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.aureus1959 – Methicillin1961 – Methicillin-resistant S.aureus1998 – 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

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• Staphylococcus aureus

• Enterococci

• E. coli

• Pseudomonas aeruginosa

• Streptococci

• Enterobacter spp.

Nosocomial outbreaks• 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

Gastmeier et al.

Nosocomial outbreaks in Norway, 2005

Internet-based outbreak reporting system

• 47 outbreaks reported from hospitals with 622 cases

Outbreaks No. of cases

Norovirus 25 463

MRSA 11 41

Gastroenteritis 5 61

Listeria 2 6

Influenza 1 22

Other 3 29

Grahek-Ogden et al. Varsler om mistenkte utbrudd av smittsomme sykdommer I Norge I 2005. MSIS rapport 2006;34:22 (in Norwegian)

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

Gastmeier et al.

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

• Threshold: – 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

80% (4/5)

2 1 3 2 4 … 5 2 1 1 31 2 3 4 5 … 22 23 24 25 26

1 1 1 2 2 … 3 3 4 4 51 2 3 4 5 … 22 23 24 25 26

IsolatesWeeks

IsolatesWeeks

Proteus mirabilis isolates

Threshold (80%)

Baseline data

0

1

2

3

4

5

6

7

8

9

1 3 5 7 9 11 13 15 17 19 21 23 25

Weeks

No

. of

iso

late

s

Evaluation of the method

• Early warning mechanism for potential outbreaks

• Either unusual and common pathogens

• Minimum effort and time

• Hospitals with limited infection control personnel

• Establishing endemic nosocomial infection rates

• Sensitive if organisms are routinely tested

• Epidemics involving

several locations might go undetected

• Epidemics with prolonged incubation might go undetected

Investigation of nosocomial outbreaks

• 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

Steps of an outbreak investigation..

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)

Controlmeasures

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

Month and year of first positive test

12

2001 2002

Cases

8 9 10 114 5 6 712 1 2 38 9 10 114 5 6 7

2000

1 2 3

50

10 11 12

10

20

30

40

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 OR

(95% CI)

P-value

Use of Dent-O-Sept swab

5.3

(2.0-14)

0.001

Been on ventilator

6.4

(2.3-17)

<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

RR

(95% CI)

Use of Dent-O-Sept 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 antimicrobial-

resistent pathogens (antibiotic-plan)

The ultimate goal: patient safety