Flexible bronchoscopy (FB) is one of the most common and useful procedures performed
by chest clinicians, with > 500,000 procedures performed annually in the United States.
1
FB is performed frequently in immunocompromised patients and in health-care settings
where exposures to increasingly virulent and drug-resistant microorganisms may occur.
Recently reported bronchoscopy-associated infections (BAIs) and pseudoinfections
1
,
2
,
3
,
4
potentially affected > 800 patients and included instances in which clinically significant
infections and fatalities might have been related to the procedure. Such outbreaks
have generated major concerns regarding the possibility of iatrogenic infections due
to FB and underscore the importance of reappraising this problem and optimizing preventive
practices.
5
,
6
The potential sequelae of BAI and pseudoinfection are enormous. In addition to the
possible morbidity and mortality associated with true infections, such events require
expenditure of considerable time and resources for the careful assessment of the vast
majority of instances in which no infections or harm to patients have occurred.
For many reasons it is remarkable that true infection due to FB appears to be an uncommon
event. During performance of the procedure, host defenses are bypassed routinely as,
most often, the bronchoscope is passed through the upper airways, which are invariably
colonized by a myriad of potential pathogens. The patient’s cough and other protective
reflexes are attenuated purposefully with a variety of medications, ensuring aspiration
of microbes, and these and other solutions are instilled routinely into progressively
more distal airways, potentially soiling peripheral lung parenchyma. Normal mucosal
barriers to infection are disrupted during lung biopsies and an increasing array of
interventional procedures. With the latter, lengthier procedure times may increase
opportunities for hematogenous as well as local infections. Simultaneously, the progressive
miniaturization of bronchoscopes and accessories introduces potential difficulties
in effective cleaning and disinfection of these structurally complex instruments.
In addition, the prevalence of HIV, multidrug-resistant tuberculosis, hepatitis B,
and newly emerging pathogens such as the coronavirus agent of severe acute respiratory
syndrome (SARS-CoV) heighten concerns about the relative risks posed by bronchoscopy.
Prakash
7
has described several scenarios in which the bronchoscope may propagate infection.
These include intrapulmonary or extrapulmonary spread of infection within the same
patient, pathogen transmission from one patient to another, and the spread of infection
from the patient to participating medical personnel.
8
,
9
Each of these possibilities poses unique challenges in implementing effective infection
control practices. Other inherent difficulties also relate to meaningful definitions
of infection or pseudoinfection, their accurate recognition in individual patients,
appropriate monitoring of cleaning and disinfection of bronchoscopy instruments, laboratory
maintenance, bronchoscopy staff education, and longitudinal monitoring and documentation
of effective practices.
The exact incidence of infections caused by FB is unknown. Their apparent low frequency
might reflect a truly uncommon occurrence. Alternatively, such events might be systematically
underrecognized because of multiple factors. For example, new infection (patient-to-patient)
induced by the procedure may be easily masked by the primary signs and symptoms for
which FB was performed. In an individual patient, interpretation of a positive culture
from a bronchoscopy specimen and the precise differentiation of colonization or infection
from instrument contamination may be extraordinarily difficult. This dilemma may be
especially problematic in the sickest patients. An individual bronchoscopist may not
have access to the comprehensive experience of a bronchoscopy suite in order to discern
whether telltale patterns of microbiologic isolates exist among patients undergoing
bronchoscopy, and whether these are inappropriate for the patient’s clinical presentation.
Other reasons for underrecognition of endemic infection may include inadequate surveillance
of outpatient procedures, asymptomatic infection or prolonged incubation period prior
to the development of the symptoms, and fear of medicolegal ramifications of reporting/publishing
device-related transmission of infection.
At an institutional level, recognition of BAI requires regular systematic review of
all aspects of the procedure, rigorous adherence, reinforcement and monitoring of
well-established infection control practices, and established mechanisms for timely
objective epidemiologic evaluation whenever a potential problem occurs. Such resources
(and the funding needed to support such efforts) may not be consistently available
at many centers.
While prevention of infection caused by FB is an obvious concern for all chest clinicians,
several observations suggest that there is a widespread unfamiliarity with practice
guidelines and that potentially consequential variations in primary and secondary
prevention of BAI and pseudoinfections may occur. Although reprocessing guidelines
exist, their focal point has been on GI endoscopes. Guidelines addressing bronchoscope
reprocessing, for the most part, are derived from those dealing with the GI endoscope.
Guidelines for reprocessing of flexible endoscopes, including bronchoscopes, have
been promulgated by the Association for Professionals in Infection Control, Association
of PeriOperative Registered Nurses, and the British Thoracic Society, among others.
10
,
11
,
12
,
13
Data suggest that these guidelines have not been effectively disseminated and/or followed
by many clinicians. A comprehensive survey
14
of bronchoscopists in the United Kingdom revealed that national guidelines for bronchoscope
reprocessing were not followed consistently. Minimum disinfection times recommended
before and after routine bronchoscopies were often (35%) not achieved, and no disinfection
was carried out in 34% of medical centers before emergency bronchoscopies and in 19%
of units after suspected cases of tuberculosis. Adequate rinsing of the bronchoscope
with sterile or filtered water was not carried out by 43% of units. Staff rarely (7%)
wore recommended protective clothing during bronchoscopy. In another survey involving
US bronchoscopists, nearly two thirds of respondents, including medical directors
of bronchoscopy suites, acknowledged that they were unfamiliar with national reprocessing
recommendations. Interestingly, this survey was carried out just following a widely
publicized Pseudomonas outbreak involving bronchoscopes.
