Nosocomial outbreaks Agnes Hajdu EpiTrain III, 24.08.2006 Jurmala, Latvia Content • • • • Nosocomial infections Health care setting Antimicrobial resistance Nosocomial outbreaks – – – – History Characteristics Outbreak database Detection, Investigation • An example: Dent-O-Sept • Summary Nosocomial infection …hospital-acquired infection, health-care associated infection… • Infection acquired in the hospital due to exposure to the pathogen in the hospital • Development of infection after 48 hours of hospital admission (CDC) Burden of nosocomial infections • Increased morbidity, mortality – 10% of in-patients acquire an infection in the hospital • Increased costs – Prolonged hospital stay, additional medical procedures and treatment • 30% preventable Health care setting • • • • Devices: endoscope, catheter, ventilator.. Medical procedures: surgery.. Medical personnel: doctors, nurses.. Patient: immunocompromised, susceptible maybe the source as carrier of pathogen • Dangerous residents: MRSA, VRSA, VRE, ESBL*, C. difficile ribotype 027 *methicillin-/vancomycin-resistant S.aureus; vancomycin-resistant Enterococcus; extended spectrum beta-lactamases Patients at risk • Immunocompromised patient – Malignancy, immunosuppressive treatment, HIV infection • Other factors – Severe underlying disease, age, obesity • Intensive care units – Medical, surgical, neonatal, burn units Antimicrobial resistance 1945 – Penicillin 1948 – Penicillin-resistant S.aureus 1959 – Methicillin 1961 – Methicillin-resistant S.aureus 1998 – Vancomycin-resistant S.aureus • Use, overuse and wrong use of antibiotics • Possible to reverse, but it takes time • Knowledge Attitude Behaviour What can be worse than a nosocomial infection? A Nosocomial Outbreak!!! An unusual increase in the number of nosocomial infections (time, place, person) History of nosocomial outbreaks • First well-documented outbreak – Puerperal (child-bed) fever in a hospital in Vienna, 1847 – Ignác Semmelweis, Hungarian physician gathered and analysed mortality data – Autopsy room Maternity wards – Handwashing intervention (chlorine solution) • Modern epidemiology – S. aureus hospital outbreaks worldwide, 1950s – CDC projects from 1970s – Intensive research from 1990s 1847 2006 Nosocomial outbreaks - examples • Unusual transmission – ESBL Klebsiella pneunomiae – Maternity wards, contaminated ultrasonography gel (France, 1993) • Rare pathogen – Malassezia pachydermatis – Neonatal ICU, associated with colonization of health care workers’ pet dogs (US, 1995) • Emergence of more virulent strain – C. difficile ribotype 027 – Increased severity of diarrhoea, recent outbreaks in US, Canada, Netherlands, England Nosocomial outbreak database • Database providing information to facilitate interventions • A learning tool – What kind of data to collect? Control selection? • Search by pathogen, ward type etc. • Osaka University, Japan • http://health-db.net/infection/top1.htm Characteristics of nosocomial outbreaks • • • • • • Location Type of infection Pathogens Source Mode of transmission Preventive/control measures Gastmeier et al. How Outbreaks Can Contribute to Prevention of Nosocomial Infections: Analysis of 1022 Outbreaks. Infection Control and Hospital Epidemiology; 2005 26(4);357-361 Gastmeier et al. Location • Hospital – 83% – 50% in intensive care units • Outpatient care – 12% • Nursing home – 5% Special problems: – Hospital staff with part-time job in nursing homes (transmissing pathogens in both ways) – Nursing home: no infection control personnel, underreporting of outbreaks, gastroenteritis, scabies Gastmeier et al. Type of infections • • • • • • • • Bloodstream – 37% Gastrointestinal* – 29% Pneumonia – 23% Urinary tract – 14% Surgical site – 12% Other lower respiratory – 10% Central nervous system – 8% Skin and soft tissue – 7% *Probable underreporting Gastmeier et al. Most frequently reported pathogens* Nosocomial infections Nosocomial outbreaks • • • • • • • • • • • • Staphylococcus aureus Enterococci E. coli Pseudomonas aeruginosa Streptococci Enterobacter spp. Staphylococcus aureus Pseudomonas aeruginosa Klebsiella pneumoniae Serratia marcescens Hepatitis B, C virus Legionella pneumophila *Probable underreporting: Salmonella spp., Campylobacter spp., norovirus, rotavirus, respiratory viral infections Nosocomial