Yuanyuan Tang

Association Between Therapeutic Hypothermia and Survival After In-Hospital Cardiac Arrest

Importance Therapeutic hypothermia is used for patients following both out-of-hospital and in-hospital cardiac arrest. However, randomized trials on its efficacy for the in-hospital setting do not exist, and comparative effectiveness data are limited. Objective To evaluate the association between therapeutic hypothermia and survival after in-hospital cardiac arrest. Design, Setting, and Patients In this cohort study, within the national Get With the Guidelines-Resuscitation registry, 26 183 patients successfully resuscitated from an in-hospital cardiac arrest between March 1, 2002, and December 31, 2014, and either treated or not treated with hypothermia at 355 US hospitals were identified. Follow-up ended February 4, 2015. Exposure Induction of therapeutic hypothermia. Main Outcomes and Measures The primary outcome was survival to hospital discharge. The secondary outcome was favorable neurological survival, defined as a Cerebral Performance Category score of 1 or 2 (ie, without severe neurological disability). Comparisons were performed using a matched propensity score analysis and examined for all cardiac arrests and separately for nonshockable (asystole and pulseless electrical activity) and shockable (ventricular fibrillation and pulseless ventricular tachycardia) cardiac arrests. Results Overall, 1568 of 26 183 patients with in-hospital cardiac arrest (6.0%) were treated with therapeutic hypothermia; 1524 of these patients (mean [SD] age, 61.6 [16.2] years; 58.5% male) were matched by propensity score to 3714 non-hypothermia-treated patients (mean [SD] age, 62.2 [17.5] years; 57.1% male). After adjustment, therapeutic hypothermia was associated with lower in-hospital survival (27.4% vs 29.2%; relative risk [RR], 0.88 [95% CI, 0.80 to 0.97]; risk difference, -3.6% [95% CI, -6.3% to -0.9%]; P = .01), and this association was similar (interaction P = .74) for nonshockable cardiac arrest rhythms (22.2% vs 24.5%; RR, 0.87 [95% CI, 0.76 to 0.99]; risk difference, -3.2% [95% CI, -6.2% to -0.3%]) and shockable cardiac arrest rhythms (41.3% vs 44.1%; RR, 0.90 [95% CI, 0.77 to 1.05]; risk difference, -4.6% [95% CI, -10.9% to 1.7%]). Therapeutic hypothermia was also associated with lower rates of favorable neurological survival for the overall cohort (hypothermia-treated group, 17.0% [246 of 1443 patients]; non-hypothermia-treated group, 20.5% [725 of 3529 patients]; RR, 0.79 [95% CI, 0.69 to 0.90]; risk difference, -4.4% [95% CI, -6.8% to -2.0%]; P < .001) and for both rhythm types (interaction P = .88). Conclusions and Relevance Among patients with in-hospital cardiac arrest, use of therapeutic hypothermia compared with usual care was associated with a lower likelihood of survival to hospital discharge and a lower likelihood of favorable neurological survival. These observational findings warrant a randomized clinical trial to assess efficacy of therapeutic hypothermia for in-hospital cardiac arrest.

Personalizing the Intensity of Blood Pressure Control: Modeling the Heterogeneity of Risks and Benefits From SPRINT (Systolic Blood Pressure Intervention Trial)

Background— In SPRINT (Systolic Blood Pressure Intervention Trial), patients with hypertension and high cardiovascular risk treated with intensive blood pressure (BP) control (<120 mm Hg) had fewer major adverse cardiovascular events (MACE) and deaths but higher rates of treatment-related serious adverse events (SAE) than patients randomized to standard BP control (<140 mm Hg). However, the degree of benefit or harm for an individual patient could vary because of heterogeneity in treatment effect. Methods and Results— Using patient-level data from 9361 randomized patients in SPRINT, we developed models to predict risk for MACE or death and treatment-related SAE to allow for individualized BP treatment goals based on each patient’s projected risk and benefit of intensive versus standard BP control. Models were internally validated using bootstrap resampling and externally validated on 4741 patients from the ACCORD-BP (The Action to Control Cardiovascular Risk in Diabetes blood pressure) trial. Among 9361 SPRINT patients, 755 patients (8.1%) had a MACE or death event and 338 patients (3.6%) had a treatment-related SAE during a median follow-up of 3.3 years. The MACE/death and the SAE model had C statistics of 0.72 and 0.70, respectively, in the derivation cohort and 0.69 and 0.65 in ACCORD. The MACE/death model had 10 variables including treatment interactions with age, baseline systolic BP, and diastolic BP, and the SAE model had 8 variables including treatment interaction with number of BP medications. Intensive BP treatment was associated with a mean 2.2±2.6% lower risk of MACE/death compared with standard treatment (range, 20.7% lower risk to 19.6% greater risk among individual patients) and a mean 2.2±1.2% higher risk for SAEs (range, 0.5%–15.8% more harm in individual patients). Conclusions— To translate the findings from SPRINT to clinical practice, we developed prediction models to tailor the intensity of BP control based on the projected risk and benefit for each unique patient. This approach should be prospectively tested to better engage patients in shared medical decision making and to improve outcomes. Clinical Trial Registration— URL: https://clinicaltrials.gov. Unique identifier: NCT01206062.

