2013 and neither the primary nor secondary or tertiary outcomes of neurological outcome were different in the 33 ◦ C and 36 ◦ C groups.15,16 A prespecified sub group analysis did not show any overall difference of the effect of TTM at 33 vs. 36 ◦ C men and women.15 Randomization was performed through a web-based system, assigning patients in a 1:1 ratio stratified for site to either 33 ◦ C or 36 ◦ C target temperature treatment.17 Active treatment was continued in all patients until blinded neurological evaluation had been performed 108 h after arrest or later.18 Four patients withdrew consent and 7 patients were excluded from the intention-to-treat population and thus 939 patients were analyzed for mortality and neurological outcome.18 Pre-hospital data such as initial rhythm, presence of bystander, bystander performing CPR and time to ROSC were collected according to Utstein guidelines.19 EEG at 12-36 h and SSEP at 48–72 h were strongly recommended by protocol. All other treatment and clinical investigations were left at the discretion of the treating physician. We assessed severity of comorbidities by creating a comorbidity index based on the Charlson comorbidity index. The Charlson index is a weighted and validated index based on 22 conditions and is used to predict short-term mortality.20
79
2.2. Ethical approval The TTM-trial is registered at ClinicalTrials.gov (Identifier: NCT01020916) and the protocol was approved by ethical committees in each participating country. Informed consent was waived or obtained from all participants or their next of kin according to national legislation. 2.3. Outcome Mortality was assessed using the survival status in the TTMdataset, obtained from hospital registry, public registry or contact with patient, relatives or physician. Survival status and neurological function was assessed 6 months after randomization with 100% of patients successfully assessed for mortality and 99% for neurological outcome. An assessor, blinded for intervention, evaluated neurological outcome of the surviving patients by Cerebral Performance Category (CPC) and modified Rankin Scale (mRS) during a follow-up visit at 180 days after OHCA. 2.4. Statistical analyses Data are presented as total number of patients and proportion (%) of patients in men and women. Continuous data are reported as mean and standard deviation (SD), median and interquartile range (IQR) or median with 25–75% quartiles for time intervals and differences were assessed by unpaired two-sample t-test for differences in means, Pearson’s x2 test was used to test for distribution differences between two categories and Monte Carlo simulation was applied upon encountering counts ≤5. Kruskal–Wallis rank sum test was used to test the distribution of variables among more than two categories. Logistic regression was used to assess factors influencing favorable vs. unfavorable neurological outcome and performance of acute CAG and PCI, defined as CAG and PCI performed ≤24 h post ROSC. Mortality was assessed by Kaplan–Meier plots and Cox regression and compared using log rank tests. Multivariate proportional hazard Cox regression was applied to assess the influence on other variables, including inclusion site (largest two sites vs. other sites as defined in the TTM main study18 ). Hazard ratio (HR) and 95% confidence interval (CI) are reported. Survival curves were adjusted to the mean values of the population and plotted for men and women in order to assess survival differences if mean values in all variables are assumed. Differences between men and women in neurological outcome were tested with x2 test, also for favorable vs. non-favorable outcome and Monte Carlo simulation was applied upon encountering counts ≤5. All analyses were performed in R.3.1.2.21 Missing values were present at levels <10% and thus we did not perform imputation. 3. Results Randomization was reasonably balanced between the two sexes. Men and women did not differ with regards to age, type of cooling device presence of bystander, bystander CPR, time from cardiac arrest to basic life support (BLS), advanced life support (ALS) and ROSC (Table 1). We also found no difference in lactate at admission, primary rhythm (shockable vs. non-shockable) and ECG at admission (Table 1). Women more often had OHCA at home than men (61% vs. 52%, p = 0.04), had a lower comorbidity index (57% of women had CI 0 vs. 49% men, p = 0.04) and bystander defibrillation was less frequent in women (4% vs. 11%, p = 0.01, Table 1). Acute CAG within 24 h after ROSC was performed less frequently in women than in men (56% vs. 66%, p = 0.02, Table 1) but delay to CAG was not different. The same was true for PCI
80
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
Table 1 Population and OHCA characteristics in men and women. OHCA: out-of-hospital cardiac arrest, CPR: cardiopulmonary resuscitation, BLS: basic life support, ALS: advanced life support, ROSC: return of spontaneous circulation, VF: ventricular fibrillation, VT: ventricular tachycardia, PEA: pulseless electrical activity, CA: cardiac arrest, ECG: electrocardiogram, LBBB: left bundle branch block, CCI: Charlson comorbidity index.
