mercredi 5 février 2020 

14h10 - 15h10


 
Espace Poster 4

Infectiologie : divers I

MODÉRATEUR(S) :  Jean-Pierre QUENOT (Dijon), Guillaume VOIRIOT (Paris)  
  

Discrepancies exist between medical and paramedical ICU health workers’ opinions on a mandatory vaccination for influenza

Orateur(s) :   Renaud PRÉVEL (Bordeaux) 

Auteur(s) :  Raphaël ENAUD (Bordeaux)   Erwan BEGOT (Bordeaux)   Alexandre MASSRI (Pau)   Chloé GISBERT-MORA (Bayonne)   Patrick BERGER (Bordeaux)   Cédric CARRIÉ (Bordeaux)   Hugues DE COURSON (Bordeaux)   Alexandre BOYER (Bordeaux)   Didier GRUSON (Bordeaux)  

14h10 - 14h17
Durée de la présentation : 4 min
Durée de la discussion : 3 min


Abstract : 
Discrepancies exist between medical and paramedical ICU health workers’ opinions on a mandatory vaccination for influenza

Introduction / Rationale :
Influenza is a potential lethal disease causing dozens of thousands excess deaths per year both in Europe and in the United states. Besides hygiene procedures, vaccination is a cornerstone of influenza prevention and guidelines recommend for vaccination among health workers (HW), especially if they are in close contact with frail people. Despite these recommendations, the vaccination coverage is low among health workers both in Europe and in the US. The relevance of a mandatory vaccination for health workers is currently a hot topic but data are scarce regarding intensive care unit health workers’ opinion.

Méthodes / Patients and Methods :
Health workers from 2 medical, 6 surgical and 2 polyvalent ICUs received a link to the electronic record of the survey.

Résultats / Results :
Among the 10 ICUs, 1637 ICU health workers (HW) (medical: 185 and paramedical: 1452) were questioned. Three hundred and forty-one ICU (21%) answered, 107 (58%) medical health workers (MHW) and 234 (16%) paramedical health workers (PHW) (p < 0,0001). Among MHW 94/107 (88%) were vaccinated vs only 119/234 (51%) PHW (p< 0,0001). Discrepancies exist between medical and paramedical ICU health workers’ opinions and beliefs about vaccination for influenza and its acceptance. Medical health workers were more prone to consider influenza as a potentially lethal disease occurring not only among frail people but also in healthy people, to consider the vaccine efficient and safe. To agree with “Vaccination for influenza is mostly related with gain for pharmaceutical industry” (OR: 11 [2.3-50]) and to disagree with “The risk of Guillain-Barré syndrome is higher after an episode of influenza than after vaccination for influenza” (OR: 4.2 [1.2-14]) were independently associated to the disagreement with a mandatory vaccination for ICU HW.

Discussion / Discussion :


Conclusion / Conclusion :
Vaccination for influenza should be strongly recommended as a tool of individual protection for ICU health workers as for general population. As confidence in vaccine efficacy and concerns about vaccine side-effects impact the vaccination rate, objective information should be provided to ICU health workers about the efficacy and the side effects of vaccination for influenza.
 

Predictive mortality factors of severe influenza A in intensive care: Experience of a tunisian referral center

Orateur(s) :   Samia AYED (Tunis, TUNISIE) 

Auteur(s) :  Amira JAMOUSSI (Tunis)   Hamdi DOGHRI (Tunis)   Takoua MERHABENE (Tunis, TUNISIE)   Amine SLIM (Tunis)   Jalila BEN KHELIL (Tunis)   Mohamed BESBES (Tunis)  

14h17 - 14h24
Durée de la présentation : 4 min
Durée de la discussion : 3 min


Abstract : 
Predictive mortality factors of severe influenza A in intensive care: Experience of a tunisian referral center

Introduction / Rationale :
Influenza A virus infection is a contagious acute respiratory infection with possible severe clinical form. In Tunisia, surveillance of influenzae was initiated in 1989 and our unit is referral center of surveillance of severe cases since 2009.
The aim of this study was to describe the epidemiological and clinical aspects of influenza A, and to determine independent predictive factors of ICU mortality.

