Aims: Staphylococcus aureus bacteremia (SAB) can be complicated by endocarditis. The aim of this study was to develop and validate a prediction score to determine endocarditis risk among SAB.
Methods: This retrospective study included adults with SAB hospitalized in Lausanne University Hospital from 2015 to 2021. Using the modified Duke Criteria for definite endocarditis as reference standard, the new LAUSTAPHEN score was compared to previous scores (VIRSTA, PREDICT, and POSITIVE).
Results: Among 821 episodes, definite endocarditis was diagnosed in 118 episodes (14.4%) using the modified Duke Criteria and 102 (12.4%) using the imaging and pathological criteria. Transthoracic and transesophageal echocardiography (TEE) were performed in 77.5% and 42.1% of episodes, respectively. Multivariate analysis established that cardiac predisposing factors (p < 0.001; OR 5.9, CI 3.3-10.5), cardiac implantable electronic devices (CIED; p < 0.001; OR 8.4, CI 4.3-16.5), prolonged bacteremia ≥ 48h (p 0.007; OR 1.3, CI 1.1-1.5), community-acquired bacteremia (p 0.011; OR 1.5, CI 1.1-2.2), and vascular phenomena (p < 0.001; OR 16.9, CI 9.5-30.3) were independently associated with endocarditis (according to modified Duke criteria). The same factors were identified when using the imaging and pathological criteria. LAUSTAPHEN score included: cardiac predisposing factors, CIED, prolonged bacteremia ≥ 48h, vascular phenomena and native bone and joint infection; presence of any parameter allowed to classify patients in the high-risk group. LAUSTAPHEN and VIRSTA scores had a negative predictive value > 98% and misclassified < 5% of endocarditis cases as low risk. Misclassification of POSITIVE and PREDICT was > 10%. If the scores were to be implemented in clinical practice, the number of TEEs required to safely exclude endocarditis would have been 66.9% and 51.6% with VIRSTA and LAUSTAPHEN, respectively.
Conclusion: We constructed a new reliable and easy to assess score with five variables known to be associated with endocarditis. Neither POSITIVE nor PREDICT should be used in clinical practice, due to the high rate of misclassification of endocarditis patients in low risk group. LAUSTAPHEN and VIRSTA scores exhibited the lowest number of misclassification of endocarditis patients in low risk group. Nonetheless, VIRSTA significantly overestimated endocarditis risk, leading to a higher number of echocardiography (increase of 29.7%) to achieve the same result as LAUSTAPHEN.

Aims
The primary objective was to describe the incidence and timing of bloodstream infections (BSI) during the 1st-year post-solid organ transplant (SOT). Secondary objectives included the: (a) epidemiology of bacterial pathogens, (b) primary site of infection, (c) risk factors, and (d) all-cause mortality in patients with BSI during the 1st-year post-SOT, using the Swiss Transplant Cohort Study (STCS) database.
Methods
All consecutive adult SOT-recipients (SOTr) who received a heart, kidney, liver, lung or kidney-pancreas transplant between 01.05.2008 and 31.12.2019 were included.
Results
Among 4’383 SOTr there were 640 (14.6%) patients with 963 BSI-events identified. Of them, 557 (557/963, 57.8%) BSI-events in 415 (415/640, 64.8%) SOTr occurred within the 1st-year post-SOT. The incidence of 1st BSI per patient at one-year post-SOT was 9.5% (95%CI: 8.6-10.4%) with a median time from SOT to occurrence of 44 days (IQR: 9-128.5; Figure 1A). By SOT type, it was 12.8% (95%CI: [9.6, 16.5], 8.3% [7.2, 9.4], 11.4% [9.6, 13.4], 9.8% [7.2, 12.7] and 5.9% [2.4, 11.6] in heart, kidney, liver, lung and kidney-pancreas recipients, respectively (p=0.003; Figure 1B). Time to 1st BSI was shortest among lung (17 days (6.25-77.25)) and longest in kidney recipients (69 days (23-171)) ("p < 0.001"). More than one pathogen was detected in 9% (36/415) of all first BSI. The cumulative incidence of mono- and poly-microbial BSI after one-year post-SOT was 8.7% [7.8, 9.5] and 0.8% [0.6, 1.1], with a trend for time to mono-microbial BSI being shorter compared to poly-microbial BSI (median: 43 (9-126.5) vs. 71 days (12-144.5); "p = 0.08"; Figure 1C). Among patients with mono-microbial BSI, BSI due to gram-positive bacilli were rare and time to occurrence was shorter (median days 4.5 (2.25-15) in 10 patients) compared to gram-positive cocci (GPC, n=123/379: 32.5%, median days to BSI: 31 (6.5-111) and gram-negative bacilli (GNB, n=246/379: 64.9%, median days to BSI: 53.5 (12-138.75); "p < 0.001"; Figure 1D).
Conclusions
BSI occur in almost 1 in 10 SOTr, predominately during the 1st year post-SOT and relatively early post-SOT, mainly due to GNB followed by GPC.

Background: Staphylococcus aureus persistent bacteraemia is only vaguely defined and the effect of different durations of bacteraemia on mortality is not well established. Our primary aim was to analyse mortality according to duration of bacteraemia and to derive a clinically relevant definition for persistent bacteraemia.
Methods: We did a secondary analysis of a prospective observational cohort study at 17 European centres (nine in the UK, six in Spain, and two in Germany), with recruitment between Jan 1, 2013, and April 30, 2015. Adult patients who were consecutively hospitalised with monomicrobial S. aureus bacteraemia were included. Patients were excluded if no follow-up blood culture was taken, if the first follow-up blood-culture was after 7 days, or if active antibiotic therapy was started more than 3 days after first blood culture. The primary outcome was 90-day mortality. Univariable and time-dependent multivariable Cox regression analysis were used to assess predictors of mortality. Duration of bacteraemia was defined as bacteraemic days under active antibiotic therapy counting the first day as day 1.
Results: Of 1588 individuals assessed for eligibility, 987 were included (median age 65 years [IQR 51–75]; 625 [63%] male). Death within 90 days occurred in 273 (28%) patients. Patients with more than 1 day of bacteraemia (315 [32%]) had higher Charlson comorbidity index and sequential organ failure assessment scores and a longer interval from first symptom to first blood culture. Crude 90-day mortality increased from 22% (148 of 672) with 1 day of bacteraemia, to 39% (85 of 218) with 2–4 days, 43% (30 of 69) with 5–7 days, and 36% (10 of 28) with more than 7 days of bacteraemia. Metastatic infections developed in 39 (6%) of 672 patients with 1 day of bacteraemia versus 40 (13%) of 315 patients if bacteraemia lasted for at least 2 days. The second day of bacteraemia had the highest HR and earliest cutoff significantly associated with mortality (adjusted hazard ratio 1·93, 95% CI 1·51–2·46; p < 0·0001).
Conclusion: We suggest redefining the cutoff duration for persistent bacteraemia as 2 days or more despite active antibiotic therapy. Our results favour follow-up blood cultures after 24 h for early identification of all patients with increased risk of death and metastatic infection. This study can serve as a basis for a prospective trial evaluating diagnostic and therapeutic interventions in persistent S.aureus bacteremia.

See Poster Nr. P74, P75, P76, P77 for Abstract Details

Aims: During the second and third wave of COVID-19 in Switzerland, an increasing number of patients exposed roommates to SARS-CoV-2. We aimed to calculate the secondary attack rates after exposure and to identify risk factors for SARS-CoV-2 transmission.
Methods: All patients with COVID-19 admitted to the University Hospital Zurich during the second and third COVID-19 wave (Week 40 in 2020 – Week 25 in 2021) were included in the analysis. We stratified COVID-19 as community-associated, healthcare-associated (definite and probable) or indeterminate according to the European Centre for Disease Prevention and Control (ECDC) criteria. Index patients were defined as patients who had contact to other patients during 48h before symptom onset (or before a positive SARS-CoV-2 test if asymptomatic). Exposed patients were defined as patients sharing a room with an index patient, for (1) ≥6 hours in intermediate or intensive care units (IMC or ICU), or (2) any time on the general ward, or when the index patient underwent an aerosol generating procedure (AGP). Transmission pathways between patients were assessed by in-depth review of symptom onset, Ct-values, patient contacts, and sequencing data if available. Secondary attack rates (SAR) were assessed by estimating the cumulative incidence using the Kaplan-Meier estimator. Risk factors for transmission were assessed in uni- and multivariable logistic regression models.
Results: In this retrospective study, of 1118 in-patients with COVID-19 at our hospital, 83 (7.4%) had probable and definite healthcare-associated COVID-19. A total of 298 patients were exposed to SARS-CoV-2 by 127 index patients. After in-hospital exposure, estimated cumulative incidence of COVID-19 was 23.3%. Transmission of SARS-CoV-2 was associated with exposure time (adjusted Odds Ratio (aOR) per hour increase: 1.02, 95%CI: 1.01-1.03) and Ct-value of the index patient (aOR per unit increase: 0.93, 95%CI: 0.87-0.98).
Conclusion: High viral loads of index patients and prolonged contact time in shared bedrooms play an important role in healthcare-associated SARS-CoV2-transmission. Even though the findings originate from a time with circulating pre-alpha and alpha-variants in a non-vaccinated population, they might again gain relevance with the emerging of new variants of concern, and waning of immunity. It seems particularly important to identify contagious patients early, assert mask adherence and adherence to regularly aerate shared patient rooms.

Aims: Following the successful pilot program “Safe urinary catheters” that ran from 2015-2018 as a joint venture between Patient Safety Switzerland and Swissnoso, the Federal Office of Public Health mandated Swissnoso to develop a national surveillance module for catheter-associated urinary tract infections (CAUTI). Here, we describe the methodology and first results of the new surveillance module.

Methods: We assessed the toolkit from the pilot program and revised the surveillance manual, the surveillance definition, and other pertinent variables. Then, we identified a database provider that would serve as the platform for data acquisition from participating hospitals, construct an algorithm for infection event calling, and present data output in a user-friendly fashion. All acute care hospitals in Switzerland were invited to participate in the new module in mid-2021, with the start of the surveillance set for 1/1/2022. The plan was to report surveillance data along with benchmark data on a quarterly basis. How data was collected (ie, manually or in an automated way) was at the discretion of each hospital.