Constraint approaches to the estimation of relative risk

In medical and epidemiologic studies, relative risk is usually the parameter of interest. However, calculating relative risk using standard log-Binomial regression approach often encounters non-convergence. A modified Poisson regression, which uses robust variance, was proposed by Zou in 2004. Although the modified Poisson regression with sandwich variance estimator is valid for the estimation of relative risk, the predicted probability of the outcome may be greater than the natural boundary 1 for the unobserved but plausible covariate combinations. Moreover, the lower and upper bounds of confidence intervals for predicted probabilities could fall out of (0, 1). Chu and Cole, in 2010, proposed a Bayesian approach to overcome this issue. Posterior median was used to get the parameter estimation. However, the Bayesian approach may provide biased estimation, especially when the probability of outcome is high. In this article, we propose an alternative constraint optimization approach for estimating relative risk. Our approach can reach similar or better performance than Bayesian approach in terms of bias, root mean square error, coverage rate, and predictive probabilities. Simulation studies are conducted to demonstrate the usefulness of this approach. Our method is also illustrated by Prospective Registry Evaluating Myocardial Infarction: Event and Recovery data.

Association of Serial Kansas City Cardiomyopathy Questionnaire Assessments With Death and Hospitalization in Patients With Heart Failure With Preserved and Reduced Ejection Fraction: A Secondary Analysis of 2 Randomized Clinical Trials

Importance While there is increasing emphasis on incorporating patient-reported outcome measures in routine care for patients with heart failure (HF), how best to interpret longitudinally collected patient-reported outcome measures is unknown. Objective To examine the strength of association between prior, current, or a change in Kansas City Cardiomyopathy Questionnaire (KCCQ) scores with death and hospitalization in patients with HF with preserved (HFpEF) and reduced (HFrEF) ejection fractions. Design, Setting, and Participants Secondary analyses of the Treatment of Preserved Cardiac Function Heart Failure With an Aldosterone Antagonist (TOPCAT) trial of 1372 patients with HFpEF, conducted between August 2006 and January 2012, and the HF-ACTION trial that included 1669 patients with HFrEF, conducted between April 2003 and February 2007. Exposures Prior, current, and change in KCCQ Overall Summary scores (KCCQ-os) in 5-point increments (higher scores indicate better health status). Main Outcomes and Measures Time to cardiovascular death/first HF hospitalization (primary outcome) and all-cause death (secondary outcome). Results Of 1767 eligible TOPCAT participants, 882 were women (49.9%), and the mean (SD) age was 71.5 (9.7) years. Of 2130 eligible HF-ACTION participants, 599 were women (28.1%), and the mean age was 58.6 (12.7) years. Each 5-point difference in prior or current KCCQ-os scores was associated with a 6% (95% CI, 4%-8%; P < .001) to 9% (95% CI, 7%-11%; P < .001) lower risk for subsequent cardiovascular death/first HF hospitalization in patients with HFpEF and 6% (95% CI, 4%-9%; P < .001) to 8% (95% CI, 5%-10%; P < .001) lower risk for subsequent cardiovascular death/first HF hospitalization in patients with HRpEF and HFrEF in unadjusted analyses. Results were similar for change in KCCQ-os. In models with the prior and current KCCQ-os, only the current KCCQ-os was significantly associated with 10% (95% CI, 7%-12%; P < .001) and 7% (95% CI, 3%-11%; P < .001) lower risk for subsequent cardiovascular death/first HF hospitalization in patients with HFpEF and HFrEF, respectively. Similar results were observed when the current and &Dgr; KCCQ-os were considered together, when adjusted for important patient and treatment characteristics, when including 3 sequential KCCQ-os scores, and when examining all-cause death as the outcome. Conclusions and Relevance In serial health status evaluations of patients with HF, the most recent KCCQ score was most strongly associated with subsequent death and cardiovascular hospitalization in HFpEF and HFrEF. Measuring serial patient-reported outcome measures in the clinical care of patients with HF can provide an updated assessment of prognosis. Trial Registration clinicaltrials.gov Identifier: NCT00094302 (TOPCAT) and NCT00047437 (HF-ACTION)

Association Between Prompt Defibrillation and Epinephrine Treatment With Long-Term Survival After In-Hospital Cardiac Arrest