Méthodes / Patients and Methods :
It was an observational cohort study over an 11-year period (June 2009-June 2019). We prospectively collected data (demographic, clinical and biological data, evolving features) of all consecutive patients diagnosed with influenza A.
Multivariate analysis of the predictive factors of ICU mortality was realized.

Résultats / Results :
During the study period, 120 patients with influenza A were admitted, 14 (11.7 %) were pregnant women. Mean age was 48 ± 16 years and a sex-ratio of 1.14. The mean scores of SAPS II, APACHE II and SOFA were respectively 30, 12 and 4.5.
A history of hypertension, diabetes, chronic obstructive pulmonary disease and obesity were recorded in respectively 25, 20.8, 12.5 and 7.5% of cases. Only 7 (5.8%) patients were vaccinated.
Acute respiratory failure was the main reason for admission in 115 patients (95.8%). Mean first PaO2/FiO2 was 181 ± 100 mmHg [45-457]. Acute respiratory distress syndrome (ARDS) was diagnosed in 82 patients, it was severe (n=44; 54%), moderate (n=27; 33%) and mild (n=11; 13%).
At admission, lymphopenia was present in 86 patients (72%) and rhabdomyolysis in 19 (15.8%).
Virus strains identified with PCR were H1N1 pdm09 (84.2%) and H3N2 (15.8%). A Co-infection with bacteria was documented in 16 cases (13.3%) and aspergillosis in 3 cases.
The most frequent complications were acute kidney injury (n=50), shock (n=48) and hospital-acquired infections (n=46).
Antiviral therapy oseltamivir* was prescribed in 88 patients.
Non-invasive ventilation (NIV) was used in 72 patients with success in 35 cases (48.6%). Endotracheal intubation was performed in 59 patients (37 after NIV failure).
Mean ICU length of stay was 12 ± 12.8 days and the overall ICU mortality was 31.6%.
Independent predictive factors of ICU mortality were : severe ARDS (OR=4.7 ; IC95% [1-20.9] ; p=0,040), acute kidney failure (OR=5.28 ; IC95% [1.2-23.3] ; p<10-3) and shock (OR=29 ; IC95% [5.3-157.5] ; p<10-3). An age under 50 years was protective against ICU mortality (OR=0.129 ; IC95% [0.027-0.606] ; p=0,009).

Discussion / Discussion :


Conclusion / Conclusion :
Influenza A in ICU characterized by high morbi-mortality, especially in patients with severe ARDS or shock.
 

Antibiotherapy of peritonitis in intensive care: Impact of the 2015 SFAR recommendations and interest of an adapted antibiotherapy.

Orateur(s) :   Adrien KRINGS (Tourcoing) 

Auteur(s) :  Hugues GEORGES (Tourcoing)   Pierre-Yves DELANNOY (Tourcoing)   Olivier LEROY (Tourcoing)  

14h24 - 14h31
Durée de la présentation : 4 min
Durée de la discussion : 3 min


Abstract : 
Antibiotherapy of peritonitis in intensive care: Impact of the 2015 SFAR recommendations and interest of an adapted antibiotherapy.

Introduction / Rationale :
Intra-abdominal infections are a major cause of morbidity and mortality. SFAR recommendations on this topic were published in February 2015. The purpose of this work is to evaluate whether our antibiotic therapy was adequate for these recommendations and whether they were adapted to our unit. The secondary objectives are to look for different risk factors for mortality, to evaluate the impact of inappropriate antibiotic therapy, to evaluate the relevance of the carbapenem prescription.

Méthodes / Patients and Methods :
This is a single-center retrospective observational study of secondary peritonitis in the Tourcoing intensive care unit. For each peritonitis, the epidemiological data and the co-morbidities of the patients were collected. Bacteriology and anti-infectious therapies were described to determine the rates of adaptation of our antibiotic therapy and that recommended by SFAR. The adequacy of our treatments to the recommendations was also quantifiable. The description of the stay, the occurrence of a death was specified.