Results: Data capturing started 1/1/2022 and a first analysis was done for the 3-month period from 1/1/2022 – 31/3/2022. Fourteen hospitals participated in the surveillance. There were 30,322 patients with 4,791 urinary catheters in place, resulting in a proportion of 14.21 (95% CI 13.82 – 14.61) percent of catheterized patients. Expressed as catheter days per 100 patient days, the rate was 9.09 (8.97 – 9.21). A total of 24 CAUTI events occurred, yielding a rate of 1.2 CAUTI per 1000 catheter days (95% CI 0.77 – 1.78), or 0.01 CAUTI per 100 patient days (95% 0.01 – 0.02). Eight of the 14 hospitals collected information on whether the catheter had an evidence-based indication, which was documented for 84.61% (95% CI 83.16 - 85.98) of catheters.

Conclusion: The new CAUTI surveillance module by Swissnoso was launched successfully in the first quarter of 2022. Due to the ongoing Covid-19 pandemic, only few hospitals were able to dedicate resources to this surveillance. Catheter utilization and CAUTI rate were comparable to published data from other healthcare systems. Over time, more hospitals are expected to participate in Switzerland’s CAUTI surveillance.

Background
Mitarbeitende (MA) im Gesundheitswesen (engl. Abk.: HCW) sind das Rückgrat hinsichtlich Erfüllung spitalhyg. relevanter Standards. Ihre Förderung ist obligatorisch und notwendig. Anforderungen für gute Akzeptanz von Standards und Richtlinien der Spitalhygiene sind u. a.: Örtlich und zeitlich flexibler Zugriff auf Wissen; Periodische Schulungen zur MA-Förderung; Inhalte verfügbarer Grundlagen friktionsfrei in Arbeitsalltag integrieren.
Aims
Spitalhygiene gewinnt mit der Digitalisierung an Modernität und verbessert seine Reichweite. Neben Verfügbarkeit fördern Zugänglichkeit und Vermittlung von Wissen eine gute Compliance. Die Strategie NOSO (2020) fordert Schlüsselkomponenten (SK). Ziel dieses Projektes ist die Einführung eines elektronischen Hygieneordners und E-Learnings zur Schulung aller HCW der Spital-Versorgungsregion Zentralschweiz. Mehrere Privatspitäler wenden einheitliche Standards der Spitalhygiene an.
Methods
Das Projekt ist in zwei Phasen gegliedert. Phase 1 (Go-live Oktober 2021): Einführung eines Sharepoint-basierten elektron. Hygieneordners. Phase 2 (Go-live Februar 2022): Implementierung eines E-Learnings zur Spitalhygiene. Das E-Learning ermöglicht den Einstieg in den Hygieneordner, vermittelt Grundlagen und motiviert zum Selbststudium. Als Serie von 6 Themen als Lehrgang aufgebaut, greift das E-Learning auf die Ressource aus Phase 1 zurück. Nutzer navigieren zwischen Spitalhygiene-Dokumenten, die durch Hyperlinks untereinander verknüpft sind. Zielvorgaben waren u.a.: Einbindung von Belegärzten; Obligatorische Absolvierung innert Jahr; Aufforderung bei Verzug; Zertifikat nach Abschluss
Results
Akzeptanz durch Nutzung des Hygieneordners ist vorhanden. Der Durchschnittswert der Treffer (Navigation Nutzer) stieg um einen Faktor 1.8 von 717 im Januar 2022 auf 1’304 im Februar 2022. Der Durchschnittswert der eindeutigen Benutzer stieg um einen Faktor 2.4 im Jan. 2022 von 192 auf 461 im Feb. 2022. Das Einführungsdatum des E-Learnings lässt einen push-Effekt feststellen. Die Nutzungsintensität wird im Jahresverlauf bis zur Sättigung gewährleistet.
Conclusion
Basierend auf ersten Ergebnissen kann von einer erfolgreichen Umsetzung des Projektes ausgegangen werden. Die exakte Bilanz zur Nutzung wird ein Jahr nach Einführung erfolgen, wenn alle HCW das E-Learning durchliefen. Mit Umsetzung des Projekts werden Ziele der Strategie NOSO umgesetzt, insbesondere betreffend den SK «Richtlinien & Weisungen» und «aufgabenorientierte Schulung».

Aims
Healthcare workers (HCW) have been recognized as sources, vectors, but also victims of pathogens causing nosocomial infections. Extending an existing HCW cohort established during the COVID-19 pandemic, we aim to explore the HCWs role in the prevention and transmission dynamics of microorganisms, and to identify modifiable individual and institutional risk factors.

Methods
During the COVID-19 pandemic, a multicentre cohort study was established involving patient-facing and non-facing HCW from different healthcare networks in Northern and Eastern Switzerland. Participants were prospectively followed to study epidemiological aspects of SARS-CoV-2. In March 2022, the established cohort was repurposed to launch a research platform (named SURPRISE+), extending the scope of the original study to other relevant infectious diseases.

Results
By May 2022, the platform had enrolled 3671 HCW from 13 healthcare networks (seven acute care networks, three psychiatric institutions, one geriatric hospital, one rehabilitation clinic, and one children's hospital) across Northern and Eastern Switzerland. Median age was 44 years, 81% were female. Most institutions (12/13) and 49% of participants had already participated in the original cohort. The new open cohort design allows to continuously include participants during the ongoing study. Within a participatory surveillance approach, a research database was built to prospectively collect data on disease exposures, symptoms, and outcomes. The monthly electronic questionnaire can be easily answered via mobile phone. Viral respiratory diseases including influenza and SARS-CoV-2 remain an important research focus: for the season 2022/2023, we will monitor the activity of viral respiratory diseases within the study population; results will be readily reported to participants, and – in aggregated form – to individual institutions. In addition, coverage of other research topics is envisioned. For September 2022, we are planning a cross-sectional study regarding HCW colonization with antibiotic resistant pathogens as well as a questionnaire on antimicrobial stewardship.

Discussion
Here, we demonstrate that research projects initiated to study SARS-CoV-2 can be leveraged and extended to a multipurpose research platform to potentially study other infectious diseases. The population of HCW is easily accessible and highly motivated to contribute to such a participatory research program.

Hintergrund
In der Schweiz sollen Präventions- und Kontrollmassnahmen eingehalten werden, um die Ausbreitung von Vancomycin-resistente Enterokokken (VRE) zu verhindern. Im USZ kam es zwischen Mai 2019 bis März 2021 zu einem VRE-Ausbruch. Hier beschreiben wir die Massnahmen, welche zur Kontrolle des Ausbruchs ergriffen wurden.
Methoden
• Kontakt-Isolation aller VRE-Patienten (bis 5 Sets negativ) und Direktexponierten (bis 3 Sets negativ); keine Isolation von Stations-Exponierten (aus Kapazitätsgründen).
• Screenings (Tag 0, 7 und 14) aller Direktexponierten; Punkt-Prävalenzscreening aller Patienten auf Ausbruchstationen (1x wöchentlich); Austrittsscreenings nach Aufenthalt auf Ausbruchsstation (punktuell), Eintrittsscreenings (punktuell)
• zweimalige Schlussdesinfektion nach VRE-Patient oder VRE-Kontaktpatient; Desinfizierende Reinigung auf Ausbruchsstationen (inkl. Betten), Toilettenreinigungen alle 2h tagsüber in gemeinsam benutzten WC’s
• Bildung interdisziplinärer und interprofessioneller VRE-Taskforce
• Bettenreduktion auf Station mit Häufung von Übertragungen (1 IMC)
• Spitalhygienische Schulungen auf Ausbruchsstationen
• Verarbeitung Abstriche nur bei sichtbarem Fäkalmaterial
• Patienteninvolvierung: Plakate für die WC-Desinfektion durch die Patienten, Informationsbroschüren zu VRE-Kolonisation, Exposition und Screenings
• Informationsschreiben an medizinische Nachbehandler
Resultate
Zwischen Mai 2019 und März 2021 (22 Monate) wurden 215 Patienten mit VRE diagnostiziert (repatriierte Patienten nicht mitgezählt). 174 (80.9%) Patienten hatten den Ausbruchsstamm mit Multilocus Sequence Typing (MLST) ST80 van B, 20 (9.3%) wurden nicht typisiert, 21 (9.8%) hatten einen anderen Stamm. Es wurden über 6500 Patientenscreenings durchgeführt. Der VRE-Ausbruch am USZ betraf hauptsächlich Patienten der Hämatoonkologie und der Viszeralchirurgie. Der Ausbruch konnte im März 2021 als beendet erklärt werden.
Schlussfolgerung
Dieser schwer zu kontroliierende Ausbruch eines VRE Van B ST80 konnte nach 22 Monaten mittels einer multimodalen Intervention und unter Involvierung von Mitarbeitenden verschiedenster Kliniken und Disziplinen (Pflegefachleute, Aerzte, Reinigung, Physio, Hotellerie etc) kontrolliert werden.

Ziele
Das Ziel dieses Projekt ist die praktische Teilumsetzung der Schlüsselkomponente 5-Audits und Monitoring-aus den «Strukturellen Mindestanforderungen für die Prävention und Bekämpfung von healthcare-assoziierten Infektionen (HAI) bei hospitalisierten Patientinnen und Patienten für Schweizer Akutspitäler» von Swissnoso.
Die Fragestellung lautet:
Welche Voraussetzungen und Instrumente werden benötigt, um ein standardisiertes Hygieneaudit zur Überprüfung der hygienerelevanten Strukturen und Abläufe zur Infektionsprävention auf den Pflegeabteilungen durchzuführen?
Methodik
Eine Literaturrecherche und Gespräche mit der Abteilung Qualitätsmanagement wurden durchgeführt. Im Rahmen eines Pilotprojektes wurden fünf Auditdokumente erstellt, die während des Projekts laufend angepasst und aktualisiert wurden. Mit den erstellten Dokumenten und Voraussetzungen wurden in einem Feldversuch auf 6 Pflegeabteilungen Hygieneaudits durchgeführt, ausgewertet und an die betroffenen Abteilungen rückgemeldet (mit Massnahmen- und Evaluationsvorschlag).
Ergebnisse
Hygiene kann man beobachten (z. B.: Einhalten der hygienerelevanten Schritte bei pflegetechnischen Massnahmen, Einhalten der Reinigungs- und Desinfektionsschritte), oder ganz konkret überprüfen (z. B.: Vorhandensein von Strukturen an den entsprechenden Orten, Erscheinungsbild der anwesenden MA bei der Überprüfung). Ein klassisches Audit hat Interviewcharakter. Audit-Fragen werden gestellt und beantwortet. Überprüfbar ist in dem Moment nur die Person, die sie beantwortet. Die bereits vorhandene Audit-Checkliste wurde komplett überarbeitet zu einer Überprüfungs-Checkliste. Die Erarbeitung des Auswertungssystems gestaltete sich komplex.
Das Pilotprojekt auf den Abteilungen ist sehr gut gelaufen und wurde von den Abteilungsleitern und Abteilungsleiterinnen als «bereichernd, als Chance» und als «interessant und spannend» erlebt. Durchweg positiv waren auch die Reaktionen auf die mündliche und schriftliche Rückmeldung nach den durchgeführten Überprüfungen.
Schlussfolgerung
Abschliessend ist eine praktisch einsetzbare Hygieneüberprüfung für die hygienerelevanten Strukturen, die auf allen Pflegeabteilungen verwendet werden kann, entstanden. Mit dem entwickelten Auswertungssystem ist sie in den internen kontinuierlichen Verbesserungsprozess integriert.

