Peri-procedural code status discussions before bronchoscopy: A cross-sectional survey of German and Swiss pulmonologists

Abstract

Introduction

In-hospital cardiac arrest is associated with poor outcomes, yet peri-procedural code status discussions before bronchoscopy may be inconsistent, particularly in patients with elevated procedural risk.

Methods

We conducted an anonymous online survey of a convenience sample of hospital-based pulmonologists in Germany and Switzerland between April and July 2025. The survey assessed routine practices and attitudes regarding peri-procedural code status discussions, risk stratification, and management of patients with pre-existing No-CPR status. Owing to distribution through multiple channels and secondary forwarding, a precise response rate could not be calculated.

Results

A total of 148 pulmonologists participated (mean age 47.8 years; 94.6% board-certified). Discussions regarding CPR preferences were conducted almost always or most of the time before high-risk bronchoscopies by 67.6% of respondents, but were rarely or almost never performed before low-risk procedures by 72.0%. Respondents more frequently reported procedural risk than patient-related factors as a trigger for these discussions. Periprocedural No-CPR status was considered controversial by 44.9%, and responsibility for these discussions varied among respondents. Only 25% reported formal institutional training.

Conclusions

Substantial variability exists in peri-procedural code status practices before bronchoscopy. Clearer responsibilities and structured guidance are needed to better align care with patient preferences.

Introduction

In-hospital cardiac arrest (IHCA) is associated with poor outcomes, with survival-to-discharge rates of 15–34% and clinically significant neurological disability in up to 28% [1, 2]. Discussions regarding whether cardiopulmonary resuscitation (CPR) should be initiated in the event of cardiac arrest are commonly referred to as code status discussions (CSDs). However, such discussions are frequently insufficient, potentially resulting in decisions that do not align with patient values and goals of care [3, 4]. Both the general population and clinicians overestimate survival likelihood, potentially influencing code status decisions [3, 5].

Surveys indicate that physicians may disregard existing orders to withhold cardiopulmonary resuscitation (No-CPR status) during sedation or anaesthesia [6]. For example, 75% of surgeons and 54% of anaesthesiologists considered that pre-existing donot- resuscitate (DNR) orders should not apply during surgery. Notably, 38% of surgeons and 18% of anaesthesiologists reported suspending pre-existing DNR orders without prior discussion [6]. According to self-reported data, U.S. gastroenterologists reverse the code status before endoscopy in all cases (41%) or in most cases (18% )[7].

Whether these findings can be directly applied to bronchoscopy remains uncertain. Bronchoscopy is generally considered a low-risk procedure. In a large prospective cohort including more than 20,000 procedures, major complications occurred in approximately 1% of bronchoscopies, with a mortality rate of 0.02% [8].

However, risks are higher in critically ill patients. In a study of 169 non-intubated ICU patients with acute respiratory failure (PaO₂/FiO₂ <300mmHg, mean 194mmHg), 59 patients (35%) required increased ventilatory support within 24 hours after bronchoscopy, including 25 (15%) who were intubated during this period; 4% required intubation within the first 2 hours. Complications included cardiac arrest (n=4), arrhythmias (n=9) and pneumothorax (n=2) [9]. In another study of 148 ICU patients with hematologic or oncologic malignancies and respiratory failure, 49% deteriorated and 27% needed intubation following bronchoscopy. The overall hospital mortality was 55% [10].

Given these risks and the associated ethical concerns regarding patient autonomy and CPR preferences, we investigated pulmonologists’ current practices and attitudes toward CSDs in patients undergoing bronchoscopy.

Methods

Between April and July 2025, an anonymous online survey was distributed to hospital-based pulmonologists across Germany and Switzerland.

The questionnaire was developed by the study team, consisting of four internists, one intensive care physician, and one pulmonologist. Content validity was assessed by two additional pulmonologists and one anaesthesiologist. The survey was pilot tested within the pulmonology department of the University Hospital Basel, and minor modifications were made based on feedback. No formal cognitive testing was performed.

All questions were mandatory, respondents could review and modify their responses before submission, and no branching logic was used. Percentages were calculated based on the number of valid responses for each item; therefore, denominators may vary owing to item nonresponse. The survey was administered using the REDCap platform and was available in German and French (Supplementary Material 1).