Résultats / Results :
131 peritonitis were included. The rate of adaptation of the SFAR antibiotic therapy was 80%. The rate of adaptation of our antibiotic therapy was 73% and its adequacy rate of 24%. The main differences in prescriptions concerned over-prescription of antifungals, molecule against gram positive bacillus and a sub-prescription of aminoglycosides and beta-lactams, in particular carbapenems. The different mortality risk factors found were SOFA score> 8 (OR 5,4 95% CI 2,15-13,7), the Charlson score> 3 (OR 5.2 95% CI 1.52-17.9), the hollow organ perforation (OR 4.2 95% CI 1.63-10.59). A comparison of the appropriate or not antibiotic groups did not reveal a significant difference in mortality, number of surgical revision and length of stay. In 55% of nosocomial peritonitis, antibiotic therapy with carbapenem was recommended. After recovery of microbiological data, it was only necessary for 7.7% of cases.

Discussion / Discussion :


Conclusion / Conclusion :
Our work showed a low rate of compliance with SFAR recommendations. These recommendations are applicable to our service by providing a particular reflection for fungal infections. Our study does not show a correlation between mortality and inadequate antibiotic therapy, surgery remaining the major treatment.
 

Epidemiological profile and antibiotic susceptibility of Acinetobacter baumannii isolates in a trauma and burn center in Tunisia

Orateur(s) :   Lamia THABET (Ben Arous, TUNISIE) 

Auteur(s) :  Emna HAMMAMI (Ben Arous)   Sarra DHRAIEF (Ben Arous, TUNISIE)   Mehdi GADDAS (Ben Arous)   Sonia BEN BEHI (Ben Arous, TUNISIE)   Amenallah MESSAADI (Ben Arous)  

14h31 - 14h38
Durée de la présentation : 4 min
Durée de la discussion : 3 min


Abstract : 
Epidemiological profile and antibiotic susceptibility of Acinetobacter baumannii isolates in a trauma and burn center in Tunisia

Introduction / Rationale :
Acinetobacter baumannii is a gram-negative opportunistic bacteria that has gained several drug resistance mechanisms over the last decades. Analysis of A.baumanii’s resistance profile helps to establish a prompt control and a prevention program.
The aim of this study was to evaluate the epidemiology and antimicrobial resistance of A.baumannii isolates in a trauma and burn center in Tunisia.

Méthodes / Patients and Methods :
Retrospectively, we studied all strains of Acinetobacter baumannii isolated over a seven-year period (from January 2012 to December 2018). Conventional methods were used for identification. Antimicrobial susceptibility testing was performed with the disk diffusion method, and susceptibility results were interpreted using clinical breakpoints according to CA-SFM guidelines. Data were analyzed using the SIR-system. Minimum inhibitory concentration (MIC) of colistin was determined using the E-test® method (bioMérieux), then using the EUCAST broth micro-dilution method ( UMIC, Biocentric®) since May 2017.

Résultats / Results :
During the study period, 1248 non-repetitive strains of Acinetobacter baumannii were isolated representing 9.9% of all isolates, 14% of gram-negative bacilli (GNB) and 40.6% of non-fermenting GNB. In our center, infections due to A.baumannii were endemic with epidemic peaks. A.baumannii was mainly isolated from burn intensive care unit (67%) and anesthesiology department (22.6%).
The most frequent sites of isolation were blood cultures (34.3%), catheters (20%), respiratory specimens (12.5%) and skin samples (10%).
The survey of antibiotic susceptibility showed high percentages of resistance to the different antibiotics: 84% to ceftazidime, 93% to imipenem, 86% to amikacin and 91,5% to ciprofloxacin.
From 2012 to 2018, the imipenem resistance rate was stable but high (over 90 % of resistant strains). Amikacin resistance rate was stable all over the study period. Whereas ceftazidime resistance rate was high and fluctuating (57% in 2016, 89% in 2013 and 2017). A.baumannii strains gained resistance to ciprofloxacin over the years (from 86 % in 2012 92.3 % in 2018). Twenty-two strains were resistant to colistin. These strains were isolated mainly from burn patients (72.7%).