Hintergrund

Gemäss den strukturellen Mindestanforderungen für die Prävention und Bekämpfung von healthcareassoziierten Infektionen (HAI) in Schweizer Akutspitälern (Swissnoso, 2021) müssen alle Gesundheitsfachpersonen mit Patientenkontakt regelmässig zu einem Thema im Bereich der Infektionsprävention geschult werden. Aufgrund der COVID-19 Pandemie war es schwierig Präsenzveranstaltungen durchzuführen. Die Herausforderung bestand darin, Informationen wie z.B. Änderungen in der COVID-Strategie, zeitnah allen Mitarbeitenden der sechs Standorte der Insel Gruppe zugänglich zu machen. Daraus entstand die Idee, Schulungen als interaktive Kurzveranstaltungen online anzubieten.

Ziel

Die Webinare* der Spitalhygiene haben zum Ziel, neue Informationen zeitnah und verständlich an möglichst viele Mitarbeitende anzubieten. Das Onlineangebot reduziert den zeitlichen und organisatorischen Aufwand für Vortragende und Teilnehmende.

Methode

Wichtige Informationen werden in einem kurzen PowerPoint gestützten Vortrag online den Mitarbeitenden der Inselgruppe drei bis vier Mal an verschiedenen Wochentagen und zu unterschiedlichen Zeiten angeboten. Die Webinare dauern dabei jeweils max.15 Minuten, damit Mitarbeitende auch bei hoher Arbeitsbelastung die Information abholen können. Über die Rubrik «Veranstaltungen» im Intranet sowie den Verteiler des Hygieneordners werden die Einladungen inkl. Zugangslink versendet. Die Präsentation wird direkt am Computer gehalten. Dadurch entfallen für die Mitarbeitenden der Spitalhygiene die Wegzeiten an die Standorte und organisatorische Abläufe wie z.B. die Raumorganisation werden erleichtert. Die Teilnehmenden können ohne grossen Zeitaufwand direkt an ihrem Computer alleine oder im Team teilnehmen.

Schlussfolgerung

Das Angebot der Webinare wird rege genutzt. Die Anzahl der Teilnehmenden schwankt je nach Thema zwischen zehn bis vierzig Mitarbeitende. Für die Spitalhygiene wird insbesondere die Organisation erleichtert. Schulungen können kurzfristig mit geringen zeitlichen und personellen Ressourcen angeboten und Informationen an die Mitarbeitenden der sechs Standorte gleichzeitig vermittelt werden. Die Spitalhygiene hat das Ziel, monatlich mindestens ein Webinar zu einem hygienerelevanten Thema anzubieten.

* Bei einem Webinar wird eine Präsentation, ein Vortrag oder eine Schulung mit Hilfe von Videotechnik über das Internet übertragen.

Aim
Chimeric antigen receptor-modified T (CAR T) cell immunotherapy for B cell malignancies results in prolonged B cell depletion. The aim of this work was to investigate pathogen specific humoral immunity after CAR T cell therapy.

Methods
We conducted a cross-sectional study of children and adults treated with CD19- or BCMA-CAR T cell therapy. Eligible patients were ≥ 6 months post-CAR T cell infusion and in remission without subsequent chemoimmunotherapy. We measured total immunoglobulin G (IgG), pathogen-specific IgG levels for 12 vaccine-preventable infections, and B cell subsets from blood. Seroprotective antibody titers were based on standard thresholds. We described the proportion of patients with seroprotective titers and tested for associations between clinical factors and seroprotection using generalized estimating equations.

Results
65 patients who received CD19- (n=54) or BCMA- (n=11) CAR T cell therapy were enrolled. Seven patients were < 18 years old. Samples were collected a median of 20 months (range, 7-68) after CAR T cell infusion. Seroprotection to vaccine-preventable pathogens was generally comparable to the U.S. population even though blood CD19+ B cell counts were low (< 20 cells/mm3) in 60% of patients. Among 30 patients without IgG replacement in the prior 16 weeks, 27 (90%) had hypogammaglobulinemia. Despite this, these individuals had seroprotection to a median of 67% (IQR, 59%-73%) of tested pathogens. The proportion of patients with seroprotection was lowest for mumps, hepatitis A and B, H. influenzae type B (Hib), S. pneumoniae, and B. pertussis. Patients receiving BCMA-CAR T cells had seroprotection to fewer pathogens than those receiving CD19-CAR T cells, but the difference did not reach statistical significance. There were no significant differences by other variables.

Conclusion
Seroprotection for vaccine-preventable infections after CD19-CAR T cell therapy was comparable to the general population. BCMA-CAR T cell recipients seem to have seroprotection to fewer pathogens.

Aim:
COVID-19 displays diverse disease severities and symptoms including acute systemic inflammation and hypercytokinemia, with subsequent dysregulation of immune cells. Bacterial superinfections in COVID-19 can further complicate the disease course and are associated with increased mortality. However, there is limited understanding of how SARS-CoV-2 pathogenesis and hypercytokinemia impede the innate immune function against bacterial superinfections.

Method:
We assessed the influence of COVID-19 plasma hypercytokinemia on the functional responses of myeloid immune cells upon bacterial challenges from acute-phase COVID-19 patients and their corresponding recovery-phase. Additionally, we assessed the impact of the dysregulated cytokine profile on the regulated cell death (RCD) programme of neutrophils.

Results:
We show that a severe hypercytokinemia status in COVID-19 patients correlates with the development of bacterial superinfections. Neutrophils and monocytes derived from COVID-19 patients in their acute-phase showed an impaired intracellular microbicidal capacity upon bacterial challenges. The impaired microbicidal capacity was reflected by abrogated MPO and reduced NETs production in neutrophils along with reduced ROS production in both neutrophils and monocytes. Moreover, we observed a distinct pattern of cell surface receptor expression on both neutrophils and monocytes, in line with suppressed autocrine and paracrine cytokine signaling. Aditionally, COVID-19 plasma induced a necroptosis-sensitive neutrophil phenotype, characterised by cell lysis, elevated release of DAMPS, RIPK1 levels and MLKL involvement. Furthermore, the impaired antibacterial effector function was mediated by synergistic effect of the cytokines TNF-α, IFN-γ and IL-4. COVID-19 patients receiving dexamethasone showed a significant reduction of overall inflammatory markers in the plasma as well as exhibited an enhanced immune response towards bacterial challenge ex vivo. Finally, broad anti-inflammatory treatment was associated with a reduction in CRP, IL-6 levels as well as length of ICU stay and ventilation-days in critically ill COVID-19 patients.

Conclusion:
Our data provides insights into the transient functional dysregulation of myeloid immune cells against subsequent bacterial infections in COVID-19 patients and describe a beneficial role for the use of dexamethasone in these patients.

Aims
Hepatitis B virus (HBV) infection is a major cause of morbidity and mortality among people living with HIV. Tenofovir disoproxil fumarate (TDF) or tenofovir alafenamide fumarate (TAF) suppress HBV viral load in the majority of patients, but HBsAg loss (functional HBV cure) is a rare event and its determinants are not well understood. We aimed to assess HBsAg loss and related risk factors in people living with HIV/HBV in the Swiss HIV Cohort Study (SHCS).
Methods
We included all participants with two positive HBsAg tests at least 180 days apart treated with TDF or TAF. We calculated the proportion of participants with HBsAg loss and estimated logit-transformed 95% confidence intervals (CI). We defined HBsAg loss as a negative HBsAg test without a positive test further on. We compared demographical, clinical and laboratory characteristics at start of treatment with TDF or TAF (baseline) of participants with and without HBsAg loss using descriptive statistics and logistic regression. For the multivariable model, we included all variables with a p-value < 0.1 in univariable analyses. In a small subset of five participants with and five without HBsAg loss, we performed serial assessments of quantitative HBsAg (qHBsAg) levels.
Results
We included 323 participants with an HBsAg test during follow-up. 74 (22.9%) were female, 195 (60.4%) were of European origin and median age was 41 years (inter quartile range (IQR) 35 – 46). The participants were followed for a median of 9.5 years (IQR 6.1 – 12.9). 43 participants (13.3%, 95% CI 10.0 – 17.5) achieved HBsAg loss. 234 participants had available qHBsAg measurements and were included in multivariable analysis. In multivariable analysis, only qHBsAg (as log10IU/ml, odds ratio (OR) 0.57, CI 0.43 – 0.74) was significantly associated with functional HBV cure. In the subset of 10 participants with longitudinal qHBsAg measurements, the participants with HBsAg loss had a median decline in qHBsAg levels of -0.35 log10IU/ml (IQR -1.87 – -0.18) compared to -0.05 (IQR -0.06 – -0.03) in those without HBsAg loss during the first year (p = 0.25). During the first two years, the median decline in qHBsAg levels was -1.24 log10IU/ml (IQR -2.00 – -0.53) in those with HBsAg loss compared to -0.15 (IQR -0.33 – -0.07) in those without (p = 0.25).
Conclusion
Functional HBV cure during tenofovir therapy occurred in 13% of persons living with HIV/HBV. Lower qHBsAg levels at tenofovir start strongly predicted functional HBV cure.

