Brain Natriuretic Peptide as a Marker of Adverse Neurological Outcomes Among Survivors of Cardiac Arrest.

BACKGROUND: Neurological prognosis after cardiac arrest remains ill-defined. Plasma brain natriuretic peptide (BNP) may relate to poor neurological prognosis in brain-injury patients, though it has not been well studied in survivors of cardiac arrest. METHODS: We performed a retrospective review and examined the association of BNP with mortality and neurological outcomes at discharge in a cohort of cardiac arrest survivors enrolled from January 2012 to December 2016 at the Wake Forest Baptist Hospital, in North Carolina. Cerebral performance category (CPC) and modified Rankin scales were calculated from the chart based on neurological evaluation performed at the time of discharge. The cohort was subdivided into quartiles based on their BNP levels after which multivariable adjusted logistic regression models were applied to assess for an association between BNP and poor neurological outcomes as defined by a CPC of 3 to 4 and a modified Rankin scale of 4 to 5. RESULTS: Of the 657 patients included in the study, 254 patients survived until discharge. Among these, poor neurological status was observed in 101 (39.8%) patients that had a CPC score of 3 to 4 and 97 patients (38.2%) that had a modified Rankin scale of 4 to 5. Mean BNP levels were higher in patients with poor neurological status compared to those with good neurological status at discharge (P = .03 for CPC 3-4 and P = .02 for modified Rankin score 4-5). BNP levels however, did not vary significantly between patients that survived and those that expired (P = .22). BNP did emerge as a significant discriminator between patients with severe neurological disability at discharge when compared to those without. The area under the curve for BNP predicting a modified Rankin score of 4 to 5 was 0.800 (95% confidence interval [CI] 0.756-0.844, P < .001) and for predicting CPC 3 to 4 was 0.797 (95% CI 0.756-0.838, P < .001). BNP was able to significantly improve the net reclassification index and integrated discriminatory increment (P < .05). BNP was not associated with long-term all-cause mortality (P > .05). CONCLUSIONS: In survivors of either inpatient or out-of-hospital cardiac arrest, increased BNP levels measured at the time of arrest predicted severe neurological disability at discharge. We did not observe an independent association between BNP levels and long-term all-cause mortality. BNP may be a useful biomarker for predicting adverse neurological outcomes in survivors of cardiac arrest.

Characteristics and outcomes of cardiac arrest survivors with acute pulmonary embolism.

INTRODUCTION: The characteristics and outcomes of patients that suffer cardiac arrest due to acute pulmonary embolism (PE) are not well studied. We compared the characteristics and outcomes of cardiac arrest survivors that suffered PE with other forms of cardiac arrest. METHODS: Consecutive cardiac arrest survivors were enrolled that were able to survive for 24 h post cardiopulmonary resuscitation. Diagnosis of PE was confirmed by CT angiogram or high-probability of PE on ventilation perfusion scan after the successful resuscitation from cardiac arrest. Survival curves were examined and predictors of mortality in PE patients were examined in an adjusted Cox proportional hazard model. RESULTS: Among the 996 cardiac arrest patients (mean age 62.6 ±â€¯14.8 years, females 39.4%), 87 (8.7%) patients were found to have acute PE. The mortality rate of cardiac arrest survivors with and without acute PE was not significant different (68.3% vs. 64%). There were no significant differences in mortality among PE patients that received thrombolytics versus those who did not. Out of 87 patients, 33 (37.9%) required transfusion and had a bleeding complication. The risk of mortality in PE patients was predicted by older age, female sex, history of diabetes mellitus, end-stage renal disease and use of targeted temperature management. CONCLUSION: Cardiac arrest survivors with PE did not have significantly better survival than patients with non-PE related cardiac arrest. In addition, use of thrombolytics did not improve survival but these patients ended up requiring transfusion that could have off set the benefit of thrombolytics.

Risk of Mortality Associated With Therapeutic Hypothermia Among Sudden Cardiac Arrest Survivors With Known Heart Failure.

Current guidelines do not inform about use of therapeutic hypothermia among heart failure (HF) patients who suffer from cardiac arrest. We assessed the risk of mortality associated with hypothermia among cardiac arrest survivors with HF. This analysis includes 1,416 comatose patients with cardiac arrest who achieved return of spontaneous circulation on admission and had a left ventricular ejection fraction (LVEF) assessment or HF admission within the previous year. HF was defined as either previous episode of HF or presence of left ventricular ejection fraction <50%. Hazard ratios (HR) and 95% confidence intervals (CI) for association of hypothermia and mortality among patients with and without HF were computed using Cox proportional hazard models adjusted for several risk factors. A propensity score matched analysis was also performed. There were 624 patients (44%) with pre-existing HF and 467 patients (33.0%) received hypothermia. The mortality rate was higher in HF patients treated with hypothermia compared with patients without hypothermia (75.4% vs 53.2%, p <0.0001). Hypothermia was associated with increased mortality among HF patients (HR 1.69; 95% CI 1.27, 2.24, p <0.001) and was not associated with mortality among non-HF patients (HR 1.21; 95% CI 0.93, 1.56, p = 0.15). The association of hypothermia with mortality was higher among HF patients who presented with shockable rhythm compared with nonshockable rhythm (interaction p value = 0.0495). Hypothermia is associated with increased mortality among cardiac arrest survivors with known HF.