Hu, Chengcheng

Background: Bystander CPR (BCPR) and telephone CPR (TCPR) are associated with improved OHCA outcomes. It has been shown that BCPR increases the proportion of patients with an initial shockable cardiac rhythm (VF/VT) when encountered by EMS. It is unknown whether TCPR does the same. Objective: To assess whether TCPR is independently associated with an increase in initial shockable cardiac rhythms during OHCA. Methods: Data from 9-1-1 audio recordings, first care EMS reports and hospital records were linked for OHCAs of presumed cardiac origin (1/2011-12/2014). Three cohorts were analyzed: TCPR, BCPR or no CPR. Using no CPR as the reference group, we assessed whether TCPR and BCPR were independently associated with initial shockable rhythms in a logistic regression model controlling for gender, event location, witness status and EMS response interval. Results: After exclusions, 2715 adult OHCA events with linked outcome data were analyzed (median age: 63; male: 66.8%; 33.3% witnessed arrest). Median response interval was 5 min. The BCPR rate (lay rescuer CPR without telephone instructions) was 24.8% and survival in this group was 15.4%. The TCPR rate (lay rescuer CPR with telephone instructions) was 40.2% and survival in this group was 11.7%. For the no-CPR group, 20.5% had an initial shockable rhythm. For the BCPR group, 34.6 % had an initial shockable rhythm. For the TCPR group, 23.2% had an initial shockable rhythm. Provision of BCPR and of TCPR were each independently associated with shockable rhythms [aOR (BCPR) = 1.7, 95% CI: 1.33-2.17, p=0.0001; aOR (TCPR) = 1.3, 95% CI: 1.04-1.63, p=0.0001]. Conclusion: The provision of both BCPR and TCPR were independently associated with an increased rate of initial shockable cardiac rhythms and an increase in survival to hospital discharge compared to no CPR.

Introduction: Coronary angiography (CA) is recommended after out-of-hospital cardiac arrest (OHCA) when ST-elevation is present but the ideal timing and benefit of CA when ST-elevation is not present are unknown. Methods: Prospective observational study of adult (age >17) OHCA of presumed cardiac etiology from 1/01/2010-12/31/2014 who were admitted to one of 40 recognized cardiac receiving centers within a state resuscitation center network. DNR patients, those who did not achieve ROSC, and/or those who died in the ED were excluded since the likelihood of percutaneous coronary intervention (PCI) impacting outcome in these patients is low. An in-hospital database was linked with a statewide OHCA database. Results: Among 6,615 cases, 1,247 remained for analysis after exclusion [DNR (1254); failed to achieve ROSC/died in ED (4,020); missing prehospital variable (92); and missing outcome (2)]. Median age was 63 and 70% were male. There were 62% witnessed arrests, 54% received bystander CPR and the initial rhythm was VF/VT in 55%. Sixty-four percent survived to hospital discharge, 82% of which had a CPC score of 1 or 2. Of the 326 STEMI cases, 298 (91%) underwent CA with resultant PCI in 206 (63%). Of the 849 non-STEMI cases, 361 (43%) underwent CA with resultant PCI in 107 (13%). Overall survival was: 54% in cases without CA; 73% in cases with CA but without PCI; and 73% in those with PCI. Survival in non-STEMI cases was: 55% in cases without CA; 82% in cases with CA but without PCI; and 80% in those with PCI. In non-STEMI cases, the adjusted odds of survival for CA without PCI vs no CA was 2.70 (95% CI: 1.81, 4.02) and for PCI versus no CA it was 2.01 (95% CI: 1.16, 3.48). Conclusion: In this statewide analysis over 90% of OHCA cases with STEMI underwent CA. In cases that were not STEMI, CA was strongly and independently associated with survival regardless of whether PCI was performed. Since there is no reason to believe that CA without PCI confers a survival benefit, these findings suggest a component of selection bias in those without STEMI who undergo CA.

Background: Quantitative End-Tidal CO2 (PETCO2) measurement during cardiac arrest is recommended for monitoring the CPR quality to detect ROSC. While low ETC02 can reflect suboptimal CPR quality and increases in ETC02 are often associated with ROSC the significance of high values of PETCO2 without ROSC remain unclear. Objectives: To describe the population of OHCA victims with high PETCO2 (>50 mm Hg) during ongoing resuscitation including demographics, rhythm characteristics, occurrence of ROSC, the later occurrence of field termination of resuscitation and the presumed etiology. Methods: An observational analysis of adult OHCAs with available PETCO2 data occurring between Oct 2010 and Nov 2014 at two sites involved in the Save Hearts in Arizona Registry and Education (SHARE) quality improvement program. CPR and PETCO2 data were recorded using ZOLL E and X Series EMS monitors. PETCO2 is reported as the mean value for each minute. The cardiac rhythm and the presence of ROSC were determined for each minute using EMS provider reports, vital signs, ECG, and accelerometer data. The first recorded minute of PETCO2 and minutes with ROSC were excluded. Results: Study population included 654 OHCAs with PETCO2 data, 96 cases (58% male, age median 65 (IQR 54-75)) with at least 1 min of high PETCO2 in the absence of ROSC were identified. Of these, 31 (32%) had ROSC at least once during their resuscitation. Cases were of presumed cardiac etiology in 81 (84%), while 12% and 4% had presumed respiratory and other etiologies, respectively. A total of 322 minutes of PETCO2 >50 mm Hg in the absence of ROSC were identified. The mean PETC02 value was 67 (IQR 54-75). Of these 296 minutes (92%) occurred during PEA. Despite high PETCO2, resuscitation was terminated in the field in 21 (22%) patients. Only one subject survived to hospital discharge. Conclusions: In this observational study, high PETCO2 was frequently found during resuscitation without ROSC and PEA was the most common cardiac rhythm. Further study is warranted to determine the significance and possible reasons for this finding.

