250 Advanced management of out of hospital cardiac arrest: a single-centre experience in a tertiary cardiac intensive care unit. (8th December 2021)

• Record Type: Journal Article
• Authors: De Michieli, Laura; Bettella, Alberto; Famoso, Giulia; Babuin, Luciano; Scarpa, Daniele; Tarantini, Giuseppe; Lorenzoni, Giulia; Gregori, Dario; Marra, Martina Perazzolo; Cacciavillani, Luisa
• Is Part Of: European heart journal supplements. Volume 23(2021)Supplement G
• Publication Date: 2021-12-08
• Subjects: Cardiology -- Periodicals; Cardiology -- Europe -- Periodicals; 616.12005
• DOI: 10.1093/eurheartj/suab126
• Languages: English
• ISSNs: 1520-765X

Abstract

Abstract: Aims: Out-of-hospital cardiac arrest (OHCA) affects around 1/1000 person-years. Following return of spontaneous circulation (ROSC), the patient can manifest neurological impairment. A targeted temperature management (TTM) protocol is recommended to prevent hypoxic–ischaemic brain damage in patients with coma after cardiac arrest. Neuro-prognostication remains substantial for the prediction of clinical outcomes. To study clinical characteristics, overall survival, and neurological outcome of patients with Glasgow Coma Scale (GCS) <8 after ROSC following an OHCA of presumed cardiac cause at our Institution. Secondly, to investigate determinants of a negative neurological outcome. Methods: Observational retrospective study evaluating all patients with OHCA of presumed cardiac cause and with GCS < 8 after ROSC treated in an intensive cardiac care unit of a tertiary centre. The study period was from January 2017 to December 2020. Results: One-hundred and five patients out of 107 patients initially selected were included in the study (77% male, mean age 67 years). At 30 days, mortality was 41% and 53% of patients had a poor neurological outcome (Cerebral Performance Category, CPC, 3–5). Sixty-nine patients (66%) underwent TTM. In regard of the circumstances of OHCA, index event in a private place [OR = 3.12 (1.43–7.11), P = 0.005], ineffective rhythm changes during resuscitation manoeuvres [OR = 2.40 (1.05–5.47), P = 0.037] and a greater amount of adrenaline administered during resuscitation [OR = 1.62 (1.27–2.06), P < 0.001] were related to a worse neurological outcome. A history of diabetes mellitus [OR = 3.35 (1.26–8.91), P = 0.015], blood lactates at presentation [OR = 1.33 (1.15—1.53), P < 0.001], neuron-specific enolase (NSE) at presentation [OR = 1.055 (1.022–1.089), P < 0.001] and as peak [OR = 1.034 (1.013–1.054), P < 0.001] were associated with a worse neurological outcome. Among the neurological examinations, the presence of status epilepticus on the EEG [OR = 13.97 (1.73–113.02), P = 0.013] was a predictor of a poor neurological outcome. Treatment with targeted temperature management did not show a significant impact in terms of outcome at univariate analysis [OR = 1.226 (0.547–2.748), P = 0.62]. Two models were developed with multivariate logistic regression for the prediction of neurological outcome. The first one, on a statistical basis, considers pupil reactivity after ROSC, NSE as peak and left ventricular ejection fraction (AUC = 92%). The second model, on a clinical basis, considers age, first blood lactate value and NSE as peak (AUC = 89 %). Finally, the performance of the multiparametric MIRACLE score was tested in our population (AUC 0.81 for neurological outcome at 30 days). Conclusions: In our population, at 30 days after cardiac arrest, survival rate and the rate of good neurological outcome were comparable to those of the major international registries and studies. Even though patients treated with TTM did not demonstrate significant differences in terms of neurological outcome, this might be related to study-sample size and patient selection. Results in the literature are still controversial on this topic. The MIRACLE score showed a good performance, making it suitable for clinical use in our population. Similarly, the proposed multivariate models are potentially useful for the elaboration of simple and effective prognostic scores in neurological risk stratification.