Discussion / Discussion :


Conclusion / Conclusion :
The dissemination of A.baumannii multidrug-resistant strains in our center must be contained by the implementation of strict isolation methods and better hygienic procedures.
 

Multicentric satisfaction survey of aScope Bronchosampler™, a new sampling accessory for aScope™4 Broncho

Orateur(s) :   Emmanuel NOVY (Vandoeuvre-Les-Nancy) 

Auteur(s) :  Claude MEISTELMAN (Vandoeuvre Les Nancy)   Clement FOURNIER (Lille)   Gilles DHONNEUR (Paris, AFGHANISTAN)   Julien POTTECHER (Strasbourg)  

14h38 - 14h45
Durée de la présentation : 4 min
Durée de la discussion : 3 min


Abstract : 
Multicentric satisfaction survey of aScope Bronchosampler™, a new sampling accessory for aScope™4 Broncho

Introduction / Rationale :
aScope BronchoSampler is a small sterile kit to be clicked on aScope 4 Broncho simplifying and securing the sampling technique using a closed process (no suction tube switch).
This prospective multicentric observational satisfaction survey had for main objective to collect the user point of view regarding BronchoSampler especially its functionality in clinical practice, such as bronchial sampling in ICU Units/Pulmonology Departments with aScope 4 Broncho versus standard sampling technique, in order to assess its potential adoption.

Méthodes / Patients and Methods :
BronchoSampler was evaluated in 4 hospitals (Lille, Nancy, Strasbourg, Curie Institute) for 4 months. Ambu provided the units free of charge to the different sites. 48 evaluation forms have been collected and consolidated involving 23 operators already using aScope 4 Broncho in their practice.
BronchoSampler regardless of the item, was evaluated in comparison to the standard sampling method used by the operators.
Results are expressed in absolute value and percentage; the sum might differ from 48 due to lack of answer from evaluator.

Résultats / Results :
BronchoSampler was mainly used for the following procedures:
Bronchial Wash - BW: 8 (17%)
Bronchial Alveolar Lavage - BAL: 36 (75%)
Usually, those procedures were done as follows:
Wall suction/specimen trap: 40 (84%)
Manual pull/syringe: 2 (4%)
Material preparation traditionally requires two dedicated persons 32 (67%) versus one operator only for BronchoSampler usage 27 (56%). It frees assistant time and enable the clinician to perform the sampling alone more often.
Detailed of the different evaluated items are listed in the grid.
Sampling duration is also reduced, improving workflow.
Evaluators consider that BronchoSampler rationalizes the cumbersome sampling process and that the closed system design reduces the risk of losing sample or sample contamination. The set-up, the suction capacity, the sampling quality and quantity have all been evaluated better or far better than that usually observed with usual sampling techniques and devices.
Finally, 36 (75%) of users prefer BronchoSampler to commonly used method.

Discussion / Discussion :


Conclusion / Conclusion :
This satisfaction survey shows that with its simple but revolutionary design, BronchoSampler brings a real effective benefit in sampling procedure enabling the clinician to perform it alone, and 39 (81%) of the survey evaluators consider that BronchoSampler should replace their current practice.
 