15
Moreover, many (39%) were not aware of the approaches to reprocessing at their own
institutions! Although specific data are unavailable, anecdotal observations suggest
that bronchoscope reprocessing techniques are inadequately emphasized during the training
of many pulmonary fellows, and many bronchoscopists may take it for granted that the
instruments they use are safe.
In order to help organize and disseminate information regarding validated practices
for bronchoscope reprocessing and infection prevention and control, this committee
was jointly convened by the American College of Chest Physicians and the American
Association for Bronchology. In this document, we summarize data from the literature
regarding the extent of the problem and specific risk factors (generally discovered
during the comprehensive investigation of outbreaks). We also present principles related
to the specific maintenance and disinfection of instruments together with provisional
recommendations. Since many important factors extend beyond the purview of individual
bronchoscopists and relate to the milieu in which the procedure is performed, we also
address institutional aspects of the infrastructure and processes essential to bronchoscopy
infection prevention and control.
The readers of this article should recognize that this is a consensus statement and
does not represent evidence-based recommendations. Because of the relative absence
of prospective investigations in this area, most of these recommendations are based
on clinical experience and consensus opinion, rather than the higher grades of evidence
generally required for true clinical practice guidelines. Accordingly, our recommendations
represent an evolving perspective that provides numerous important opportunities for
clinical outcomes research and that will require future critical refinement. In addition
to general principles, the implementation of effective programs is institution specific
and requires local modification through continued active dialogue among bronchoscopists,
staff, and infection control teams.
16
We would also like to highlight that these recommendations are not intended to ensure
the inactivation or removal of causative agents of transmissible spongiform encephalopathies
(prion proteins).
Literature Review
The true incidence of BAI is unknown, due in part to episodic reporting and lack of
specific monitoring or institutional surveillance of such events. A review of the
English-language literature from 1970 to 2003 revealed over three hundred references
to endoscopy-related transmissions of infection. Sixty-two were specific to FB. Despite
published guidelines for the reprocessing of endoscopes, reports of “true infections”
and “pseudoinfections” or “pseudoepidemics” related to FB seem to be increasing. Most
citations refer to pseudoinfections and pseudoepidemics as the isolation of organisms
in bronchoscopy specimens due to colonization or contamination of the bronchoscope
rather than true patient-to-patient transmission (“true infection”) [Table 1
].
1
,
2
,
3
,
4
,
17
,
18
,
19
,
20
,
21
,
22
,
23
,
24
,
25
,
26
,
27
,
28
,
29
,
30
,
31
,
32
,
33
,
34
,
35
,
36
,
37
,
38
,
39
,
40
,
41
,
42
,
43
,
44
,
45
,
46
,
47
,
48
,
49
,
50
,
51
,
52
,
53
,
54
,
55
,
56
,
57
,
58
,
59
,
60
,
61
,
62
,
63
,
64
,
65
,
66
,
67
,
68
,
69
,
70
,
71
,
72
The majority of reports are descriptive, with few case-controlled investigations.
Typically, episodes can be traced to inadequate cleaning techniques or disinfection
processes. Occasionally, the infection is due to contamination of water supplies,
reprocessing equipment, or accessories such as stopcocks or cleaning brushes. Defects
within the bronchoscope itself (suction valve port) have also been implicated in transmission
of organisms from patient to patient (Table 2
).
17
,
18
,
19
,
20
,
21
,
22
,
23
,
24
,
25
,
26
,
27
,
28
,
29
,
30
,
31
,
32
,
33
,
34
,
35
,
36
,
37
,
38
,
39
,
40
,
41
,
42
,
69
,
70
,
71
,
73
,
74
Table 1
Bronchoscopy-Related Pseudoinfections*
Organisms
Total Reports, No.
Affected Patients, No.
Reference(s)
Bacteria
Proteus sp
2
8
3
,
43
Bacillus sp
2
23
44
,
45
Serratia marcescens
5
33†, ‡
40
,
42
,
46
,
47
,
48
Pseudomonas aeruginosa
8
220†, §
1
,
2
,
48
,
49
,
66
,
67
,
68
,
69
,
70
,
71
Legionella pneumophilia
1
5
38
Klebsiella pneumonia
2
19
3
,
48
Methylobacterium mesophilicum
2
25
27
,
34
Morganella morganii
1
1
3
Fungi
Aureobasidium sp
1
9
23
Rhodotorula rubra
3
56
21
,
28
,
41
Blastomyces dermatitidis
1
2
17
Trichosporon cutaneum
1
8
4
,
5
,
29
Penicillium sp
1
8
29
Cladosporium sp
1
1
29
Phialospora sp
1
1
29
Mycobacteria
Mycobacterium tuberculosis
9
24†, ‡
18
,
20
,
31
,
32
,
49
,
50
,
51
,
52
,
53
Mycobacterium avium-intracellulare
4
11†, ‡
20
,
22
,
39
,
49
Mycobacterium xenopi
2
13†, ‱
25
,
35
Mycobacterium chelonae
15
304†, ‡, ‱
19
,
24
,
25
,
26
,
27
,
33
,
54
,
55
,
56
,
57
,
58
,
59
,
60
,
61
,
62
Mycobacterium fortuitum
2
4†, ‱
25
,
61
Mycobacterium gordonae
3
59†, ¶
24
,
30
,
36
,
72
Mycobacterium abscessus
2
33
63
,
64
Various nontuberculous mycobacteria
3
17†
37
,
60
,
65
*
Modified from Culver et al.
