This cohort study examines demographic and clinical factors associated with a tracheotomy being performed by an otolaryngologist vs another specialist and with open vs percutaneous tracheotomy among patients with cardiopulmonary failure.
Key Points
Question
Are there patient-centric clinical and demographic characteristics associated with undergoing a tracheotomy performed by an otolaryngology service?
Findings
In this cohort study of 2929 patients from 8 academic institutions, self-identified Black race, history of neck surgery, morbid obesity, and ongoing antiplatelet and/or anticoagulation therapy were associated with greater odds of an otolaryngologist service vs other services performing tracheotomy in patients with cardiopulmonary failure.
Meaning
The findings suggest that otolaryngologists are significantly more likely than other specialists to perform a tracheotomy for patients with history of neck surgery, morbid obesity, and ongoing anticoagulation therapy—3 factors associated with a high level of surgical complexity.
Abstract
Importance
Tracheotomies are frequently performed by nonotolaryngology services. The factors that determine which specialty performs the procedure are not defined in the literature but may be influenced by tracheotomy approach (open vs percutaneous) and other clinicodemographic factors.
Objective
To evaluate demographic and clinical characteristics associated with tracheotomies performed by otolaryngologists compared with other specialists and to differentiate those factors from factors associated with use of open vs percutaneous tracheotomy.
Design, Setting, and Participants
This multicenter, retrospective cohort study included patients aged 18 years or older who underwent a tracheotomy for cardiopulmonary failure at 1 of 8 US academic institutions between January 1, 2013, and December 31, 2016. Data were analyzed from September 2022 to July 2023.
Exposure
Tracheotomy.
Main Outcomes and Measures
The primary outcome was factors associated with an otolaryngologist performing tracheotomy. The secondary outcome was factors associated with use of the open tracheotomy technique.
Results
A total of 2929 patients (mean [SD] age, 57.2 [17.2] years; 1751 [59.8%] male) who received a tracheotomy for cardiopulmonary failure (652 [22.3%] performed by otolaryngologists and 2277 [77.7%] by another service) were analyzed. Although 1664 of all tracheotomies (56.8%) were performed by an open approach, only 602 open tracheotomies (36.2%) were performed by otolaryngologists. Most tracheotomies performed by otolaryngologists (602 of 652 [92.3%]) used the open technique. Multivariable regression analysis revealed that self-reported Black race (odds ratio [OR], 1.89; 95% CI, 1.52-2.35), history of neck surgery (OR, 2.71; 95% CI, 2.06-3.57), antiplatelet and/or anticoagulation therapy (OR, 1.74; 95% CI, 1.29-2.36), and morbid obesity (OR, 1.54; 95% CI, 1.24-1.92) were associated with greater odds of an otolaryngologist performing tracheotomy. In contrast, history of neck surgery (OR, 1.36; 95% CI, 0.96-1.92), antiplatelet and/or anticoagulation therapy (OR, 0.80; 95% CI, 0.56-1.14), and morbid obesity (OR, 0.94; 95% CI, 0.74-1.19) were not associated with undergoing open tracheotomy when performed by any service, and Black race (OR, 0.56; 95% CI, 0.44-0.71) was associated with lesser odds of an open approach being used. Age-adjusted Charlson Comorbidity Index score greater than 4 was associated with greater odds of both an otolaryngologist performing tracheotomy (OR, 1.26; 95% CI, 1.03-1.53) and use of the open tracheotomy technique (OR, 1.48, 95% CI, 1.21-1.82).
Conclusions and Relevance
In this study, otolaryngologists were significantly more likely than other specialists to perform a tracheotomy for patients with history of neck surgery, morbid obesity, and ongoing anticoagulation therapy. These findings suggest that patients undergoing tracheotomy performed by an otolaryngologist are more likely to present with complex and challenging clinical characteristics.
