Flexible bronchoscopy is a minimally invasive procedure that has become an integral part of pulmonology. Reference Miller, Casal, Lazarus, Ost and Eapen1 With the ability to perform bronchoscopy in settings ranging from dedicated endoscopy suites to the bedside, hundreds of thousands of procedures are carried out globally each year. Reference Muscarella2,Reference Lee, Driver and Prekker3
Due to the risk of infection transmission, a dedicated disinfection protocol is essential for a successful bronchoscopy service. The protocol should incorporate a structured approach that includes specific hardware, such as automatic endoscope reprocessors (AERs), specific reprocessing and turnaround times, strict adherence to the manufacturer’s instructions, and dedicated personnel. Reference Ofstead, Quick and Wetzler4 Despite adherence to rigorous protocols, there is a risk of encountering various flaws that can result in bronchoscopy-related complications, including outbreaks and pseudo-outbreaks. Reference Terjesen, Kovaleva and Ehlers5
An outbreak is defined as the increase in identified infections above the baseline rate, whereas a pseudo-outbreak refers to the isolation of microorganisms in specimens without any indication of infection. Reference Sood and Perl6 They occur as a result of contamination or colonization of the bronchoscope and not an actual infection. Although pseudo-outbreaks are not genuine infections, they can still have adverse effects because patients may receive unnecessary antimicrobial therapy. Reference Galdys, Marsh and Delgado7
Given the burden of these complications, we conducted a systematic review to identify and elucidate all published reports of bronchoscopy-related outbreaks and pseudo-outbreaks. We also sought to uncover any correlations between specific pathogens and sources of contamination and to determine whether they are linked to the categorization of a study event as an outbreak or a pseudo-outbreak.
Materials and methods
This systematic review was conducted according to the international Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. Reference Page, McKenzie and Bossuyt8 The study protocol has been published in the international prospective register of systematic reviews PROSPERO (https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=375610, registration no. CRD42022375610).
Data sources and searches
A systematic literature review of articles published in PubMed was conducted from inception to December 31, 2022, using the following search terms: bronchoscopy [title/abstract] AND outbreak[title/abstract]; bronchoscope [title/abstract] AND outbreak[title/abstract]; bronchoscopy [title/abstract] AND pseudo-outbreak[title/abstract]; bronchoscope [title/abstract] AND pseudo-outbreak[title/abstract]. No restrictions were applied regarding the date of publication or language. Article titles and abstracts were screened based on the selection criteria presented in Table 1. References from the articles extracted were reviewed to identify further relevant publications.
Table 1. Inclusion and Exclusion Criteria of Study Selection
Study selection
The study selection process was conducted by 2 reviewers (L.K. and S.E.) who applied the selection criteria summarized in Table 1 to determine whether a study was eligible for inclusion in the review. Studies that met the following 4 inclusion criteria were then further evaluated for their relevance and quality:
-
(1) Study design: Case-control studies, case series, and cohort studies describing outbreaks or pseudo-outbreaks following bronchoscopy.
-
(2) Type of participants: Adult patients having positive cultures from samples obtained during bronchoscopy.
-
(3) Type of exposure: Bronchoscopy regardless of indication.
-
(4) Type of outcome: Primary outcomes were the source(s) of contamination and the organisms involved in bronchoscopy-associated outbreaks and pseudo-outbreaks, as well as any correlations between them. Secondary outcomes of interest included whether individual pathogens were more likely to be associated with outbreaks or pseudo-outbreaks, whether contamination sources were associated with specific organisms, and the use of antibiotics among patients involved in outbreaks and pseudo-outbreaks.
Data extraction and quality assessment
The following data were extracted from full-text articles included in this review: eligibility criteria, study design, study event (whether the study event represented an outbreak or a pseudo-outbreak), pathogens involved, the duration of the study event and the source of the contamination.
The risk of bias was assessed independently by 3 authors (L.K., S.E., and F.S.) using the Murad scale, a modified version of the Newcastle-Ottawa Scale (NOS) evaluating nonrandomized tirals, Reference Wells, Shea and O’Connell9 using 6 evidence-based criteria evaluating for selection, representativeness of cases, ascertainment of outcomes and exposure, and adequate reporting. Reference Murad, Sultan, Haffar and Bazerbachi10 The questions utilized for the assessment of each study are presented in Supplementary Table S1 (online). In cases of disagreement between the reviewers, studies were discussed among the group until an agreement was reached. A study was considered to have low risk of bias if all 6 criteria were met, moderate risk of bias when 4 or 5 criteria and met and high risk of bias when 3 or fewer criteria were met. Reference Salah11 The scores of each study are presented in Supplementary Table S2 (online).
