Transcript Neutropenia Febril

What are we looking at?
Challenges in the diagnosis of
Invasive Mould Diseases
Alessandro C. Pasqualotto
[email protected]
Porto Alegre, Brazil
Potential conflicts of interest
• Research Grants
 Myconostica, Pfizer, Merck, Sigma-Tau, CAPES,
CNPq, Fungal Research Trust
• Travel Grants
 Pfizer, United Medical, Schering (now Merck), Bagó,
Merck
• Speaker honoraria
 Pfizer, United Medical, Merck, Schering (now Merck),
Biometrix
First assumption:
IFD are highly lethal diseases
p<0.001
Incidence 13.3% in lung
transplant recipients
Xavier MO, Pasqualotto AC, et al. ECCMID 2009
Rapidly evolving diseases
4 days later
www.aspergillus.org.uk
Disseminated infection
www.aspergillus.org.uk
Second assumption:
We need to intervene asap
Early versus late intervention
Mortality rate (%)
100
80
60
40
20
0
Within 10 days
Von Eiff, et al. Respiration 1995; 62: 241-7
Early versus late intervention
Mortality rate (%)
100
80
60
40
20
0
Within 10 days
> 11 days
Von Eiff, et al. Respiration 1995; 62: 241-7
But how can we achieve such an
early diagnosis?
A small black
scar
2 days earlier
+ serum GM
Patient died
1 day after this
picture was
taken
A small black
scar
2 days earlier
+ serum GM
Patient died
1 day after this
picture was
taken
MD Anderson Cancer Centre
Necropsy study over a 15-years period
• IFD detected in 31% over 1,017
necropsies
• Antemortem diagnosis in only 25%
Chamilos G, et al. Haematologica 2006; 91: 986-9
MD Anderson Cancer Centre
Necropsy rate has reduced over time
100
80
%
60
40
20
0
1989-1993
1994-1998
1999-2003
Chamilos G, et al. Haematologica 2006; 91: 986-9
By the way, what is the necropsy
rate in your institution?
1. >40%
2. 10-39%
3. 1-10%
4. <1%
5. Are you kidding?
MD Anderson Cancer Centre
Trends in the prevalence of IFD
100
Invasive
fungal disease
80
%
60
Invasive
mould disease
40
20
0
1989-1993 1994-1998 1999-2003
Mixed
infections
Chamilos G, et al. Haematologica 2006; 91: 986-9
No need to worry!
CT scan and
galactomannan
are there to help us out!
‘Halo sign’ surrounding a nodule
Day 0:
halo
Day 4:
size, halo
Day 7:
air crescent
Caillot, et al. J Clin Oncol 1997; 15: 139-47
The sign is not specific for IA
• Vasculitis
• Metastasis
• Pseudomonas infections
• Zygomycosis and other angio-invasive
infections
Greene RE, et al. Clin Infect Dis 2007; 44: 373-9
Absence of typical findings at
chest CT scan
• COPD
• Steroids
• Other non-neutropenic patients / ICU
• Lung transplant recipients
• ? Monoclonal antibodies
‘Reversed halo sign’
Organising cryptogenic pneumonia
Wahba H, et al. Clin Infect Dis 2008; 46: 1733-7
‘Reversed halo sign’
• Review of 189 cases of invasive mould
disease
 Overall frequency 4%
Wahba H, et al. Clin Infect Dis 2008; 46: 1733-7
‘Reversed halo sign’
• Review of 189 cases of invasive mould
disease
 Overall frequency 4%
 Zygomycosis 19%
 Aspergillosis <1%
 Fusariosis 0%
(p<0.01)
Wahba H, et al. Clin Infect Dis 2008; 46: 1733-7
Other predictors of zygomycosis
• >10 nodules
• Pleural effusion
• Concomitant sinusitis
• Treatment with voriconazole
Chamilos G, et al. Clin Infect Dis 2005; 41: 60-6
Meta-analysis of GM testing
Low PPV
High NPV
Pfeiffer CD, et al. Clin Infect Dis 2006; 42: 1417-27
Reproducibility
Caution with low
+ve indexes!
