Percutaneous dilatational tracheostomy with versus without bronchoscopic guidance: a systematic review and meta-analysis of randomized controlled trials

Abstract

Background

Percutaneous dilatational tracheostomy (PDT) is a critical intervention for patients with respiratory failure. However, the value of bronchoscopic guidance during PDT remains clinically debated. We conducted a systematic review and meta-analysis to evaluate the role of bronchoscopy during PDT by assessing its impact on complication rates and other clinically relevant outcomes.

Methods

A systematic search was conducted in PubMed, Embase, the Cochrane library and Web of Science. Randomized controlled trials (RCTs) comparing PDT performed with versus without bronchoscopic guidance were included. The risk of bias of the RCTs was rated using the revised Cochrane risk of bias, version 2 tool (RoB 2) and the certainty of evidence was assessed by the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework. The primary outcomes were complications including bleeding, misplacement (paratracheal placement of the tube), and posterior tracheal wall injury. The secondary outcomes included operating time and first-attempt success of tracheal puncture. Pooled effect estimates were calculated using risk ratios for dichotomous outcomes and mean differences for continuous outcomes, with 95% confidence intervals.

Results

Seven studies (n = 639 patients) were included. The bronchoscopic guidance group showed reduced incidence of bleeding (RR 0.57, 95% CI, 0.37 to 0.88), misplacement (RR 0.17, 95% CI, 0.06 to 0.49), and posterior tracheal wall injury (RR 0.08, 95% CI, 0.02 to 0.32) compared to the non-bronchoscopic group with moderate-certainty evidence. Additionally, the bronchoscopic guidance group had a higher first-attempt success rate of tracheal puncture (RR 1.33, 95% CI, 1.21 to 1.46, high-certainty evidence). No significant difference in operating time was found between the two groups (MD 1.14, 95% CI, -1.52 to 3.79, very low-certainty evidence).

Conclusion

Compared with the non-bronchoscopic method, bronchoscopic guidance during PDT may improve procedural safety and efficacy. However, high-quality, larger RCTs are required to draw more definitive conclusions.

Trial registration

PROSPERO (registration number: CRD42024508516).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12871-026-03759-2.

Introduction

Since its initial description by Ciaglia et al. in 1985 [1], percutaneous dilatational tracheostomy (PDT) has become a widely adopted bedside procedure for managing mechanically ventilated patients in the intensive care unit (ICU) [2]. It offers not only clinical benefits, but also represents a resource-efficient alternative to surgical tracheostomy [3]. Nevertheless, the procedure is associated with several complications. Early complications include bleeding, stomal infection, subcutaneous emphysema, posterior tracheal wall injury, and tube obstruction or dislodgement [4, 5]. Late complications include tracheal stenosis, malacia, tracheoinnominate artery fistula, tracheoesophageal fistula, and tracheocutaneous fistula [5, 6].

Bronchoscopy offers real-time luminal visualization during dilation, allowing accurate needle placement by guiding intercartilaginous space selection and midline puncture, thereby minimizing the risk of blind tracheal injury [7, 8]. Consequently, bronchoscopic guidance is widely regarded as the standard technique for PDT [9]. However, several studies suggest that the incidence of complications is low with or without bronchoscopic guidance [10–12], raising the question of whether visual guidance is essential. Additionally, bronchoscopy adds to the procedural cost and requires the involvement of specialists, such as bronchoscopists and intensivists, thereby increasing logistical complexity. Thus, the necessity of routine bronchoscopic guidance for PDT remains controversial.

To address this uncertainty, we conducted a systematic review and meta-analysis to evaluate the utility of bronchoscopy during PDT by assessing its impact on complication rates and other clinically relevant outcomes.

Methods

This systematic review and meta-analysis was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [13]. The study protocol was registered in PROSPERO (registration number: CRD42024508516).

Data sources and search strategy

Detailed search strategies are provided in the Supplementary Materials: Table S1. PubMed, Web of Science, EMBASE, and the Cochrane Library were searched from their inception to 20 August 2025. Only articles published in English or Chinese were included.

Study selection

Eligible studies met the following criteria: (1) randomized controlled trials (RCTs); (2) adult patients (older than 18 years); (3) comparison between PDT with versus without bronchoscopic guidance. The exclusion criteria were: (1) abstracts, case reports, letters, editorials, or conference articles; (2) repeated studies and data; (3) studies with unavailable data.

Data extraction

Search results were merged, and duplicates were removed. Two reviewers (XD and XG) independently screened titles and abstracts for eligibility and subsequently assessed the full texts of potentially relevant studies. Data were extracted using a predefined form. Any disagreements were resolved through discussion with a third reviewer (YZ).