6
†
The precise number of pseudoinfections is not specified. The number of affected cases
are estimated from the excess positive bronchoscopy culture results compared to control
periods.
2
,
48
,
60
,
69
‡
The exact number of total pseudoinfections is unclear in these reports.
39
,
42
,
49
,
52
,
61
§
One report
69
described 35 excess cases but did not differentiate the proportion of pseudoinfections
and true infections. The same outbreak is described elsewhere.
49
,
70
∥
One report
25
described 15 patients with M xenopi, M chelonae, and/or M fortuitum pseudoinfections
but did not specify the numbers of each.
¶
Six patients with culture-positive M gordonae and two additional patients with smear-positive
acid-fast bacilli only.
36
.
Table 2
Major Sources of Contamination
Source of Contamination
Reference(s)
Ineffective cleaning
Inadequate cleaning
17
,
71
Damaged internal channel
18
,
19
Poorly mated internal components
74
Reusable suction valve
20
Suction channel
20
,
21
Biopsy port
71
Accessories
Sample collection tubing
22
Reused stopcocks for BAL fluid aspiration
23
Contaminated reprocessing equipment
Automated washer
69
Rinsing tank
24
Tubing
25
Filter
26
Biofilm in reprocessor
27
,
69
Cleaning brushes
28
Instilled solutions
Topical anesthesia (cocaine)
29
Green dye (additive to anesthetic)
30
Atomizer
31
Disinfectant
Inadequate activity
32
Incorrect disinfectant concentration dispensed by automated reprocessor
73
Contaminated glutaraldehyde
23
,
33
Improper connector to reprocessor
70
Recontamination after disinfection
Rinsing tap water (hospital supply)
34
,
35
,
36
,
37
Contaminated water filters
38
,
39
Reuse of “sterile water” for rinsing
40
Reassembly of valves prior to storage
41
Storage in coiled position/in cases:
42
Only 18 publications have suggested “true” infection: the transmission of a specific
pathogen associated with a clinically significant illness in a patient undergoing
FB (Table 3
).
1
,
2
,
18
,
19
,
20
,
31
,
32
,
42
,
46
,
49
,
53
,
66
,
67
,
68
,
69
,
70
,
73
,
75
,
76
Two of the accounts were reported in 2003, involving 33 patients with three possible
deaths.
1
,
2
The use of DNA probes was helpful in identifying patterns of transmission. In these
reports, the FBs were cleaned using automated endoscope reprocessors (AERs) and the
recommendations for disinfection of endoscopic equipment were strictly followed.
Table 3
Bronchoscopy-Related True Infections*
Organism
Mechanism
Outcome
Reference(s)
Year
S marcescens
Inadequate cleaning and disinfectant (alcohol)
Three true infections, 1 probable death, and 103 pseudoinfections
46
1975
Psudomonas sp
Suction attachment not detached prior to attempted disinfection
One true infection and five pseudoinfections
66
1978
Burkholderia pseudomallei
†
Unknown (rigid bronchscope)
Causality tenuous
67
1979
P aeruginosa
Inadequate disinfectant (povidone-iodine) Insufficient disinfection time (5 min);
reintroduction of cleaning brush after disinfection
One true infection and 10 pseudoinfections
68
1982
M tuberculosis
Inadequate disinfectant (povidone-iodine)
One patient each with true infection and pseudoinfection
32
1983
M chelonae
Damaged suction channel prevented adequate disinfection
Two true infections and 70 pseudoinfections
19
1983
M tuberculosis
Use of nondisposable suction valves
One patient acquired active TB, and two pseudoinfections each with M tuberculosis
and MAI
20
1989
S marcescens
Regimen inadequate at multiple steps
Six cases (five possible true infections); causality tenuous
42
1993
M tuberculosis
Multiple deviations from APIC guidelines
One patient acquired active TB
75
1997
Multidrug-resistant M tuberculosis
Multiple deviations from APIC guidelines
One patient each with active TB (died due to TB) and skin test conversion, and one
patient with pseudoinfection
53
1997
P aeruginosa
Contaminated AER
Number of true infections and pseudoinfections not specified
69
1997
P aeruginosa
Contaminated AER/not routinely serviced or cleaned
Two or more of eight total patients with true infection
76
2000
P aeruginosa
Inadequate disinfectant concentrations from automatic dispenser
Six ICU patients with colonization
73
2001
M tuberculosis
Reuse of lidocaine atomizers (?)