Introduction
Tracheotomy is a commonly performed procedure for patients with upper airway obstruction and prolonged mechanical ventilation. More than 100 000 tracheotomies are performed in the US annually, with more than two-thirds indicated for cardiopulmonary failure (CPF).1,2 In patients with CPF, tracheotomies provide several benefits, including improved ventilation and oxygenation. However, there are significant risks associated with performing tracheotomy, including infection, bleeding, and injury to surrounding structures, that pose notable challenges. Both surgical and nonsurgical services are equipped to perform tracheotomy in the hospital using either an open or a percutaneous technique. Traditional open tracheotomies continue to be readily performed, although increased experience with percutaneous tracheotomies has led to the use of both techniques for most critically ill patients.3,4
Factors that influence consulting otolaryngologists to perform a tracheotomy remain unknown. Otolaryngologists at many institutions do not regularly perform tracheotomies but are consulted for a subset of patients. Multiple preoperative factors may influence this decision, including institutional experience, patient-centric features, history, and clinical characteristics. We hypothesized that the preoperative decision to consult an otolaryngologist for a tracheotomy relies both on institutionally dependent consultation trends and other clinicodemographic factors, such as morbid obesity, prior head and neck surgery, and comorbidities, that are not well defined in the current literature. We further hypothesized that both otolaryngologists and open tracheotomy technique would be involved more often for patients with complex and challenging clinical characteristics. Taking this into account, we aimed to assess clinicodemographic factors associated with tracheotomies performed by otolaryngologists compared with other specialists. Then, we aimed to identify factors associated with performing open tracheotomy vs a percutaneous approach by all services to understand whether the decision to consult an otolaryngologist for the tracheotomy is distinct from the decision of which approach to use.
Methods
In this cohort study, electronic medical records at participating sites were queried for Common Procedural Terminology codes 31600, 31601, 31603, 31605, 31610, and 31612. Our protocols have been previously described.5 We chose to include Common Procedural Terminology codes that met our study criteria of patients aged 18 years or older who received a tracheotomy while cross-referencing the database to confirm eligibility. We excluded tracheotomies performed with other procedures and revision cases. Institutional review board approval was obtained at all participating sites (University of California San Diego, Mayo Clinic Scottsdale, University of California Davis, University of Wisconsin–Madison, Stanford University, University of Michigan, Medical University of South Carolina, University of Colorado, Cleveland Clinic, and Loma Linda University) that contributed to the multi-institutional study on tracheotomy. Informed consent was waived because all data were deidentified. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Study Population
A retrospective multi-institutional cohort study of adult patients aged 18 years or older who received a tracheotomy between January 1, 2013, and December 31, 2016, was conducted. Our analysis specifically evaluated patients receiving a tracheotomy for CPF, as we believed it offered the best avenue to understand patient-centric characteristics associated with the specialty asked to perform a tracheotomy procedure (Table 1). Other indications for tracheotomy, such as stenosis or malignant obstruction, were excluded as they are less likely to accurately represent clinicodemographic factors that influence decision-making and the majority of these procedures are performed by otolaryngology services. Only institutions where tracheotomy was performed by both otolaryngology and other services and where both open and percutaneous tracheotomies were performed were included in the present analysis (eTable 1 in Supplement 1).
Table 1. Indications for Tracheotomy.
| Indication | Patients, No. (%) | ||
|---|---|---|---|
| Otolaryngology service (n = 985) | Other services (n = 2429) | Total (N = 3414) | |
| Cardiopulmonary failure | 652 (66.2) | 2277 (93.7) | 2929 (85.8) |
| Obstructive malignancy | 178 (18.1) | 53 (2.2) | 231 (6.8) |
| Benign progressive airway obstruction | 142 (14.4) | 81 (3.3) | 223 (6.5) |
| Conversion from percutaneous tracheotomy or cricothyroidotomy | 13 (1.3) | 18 (0.7) | 31 (0.9) |
Data Collected
Variables collected included demographic data at the time of tracheotomy (age, biological sex, and race and ethnicity), comorbidities included in the Charlson Comorbidity Index (CCI), tracheotomy technique (open or percutaneous), time from intubation to tracheotomy (in days), history of neck surgery, anticoagulation and antiplatelet therapy status, reason for tracheotomy (eg, obstructive malignant neoplasm, CPF), and performing service. All variables in the study were collected by retrospective medical record review. Indication for tracheotomy was determined by review of individual operative notes. Patient race and ethnicity were self-identified and obtained from the patient’s electronic health record; categories were Black, Hispanic, non-Hispanic White, other (patients’ medical records across multiple institutions included this category without additional specification), and unknown. Race and ethnicity were included in the analysis because we wanted to highlight any potential differences in performance of tracheotomy, although we recognized that conclusions derived from our study would be limited and possibly oversimplified. Age-adjusted CCI score was assessed as a dichotomized variable (high vs low, contingent on the median CCI score of the population).