Outcomes

The primary outcomes were procedure-related complications, including bleeding, misplacement (paratracheal placement of the tube), and posterior tracheal wall injury. Secondary outcomes included operating time and first-attempt success of tracheal puncture.

Risk of bias and evidence quality assessment

Two reviewers (XD and XG) independently assessed the risk of bias of included RCTs using the revised Cochrane risk of bias, version 2 (RoB 2) tool [14]. The RoB 2 judgment for each domain was generated by the tool’s algorithm, which is based on responses to a series of signaling questions. Disagreements were resolved by consensus.

The overall certainty of evidence for each outcome was evaluated using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework [15]. The GRADEprofiler (GRADEpro version: 3.6) was used to generate the results. Two investigators independently performed the assessments and reached consensus through discussion.

Statistical analysis

Dichotomous outcomes were pooled using the Mantel-Haenszel method and reported as risk ratios (RRs) with 95% confidence intervals (CIs). Continuous outcomes were pooled using the inverse variance method and expressed as mean differences (MDs) with 95% CIs. When necessary, medians and interquartile ranges were converted to means and standard deviations using the methods proposed by Luo et al. and Wan et al. [16, 17].

Statistical heterogeneity was assessed through the chi-squared (χ²) test (Cochran’s Q) and inconsistency index (I²) [18]. χ²p value < 0.05 or I² > 50% indicated significant heterogeneity. A random-effect model was applied to estimate the combined MD or RR in cases of substantial heterogeneity. Otherwise, the fixed-effect model was used. Publication bias was assessed visually using funnel plots. For sensitivity analysis, we employed the leave-one-out remove method to explore the impact of individual studies on the overall results for outcomes exhibiting significant heterogeneity [19].

All statistical analyses were performed using Review Manager 5.4 version (Cochrane Collaboration, Oxford, UK) or Stata/MP 16.0 version (Stata Corp, College Station, TX, USA). A P value < 0.05 was considered statistically significant.

Results

A total of 1317 records were identified through our search strategy. After screening, 24 studies underwent full-text review. The study selection process is summarized in Fig. 1. Among these, seven studies met the inclusion criteria and were included in the final quantitative synthesis [12, 20–25], encompassing a total of 639 patients.

Fig. 1.

Flowchart of the selection process for the included studies

Characteristics of included studies

The key characteristics of the included studies and patients are summarized in Table 1. All studies were published between 2013 and 2022 and collectively enrolled 639 patients from five countries. Of these, 320 patients underwent PDT with bronchoscopic guidance, and 319 underwent PDT without bronchoscopy.

Table 1.

Baseline characteristic of studies included in the meta-analysis

Authors (year)	Study period	Country	Study type	Setting	No. patients (BGG/NGB)	Age (mean, year) (BGG/NGB)	Male (BGG/NGB)	Days intubated (mean, day) (BGG/NGB)
Zhang (2017) [24]	2011–2016	China	Single-center RCT	ICU	60/60	60.5/59.2	29/28	NA
Zhao (2022) [25]	2020–2022	China	Single-center RCT	Anesthesiology	30/30	60.9/60.9	25/22	NA
Hassanin (2013) [20]	NA	Egypt	Single-center RCT	ICU	15/15	58.7/53.5	9/7	12.7/11.2
Saritas (2016) [22]	2013–2014	Turkey	Single-center RCT	Anesthesiology	30/30	68.7/74.5	12/13	NA
Naqv (2017) [21]	2015–2016	Pakistan	Single-center RCT	ICU	53/50	44/45	24/24	10/9
Taha (2017) [12]	2008–2011	Abu Dhabi	Single-center RCT	ICU	87/89	45.9/46.3	52/49	9.6/10.2
Shen (2019) [23]	2017–2018	China	Single-center RCT	ICU	45/45	61/63	10/13	7/7

BGG bronchoscopic guidance group, ICU intensive care unit, NA not applicable, NBG non-bronchoscopic group, RCT randomized controlled trial

Primary outcomes

All seven studies reported bleeding. Meta-analysis indicated a significantly lower incidence of bleeding in the bronchoscopic guidance group (RR 0.57, 95% CI, 0.37 to 0.88, P = 0.01, I² = 0%; Fig. 2). Misplacement was reported in 6 trials with a significantly lower incidence in the bronchoscopic guidance group (RR 0.17, 95% CI, 0.06 to 0.49, P = 0.001, I² = 0%; Fig. 3). Posterior tracheal wall injury was reported in four studies, and occurred less frequently in the bronchoscopic guidance group (RR 0.08, 95% CI, 0.02 to 0.32, P = 0.0004, I² = 0%; Fig. 4).