This study was primarily descriptive. Categorical variables are presented as counts and percentages, and continuous variables as mean ± SD and median (IQR). For Figure 1 and Table 3, responses were dichotomized into yes (“almost always” or “most of the time”) and no (“rarely” or “almost never”). For patient-related factors, responses of “always necessary” or “often necessary” were considered affirmative. Percentages were calculated based on valid responses for each item. Exploratory subgroup analyses were based on available cases for each variable; respondents with missing data for the respective variable were excluded from that specific analysis. Continuous variables included in exploratory subgroup analyses were dichotomized at the median for analysis. Exploratory subgroup analyses were performed using chi-square tests. Statistical analyses were conducted using Stata (StataCorp, College Station, TX, USA).

Figure 1.

Frequency of discussions regarding CPR preferences, ICU admission, and treatment limitations before low- and high-risk bronchoscopy. † Rarely discussed: Rarely + Almost never. * Usually discussed: Almost always + Most of the time.

Table 3.

Bivariate analysis of factors associated with discussing cardiopulmonary resuscitation (CPR) prior to high-risk bronchoscopy, including sex, clinical experience, weekly bronchoscopy volume, estimated survival after cardiac arrest during bronchoscopy, and concern that CPR discussions may increase patient anxiety.

Variable	Category	Discuss CPR status n (%)	Does not discuss CPR status n (%)	P value
Sex	Male	45 (51.7)	42 (48.3)	0.092
	Female	18 (36.7)	31 (63.3)
Clinical experience	<19 years	33 (45.2)	34 (54.0)	0.308
	≥19 years	29 (46.0)	40 (54.8)
Weekly bronchoscopy volume	<6	34 (46.6)	32 (50.8)	0.624
	≥6	39 (53.4)	31 (49.2)
Estimated survival after cardiac arrest during bronchoscopy	<70%	31 (42.5)	20 (31.8)	0.198
	≥70%	42 (57.5)	43 (68.2)
Perceived patient anxiety related to CPR discussions	increases anxiety	15 (42.9)	20 (57.1)	0.572
	does not increase anxiety	36 (48.6)	38 (51.4)

Legend: P values are based on exploratory chi-square tests and should be interpreted descriptively; Analyses were based on available cases for each variable; denominators may therefore vary because of incomplete items. Continuous variables were dichotomized at their respective medians (clinical experience: 19 years; weekly bronchoscopy volume: 6 bronchoscopies/week; estimated survival after cardiac arrest during bronchoscopy: 70%).

The survey was distributed through the newsletters of the German and Swiss respiratory societies, which were sent to approximately 4,000 and 500 physicians, respectively. According to metrics provided by the German Respiratory Society, 42% of recipients opened the newsletter. Because the survey was disseminated through multiple channels, including secondary forwarding via department heads, the exact number of eligible physicians who received the invitation could not be determined; therefore, a formal response rate could not be calculated (Figure S1).

The Ethics Committee of Northern and Central Switzerland reviewed the study design and waived the need for formal ethical approval (Req-2024-01571). This study is reported in accordance with the CHERRIES recommendations [11] (Supplementary Material 2).

Within the survey, low-risk bronchoscopy was described as a procedure without an anticipated increased risk of severe hypoxemia, whereas high-risk bronchoscopy was described as a procedure associated with an increased risk of severe hypoxemia, intubation, or circulatory arrest. Examples of high-risk situations provided in the survey included pre-existing respiratory failure, severe infection, advanced COPD, and hypotension. All survey scenarios implicitly assumed that sufficient time was available to conduct a CPR status discussion before bronchoscopy if considered appropriate. In addition, no distinction was made between diagnostic and therapeutic procedures.

Results

A total of 148 pulmonologists were included in the analysis (mean age ±SD: 47.8 ±10.9 years; 35.8% female; 94.6% board-certified). Most were attending physicians (41.9%) or consultant physicians (48.0%), with an average of 20.1 ±10.0 years of clinical experience. Definitions of these positions are provided in the table footnotes. A notable finding was that 87.8% of respondents reported having witnessed or performed more than 10 unexpected CPRs during bronchoscopy (Table 2).

Table 2.

Characteristics of survey respondents.