Data synthesis and analysis
Data pertaining to the primary and secondary end points were extracted; they have been presented using a narrative and analytical synthesis. SPSS for Windows version 28 (IBM, Armonk, NY) was used for statistical analysis. For continuous variables, means were reported for parametric variables and medians were used for nonparametric variables.
Statistical analyses were also conducted to evaluate whether the presence of a specific pathogen affected the possibility that the study event would be considered an outbreak or a pseudo-outbreak and whether the source of contamination was associated with a specific type of pathogen. This determination was achieved by extracting data as binary variables and using Pearson correlation to identify correlations between variables. P values were calculated using the Fisher exact test (2-sided). Only P values <.05 were considered significant.
Results
The initial PubMed search yielded 206 articles, and an additional 64 articles were identified through citation searching. Following the screening process depicted in Figure 1, a total of 74 studies were deemed to meet the inclusion criteria and were included in this review. Data extracted are summarized in Table 2, and the quality assessments of all studies are presented in Supplementary Table S2 (online).
Figure 1. A diagram representing the assessed studies in accordance with the globally recognized Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. Reference Page, McKenzie and Bossuyt8
Table 2. Parameters Extracted From Each Study, Outlining the Patient Population, Type of Outbreak and Duration, Pathogen Identified and Source of Contamination a
Note. AER, automated endoscope reprocessor; OR, odds ratio; CI, confidence interval.
a Numbers presented here include the index case as part of the affected patients.
Population characteristics
Overall, 74 studies describing 23 outbreaks and 52 pseudo-outbreaks were included; one study described 2 separate pseudo-outbreaks within their institution from different pathogens, which were counted as separate entries. Reference Cêtre, Nicolle and Salord12 Study characteristics are presented in Table 3. During the defined event period, 7,105 patients had undergone bronchoscopy. Among them, 1,521 (20.3%) were part of an outbreak or a pseudo-outbreak and were classified as having been affected by the study event. The median number of patients described in studies of outbreaks was 15, with a median 10 being affected. The corresponding medians for pseudo-outbreaks were 21 and 14, respectively (Table 3). The median duration of was 91.3 days for outbreaks and 94.1 for pseudo-outbreaks.
Table 3. Study Characteristics and Findings, Divided Based on Study Event
Note. NTM, nontuberculous mycobacteria; AER, automated endoscope reprocessor.
a Units unless otherwise specified.
b Total patients: Number of patients described by the studies, which include both patients affected by outbreaks or pseudo-outbreaks, as well as control patients that had undergone bronchoscopy during the same period but remained unaffected by the study event.
c Affected patients: patients that were part of an outbreak or a pseudo-outbreak following bronchoscopy.
Sources of contamination
We identified 5 major categories of contamination sources during the review: (1) use of contaminated water or ice (34.7% of studies), (2) defect in the bronchoscope itself or one of its parts (32%), (3) defect or contamination of the AER (28%), (4) inadequate disinfection due to deviation from disinfection protocol (18.7%), and (5) use of contaminated aerosolized topical anesthetics in preparation for the procedure (5.4%).
Water contamination was the most commonly identified source of contamination; it was reported in 26 of 74 studies, Reference Schuetz, Hughes and Howard13–Reference Cox, deBorja and Bach39 24 (92.3%) of which were pseudo-outbreaks. Reference Schuetz, Hughes and Howard13–Reference Zhang, Zhou, Jiang, Wang, Li and Huang15,Reference Schaffer, Fitzgerald, Commane, Maguiness and Fenelon17–Reference Richardson, Rothburn and Roberts20,Reference Rosengarten, Block and Hidalgo-Grass22–Reference Cox, deBorja and Bach39 In 13 (50%) of these studies, the pathogens were nontuberculous mycobacteria (NTM). Reference Guy, Vanhems and Dananché16,Reference Kiely, Sheehan, Cryan and Bredin23,Reference Gubler, Salfinger and von Graevenitz25,Reference Bennett, Peterson, Johnson, Hall, Robinson-Dunn and Dietrich27–Reference Nye, Chadha, Hodgkin, Bradley, Hancox and Wise29,Reference Stine, Harris, Levin, Rivera and Kaplan31–Reference Cox, deBorja and Bach39 Notably, 4 pseudo-outbreaks were attributed to the use of contaminated ice. The ice was utilized in the preparation of cold saline solutions, which were used to control bleeding during interventional bronchoscopy procedures. Reference Schuetz, Hughes and Howard13,Reference Kioski, Montefour and Saubolle14,Reference Blake, Embil, Trepman, Adam, Myers and Mutcher19,Reference Bringhurst, Weber and Miller32 In all 4 cases, contaminated ice (obtained from a nonsterile, common-use ice machines) came in contact with sterile normal saline, while in all 4 cases, the pseudo-outbreaks were terminated by changing this practice. Reference Schuetz, Hughes and Howard13,Reference Kioski, Montefour and Saubolle14,Reference Blake, Embil, Trepman, Adam, Myers and Mutcher19,Reference Bringhurst, Weber and Miller32 The identification of a contaminated water as the source of contamination was significantly correlated with the characterization of the study as a pseudo-outbreak (r = 0.363; P = .001), and by the identification of Legionella pneumophila (r = 0.280; P = .015) or NTM (r = 0.331; P = .004) as the pathogens of interest (Table 4).