Fig. 2.

Forest for bleeding. BGG: bronchoscopic guidance group; NBG: non-bronchoscopic group

Fig. 3.

Forest for misplacement. BGG: bronchoscopic guidance group; NBG: non-bronchoscopic group

Fig. 4.

Forest for posterior tracheal wall injury. BGG: bronchoscopic guidance group; NBG: non-bronchoscopic group

Secondary outcomes

All seven studies reported operating time. Pooled analysis showed no significant difference between the two groups (MD 1.14, 95% CI, -1.52 to 3.79, P = 0.40, I² = 98%; Fig. 5). Five studies reported first-attempt success of tracheal puncture. The bronchoscopic guidance group had a significantly higher first-attempt success rate (RR 1.33, 95% CI, 1.21 to 1.46, P < 0.00001, I² = 46%; Fig. 6).

Fig. 5.

Forest for operating time. BGG: bronchoscopic guidance group; NBG: non-bronchoscopic group

Fig. 6.

Forest for first-attempt success of tracheal puncture. BGG: bronchoscopic guidance group; NBG: non-bronchoscopic group

Sensitivity analysis

A leave-one-out sensitivity analysis was performed for operating time to assess the stability of the pooled estimate (Supplementary Materials: Fig. S1). The results remained consistent after sequentially excluding each study, indicating that the overall effect was not driven by any single study.

RoB 2 and GRADE assessment

The RoB 2 judgment for each domain is presented in the Supplementary Materials (Table S2). Six of the seven studies were judged to have a low risk of bias across all domains. One study (Taha et al.) was judged to have “some concerns” overall because its randomization process [12], which used an odd-even allocation method, did not conceal allocation and is therefore at risk for selection bias. Visual inspection of the funnel plots were not interpretable due to the limited number of trials included (Supplementary Materials: Fig. S2).

According to the GRADE framework, the certainty of evidence was rated as high for first-attempt success of tracheal puncture, moderate for outcomes such as misplacement, bleeding, and posterior tracheal wall injury, and very low for operating time (Table 2).

Table 2.

GRADE evidence profile for the studies in the meta-analysis

Quality assessment							Patients, number		Effect		Quality
Studies, number	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	BGG	NBG	Relative(95% CI)	Absolute
Misplacement
6	RCT	Not serious	Not serious	Not serious	Seriousa	None	1/233 (0.43%)	20/230 (8.7%)	RR 0.17 (0.06 to 0.49)	72 fewer per 1000 (from 44 fewer to 82 fewer)	Moderatea
Bleeding
7	RCT	Not serious	Not serious	Not serious	Seriousa	None	27/320 (8.4%)	47/319 (14.7%)	RR 0.57 (0.37 to 0.88)	63 fewer per 1000 (from 18 fewer to 93 fewer)	Moderatea
Posterior tracheal wall injury
4	RCT	Not serious	Not serious	Not serious	Seriousa	None	0/135 (0%)	24/135 (17.8%)	RR 0.08 (0.02 to 0.32)	164 fewer per 1000 (from 121 fewer to 174 fewer)	Moderatea
Operating time
7	RCT	Not serious	Very seriousb	Not serious	Seriousc	None	320	319	-	MD 1.14 higher (1.52 lower to 3.79 higher)	Very lowb, c
First-attempt success of tracheal puncture
5	RCT	Not serious	Not seriousd	Not serious	Not serious	None	225/252 (89.3%)	171/254 (67.3%)	RR 1.33 (1.21 to 1.46)	222 more per 1000 (from 141 more to 310 more)	Highd

BGG bronchoscopic guidance group, CI confidence interval, MD mean difference, NBG non-bronchoscopic group, RR risk ratio, RCT randomized controlled trial. aWe downgraded by one level, for imprecision. The number of events was lower than the optimal information size. bWe downgraded by two levels, for inconsistency. Significant inconsistency was present (I2=98%). cWe downgraded by one level for imprecision; the confidence interval contained significant benefit and harm. dSignificant inconsistency was not present (I2=43%)

Discussion

This systematic review and meta-analysis of seven RCTs, encompassing 639 patients, provides a comprehensive evaluation of the role of bronchoscopic guidance during PDT. Based on moderate-certainty evidence, compared with the non-bronchoscopic approach, bronchoscopic guidance significantly reduced the incidence of bleeding, misplacement, and posterior tracheal wall injury, and improved the first-attempt needle success rate (high-certainty evidence), without significantly prolonging operating time (very low-certainty evidence).