Characteristic	Category	Respondents (n)	n (%)
Age, mean (± SD); median (IQR 25, 75)	Years	148	47.8 (10.9); 46.5 (39.0, 55.0)
Sex, n (%)	Male	148	95 (64.2)
	Female		53 (35.8)
Workplace, n (%)	Tertiary hospital (e.g., university hospital)	148	67 (45.3)
	Non-tertiary hospitals (e.g., regional hospital)		81 (54.7)
Clinical experience, mean (± SD), median (IQR 25, 75)	years	148	20.1 (10.0) 19.0 (12.0, 26.3)
Current position, n (%)	Resident	148	15 (10.1)
	Attending physician		62 (41.9)
	Consultant		71 (48.0)
Board-certified pulmonologist, n (%)		148	140 (94.6)
Bronchoscopies per week, mean (± SD), median (IQR 25, 75)		148	10.9 (11.2) 6.0 (4.0, 15.0)
Number of peri-procedural CPRs performed, n (%)	More than 10	148	130 (87.8)
	Fewer than 10		18 (12.2)
Total number of CPRs performed, n (%)	Fewer than 10	148	25 (16.9)
	10 – 50		66 (44.6)
	More than 50		57 (38.5)
Estimated survival rate after peri-procedural CPR with good neurological outcome in %, mean (± SD), median (IQR 25, 75)		111	61.1 (25.2) 70.0 (50.0, 80.5)

Values are presented as n (%), unless otherwise indicated in the first column. Continuous variables are reported as mean (SD) and median (IQR), as specified.

Discussions regarding CPR status were reported to be conducted almost always (43.9%) or most of the time (23.7%) before high-risk bronchoscopies. In contrast, such discussions were conducted rarely (21.6%) or almost never (50.4%) before low-risk bronchoscopies (Figure 1; Table S1).

Beyond CPR preferences, ICU admission preferences (89.4%) and therapy limitations (56.6%) were also frequently discussed before high-risk bronchoscopies (Figure 1; Table S1). Prior complications (85.6%) were frequently cited as a reason for initiating these discussions, although patient-related factors such as advanced age (71.4%) and cognitive impairment (71.8%) were also commonly reported (Table S2).

In exploratory bivariate analyses using the chisquared test, no significant differences in the frequency of CPR status discussions prior to high-risk bronchoscopy were observed according to sex, clinical experience, weekly bronchoscopy volume, estimated survival following cardiac arrest during bronchoscopy, or perceived patient anxiety related to CPR discussions (Table 3).

44.9% of respondents considered the presence of a No-CPR status during bronchoscopy controversial. Among respondents, 67.7% reported that the risk of hypoxia justified full resuscitative measures, including CPR, despite a No-CPR status.

In addition, 45.9% supported its routine temporary suspension. Nevertheless, 66.9% reported re-discussing resuscitation preferences with patients who had an existing No-CPR status. There was no clear agreement on the responsibility for these discussions: 53.1% assigned it to the referring team, and 46.8% to the pulmonologist. Although 88.6% acknowledged the growing importance of peri-procedural CSDs, only 25% reported the presence of formal institutional training (Table 1).

Table 1.

Main findings on peri-procedural CPR discussions and No-CPR status before bronchoscopy.

Factor / Variable	Category	Respondents (n)
Primary responsibility for inpatient code status discussions, n (%)	Ward / ICU physician	126	67 (53.1%)
	Pulmonologist		59 (46.8%)
Formal training on code status discussion for residents, yes, n (%)		112	28 (25.0%)
Re-discussion of pre-existing No-CPR status prior to bronchoscopy, yes, n (%)		109	73 (66.9%)
No-CPR status during bronchoscopy considered controversial due to hypoxia risk, yes, n (%)		98	44 (44.9%)
Hypoxia risk requires full intervention, including CPR, despite No-CPR status, yes, n (%)		96	65 (67.7%)
Temporary suspension of No-CPR status considered justified due to procedural risk, yes, n (%)		96	44 (45.9%)
Code status discussion should be routine prior to bronchoscopy, yes, n (%)		96	69 (71.9%)
Discussion of ICU measures and intubation before bronchoscopy should be routine, yes, n (%)		96	79 (82.3%)

Abbreviations: IQR, interquartile range; CPR, cardiopulmonary resuscitation; ICU, intensive care unit; advance directives (AD), documents that record a patient’s preferences for medical care if they become unable to decide; DNR, do not resuscitate; ROSC, return of spontaneous circulation. Percentages were calculated using the number of valid responses for each item. Denominators vary because some participants did not complete all sections of the survey. Attending physician: equivalent to a board-certified physician with independent clinical responsibility and supervision of resident physicians. For complex procedures or clinical decisions, support from a consultant may be required. (approximately equivalent to a Consultant in the UK). Consultant: equivalent to a senior physician with departmental leadership responsibilities, including supervision of attending physicians and strategic oversight of clinical services (approximately equivalent to a Senior Consultant in the UK and a Division Chief or Department Chair in the USA).