Age (years) Temperature 33 36 Cooling by surface device OHCA at home Witnessed OHCA Bystander CPR Bystander defibrillation Time from OHCA to BLS (min) Time from OHCA to ALS (min) Time from OHCA to ROSC (min) Lactate at admission (mmol/L) First monitored rhythm VF/VT PEA Asystole ROSC—bystander CPR/defibrillation Acute coronary angiography (<24 h) Time to CAG (h) Acute PCI (<24 h) Time to PCI (h) ECG findings at admission Acute ST-elevation Acute LBBB Normal or unchanged Other abnormality CAG > 24 h post ROSC Time to CAG (days) PCI > 24 h post ROSC Time to PCI (days) CT of the brain Time to CT (h) EEG Time to EEG (h) MRI of the brain Time to MRI (h) SSEP Time to SSEP (h) Pacemaker Time to Pacemaker (days) ICD Time to ICD (days) Cause of death Cerebral Cardiovascular Multi organ failure Other/unknown Charlson comorbidity index CI 0 CI 1 CI 2 CI ≥ 3
Men (n = 761)
Women (n = 178)
65 (56–72)
66 (57.25–73)
0.66
393 (52%) 368 (48%) 567 (75%) 392 (52%) 685 (90%) 555 (73%) 86 (11%) 1 (0–2) 9 (6–13) 25 (17–39) 6 (3.0–9.4)
80 (45%) 98 (55%) 127 (71%) 108 (61%) 153 (86%) 128 (72%) 7 (4%) 1 (0–3) 10 (6–14) 24 (17–40) 6.15 (3.1–9.93)
0.13
600 (79%) 49 (6%) 86 (11%) 12 (2%) 500 (66%) 2 (1–3) 330 (43%) 2 (2–3)
129 (72%) 16 (9%) 27 (15%) 1 (1%) 99 (56%) 2 (1–3) 59 (33%) 2 (2–3)
0.08
321 (42%) 40 (5%) 179 (24%) 212 (28%) 81 (11%) 8 (5–17) 44 (6%) 9 (4.75–17.25) 263 (35%) 14 (3–88.75) 325 (43%) 72 (51–99.5) 30 (4%) 213 (140.5–307.2) 164 (22%) 93 (70–117) 14 (2%) 19 (15–22.75) 114 (15%) 16 (11–24)
63 (35%) 15 (8%) 39 (22%) 59 (33%) 21 (12%) 2 (1–7) 12 (7%) 6 (3–9) 79 (44%) 32 (3–97) 64 (36%) 84.5 (57.25–125.5) 5 (3%) 309 (184–411) 34 (19%) 99 (71–109) 6 (3%) 14.5 (7.25–20.25) 25 (14%) 16 (12–20)
0.42
213 (59%) 81 (23%) 41 (11%) 23 (6%)
53 (52%) 30 (29%) 15(15%) 4 (4%)
374 (49%) 226 (30%) 121 (16%) 40 (5%)
102 (57%) 34 (19%) 32 (18%) 10 (6%)
within 24 h after ROSC (33% vs. 43%, p = 0.02, Table 1) with similar delay. However, female sex was not associated with odds of having CAG or PCI within 24 h post ROSC in logistic regression when adjusting for confounding factors (CAG: OR = 0.49, CI = 0.14–1.97, p = 0.28, PCI: OR = 0.61, CI = 0.23–1.68 p = 0.32) in the total patient population. There was no interaction between female sex and STEMI, p = 0.57. However, when stratifying by STEMI and No-STEMI subgroups female sex was associated with lower odds of having CAG performed in the STEMI group (ORcagstemi = 0.46, CIcagstemi = 0.21–1.00, pcagstemi = 0.045), but not with statistically significant lower odds of PCI in the STEMI group (ORpcistemi = 0.56, CIpcistemi = 0.30–1.09, ppcistemi = 0.08) nor CAG/PCI in the No-STEMI group, (ORcagnostemi = 0.71, CIcagnostemi = 0.44–1.13, pcagnostemi = 0.15, ORpcinostemi = 0.78, CIpcinostemi = 0.42–1.37, ppcinostemi = 0.40).