Bronchoscopic guidance is widely recommended during PDT to improve visualization, confirm correct tube placement through direct carina identification, and potentially reduce complications [7, 26, 27]. Our pooled analysis demonstrated that, compared to the control, bronchoscopic guidance significantly reduced the risk of bleeding, posterior tracheal wall injury, and misplacement, while simultaneously increasing the rate of first-attempt success. This strongly supports the hypothesis that real-time visualization afforded by bronchoscopy enhances procedural accuracy, enabling precise needle puncture and dilator placement, which minimizes traumatic injury to adjacent structures and vasculature. However, the robustness of these findings is challenged by the overall low event rates of complications observed—a finding consistent with the existing literature—which resulted in serious imprecision and downgrading of the evidence quality within our meta-analysis [28–31]. Therefore, while these outcomes align well with the postulated benefits of bronchoscopic guidance, they warrant further validation through larger-scale RCTs to draw more definitive conclusions.

Our analysis found no statistically significant difference in overall operating time between the two groups. As previously suggested, during landmark-based techniques, operators often need to make repeated adjustments or attempts due to inaccurate localization or the occurrence of complications, which can consume considerable time. Although bronchoscopy introduces additional steps for insertion and positioning, the real-time visual feedback it provides may improve overall precision and efficiency. Nevertheless, substantial heterogeneity was observed for this outcome, and the overall certainty of evidence was very low, limiting the reliability of this finding. Although sensitivity analysis suggested the result was robust, the source of heterogeneity remained unclear. This may be attributed to variations in the definition and measurement of operating time across studies. For example, Zhao et al. defined operating time as the period from skin incision to bronchoscopic confirmation of tube placement [25], while other studies did not provide a clear definition. Additionally, differences in operator experience and procedural protocols may have contributed to the variability. Thus, no definitive conclusion can be drawn regarding operating time, underscoring the need for further research.

Beyond the reduction in short-term complications demonstrated in this analysis, bronchoscopic guidance may also decrease the incidence of long-term sequelae. For instance, Zhang et al. observed a notable reduction in tracheoesophageal fistula rates when bronchoscopy was employed [24]. However, the included studies primarily assessed early outcomes, and evidence regarding long-term benefits remains limited. Future trials with extended follow-up are needed to clarify this potential advantage. Conversely, bronchoscopy itself carries inherent risks, including transient airway obstruction, hypoxemia, arrhythmias, infection introduction, and airway reactivity [32–35]. These factors may offset some procedural benefits, particularly in high-risk patients. Thus, while our results support the safety and efficacy of bronchoscopic guidance during PDT, clinicians should individualize its use by weighing visual guidance benefits against procedural risks, operator experience, and patient-specific factors.

Several limitations of this meta-analysis should be acknowledged. (1) Only seven studies with relatively small sample sizes were included. (2) There was variability in PDT protocols among the studies. (3) Operating time was not clearly defined in most studies. (4) The restriction to English and Chinese publications may have introduced language bias. (5) The exclusion of grey literature (e.g., trial registries, conference abstracts) from our search strategy may limit the comprehensiveness of our findings. (6) The enrolled populations differed with respect to underlying diagnoses, comorbidities, and indications for tracheostomy.

Conclusion

Based on the available evidence, our meta-analysis suggests that bronchoscopic guidance during percutaneous dilatational tracheostomy may significantly reduce the risk of common clinical complications, including bleeding, tube misplacement, and posterior tracheal wall injury, and improve the first-attempt success rate of the initial needle puncture. However, robust, large trials are warranted to make definitive statements.

Abbreviations

BGG

Bronchoscopic Guidance Group

CIs

Confidence Intervals

GRADE

Grading of Recommendations Assessment, Development and Evaluation

ICU

Intensive Care Unit

MDs

Mean Differences

NA

Not Applicable

NBG

Non-bronchoscopic Group

PDT

Percutaneous Dilatational Tracheostomy

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

RCTs

Randomized Controlled Trials

RoB 2

Cochrane risk of bias, version 2 tool

RRs

Risk Ratios

Authors’ contributions

YZ served as the lead investigator and is the guarantor of this paper. DX conceived the study and designed the protocol. XG was responsible for data collection, performed the analysis, and contributed to data interpretation. YZ drafted the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by the Construction Fund of Key Medical Disciplines of Hangzhou (Grant number 2025HZZD04).

Data availability

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

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.