Aims: Human Pegivirus-1 (HPgV-1) is a so-called commensal virus [1] for which no known associated organ disease has been found to date. Yet, its immunomodulatory role was studied among HIV population, in whom active co-infection with HPgV-1 can provide a protective effect against evolution to AIDS and mortality [2]. Little is known on the immunomodulatory effect of HPgV-1 in allogeneic stem cell transplant (allo-HSCT) recipients. The aim of this study was to compare the immune reconstitution after allo-HSCT among HPgV-1-viremic and HPgV-1-non-viremic patients.
Methods: Within a cohort study of allo-HSCT patients, 20 allo-HSCT recipients positive in plasma sample for HPgV-1 by rRT-PCR at predefined timepoints (30 days, 3, 6 and/or 12 months) after transplantation were matched with 20 allo-HSCT recipients negative for HPgV-1 in plasma at the same timepoints. Peripheral blood samples were obtained from these allo-HSCT recipients at the same timepoints and immune reconstitution was monitored by extracellular and intracellular flow cytometry to assess any difference between both groups. 24 healthy blood donors were included as controls.
Results: There was no significant difference in the absolute number and subsets proportion of CD4 and CD8 T cells between both groups at any analysed timepoint. Nevertheless, there was a significantly higher absolute number of NK cells at 3 months among HPgV-1-viremic patients (median 230 cells/µl vs 94 cells/µl, p = 0.022). Analysis of NK cells subsets showed a significantly higher proportion of NK bright among HPgV-1-positive patients at day 30 (median 54% vs 39%, p = 0.032) 3 months (median 58% vs 40%, p = 0.038) and 6 months (median 60% vs 27%, p = 0.024). Analysis of NK dim proportion showed the inverse relationship. At 6 months post-HSCT, proportion of NK cells expressing CD57 antigen was also lower among HPgV-1-viremic patients (median 26% vs 45%, p = 0.0068). In accordance with their less differentiated phenotype, we detected a significantly reduced expression of granzyme B in NK cells in HPgV-1-viremic patients at 6 months (median 58% vs 76 %, p = 0.0045).
Conclusion: We found a significant different NK cell, but not T cell, immune-reconstitution after allo-HSCT between HPgV-1-viremic and HPgV-1-non-viremic patients. HPgV-1-viremic patients exhibited a less differentiated NK cells profile than HPgV-1-non-viremic patients during the first months after transplantation.

Background: The optimal duration of antibiotic treatment in spinal infections, with or without an infected implant, is unknown. We display the results of the second interim analysis of the prospective randomized multicenter trial (SASI).

Methods: A multicenter prospective cohort of 236 spine infections undergoing surgical treatment is included into the SASI trial. A 1:1 randomization allocated patients into a short or long postoperative antibiotic treatment group. The short and long treatment group were defined as 3 vs. 6 weeks in absence of a residual infected implant and 6 vs. 12 weeks in case of remaining implants. Non-inferiority was defined at 10%. For the interim analysis a minimal follow-up of 1 year was defined to investigate two distinct outcomes: "clinical failure" (reoperation due to known, or new infections or mechanical problems) and "microbiological recurrence" (microbiologically-identical relapses with the same pathogen after therapy). 96 (56%) Patient could be included in this interim analysis.

Results: Clinical failures were found in four (4/96; 4%), of which two were true "microbiological recurrence" due to Staphylococcus aureus; occurring in the same patient; and equally distributed in the short and long arms. Two other clinical failures are a secondary endocarditis during treatment and clinical failure without the proof of recurrent pathogens, also distributed equally in both groups. Adverse events of antibiotic therapy were equally distributed between the treatment groups: Among all 59 single adverse events, 36 (61%) yielded a link to the antibiotic treatment. In the Per-Protocol analysis, only the outcome “microbiological recurrence” fulfills the statistical non-inferiority (5.6% percentage points [90%CI 9.2% to 9.8%]).

Conclusion: In this second interim analysis, a shorter period of antibiotic treatment was noninferior to longer treatment duration in terms of remission of spine infections after surgical debridement, also with no difference of adverse events.

Background
Monitoring of trough plasma concentrations may guide antifungal (AF) therapy. However, data are lacking about the actual bioavailability of the drug at the site of infection in real clinical conditions. The objective of this study was to measure the AF concentrations in infected body fluids or tissues in comparison to simultaneously measured plasma concentrations.
Methods
Patients with a proven or suspected invasive fungal infection were eligible if they underwent a surgical or interventional procedure at the site of infection while antifungal therapy with a triazole or an echinocandin was ongoing for ≥48h. AF concentrations were measured by ultra-performance liquid chromatography-tandem mass spectrometry in the infected tissue or fluid (= infection site, IS) and in two plasma samples drawn at the time of the intervention and just before the next dose (trough level).
Results
21 patients were enrolled: 16 cases of invasive candidiasis treated by caspofungin (n=12), anidulafungin (n=1) or fluconazole (n=3), and 5 cases of invasive aspergillosis treated by voriconazole (n=3) or isavuconazole (n=2). Interventions and measurements were performed at a median of 13.3 days (range 2-63) from the start of antifungal therapy. Sites of infection were intra-abdominal collections (n=10), hepatic abscesses (n=3), pleural fluids or fragments (n=5), pulmonary nodules or masses (n=3), and cerebral abscess (n=1). AF trough concentrations were all within the expected therapeutic range. Echinocandins were not detected in liver abscesses and lung lesion. However, the median IS/plasma ratio in intra-abdominal or pleural collections was 0.24 (0.11-0.75) with concentrations at the infected site that were equal or above the minimum inhibitory concentration (MIC) of Candida spp (MIC90: 0.12-0.25 mg/l). Fluconazole achieved similar concentrations in abdominal or pleural collections compared to plasma. IS/plasma ratios for voriconazole and isavuconazole were >1 in lung and brain lesions, but undetectable in pleural fluid.
Conclusion
This pilot study of AF drug concentrations in infected human tissues or fluids showed acceptable penetration of echinocandins in abdominal or pleural collections with the exception of hepatic abscesses and lung lesion. Penetration of azoles in deep tissues was excellent (IS/plasma ratios ≥1), with the exception of voriconazole in pleural fluid.

In response to globally increasing antiretroviral resistance, the WHO recommends dolutegravir (DTG) based antiretroviral therapy (ART). DTG resistance is rare, even in people failing on a DTG-based regimen. However, (functional) DTG monotherapy and pre-existing InSTI resistance increase the risk of resistance. We aim to analyse risk factors for DTG resistance and the prevalence of InSTI drug resistance mutations (DRMs) in people living with HIV with a genotypic resistance test (GRT) during a DTG-based regimen.
We pooled data from 4 European and one Canadian cohort (AHS, ATHENA, CBC, ICONA, SHCS) and the South African AID for Aids (AfA) program. We identified drug resistance levels to DTG and DRMs using the Stanford resistance algorithm. Using ordinal logistic regression, we assessed the association of DTG resistance with covariables chosen based on case reports and clinical relevance, particularly prior InSTI exposure, HIV subtype, HIV viral load, ART regimen, and resistance to reverse transcriptase inhibitors (RTIs).
We identified 372 patients with a GRT covering integrase and sampled during DTG-based ART; 271/372 patients were from European cohorts. Resistance levels were as follows: susceptible, 349 (94 %); potential low, 3; low, 5; intermediate, 10; high, 5. At least low level DTG resistance was found in 5 / 128 (3.9 %) patients with VL < 400 copies / ml, 12 / 210 (5.7 %) with VL > 400 copies / ml, and 9 / 174 (5.2 %) with VL > 1000 copies / ml (thresholds commonly used in resource-limited countries). We found a strong association of resistance against RTIs with DTG resistance (adjusted OR, 47. 5; 95 % CI, 9.9 - 226.8). Furthermore, patients on a DTG dual- or monotherapy were at increased risk for DTG resistance (adjusted OR, 6.4; 95 % CI, 2.5 - 16.5). The most frequently observed DRMs occurred on integrase position 140 (8 patients), 155 (7 patients), and 263 (6 patients).
DTG resistance is rare, even in patients failing on a DTG regimen. In the coming years with an increasing number of people on DTG-based regimens, especially in low- and middle-income countries where pre-existing resistances are more common and people are often switched with unknown VL, careful prospective surveillance is needed to determine the effect of functional DTG monotherapy. DTG DRM accumulation may follow different mutational pathways in InSTI naïve patients. Mutational patterns affecting DTG resistance may differ from known patterns and require further investigation.

Aims.
Standard diagnostics test for urethritis or proctitis, when a sexually transmitted infection (STI) is suspected, is a polymerase chain reaction (PCR) for N. gonorrhea (NG) and C. trachomatis (CT) and, in case of a negative result a further PCR test for Mycoplasma genitalium (MG) 1. According to current guidelines, first-line treatment for proven infection with NG is ceftriaxone and doxycycline for CT infections 2-4. As the time to receive the results of the PCR varies, so-called blind treatments are often done due to urgent request from symptomatic patients, in which suspected NG and/or CT infections are treated before test results are available. The aim of our study is to collect data from the largest STI clinic in Switzerland regarding the number of blind treatments administered before and after on-site PCR testing was implemented in the clinic.

Methods. Until the end of 2020, swab samples from patients with symptoms of urethritis and/or proctitis of Checkpoint Zurich were sent to an external laboratory for PCR testing for NG and CT, with test results available 24 to 48 hours later. In case of a negative result, the laboratory needed to be contacted to perform a further PCR test on MG.
In 2021 the process was changed by using the GeneXpert Dx system GX-IV from Cepheid 5. Samples were analyzed on-site by rtPCR (Xpert CT/NG assay) 6. In this setting a test for NG/CT can be completed within 90 minutes . In case of a negative result, a further PCR test on MG (SpeeDx ResistancePlus® MG-Assay) was performed from the same sample within the same day. Treatment documentation and swab samples between the 1st of January 2020 and 31thOf December 2021 were evaluated retrospectively for this study.

Results. In 2020, a total of 1538 infections with either CT, NG and MG were detected at the clinic. One hundred-and-nine (7.1%) received a blind treatment prior to receiving the results of the swab specimens. In 2021, 2209 infections with either CT, NG and MG were detected. Only 35 (1.6%) of these received a blind treatment in advance. The rate of blind treatments could be reduced by a factor of 4.4 between 2020 and 2021 (p < 0.001).

Conclusions. By introducing on-site PCR testing in our clinic, with results for NG, CT and MG usually available on the same day, the number of blind treatments with ceftriaxone and doxycycline was significantly and substantially reduced.

Aims
Within a prospective healthcare worker (HCW) cohort, we determined the risk and severity of SARS-CoV-2 (re)-infection during the Delta- and Omicron dominating waves, depending on baseline immune status and booster vaccination.