Discussion

These findings provide exploratory insights into current practices among survey respondents. According to this survey, discussions regarding CPR preferences are infrequent prior to low-risk bronchoscopies. Over 70% reported rarely or almost never initiating them. Responses were more variable for high-risk procedures, with 23.7% and 8.6% reporting that such discussions were conducted rarely or almost never, respectively. Anticipated procedural risk, rather than patient characteristics (e.g., frailty, age, comorbidities) emerged as the primary trigger for initiating these discussions.

Roughly half of the respondents considered the referring ward or ICU physician, rather than the pulmonologist performing the procedure, to be responsible for discussing CPR status.

In contrast, a survey among anaesthesiologists and surgeons demonstrated a remarkably high degree of agreement regarding responsibility for perioperative code status discussions. Anaesthesiologists considered themselves responsible in 95.8% of cases [12], while surgeons identified anaesthesiologists as responsible in 95.6% of cases [13], either independently or jointly with surgeons. Such clarity regarding responsibility may help ensure that these discussions are conducted more consistently.

Two-thirds of pulmonologists reported rediscussing code status with patients who had an existing No-CPR status. This proportion is lower than that observed in a survey of anesthesiologists, in which 76.3% of respondents reported re-evaluating CPR preferences before surgery in patients with a pre-existing No-CPR status [12]. This practice is particularly relevant for patients with advanced illness, as some may wish to avoid resuscitation due to concerns about neurological outcomes, yet might prefer temporary full intervention, including CPR, in the event of complications occurring in a setting such as bronchoscopy, where immediate resuscitation efforts are possible. Integrating mandatory documentation of ward and procedure-specific CPR status into the bronchoscopy referral workflow may facilitate interdisciplinary discussion and should be evaluated in future studies.

Over two-thirds of respondents reported that CPR preference discussions might cause anxiety in patients. This observation may be clinically relevant, as a survey of anaesthesiologists found that physicians who were concerned about causing patient anxiety were about half as likely to conduct CPR discussions (adjusted OR 0.53) [12]. Communication training may help clinicians address CPR preferences more confidently without increasing patient anxiety [14].

Another notable finding was the wide variability in physicians’ estimated survival with good neurological outcome following peri-procedural CPR. On average, respondents estimated this probability at 61.1%, with a standard deviation of 25.2%. Similar variability has been reported among anaesthesiologists and surgeons, who estimated survival rates of 40.0% and 44.7%, respectively. This variability may reflect differences in personal experience with CPR, procedural complications, and clinical case mix.

This variation reflects differing perceptions of prognosis, which may influence how physicians communicate risks, frame recommendations, or interpret existing DNR orders. Although specific survival data for cardiac arrest during bronchoscopy are lacking, respiratory failure is a frequent precipitant of monitored IHCA. In a large observational study, 49% of patients who experienced a monitored and witnessed IHCA had acute pre-existing respiratory failure, 39% had an invasive airway in place at the time of event, and 32% were receiving supplemental oxygen. Acute respiratory insufficiency was the third most common immediate clinical feature in monitored and witnessed IHCA (40%), after arrhythmia (67%) and shock (50%). Even when cardiac arrests were witnessed and monitored, survival with good neurological outcomes remained low (14% vs. 5% for unwitnessed and unmonitored arrests) [15].

How these conversations are framed matters: if presented solely as a warning of possible death, they may indeed feel distressing. But when incorporated into a broader process of shared decision-making, they can help align care with patient values—particularly in populations with limited life expectancy.

Our findings suggest considerable variability in the availability of institutional policies and formal training regarding code status discussions in the setting of bronchoscopy. Only 25% of respondents reported having received such training. Similarly, a survey of gastroenterologists demonstrated limited awareness of institutional policies, while 89% of respondents reported that procedure-specific guidelines would be helpful [7].

This study has limitations. The survey represents a convenience sample and was based on voluntary participation, which may introduce selection bias. Also, as the study was conducted exclusively in Germany and Switzerland, the findings may not be generalizable to other healthcare systems with different clinical practices or institutional frameworks. In addition, respondents were not asked to specify their country of practice, which precluded country-specific analyses and limits interpretation of potential differences between Germany and Switzerland. Despite these limitations, it provides exploratory insights and may help inform future research and training initiatives. These findings may inform future quality-improvement initiatives. Structured code status discussions before bronchoscopy, supported by institutional policies or checklists, could help ensure that CPR preferences and responsibilities for these conversations are addressed consistently, particularly in patients undergoing high-risk procedures.