Upton A, et al. J Clin Microbiol 2005; 43: 4796-800
GM release by non-Aspergillus fungi
• Penicillium marneffei • Paecilomyces variotii
• Geotricum capitatum • Botrytis tulipae
• Acremonium species • Cladosporium species
• Alternaria alternata
• Exophiala dermatitidis
• Rhodotorula rubra
• Trichophyton species
Aquino VR, Goldani LZ, Pasqualotto AC. Mycopathologia 2007; 163: 191-202
Cross-reaction with GM testing
100
90
80
70
60
% 50
40
30
20
10
0
Paracocci
Histo
Crypto
neoformans
C gattii
Xavier MO, Pasqualotto AC, Severo LC. Clin Vaccin Immunol 2009; 16: 132-3
Clinical case
• 19 year-old man, refractory leukaemia
• Febrile neutropenia
• Amox-clav for E. Coli bacteremia
Maertens J, et al. Clin Infect Dis 2004; 39: 289-90
Clinical case
• 19 year-old man, refractory leukaemia
• Febrile neutropenia
• Amox-clav for E. Coli bacteremia
• Daily GM determination
– D1 after antibiotic: GM index of >1.5
– 5 +ve tests afterwards
Maertens J, et al. Clin Infect Dis 2004; 39: 289-90
Clinical case
• 19 year-old man, refractory leukaemia
• Febrile neutropenia
• Amox-clav for E. Coli bacteremia
• Daily GM determination
– D1 after antibiotic: GM index of >1.5
– 5 +ve tests afterwards
• Fluoroquinolone: gradual reduction in GM index
Maertens J, et al. Clin Infect Dis 2004; 39: 289-90
Clinical case
• 1 wk later
– Pipe-tazo for appendicitis
– GM >2.5; bilateral nodular infiltrate
Maertens J, et al. Clin Infect Dis 2004; 39: 289-90
Clinical case
• 1 wk later
– Pipe-tazo for appendicitis
– GM >2.5; bilateral nodular infiltrate
• Probable IA (EORTC / MSG)
– Antifungal therapy + meropenem
– Gradual ↓ in GM index
Maertens J, et al. Clin Infect Dis 2004; 39: 289-90
Clinical case
• 1 wk later
– Pipe-tazo for appendicitis
– GM >2.5; bilateral nodular infiltrate
• Probable IA (EORTC / MSG)
– Antifungal therapy + meropenem
– Gradual ↓ in GM index
• Necropsy: leukaemia infiltrate
– Absence of IA
Maertens J, et al. Clin Infect Dis 2004; 39: 289-90
Meta-analysis of GM testing
Proven or probable IA
Haematological
malignancies
Solid organ
transplantation
Sensitivity
Specificity
0.58 (52-64)
0.95 (94-96)
0.41 (21-64)
0.85 (80-89)
Pfeiffer CD, et al. Clin Infect Dis 2006; 42: 1417-27
Specificity 94%
BAL PCR testing
Sensitivity 79%
Marked heterogeneity
(particularly for sensitivity)
Tuon FF. Rev Iberoam Micol 2007; 24: 89-94
PCR Critical points
• Variable sensitivity / specificity
• Lack of standardised targets / reagents
• Extraction method
• Platform (conventional PCR vs Real time)
• Poor understanding of DNA kinetics
• Not yet part of the EORTC/MSG criteria
De Pauw B, et al. Clin Infect Dis 2008; 46: 1813-21
Beta-Glucan
• Detected in IFDs caused by
– Candida and Aspergillus
Ergosterol
Yoshida M, et al. J Med Veter Mycol 1997; 35: 371-4
Beta-Glucan
• Detected in IFDs caused by
– Candida and Aspergillus
– Trichosporon
– Fusarium
– Acremonium
– Saccharomyces
– Pneumocystis
Yoshida M, et al. J Med Veter Mycol 1997; 35: 371-4
There he is again,
speaking about a test that nobody
uses in Brazil …
The reality in Brazil
• A survey performed in collaboration
with ANVISA
140 hospitals
 >42,000 beds
 65% teaching hospitals
 90% belonging to the Sentinel Network
N
B
eu
M
ro
T
su
rg
er
y
N
A
eo
na IDS
ta
H
lc
ae
a
m
od re
ya
lis
is
B
ur
n
T
%
SO
nc
H
ea olo
gy
rt
su
rg
er
y
O
Complexity in hospital care
100
80
60
40
20
0
Is IFD a problem in your centre?