Table 4. Results of Pearson Correlation Analysis
Note. NTM, nontuberculous mycobacteria; AER, automated endoscope reprocessor.
AER contamination was identified in 21 studies. Reference Rosengarten, Block and Hidalgo-Grass22–Reference Gubler, Salfinger and von Graevenitz25,Reference Takigawa, Fujita and Negayama28,Reference Seidelman, Wallace and Iakhiaeva33,Reference Scorzolini, Mengoni and Mastroianni34,Reference Chroneou, Zimmerman and Cook36–Reference Fernandes Garcia de Carvalho, Rodrigues Mestrinari and Brandão38,Reference Botana-Rial, Leiro-Fernández and Núñez-Delgado40–Reference Brown, Reeves, Hellyar and Harvey50 Among them, 10 pertained to a combination of AER and water source contamination. Reference Rosengarten, Block and Hidalgo-Grass22–Reference Gubler, Salfinger and von Graevenitz25,Reference Takigawa, Fujita and Negayama28,Reference Seidelman, Wallace and Iakhiaeva33,Reference Scorzolini, Mengoni and Mastroianni34,Reference Chroneou, Zimmerman and Cook36–Reference Fernandes Garcia de Carvalho, Rodrigues Mestrinari and Brandão38 In the majority of these, the contaminated water source led to the contamination of the AER itself. Presence of AER contamination was significantly correlated with pseudo-outbreaks (r = 0.286; P = .013) and with the presence of NTM (r = 0.446; P < .001).
In 24 studies, the source of contamination was attributed to a defect in the bronchoscope itself. Reference Galdys, Marsh and Delgado7,Reference Cêtre, Nicolle and Salord12,Reference Guy, Vanhems and Dananché16,Reference Guimarães and Chimara43,Reference Campagnaro, Teichtahl and Dwyer46,Reference Blanc, Parret, Janin, Raselli and Francioli49–Reference Cosgrove, Ristaino and Caston-Gaa66 Bronchoscope segments identified as being defective or damaged included the lumen, the suction valve and channel, the biopsy port and its cap, as well as the biopsy channel itself. Notably, in 1 study the defect at the entry port of the biopsy channel was so significant that it led to a recall of the specific model by the manufacturer. Reference Cêtre, Nicolle and Salord12 The identification of Pseudomonas aeruginosa (r = 0.351; P = .002) or Klebsiella pneumoniae (r = 0.346; P = .002) as the causative agents were correlated with the presence of a bronchoscope defect.
Another source of contamination identified was the inadequate disinfection of the bronchoscope. This was reported in 14 studies, Reference Zhang, Zhou, Jiang, Wang, Li and Huang15,Reference Guy, Vanhems and Dananché16,Reference Blake, Embil, Trepman, Adam, Myers and Mutcher19,Reference Young, Sabel and Epidemiologic21,Reference Kressel and Kidd42,Reference Schelenz and French44,Reference Goldstein and Abrutyn59,Reference Jereb, Burwen and Dooley67–Reference Larson, Lambert, Stricof, Driscoll, McGarry and Ridzon73 50% of which pertained to outbreaks. Reference Guy, Vanhems and Dananché16,Reference Young, Sabel and Epidemiologic21,Reference Schelenz and French44,Reference Jereb, Burwen and Dooley67,Reference Bou, Aguilar and Perpiñán68,Reference Vandenbroucke-Grauls, Baars, Visser, Hulstaert and Verhoef72,Reference Larson, Lambert, Stricof, Driscoll, McGarry and Ridzon73 In 6 studies, an epidemiologic review identified deviations from the appropriate disinfection protocol Reference Zhang, Zhou, Jiang, Wang, Li and Huang15,Reference Schelenz and French44,Reference Bou, Aguilar and Perpiñán68,Reference Larson, Lambert, Stricof, Driscoll, McGarry and Ridzon73–Reference Kressel and Kidd75 that led to the contamination. Examples included deviation from the disinfection protocol by the team maintaining the bronchoscopes on the weekends, Reference Bou, Aguilar and Perpiñán68 inadequate maintenance of the AER, Reference Schelenz and French44 lack of a dedicated technician for bronchoscope disinfection, and workarounds utilized by staff when the suite’s AER had broken down. Reference Kolmos, Lerche, Kristoffersen and Rosdahl74
