Transcript Slide 1

TB Infection Control:
Principles, Pitfalls, and Priorities
Kevin P. Fennelly, MD, MPH
Interim Director
Division of Pulmonary & Critical Care Medicine
Center for Emerging & Re-emerging Pathogens
UMDNJ-New Jersey Medical School
[email protected]
Objectives
1. To review basic principles underlying TB
transmission and TB Infection Control
policies.
2. To review the recent history of TB Infection
Control.
3. To discuss personal observations and offer
practical solutions to common problems in
TB Infection Control.
Is TB an Occupational Disease of HCWs?
LTBI
Low- & middleincome
countries
63% (33-79%)
High-income
countries
24% (4-46%)
5.8% (0-11%)
1.1% (0.2-12%)
??
1.18 (1.04-1.35)
3.04 (1.62-5.19)
(prevalence)
TB disease
(annual incidence)
TB mortality (inpt)
(PMR)
(outpt)
- Menzies D et al. IJTLD 2007; 11:593
HCW Deaths due to
Nosocomial Transmission of DR-TB
• MDR outbreaks U.S. 1980s-1990s
– 9 HCWs died
• All immunocompromised, 8 with HIV
– Sepkowitz KA, EID 2005
• XDR-TB outbreak, So Africa, 2006
– 52/53 died of unrecognized XDR-TB
• 44/44 tested were HIV+
• Median survival from sputa collection=16 days
• 2 HCWs died; 4 others sought care elsewhere
– Gandhi N, Lancet 2006
Personal Respiratory Protection Against
M. tuberculosis: Contentious Controversy
from Sol Permutt, 2004
Wells-Riley Equation:
Mathematical model of airborne infection
Pr{infection}=C/S=1-e(-Iqpt/Q)
Where
C=# S infected
S=# susceptibles exposed
I = # infectors (# active pulm TB cases)
q = # infectious units produced/hr/Infector
p = pulm ventilation rate/hr/S
t = hours of exposure
Q = room ventilation rate with fresh air
Control Measures are Synergistic & Complementary
Assumptions: Homogenous distribution of infectious
aerosol over 10 hours; uniform susceptibility.
- Fennelly KP & Nardell EA. Infect Control Hosp Epidemiol 1998; 19;754
Wells-Riley Mathematical Model of Airborne Infection
Probability of MTB Infection:
Isolation Room with 6 ACH:
Infectiousness and Duration of Exposure
1
Risk of MTB Infection
0.9
0.8
0.7
0.6
0.5
1
0.4
10
0.3
100
0.2
1000
0.1
0
0.1
1
10
100
Duration of Exposure (hours)
1000
TB is Spread by Aerosols,
NOT sputum
Particle size* & suspension in air
(* NOT size of bacilli)
• Particle size &
deposition site
–
–
–
–
• Time to fall the height
of a room
100 
20 
10  – upper airway
1 - 5  – alveolar
deposition
–
–
–
–
10 sec
4 min
17 min
Suspended indefinitely
by room air currents
- Courtesy of Sol Permutt, 2004
*NOT organism size
Six-stage Andersen cascade impactor
Andersen AA. J Bacteriol 1958;76:471.
Cough Aerosol Sampling System
- Fennelly KP et al. Am J Resp Crit Care Med 2004; 169; 604-9
Cough-generated aerosols of Mtb:
Initial Report from Denver, CO
4 of 16 (25%) of SS+ subjects
- Fennelly KP et al. Am J Resp Crit Care Med 2004; 169; 604-9
Variability of Infectiousness in TB:
Epidemiology
 Rotterdam, 1967-69: Only 28% of smear
positive patients transmitted infections.
 Van Geuns et al. Bull Int Union Tuberc 1975; 50:107
• Case control study 796 U.S. TB cases
– Index cases tended to infect most (or all) or
few (or none) of their contacts
– Snider DE et al. Am Rev Respir Dis 1985; 132:125
• Ability to publish outbreaks suggests that they
are episodic.
Variability of Infectiousness in TB:
Experimental
• All infections attributed to 8 of 61 (13%) patients.
of infections due to one patient with TB laryngitis.
50%
 Riley RL et al. Am Rev Respir Dis 1962; 85:511.
• 3 (4%) of 77 patients produced > 73% of the infections in
the guinea pigs.