Methods
We determined immune status based on vaccination/infection history and serology results as of September 2021: Group N (reference): No reported infection, anti-N/-S negative, no previous SARS-CoV-2 vaccination; Group V (vaccinated): no reported infection and anti-N negative, twice vaccinated; Group I (infected): infection reported or anti-N positive, no vaccination; Group H (hybrid immunity): reported infection or anti-N positive, vaccination (≥1 dose). Participants were followed until March 2022; date/symptoms of SARS-CoV-2 (re-)infections and receipt of booster vaccinations (for groups V and H) were recorded. The period before December 27th 2021 was defined as Delta, the period after this date as Omicron-dominant. Cox-regression analysis was performed to compare the main outcome “time to positive swab” according to immune status and time period, adjusted for multiple confounders. Poisson regression was used to assess the impact of immune status/booster vaccination on number of symptoms (i.e. disease severity) reported.

Results
Among 2451 participants (median follow-up 171 days), we observed 705 (29%) (re-)infections, thereof 541 during the Omicron period. Overall, the hazard ratio (HR) for (re-)infection was 0.6 (95% confidence interval (CI) 0.4-0.7) for V (number of HCW=1561), 0.5 (95% CI 0.3-0.7) for I (n=184), and 0.3 (95% CI 0.2-0.4) for H (n=519) compared to group N (n=188). This effect was mainly driven by the Delta period and decreased during Omicron for all groups. This was particularly true for group V, which showed similar infection risk during Omicron as group N in adjusted analysis (adjusted HR 1.0, 95% CI 0.7-1.5). Booster vaccination reduced the (re-)infection risk in groups V (aHR 0.7, 95% CI 0.5-0.9) and H (aHR 0.5, 95% CI 0.3-0.8). Group H (vs. N, risk ratio (RR) 0.7, 95% CI 0.5-0.9) and those with booster vaccination (vs. non-boosted, RR 0.8, 95% CI 0.7-0.9) reported less symptoms during (re-)infection.

Conclusion
Hybrid immunity and booster vaccination reduce the risk and severity of SARS-CoV-2 (re-)infection during Delta- and Omicron-dominant periods. For non-infected individuals with two vaccinations, booster vaccination is necessary to maintain the protective effect during the Omicron-dominan

Aims
The growing crisis of antimicrobial resistance (AMR) in Gram-negative bacteria is being accelerated by plasmid-mediated AMR dissemination. Thus, understanding and tracking plasmid dissemination has become increasingly necessary in hospital settings.

Methods
In our institution, a carbapenemase-producing blaOXA-48 Citrobacter freundii was isolated in blood and urine cultures of a patient. A rectal swab revealed the carriage of three blaOXA-48 producing Enterobacterales: C. freundii, Escherichia coli and Klebsiella pneumoniae. Five bacterial isolates were obtained, identified (MALDI-TOF) and tested blaOXA-48 positive by Amplex. Each strain was cultured, total DNA was extracted, and sequenced by long-read Nanopore MinION and short-read Illumina technologies. Long-read data was obtained using R10.3 flow cells and basecalled using high accuracy model implemented in Guppy software. The assembly was performed with flye software and corrected with illumina short reads. The assembled sequences were annotated to include sequence types, resistance genes and plasmid type determination.

Results
The flye assembly produced closed chromosomes and plasmids for the five strains. The three C. freundii strains contained two plasmids, E. coli contained four plasmids and K. pneumoniae six plasmids. Nanopore long-read sequencing coverage was high, ranging between 56 and 102x. All strains contained a 74.3 Kb IncM1 plasmid harboring blaOXA-48 within a tn1999.1 insertion sequence, inverted in the K. pneumoniae IncM1 plasmid. Otherwise, IncM1 plasmids were 99% identical. Besides blaOXA-48, all strains carry blaCTX-M-14b resistance gene.

Conclusions
Nanopore long-read technology allows for the closure of full bacterial chromosomes and plasmids, which is essential for refined analysis. Additionally, it resolved high complexity regions often containing repetitive DNA and structural variations associated with mobile elements, transposons, and gene duplications. In this way, we could detect an inversion of tn1999-pOXA-48 in the K. pneumoniae plasmid. Such transposition events have been shown to happen at high frequency for this IS. Our results show that the blaOXA-48 can be carried by IncM1 plasmids, while it is classically known to be associated with IncL plasmids. Our analysis strongly suggests within-patient dissemination of the IncM1 plasmid in the gut microbiota across the three Enterobacterales species.

Background: Patients with hematological neoplasias such as multiple myeloma (MM) have impaired immune responses after primary 2-dose immunisation with mRNA SARS-CoV-2 vaccines. We studied humoral and cellular immunity and efficacy of a booster dose in MM patients with low antibody titers after primary immunisation.

Materials: MM patients and healthy age-matched controls were enrolled after they had received two doses of a SARS-CoV-2 mRNA vaccine. Persons with prior SARS-CoV-2 infection were excluded. SARS-CoV-2 anti-spike (S) antibodies were measured at 3 (t1) and 6 (t2) months after second vaccination and at 1 month (t3) after booster dose 6-7 months after second vaccination by electro-chemiluminescence-immunoassay (ECLIA; Roche). T cell response was measured as IFN-γ release after incubation with SARS-CoV-2 spike peptide pools using enzyme-linked immunosorbent spot (ELISpot, Mabtech) assay.

Result: mRNA-1273 vaccination (n=13) resulted in higher anti-S titers (1235 U/l vs. 166 U/l, p=0.001) than BNT162b2 (n=59) in MM patients. Subsequent analysis was therefore restricted to those 59 MM patients (34% female, 24% "≥ 75 years") vaccinated with BNT162b2. Those had lower median anti-S concentrations (at t1: 166 U/L vs. 929 U/l, "p < 0.001") than controls. T cell response and anti-S levels correlated weakly (at t1 r=0.28, p=0.02) in MM patients. MM patients with "anti-S < 250 U/l" had lower T cell response than those with "anti-S ≥ 250 U/l" (p=0.004) but overall T cell responses were similar between MM patients and controls (p=0.31). After booster vaccination (given to 25 patients with t1 "anti-S < 250 U/l"), t3 anti-S significantly increased ("p < 0.001") and were higher than t2 anti-S levels for those not initially qualifying for a booster dose (i.e. t1 "anti-S ≥ 250 U/l") (p=0.001). Of 5 patients without seroconversion at t1, 4 seroconverted after the booster; the single persistently seronegative MM patient was the only patient without T cell response at t1 and t2.

Conclusion: Compared to controls, humoral response was impaired in MM patients. While T cell response to mRNA SARS-CoV-2 vaccines was preserved in patients with MM, it was reduced in those with lower anti-S levels. A booster dose resulted in seroconversion in most initially seronegative vaccinees, but impaired T cell response might predict failure.

AIMS Comparative safety data for transcatheter (TAVR) and surgical aortic valve replacement (SAVR) in patients with aortic valve stenosis are limited. Therefore, we aimed to compare rates of infective endocarditis (IE) and major adverse cardiac events (MACE) in patients with TAVR, biological (bio), and mechanical (mech) SAVR.
METHODS Population-based cohort study of patients receiving TAVR or SAVR in Switzerland between 2012-2018 using nation-wide samples from the Federal Statistical Office. Three pairwise 1:1 propensity-score-matched cohorts of adults (≥18 years) with symptomatic aortic valve stenosis who received TAVR, bioSAVR or mechSAVR were explored. The pre-specified primary outcome was the incidence of post-surgical IE. Secondary outcomes included MACE (myocardial infarction, hospitalization for heart failure, cardiac arrest, stroke), in-hospital mortality, and length of stay among others. Incidence rates, hazard ratios (HR), and risk ratios with 95% confidence intervals (CI) were calculated in each propensity-score matched cohort.
RESULTS For the final matched analysis 13’825 patients were eligible. Before propensity-score-matching, cumulative incidence of IE per 1’000 PY was highest in the peri-operative phase (≤3 months) for TAVR (42.3), bioSAVR (39.8), and mechSAVR (23.0) and decreased thereafter. After matching, the HR for overall IE was 1.32 (95%CI 0.94 to 1.86) between TAVR and bioSAVR, 1.91 (95%CI 1.04 to 3.50) between TAVR and mechSAVR, and 1.65 (95%CI 1.11 to 2.47) between bioSAVR and mechSAVR. The HR for MACE was 1.59 (95%CI 1.39 to 1.83) between TAVR and bioSAVR, 0.97 (95%CI 0.80 to 1.17) between TAVR and mechSAVR, and 0.77 (95%CI 0.65 to 0.91) between bioSAVR and mechSAVR. TAVR patients had a higher risk for in-hospital mortality for any cause compared with bioSAVR patients (2.8% vs. 1.7%, risk ratio 1.59 [95%CI 1.03 to 2.47]) but were shorter hospitalized during index hospitalization where aortic valve replacement was performed compared with both types of SAVR (10.0 vs. 13.0 days in bioSAVR, -2.95 days [95%CI -3.46 to -2.44] and 11.2 vs. 14.0 days in mechSAVR, -2.81 days [95%CI -3.80 to -1.82]).
CONCLUSIONS In this large cohort study, TAVR and bioSAVR patients had a higher risk of IE compared with mechSAVR, whereas the risk was similar for TAVR and bioSAVR patients. The risk of MACE was higher among patients receiving TAVR or mechSAVR compared with bioSAVR. Hospitalization was shorter in TAVR than SAVR patients.

Aims
Bone and joint infection (BJI) epidemiology and outcomes in solid organ transplant recipients (SOTr) remain largely unknown. We aim to describe BJI in a multi-center cohort of SOTr (Swiss Transplant Cohort Study).

Methods
All consecutive SOTr with BJI (01.05.2008-31.12.2019) were included. A nested case-control study to identify risk factors for BJI was performed.

Results
Among 4,482 patients, 61 SOTr with 82 BJI were included, at an incidence of 1.4% (95%CI 1.1-1.7), higher in heart and kidney-pancreas SOTr (Gray’s test p < 0.01). Timing, BJI type and pathogens involved varied across different SOT categories. Although BJI were predominately late events (median of 18.5 months post-SOT), most infections occurred during the first year post-transplant in thoracic SOTr. Diabetic foot osteomyelitis was the most frequent infection (38/82, 46.3%), followed by non-vertebral osteomyelitis (26/82, 31.7%). Pathogens included Gram-positive cocci (70/131, 53.4%), Gram-negative bacilli (34/131, 26.0%), and fungi (9/131, 6.9%). BJI predictors included male gender (OR 2.94, 95% CI 1.26-6.89) and diabetes (OR 2.97, 95% CI 1.34-6.56). Treatment failure was observed in 25.9% (21/81) patients and 1-year mortality post-BJI diagnosis was 14.8% (9/61).

Conclusion
BJI among SOTr remain a rare event, albeit associated with subtle clinical presentation, high morbidity and relapses, requiring additional studies in the future.