Finally, in 4 studies, the source of contamination was related to the use of topical anesthetics. Reference Steere, Corrales and Von Graevenitz76–Reference Southwick, Hoffmann, Ferree, Matthews and Salfinger79 Mechanisms of contamination included contamination of the anesthetic solution itself, Reference Steere, Corrales and Von Graevenitz76 cleaning of anesthetic sprayers with contaminated tap water, Reference Cox, deBorja and Bach78 not sterilizing the bottles containing the anesthetic, Reference Schleupner and Hamilton77 or contamination of the atomizer used to apply the anesthetic. Reference Southwick, Hoffmann, Ferree, Matthews and Salfinger79 Although 3 of the studies reported pseudo-outbreaks caused by NTM Reference Steere, Corrales and Von Graevenitz76,Reference Cox, deBorja and Bach78 and fungi, Reference Schleupner and Hamilton77 1 study described an outbreak where 3 patients were infected with Mycobacterium tuberculosis. Reference Southwick, Hoffmann, Ferree, Matthews and Salfinger79 This organism was traced back to a contaminated atomizer that had been used to spray lidocaine in preparation for bronchoscopy. Reference Southwick, Hoffmann, Ferree, Matthews and Salfinger79
Pathogens
Pathogens identified as causing outbreaks or pseudo-outbreaks included bacteria in 54.6% of studies, mycobacteria in 41.3%, fungi in 10.6%, and viruses in 2.7%. Bacteria included P. aeruginosa (identified in 24% of studies), Stenotrophomonas maltophilia (5.3%), K. pneumoniae (5.3%), Serratia marcescens (6.7%), L. pneumophila (4%), and Acinetobacter spp (4%). Notably, 2 studies reported outbreaks with carbapenem-resistant strains of K. pneumoniae Reference Zweigner, Gastmeier, Kola, Klefisch, Schweizer and Hummel55 and P. aeruginosa. Reference Sorin, Segal-Maurer, Mariano, Urban, Combest and Rahal85 Mycobacteria consisted of M. tuberculosis, identified in 12% of studies and NTM, identified in 29.3%.
The most common type of pathogen identified overall were NTM (22 studies, 29.3%). Reference Kiely, Sheehan, Cryan and Bredin23,Reference Gubler, Salfinger and von Graevenitz25,Reference Bennett, Peterson, Johnson, Hall, Robinson-Dunn and Dietrich27–Reference Nye, Chadha, Hodgkin, Bradley, Hancox and Wise29,Reference Stine, Harris, Levin, Rivera and Kaplan31–Reference Cox, deBorja and Bach39,Reference Kressel and Kidd42,Reference Guimarães and Chimara43,Reference Campagnaro, Teichtahl and Dwyer46,Reference Maloney, Weibel and Daves48,Reference Brown, Reeves, Hellyar and Harvey50,Reference Wheeler, Lancaster and Kaiser60,Reference Steere, Corrales and Von Graevenitz76,Reference Wang, Liaw, Yang, Kuo and Luh80 The most commonly isolated species was M. chelonae, Reference Kiely, Sheehan, Cryan and Bredin23,Reference Gubler, Salfinger and von Graevenitz25,Reference Takigawa, Fujita and Negayama28,Reference Nye, Chadha, Hodgkin, Bradley, Hancox and Wise29,Reference Chroneou, Zimmerman and Cook36,Reference Fraser, Jones, Murray, Medoff, Zhang and Wallace37,Reference Kressel and Kidd42,Reference Campagnaro, Teichtahl and Dwyer46,Reference Cox, deBorja and Bach78,Reference Wang, Liaw, Yang, Kuo and Luh80 (10 studies, 13.5%) followed by M. abscessus, Reference Fernandes Garcia de Carvalho, Rodrigues Mestrinari and Brandão38,Reference Guimarães and Chimara43,Reference Maloney, Weibel and Daves48 M. fortuitum, Reference Campos-Gutiérrez, Ramos-Real, Abreu, Jiménez and Lecuona35,Reference Fernandes Garcia de Carvalho, Rodrigues Mestrinari and Brandão38,Reference Cox, deBorja and Bach78 and M. gordonae Reference Gubler, Salfinger and von Graevenitz25,Reference Scorzolini, Mengoni and Mastroianni34,Reference Cox, deBorja and Bach78 (3 studies each, 4%). With the exception of a single study by Wheeler et al, Reference Wheeler, Lancaster and Kaiser60 which involved isolation