Browse the corpus

Principles for Guideline Application and Implementation These recommendations are intended to support, but not replace, individual clinical judgment, which must always remain the foundation of clinical decision-making. In all cases, the potential benefits of the bronchoscopic procedure should be carefully weighed against the associated risks. Patient preferences should also be considered as part of the decision-making process.

A range of diagnostic and therapeutic procedures can be performed during bronchoscopy. This guideline focuses on the most commonly used diagnostic interventions in routine clinical practice: Collection of bronchial washings for microbiological, cytological, and other tests; Transbronchial needle aspiration (TBNA), most commonly performed under endobronchial ultrasound guidance (EBUS), i.e., EBUS-TBNA; Transbronchial lung biopsy using forceps (TBLB) and cryoprobe (TBLC). These guidelines apply to all patients undergoing the above bronchoscopic procedures. This document does not apply to patients undergoing rigid bronchoscopy nor therapeutic bronchoscopic procedures.

This guideline is intended for pulmonologists, thoracic surgeons, and physicians in specialty training in these fields. It is also relevant for physicians from other specialties—such as anesthesiology, internal medicine, intensive care, and oncology—who perform bronchoscopic procedures. Additionally, the recommendations apply to nurses working in pulmonary wards and endoscopy units involved in the care of patients undergoing bronchoscopy.

The Working Group that developed this guideline included pulmonary specialists with extensive experience in bronchoscopy, as well as a thoracic surgeon, cardiologist, hematologist, anesthesiologist, critical care specialist, procedural nurse, and methodologists.

The guideline was developed in accordance with the principles of evidence-based medicine (EBM), incorporating some elements of the GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodology []. Five key clinical questions were formulated: In patients with abnormal platelet count, is it safe to perform bronchoscopy, BAL, EBB, EBUS-TBNA, TBLB, and TBLC? What is the minimum platelet count that allows these procedures to be performed safely? In patients with abnormal coagulation parameters—specifically, prolonged prothrombin time (PT), expressed as international normalized ratio (INR), and activated partial thromboplastin time (APTT)—is it safe to perform bronchoscopy, BAL, EBB, EBUS-TBNA, TBLB, and TBLC? What are the maximum acceptable INR and APTT values that allow these procedures to be performed safely? In patients taking 1 or 2 antiplatelet agents, is it safe to perform bronchoscopy, BAL, EBB, EBUS-TBNA, TBLB, and TBLC? In patients receiving oral anticoagulants, is it safe to perform bronchoscopy, BAL, EBB, EBUS-TBNA, TBLB, and TBLC? If not, when should the anticoagulant be discontinued prior to the procedure, and when should it be resumed afterward?

In patients receiving low-molecular-weight heparin (LMWH) at prophylactic or therapeutic doses, is it safe to perform bronchoscopy, BAL, EBB, EBUS-TBNA, TBLB, and TBLC? If not, when should low-molecular-weight heparin (LMWH) be discontinued before the procedure, and when should it be resumed afterward? Based on these key questions, a list of detailed clinical questions was developed and is presented in . These questions served as the basis for the literature search. For Question 4, the original formulation and literature search focused only on direct oral anticoagulants (DOACs). Following panel discussion, the clinical question was expanded to include vitamin K antagonists (VKAs), and the literature search was updated accordingly. Given the scope of the guideline, the only primary outcome assessed was bleeding during bronchoscopy. For the purpose of this guideline, a bleeding severity scale was developed based on the most commonly used classifications in the literature () [,,,]. The table also includes weights assigned to each bleeding severity grade as outcome measures.