p Value
0.44 0.04 0.27 0.53 0.01 0.27 0.87 0.97 0.90
0.02 0.66 0.02 0.99
0.71 0.049 0.73 0.21 0.02 0.95 0.12 0.19 0.62 0.82 0.54 0.13 0.39 0.26 0.93 0.14
0.01
0.04
For CAG and PCI performed after 24 h post ROSC, there was no difference between men and women (p = 0.71, Table 1) but delay to CAG was shorter in women (2 vs. 8 days, p < 0.05, Table 1). For neurological prognostication, CT of the brain was less often performed in men (35% vs. 44%, p = 0.02, Table 1) but delay was not different (Table 1). EEG, MRI and SSEP frequency did not differ. There was also no difference between pacemaker and implantable cardioverter defibrillator (ICD) implantation and delay to implantation. More women died from cardiovascular causes (29% vs. 23%, p < 0.01, Table 1) and more men died from cerebral causes (59% vs. 52%, p = 0.01, Table 1). More women died from multi organ failure (15% vs. 11%, p = 0.01). Death from unknown causes was 6% in men and 4% in women Hazard ratios (HR) for death in women was significantly higher in univariate analysis at HR: 1.29, CI: 1.04–1.61, p = 0.02, Table 2,
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
81
Table 2 Hazard ratios for female sex from Cox regression. Univariate
Female sex
Model 1
Model 2
Model 3
HR (95% CI)
p
HR (95% CI)
p
HR (95% CI)
p
HR (95% CI)
p
1.29 (1.04–1.61)
0.02
1.26 (1.01–1.57)
0.04
1.13 (0.90–1.42)
0.29
1.11 (0.87–1.41)
0.42
Model 1, basic demographic factors: Female sex, age at arrest, Comorbidity index. Model 2, risk factors associated with OHCA and resuscitation: female sex, arrest at home, age at arrest, inclusion at 2 largest sites, bystander CPR, bystander defibrillation, time to ROSC (min), Comorbidity index, shockable primary rhythm. Model 3, Demographic, Pre-hospital–and in-hospital risk factors: female sex, TTM 36 C, age at arrest, public arrest, inclusion at 2 largest sites, witnessed arrest, bystander CPR, bystander defibrillation, time to ROSC (min), lactate at admission, CAG first 24 h post ROSC, Comorbidity index, cooling device, STEMI at admission ECG, shockable primary rhythm.
Fig. 1. Survival plots: (A) Observed mortality in men and women, full study period. Differences tested by log-rank test, p < 0.02. (B) Survival curves adjusted to the mean values of the total population.
Fig. 1A, and this was still significant when adjusting for demographic variables (age and comorbidity) (HR: 1.26, CI: 1.01–1.57, p = 0.04, Table 2). When adding OHCA circumstances to the model the risk was still increased; but was not significant (HR; 1.13, CI: 0.90–1.42, p = 0.29, Table 2). This did not change when adjusting for in-hospital treatment (HR: 1.11, CI: 0.87–1.41, p = 0.42, Table 2). When adjusting men and women to the mean values of the population, there was no difference in mortality, Fig. 1B. No significant interaction between sex and TTM group was found (p = 0.73) (Fig. 2), or between sex and age (p = 0.35). Neither was there any interaction between sex and primary rhythm (p = 0.61). Specific sub-analysis did not reveal any interaction between sex and age <50, p = 0.63. Neurological outcome including death was significantly different between men and women for both CPCmen = 3 (1–5), CPCwomen = 5 (1–5), p = 0.047) and mRSmen = 4 (1–6), mRSwomen = 6 (2–6), p = 0.01), Fig. 3 implying that both are worse in women. When limiting analysis to survivors; CPC did not differ between men and women CPCmen = 1 (1–1), CPCwomen = 1 (1–2), p = 0.19, Fig. 3) but mRS remained significantly different mRSmen = 1 (0–2), mRSwomen = 1 (1–1), p = 0.01, Fig. 3). In linear regression, female sex was not associated with unfavorable outcome in survivors when adjusting for other confounders, neither assessed by CPC score (p = 0.27) or mRS (p = 0.83). Sex and target temperature did not interact neither for CPC (p = 0.77) or mRS (p = 0.65).