Statistical Analysis
Baseline clinicodemographics for each cohort (otolaryngology vs other services, percutaneous vs open tracheotomy) were reported using means and SDs for continuous variables and frequencies and percentages for categorical variables, with differences and Cohen d effect size and corresponding 95% CIs reported between groups. A logistic regression model was fit to identify clinical and demographic factors associated with otolaryngology services vs all other services performing a tracheotomy. Odds ratios (ORs) with corresponding 95% CIs were used to quantify estimated associations. Parameters found to be associated by univariable regression were included in the multivariable model. Similarly, a logistic regression model was fit to identify factors associated with open vs percutaneous tracheotomy being performed. Stata/IC, version 28.0 (StataCorp LLC) was used for statistical analyses. Data were analyzed from September 2022 to July 2023.
Results
Baseline Characteristics
A total of 2929 patients who underwent a tracheotomy for CPF (652 tracheotomies [22.3%] were performed by otolaryngology services and 2277 [77.7%] by other services) met the inclusion criteria (Table 2). Mean (SD) age of the study population was 57.2 (17.2) years; 1178 patients (40.2%) were female, and 1751 (59.8%) were male. A total of 578 patients (19.7%) were Black; 373 (12.7%), Hispanic; 1655 (56.5%), non-Hispanic White; 212 (7.2%), other race and ethnicity; and 111 (3.8%), unknown race and ethnicity. Performing services in our analysis included otolaryngology (652 tracheotomies [22.3%]), pulmonology (461 [15.7%]), cardiothoracic surgery (459 [15.7%]), general surgery (710 [24.2%]), burn surgery (12 [0.4%]), trauma surgery (590 [20.1%]), or other unspecified service (45 [1.5%]). Compared with patients who underwent tracheotomy from other services, patients who had a tracheotomy performed by an otolaryngologist were older (mean [SD] age, 59.1 [16.5] years vs 56.8 [17.3] years; Cohen d, 0.14; 95% CI, 0.05-0.22), more likely to be female (295 [45.3%] vs 883 [38.8%]; difference, 6.47 percentage points; 95% CI, 4.61-8.34 percentage points) and to identify as Black (191 [29.3%] vs 387 [17.0%]; difference, 12.30 percentage points; 95% CI, 10.34-14.35 percentage points), and less likely to identify as Hispanic (32 [4.9%] vs 341 [15.0%]; difference, 10.07 percentage points; 95% CI, 9.65-10.15 percentage points). Patients who had tracheotomy performed by an otolaryngologist had more comorbidities (age-adjusted CCI score >4, 281 patients [43.1%] vs 859 patients [37.7%]; difference, 5.37 percentage points; 95% CI, 3.53-7.25 percentage points) but were less likely to have peripheral vascular disease or liver disease. Compared with other specialists, otolaryngologists performed tracheotomies for a greater proportion of patients who had morbid obesity (190 [29.1%] vs 443 [19.5%]; difference, 9.69 percentage points; 95% CI, 7.83-11.65 percentage points), were taking antiplatelet and/or anticoagulation therapy (86 [13.2%] vs 166 [7.3%]; difference, 5.90 percentage points; 95% CI, 4.43-7.60 percentage points), and had a history of neck surgery (120 [18.4%] vs 165 [7.2%]; difference, 11.16 percentage points; 95% CI, 9.29-13.21 percentage points). Patients who received tracheotomy from otolaryngologists were slightly more likely to be discharged with a tracheostomy than were patients who received tracheotomy from other services (465 of 652 [71.3%] vs 1498 of 2277 [65.8%]; difference, 5.53 percentage points; 95% CI, 3.88-7.03 percentage points).