Background: Previous studies have described modest correlation between end-tidal CO 2 (ETCO 2 ) and CPR quality during resuscitation of cardiac arrest patients, but it is unclear whether ETCO 2 alone can indicate CPR quality. The present study investigated whether ETCO 2 adequately identifies the quality of CPR provided during out-of-hospital cardiac resuscitation. Methods: ETCO 2 was monitored with side-stream CO 2 (Philips/Respironics) and CPR quality measured with an accelerometer-based system (E Series, ZOLL Medical) during the treatment of consecutive adult OHCA patients with presumed cardiac etiology by 2 EMS agencies in the Arizona SHARE QI Program between 10/08-06/13. Minute-by-minute ETCO 2 and CPR quality were extracted. ETCO 2 values were log transformation to achieve approximate normality. Linear mixed effect models were fitted to use (transformed) ETCO 2 level to predict four CPR variables: chest compression (CC) depth, CC rate, CC release velocity (CCRV), and ventilation rate (VR). A random intercept for each case was included and a spatial power covariance structure assumed for measurements over time. Results: 230 subjects (median age 69 yrs, 69% male) with 1581 minutes of data were studied. Transformed ETCO 2 was significant for CC depth (p< 0.0001), CCRV (p=0.003) and VR (p<0.0001), but only explained 3.7%, 2.7%, and 10.0% of the total variance for these variables, respectively. Transformed ETCO 2 was not a significant predictor for CC rate (p=0.89). The Figure illustrates the overlap in CC depth over quartiles of ETCO 2 , demonstrating that any specific ETCO 2 level could be found over a wide range of CC depths. Conclusion: In this secondary analysis, ETCO 2 was not an independent indicator of CC rate but was a weak predictor for CC depth, CCRV and VR. These findings suggest that ETCO 2 may be not be an adequate substitute for CPR quality monitoring. Future studies should investigate how ETCO 2 and CPR quality monitoring can be used in conjunction to optimize CPR.

Background: Higher chest compression release velocity (CCRV) is associated with better outcomes after OHCA. Patient age and gender have been associated with variations in chest wall compliance and compressibility which may impact outcomes. Hypothesis: To evaluate the association between CCRV and OHCA patient age, weight, and gender. Methods: Observational study of prospectively collected OHCA quality improvement data in two suburban EMS agencies in Arizona between 10/1/2008 and 12/31/2016. Patient age and weight were categorized into quartiles for the analysis. CCRV was summarized at each level of the categorical variable by median and inter-quartile range (IQR), and was then compared between different levels by the Kruskal-Wallis test. Results: During the study period, 2,661 OHCA cases were treated. After exclusion criteria, 1,140 cases remained for analysis. Median duration of compressions was 8.70 minutes. Mean CCRV was negatively correlated with age and positively correlated with weight (Table). Male patients exhibited a greater mean CCRV compared to female patients [344.4 mm/s (IQR 307.3-384.6) vs. 331.5 (285.3-385.5), p=0.0133]. Conclusion: Patient characteristics including younger age, male gender, and increased weight were associated with a higher CCRV during OHCA resuscitation.

Background: Extended care facilities (ECFs) house unique patient populations with increased risk of out of hospital cardiac arrest (OHCA) and with unique, generally medically trained first responders initiating care prior to arrival of emergency medical services (EMS). However, little is known about bystander cardiopulmonary resuscitation (CPR) and resuscitation outcomes in this population. Our aim is to assess rate of CPR delivery prior to EMS arrival in cases of OHCA among residents ECFs across the state of Arizona. Methods: Cases of OHCA between 2016-2018, cared for by participating Arizona EMS agencies, were included if OHCA occurred at an ECF, age 18+, and was of presumed cardiac etiology. Association between survival status at discharge and bystander CPR was examined first by Chi-squared test then by logistic regression, which adjusted for important risk factors and potential confounders including: age, gender, witnessed arrest, shockable initial rhythm, and return of spontaneous circulation (ROSC) in EMS care. Results: There were 1,046 cases included in this study after excluding 32 cases with DNR and 11 with unknown bystander CPR status. Median patient age was 74 (interquartile range 64, 83). The overall rate of CPR prior to EMS arrival was 78.4% (95% CI, 75.8%, 80.9%). Survival to hospital discharge and survival with favorable neurologic outcome were not different between nursing home patients who received CPR prior to EMS arrival and those who did not [6.2% vs. 4.4% (p = 0.34) and 3.1% vs. 1.8% (p = 0.29)]. There was no statistically significant difference in the adjusted analysis. Conclusions: In this study population CPR was commonly delivered prior to EMS arrival. However, delivery of CPR prior to EMS arrival was not associated with survival to hospital discharge or favorable neurologic outcome. The lack of positive association between CPR prior to arrival and survival to hospital discharge may be related to comorbid factors present in ECF patients or differences in the time between collapse and initiation of CPR. Further study is needed to identify factors that might improve survival in this unique population.