Based on the detailed clinical questions, a systematic literature search of PubMed and EMBASE databases was conducted in March 2025 without restrictions on the start date. The search strategy combined free-text terms with controlled vocabulary from the MeSH and EMTREE thesauri. Studies relevant to the clinical questions were included regardless of methodological quality. Additional publications, including guidelines and secondary studies (systematic reviews, meta-analyses, or review articles), were also included. The detailed search strategy is described in . After duplicate removal, the remaining records were independently screened by 2 authors (FM and JZ). Any discrepancies were resolved through discussion. The database search was supplemented by manual screening of the reference lists of included publications, including identified guidelines, as well as relevant scientific society websites. The overall quality of the available evidence was assessed according to the GRADE classification. Primary and secondary studies were independently graded by 2 authors (FM and JZ) according to study or document type; however, no formal quality assessment process was applied to these publications. An exception was made for the guidelines, which were appraised using the AGREE II Global Rating Scale. In the absence of direct evidence, indirect data were considered when deemed relevant. Existing English-language guidelines on bronchoscopy were also reviewed [,,,,,,,,]; two documents available only in Chinese were excluded [,]. Additionally, given the scope of the guideline, documents addressing the use of anticoagulants and antiplatelet agents, as well as perioperative management, were also considered. These included the most recent guidelines from the American College of Chest Physicians (ACCP) on anticoagulant therapy and from the British Society for Haematology on bleeding risk assessment prior to invasive procedures [,]. A synthesis of the scientific evidence related to each clinical question, including an assessment of evidence quality, was prepared and presented to the panel members. Based on the available evidence, the panel members formulated individual recommendations during joint meetings. Recommendations were prepared based on evidence synthesis and further refined through discussion until full consensus was reached among all panel members. No formal consensus method was used.

In accordance with the GRADE methodology, recommendations were classified as either strong or weak (conditional) (), and the quality of the supporting evidence was rated as high, moderate, low, or very low (). In the absence of high-quality direct evidence, good practice points were formulated by extrapolating from non–bronchoscopy-related guidelines, narrative reviews, and the clinical experience of the panel members.

This document does not evaluate the risk–benefit balance for each clinical question individually. Instead, it focuses on assessing bleeding risk and the feasibility of safely performing the procedure. This approach reflects the fact that the potential benefit of bronchoscopy depends on the clinical indication (e.g., diagnostic value from obtaining a histopathological sample). Furthermore, in certain clinical scenarios, decision-making must also consider the potential cardiovascular risks associated with interrupting or modifying antiplatelet or anticoagulant therapy. Therefore, the decision to proceed with bronchoscopy should always be individualized, made in collaboration with a well-informed patient, and, when appropriate, based on a consultation with other specialists (e.g., a cardiologist or hematologist).

Bleeding during bronchoscopy occurs in 0.2–5% of cases; however, the risk of severe bleeding is much lower, estimated at around 0.031% [,,,]. Several factors influence bleeding risk. Key determinants include: Patient-related factors (e.g., coagulation status, use of antiplatelet or anticoagulant agents, and other bleeding risk factors); Location and type of structural lesions; Scope and characteristics of the procedure () [].

Primary data: Several observational studies were identified, most of which included relatively small patient cohorts. It should be noted that significant confounding factors were present in many of these studies. In particular, a substantial proportion of patients received platelet transfusions, limiting the ability to draw reliable conclusions regarding platelet count thresholds associated with bleeding risk. One retrospective analysis described 2053 bronchoscopic procedures without biopsy, performed in 1711 patients with cancer. Among these patients, 990 had a platelet count > 100,000/μL (control group), 301 had 50–100,000/μL, 413 had 20–50,000/μL, 218 had 10–20,000/μL, and 85 had <10,000/μL []. The overall bleeding risk was 1.1%, while the risk of severe bleeding was 0.2%. Bleeding risk was not associated with platelet count. Severe bleeding included 2 cases of epistaxis, one in a patient with a central airway tumor and one in a patient with breast cancer and radiation-induced pneumonitis. Both non-nasal bleeding events occurred in patients with normal platelet count. The value of this study is significantly limited by the high proportion of patients who received platelet transfusions (90.6% of those with a platelet count < 10,000/μL and 55.5% of those with 10–20,000/μL), which was not adequately accounted for in the analysis. Additionally, 7.7% of patients were taking aspirin, and 10.8% were receiving anticoagulants. Another factor complicating the interpretation of the results is that in 92.4% of cases, bronchoscopy was performed via the nasal route.