Conclusions

In conclusion, these exploratory findings suggest substantial variability in how responding pulmonologists approach discussions about CPR preferences before bronchoscopy. Procedural risk appeared to be a more common trigger for initiating these conversations than patient-specific factors. These findings contribute to a better understanding of the heterogeneity of approaches among responding pulmonologists. Future studies should evaluate whether structured discussions, institutional guidance, or checklist-based approaches improve consistency in clinical practice and alignment with patient preferences.

Appendix. Supplementary files

Table S1.

Detailed survey results regarding CPR preferences, ICU admission, and treatment limitations before low- and high-risk bronchoscopy.

Code status discussion prior to low-risk bronchoscopy, n (%)	Almost always	139	24 (17.3%)
	Most of the time		15 (10.8%)
	Rarely		30 (21.6%)
	Almost never		70 (50.4%)
Code status discussion prior to high-risk bronchoscopy, n (%)	Almost always	139	61 (43.9%)
	Most of the time		33 (23.7%)
	Rarely		33 (23.7%)
	Almost never		12 (8.6%)
ICU admission preferences discussed prior to low-risk bronchoscopy, n (%)	Almost always	132	38 (28.8%)
	Most of the time		28 (21.2%)
	Rarely		37 (28.0%)
	Almost never		29 (22.0%)
ICU admission preferences discussed prior to high-risk bronchoscopy, n (%)	Almost always	132	92 (69.7%)
	Most of the time		26 (19.7%)
	Rarely		11 (8.3%)
	Almost never		3 (2.3%)
Therapy limitations discussed prior to low-risk bronchoscopy, n (%)	Almost always	129	10 (7.8%)
	Most of the time		9 (7.0%)
	Rarely		29 (22.5%)
	Almost never		81 (62.8%)
Therapy limitations discussed prior to high-risk bronchoscopy, n (%)	Almost always	129	43 (33.3%)
	Most of the time		30 (23.3%)
	Rarely		37 (28.7%)
	Almost never		19 (14.7%)

Table S2.

Attitudes, decision-making, and factors influencing peri-procedural CPR discussions before bronchoscopy

Routine assessment of advance directives (AD) prior to bronchoscopy in outpatients, yes, n (%)		112	32 (28.6%)
Routine assessment of AD prior to bronchoscopy in inpatients, yes, n (%)		112	91(81.3%)
Code status discussions can frighten patients, yes, n (%)		112	77 (68.7%)
Peri-procedural code status discussions considered increasingly relevant in aging population, yes, n (%)		96	85 (88.6%)
Preferred decision-making approach, peri-procedural setting, n, (%)	Physicians’ decision	125	15 (12.0%)
	Shared decision		93 (74.4%)
	Patients’ decision		17 (13.6%)
AD considered relevant for bronchoscopy, yes, n (%)		112	76 (67.9%)
Proportion of patients with No-CPR status prior to bronchoscopy <25%, n (%)		109	96 (88.1%)
Patient-related factors for initiating code status discussions
Advanced age (e.g.,>80 years), yes. n (%)		112	80 (71.4%)
Full care dependency, yes, n (%)		110	86 (78.2%)
Cognitive impairment / dementia, yes, n (%)		110	79 (71.8%)
Severe cardiopulmonary disease (e.g. CAD, advanced COPD), yes, n (%)		107	84 (78.5%)
Prior bronchoscopy complications (e.g., bronchospasm), yes, n (%)		111	95 (85.6%)
High bronchoscopy risk, yes, n (%)		110	99 (90.0%)
Monitored setting considered justification for overriding DNR order, yes, n (%)		97	19 (19.6%)
No-CPR status must be respected, as decisions reflect patient values, not medical judgment alone, yes, n (%)		97	64 (65.9%)
ROSC after bronchoscopy warrants more aggressive care, yes, n (%)		94	50 (53.1%)

Abbreviations: IQR, interquartile range; CPR, cardiopulmonary resuscitation; ICU, intensive care unit; advance directives (AD), documents that record a patient’s preferences for medical care if they become unable to decide; DNR, do not resuscitate; ROSC, return of spontaneous circulation. Percentages were calculated using the number of valid responses for each item. Denominators vary because some participants did not complete all sections of the survey. Attending physician: equivalent to a board-certified physician with independent clinical responsibility and supervision of resident physicians. For complex procedures or clinical decisions, support from a consultant may be required. (approximately equivalent to a Consultant in the UK). Consultant: equivalent to a senior physician with departmental leadership responsibilities, including supervision of attending physicians and strategic oversight of clinical services (approximately equivalent to a Senior Consultant in the UK and a Division Chief or Department Chair in the USA).