4. Discussion Female sex was associated with higher overall mortality. This difference may be explained by differences in OHCA circumstances and maybe also in in-hospital use of CAG and PCI, despite female sex only being associated with use of CAG in the STEMI group. Thus, when adjusting for other confounders, there was no longer a statistically significant mortality difference, even though the point estimate continued to suggest an increased risk in women. In this
data set, female sex did not modify the effect of 33 or 36 ◦ C and thus the data neither suggest a benefit of TTM at 33 ◦ C compared to 36 ◦ C in women nor in men, nor in any age groups in the two sexes. 4.1. Pre-hospital factors and acute in-hospital management OHCA in women occurred less often in public places, but in the present study this did not affect the rate of CPR administered by bystanders in the cohort, contrary to other findings.1,7 However, the higher proportion of men that received defibrillation may in part be explained by the higher proportion of cardiac arrest in public places, since AEDs are predominantly available in public spaces.22 Women are often found to be older than men at arrest,1,6,7 but we did not find age differences in our study. Contrary to the investigations previously discussed,1,4–10 the current study only deals with successfully resuscitated patients who survived long enough for ICU admission and enrollment in the TTM trial. In case older age of women in studies that encompass the entire cardiac arrest population is driven by unsuccessfully resuscitated patients, this would explain the lack of difference in the current study. We did, as others, find significant differences between the use of acute CAG12,13 and PCI,14 and in the group of patients presenting with STEMI, female sex was associated with lower chance of CAG being performed in multivariate linear regression. The reason for not performing CAG/PCI was not recorded in our dataset, so the difference in utilization and to which extent it was justified cannot be further explored. It should be noted that the difference did not seem to be rooted in a different proportion of STEMI, age or severity of comorbidities, since these factors did not differ between men and women. 4.2. Outcome When comparing studies regarding survival in cardiac arrest patients it is crucial to keep in mind whether the population studied
82
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
Fig. 2. Bar plot of % mortality in women and men stratified by TTM allocation group and age quartiles with 95% confidence intervals. Differences in quartiles were tested by log-rank test. There is no significant interaction between sex and TTM allocation group, p = 0.35.
Fig. 3. Neurological outcome: (A) Neurological outcome assessed by CPC in men and women. (B) Neurological outcome assessed by mRS in men and women. Outcome was significantly different, CPC: p = 0.047, mRS: p = 0.01. In survivors, CPC was not different between men and women (p = 0.19) but mRS was with a higher rate of favorable outcome in men (p = 0.01).