of both M. avium and M. tuberculosis, all other studies involving NTMs were classified as pseudo-outbreaks by their investigators. Reference Kiely, Sheehan, Cryan and Bredin23,Reference Gubler, Salfinger and von Graevenitz25,Reference Bennett, Peterson, Johnson, Hall, Robinson-Dunn and Dietrich27–Reference Nye, Chadha, Hodgkin, Bradley, Hancox and Wise29,Reference Stine, Harris, Levin, Rivera and Kaplan31–Reference Cox, deBorja and Bach39,Reference Kressel and Kidd42,Reference Guimarães and Chimara43,Reference Campagnaro, Teichtahl and Dwyer46,Reference Maloney, Weibel and Daves48,Reference Brown, Reeves, Hellyar and Harvey50,Reference Steere, Corrales and Von Graevenitz76,Reference Wang, Liaw, Yang, Kuo and Luh80 Also, in the vast majority of studies that NTM were identified, they were associated with either use of contaminated water (5 studies), Reference Bennett, Peterson, Johnson, Hall, Robinson-Dunn and Dietrich27,Reference Nye, Chadha, Hodgkin, Bradley, Hancox and Wise29,Reference Stine, Harris, Levin, Rivera and Kaplan31,Reference Campos-Gutiérrez, Ramos-Real, Abreu, Jiménez and Lecuona35,Reference Cox, deBorja and Bach78 deficiencies in the AER (5 studies), Reference Kressel and Kidd42,Reference Guimarães and Chimara43,Reference Campagnaro, Teichtahl and Dwyer46,Reference Maloney, Weibel and Daves48,Reference Brown, Reeves, Hellyar and Harvey50 or both (8 studies). Reference Kiely, Sheehan, Cryan and Bredin23,Reference Gubler, Salfinger and von Graevenitz25,Reference Takigawa, Fujita and Negayama28,Reference Seidelman, Wallace and Iakhiaeva33,Reference Scorzolini, Mengoni and Mastroianni34,Reference Chroneou, Zimmerman and Cook36–Reference Fernandes Garcia de Carvalho, Rodrigues Mestrinari and Brandão38 The presence of NTM was significantly associated with pseudo-outbreaks (r = 0.365; P = .001), water contamination (r = 0.331; P = .004), and AER dysfunction (r = 0.446; P < .001).
In contrast, M. tuberculosis was a considerably more significant pathogen. Among the 9 studies Reference Gubler, Salfinger and von Graevenitz25,Reference Prigogine, Glupczynski, Van Molle and Schmerber47,Reference Ramsey, Oemig, Davis, Massey and Török54,Reference Wheeler, Lancaster and Kaiser60,Reference Jereb, Burwen and Dooley67,Reference Larson, Lambert, Stricof, Driscoll, McGarry and Ridzon73,Reference Southwick, Hoffmann, Ferree, Matthews and Salfinger79,Reference Agerton, Valway and Gore81,Reference Nelson, Larson, Schraufnagel and Jackson82 in which M. tuberculosis was detected, 7 were considered outbreaks. Reference Ramsey, Oemig, Davis, Massey and Török54,Reference Wheeler, Lancaster and Kaiser60,Reference Jereb, Burwen and Dooley67,Reference Larson, Lambert, Stricof, Driscoll, McGarry and Ridzon73,Reference Southwick, Hoffmann, Ferree, Matthews and Salfinger79,Reference Agerton, Valway and Gore81,Reference Nelson, Larson, Schraufnagel and Jackson82 Regarding sources of contamination, inadequate disinfection was identified as the cause in 2 cases, Reference Jereb, Burwen and Dooley67,Reference Larson, Lambert, Stricof, Driscoll, McGarry and Ridzon73 the presence of a bronchoscope defect was identified in another 2 cases, Reference Ramsey, Oemig, Davis, Massey and Török54,Reference Wheeler, Lancaster and Kaiser60 water contamination in 1 case, Reference Gubler, Salfinger and von Graevenitz25 AER dysfunction in 2 cases, Reference Gubler, Salfinger and von Graevenitz25,Reference Prigogine, Glupczynski, Van Molle and Schmerber47 and the use of a contaminated atomizer in 1 case. Reference Southwick, Hoffmann, Ferree, Matthews and Salfinger79 In 2 cases, a source was not identified. Reference Agerton, Valway and Gore81,Reference Nelson, Larson, Schraufnagel and Jackson82 The only significant correlation regarding M. tuberculosis was with the study event being identified as an outbreak (r = 0.377; P = .001).