Objectives: Severe COVID-19 is associated with exuberant inflammation, but long COVID patients show signs of immunosuppression. Polymorphonuclear and monocytic myeloid-derived suppressor cells (PMN-MDSCs, M-MDSCs) are immunosuppressive cells rising during inflammation and infections. We aimed to characterize the dynamic of MDSCs in relation with immune parameters in COVID-19 patients followed for 3 months.

Methods: Fifty-six SARS-CoV-2 infected adult patients hospitalized at CHUV were included in the study. Blood was obtained at study inclusion and 3 months later in 21 patients (14 moderate and 7 severe COVID-19), and from 10 healthy controls. Blood was stimulated with TLR ligands (lipopolysaccharide, R848, Pam3CSK4 and CpG). Leukocyte populations and cytokines were analyzed by flow cytometry, mass cytometry, multiplex bead assay and ELISA.

Results: Forty-five patients (80%) had moderate COVID-19, and 11 (20%) severe COVID-19. Two patients died. At hospital admission, PMN-MDSCs and M-MDSCs counts were increased 2-4-fold in COVID-19 patients when compared to healthy controls (P=0.01-0.03). PMN-MDSCs and M-MDSCs counts were higher in severe than in moderate COVID-19 patients (P < 0.005). PMN-MDSCs counts inversely correlated with T cell counts (P=0.025). PMN-MDSCs and M-MDSCs positively correlated with blood levels of epidermal growth factor and hepatocyte growth factor (P=0.01-0.02). M-MDSCs also correlated positively with IL-1β, IL-7, platelet-derived growth factor and vascular endothelial growth factor (P=0.0001-0.03). In whole blood stimulated with TLR ligands, the proportions of TNF and IL-6-producing monocytes and dendritic cells (DCs) were reduced in patients when compared to controls. Three months after COVID-19 diagnosis, leukocyte blood counts were back to normal levels, while the production of cytokines by blood, monocytes and DCs was still largely affected.

Conclusions: PMN-MDSCs and M-MDSCs were elevated and correlated with serum cytokine concentrations and disease severity in COVID-19 patients analyzed at hospitalization. Innate immune blood responses evaluated by cytokine production were impaired in patients, which persisted for up to 3 months. Overall, our results suggest that COVID-19 induces rapid and long-standing innate immune dysregulation.

Acknowledgement: Swiss National Science Foundation, European Union, Société Académique Vaudoise and Porphyrogenis Foundation.

Objectives: Implant-associated infections (IAI) lead to considerable morbidity. Staphylococcus epidermidis (SE) accounts for at least one third of all IAI. Its emerging resistance against common antibiotics is an increasing challenge. To date, very little is known about the consequences of antibiotic resistance of SE on treatment outcome in IAI.
Methods: Retrospective cohort study including patients with IAI due to SE at the University Hospital Basel between 2013 and 2020. IAI with SE was diagnosed if at least two deep tissue samples grew SE with identical antibiotic resistance profiles at a clinically infected site of an implant requiring antibiotic and/or surgical treatment. Treatment failure was defined as unplanned follow-up surgery, infection-related change in antibiotic regimen, or death due to IAI.
Results: Of 1776 patients screened with SE, 319 (18%) had a diagnosis of IAI. Median age was 65 years [IQR 53, 74] and 41% were female (n = 131). Infections were classified as acute in 137 patients (42.9%) and as chronic in 164 (51.4%). Most IAI involved bone and joints (n = 260; 82%), located at the lower extremities (n = 122; 38%). Other infections affected soft tissue (n = 14; 4%), cardiovascular (n = 23; 7%) or neurologic (n = 20; 6%) implants. Two different morphotypes of SE were found in 78 patients. Resistance against methicillin was detected in 309 (78%) isolates, against cotrimoxazole in 140 (35%), ciprofloxacin in 218 (55%), clindamycin in 192 (48%), gentamicin in 186 (47%), daptomycin in 2 (1%), and rifampicin in 45 (11%) – according to corresponding EUCAST interpretation.
Treatment failure occurred in 98 patients (31%). Patients with failure had a higher median Charlson comorbidity index (2 IQR [0, 3]) than those without failure (1 IQR [0, 3]). SE from patients with failure showed higher rates of antibiotic resistance compared to the non-failure group, particularly against cotrimoxazole (49% vs. 29%, p < 0.001), ciprofloxacin (68% vs. 49%, p < 0.001), gentamicin (61% vs. 41%, p < 0.001), and rifampicin (18% vs. 8%, p = 0.003). Moreover, resistance against ≥ 2 oral antibiotics was associated with treatment failure (55% of patients in non-failure vs. 76% in the failure group, p = 0.001).
Conclusions: This is one of the largest studies so far investigating the association of antibiotic resistance of SE on treatment outcome in IAI. These preliminary data suggest an increased risk of treatment failure in patients infected with multi-resistant SE.

Aims. Whole-genome sequencing (WGS) represents the methodology of choice for SARS-CoV-2 lineage classification and characterization in diagnostic laboratories. The rapid, near-full-length sequencing of the RNA virus genome is enabled by high-throughput sequencing of PCR amplicons derived from the amplification of the cDNA molecules. The method is easy to implement, but essentially relies on well-designed primers to produce enough target molecules and well-balanced PCR conditions. To bypass the PCR step, one could use a shotgun metatranscriptomics approach, which, however, would be rather costly and would require rather longer and deeper sequencing run to account for the smaller fraction of viral molecules found in a typical clinical sample.
Methods. In this study, we tested Nanopore, long-read adaptive sampling as a way to bypass the PCR amplification step for SARS-CoV-2 genome sequencing. Adaptive sampling is a Nanopore-specific, software-based approach to enrich or deplete for specific sequences while the sequencing machine operates, i.e. during the translocation of the molecules through the nanopores. The approach hence limits the production of non-target sequences. We tested the approach on clinical samples positive for SARS-CoV-2 during the omicron wave, and for which a whole-genome sequence was available using the conventional, PCR-based approach. We compared different protocols using either Nanopore ligation-based vs. transposase-based (rapid) library preparation strategies and different levels of sample multiplexing per flow cell.
Results. Our results indicate improved target:non-target ratio as compared to shotgun metatranscriptomic approach without adaptive sampling enrichment, leading to up to 50-100x coverage of the genome sequences, thus providing accurate genome sequence reconstruction. We discuss further the applications this technology may have for clinical diagnostic laboratories in terms of turnaround time, costs and accuracy when compared to other currently available sequencing strategies.
Conclusion. The new approach may have applications, well beyond SARS-CoV-2 genome sequencing,
for other DNA or RNA viruses in clinical samples, but also for all kinds of pathogens.

Aims
Covid-19 outbreaks in long-term care facilities (LTCF) are difficult to control and are associ-ated with high case fatality rates for residents. Therefore, effective infection prevention and control strategies are crucial. We describe a Covid-19 outbreak in a 90-bed LTCF in Switzer-land, in which the main intervention was rapid application of the mRNA vaccine booster. At the time the outbreak occurred, B.1.617.2 (delta) variant was the predominant lineage, and a mRNA vaccine booster was not yet recommended by the vaccination commission.

Methods
We rapidly offered and administered a third vaccination (BNT162b2 vaccine booster) as the main intervention at day 6 after the outbreak had started. Cantonal authorities mandated the outbreak investigation. Targeted and repetitive (5-daily) universal screening testing was per-formed for residents and healthcare workers (HCW).

Results
Totally 82 residents lived in the LTCF. There were no known Covid cases before the out-break. The initial case was detected on Nov 2, additionally affected wards with symptomatic cases detected on Nov 4. During the outbreak there were 39/82 (48%) Covid-positive resi-dents. The median age was 84.5 years [IQR 81-90.2], 27 (33%) were male. 79 (96%) had pre-vious Covid vaccination. Of the 39 outbreak cases, 38 (97%) had obtained previous vaccina-tion and 37 (95%) complete baseline vaccinations with two mRNA vaccines. The booster was administered on 8 Nov in 58/82 (71%) residents. 38/43 (88%) of subsequently Covid-negative cases received the booster. Five residents diagnosed between 4 Nov and 18 Nov (including two who received the booster) had fatal courses.
We evaluated the rate of infection per disease-free resident day during the outbreak before (Nov 2 – 20, period I) and at least day 12 after the booster (period II, 21 Nov – 11 Dec, when the outbreak was declared over) with a Poission regression model. The infection incidence rate ratio for period II versus I was 0.04 (95 % CI, 0.00 0.17, p = 0.001).

Conclusion
The administration of a Covid mRNA vaccine booster may play a role in getting a Covid-outbreak (delta variant) in a LTCF under control.

Background
Measuring the appropriateness of antibiotic use is crucial for antibiotic stewardship programs to identify targets for intervention and improvement of antibiotic prescription. The introduction of clinical information systems (CIS) facilitates continuous monitoring of antibiotic use and may allow to measure the appropriateness of antibiotic use.

The aim of this project is to assess the feasibility of converting electronic patient records data into antibiotic stewardship quality indicators suggested by literature.

Methods
In this observational feasibility study, we retrospectively extracted patient-specific electronic data from a single cantonal hospital in Switzerland, starting three months after its introduction of the CIS "Epic". Health records of all patients hospitalized between 01.10.2019 and 30.09.2021 and receiving at least one dose of an antibiotic were included.
We defined the dataset needed for calculation of quality indicators suggested by literature and calculated these indicators. Validity of the measure was assessed and categorized as data complete ("1") or missing data possible due to incomplete documentation ("2 A") or data processing issues ("2 B"), i.e. that some information could not be used anymore after data were processed by the CIS.

Results
In total data of 25'333 hospitalisations from 20'723 individual patients were analysed. Data extraction allowed to calculate 94 % (34/36) of pre-defined quality indicators, including 23 indicators measuring the appropriateness of antibiotic therapy (Table I and II). Data was complete for 75% (27/36) of all quality indicators.
The most important issues according to the data quality observed were: A) missing (58%) or meaningless (30%) information on indication (e.g. general indication, infection), including prophylaxis and B) data processing issues for 11 % (4/36) of all quality indicators (e.g. the categories "change of therapy" and "prescribe therapy again" caused a change in an underlying "order identifier" whereas "prolongation of therapy", the 3rd category available, did not cause a change).

Conclusion
Calculation of quality indicators reflecting the appropriateness of antibiotic prescription out from electronic patient health records was feasible. However, a better data structure and processing within the CIS is crucial for improving the validity of the results.