Epidemiological profile and antibiotic susceptibility of Staphylococcus aureus isolates in a trauma and burn center in Tunisia

Orateur(s) :   Lamia THABET (Ben Arous, TUNISIE) 

Auteur(s) :  Karim MECHRI (Ben Arous, TUNISIE)   Sarra DHRAIEF (Ben Arous, TUNISIE)   Hana FREDJ (Ben Arous)   Kawther FALEH (Ben Arous)   Amenallah MESSAADI (Ben Arous)  

14h45 - 14h52
Durée de la présentation : 4 min
Durée de la discussion : 3 min


Abstract : 
Epidemiological profile and antibiotic susceptibility of Staphylococcus aureus isolates in a trauma and burn center in Tunisia

Introduction / Rationale :
Staphylococcus aureus is one of the main bacteria involved in nosocomial infections. The spread of methicillin-resistant S. aureus (MSRA) has led to considerable difficulties in the treatment of infections due to this pathogen.

Méthodes / Patients and Methods :
The aim of our study was to evaluate epidemiological profile and antibiotic resistance of S. aureus isolates in a trauma and burn center in Tunisia.
Retrospectively, we studied all strains of S. aureus isolated over a seven-year period (from January 2012 to December 2018). Conventional methods were used for identification. Antimicrobial susceptibility testing was performed with disk diffusion method and susceptibility results were interpreted using clinical breakpoints according to CA-SFM guidelines. Data were analyzed using the SIR-system.

Résultats / Results :
During study period, 1875 non-repetitive strains of S. aureus were isolated, representing 14.8% of all isolates. S. aureus was mainly isolated from burn intensive care unit (ICU) (56%), orthopedics (14.4%) and anesthesiology department (12.3%). The most frequent sites of isolation were blood cultures (35.8%) and skin samples (24.4%). The survey of antibiotic susceptibilty showed that MRSA rate was variable depending on the ward involved : The highest rates were observed in burn ICU (61%) and anesthesiology department (48.7%). The lowest rate was observed in orthopedics (22.4%). The overall resistance to methicillin was 47.5%. Evolution of MRSA rate was marked by a decrease from 55% in 2012 to 30.2% in 2018. Regarding other families of antibiotics, the resistance rates have also decreased during the study period: from 53.6% in 2012 to 25.3% in 2018 for gentamicin, from 45.3% and 16.7% in 2012 to 11% and 1.7% in 2018 for erythromycin and clindamycin, respectively, and from 44% in 2012 to 21.8% in 2018 concerning ciprofloxacin. Resistance to tigecycline fell from 20.5% in 2012 to 0.52% in 2018. Resistance to linezolid was rare. It concerned only one strain. All strains were susceptible to glycopeptides.

Discussion / Discussion :


Conclusion / Conclusion :
The decrease in MRSA rate and other associated resistances could be explained by a strengthening of hygiene measures and a rationalization of antibiotics use in our center.
 

INTEREST OF PROCALCITONIN IN INTENSIVE CARE (About 120 cases)

Orateur(s) :   Amine RAJA (Casablanca, MAROC) 

Auteur(s) :  Mohamed ELAIASSI (Casablanca)   Boubaker CHARRA (Casablanca)  

14h52 - 14h59
Durée de la présentation : 4 min
Durée de la discussion : 3 min


Abstract : 
INTEREST OF PROCALCITONIN IN INTENSIVE CARE (About 120 cases)

Introduction / Rationale :
The possibility of having a sensitive, specific and prognostic biological marker for bacterial infections is a considerable challenge. A step was taken with the discovery of pracalcitonin.

Méthodes / Patients and Methods :
This is a prospective observational cohort study of 120 patients in the Medical Resuscitation Department of the University Hospital of Casablanca during the 6-month period, Including patients in whom the PCT was dosed, The data collected allowed us to form two groups according to the PCT value: PCT + group with PCT> 2ng / ml and PCT- group with PCT <2ng / ml.
The statistical analysis of these different data was carried out using Epi Info software version 3.5.3.