Figure S1

Flow diagram.

Numbers are approximate and based on available distribution metrics. Department heads at the following pulmonary centers were contacted and asked to distribute the survey within their institutions: Germany: 1) University Hospital Heidelberg; 2) Asklepios Munich; 3) RBK Lungenzentrum Stuttgart; 4) Johannesstift Diakonie (Berlin); 5) Heilos Gesundheit (Berlin); 6) Lungen Clinic Grosshansdorf (Hamburg); 7) Universitätsspital Giessen; 8) Ruhrlandklinik (Essen); 9) Rhön Klinikum (Giessen/Marburg); 10) Klinikum Chemnitz; 11) Klinikum Nürnberg; 12) Uniklinik Freiburg; Switzerland: 1) University Hospital Zürich; 2) Kantonsspital Aarau; 3) University Hospital Geneva; 4) University Hospital Lausanne; 5) Health Ostschweiz; 6) University Hospital Inselspital Bern; 7) Luzerner Kantonsspital; 8) Kantonsspital Winterthur.

Supplementary Material 1.

Questionnaire. Peri-procedural code status discussions before bronchoscopy: A cross-sectional survey of German and Swiss pulmonologists.

1. Introduction
Dear Participant,
This survey addresses peri-procedural CPR status discussions in patients undergoing bronchoscopy and aims to provide insight into current practices and attitudes regarding CPR status discussions, intensive care preferences, treatment limitations, and the management of patients with pre-existing No-CPR orders.
Please note the following information: - This questionnaire is intended for pulmonologists who perform bronchoscopies. - The questions relate to discussions held before bronchoscopy regarding CPR preferences, intensive care treatment, treatment limitations, and the management of severe procedural complications. - The survey takes approximately 8 minutes to complete. - Please complete the survey only once.
Thank you for your valuable participation. Your contribution will help improve our understanding of current approaches to peri-procedural CPR status discussions and decision-making in bronchoscopy.
Sincerely,
The Study Team

2. Personal Details (10 items)
2.1	What is your age (in years)?	_________________________
2.2	Your gender	Female Male Prefer not to say
2.3	Where do you work?	Tertiary Care Hospital (e.g. university hospital)/hospital with maximum care Secondary Care Hospital Regional Hospital (smaller Secondary Care Hospital) Outpatient pulmonology practice (private practice) Other
2.4	How many years of professional medical experience do you have?	_________________________
2.5	What is your current position?	Resident Senior physician / Attending physician Consultant Chief physician
2.6	Do you hold a board-certified medical specialty?	Yes No
2.7	Which specialties are you board-certified in?	Pulmonology Internal Medicine Anesthesiology Intensive Care Medicine other
2.8	How many bronchoscopies do you perform per week?	_________________________
2.9	How many unexpected cardiopulmonary resuscitations (CPRs) during bronchoscopy have you personally performed or witnessed? Unexpected refers to patients who were hemodynamically stable before bronchoscopy but unexpectedly required CPR during the procedure, for example because of sedation-related complications or bleeding.	< 10 10–50 > 50
2.10	How many resuscitations (in any setting) have you performed or witnessed?	< 10 10–50 > 50