Among bacteria, the most commonly isolated was P. aeruginosa, which was reported in 18 studies. Reference Galdys, Marsh and Delgado7,Reference Cêtre, Nicolle and Salord12,Reference Zhang, Zhou, Jiang, Wang, Li and Huang15,Reference Guy, Vanhems and Dananché16,Reference Schelenz and French44,Reference Sorin, Segal-Maurer, Mariano, Urban, Combest and Rahal45,Reference Blanc, Parret, Janin, Raselli and Francioli49,Reference Kirschke, Jones and Craig52,Reference Severino and Magalhães53,Reference Srinivasan, Wolfenden and Song56,Reference DiazGranados, Jones and Kongphet-Tran57,Reference Sammartino, Israel and Magnussen63,Reference Cosgrove, Ristaino and Caston-Gaa66,Reference Bou, Aguilar and Perpiñán68,Reference Kolmos, Lerche, Kristoffersen and Rosdahl69,Reference Silva, Magalhães, Pereira, Kawagoe, Ikura and Ganc71,Reference Alipour, Karagoz and Taner83 It was not uncommon for P. aeruginosa to be isolated along with other bacteria, such as K. pneumoniae, Reference Galdys, Marsh and Delgado7,Reference Cetre, Salord and Vanhems61 S. marcescens, Reference Severino and Magalhães53,Reference Silva, Magalhães, Pereira, Kawagoe, Ikura and Ganc71,Reference Kirschke, Jones and Craig84 or S. maltophilia. Reference Cosgrove, Ristaino and Caston-Gaa66 P. aeruginosa caused outbreaks in 66.6% of studies, Reference Galdys, Marsh and Delgado7,Reference Guy, Vanhems and Dananché16,Reference Schelenz and French44,Reference Severino and Magalhães53,Reference Srinivasan, Wolfenden and Song56,Reference DiazGranados, Jones and Kongphet-Tran57,Reference Cetre, Salord and Vanhems61,Reference Sammartino, Israel and Magnussen63,Reference Bou, Aguilar and Perpiñán68,Reference Alipour, Karagoz and Taner83–Reference Sorin, Segal-Maurer, Mariano, Urban, Combest and Rahal85 and its presence was associated with bronchoscope defects Reference Galdys, Marsh and Delgado7,Reference Cêtre, Nicolle and Salord12,Reference Guy, Vanhems and Dananché16,Reference Blanc, Parret, Janin, Raselli and Francioli49,Reference Severino and Magalhães53,Reference Srinivasan, Wolfenden and Song56,Reference DiazGranados, Jones and Kongphet-Tran57,Reference Cetre, Salord and Vanhems61,Reference Sammartino, Israel and Magnussen63,Reference Cosgrove, Ristaino and Caston-Gaa66,Reference Kirschke, Jones and Craig84 or inadequate disinfection of the bronchoscope. Reference Zhang, Zhou, Jiang, Wang, Li and Huang15,Reference Schelenz and French44,Reference Bou, Aguilar and Perpiñán68,Reference Silva, Magalhães, Pereira, Kawagoe, Ikura and Ganc71,Reference Kolmos, Lerche, Kristoffersen and Rosdahl74 Significant correlations were identified between the presence of P. aeruginosa and the study event being an outbreak (r = 0.439; P < .001) and with a bronchoscope defect being identified as the source of contamination (r = 0.351; P = .002).
Other bacteria identified included L. pneumophila Reference Schuetz, Hughes and Howard13,Reference Kioski, Montefour and Saubolle14,Reference Mitchell, Hicks, Chiew, Montanaro and Chen24 in 3 cases, K. pneumoniae Reference Galdys, Marsh and Delgado7,Reference Cêtre, Nicolle and Salord12,Reference Zweigner, Gastmeier, Kola, Klefisch, Schweizer and Hummel55,Reference Cetre, Salord and Vanhems61 in 4 cases, S. marcescens Reference Siegman-Igra, Inbar and Campus30,Reference Kirschke, Jones and Craig52,Reference Severino and Magalhães53,Reference Silva, Magalhães, Pereira, Kawagoe, Ikura and Ganc71,Reference Vandenbroucke-Grauls, Baars, Visser, Hulstaert and Verhoef86 in 5 cases, S. maltophilia Reference Botana-Rial, Leiro-Fernández and Núñez-Delgado40,Reference Ece, Erac, Limoncu, Baysak, Oz and Ceylan41,Reference Waite, Georgiou, Abrishami and Beck51,Reference Cosgrove, Ristaino and Caston-Gaa66 in 4 cases and A. baumannii in 1 case. Reference Ardoino, Zangirolami and Iemmi98 Significant correlations were identified between the identification of K. pneumoniae and the presence of a bronchoscope defect (r = 0.346; P = .002) or the study event being an outbreak (r = 0.228; P = .049) and between L. pneumophila and water contamination (r = 0.280; P = .015).