One patient each with lung and ocular TB
31
2001
P aeruginosa
Wrong connectors used for lumen disinfection by AER
Three true infections, and 14 pseudoinfections
70
2001
M tuberculosis
Punctured sheath/leak test not done
Two patients acquired active infection, and six patients had pseudoinfection
18
2002
P aeruginosa
Unclear (?) loose biopsy port cap prevented cleaning and disinfection
Twenty to 43 possible infections; pneumonias, sinusitis, bacteremias; and three possible
deaths
2
2003
P aeruginosa
Same as Srinivasan (above)
Probable pneumonia; details scarce and/or causality tenuous in these reports.
1
2003
*
APIC = Association for Professionals in Infection Control and Epidemiology; MAI =
M avium-intracellulare. Used with permission from Culver et al.
6
†
Formerly known as Pseudomonas pseudomallei.
The most common organisms implicated in bronchoscopy-related pseudoinfections include
bacterial pathogens such as P aeruginosa or S marcescens, contagious and noncontagious
mycobacteria, and environmental fungi
4
, (Table 1). Pseudomonas and Serratia species, and M tuberculosis are among the most
common organisms reported in true infections (Table 3).
Viruses can be nominally classified into two groups based on the presence or absence
of a lipid bilayer envelope. The latter provides a barrier to external digestive enzymes
and therefore more resistance to disinfection. Patient-to-patient transmission of
viral infections during bronchoscopy has not been reported, however. Interestingly,
transmission of hepatitis B as well as hepatitis C has been previously reported in
the setting of inadequately disinfected gastroendoscopes.
77
While HIV is readily inactivated using standard disinfection techniques, HIV-RNA can
be isolated from the bronchoscope after its use in patients with the virus. In addition,
intact papilloma virus DNA can be isolated in the vapor plume from laser photoresection.
78
,
79
,
80
,
81
These observations further illustrate the potential for viral transmission during
the FB procedure.
Reports of the transmission of airborne infections such as M tuberculosis or influenza
to health-care providers suggest that there may be an additional occupational risk
for bronchoscopy personnel. Recently, the outbreak of SARS-CoV afflicting health-care
providers during the severe acute respiratory syndrome epidemic illustrated the risks
of communicable airborne diseases. New evidence suggests that although transmission
appears to occur from infectious droplets, there are occasional episodes in which
airborne transmission cannot be excluded, including “aerosol”-generating procedures
such as intubation, bilevel positive airway pressure ventilation, and bronchoscopy.
82
,
83
,
84
,
85
Cases of direct transmission of infectious diseases to staff during bronchoscopy are
apparently rare. However, the increased incidence of latent tuberculosis among respiratory
therapists and pulmonary fellows compared with other health-care providers suggests
that there may be an increased risk to health-care providers who are involved in bronchoscopy
procedures.
86
,
87
Recommendations
We reemphasize that our review of the literature reveals that the infrequent and sporadic
recognition of bronchoscopy-related infection hinders development of evidence-based
guidelines for this topic. However, review of the available literature suggests that
all episodes are preventable. Many of the following recommendations are based on accumulated
clinical experience and expert opinions involving several disciplines. The following
recommendations have taken into account guidelines published by other societies and
organizations.
6
,
10
,
11
,
12
,
88
,
89
Review of the literature also points out several flaws in current practices; we highlight
measures to avoid such mistakes.
General Recommendations
All bronchoscopy personnel, including technical staff, physicians, and fellows should
be educated in infection control practices including “sharp precautions,” all reprocessing
steps, and material handling. All bronchoscopy personnel should be vaccinated against
influenza as well as hepatitis B and should undergo a surveillance purified protein
derivative test every 6 months as long as they are not tested positive at a prior
testing interval.
All bronchoscopes should be properly maintained according to the recommendations of
the manufacturer. The user manual should be easily accessible and provide information
on the use of specific bronchoscope models. Use of bronchoscopes that are fully immersible
and have disposable suction and biopsy valves is highly recommended. Nonimmersible
bronchoscopes and those with a reuseable valve should be replaced as soon as possible.
There are no reports of subacute bacterial endocarditis or that of joint infections
resulting from a bronchoscopy procedure. However, prophylactic antibiotic therapy
should be considered for patients at high risk for these complications (Table 4, Table
5
).
7
,
90
Patients with joint replacement within the past 2 years, history of previous prosthetic
joint infection, inflammatory arthropathy, hemophilia, malnutrition, insulin-dependent
diabetes mellitus, and immunocompromised status are susceptible to hematogenous total
joint infection and, therefore, may benefit from prophylactic antibiotics.
91
In view of the lack of data, application of such a practice could be recommended only
on an individual basis.