Background: Prior studies have reported higher in-hospital survival with prompt defibrillation and epinephrine treatment in patients with in-hospital cardiac arrest (IHCA). Whether this survival benefit persists after discharge is unknown. Methods: We linked data from a national IHCA registry with Medicare files and identified 36 961 patients ≥65 years of age with an IHCA at 517 hospitals between 2000 and 2011. Patients with IHCA caused by pulseless ventricular tachycardia or ventricular fibrillation were stratified by prompt (⩽2 minutes) versus delayed (>2 minutes) defibrillation, whereas patients with IHCA caused by asystole or pulseless electric activity were stratified by prompt (⩽5 minutes) versus delayed (>5 minutes) epinephrine treatment. The association between prompt treatment and long-term survival for each rhythm type was assessed with multivariable hierarchical modified Poisson regression models. Results: Of 8119 patients with an IHCA caused by ventricular tachycardia or ventricular fibrillation, the rate of 1-year survival was higher in those treated with prompt defibrillation than with delayed defibrillation (25.7% [1466 of 5714] versus 15.5% [373 of 2405]; adjusted relative risk [RR], 1.49; 95% confidence interval [CI] 1.32–1.69; P<0.0001). This survival advantage persisted at 3 years (19.1% versus 11.0%; adjusted RR, 1.45; 95% CI, 1.23–1.69; P<0.0001) and at 5 years (14.7% versus 7.9%; adjusted RR, 1.50; 95% CI, 1.22–1.83; P<0.0001). Of 28 842 patients with an IHCA caused by asystole/pulseless electric activity, the rate of 1-year survival with prompt epinephrine treatment was higher than with delayed treatment (5.4% [1341 of 24 885] versus 4.3% [168 of 3957]; adjusted RR, 1.20; 95% CI, 1.02–1.41; P=0.02), but this survival benefit was no longer present at 3 years (3.5% versus 2.9%; adjusted RR, 1.17; 95% CI, 0.95–1.45; P=0.15) and at 5 years (2.3% versus 1.9%; adjusted RR, 1.18; 95% CI, 0.88–1.58; P=0.27). Conclusions: Prompt defibrillation for IHCA caused by ventricular tachycardia or ventricular fibrillation was associated with higher rates of long-term survival throughout 5 years of follow-up, whereas prompt epinephrine treatment for asystole/pulseless electric activity was associated with greater survival at 1 year but not at 3 or 5 years. By quantifying the greater survival associated with timely defibrillation and epinephrine administration, these findings provide important insights into the durability of survival benefits for 2 process-of-care measures in current resuscitation guidelines.

Implications of a recurrent in-hospital cardiac arrest on survival and neurological outcomes

Background Despite the high incidence of in‐hospital cardiac arrest (IHCA) in US hospitals, the prognosis and end‐of‐life decision‐making patterns of a patient with a recurrent IHCA are unknown. Methods Within Get‐With‐The‐Guidelines‐Resuscitation, we identified 192,250 patients from 711 hospitals with an IHCA from 2000 to 2015. Patients were categorized as having no recurrent IHCA (only 1 event), recurrent IHCA (≥2 IHCAs), and recurrent out‐of‐hospital cardiac arrest (OHCA), defined as an IHCA after an OHCA. Using multivariable hierarchical logistic regression, rates of survival to discharge and favorable neurological survival (mild or no disability) between the 3 groups were compared. Rates of de novo “do not attempt resuscitation” (DNAR) and withdrawal of care orders among successfully resuscitated patients were also evaluated. Results Overall, 165,446 (86.1%) had no recurrent IHCA, 23,643 (12.3%) had recurrent IHCA, and 3162 (1.6%) had recurrent OHCA. Compared with patients with no recurrent IHCA, patients with recurrent IHCA were less than half as likely to survive to discharge (12.7% vs 22.1%; adjusted OR: 0.46 [0.44‐0.48], P < .001) and have favorable neurological survival (7.0% vs 13.1%; adjusted OR: 0.44 [0.42‐0.47], P < .001). Compared with patients with recurrent OHCA, patients with recurrent IHCA also had lower rates of survival to discharge (12.7% vs 16.1%; adjusted OR: 0.81 [0.71‐0.94], P = .005) and favorable neurological survival (7.0% vs 8.9%; adjusted OR: 0.66 [0.54‐0.81], P < .001). Despite worse survival outcomes, patients with recurrent IHCA were least likely to adopt DNAR orders within the first 24 hours after successful resuscitation compared with patients with no recurrent IHCA or recurrent OHCA (17.2% vs 18.9% and 26.6%, respectively) or withdraw care at any time (17.7% vs 24.4% and 31.2%, respectively). Conclusions Nearly 1 in 8 patients with an IHCA has a recurrent IHCA, and these patients have worse outcomes than patients with only a single IHCA and those with an IHCA after being hospitalized for an OHCA. Despite worse survival, rates of DNAR and withdrawal of care were lowest for patients with recurrent IHCA. These findings provide important prognostic information for clinicians caring for patients with recurrent IHCA and suggest the need to better align resuscitation and end‐of‐life decisions with patients’ prognoses after IHCA.