consists of all EMS-attended OHCA patients, or those successfully resuscitated. Since female OHCA patients are often older,1,6,7 the risk of being widowed and live alone is higher, and it is likely that assessing survival in women compared to men in these two different populations represent completely different scenarios, because severe cases with longer time to ALS/BLS may not be equally distributed among the two sexes. In addition, different end points in studies can make comparisons challenging. Studies on EMS-attended cardiac arrest assess different combinations of endpoints such as ROSC,1,4–9 survival to discharge,1,6,8–10 neurological outcome3,9 or 30-day survival.3,4,7,9 Results in these studies point in different directions, with women more frequently being resuscitated1,5,6,8 but with no difference in survival,1,6,8 worse3 as well as better survival compared to male OHCA patients.4,7 The population of successfully resuscitated patients has been studied with survival to admission11 and 30-day survival2 but still results differ with men having better survival in univariate analysis2 and the excess risk found in women being smaller with increasing age.11 Our study contributes with data on a minimum of 6-month survival as well as neurological outcome in a resuscitated population of OHCA patients, and we find that the higher mortality found in women is not statistically significant when adjusting for confounders. When adjusting men and women to the background population the risk difference is also not significant. Survival is increased in adjusted curves (Fig. 1B) compared to univariate curves (Fig. 1A), as the adjustment uses mean values in analysis and thus does not take into account differences in OHCA circumstances between men and women. As such, the
mortality risk found in women may to some extent be explained by the lower frequency of public arrest and lower rate of utilization of PCI. Neurological outcome was distributed similarly between men and women as measured by CPC or mRS, but when limiting the analysis to survivors, a higher rate of unfavorable outcome in women when assessed by mRS was found. However, female sex was not associated with outcome when adjusting for other factors. Other studies have suggested that women have poorer outcome after stroke,23,24 which may, to some extent be explained by women being less often independent and more often widowed.23,24 Civil status was not recorded in our study, but patients who were CPC 3 or 4 pre-arrest were excluded. The CPC score has previously been criticized because it is not a reflection of the patient’s own assessment of function.25 It is weighted towards mental function and may overestimate the number of patients with favorable outcome.26,27 The TTM trial also performed mRS to detect differences in physical ability and increase differentiation among the good outcomes group.26 Both scales have previously been applied in men and women. Neurological outcome including death was different among men and women, but in survivors female sex is not associated with lower odds of favorable outcome in multivariate linear regression. This does not rule out that further testing would reveal functional differences between men and women. We did not detect differences related to age between men and women for mortality or neurological outcome. We also could not replicate a survival advantage in women of reproductive age (<50) that others report for OHCA9,28 as well as for in-hospital cardiac
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
arrest.29 Others again have not been able to detect survival benefit in women of reproductive age,8 and as only 29 women and 89 men in our data were <50, the power to detect interactions is likely limited. 4.3. Effect of target temperature There was no difference in the effect of target temperature in men and women overall, even though Fig. 2a appears to show a different effect in women in the two oldest age quartiles. However, as these subgroups are small this is likely due to random error. Further analysis showed no influence of age in mortality, but again, the sample size at younger and older ages is small and thus the power to detect differences in subgroups is limited. This should be further explored in a larger sample, designed to detect differences in age groups in the two sexes. 5. Limitations The power calculation for the TTM trial was based on the assumption that there are no mortality differences between men and women. Thus power is not sufficient in a stratified analysis on mortality differences between men and women at younger or older ages, where sample size is smaller in the strata. The lack of registration of reasons for not performing CAG/PCI hinders detection of patterns behind these decisions other that the crude demographic data reported. 6. Conclusion Comatose female and male OHCA survivors differ from each other with regards to arrest circumstances and in-hospital treatment. Female sex is associated with worse outcome, but this seems to be explained to large extent by differences in arrest circumstances and perhaps in-hospital management. Women received fewer CAG and PCI than men, but female sex was not associated with lower utilization of CAG/PCI in multivariate analysis. It is not clear whether the fewer CAG and PCI are justified by a lower likelihood of indications for catheter related interventions. Target temperature management at 33 ◦ C does not seem to be associated with benefit compared to 36 ◦ C in neither men nor women. Conflict of interest statement MSc. Winther–Jensen, Drs. Åneman, Cronberg, Pellis, Hovdenes, Horn,Kuiper, Wanscher, Gasche, Stammet, Koopmans, Hassager and Erlinge report no conflicts of interest. Dr. Friberg reports lecture fees from BARD Medical and NatusInc., outside the submitted work. Dr. Kjaergaard reports grants from The EU Interreg IV A pro-gramme funding for establishing ‘Centre for Resuscitation Sciencein the Oresund Region’ 2013–2014, grants from Danish HeartFoundation: co-funding for the ‘Centre for Resuscitation Science in the Oresund Region’ 2013–2014, during the conduct of the study. Dr. Nielsen reports Speakers Honorarium for BARD Medical. Dr. Wise reports personal fees from BARD medical, outside the submitted work. Funding This work was supported by the European regional development fund through the Interreg IV A OKS programme [NYPS ID: 167157] and the Danish Heart Foundation [grant no: 13-04-R94A4516-22755] The TTM main study was supported by independent research grants from the Swedish Heart–Lung Foundation,
83
Arbetsmarknadens Försäkringsaktiebolag Insurance Foundation, Swedish Research Council, RegionSkåne (Sweden), National Health Service (Sweden), Thelmaoega Foundation, Krapperup Foundation, Thure Carlsson Foundation, Hans-Gabriel and Alice Trolle-Wachtmeister Foundation for Medical Research, Skåne University Hospital, TrygFonden (Denmark), and European Clinical Research Infrastructures Network.