Table 2. Baseline Clinicodemographics of Patients With Tracheotomy Performed by Otolaryngology Services vs Other Services.
| Characteristic | Patients, No. (%) | Difference, percentage points (95% CI) | |
|---|---|---|---|
| Otolaryngology services (n = 652) | Other services (n = 2277) | ||
| Age, mean (SD), y | 59.1 (16.5) | 56.8 (17.3) | 0.14 (0.05-0.22)a |
| Sex | |||
| Female | 295 (45.3) | 883 (38.8) | 6.47 (4.61-8.34) |
| Male | 357 (54.8) | 1394 (61.2) | |
| Race and ethnicity | |||
| Black | 191 (29.3) | 387 (17.0) | 12.30 (10.35-14.35) |
| Hispanic | 32 (4.9) | 341 (15.0) | 10.07 (9.65-10.15) |
| Non-Hispanic White | 383 (58.7) | 1272 (55.9) | 2.88 (1.06-4.64) |
| Otherb | 34 (5.2) | 178 (7.8) | 2.60 (1.79-3.11) |
| Unknown | 12 (1.8) | 99 (4.3) | 2.51 (2.08-2.59) |
| Duration of mechanical ventilation prior to tracheotomy, mean (SD), d | 10.3 (7.9) | 10.8 (7.9) | 0.07 (0.02-0.15)a |
| Urgent or emergent tracheotomy | 38 (5.8) | 162 (7.2) | 1.29 (0.34-1.94) |
| History of neck surgery | |||
| Any | 120 (18.4) | 165 (7.2) | 11.16 (9.29-13.21) |
| Tracheotomy | 53 (8.1) | 27 (1.2) | 6.94 (5.37-8.78) |
| Neck dissection | 23 (3.5) | 40 (1.8) | 1.77 (0.99-2.86) |
| Other | 60 (9.2) | 125 (5.5) | 3.71 (2.51-5.18) |
| Prior radiation | |||
| No | 627 (96.2) | 2228 (97.9) | 1.77 (0.98-2.86) |
| Yes | 25 (3.8) | 47 (2.1) | |
| Prior solid tumor | |||
| Localized | 81 (12.4) | 194 (8.5) | 3.90 (2.58-5.46) |
| Metastasized | 37 (5.7) | 85 (3.7) | 1.94 (1.03-3.14) |
| Antiplatelet and/or anticoagulation therapy | |||
| No | 566 (86.8) | 2111 (92.7) | 5.90 (4.43-7.60) |
| Yes | 86 (13.2) | 166 (7.3) | |
| Morbid obesityc | 190 (29.1) | 443 (19.5) | 9.69 (7.83-11.65) |
| Age-adjusted CCI score >4 | 281 (43.1) | 859 (37.7) | 5.37 (3.53-7.25) |
| Congestive heart failure | |||
| No | 516 (79.1) | 1761 (77.4) | 1.80 (0.25-3.18) |
| Yes | 135 (20.7) | 514 (22.6) | |
| Peripheral vascular disease | |||
| No | 620 (95.1) | 2082 (91.5) | 3.61 (2.88-4.02) |
| Yes | 32 (4.9) | 194 (8.5) | |
| COPD | |||
| No | 522 (80.1) | 1919 (84.3) | 4.22 (2.69-5.93) |
| Yes | 130 (19.9) | 358 (15.7) | |
| Chronic kidney disease | |||
| No | 562 (86.2) | 1918 (84.2) | 1.96 (0.64-3.04) |
| Yes | 90 (13.8) | 359 (15.8) | |
| Liver disease | |||
| Mild | 12 (1.8) | 86 (3.8) | 1.94 (1.44-2.08) |
| Moderate | 9 (1.4) | 27 (1.2) | 0.19 (0.15-0.88) |
| Severe | 5 (0.8) | 76 (3.3) | 2.57 (2.38-2.39) |
| Diabetes | |||
| Uncomplicated | 168 (25.8) | 475 (20.9) | 4.91 (3.24-6.72) |
| End-organ damage | 32 (4.9) | 162 (7.1) | 2.21 (1.39-2.71) |
Abbreviations: CCI, Charlson Comorbidity Index; COPD, chronic obstructive pulmonary disease.