outbreaks in Norway, 2005 Internet-based outbreak reporting system • 47 outbreaks reported from hospitals with 622 cases Norovirus MRSA Gastroenteritis Listeria Influenza Other Outbreaks 25 11 5 2 1 3 No. of cases 463 41 61 6 22 29 Grahek-Ogden et al. Varsler om mistenkte utbrudd av smittsomme sykdommer I Norge I 2005. MSIS rapport 2006;34:22 (in Norwegian) Gastmeier et al. Source of outbreak • • • • • • • Patient – 26% Medical equipment / device – 12% Environment – 12% Medical personnel – 11% Contaminated drug – 4% Contaminated food – 3% Care equipment – 3% • Unclear source – 37% Gastmeier et al. Mode of transmission • Contact – 45% • Invasive technique – 16% • Inhalation – 15% (droplet, airborne) • Ingestion – 4% • Unclear mode of transmission – 28% Gastmeier et al. Managing hospital outbreaks • • • • • • • • • Patient, health personnel screening, surveillance Isolation, cohorting Handwashing, hand disinfection Sterilisation, disinfection (Change) antibiotic therapy Modification of care / equipment Protective clothing Restriction of work load Vaccination CDC guidelines • Standard Precautions • Contact / Droplet / Airborne Precautions http://www.cdc.gov/ncidod/dhqp/gl_isolation_ standard.html Detection of nosocomial outbreaks • Alert from an effective surveillance system • Alert from – the physician – the nurse – the hospital microbiologist – the hospital epidemiologist Nosocomial transmission? • Similar cases at one department / among similar patients • Cases associated with invasive device • Health personnel and patients with same infection • Nosocomial pathogen Problems with detecting outbreaks • No detection – 2-3 patients with pneumonia in intensive care unit • Detection No investigation – Nursing homes • Detection Investigation No reporting – If sanctions against reporting doctors, nurses • False detection: pseudo-epidemics (artefacts) – E.g. consequent laboratory contamination – May lead to unnecessary antibiotic treatment A method for early detection • Reports from antibiotic susceptibility tests from each medical ward (at least 85% culturing proportion) • Baseline data: frequency of each pathogen isolated from specimens over a 26-week observation period Isolates Weeks 2 1 1 2 3 3 2 4 4… 5… 5 2 1 1 3 22 23 24 25 26 Isolates Weeks 1 1 1 2 1 3 2 4 2… 5… 3 3 4 4 5 22 23 24 25 26 • Threshold: 80% (4/5) – Isolates ranked from the lowest to highest – Divide the distribution into quintiles (fifths), and set the cut-off between the 4th and 5th quintile – The number of isolates represented by the 22nd week item becomes the threshold value Proteus mirabilis isolates 9 8 No. of isolates 7 6 5 4 Threshold (80%) 3 2 Baseline data 1 0 1 3 5 7 9 11 13 15 17 19 21 23 25 Weeks Evaluation of the method • Early warning mechanism for potential outbreaks • Sensitive if organisms are routinely tested • Either unusual and common pathogens • Epidemics involving several locations might go undetected • Minimum effort and time • Hospitals with limited infection control personnel • Establishing endemic nosocomial infection rates • Epidemics with prolonged incubation might go undetected Investigation of nosocomial outbreaks Steps of an outbreak investigation.. • Asset – Diagnosis can usually be made rapidly – Direct access to medical care, laboratory – Patient’s records are available – Easy cohorting of the cases • Disadvantage – Multidrug-resistent pathogens – Complex environment – Intra – interhospital transfer – Temporary staff, working in shifts Case ascertainment • The investigation is dependent on clearly defined case definitions and case ascertainment strategies • Molecular diagnostics – PFGE, PCR.. – Demonstrating clonality among epidemic isolates • Combinative approach – Epidemiological study – Genetic typing method Steps of an outbreak investigation • • • • • • • • Have an outbreak control plan Confirm outbreak diagnosis Define a case Identify cases and obtain information Descriptive data collection and analysis Develop hypothesis Analytical studies to test hypothesis Communication (outbreak report) Control measures www.mcht.nhs.uk/documents/policies/Infection_Control/A12%20-%20Hospital%20Outbreak%20Policy.pdf A Norwegian example: Dent-O-Sept • Antiseptic non-sterile single-use swab for mouth hygiene • Largest Pseudomonas outbreak ever recorded in Norway • 231 confirmed cases, 34 deaths • Genotypically identical strains in production plant, swabs and patients Alert to NIPH • Late February 2002 • Notification from the infection control personnel (not quantified by lab statistics) • Impression of possible increase in the number of pseudomonas infections in clinical wards of Norwegian hospitals (ICU) outbreak investigation Outbreak investigation • Objectives: – Describe the outbreak – Identify the cause – Make recommendations for future prevention • Outbreak Control Team: – Members from NIPH and the hospitals Initial investigation • March 2002: identical outbreak strain is shown in patients in 3 different hospitals • No national surveillance system for P. aeruginosa in Norway • Inquiry for increased awareness in regional centers for infection control • Patient interviews/case notes with trawling questionnaires in the affected hospitals (common exposure?) • Suspect: fluids and moist products samples Environmental investigation • April 2002: laboratory identifies genotypically identical strains of P. aeruginosa in swabs control measures: information to the producer, hospitals, authorities and the public, product recalled, production ceased • Hospitals were asked to report which batches of the product they had in store samples • Inspection and sampling at the production plant – Outbreak strain found in packing machine spraying moisturizing liquid into the bags – Violation of regulations (no documented quality assurance system with microbiological testing..) Case finding and descriptive study • Norway: routine storage of all clinical bacterial isolates from blood and cerebrospinal fluid • Isolates of P. aeruginosa from 1999-2002 to reference labs for genotyping (PFGE) • For patients with P.aeruginosa (outbreak strain or other) questionnaire: – Demographic and clinical data Descriptive results • 231 patients with outbreak stain from 27 health care institutions – – – – Median age: 65 yrs, 61% men 87 pneumonia, 42 sepsis, 70 colonization 31% died (all had severe underlying disease) 31% had not or probably not used the swab Dent-O-Sept outbreak – epidemic curve 50 40 Cases 30 20 10 10 11 12 1 2000 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 2001 Month and year of first positive test 5 6 2002 7 8 9 10 11 12 Analytical study 1. • Case-control study • Case definition: person with the outbreak strain isolated from blood or CSF during Oct 2001-Dec 2002 • Control definition: person with another strain of P. aeruginosa isolated from blood or CSF in the same period • To identify risk factors for having the outbreak strain Results – CC study Risk factor Use of Dent-OSept swab Been on ventilator OR (95% CI) 5.3 (2.0-14) 6.4 (2.3-17) P-value 0.001 <0.001 Patients with the outbreak strain were more likely to be on ventilator / use the swab Analytical study 2. • Cohort study • Including all patients in the CC study • To identify risk factors for a fatal outcome during hospital stay for patients with invasive pseudomonas infection Results – Cohort study Risk factors for dying Use of Dent-O-Sept RR (95% CI) 2.2 (1.4-3.5) Patients with fatal outcome were more likely to be exposed to the swab Confounding? (severely ill patients –> mouth swab instead of tooth brush) Lessons learned • Direct transmission from swabs and indirect transmission through health personnel and contaminated environment • Need to strengthen infection control routines and standard precautions – E.g. for patient with severe underlying disease should only use sterile products • Adherence to regulations in production of medical equipment Oakland Daily Evening Tribune, 06 Nov 1939 Media • The outbreak may cause serious damage to the hospital’s reputation • Tool for education: ’message’ – Public: antibiotics = anti-cold drugs – Professionals: every hospital need a plan for antibiotic use • Increased awareness: time to evaluate routines, existing guidelines (local/national/international level) Summary • Detection – Effective surveillance system, vigilant hospital personnel • Investigation – Skilled hospital infection control practicioner, epidemiologist, microbiologist • Prevention / Control – Appropriate infection control practices – Strategies to prevent and control antimicrobialresistent pathogens (antibiotic-plan) The ultimate goal: patient safety