In a second large retrospective study, bronchoscopy with BAL was performed in 507 patients with hematologic diseases and a platelet count < 50,000/μL, including 281 patients with very severe thrombocytopenia < 20,000/μL []. Platelet transfusions were administered in 81.5% of patients with platelet count < 20,000/μL and in 22.5% of those with platelet count > 20,000/μL. The risk of severe bleeding and hemoglobin decline was similar between the groups with a platelet count of 20–50,000/μL and <20,000/μL. The risk of non-severe bleeding was slightly higher in the <20,000/μL group (4.6% vs. 1.8%), but this difference was not statistically significant ( p = 0.076). Clinically significant bleeding occurred in only 2 patients, both with platelet count < 20,000/μL. Multivariate analysis did not identify platelet count as a factor associated with bleeding risk.

In another single-center retrospective study involving 150 patients with platelet count < 100,000/μL, bronchoscopy (with BAL in some cases) was performed safely, with only 1 case of mild bleeding reported in a patient with a platelet count of 61,000/μL []. Prior to the procedure, 72 patients had platelet count < 50,000/μL and 15 had <20,000/μL. Platelet transfusions were administered to 65% of patients with platelet count < 50,000/μL and 8% of those with >50,000/μL. Several small studies focusing on specific patient populations were also identified. Kim et al. [] performed bronchoscopy with BAL ( n = 33) and TBLB ( n = 5) in mechanically ventilated patients with platelet count < 50,000/μL, most of whom had hematologic malignancies. One episode of severe bleeding occurred, resulting in death within 24 h. In a prospective study of patients within 6 months following bone marrow transplantation, 66 bronchoscopies were performed in 47 patients with platelet count < 50,000/μL (including 13 with platelet count < 20,000/μL) []. There were 3 episodes of minor and self-limiting bleeding and 1 episode of severe epistaxis.

Insufficient data were available to draw conclusions about the safety of biopsy procedures. No studies were identified regarding EBB or TBLC. TBLB was reported in 5 patients in the study by Kim et al. [] and in another study involving 22 patients with hematologic disorders []. In the latter study, biopsies were performed using a radial EBUS probe with a guide sheath. The mean platelet count was 12,000/μL (range: 7500–23,600/μL). No severe bleeding events were observed; however, all patients received platelet transfusions before or during the procedure. A conference abstract reported robot-assisted EBUS-TBNA in 35 patients with platelet count < 150,000/μL []. Two severe bleeding events were described in patients with platelet count < 50,000/μL, and an increased bleeding risk was noted in this subgroup compared to patients with platelet count > 50,000/μL.

The 2013 British Thoracic Society (BTS) guidelines state that bronchoscopy with BAL can be performed in patients with platelet count > 20,000 per μL, and that the local hematology team should be consulted regarding the need for platelet transfusion if EBB or TBLB is planned []. This is a grade D recommendation, the lowest level in the 4-tier system (A to D). The guidelines also recommended assessing platelet count when clinical risk factors suggest a likelihood of abnormal coagulation (also grade D). In addition, the evidence statement notes that “bleeding complications in patients with thrombocytopenia undergoing bronchoscopy and lavage are approximately 7%. No data are available regarding the safety of TBLB or EBB in thrombocytopenia but the majority of bleeding complications relate to epistaxis.” [] (p. i10). The 2019 guidelines from the Indian Chest Society, National College of Chest Physicians, and Indian Association for Bronchology recommend a minimum platelet count of at least 20,000/μL for BAL and 50,000/μL for EBB or TBLB []. These are conditional recommendations based on low-quality evidence. The same document also states that “BAL can be performed in patients with platelet count <20,000 per mm 3 if clinically indicated, after careful risk–benefit analysis. In patients with thrombocytopenia, oral route is preferred for performing bronchoscopy (UPP) [translator’s note: good practice point]” [] (p. S47). According to the 2024 guidelines from the British Society of Haematology, platelet count alone should not be used as a predictor of bleeding risk. The decision to perform prophylactic platelet transfusion should take into account the cause of thrombocytopenia, platelet function, and the procedure-related risk, including both the patient’s condition and the nature of the intervention [].