Cohen d statistics are given for age and duration of mechanical ventilation prior to tracheotomy.
Patients’ medical records included the category “other” without additional specification.
Defined as a body mass index greater than 35 (calculated as weight in kilograms divided by height in meters squared).
In the study cohort, more patients underwent open tracheotomy (1664 [56.8%]) compared with percutaneous tracheotomy (1265 [43.2%]) (eTable 2 in Supplement 1). Compared with patients who underwent percutaneous tracheotomy, patients who received open tracheotomy were older (mean [SD] age, 58.5 [16.9] years vs 55.7 [17.4] years; Cohen d, 0.16; 95% CI, 0.09-0.24), had more comorbidities (age-adjusted CCI score >4, 695 [41.8%] vs 445 [35.2%]; difference, 6.59 percentage points; 95% CI, 6.30-6.84 percentage points), and were more likely to have morbid obesity (384 [23.1%] vs 248 [19.7%]; difference, 3.39 percentage points; 95% CI, 3.19-3.55 percentage points) and a history of neck surgery (212 [12.7%] vs 73 [5.8%]; difference, 6.97 percentage points; 95% CI, 6.62-7.24 percentage points). There were no meaningful differences in sex and rates of antiplatelet and/or anticoagulation therapy between groups.
Distribution of Tracheotomies Performed by Institution
In patients receiving a tracheotomy for CPF, the majority of otolaryngologists used the open tracheotomy technique (602 of 652 [92.3%]). Among all institutions, the percentage of tracheotomies performed by an otolaryngologist ranged from 7.7% (25 of 323) to 41.4% (237 of 572); proportions of open tracheotomies ranged from 29.1% (88 of 302) to 92.5% (135 of 146) (Figure). Although 1664 tracheotomies (56.8%) were performed by an open approach, only 602 of the open tracheotomies (36.2%) were performed by an otolaryngologist (eTable 2 in Supplement 1). Thus, we identified factors associated with performing tracheotomy using the open technique compared with a percutaneous approach to assess whether characteristics associated with an otolaryngologist performing tracheotomy were also associated with an open surgical approach.
Figure. Specialist Service Performing Tracheotomy for Cardiopulmonary Failure and Tracheotomy Technique Used.
Clinicodemographic Factors Associated With an Otolaryngologist Performing Tracheotomy
Multivariable analysis revealed that self-reported Black race (OR, 1.89; 95% CI, 1.52-2.35), history of neck surgery (OR, 2.71, 95% CI, 2.06-3.57), antiplatelet and/or anticoagulation therapy (OR, 1.74; 95% CI, 1.29-2.36), and morbid obesity (OR, 1.54; 95% CI, 1.24-1.92) were associated with greater odds of an otolaryngologist performing tracheotomy, while Hispanic ethnicity was associated with lesser odds of an otolaryngologist performing tracheotomy (OR, 0.39; 95% CI, 0.26-0.57) (Table 3). In contrast, history of neck surgery (OR, 1.36; 95% CI, 0.96-1.92), antiplatelet and/or anticoagulation therapy (OR, 0.80; 95% CI, 0.56-1.14), and morbid obesity (OR, 0.94; 95% CI, 0.74-1.19) were not associated with undergoing open tracheotomy performed by any service, and Black race (OR, 0.56; 95% CI, 0.44-0.71) was associated with lesser odds of an open approach being used. Presence of more comorbidities (age-adjusted CCI score >4) was associated with greater odds of both an otolaryngologist performing tracheotomy (OR, 1.26; 95% CI, 1.03-1.53) and the open tracheotomy technique being used (OR, 1.48; 95% CI, 1.21-1.82) (Table 3 and eTables 3-4 in Supplement 1). Urgent or emergent tracheotomy was not associated with an otolaryngologist performing tracheotomy (OR, 0.81; 95% CI, 0.56-1.17) or with use of the open tracheotomy technique (OR, 1.03; 95% CI, 0.77-1.38).
