Utility of bronchoalveolar lavage in diagnosing respiratory tract infections in patients with hematological malignancies: are invasive diagnostics still needed?

Figures & data

Table 1. Patient characteristics.

Characteristics	n = 151 (%)
Unique patients	n = 133
Age, years, median (range)	59 (19–84)
Gender
Male	79 (59)^a
Female	53 (41)
Diagnosis
AML	69 (46)
B-ALL^b	18 (12)
Lymphoma^c	17 (11)
Myeloma	15 (10)
CLL	13 (9)
CML	8 (5)
MDS	6 (4)
MPN	8 (5)
Other	10 (7)
Deceased
At end of study period	77^a (58)
Within 30 days from time of BAL	5 (3)
Within 1–6 months from time of BAL	35 (23)
Still living	56^a (42)
Allo-SCT
Yes	53 (35)
No	98 (65)
Time from allo-SCT to BAL
Within 1 month	8
Within 1–3 months	8
Later than 3 months	37
Active GVHD in allo-SCT patients
Yes	19
No	34
Auto-SCT
Yes	15 (10)
No	136 (90)
Time from auto-SCT to BAL
Within 1 month	6
Within 1–3 months	0
Later than 3 months	9
Infectious symptoms at time of BAL
Fever	103 (68)
Cough	41 (27)
Hypoxia and dyspnea	17 (11)
Fever only	11 (7)
Other	9 (6)
Time (days) from onset of infectious symptoms to time of BAL, median (range)	9 (0–371)
Platelets at time of BAL, ×10^9/L, median (range)	61 (<5–825)
Neutropenia^d at time of BAL
Yes	54 (36)
No	97 (64)
Ongoing treatment at time of BAL
Systemic broad-spectrum antibiotics	98 (65)
Systemic broad-spectrum antimycotics	53 (35)
Antibiotics and antimycotics^e	44 (29)
Time (days) of treatment before BAL, median (range)
Systemic broad-spectrum antibiotics	5 (1–24)
Systemic broad-spectrum antimycotics	3 (1–48)
Pulmonary infiltrates on CT scan and/or X-ray
Yes	139 (92)
No	12 (8)
Ongoing immunosuppressive treatment at time of BAL
Low-dose steroids	32 (21)
High-dose steroids^f	18 (12)
Cyclosporin	29 (19)
No	90 (60)

^a Numbers calculated on 133 unique patients.

^b Including Burkitt’s lymphoma and lymphoblast lymphoma.

^c Including high-grade non-Hodgkin’s lymphoma, low-grade non-Hodgkin’s lymphoma, Hodgkin’s lymphoma, and T-cell lymphoma.

^d Absolute neutrophil count (ANC) below 0.5 × 10⁹.

^e Systemic intravenous or oral agents, excluding prophylaxis, with a wide therapeutic range.

^f A daily dose of ≥20 mg prednisone (or equivalent).

Table 2. Impact on diagnosis and clinical management.

Impact of BAL on clinical management	n = 151 (%)
Diagnosis based solely on BAL	27 (18)
Diagnosis based on BAL and other diagnostic methods	11 (7)
Negative findings from BAL leading to cessation of   initiated therapy	6 (4)
No impact	107 (71)
Method for establishing final diagnosis	n = 151 (%)
Diagnosis based solely on BAL	27 (18)
Diagnosis based on BAL and other diagnostic methods	11 (7)
Diagnosis based on other methods than BAL	21 (14)
No diagnosis established	92 (61)
Final diagnosis^a	n = 157 (%)
Aspergillus pneumonia	23 (15)
Pneumocystis pneumonia	14 (9)
Other specified cause^b	26 (17)
Pneumonia with unknown cause	85 (54)
Neutropenic fever and/or respiratory tract symptoms   with unknown cause	9 (6)

^a More than one diagnosis in some cases.

^b Bacterial pneumonia with specified cause, invasive candidosis, Mycobacterium tuberculosis, RS-virus, CMV pneumonitis, pulmonary GVHD, bacterial sinusitis with established cause, drug reaction, heart failure, Legionella pneumonia, Rhino virus, pulmonary embolism, COP.

Table 3. Methods for establishing Aspergillus pneumonia and Pneumocystis pneumonia.

Method for establishing diagnosis Aspergillus pneumonia^a (n = 23)	n (%)
Aspergillus infection proven	2 (9)
Aspergillus infection probable	8 (35)
Aspergillus infection possible^b	13 (57)
Diagnosis based solely on BAL	10 (43)
Diagnosis based on BAL and other diagnostic methods	5 (22)
Diagnosis based on other methods	8 (35)
Positive antigen for Aspergillus in BAL^c	3 (13)
Positive PCR for Aspergillus in BAL	4 (17)
Positive direct microscopy for Aspergillus in BAL	1 (4)
Positive cultivation for Aspergillus in BAL	8 (35)
Positive antigen for Aspergillus in blood^c	9 (39)
Positive PCR for Aspergillus in blood	0 (0)
Other methods (not based on BAL) contributing to diagnosis	2 (9)
Typical X-ray finding for Aspergillus^d	8 (35)
Method for establishing diagnosis Pneumocystis pneumonia^a (n = 14)	n (%)
Diagnosis based solely on BAL	11 (79)
Diagnosis based on BAL and other diagnostic methods	2 (14)
Diagnosis based on other methods	1 (7)
Positive immune morphology for Pneumocystis jirovecii in BAL	4 (29)
Positive PCR for Pneumocystis jirovecii in BAL	6 (43)
Positive for both immune morphology and PCR in BAL	3 (21)
Other methods (not based on BAL) contributing to diagnosis	3 (21)

^a Excluding clinical assessment.

^b According to the ‘Revised Definitions of Invasive Fungal Disease from the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) Consensus Group’.

^c Optical density cutoff value 1.0.

^d Halo sign and/or air crescent sign.

Table

All hematology patients:

• General culture (quantitative, aerobe, anaerobe)

• Fungal culture, direct fungal microscopy

• Respiratory virus PCR (RS-, adeno-, para influenza-, influenza A and B)^a

• Respiratory virus immunofluorescence (RS-, adeno-, para influenza-, influenza A and B)^b

• Legionella PCR (cultivation is performed automatically in case of positive PCR)^c

• Legionella immunofluorescence^d

• Protected specimen brush: general culture, fungal culture, Legionella PCR (culture is performed automatically in case of positive PCR)^c

• Pneumocystis jirovecii immunofluorescence (PCR is performed automatically in case of negative result)

• Mycobacteria direct microscopy and general culture

• General virus culture

• Papanicolaou cytology

Immunocompromised patients:

• Cytomegalovirus PCR

• Herpes simplex type 1 and 2 PCR^c

• Aspergillus antigen and Aspergillus PCR^c

In selected cases:

• Pneumococci PCR^c

• Mycoplasma + Chlamydophilia pneumonia PCR

• M. tuberculosis complex PCR^c

• Cytomegalovirus Antigen^b

^a Added 2012.

^b Excluded 2012.

^c Added 2007.

^d Excluded 2007.