Nosocomial infections in patients requiring short-term mechanical circulatory support: characteristics, risk factors and prognosis. (3rd May 2023)

• Record Type: Journal Article
• Authors: Hernandez Martos, A V; Alonso Fernandez De Gatta, M; Martin Herrero, F; Moreta Matilla, M; Diego Nieto, A; Merchan Gomez, S; Gonzalez Cebrian, M; Toranzo Nieto, I; Barrio Rodriguez, A; Cid Menendez, A; Gonzalez Calle, D; Barreira De Sousa, G J; Cruz Galban, A; Hernandez Hidalgo, M; Sanchez Fernandez, P L
• Is Part Of: European heart journal. Volume 12(2023)Supplement 1
• Publication Date: 2023-05-03
• Subjects: 616.1205
• DOI: 10.1093/ehjacc/zuad036.109
• Languages: English
• ISSNs: 2048-8726

Abstract

Abstract: Funding Acknowledgements: Type of funding sources: None. Background: The use of short-term mechanical circulatory support (MCS) has been increasing in recent years. Infection is a frequent complications and seems to contribute to increased morbidity and mortality. Purpose: Our objective was to describe the profile of infections in patients with short-term MCS, as well as factors related to their appearance and their prognostic influence. Methods: Retrospective analysis of consecutive cases of venoarterial extracorporeal membrane oxygenator (VA-ECMO) or Impella CP® implantation in a referral center. We characterized the infections during the entire hospitalization and analyzed their relationship with different variables and survival. Infection as microbiological isolation (no colonization) or clinical suspicion that motivates antimicrobial therapy with therapeutic intent. Results: 232 patients between 2014-sep 2022 (Table). A total of 104 (56.5%) patients presented infection, respiratory tract (other than pneumonia) (23.3%) was the most prevalent. The most frequently isolated microorganisms were Gram-negative bacteria and Staphylococcus aureus, with 8.2% multidrug-resistant bacteria (Figure). Infection appeared during MCS in 25% of patients, 13% after its removal, 11% at both times, and 6.1% of patients were already infected at MCS implant. Baseline characteristics and admission evolution depending on the presence of infection are described in table. Infections were significantly more frequent in cardiogenic shock indication and bridge to recovery intention, and non-percutaneous access. We observed a non-statistically significant trend towards more infections under VA-ECMO vs Impella CP® device. The following factors were associated with the appearance of infection: a longer time under MCS, endotracheal intubation (especially out-of-hospital), longer invasive mechanical ventilation (MV) and tracheostomy, right ventricular dysfunction, the need for vasoactive drugs, and corticosteroids treatment during more than 3 days. The appearance of infection was significantly associated with other complications (Table). At multivariate analysis a longer time under MCS (HR 1.15, IC95% 1.06-1.26, p=0.002) and MV (HR 1.06, IC95% 1.02-1.09, p=0, 001) were independently related to infections. Overall in-hospital mortality was 59.8% (n=137), 3.9% due to infection. No direct relationship was observed between in-hospital mortality and the presence of infection (57.4% in infected patients vs 61.7% without infection, p=0.587). Patient already infected at MCS implantation had higher mortality (87.7% vs 57% during MCS vs 41.9% after MCS, p=0.030). Conclusions: Infections in patients under short-term SCM are frequent, highlighting the upper respiratory tract infection. Longer support and ventilation time were independently related to infections. The development of infection was associated with other complications, but not with higher in-hospital mortality. Short-term MCS in already infected patients was related to high mortality. Table Figure