Table 4
Factors Associated With Bacterial Endocarditis*
Patients susceptible to bacterial endocarditis
Individual factors†
History of bacterial endocarditis
Prosthetic heart valves including bioprosthetic and homograft valves
Cyanotic congenital heart diseases
Rheumatic valve disease
Hypertrophic cardiomyopathy
Mitral valve prolapse with regurgitation
Surgically corrected systemic-pulmonary shunts or conduits
Immunocompromised patients with lower respiratory tract infection
Procedural factors
Diagnostic tests that would induce mucosal trauma (brushing, endobronchial or bronchoscopic
lung biopsy, transbronchial needle aspiration)
Therapeutic bronchoscopy causing mucosal trauma: laser photoresection, endobronchial
electrosurgery, balloon bronchoplasty, stent placement
Patients susceptible to hematogenous total joint infection
Joint replacement within past 2 yr
Previous prosthetic joint infection
Inflammatory arthropathy
Immunocompromized patient
Hemophilia
Malnutrition
Insulin-dependent diabetes mellitus
*
There are no reports of bacterial endocarditis or joint infections caused by the FB.
Prophylaxis against these conditions should be considered in susceptible patients
on individual basis.
†
Many center administer antibiotics during or immediately after bronchoscopic lung
biopsy in lung transplant recipients to prevent BAI. To date, there are no data to
support such practice.
Table 5
FDA and CDC Public Health Advisory*
1. Follow the instructions for cleaning the endoscope provided by the manufacturer.
2. Check with the manufacturers of the endoscope to determine whether the endoscope
can be reprocessed in an AER. Also, check with the manufacturer to determine whether
any specific steps are required prior to a specific type of endoscope being reprocessed
in an AER.
3. Compare the reprocessing instructions provided by the endoscope and AER manufacturers
and resolve any conflicting recommendations.
4. In the absence of specific technical instructions on automated reprocessing for
each model of endoscope used in the facility, be sure to follow the manual reprocessing
instructions of the endoscope manufacturer as well as the recommendations of the manufacturer
of the chemical germicides at the facility.
5. Regardless of manual reprocessing of the endoscope or use of an AER, consider incorporating
a final drying step in the protocol.
6. Ascertain that the instructions of the facility for preparing endoscopes for patient
contact are appropriate and that the staff is adhering to these instructions.
7. Provide comprehensive and intensive training for all staff assigned to reprocessing
endoscopes to ensure that they understand the importance of proper reprocessing of
all endoscopes used in the facility.
8. Implement a comprehensive quality control program.
*
Infections from endoscopes inadequately reprocessed by an AER system.
Infection control precautions (“standard precautions”) should be mandated for every
procedure and should include full barrier clothing (gown, gloves, mask, and eye shields)
and needle precautions for the bronchoscopist (Fig 1
). Use of a fit-tested N95 particulate respirator by the bronchoscopist or a higher-grade
respiratory precaution are recommended when mycobacterial infection is suspected.
For highly contagious agents (eg, SARS-CoV), a power air-purifying respirator (PAPR)
hood should be used (Fig 2
).
82
,
83
,
84
,
85
,
92
Prior to their use, medical clearance of proper fit-testing and familiarity with these
masks is essential for all bronchoscopy personnel. There must be compelling indications
for bronchoscopy in patients suspected of having highly contagious infections (see
“Patient Selection”). When bronchoscopy is to be performed in such patients, it should
be performed in a negative pressure-ventilated room, if one is available. These patients
should also be required to wear a surgical mask so that the risk of dissemination
of airborne infection can be minimized (Fig 3
).
Figure 1
Standard precautions for bronchoscopy, showing bronchoscopist with gown, gloves, mask,
and eye gear.
Figure 2
Facial masks used during bronchoscopy. Left, A, and right, A: The surgical mask helps
prevent spread of droplet infection from the surgeon into the open surgical wounds.
Left, B, and center, B: N95 particulate respirators help prevent spread of airborne
infection from patient to the bronchoscopy personnel, and filter 95% of particles
> 0.3 μm. Prefitting is required. Top, C: PAFR hood.
Figure 3
Bronchoscopy personnel wearing a PAPR. The respirator circulates filter air through
the hood.
Administration of adequate topical anesthetic agents or cough suppressants is recommended
to minimize coughing resulting in dissemination of airborne pathogens. Sharp metal
objects or needles should not be used to remove biopsy specimen from the forceps,
because this may increase the risk of transmission of blood borne pathogens.
77
A procedure log should be kept that includes the patient’s name and medical record
number, the bronchoscope used, the name of the bronchoscopist(s), and the automated
endoscope reprocessor (AER), if used, to assist in outbreak investigation. In addition,
a maintenance record must be maintained for each bronchoscope and AER in use.
Patient Selection
To prevent contamination of the instrument and aerosolization of highly pathogenic
organisms such as M tuberculosis or SARS-CoV,
82
,
83
,
84
,
85
every effort should be made to confirm the diagnosis by other techniques. Bronchoscopy
for the purpose of diagnosis should be postponed until at least three sputum or gastric
aspirate smear findings are negative for acid-fast bacilli in patients suspected of
having pulmonary tuberculosis.
93
Bronchoscopy Suite
The bronchoscopy suite as well as the postprocedure recovery areas should have engineering
controls that will allow ≥ 12 air exchanges per hour for new construction as of year
2001 or 6 exchanges per hour for construction before 2001 and be under negative pressure.