Résultats / Results :
60% of our patients had a bacterial infection and 40% did not have one.
We also distinguished community infections (42% of I + patients) and nosocomial infections (58% of I + patients). We found that the highest rates of PCT were in nosocomial infections and the lowest PCT rates were found in community-acquired infections. Then, in each type of organ involvement we tried to vary the PCT thresholds to 0.5-2 and 10 ng / ml in order to find the best threshold for which PCT allowed to diagnose bacterial infection, justifying our choice of departure.
We concluded that the best PCT cut-off value in general was 2ng / ml, because it gave us the best sensitivity / specificity ratio (67% and 50% respectively) with a positive predictive value of 65% and a negative predictive value of 52%. The link between PCT and bacterial infection was moderate (Yule Q-factor at 0.34). By analyzing the different therapeutic aspects, we showed that 71% of our patients had been treated with ATB before the PCT assay and that the broadest spectrum antibiotics available to our service were used in patients with PCT levels. the highest. Finally, concerning the evolution, the higher the rate of PCT, the higher the death rate, especially since 100% of patients with PCT> 10ng / ml died.

Discussion / Discussion :


Conclusion / Conclusion :
Procalcitonin is considered to be one of the best markers of systemic bacterial infection. Indeed, its elevation is earlier than that of CRP and its specificity is better compared to IL-6 and IL-8. The rate of procalcitonin remains low in the presence of viral infection. Procalcitonin is also a prognostic marker, its elevation is correlated with the severity of the infection, and its decrease is a good indicator of the effectiveness of antibiotic therapy.
 

Worldwide clinical practices in perioperative antibiotic therapy for lung transplantation

Orateur(s) :   Sami HRAIECH (Marseille) 

Auteur(s) :  Benjamin COIFFARD (Marseille)   Eloi PRUD'HOMME (Marseille)   Nadim CASSIR (Marseille)   Laurent PAPAZIAN (Marseille)  

14h59 - 15h06
Durée de la présentation : 4 min
Durée de la discussion : 3 min


Abstract : 
Worldwide clinical practices in perioperative antibiotic therapy for lung transplantation

Introduction / Rationale :
Due to induction immunosuppression infection is the most common cause of mortality within the first year after lung transplantation (LTx). The management of perioperative antibiotic therapy is a major issue, but little is known about worldwide practices.

Méthodes / Patients and Methods :
We sent by email a survey to 180 LTx centers around the world dealing with 5 daily clinical vignettes concerning perioperative antibiotic therapy. We considered perioperative period as the period of the transplant surgery (per operative) and the post-surgery time before any infection occurrence (postoperative). After general questions on local practices, we asked each center for colonization definition and their diagnostic methods for microbial screening in recipients and donors. The clinical cases were related to specific issues concerning the management of antibiotic therapy in different clinical situations, including no prior colonization, prior colonization with sensitive or multi-drug resistant (MDR) microorganisms including prior colonization with MDR bacteria not sensitive to beta-lactams. The invitation and a weekly reminder were sent to lung transplant specialists for a single consensus answer per center between June and September 2018.

Résultats / Results :
We received a total of 99 responses from 24 countries, mostly from Western Europe (n=46) and the USA (n=34), Figure 1. Systematic screening for bronchial colonization before LTx was mostly performed with sputum samples (72%), regardless of the underlying lung disease. Definition of colonization was very heterogeneous and the delay between the last bacterial isolation in pre-transplant and the LTx to consider if the therapy should target these bacteria varied between 1 week and more than one year. In recipients without colonization, antibiotics with activity against gram-negative bacteria resistant strains (piperacillin / tazobactam, cefepime, ceftazidime, carbapenems) were reported in 72% of the centers, and antibiotics with activity against methicillin-resistant Staphylococcus aureus (mainly vancomycin) were reported in 38% of the centers. For these recipients, the duration of antibiotics reported was 7 days (33%) or less (26%) or stopped when cultures of donor and recipients were reported negatives (12%). In recipients with pre-transplant colonization, antibiotics were adapted to the susceptibility of the most resistant strain isolated in pre-transplant samples and given for at least 14 days (67%).

Discussion / Discussion :


Conclusion / Conclusion :
Practices vary widely around the world, but resistant bacterial strains are mostly targeted even if no colonization occurs. The antibiotic duration reported was longer for colonized recipients.