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 fennelkp@umdnj.edu

2. 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.

3. Is TB an Occupational Disease of HCWs? Low- & middle- income countries High-income countries LTBI (prevalence) 63% (33-79%)24% (4-46%) TB disease (annual incidence) 5.8% (0-11%)1.1% (0.2-12%) TB mortality (inpt) (PMR) (outpt) ??1.18 (1.04-1.35) 3.04 (1.62-5.19) - Menzies D et al. IJTLD 2007; 11:593

4. 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

5. Personal Respiratory Protection Against M. tuberculosis: Contentious Controversy

6. from Sol Permutt, 2004

7. 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

8. Assumptions: Homogenous distribution of infectious aerosol over 10 hours; uniform susceptibility. - Fennelly KP & Nardell EA. Infect Control Hosp Epidemiol 1998; 19;754 Control Measures are Synergistic & Complementary

9. Wells-Riley Mathematical Model of Airborne Infection

10. TB is Spread by Aerosols, NOT sputum

11. *NOT organism size Particle size* & suspension in air (* NOT size of bacilli) Particle size & deposition site –100  –20  –10  – upper airway –1 - 5  – alveolar deposition Time to fall the height of a room –10 sec –4 min –17 min –Suspended indefinitely by room air currents - Courtesy of Sol Permutt, 2004

12. Six-stage Andersen cascade impactor Andersen AA. J Bacteriol 1958;76:471.

13. Cough Aerosol Sampling System - Fennelly KP et al. Am J Resp Crit Care Med 2004; 169; 604-9

14. 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

15. 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.

16. Variability of Infectiousness in TB: Experimental All infections attributed to 8 of 61 (13%) patients. 50% of infections due to one patient with TB laryngitis.  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

17. 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

18. What is the magnitude and variability of infectious aerosols of M. tuberculosis? (Can we better identify the most infectious?) Hypothesis 1: Cough- generated aerosols of Mtb can be measured in resource-limited settings. Hypothesis 2: Cough- generated aerosols will be detected in approximately 25-30% of patients with PTB.

19. Cough Aerosol Sampling System v.2

20. Frequency Distribution of Cough-generated Aerosols of M. tuberculosis and Relation to Sputum Smear Status 31/112 (28%) SS+ subjects

21. Cough-generated Aerosols of M. tuberculosis: Normalized Particle Sizes 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.

22. 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: 1 st AFB smear ‘STAT’ Decisions re: infectiousness falls onto clinicians with variable expertise

23. 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

24. 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

25. 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

26. 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