In addition, 8 studies have reported pseudo-outbreaks related to fungi. Among the isolated organisms, Rhodotorula rubra was identified in 3 studies, whereas all other species were reported in only 1 study each. Reference Hoffmann, Weber and Rutala26,Reference Hagan, Klotz, Bartholomew, Potter and Nelson62,Reference Whitlock, Dietrich, Steimke and Tenholder87 These included Fusarium solani, Reference Schaffer, Fitzgerald, Commane, Maguiness and Fenelon17 Rhinocladiella similis, Reference Abdolrasouli, Gibani and de Groot58 Phaeoacremonium parasiticum, Reference Blake, Embil, Trepman, Adam, Myers and Mutcher19 Aureobasidium spp. Reference Wilson, Everts, Kirkland and Sexton88 In 1 study, both Trichosporon cutaneum and Penicillium spp were isolated. Reference Schleupner and Hamilton77 The only significant correlation identified was between the presence of fungi and pseudo-outbreaks (r = 0.230; P = .047).
Lastly, 2 studies reported viral pseudo-outbreaks. Both pseudo-outbreaks involved adenovirus and were attributed to a bronchoscope defect. Reference Hellinger, Parth and Smith64,Reference Seidelman, Akinboyo and Taylor65
Use of antibiotics
Information regarding antibiotic administration was available from 56 studies. Among them, 18 pertained to outbreaks and 38 to pseudo-outbreaks. From a total of 489 patients that were involved in outbreak studies, data on antibiotic use were available for 273, of whom 169 (61.9%) received antibiotics. Among 1,032 patients described in pseudo-outbreak studies, information on antibiotic use was available for 194, of whom 65 (33.5%) received antibiotics. Notably, among the 18 outbreak studies with available antibiotic usage data, all reported at least 1 patient receiving antibiotics. Reference Galdys, Marsh and Delgado7,Reference Guy, Vanhems and Dananché16,Reference Ramsey, Oemig, Davis, Massey and Török54,Reference Zweigner, Gastmeier, Kola, Klefisch, Schweizer and Hummel55,Reference DiazGranados, Jones and Kongphet-Tran57,Reference Wheeler, Lancaster and Kaiser60,Reference Cetre, Salord and Vanhems61,Reference Sammartino, Israel and Magnussen63,Reference Jereb, Burwen and Dooley67,Reference Vandenbroucke-Grauls, Baars, Visser, Hulstaert and Verhoef72,Reference Larson, Lambert, Stricof, Driscoll, McGarry and Ridzon73,Reference Southwick, Hoffmann, Ferree, Matthews and Salfinger79,Reference Alipour, Karagoz and Taner83,Reference Sorin, Segal-Maurer, Mariano, Urban, Combest and Rahal85,Reference Srinivasan, Wolfenden and Song89–Reference Dennison, Compston and Flahive91 Among the 38 studies describing pseudo-outbreaks that reported data on antibiotic use, 14 described at least 1 patient being on antibiotics. Reference Kioski, Montefour and Saubolle14,Reference Gubler, Salfinger and von Graevenitz25,Reference Takigawa, Fujita and Negayama28,Reference Scorzolini, Mengoni and Mastroianni34,Reference Campos-Gutiérrez, Ramos-Real, Abreu, Jiménez and Lecuona35,Reference Botana-Rial, Leiro-Fernández and Núñez-Delgado40,Reference Ece, Erac, Limoncu, Baysak, Oz and Ceylan41,Reference Guimarães and Chimara43,Reference Brown, Reeves, Hellyar and Harvey50,Reference Waite, Georgiou, Abrishami and Beck51,Reference Cosgrove, Ristaino and Caston-Gaa66,Reference Weinstein, Bone and Ruth70,Reference Kressel and Kidd75,Reference Peaper, Havill and Aniskiewicz92 Among these studies, 7 pertained to pseudo-outbreaks by NTM. Reference Gubler, Salfinger and von Graevenitz25,Reference Takigawa, Fujita and Negayama28,Reference Scorzolini, Mengoni and Mastroianni34,Reference Campos-Gutiérrez, Ramos-Real, Abreu, Jiménez and Lecuona35,Reference Guimarães and Chimara43,Reference Brown, Reeves, Hellyar and Harvey50,Reference Kressel and Kidd75 Factors positively correlated with the percentage of patients receiving antibiotics were the characterization of the study event as an outbreak (r = 0.617; P < .001) and the identification of M. tuberculosis (r = 0.479; P < .001) as the causative agent.