 Sultan L. Am Rev Respir Dis 1967; 95:435.
 Recent replication of this model in Peru
 118 hospital admissions of 97 HIV-TB coinfected patients
 8.5% caused 98% of secondary GP infections
 90% due to inadequately treated MDR-TB
 Escombe AR et al. PLoS Medicine 2008; 5:e188
Occupational TB
in Sub-Saharan Africa
• Malawi
– 25% mortality
– Harries AD, Tran R Soc Trop Med Hyg 1999; 93: 32
• Ethiopia
• South Africa
• Nigeria
– 32 of 2,173 HCWs
• 15 (47%) as HIV-TB
– Salami AK, Nigerian J Clin Prac 2008; 11: 32
What is the magnitude and variability of infectious
aerosols of M. tuberculosis?
(Can we better identify the most infectious?)
Hypothesis 1: Coughgenerated aerosols of Mtb
can be measured in
resource-limited settings.
Hypothesis 2: Coughgenerated aerosols will be
detected in approximately
25-30% of patients with
PTB.
Cough Aerosol Sampling System
v.2
Frequency Distribution of Cough-generated
Aerosols of M. tuberculosis and
Relation to Sputum Smear Status
31/112 (28%) SS+ subjects
3.5
5
3
4
3
2
1.5
2
1
1
0.5
0
0
1
7
13
19
25
31
37
43
49
55
61
67
73
79
85
91
Subjects Sorted by Aerosol CFU then by Sputum AFB
Aerosol Log 10 CFU
Sputum AFB
97 103 109
Sputum AFB
Aerosol Log CFU
2.5
Cough-generated Aerosols of
M. tuberculosis:
Normalized Particle Sizes
50
P er C ent C FU
40
30
20
10
0
1
2
3
4
5
6
S t ag e o f An d er sen C ascad e Im p act o r
Lower limit of size range(µ)
7.0
4.7
3.3
2.1
1.1
0.65
Anatomical deposition: Upper airway -- bronchi -- alveoli
Abstract, ATS International Conference, 2004.
Pitfalls in Administrative Controls
• TB Mortality not prioritized or under
surveillance (i.e., no data collection)
• HIV screening of HCWs not prioritized
– major risk factor for TB disease & death
– HAART now feasible in much of world
– HIV screening advocated for adm’t patients in US
• TB laboratory personnel often not involved in
TB infection control efforts
– Botswana: 1st AFB smear ‘STAT’
• Decisions re: infectiousness falls onto
clinicians with variable expertise
Pitfalls in Environmental Controls
• Little or no engineering expertise and support for
hospitals & HCFs
– No systems of communication / interaction
– Different ‘cultures’ and mind-sets
• TB nurses or administrators subject to sales
pitches from commercial vendors
– UVGI lamps in SANTA facilities
– Mobile air filters in Newark, NJ
• Lack of appreciation of natural ventilation…and
its limitations!
– Low rate of nosocomial infection in Uganda project
– High rate in Tugela Ferry
Pitfalls in Personal Respiratory Protection
• Too much attention paid to ‘masks’ at expense
of administrative and environmental measures
• Rizdon R et al: Renal unit with poor ventilation
• Inappropriate use on patients
• Focus on fit-testing and regulation rather than on
follow up on use in field
• Lack of appreciation that not all respirators
provide the same level of protection
– Need for more protection in high-risk aerosol-inducing
procedures, e.g., bronchoscopies
TB-IC Practices
for Community Programs
• Best administrative control:
– Suspect and separate until diagnosed
– Surveillance of HCWs with TST (and/or IGRAs) and rapid
treatment of LTBI if conversions occur
• Best environmental control: Ventilation
– Do as much as possible outdoors
– Use directional airflow when possible
• Natural breeze or fans: HCW ‘upwind’; patient ‘downwind’
• Personal respiratory protection
– N95 respirators when indoors or very close (procedures)
– Surgical masks on patients to control source
Summary: TB-IC
• Administrative controls most important
component of TB-IC
– ‘Suspect and separate!’
– Prioritize screening HIV in HCWs
• Prioritize good ventilation in all areas
– Back-up in areas with poor ventilation
• Fans, mechanical ventilation, UVGI
• Prioritize personal respiratory protection for
high risk settings, esp where admin and
environ controls limited