94
The air must be either discharged directly to the outside or to a monitored high-efficiency
particulate air (HEPA) filtration system before recirculation. The adequacy of these
air exchanges should be monitored on a regular basis with appropriate documentation
by the engineering and/or infection control personnel of the institution. Designated
“clean” and “dirty” (reprocessing) areas should be maintained in the bronchoscopy
suite to separate used instruments from clean ones. Bronchoscopy supervisory staff
should strictly implement and monitor such practices. No special preparation is required
prior to the reuse of a well-ventilated facility after performing a procedure in a
patient suspected of having infection with a virulent, airborne pathogen. All furniture
and equipment used during the procedure should be wiped down following each procedure
using hospital-approved cleaning solutions. The bronchoscopy suite floor should also
be wiped with hospital approved cleaning solutions at the end of each working day,
following the procedure appropriate for the soiling. If a procedure is being performed
in a nondesignated area, a portable HEPA filter should be used when appropriate (high
suspicion for airborne pathogens).
Reprocessing of the Bronchoscope
The sequence of reprocessing should be as follows.
Mechanical Cleaning
Mechanical cleaning begins immediately after the procedure to prevent drying or hardening
of organic debris.
10
,
95
Personal protective equipment (gown, gloves, mask, and eye shields) must be worn while
processing the contaminated bronchoscope. The outside of the bronchoscope should be
wiped with a detergent-soaked gauze piece and detergent solution suctioned through
the channel. All suction ports or biopsy attachments should be detached prior to leak
testing, further cleaning, and inspecting the instrument for any damage. All disposable
items must be discarded. The instrument should then be pressurized with a leak tester
to detect any damage that could have occurred during the procedure. The bronchoscope
is then fully immersed in water to confirm or to rule out any air leaks. Flexion and
extension of the bending section should be performed under water to detect any minute
leaks that may not be visualized otherwise. The presence of a leak indicates a breach
in the integrity of the external or the luminal surface of the instrument. Such puncture
sites and breaches will develop concretions of debris (blood, mucus) that cannot be
disinfected. Any instrument with a positive leak test result should not be reused
until fully repaired (Fig 4
). Only after ensuring that there are no air leaks should an enzymatic cleaner be
added to the water. Next, the bronchoscope should be soaked in cleaning solution for
approximately 5 min. The external surface of the bronchoscope should be cleaned/wiped
manually with an enzymatic detergent. Detergent solution or water is then flushed
through all ports to loosen organic debris. The detergent preparation should not be
reused. A cleaning brush of an appropriate size then should be passed multiple times
through all ports of the instrument. Meticulous cleaning alone achieves a 3.5-to 4-log
reduction in the organism load.
96
After brushing, channels should be flushed again to remove all loosened material.
Cleaning brushes should either be for single use or should receive mechanical cleaning
followed by sterilization or high-level disinfection after each use.
28
Finally the instrument and its channel should be rinsed with water to remove the enzymatic
cleaner and prepare it for disinfection.
Figure 4
Positive leak test result. Air bubbles emitting from the surface of the bronchoscope
indicate a breach in its exterior.
Following a procedure performed in a nondesignated area (ie, in the ICU), the external
surface of the bronchoscope should be wiped and its channel flushed with water. The
instrument should then be placed in a water-tight polyethylene bag and returned to
the bronchoscopy suite as soon as possible for a formal reprocessing. The bronchoscope
reprocessing should not be performed in the procedure area.
The decision to process a bronchoscope with high-level disinfection vs sterilization
is based on a classification system for medical devices that divides them by the risk
of infection transmission. The Spaulding classification
97
groups medical devices into three groups, allowing determination of the level of disinfection
for each. The following is a description of each group.
Critical Devices
Devices that normally enter sterile tissue or the vascular system (eg, cardiac catheters,
biopsy forceps). They should undergo sterilization, defined as the destruction of
all microbial life.
Semicritical Devices
Devices that come in contact with intact mucous membranes and do not normally penetrate
sterile tissue (eg, endoscopes). They should undergo high-level disinfection, defined
as the destruction of all vegetative microorganisms, mycobacteria, viruses, fungal
spores, and some, but not all, bacterial spores.
Noncritical Devices
Devices that do not ordinarily touch the patient or touch only intact skin (eg, stethoscopes
or patient carts). These items may be cleaned by low-level disinfection.
Disinfection
According to the above scheme, the FB is a semicritical device, as it seldom comes
in contact with breached mucosa or open surgical wounds.
98
A minimum of high-level disinfection is required before its reuse. Sterilization with
ethylene oxide gas, while highly effective, is likely to be associated with unacceptable
delays between procedures due to the prolonged sterilization process and the need
to aerate the instrument afterwards.
Disinfection is carried out either manually or by using an AER. Activated alkaline
gluteraldehyde, peracetic acid, and orthophthaldehyde are the acceptable chemicals
for disinfection.
99
,
100
,
101
A complete list of approved disinfectant formulations and maximum reuse time can be
found at www.fda.gov/cdrh/ode/germlab.html.
89
All currently approved agents are effective high-level disinfectants in experimental
conditions, by definition achieving a 6-log reduction in mycobacterial burden.
95
The choice of specific disinfectant can vary by institution, depending on cost, volume
of procedures, availability of AERs, and the number of bronchoscopes in use. Disinfection
for 20 min in 2% alkaline gluteraldehyde at 20°C (“20–2-20”) provides adequate disinfection
if cleaning with detergent precedes disinfection.