Discussion
This review presents an analysis of studies describing outbreaks and pseudo-outbreaks following bronchoscopy. Based on the studies identified, pseudo-outbreaks were more than twice as prevalent as outbreaks. Distinct patterns of pathogens and sources of contamination appeared to be linked to either outbreaks or pseudo-outbreaks. The isolation of P. aeruginosa, K. pneumoniae, and M. tuberculosis was more prevalent in outbreaks, as was the recognition of bronchoscope dysfunction and inadequate disinfection as sources of contamination. Conversely, the isolation of NTM was significantly more common in pseudo-outbreaks, whereas S. maltophilia, L. pneumophila, viruses, and fungi were exclusively isolated from studies describing pseudo-outbreaks. Water contamination and AER dysfunction were the sources of contamination more commonly associated with pseudo-outbreaks.
Furthermore, we were able to identify correlations between pathogens and sources of contamination. Specifically, P. aeruginosa and K. pneumoniae were correlated with primary bronchoscope defects, whereas NTM and L. pneumophila were correlated with the presence of a contaminated water source. Although these correlations are by no means absolute, they can serve as an initial guide to clinicians, infection preventionists, and hospital epidemiologists tasked with identifying the source of an outbreak or pseudo-outbreak.
The findings of this study also have significant implications from an antimicrobial stewardship perspective. Firstly, among studies that reported data on antibiotic use, 33.5% of patients involved in pseudo-outbreaks had been prescribed antibiotics. This finding implies that several patients had received unnecessary antibiotics as a result of a pseudo-outbreak, exposing them to the adverse effects of these agents, as well as leading to the development of antibiotic-resistant bacteria. Reference Galdys, Marsh and Delgado7,Reference Llor and Bjerrum93 Secondly, 2 studies described outbreaks involving highly resistant bacteria, such as carbapenem-resistant strains of K. pneumoniae Reference Zweigner, Gastmeier, Kola, Klefisch, Schweizer and Hummel55 and P. aeruginosa, Reference Sorin, Segal-Maurer, Mariano, Urban, Combest and Rahal85 which pose a substantial threat to public health.
In addition, in 14 (18.9%) of studies, the source of contamination was identified as being inadequate disinfection, highlighting the importance of proper bronchoscope reprocessing. This process necessitates the use of high-level disinfection techniques, defined as procedures that lead to the complete elimination of all microorganisms with the exception of a few bacterial spores. 94 Device-specific instructions should always be followed, and staff should be regularly trained on device-specific instructions and should have access to cleaning and disinfection protocol manuals. Reference Mehta and Minai95,Reference Culver, Gordon and Mehta96 The recommendations for bronchoscope reprocessing comprise of several steps to ensure that the instrument is properly cleaned and disinfected between uses. Reference Culver, Gordon and Mehta96 These include regular inspection and cleaning of the instrument after each procedure, cleaning external surfaces and thoroughly brushing all internal channels to reduce bioburden, using approved disinfectants and regularly testing their concentrations, ensuring compatibility between the bronchoscope and AER, rinsing and subsequently thoroughly drying the bronchoscope before storage, and maintaining a log of use and maintenance. Reference Mehta and Minai95–Reference Weber and Rutala97
This study had several limitations. The studies evaluated exhibited a high degree of heterogeneity in terms of their reported outcomes, making it impossible to conduct a random-effects–based meta-analysis. We did not record the mortality rates of the patients included in each outbreak because it was not part of the original study protocol. Most studies identified a contamination source, suggesting a potential publication bias, as studies that did not identify a source may have been less likely to be published. Finally, the reported correlations were based on binary variables regarding the presence or absence of the evaluated factors, which may not consider possible confounders.
In conclusion, this review provides valuable insights into the sources of contamination and pathogens associated with outbreaks and pseudo-outbreaks in the bronchoscopy suite. These findings highlight the importance of strict adherence to disinfection and maintenance guidelines specific to each device involved in these procedures and not limited to bronchoscopes alone. Furthermore, the identification of correlations between specific pathogens and sources of contamination can provide valuable guidance to clinicians, public health investigators, infection preventionists and hospital epidemiologists in identifying the source of an outbreak or pseudo-outbreak. Ongoing vigilance and attention to infection control practices both inside and outside the bronchoscopy suite are critical to ensure patient safety and prevent future outbreaks.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/ice.2023.250
Acknowledgments
All the relevant data obtained and analyzed for the purposes of this study are presented in the current manuscript.
Financial support
No financial support was provided relevant to this article.
Competing interests
All authors report no conflicts of interest relevant to this article.