Abstract Background: Germline cancer testing panels provide an effective method for identifying individuals at increased risk for breast cancer. However, estimates of risk for triple negative breast cancer (TNBC) (estrogen receptor-negative, progesterone receptor-negative, HER2-negative) associated with pathogenic mutations in panel genes have not been established. We sought to define the genes that contribute to TNBC. Methods:Germline hereditary cancer multigene panel testing results were obtained for 8,753 TNBCs evaluated by a clinical testing laboratory. Associations between pathogenic mutations in individual genes and TNBC were assessed by comparing mutation frequencies in TNBCs and in the Exome Aggregation Consortium, non-Finn European, non-Cancer Genome Atlas reference controls. Results: Inactivating mutations in 21 known cancer predisposition genes were identified in 14.6% of TNBCs. BRCA1, BRCA2, PALB2, BARD1, and RAD51D alterations were associated with high risks (odds ratio(OR)&gt;5.0) of TNBC and variants in BRIP1, RAD51C, MSH6, and TP53 were associated with moderate risks (OR&gt;2). In contrast, ATM, CHEK2, NBN, and RAD50 yielded no clinically relevant risks of TNBC. Pathogenic mutations in these established non-BRCA1/2 TNBC susceptibility genes were detected in 6.3% of TNBCs. Similar trends were observed among African American TNBCs. Overall, 5.5% of TNBCs with pathogenic mutations did not meet NCCN clinical testing criteria for BRCA1/2 due to a lack of significant family history and diagnosis over the age of 60. Conclusions: The identification of genes associated with elevated risk of TNBC will improve understanding of the etiology of this aggressive form of breast cancer and inform risk management of individuals receiving panel testing. The high frequency of pathogenic variants suggests that all patients with TNBC, regardless of age of diagnosis or family history of cancer, should be considered for multigene panel testing. Citation Format: Couch FJ, Shimelis H, LaDuca H, Hu C, Hart SN, Polley EC, Pesaran T, Tippin-Davis B, Goldgar DE, Dolinsky JS. Triple negative breast cancer predisposition genes [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr PD1-01.

Abstract Background: Clinical panel testing has become routine practice for patients that are diagnosed with breast cancer at a young age and/or have a personal or family history of cancer. Associations with known breast cancer genes and breast cancer subtypes have been previously identified, such as BRCA1 associations with estrogen receptor negative (ER-) and triple negative (ER-/PR-/HER2-) breast cancers. However, the cancer predisposition genes associated with each of the four clinical subtypes of breast cancer have not been fully defined. We evaluated 24,901 Caucasian female breast cancer cases receiving clinical panel testing for 23 cancer predisposition genes and assessed associations between mutations in each gene and breast cancer subtypes. Methods: Germline hereditary cancer multigene panel testing results for cancer predisposition genes were obtained for 24,901 Caucasian female breast cancer cases evaluated by a clinical testing laboratory. Information on tumor histology, personal and family history of cancer, age at diagnosis, and previous genetic testing was provided by clinical care providers of patients receiving clinical cancer genetic testing. Breast cancer cases were classified into clinical breast cancer subtypes based on estrogen/progesterone hormone receptor status (HR) and HER2 status: Luminal A (HR+/HER2-), Luminal B (HR+/HER2-), HER2 subtype (HR-/HER2+), and Triple Negative (HR-/HER2-). The frequency of pathogenic or likely pathogenic mutations observed in each subtype was compared against the Exome Aggregation Consortium (ExAC) non-TCGA non-Finnish European population to estimate risks. Results: ATM was associated with moderate risks (odds ratio (OR)&gt;2.0) of Luminal A, Luminal B, and HER2 subtypes of breast cancer, but was not associated with the Triple Negative subtype. PALB2 was associated with moderate risk for Luminal B subtype, but high risk (OR&gt;5.0) for Luminal A, HER2, and triple negative subtypes. TP53 was associated with high risks for Luminal B and HER2 tumors. NBN, MRE11A, and RAD50 were not associated with any subtype of breast cancer. Conclusions: Identifying associations between inherited mutations (odds ratio (OR)&gt;2.0) and breast cancer subtypes can inform clinical risk management, treatment options, and therapeutic development efforts. Citation Format: Lilyquist J, Laduca H, Hu C, Na J, Polley EC, Hart SN, Pesaran T, Tippin-Davis B, Goldgar DE, Dolinsky JS, Couch FJ. Associations between hereditary cancer panel predisposition genes and breast cancer histological subtypes [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P6-09-02.