3. Discussion of CPR status before bronchoscopy (24 items)
How often do you discuss CPR status in the following situations?
3.1	For a low-risk bronchoscopy? Low-risk bronchoscopy = A bronchoscopy in which severe hypoxemia or circulatory arrest is not anticipated based on the planned procedure and the patient’s clinical condition.	Almost always Often Rarely Almost never
3.2	For a high-risk bronchoscopy? High-risk bronchoscopy = A bronchoscopy associated with an increased risk of severe hypoxemia, intubation, or circulatory arrest based on the patient’s clinical condition (e.g., pre-existing respiratory failure, severe infection, advanced COPD, or hypotension)	Almost always Often Rarely Almost never
How often do you discuss severe complications that may require intensive care support in the following situations? Severe complications = e.g., severe peri-procedural hypoxemia or major bleeding requiring intubation.
3.3	For a low-risk bronchoscopy? Low-risk bronchoscopy = A bronchoscopy in which severe hypoxemia or circulatory arrest is not anticipated based on the planned procedure and the patient’s clinical condition.	Almost always Often Rarely Almost never
3.4	For a high-risk bronchoscopy? High-risk bronchoscopy = A bronchoscopy associated with an increased risk of severe hypoxemia, intubation, or circulatory arrest based on the patient’s clinical condition (e.g., pre-existing respiratory failure, severe infection, advanced COPD, or hypotension)	Almost always Often Rarely Almost never
Discussion of treatment limitations How often do you discuss post-procedural treatment limitations in the event of severe complications in the following situations? Post-procedural treatment limitations = e.g., no prolonged intubation and mechanical ventilation, or no initiation of specific organ support therapies such as hemodialysis.
3.5	For a low-risk bronchoscopy? Low-risk bronchoscopy = A bronchoscopy in which severe hypoxemia or circulatory arrest is not anticipated based on the planned procedure and the patient’s clinical condition.	Almost always Often Rarely Almost never
3.6	For a high-risk bronchoscopy? High-risk bronchoscopy = A bronchoscopy associated with an increased risk of severe hypoxemia, intubation, or circulatory arrest based on the patient’s clinical condition (e.g., pre-existing respiratory failure, severe infection, advanced COPD, or hypotension)	Almost always Often Rarely Almost never

Discussion of the CPR status
3.7	When you talk to patients about the possibility of cardiac arrest during bronchoscopy: How should the CPR status be determined?	I determine the CPR status, considering the patient’s medical condition and preferences (= physician decision) I discuss the CPR status with the patient and make a shared decision-(=shared decision making) I inform the patient about CPR so that he/she can make his/her own decision (= patient decision)
3.8	Which member of the medical team is responsible for determining the patient’s CPR status before bronchoscopy?	Ward physician / intensive care physician Physician who documented the CPR status in the electronic medical record Physician performing the bronchoscopy Other

In which of the following situations do you consider it necessary to discuss CPR status and potential intensive care measures before bronchoscopy?
		Not necessary	Often not necessary	Rarely necessary	Sometimes necessary	Often necessary	Always necessary
3.9	Advanced age (e.g. >80 years)
3.9	Young age (e.g. < 50 years)
In which of the following situations do you consider it necessary to discuss CPR status and potential intensive care measures before bronchoscopy?
3.9	Pre-existing complete dependency (e.g. nursing home)
3.9	Cognitive impairment or dementia
3.9	Documented Advance Directives
3.9	Low anticipated risk of severe peri-procedural complications during a planned diagnostic bronchoscopy
3.9	High iatrogenic risk of circulatory arrest (e.g., severe pre- existing tracheal stenosis)
3.9	Significant comorbidities (e.g. severe heart or lung disease)
3.9	History of complications during previous bronchoscopies
3.9	Advanced cancer or malignant hematologic disease
3.10		What is your estimate of the average survival rate with a good neurological outcome at hospital discharge after cardiac arrest and resuscitation during bronchoscopy? Good neurological outcome, CPC 1–2 = The patient can live independently. Certain neurological impairments (e.g. hemiplegia) may be present.			Your estimate: 0%–100%: ___________%
3.11		Do you routinely ask outpatients whether they have an Advance Directive?			Yes Rather yes Rather no No
3.12		Do you routinely ask inpatients whether they have an Advance Directive?			Yes Rather yes Rather no No
3.13		In your opinion, is the CPR status documented in an Advance Directive also relevant during bronchoscopy?			Yes Rather yes Rather no No
3.14		Do you think that discussing CPR status before bronchoscopy causes patients to feel anxious?			Yes Rather yes Rather no No
3.15		Does your institution provide training on how to discuss CPR status with patients?			Yes No

4. Pre-existing No-CPR order. (9 items) The following questions address how to manage patients with documented No-CPR status prior to bronchoscopy (e.g. based on an Advance Directive or prior hospitalization).
4.1	Approximately what proportion have a pre-existing No-CPR status?	Less than 25% 25–50% 50–75% More than 75%
4.2	Do you address or re-discuss the perioperative CPR preferences with patients who have a pre-existing No-CPR status? (e.g., from a previous hospitalization)?	Yes Rather yes Rather no No