7
,
96
,
98
,
100
Meticulous cleaning and assiduous adherence to an appropriate protocol are much more
important determinants of successful disinfection than the choice of a specific approved
agent.
100
It is important to note that dilution of disinfectants occurs over time.
103
Depending on the formulation, the solutions may be reused for 14 to 28 days.
89
Potency of the solution should be periodically tested with commercially available
test kits and documented for proper record keeping.
Gluteraldehyde solution should be discarded after 14 days or 20 cycles and should
not be used if the concentration is < 2%; the solution should be tested at the beginning
of every day of use. Even if the concentration is adequate, the disinfectant should
not be used longer than the recommended period, since the aldehyde moeity will polymerize
over time attenuating its microbiocidal activity.
102
If the AER use is the primary mode of disinfection, the advisory issued by the US
Food and Drug Administration (FDA) and Centers for Disease Control and Prevention
(CDC) should be followed closely (Table 4).
89
It is essential to ensure compatibility between bronchoscopes and AERs.
49
,
65
,
70
Chemical and biological indicators (sporicidal tests) should be strictly followed
during the use of AERs. Automated reprocessors can be used for endoscopes other than
FB, provided all AER users adhere to acceptable reprocessing protocol. All AERs should
be properly maintained according to the recommendations of the manufacturer. User
manuals should be easily accessible and provide information on which specific bronchoscope
models have been tested for compatibility with the AER.
Postprocessing Procedure
Since many pathogens isolated from the bronchoscopes are from recontamination after
disinfection,
34
,
35
,
36
,
37
,
38
,
39
,
40
it is imperative to properly dry and store the instrument. Following disinfection
or sterilization, thorough rinsing of the internal channel with sterile or filtered
tap water is essential to prevent toxic effects of residual chemicals. Ideally, the
instrument is dried by purging the channel with 70% alcohol and compressed air.
97
Proper storage of the bronchoscope is an important step to prevent pathogen growth.
Flexible bronchoscopes should be hung vertically without valves attached, in a roomy
cabinet with adequate ventilation to prevent moisture accumulation. An additional
step to decrease moisture is to place the bronchoscope in a drying cabinet that utilizes
a desiccant to reduce relative humidity. The instrument should not be stored in its
carrying case, which should be used only for long-distance transportation. Since the
carrying case itself cannot be disinfected, it is unsuitable to maintain the bronchoscope
in a “patient-ready” condition. All instruments transported in the case must be reprocessed
before and after being placed in the case. For procedures performed at locations other
than the bronchoscopy suite, a sterile sealed polyethylene bag should be used for
transportation of the disinfected instrument.
Bronchoscopic Accessories
All nondisposable bronchoscopic accessories (such as forceps) that breach the bronchial
mucosa require sterilization following a thorough mechanical cleaning.
20
,
23
Because they cannot be properly sterilized, needles used for aspiration biopsy are
for single-patient-use only. All ancillary equipment (atomizers, filters, washers)
must be cleaned and maintained according to the recommendation of the manufacturer.
Reuse of atomizers between patients is unacceptable.
31
Multiuse vials (eg, canisters of benzocaine) used for topical anesthesia should be
wiped down with a hospital-approved disinfectant between use.
Rigid Bronchoscope
The rigid bronchoscope and ancillary equipment are made of durable steel or other
metals. They can be steam-sterilized using standard autoclave methods of sterilization.
Therefore, the potential for a rigid bronchoscope to be a vector of infection is minimal
to nonexistent.
Secondary Measures of Infection Prevention
Secondary measures of infection prevention are often overlooked but become increasingly
important for infection control during FB since total interdiction of pathogen transmission
is impossible. Factors that increase the propensity for infection to be spread by
bronchoscopy include increasing frequency of the procedure, duration and complexity
of the procedure, the expanding population of immunosuppressed patients, economic
pressures, and the hardiness of the pathogens. A review of prior outbreaks suggests
that a large proportion of cases in each instance might have been prevented with earlier
recognition of the problem. The mainstay of recognition, therefore, is effective pathogen
surveillance. Each institution should develop bronchoscopy-specific protocols for
surveillance for microbiological isolates. Input from bronchoscopists, infection control
specialists, and microbiologists is ideal. Relying solely on individual practitioners
to quickly identify trends in isolate patterns will prolong outbreaks, especially
for users of shared facilities.
Because the diagnostic bronchoscopy is often performed to evaluate patients with fever,
abnormal radiographic findings, and/or other signs of possible infection, it is not
always possible to conclude with confidence that FB has caused an infection since
these clinical indexes may wax and wane. If either a “true” or a “pseudo” infection
is encountered, the bronchoscopy team must inform the institutional infection control
officer, the bronchoscope manufacturer, the state health department, the FDA MedWatch
program, and the CDC, as well as the patient(s) and referring physicians. If contamination
is suspected, the instrument must be removed from service immediately and an investigation
begun by culturing parts of the bronchoscope, hospital tap water, and reprocessing
equipment. On the basis of this initial assessment, the infection control team and
bronchoscopy personnel should proceed as needed to assess and ameliorate any breach
in infection control practices. The use of routine environmental microbiological testing
of bronchoscope for quality assurance has not yet been established.