Aims: Hospital outbreaks of multidrug-resistant Pseudomonas aeruginosa (PA) are often caused by PA clones which produce metallo-β-lactamases, such as Verona integron-encoded metallo-β-lactamase (VIM). Although different sources have been identified, the exact transmission routes often remain unknown. Quantifying the role of different transmission routes of VIM-PA is important for tailoring infection prevention and control measures. The aim of this study is to quantify the relative importance of different transmission routes by applying a mathematical transmission model using admission and discharge dates as well as surveillance culture data of patients.
Methods: We analyzed VIM-PA surveillance data collected between 2010 and 2018 of two intensive care unit (ICU) wards for adult patients of the Erasmus University Medical Center Rotterdam, the Netherlands, using a mathematical transmission model. We distinguished two transmission routes: direct cross-transmission proportional to the fraction of colonized patients in the ward and a route independent of patients in the ward. By exclusion, we assign the latter to persistent environmental contamination. Based on admission, discharge dates, and surveillance cultures, we applied a data-augmented Markov chain Monte Carlo method to estimate the proportion of transmissions assigned to each of the routes.
Results: Our study shows that only 13.7% (95% credibility interval: 1.4%, 29%) of the transmissions that occurred in these two ICU wards were likely caused by cross-transmission. By exclusion, the vast majority of transmissions (86.3%, 95% credibility interval: 71%, 98.6%) due to persistent environmental contamination.
Conclusion: Our results emphasize that persistent contamination of the environment may be an important driver of nosocomial transmissions of VIM-PA in ICUs. To minimize the transmission risk from the environment, potential reservoirs should be regularly and thoroughly cleaned and disinfected, or redesigned.

Aims
We investigated the incremental infection risk for HCW depending on cumulative COVID-19-patient exposure and its modifiability by respirator vs. surgical mask use.

Methods
In this prospective multicentre cohort study, HCW from seven Swiss healthcare networks were followed from 08/2020-09/2021. Data on SARS-CoV-2 exposure, results of SARS-CoV-2 nasopharyngeal swabs, type of mask used during contact with COVID-19 patients outside of aerosol-generating procedures, and potential confounders were obtained. Cumulative patient exposure was assessed by multiplying self-reported number of COVID-19 patient contacts and average contact duration. At baseline, in 01/2021 and 09/2021, participants were screened for anti-nucleocapsid (anti-N) antibodies. Main outcome was SARS-CoV-2 infection, defined as self-reported positive swab and/or anti-N-seroconversion from baseline. Stratified by mask type, odds ratios (ORs) for the increase in SARS-CoV-2 positivity per doubling of contact time were calculated. Independent impact of cumulative COVID-19-patient exposure and mask type on infection was assessed by logistic regression, adjusting for important co-variables and including networks as random effects.

Results
We included 2919 HCW (median age 43 years, range 18-73); 749 (26%) were SARS-CoV-2 infected. SARS-CoV-2 positivity was 13% in HCW without patient exposure. For those exposed to patients, positivity was 21% for respirator and 35% for surgical/mixed mask users, respectively (OR 0.49, 95%CI 0.39-0.61), showing a gradual increase for surgical/mixed mask users (OR 1.21, 95%CI 1.15-1.28) and, less pronounced, for respirator users (OR 1.15, 95%CI 1.05-1.27) across categories of patient exposure. In multivariable analysis, patient exposure (OR 1.2 per quantile, 95%CI 1.2-1.3) and respirator use (OR 0.6, 95%CI 0.4-0.8) remained significantly associated with SARS-CoV-2 infection.

Discussion
SARS-CoV-2 positivity in HCW increased gradually with cumulative COVID-19 patient exposure. This effect was reduced by 40% in those using respirator masks, even after adjusting for multiple work and non-work related co-variables. These data suggest a striking dose-response effect of COVID-19-patient exposure and risk of SARS-CoV-2 infection in HCW, consequent use of respirators might substantially decrease the work-related risk for HCW exposed to COVID-19-patients.

Aim

Mortality in residents of long-term care facilities (LTCF) has been stable prior to the COVID-19 pandemic and is mainly influenced by age and comorbidities. Seasonal influenza is another important influencing factor, with an excess mortality during winter season despite influenza vaccination. As we know from data from the federal office of statistics (BFS), the COVID-19 pandemic has led to an excess mortality of 16% in 2020 compared to 20191 in Swiss LTCF. We aimed to describe all-cause mortality in LTCF during the two Sars-CoV-2 pandemic years by canton (2020 and 2021) and compare them with the 2019 pre-pandemic all-cause mortality.

Methods

This is a secondary analysis of data from an ongoing prospective multicentre cohort of LTCFs (n=59) from cantons of St Gallen (SG) and Vaud (VD) during the COVID-19 pandemic. Participating institutions reported numbers of deaths per year per 100 occupied beds. All-cause mortality was analysed from the start of the pandemic until December 2021 and compared with the pre-pandemic all-cause mortality of 2019. We choose the all-cause mortality instead of the Covid-19 related mortality because of the heterogeneous testing strategies between institutions and cantons. Mortality rates were compared using the quasi-Poisson regression method.

Results

Overall the all-cause mortality of the residents in the 59 institutions of the two cantons was 36.9% in 2020 (VD 39 vs.SG 35.2%) and 26.9% in 2021 (VD 25.8% vs. SG 27.8%) (p-value 0.009). The pre-pandemic all-cause mortality in 2019 was 28.7% (VD 26.5% vs. SG 30.4%) with 2019 vs 2020 a p-value of 0.014 and 2019 vs.2021 a p-value of 0.544. In 2020, 29% more residents died in our current cohort than in 2019. The difference between the two cantons was not statistically significant neither in the pre-pandemic period nor during the pandemic.


Discussion
We confirm data from the BFS, showing an even higher excess all-cause mortality in 2020 compared to 2019, most likely due to COVID-19. Despite ongoing SARS-CoV-2 circulation, the mortality in 2021 returned to pre-pandemic levels. Potential reasons include the absence of seasonal influenza and the high SARS-CoV-2 vaccination coverage in Swiss LTCF residents.

Reference
1https://dam-api.bfs.admin.ch/hub/api/dam/assets/20444121/master

Background
Vancomycin resistant enterococci (VRE) are multidrug-resistant organism (MDRO) responsible of nosocomial outbreaks. Recommendations for preventing their spread include contact precautions. The COVID-19 pandemic caused an overload of hospitals and required a reorganization of health care facilities and systematic contact and droplet precautions implementation in dedicated wards. We describe VRE nosocomial transmissions during the COVID-19 pandemic.

Methods
Lausanne University Hospital is a 1100-bed tertiary care hospital, which followed the national guidelines for VRE control (including a weekly sentinel screening in selected wards, identification and screening of contact patients). One VRE isolate per patient was analysed by whole genome sequencing using wgMLST.

Results
Between December 2020 and June 2021, 11 cases of VRE were identified, among whom five were identified through weekly sentinel screening in intensive care unit (ICU). wgMLST revealed five patients with unique isolates (>200 loci differences) and a cluster of six isolates genetically closely related (0-8 loci differences). Epidemiological investigation allowed identification of two chains of transmission. The first chain occurred in a intermediate care unit (InCU) dedicated to COVID-19 patients and included an index case and two secondary cases identified retrospectively. The second chain included an index case and two contact patients screened positive when readmitted in hospital several weeks later. In the first chain of transmission, the patients in the InCU were not considered as contacts as they were in individual boxes and isolated for COVID-19. Secondary cases were identified later by weekly screening in ICU. Retrospectively, we identified a patient who could be the link between both chains as he was hospitalized in the same units as several
cases of the cluster, but never screened.

Conclusions
In the context of the overload due to the pandemic and the focus on respiratory measures, health care workers may have been less observant with material and environment disinfection, and infection control teams less available to supervise thorough investigations. Our study pinpoint the importance to apply good practices even in an unusual context such as a pandemic in order to limit the risk of transmission of MDRO.

Aim We aimed to investigate the incidence of and risk factors for healthcare-associated infections (HAI) among intensive care unit (ICU) patients with and without COVID-19.
Methods We conducted a single center prospective HAI surveillance from September 1st 2021 until April 30th 2022. We included patients who had an ICU length of stay (LOS) of > 2 calendar days. Electronic medical records were screened three times per week by an infection prevention and control expert; records of patients with HAI suspicion (defined according to ECDC criteria) were reviewed by an infectious disease physician. Baseline characteristics, COVID-19 status at ICU entry, insertion and removal date of venous lines and urinary catheter, mechanical ventilation (MV) and antibiotic treatment, as well as ICU-LOS, and ICU mortality were recorded. After ICU discharge patients were followed concerning HAI (at most for 48h in case of MV and venous lines, 7 days in case of urinary catheters. We performed descriptive statistics and multivariable logistic regression to identify independent risk factors for HAI.
Results We included 168 patients, 60 (35.7%) COVID-19 and 108 (64.3%) non-COVID-19 patients. COVID-19 patients were younger (median 59.5, IQR 49-64.8, vs. 66 years, IQR 57-72, p < 0.001), had similar SAPS II (median 43, IQR 32-57.5, vs. 44, IQR 34-54.75, p = 0.89), and a longer ICU-LOS (median 12, IQR 6-22.8, vs. 5 days, IQR 4-8, p < 0.001) compared to non-COVID-19 patients. We recorded 63 HAI in 43 patients (46 (73%) in COVID-19 and 17 (27%) in non-COVID-19 patients). 22 (34.9%) ventilator-associated pneumonia (VAP) and 15 (23.8%) bloodstream infections (BSI). HAI incidence per 1000 patient-days was 50.2 in COVID-19 and 20.1 in non-COVID-19 patients (risk ratio 2.5, 95% CI 1.5-4.5). This result was mainly due to different VAP rates (26.9 vs. 2.1/1000 ventilator days, p < 0.00), whereas BSI rates were similar in both groups (9.8 vs. 7.1/1000 patient days, p = 0.55). There was no effect on ICU mortality (20% vs. 14.8%, p = 0.4). In the multivariate analysis, positive COVID-19 status (OR 3.1, 95% CI 1.1-9.0) and ICU-LOS (OR per additional day 1.2, 95% CI 1.1-1.3) were associated with HAI, whereas age and MV were not.
Conclusion Incidence of VAP, the most common HAI in our analysis, was considerably higher in COVID-19 compared to non-COVID-19 patients, whereas BSI rates were similar. COVID-19 status was independently associated with HAI occurrence, even after adjusting for ICU-LOS and MV.