RELIABILITY AND RESPONSIVENESS OF SEATTLE ANGINA QUESTIONNAIRE IN WOMEN WITH CORONARY ARTERY DISEASE

Background: Although the Seattle Angina Questionnaire (SAQ) has been widely used to assess disease-specific health status in patients with CAD, the original psychometric work was done in a VA cohort. As such, scarce data exist on the reliability and responsiveness of SAQ in women with CAD. Methods

Bayesian adjustment for unidirectional misclassification in ordinal covariates

ABSTRACT In this paper, we study the identification of Bayesian regression models, when an ordinal covariate is subject to unidirectional misclassification. Xia and Gustafson [Bayesian regression models adjusting for unidirectional covariate misclassification. Can J Stat. 2016;44(2):198–218] obtained model identifiability for non-binary regression models, when there is a binary covariate subject to unidirectional misclassification. In the current paper, we establish the moment identifiability of regression models for misclassified ordinal covariates with more than two categories, based on forms of observable moments. Computational studies are conducted that confirm the theoretical results. We apply the method to two datasets, one from the Medical Expenditure Panel Survey (MEPS), and the other from Translational Research Investigating Underlying Disparities in Acute Myocardial infarction Patients Health Status (TRIUMPH).

Precision Medicine for Cardiac Resynchronization: Predicting Quality of Life Benefits for Individual Patients—An Analysis From 5 Clinical Trials

Background: Clinical trials have established the average benefit of cardiac resynchronization therapy (CRT), but estimating benefit for individual patients remains difficult because of the heterogeneity in treatment response. Accordingly, we created a multivariable model to predict changes in quality of life (QoL) with and without CRT. Methods and Results: Patient-level data from 5 randomized trials comparing CRT with no CRT were used to create a prediction model of change in QoL at 3 months using a partial proportional odds model for no change, small, moderate, and large improvement, or deterioration of any magnitude. The C statistics for not worsening or obtaining at least a small, moderate, and large improvement were calculated. Among the 3614 patients, regardless of assigned treatment, 33.3% had a deterioration in QoL, 9.2% had no change, 9.2% had a small improvement, 13.5% had a moderate improvement, and the remaining 34.9% had a large improvement. Patients undergoing CRT were less likely to have a decrement in their QoL (28.2% versus 38.9%; P<0.001) and more likely to have a large QoL improvement (38.7% versus 30.6%; P<0.001). A partial proportional odds model identified baseline QoL, age, and an interaction of CRT with QRS duration as predictors of QoL benefits 3 months after randomization. C statistics of 0.65 for not worsening, 0.68 for at least a small improvement, 0.69 for at least a moderate improvement, and 0.73 for predicting a large improvement were observed. Conclusions: There is marked heterogeneity of treatment benefit of CRT that can be predicted based on baseline QoL, age, and QRS duration.

Noncardiac chest pain after acute myocardial infarction: Frequency and association with health status outcomes

Background The frequency of noncardiac chest pain (CP) hospitalization after acute myocardial infarction (AMI) is unknown, and its significance from patients’ perspectives is not studied. Objectives To assess the frequency of noncardiac CP admissions after AMI and its association with patients’ self‐reported health status. Methods We identified cardiac and noncardiac CP hospitalizations in the year after AMI from the 24‐center TRIUMPH registry. Hierarchical repeated‐measures regression was used to identify the association of these hospitalizations with patients’ self‐reported health status using the Seattle Angina Questionnaire Quality of Life domain (SAQ QoL) and Short Form 12 (SF‐12) physical (PCS) and mental (MCS) component summary scores. Results Of 3,099 patients, 318 (10.3%) were hospitalized with CP, of whom 92 (28.9%) were hospitalized for noncardiac CP. Compared with patients not hospitalized with CP, noncardiac CP hospitalization was associated with poorer health status (SAQ QoL–adjusted differences: −8.9 points [95% CI −12.1 to −5.6]; SF‐12 PCS: −2.5 points [95% CI −4.2 to −0.8] and SF‐12 MCS: −3.5 points [95% CI −5.1 to −1.9]). The SAQ QoL for patients hospitalized with noncardiac CP was similar to patients hospitalized with cardiac CP (adjusted difference: 0.6 points [95% CI −3.2 to 4.5]; SF‐12 PCS (0.9 points [95% CI −1.1 to 2.9]), but was worse with regard to SF‐12 MCS (adjusted difference: −2.0 points [95% CI −3.9 to −0.2]). Conclusions Noncardiac CP accounted for a third of CP hospitalizations within 1 year of AMI and was associated with similar disease‐specific QoL as well as general physical and mental health status impairment compared with cardiac CP hospitalization.