References 1. Kim C, Fahrenbruch CE, Cobb La, Eisenberg MS. Out-of-hospital cardiac arrest in men and women. Circulation 2001;104:2699–703, http://dx.doi.org/ 10.1161/hc4701.099784. 2. Søholm H, Møller JE, Pedersen F, et al. Tertiary centres have improved survival compared to other hospitals in the Copenhagen area after outof-hospital cardiac arrest. Resuscitation 2012;84:162–7, http://dx.doi.org/ 10.1016/j.resuscitation.2012.06.029. 3. Akahane M, Ogawa T, Koike S, et al. The effects of sex on out-of-hospital cardiac arrest outcomes. Am J Med 2011;124:325–33, http://dx.doi.org/ 10.1016/j.amjmed.2010.10.020. 4. Adielsson A, Hollenberg J, Karlsson T, et al. Increase in survival and bystander CPR in out-of-hospital shockable arrhythmia: bystander CPR and female gender are predictors of improved outcome. Experiences from Sweden in an 18-year perspective. Heart 2011;97:1391–6, http://dx.doi.org/10.1136/hrt.2011.222711. 5. Wigginton JG, Pepe PE, Bedolla JP, DeTamble L a, Atkins JM. Sex-related differences in the presentation and outcome of out-of-hospital cardiopulmonary arrest: a multiyear, prospective, population-based study. Crit Care Med 2002;30:S131–6. 6. Perers E, Abrahamsson P, Bång a, Engdahl J, Lindqvist J, Karlson BW. There is a difference in characteristics and outcome between women and men who suffer out of hospital cardiac arrest. Resuscitation 1999;40:133–40. 7. Herlitz J, Engdahl J, Svensson L, Young M, Angquist K-A, Holmberg S. Is female sex associated with increased survival after out-of-hospital cardiac arrest? Resuscitation 2004;60:197–203, http://dx.doi.org/10.1016/ j.resuscitation.2003.09.012. 8. Bray JE, Stub D, Bernard S, Smith K. Exploring gender differences and the oestrogen effect in an Australian out-of-hospital cardiac arrest population. Resuscitation 2013;84:957–63, http://dx.doi.org/10.1016/j.resuscitation.2012.12.004. 9. Kitamura T, Iwami T, Nichol G, et al. Reduction in incidence and fatality of out-of-hospital cardiac arrest in females of the reproductive age. Eur Heart J 2010;31:1365–72, http://dx.doi.org/10.1093/eurheartj/ehq059. 10. Teodorescu C, Reinier K, Uy-Evanado A, Ayala J, Mariani R, Wittwer L, et al. Survival advantage from ventricular fibrillation and pulseless electrical activity in women compared to men: the oregon sudden unexpected death study. J Interv Card Electrophysiol 2012;34:219–25. 11. Arrich J, Sterz F, Fleischhackl R, et al. Gender modifies the influence of age on outcome after successfully resuscitated cardiac arrest: a retrospective cohort study. Medicine (Baltimore) 2006;85:288–94, http://dx.doi.org/ 10.1097/01.md.0000236954.72342.20. 12. Waldo SW, Armstrong EJ, Kulkarni A, et al. Comparison of clinical characteristics and outcomes of cardiac arrest survivors having versus not having coronary angiography. Am J Cardiol 2013;111:1253–8, http://dx.doi.org/ 10.1016/j.amjcard.2013.01.267. 13. Alfredsson J, Stenestrand U, Wallentin L, Swahn E. Gender differences in management and outcome in non-ST-elevation acute coronary syndrome. Heart 2007;93:1357–62, http://dx.doi.org/10.1136/hrt.2006.102012. 14. Radovanovic D, Erne P, Urban P, Bertel O, Rickli H, Gaspoz J-M. Gender differences in management and outcomes in patients with acute coronary syndromes: results on 20,290 patients from the AMIS Plus Registry. Heart 2007;93:1369–75, http://dx.doi.org/10.1136/hrt.2006.106781. 15. Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature management at 33 ◦ C versus 36 ◦ C after cardiac arrest. N Engl J Med 2013;369:2197–206. 16. Lilja G, Nielsen N, Friberg H, et al. Cognitive function in survivors of outof-hospital cardiac arrest after target temperature management at 33 ◦ C versus 36 ◦ C. Circulation 2015, http://dx.doi.org/10.1161/CIRCULATIONAHA. 114.014414. 17. Nielsen N, Hovdenes J, Bishop G, et al. Target temperature management after out-of-hospital cardiac arrest—a randomized, parallel-group, assessor-blinded clinical trial—rationale and design. Am Heart J 2012;163:541–8. 18. Nielsen N, Wetterslev J, Cronberg T, et al. Supplement to: targeted temperature management at 33 ◦ C versus 36 ◦ C after cardiac arrest. N Engl J Med 2013;369:2197–206, http://dx.doi.org/10.1056/NEJMoa1310519. 19. Langhelle A, Nolan J, Herlitz J, et al. Recommended guidelines for reviewing, reporting, and conducting research on post-resuscitation care: the Utstein style. Resuscitation 2005;66:271–83. 20. Charlson ME, Pompei P, Ales KLMC. A new method of classifying prognostic in longitudinal studies: development. J Chronic Dis 1987;40:373–83. 21. R Core Team. R foundation for statistical computing. R.3.01. Vienna, Austria: R Core Team; 2013 http://www.R–project.org/
84
M. Winther-Jensen et al. / Resuscitation 96 (2015) 78–84
22. Folke F, Gislason GH, Lippert FK, et al. Differences between out-of-hospital cardiac arrest in residential and public locations and implications for public-access defibrillation. Circulation 2010;122:623–30, http://dx.doi.org/ 10.1161/CIRCULATIONAHA.109.924423. 23. Kapral MK, Fang J, Hill MD, et al. Sex differences in stroke care and outcomes: results from the registry of the Canadian Stroke Network. Stroke 2005;36:809–14, http://dx.doi.org/10.1161/01.STR.0000157662.09551.e5. 24. Reeves MJ, Bushnell CD, Howard G, et al. Sex differences in stroke: epidemiology, clinical presentation, medical care, and outcomes. Lancet Neurol 2008;7:915–26, http://dx.doi.org/10.1016/S1474-4422(08)70193-5. 25. Hsu JW, Madsen CD, Callaham ML. Quality-of-life and formal functional testing of survivors of out-of-hospital cardiac arrest correlates poorly with traditional neurologic outcome scales. Ann Emerg Med 1996;28:597–605.
26. Rittenberger JC, Raina K, Holm MB, Kim YJ, Callaway CW. Association between cerebral performance category, modified Rankin Scale, and discharge disposition after cardiac arrest. Resuscitation 2011;82:1036–40, http://dx.doi.org/10.1016/j.resuscitation.2011.03.034. 27. Raina KD, Callaway C, Rittenberger JC, Holm MB. Neurological and functional status following cardiac arrest: method and tool utility. Resuscitation 2008;79:249–56. 28. Johnson MA, Haukoos JS, Larabee TM, et al. Females of childbearing age have a survival benefit after out-of-hospital cardiac arrest. Resuscitation 2012;84:639–44, http://dx.doi.org/10.1016/j.resuscitation.2012.09.011. 29. Topjian Aa, Localio AR, Berg Ra, et al. Women of child-bearing age have better in hospital cardiac arrest survival outcomes than do equal-aged men. Crit Care Med 2010;38:1254–60, http://dx.doi.org/10.1097/CCM.0b013e3181d8ca43.