Aims
Since the emergence of SARS-CoV-2, healthcare systems face the challenge of preventing nosocomial transmissions. We sought to decipher transmission pathways in all healthcare-associated infections with SARS-CoV-2 by thorough epidemiological work-up and complementary next generation sequencing (NGS) analysis. We report the findings of the four largest epidemiologic clusters of SARS-CoV-2 transmission, involving 101 patients and healthcare workers (HCWs), occurring during the second wave of the pandemic from 11/2020-12/2020.

Methods
At the University Hospital Basel, Switzerland, systematic outbreak investigation is initiated at detection of any nosocomial case of COVID-19, defined as SARS-CoV-2 infection occurring more than five days after admission. A cluster of nosocomial infections was defined as the detection of at least two positive patients and/or HCW within one week with an epidemiological link. Whole genome sequencing of SARS-CoV-2 was performed to analyze clusters of suspected nosocomial transmission. We categorized members of such clusters into three subgroups: proven transmission (≤ 1 single-nucleotide polymorphism (SNP) difference between strains), exclusion of transmission (>1 SNP difference between strains) and missing sequencing information.

Results
During the study period, four epidemiologic clusters were identified including 40 patients and 61 HCWs. Sequencing data was available for 28 patients and 42 HCWs (70/101, 69.3% of all involved cases) and confirmed epidemiologically suspected transmission in 35 cases (50% of sequenced cases). Among cases with identical strains, epidemiologic work-up suggested transmission mainly among HCWs (15/35, 42.9%), followed by transmission between HCWs and patients (13/35, 37.1%), more commonly affecting patients (9/13, 69.2%) than HCWs (4/13, 30.8%), and between patients (7/35, 20.0%). Sequencing excluded ward-based transmission events despite presence of an epidemiological link in the remaining 35 cases (12 patients and 23 HCWs).

Conclusion
Phylogenetic analyses revealed important insights into transmission pathways supporting only 50% of epidemiologically suspected SARS-CoV-2 transmissions. The remainder of cases most likely reflects community-acquired infection randomly detected by outbreak investigation. Notably, most transmissions occurred between HCWs, possibly indicating lower perception of the risk of infection among HCWs.

Aim: We aim to describe two distinct outbreaks of ESBL-producing Klebsiella sp. occurring in the same paediatric intensive care unit (PICU) within six months: the first outbreak was detected after identification of K. michiganensis in clinical swabs from two neonates and the second outbreak subsequently to a case colonised with ESBL-producing K. pneumonia.
Methods: During both outbreaks, we performed weekly screenings for ESBL-producing Enterobacteriaceae by using rectal swabs of all patients. Concomitantly we analysed several environmental swabs. Whole genome sequencing (WGS, Illumina) was performed on Klebsiella isolates from both outbreaks, using Ridom Seqphere+ to define MLST and core genome MLST (cgMLST) relatedness. Selected strains were additionally sequenced using Oxford Nanopore Technology to explore plasmid transmission.
Results: During the first outbreak, we detected a colonisation with K. michiganensis in 16.5% (19/115) of screened patients and in five environmental swabs (three sink drains and fridge) within eight weeks. During the second outbreak K. pneumoniae was detected in 14% (17/121) of screened patients and in three environmental swabs (sink drain, bedside table and CPAP device) within 11 weeks. All clinical and four of five environmental K. michiganensis isolates were sequence type ST-231. All but one clinical and environmental K. pneumoniae isolates were identified as ST-469. Strains of each ST were closely related with 0-4 allelic difference between isolate genomes in the cgMSLT. Despite the occurrence of two outbreaks within the same PICU only six months apart, there was no molecular link between the two outbreaks. Plasmid transmission was also excluded. Of note, all colonisation affected neonates and no infection with Klebsiella spp. occurred. Outbreaks were controlled by improving standard precautions and implementing regular sink cleaning three times daily. One contaminated sink was removed. The second outbreak was declared as resolved after no new colonisations with K. pneumoniae were detected for ten consecutive weeks and three sets of environmental swabs of previously contaminated areas were negative.
Conclusions: We demonstrated that standard epidemiological investigation in conjunction with WGS could confirm two distinct outbreaks with ESBL-producing Klebsiella spp. and helped to implement effective measures in interrupting transmission. Combination with long-read sequencing helped to exclude a plasmid transmission event.

Aims: We report the investigation of an outbreak of COVID-19 associated pulmonary aspergillosis (CAPA) in an adult intensive care unit (ICU) in Switzerland while major construction work was being carried out outside the hospital.

Methods: We defined CAPA as possible, probable or proven according to criteria by Koehler et al (1). CAPA was considered ICU-acquired if the first positive sample was collected more than 48h after ICU admission. We conducted environmental investigations, including air and surface sampling, in ICU bedrooms and preparatory areas.

Results: Between October 2021 and December 2021, we observed 5 probable cases of CAPA acquired in the ICU. The 15 bedrooms of the ICU are equipped with HEPA filter with an air change rate between 8 and 15 per hour. Other areas of the ICU (office, preparation areas) have an air change of 2 per hour. To prevent COVID transmission, settings of the ventilation system had been slightly modified to put ICU bedrooms on a slight depression in March 2020. All air samples performed in ICU patient bedroom were positive for Aspergillus sp (5 UFC / L). Surface sampling were also positive for Aspergillus sp. Since the first wave, ICU healthcare workers (HCW) had taken the habit of frequently opening the windows of the COVID-19 patient rooms to optimise air renewal. Patient rooms were also left open while no HCW are in the room to keep an eye-contact on the patient. From September 2021, major construction work with excavation was carried out outside the hospital and the ICU bedrooms were located directly above the excavation area. Instructions were given to keep windows closed at all times, an extensive environmental cleaning and disinfection procedure with a spore-active product was undertaken and ventilation system was reset to previous settings. Weekly CAPA screening of ICU patients (including biomarkers sampling) was also implemented. A second environmental sampling survey was performed after introduction of these measures and no Aspergillus sp were found in ICU bedrooms. No new CAPA case was detected.

Conclusion: This outbreak highlights the importance of simple infection control measures such as maintaining the windows and doors closed when air renewal is adequate and rooms are HEPA-equipped to prevent environmental airborne transmission, particularly in the context of high spore concentration outside. Environmental sampling can play a didactic role to convince HCW of the environmental risk for the patients.

Aims: During the COVID-19 pandemic SARS-CoV-2 admission screening and repetitive SARS-CoV-2 screening after admission were initiated in order to (1) diagnose COVID-19 in asymptomatic hospitalized patients early, and (2) prevent the spread of SARS COV 2 in the hospital. Here we report how many patients with COVID-19 were diagnosed by admission screening and repetitive screening on or after day 3 of hospitalization in our center.

Methods: In addition to the already established systematic testing for SARS-CoV-2 by nasopharyngeal swab at hospital admission, repetitive screening on day 3 with subsequent swabs every 5 days was started at USZ on December 30, 2021. Additionally, as a standard practice, patients who were exposed to SARS-CoV-2 in the hospital or the community were isolated and regularly tested during the 10 days after exposure. Also, patients were tested if symptomatic. Positive results for SARS CoV2 were defined as cycle threshold values for E-Gene below 35. Nosocomial infections were classified according to the ECDC criteria in “definite- , probable-, indeterminate-, and health-care associated COVID-19”.

Results: From January 1st, 2022 to April 20th, 2022, a total of 749 patients were tested positive for SARS-CoV-2 in our hospital. Of these, 119 (16 %) patients without symptoms of COVID-19 (or with symptoms not attributed to COVID-19) were diagnosed in the admission screening. Another 89 (12%) patients were diagnosed on day 3 or later, 22 (3%) of whom underwent quarantine after being exposed to a SARS-CoV-2-positive index patient. Of the remaining 67 patients, 29 were tested because of symptoms, but 38 (5%) were asymptomatic and only diagnosed due to repetitive screening after admission. Of all patients with COVID-19, 28 (3.7%), 29 (3.9%) and 32 (4.3%) had definite, probable, and indeterminate hospital-associated COVID-19, respectively.

Conclusions: Admission screening and repetitive inpatient screening after admission identified a relevant percentage (16% and 5%) of all COVID-19 cases in our hospital. It allowed early detection of both community- and health-care associated COVID-19 in asymptomatic patients and probably reduced the risk of transmission. Repetitive screening might be particularly important when COVID-19 incidence in the community is high. Further analysis is required to determine whether the approach is cost-effective.

Aims
To assess the effect of the COVID-19 pandemic on the outpatient antibiotic consumption in Switzerland.

Methods
Sales data were used for all Switzerland (J01 ATC code) from IQVIA® for the period 2018-2021 (expressed in defined daily doses per 1000 inhabitants per day (DID)). A generalized regression-based interrupted time series (ITS) model assessed how much the COVID-19 pandemic changed the monthly sales data of antibiotics. The first wave was defined from March to May 2020 and the second wave from October 2020 to April 2021 according to hospitalizations due to COVID-19. We also used prescription data from all consultations with antibiotic treatments reported from 146 practitioners from general medicine participating to the "Sentinella" network during 2018 and 2020 (expressed in number of antibiotic prescriptions per 1000 consultations).

Results
Global antibiotic consumption in Switzerland was 7.4 DID in 2021 (7.6 in 2020, 9.1 in 2019, 9.1 in 2018). The ITS analysis showed the most significant decrease of use for combinations of penicillins and beta-lactamase inhibitors (BLI) (-0.05, p < 0.001) followed by penicillins with extended spectrum (-0.03, p < 0.001), macrolides (- 0.03, p < 0.01) and fluoroquinolones (-0.02, p < 0.0001) during the first COVID-19 wave. The most significant decrease of use during the first wave was observed in the Italian- and French-speaking regions (- 0.25 DID per month, p < 0.001; -0.22 DID per month, p < 0.001, resp) followed by the German-speaking region (- 0.12 DID per month, p < 0.01). No statistically significant change was observed during the second COVID-19 wave. Regarding prescription data, the overall number of antibiotic prescriptions per 1000 consultations between 2018 and 2020 decreased from 28.5 to 23.6 (-16.9%) and for respiratory tract infections from 13.3 to 7.9 (-40.7%).

Conclusion
We observed a significant decrease in outpatient antibiotic use during the first pandemic wave, especially in the Italian- and French-speaking regions. Probably lockdowns measures contributed to this effect by several mechanisms as i) decrease of overall respiratory infections due to hygiene measures and home-office, ii) fear of visiting physicians even for other illnesses, iii) obligation to restrict ambulatory consultations to emergency from March 16 and iv) early hospitalization of COVID-19-infected patients with more severe symptoms [2, 3].