Use of Xpert MTB/RIF for Diagnosing Extrapulmonary TB Stephen D. Lawn Improving health worldwide www.lshtm.ac.uk Global Forum of Xpert MTB/RIF Implementers: Geneva, May 1 st 2014 Department of Clinical Research, London School of Hygiene & Tropical Medicine, London, UK Desmond Tutu HIV Centre, University of Cape Town, Cape Town, South Africa
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Use of Xpert MTB/RIF for
Diagnosing Extrapulmonary TB
Stephen D. Lawn
Improving health worldwide www.lshtm.ac.uk
Global Forum of Xpert MTB/RIF Implementers:
Geneva, May 1st 2014
Department of Clinical Research, London School of Hygiene & Tropical Medicine, London, UK
Desmond Tutu HIV Centre, University of Cape Town, Cape Town, South Africa
Outline
• The problem of EPTB
– Frequency
– Diagnostic challenge
• Evidence base for use of Xpert MTB/RIF for
EPTB diagnosis
• Further data needed
• New recommendations: policy update
What proportion of TB is
extrapulmonary?
• ECDC EuroSurveillance report (Solovic I et al. 2013):
– Notified TB cases in the EU in 2011 (n=72,334)
– 22% was EPTB
– Range 4% to 48% in different countries (!)
• WHO Global TB Control Report 2013
– 5.4 M notified new cases in 2012
– 0.8 M (14.8%) had EPTB
TB Notifications in Cape Town
in 2009 (n=29,478)
0
1000
2000
3000
4000
5000
TB
no
tifi
cati
on
s
Age strata
HIV positive
HIV unknown
HIV negative
Wood and Lawn PLoS ONE 2011
HIV-Associated TB in Cape Town
0
500
1,000
1,500
2,000
TB
No
tifi
cati
on
s
CD4 count (cells/µL)
PTB PTB + EPTB EPTB
Gupta / Lawn 2012; IJTLD 2013
(n=11,816)
TB Diagnosis in Unselected HIV+ Medical
Admissions in Cape Town
427 new admissions
2,391 samples (median – 3 body compartments)
3,471 diagnostic tests for TB
Lawn et al CROI 2014
TB Prevalence
• TB diagnoses = 139 (Xpert or culture of any sample)
• TB Prevalence = 32.6% (28.1-37.2)
TB
No TB
Median CD4 Count in TB cases = 80 cells/µL
Cough >2 weeks in just 13% of TB cases
WHO symptom screen NOT predictive for TB
Pulmonary vs Extrapulmonary Disease
36.7%(n=51)
17.3%(n=24)
46.0%(n=64)
PTB in 54.0%
EPTB in 82.7%
Lawn et al CROI 2014
Challenges in EPTB Diagnosis
• Sampling
– invasive procedures
– special expertise
• Frequently paucibacillary
– Mostly smear-negative
– Limited sensitivity of all microbiological tests
– Prolonged time to culture-positivity
– DST even further delayed
• Reliance on histology and imaging
• Empirical treatment
1. Testing non-respiratory samples to diagnose
pulmonary TB, especially in children
2. Diagnosis of TB at extrapulmonary sites ����
Xpert MTB/RIF Testing of
Non-Respiratory Samples
Diagnostic Accuracy of Xpert MTB/RIF When
Testing Non-Respiratory Samples vs Culture
(n=27 studies with 6026 samples)
Challenges in Assessing Literature
on Diagnostic Accuracy for EPTB
• Laboratory-based studies with convenience sampling
+ mixture of studies with prospective clinical
recruitment + a few case control
• Range in size (range 7 – 1476 samples)
• Range in TB prevalence: 0% - 81%
• Multiple sample types
• Variations in sample processing
• Reference standard: microbiological vs composite
• Disaggregation of data by sample type / age / HIV-