38.7%
Yes
No
n=140
Do you know your local
epidemiology?
40.1%
Yes
No
n=140
Specialised media for fungi
Yes
19.7%
No
n=140
Aspergillus identification
at the species level
58.5%
Yes, always
Yes, occasionally
Never
n=140
Fungal staining - biopsies
Yes, always
51.1%
Yes, occasionally
No
n=140
Biopsy specimens are sent in
formalin only?
Yes, always!
26.0%
Occasionally
Never
n=140
Access to high resolution CT
34.3%
Yes
No
n=140
Galactomannan
Sim
83.6%
Não
n=140
The appropriateness or
inappropriateness of feelings is
relative to the ground and to the
circumstances of those feelings
Aristotle
The Doctrine of the Mean, 384-322 BC
Fight fire with fire
Metallica
Ride the Lightning, 1984 AD
Invasive diagnostic interventions
• CT-guided percutaneous lung biopsy:
yield of 70-100%
Nosari A, et al. Haematologica 2003; 88: 1405-9
Crawford SW, et al. Transplantation 1989; 48: 266-71
Hoffer FA, et al. Pediatr Radiol 2001; 31: 144-52
Lass-Florl C, et al. Clin Infect Dis 2007; 45: e1001-4
Invasive diagnostic interventions
• CT-guided percutaneous lung biopsy:
yield of 70-100%
 Platelets >60,000/ml are required
 Pneumothorax 18%
 Haemoptysis 3%
Nosari A, et al. Haematologica 2003; 88: 1405-9
Crawford SW, et al. Transplantation 1989; 48: 266-71
Hoffer FA, et al. Pediatr Radiol 2001; 31: 144-52
Lass-Florl C, et al. Clin Infect Dis 2007; 45: e1001-4
Invasive diagnostic interventions
• Transbronchial biopsies:
 False-negative results are frequently seen
Lass-Florl C, Freund MC. In: Aspergillosis: from diagnosis to prevention.
Pasqulaotto AC, ed. Springer, 2009
Invasive diagnostic interventions
• Open lung biopsies:
 Provide larger samples of tissue with
improved accuracy and specificity
Lass-Florl C, Freund MC. In: Aspergillosis: from diagnosis to prevention.
Pasqulaotto AC, ed. Springer, 2009
Invasive diagnostic interventions
• Open lung biopsies:
 Provide larger samples of tissue with
improved accuracy and specificity
 Contradictory results
 Complication rate of 10-15%
Lass-Florl C, Freund MC. In: Aspergillosis: from diagnosis to prevention.
Pasqulaotto AC, ed. Springer, 2009
Invasive diagnostic interventions
Peripheral lesions
Bilateral /
multifocal disease
Focal lesions near the
hilum / great vessels
Needle biopsy /
Surgical resection
BAL
Urgent thoracotomy
and resection
Lass-Florl C, Freund MC. In: Aspergillosis: from diagnosis to prevention.
Pasqulaotto AC, ed. Springer, 2009
Conclusion
• It is mostly but not all about IA
 Invasive mould diseases have to be
differentiated from each other
Conclusion
• It is mostly but not all about IA
 Invasive mould diseases have to be
differentiated from each other
• Diagnosis is the most challenging step in
infectious diseases
Conclusion
• It is mostly but not all about IA
 Invasive mould diseases have to be
differentiated from each other
• Diagnosis is the most challenging step in
infectious diseases
• We need a better understanding on the
performance of the available tests
Acknowledgments
• Mycology team
 Luiz Carlos Severo
 Valerio R Aquino
 Cecilia B Severo
 Luciana Guazelli
 Melissa Xavier
• Infection Control Dept
 Teresa Sukiennik