We suggest that the minimum platelet count required to safely perform bronchoscopy or bronchoscopy with BAL (without biopsy) should be ≥20,000/μL [conditional recommendation/low quality of evidence]. For patients undergoing biopsy procedures (e.g., EBB, EBUS-TBNA, TBLB, TBLC), the platelet count should be ≥50,000/μL [conditional recommendation/very low quality of evidence].

These recommendations apply to cases of isolated thrombocytopenia, i.e., thrombocytopenia without coexisting hemostatic abnormalities that may increase bleeding risk. A platelet count ≥ 50,000/μL is generally considered sufficient for the safe performance of endoscopic procedures. A platelet count < 50,000/μL requires careful evaluation of the risk–benefit balance of bronchoscopy and the associated bleeding risk. In such cases, clinical decision-making should always consider the underlying cause of thrombocytopenia (e.g., immune thrombocytopenia, bone marrow suppression, bone marrow infiltration by malignancy, portal hypertension, or hypersplenism without splenomegaly). It is important to recognize that not all forms of thrombocytopenia are bleeding disorders—some represent prothrombotic states, such as heparin-induced thrombocytopenia, vaccine-induced immune thrombotic thrombocytopenia, thrombocytopenia in antiphospholipid syndrome, or thrombotic microangiopathies []. In some cases, laboratory results may reflect pseudothrombocytopenia caused by EDTA-dependent platelet clumping—this is an in vitro artifact that can lead to mismanagement. If suspected, repeat testing using blood collected in a heparin- or citrate-containing tube is recommended [].

The recommendation of a platelet threshold of 20,000/μL for bronchoscopy without biopsy is not definitive (it is a conditional recommendation). The risk of significant bleeding also depends on several additional factors, including the endoscopist’s skill, the route of bronchoscope insertion (oral insertion is now preferred, as nasal insertion is associated with a higher risk of hemorrhagic complications), patient cooperation during the procedure, and other clinical variables. All of these factors should be carefully considered and weighed against the potential benefits of the procedure. When the expected clinical benefit outweighs the risks, bronchoscopy may be performed even in patients with a platelet count below 20,000/µL. In patients with a platelet count below the recommended thresholds, it is advisable to consider postponing the procedure, implementing measures to increase the platelet count, and consulting a hematologist. Interventions aimed at increasing platelet count depend on the underlying cause of thrombocytopenia and most commonly include platelet transfusion, but may also involve pharmacologic treatments such as corticosteroids, intravenous immunoglobulins, thrombopoietin receptor agonists (romiplostim, eltrombopag, avatrombopag), and other immunosuppressive agents [].

Primary data: No primary studies directly addressing the clinical question were identified. However, several studies evaluated the predictive value of coagulation parameters (INR and APTT) for bleeding risk during bronchoscopy. In a prospective study evaluating 104 consecutive TBLB procedures performed in 51 patients (including some lung transplant recipients), clinically significant bleeding—defined as >20 mL of blood—occurred in 8 cases (7.7%). Bleeding risk did not correlate with coagulation parameters, although most patients likely had normal APTT and INR values []. In another retrospective study, 305 bronchoscopies were performed in 274 patients, of whom 28 had elevated APTT and/or INR (although the degree of abnormality was not reported). The incidence of bleeding in patients with abnormal coagulation parameters was not higher than in those with normal values []. A third study evaluated potential bleeding risk factors based on the volume of blood aspirated during bronchoscopy in 234 patients. It found that an INR > 1.3 was not associated with an increased risk of bleeding [].