Conclusions
Spread of infection during FB is underrecognized and underreported. Unless proper
infection control practices are observed, the increasing numbers of procedures performed
are likely to be associated with more frequent episodes of infections attributed to
bronchoscopy. Even though the reported incidence is quite low, lethal outcomes can
result. Prevention of BAI requires increased vigilance by physicians, assiduous implementation
of reprocessing protocols, and closer collaboration between bronchoscopy personnel,
infection control practitioners, microbiology laboratories, and instrument manufacturers.
Formalized institutional monitoring of isolate patterns and epidemiologic analysis
may aid in the early detection of potential epidemics. The use of molecular biology
techniques such as DNA fingerprinting can help confirm such occurrences. Finally,
future innovations in bronchoscope design will require attention to the principles
of infection control. The bronchoscopist, health-care providers involved in bronchoscopy,
and equipment manufacturers should actively continue to develop techniques and systems
to prevent infections from FB. The following definitions and discussion are useful.
104
Definitions
Critical Devices
Devices that normally enter sterile tissue or the vascular system (eg, cardiac catheters,
biopsy forceps).
Semicritical Devices
Devices that come in contact with intact mucous membranes and do not normally penetrate
sterile tissue (eg, endoscopes).
Noncritical Devices
Devices that do not ordinarily touch the patient or touch only intact skin (eg, stethoscopes
or patient carts).
True Infection
Transmission of organisms during bronchoscopy causing new related illness in the patient
under going the procedure.
Pseudoinfection and Pseudoepidemic
Isolation of transmitted organisms in the bronchoscopy specimens obtained from the
patient or patients, without evidence of specific infection.
Low-Level Disinfection
Process that inactivates most bacteria, some viruses, and some fungi.
Intermediate-Level Disinfection
Process that inactivates M tuberculosis, vegetative bacteria, most viruses, and fungi
High-Level Disinfection
Process that achieves 6-log reduction in mycobacterial burden.
Sterilization
Complete elimination of all forms of microbial life, including bacterial and fungal
spores.
Recommendations for Precleaning, Disinfection, and Postprocessing of the FB
The following prerequisites are required: (1) proper education and training, and (2)
personal protective equipment must be worn.
Precleaning
Immediately After Use (at the Bedside)
(1) Flush water or saline solution through the channel for 20 s, making sure that
the distal tip of the scope does not rest in the fluid without suction being applied;
(2) wipe external surface of scope with wet gauze to remove any loose debris; and
(3) place contaminated scope loosely in a sealed water-tight bag labeled biohazard,
for transportation to the cleaning area, if necessary.
Cleaning, Disinfection, and Preparation
The following procedures should be followed: (1) Immediately after use (at beside),
flush water or saline through the channel for 20 s making sure that the distal tip
of the scope does not rest in the fluid without applying suction. (2) Transport the
instrument to a processing area as soon as possible to avoid drying of debris over
the instrument. (3) With proper personal protective equipment worn, remove the instrument
from the bag and place it in the basin for precleaning after placing water-tight caps
on to protect electrical components. (4) Remove all disposable parts (suction/biopsy
valve). (5) Perform leak test before immersing the instrument in water. (6) Any instrument
that fails the leak test must be removed from service until repaired. (7) Add enzymatic
cleaner to the water and soak the instrument for 5 min. (8) Using the enzymatic solution,
wipe external surfaces with wet gauze and flush the suction channel. (9) Insert an
appropriate-size cleaning brush through the channel of the instrument and brush all
ports until there is no more visible debris being removed from the instrument. Flush
the channel again to remove all loosened material. (10) Drain the enzymatic solution
from the basin. (11) Rinse all internal as well as external surfaces with water to
prepare the instrument for disinfection.
High-Level Disinfection
For high-level disinfection, do the following: (1) Place the bronchoscope in either
an AER or a basin used for manual disinfection. (2) Use only detergents and FDA-cleared
disinfectants that are compatible with the bronchoscope as well as the reprocessor.
Minimum effective concentration of the disinfectant solution must be checked with
each cycle using the available test strips. (3) While using the AER, ascertain that
proper connections are made with the internal channel of the instrument for the flushing
of the disinfectant. (4) The bronchoscope is fully immersed in the disinfectant solution,
exposing all surfaces to the solution for the proper time required. With the manual
system, the channel is filled with the disinfectant using a syringe containing the
solution.
Postprocessing Procedure
The following steps should be followed: (1) Once the proper time for the is met for
high-level disinfection, the bronchoscope and its channel are rinsed with either sterile
or filtered tap water according to the recommendations disinfectant supplier. (2)
Proper drying of the channel is accomplished by purging 70% alcohol with forced air.
(3) Remove the water-tight caps from the instrument and hang vertically in a storage
cabinet devoid of any valves. (4) Proper documentation related to the use and the
disinfection of the instrument (medical record number of the patient, date of the
procedure, bronchoscopist performing the procedure, model and the serial number of
the scope, and the date of reprocessing) is maintained for infection control purposes.