To what extent do you agree with the following statements regarding management of patients with a No-CPR status?
		Absolutely disagree	Disagree	Rather disagree	Tend to agree	Agree	Absolutely agree
4.3	A No-CPR status during bronchoscopy is controversial because bronchoscopy inherently carries a risk of hypoxemia and, consequently, the potential need for CPR.
To what extent do you agree with the following statements regarding management of patients with a No-CPR status?
4.3	Because bronchoscopy involves direct intervention in the respiratory system, the procedural team must be able to provide all necessary interventions regardless of a patient’s pre-existing No-CPR status.
4.3	A No-CPR order must be respected during bronchoscopy, as resuscitation preferences should reflect the patient’s values and not only the clinical judgment of the medical team.
4.3	A No-CPR status should not apply during bronchoscopy because the likelihood of successful resuscitation is higher in a monitored setting.
4.3	A No-CPR order should be temporarily suspended during bronchoscopy because it may be difficult to distinguish between cardiorespiratory arrest resulting from potentially reversible procedure- or anesthesia-related complications (e.g., airway obstruction, pneumothorax, or medication-related adverse events) and arrest due to irreversible deterioration from underlying conditions (e.g., massive tumor hemorrhage, refractory hypoxemia due to severe pneumonia).
4.3	If a patient wishes to maintain a No-CPR status during bronchoscopy, or if CPR preferences cannot be discussed (e.g., because of dementia), bronchoscopy should be withheld.
4.4	In your opinion, how should pulmonologists approach a documented No-CPR status? You may answer in keywords or brief statements:				__________________________

5. Personal attitudes towards peri-procedural CPR status. To what extent do you agree with the following statements? (6 items)
		Absolutely disagree	Disagree	Rather disagree	Tend to agree	Agree	Absolutely agree
5.1	If CPR is performed during bronchoscopy and return of spontaneous circulation (ROSC) is achieved, post- procedural intensive care should be continued more aggressively because the complication occurred in a procedural setting.
5.1	Discussion and determination of peri-procedural CPR status before bronchoscopy should be standard practice.
5.1	Discussion of intensive care treatment and intubation in the event of a severe complication should be standard practice before bronchoscopy.
5.1	Given the aging population, peri-procedural CPR status should be addressed more frequently.
5.1	Discussions regarding CPR status should primarily be conducted by the referring physician.
5.2	In your opinion, how should the periprocedural CPR status be determined and discussed? You may answer in keywords or brief statements:				__________________________

Supplementary Material 2.

CHERRIES-Checklist: Peri-procedural code status discussions before bron-choscopy: A cross-sectional survey of German and Swiss pulmonologists.

Reference: Eysenbach G. (2004) Improving the quality of Web surveys: the Checklist for Reporting Results of Internet E-Surveys (CHERRIES) J Med Internet Res 6(3):e34doi:10.2196/jmir.6.3.e34.

Item Category	Checklist Item	Explanation
Design	Describe survey design	Cross-sectional anonymous online survey of pulmonologists in Germany and Switzerland.
IRB	IRB approval	Yes
	Informed consent	Participants were informed about the purpose of the study, voluntary participation, and data handling on the survey landing page. Proceeding to the questionnaire was considered implied consent.
	Data protection	Guaranteed by the responsible researchers’ personal login to the REDCap account with username and password
Development and pretesting	Development and pretesting	Questionnaire developed by the study team (internists, ICU physician, pulmonologist), reviewed by pulmonologists and an anesthesiologist, and pilot-tested at the University Hospital Basel with minor revisions based on feedback.
Recruitment process and description of the sample having access to the questionnaire	Open survey versus closed survey	Open survey
	Contact mode	Newsletter -emailing and targeted e-mailing to Heads of Departments
	Advertising the survey	No advertisement
Survey administration	Web/E-mail	REDCap
	Context	The survey was accessible through a web link provided in the invitation e-mails.
	Mandatory / voluntary	Voluntary
	Incentives	No incentives were offered.
	Time/Date	April 2025 – July 2025.
	Randomization of items or questionnaires	No randomization of survey items was used.
	Number of items	49
	Number of screens (pages)	1
	Completeness check	Yes
	Review step	Participants were able to review and modify their responses before final submission.
Response rates	Unique site visitor	Not available.
	Completion rate	A total of 154 accessed the survey, 6 were excluded before baseline characteristics leaving 148 for analysis (96 complete, 52 incomplete).
Preventing multiple entries from the same individual	Cookies used	No
	IP check	No
	Log file analysis	No
	Registration	No registration required (open survey).
Analysis	Handling of incomplete questionnaires	Partially completed questionnaires were retained. Incomplete items were not imputed, and percentages were calculated based on available responses for each question.
	Questionnaires submitted with an atypical timestamp	No exclusion based on completion time was performed.
	Statistical correction	No statistical correction was applied.

Availability of data and material

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.