The 2019 guidelines of the Indian Chest Society, National College of Chest Physicians, and Indian Association for Bronchology do not recommend routine testing of coagulation parameters, platelet count, or hemoglobin levels prior to bronchoscopy (conditional recommendation) []. However, they suggest—as a good practice point—that such tests should be performed prior to bronchoscopy in patients with clinical risk factors for bleeding, including the use of anticoagulants, known bleeding disorders, or chronic liver or kidney disease. Similarly, the 2024 guidelines of the British Society of Haematology include a conditional recommendation advising against routine coagulation or platelet function testing to assess bleeding risk prior to invasive procedures [].

Available scientific data and clinical experience do not support the routine assessment of INR and APTT in patients undergoing bronchoscopy who are not receiving anticoagulant therapy, have no comorbidities warranting coagulation testing, and no clinical history suggestive of a bleeding tendency. Assessment of clinical bleeding risk factors is essential for evaluating the risk of hemorrhagic complications during bronchoscopy.

The authors of this guideline believe that current evidence and clinical practice do not justify routine INR and APTT testing in all patients undergoing evaluation for bronchoscopy. This applies to patients who: (1) do not have comorbid conditions associated with coagulopathy; (2) do not exhibit clinical signs suggestive of a bleeding disorder (e.g., a HEMSTOP score ≥ 2); and (3) are not taking anticoagulant drugs. A laboratory test should be considered clinically useful if its result positively impacts diagnostic or therapeutic outcomes []. Current evidence indicates that routine coagulation screening tests—including APTT and PT, from which INR is calculated—do not meet this criterion of clinical usefulness in patients undergoing invasive procedures or surgery. In other words, estimating bleeding risk based solely on routine preprocedural APTT and INR testing lacks clinical justification, as prolonged APTT or elevated INR values do not correlate with an increased risk of periprocedural bleeding. More importantly, normal APTT and INR values do not exclude the presence of a bleeding disorder, such as von Willebrand disease, platelet function disorders, or mild clotting factor deficiencies.

Therefore, in patients without conditions known to significantly increase bleeding risk—such as advanced chronic liver or kidney disease, severe systemic infections, or other serious conditions that may cause coagulopathy (e.g., malnutrition or long-term antibiotic use)—and who are not receiving chronic anticoagulant therapy (e.g., VKAs) or DOACs, bleeding risk assessment should primarily rely on a structured personal and family history (e.g., using the HEMSTOP questionnaire []; ), without simultaneous APTT or INR testing. If a patient scores < 2 points on the HEMSTOP questionnaire, invasive procedures, particularly those associated with a low risk of bleeding, can be performed without additional laboratory testing of plasma coagulation. The HEMSTOP questionnaire should not be viewed as a tool to quantify the bleeding risk before bronchoscopy (it was not validated in population of patients qualified for bronchoscopy) but rather as a tool to identify subjects who have increased bleeding risk and warrant additional investigation.

In all patients with signs or symptoms suggestive of a bleeding disorder and/or a HEMSTOP score ≥ 2, as well as in those with comorbidities that increase the risk of periprocedural bleeding, additional laboratory evaluation is warranted. In many cases, this extends beyond routine hemostasis screening tests. Therefore, consultation with a hematologist—preferably one specializing in bleeding and coagulation disorders—is recommended in such situations. Hematology consultation is also appropriate when abnormal hemostasis screening test results (e.g., prolonged APTT) are incidentally found in patients with a negative personal and family bleeding history, a low HEMSTOP score, and no significant comorbidities. This is important because some causes of prolonged APTT—such as the presence of lupus anticoagulant or deficiencies in factor XII, high-molecular-weight kininogen, or prekallikrein—are not associated with an increased risk of bleeding []. Appropriate guidance on further laboratory workup is necessary to promptly confirm such findings.

This practice point should not be viewed as a contraindication for performing coagulation tests in subjects qualified for diagnostic bronchoscopy, but rather as a suggestion that a clinical-based approach should be used here instead of relying solely on laboratory test results. When evaluating a patient for bronchoscopy, special attention should also be given to those with conditions known to increase the risk of bleeding during invasive procedures, particularly patients: With chronic liver disease, especially cirrhosis [,,];