Table 3. Logistic Regression Analyses.
| Variable | Odds ratio (95% CI) | |
|---|---|---|
| Univariable analysis | Multivariable analysis (n = 2667) | |
| Factors associated with otolaryngology services vs other services performing tracheotomy | ||
| Age | 1.01 (1.00-1.01) | NA |
| Male sex | 0.77 (0.64-0.91) | 0.84 (0.69-1.02) |
| Race | ||
| Black | 1.64 (1.33-2.02) | 1.89 (1.52-2.35) |
| Hispanic | 0.31 (0.21-0.46) | 0.39 (0.26-0.57) |
| Non-Hispanic White | 1 [Reference] | 1 [Reference] |
| Othera | 0.63 (0.43-0.93) | 0.80 (0.54-1.20) |
| Urgent or emergent tracheotomy | 0.81 (0.56-1.17) | NA |
| History of neck surgery | 2.89 (2.24-3.72) | 2.71 (2.06-3.57) |
| Age-adjusted CCI score >4 | 1.39 (1.16-1.67) | 1.26 (1.03-1.53) |
| Antiplatelet and/or anticoagulation therapy | 1.93 (1.47-2.55) | 1.74 (1.29-2.36) |
| Morbid obesityb | 1.70 (1.40-2.07) | 1.54 (1.24-1.92) |
| Factors associated with open tracheotomy vs percutaneous tracheotomy | ||
| Age | 1.01 (1.01-1.01) | NA |
| Male sex | 0.93 (0.80-1.08) | 1.15 (0.94-1.40) |
| Race and ethnicity | ||
| Black | 0.77 (0.64-0.93) | 0.56 (0.44-0.71) |
| Hispanic | 1.03 (0.82-1.30) | 4.08 (2.86-5.82) |
| Non-Hispanic White | 1 [Reference] | 1 [Reference] |
| Othera | 0.71 (0.53-0.94) | 0.91 (0.63-1.30) |
| Urgent or emergent tracheotomy | 1.03 (0.77-1.38) | NA |
| History of neck surgery | 2.38 (1.81-3.14) | 1.36 (0.96-1.92) |
| Age-adjusted CCI score >4 | 1.52 (1.30-1.77) | 1.48 (1.21-1.82) |
| Antiplatelet and/or anticoagulation | 1.11 (0.85-1.44) | 0.80 (0.56-1.14) |
| Morbid obesityb | 1.22 (1.02-1.47) | 0.94 (0.74-1.19) |
| Service | ||
| Otolaryngology | 1 [Reference] | 1 [Reference] |
| General surgery | 0.17 (0.12-0.23) | 0.15 (0.11-0.21) |
| Burn surgery | 0.08 (0.03-0.27) | 0.05 (0.01-0.17) |
| Trauma surgery | 0.09 (0.06-0.12) | 0.07 (0.05-0.09) |
| Cardiothoracic surgery | 0.12 (0.08-0.17) | 0.11 (0.07-0.15) |
| Pulmonology | 0.00 (0.00-0.00) | 0.00 (0.00-0.00) |
| Other | 0.02 (0.01-0.04) | 0.01 (0.00-0.02) |
Abbreviations: CCI, Charlson Comorbidity Index; NA, not applicable; OR, odds ratio.
Patients’ medical records included the category “other” without additional specification.
Defined as a body mass index greater than 35 (calculated as weight in kilograms divided by height in meters squared).
Factors Associated With Death Within 30 Days of Tracheotomy
A total of 329 patients (11.2%) died within 30 days of tracheotomy. On multivariable analysis, performance of tracheotomy by cardiothoracic surgery (OR, 2.57; 95% CI, 1.74-3.80) and pulmonology (OR, 2.67; 95% CI, 1.63-4.38) services was associated with greater odds of death within 30 days of tracheotomy (eTable 5 in Supplement 1). Patients who self-identified as Black (OR, 0.41; 95% CI, 0.28-0.61) were less likely to die within 30 days in this cohort. There was no association of 30-day mortality with open tracheotomy technique (OR, 1.18; 95% CI, 0.87-1.62).
Discussion
In this study, we found that otolaryngologists performed tracheotomy in a subset of patients with patient demographics and clinical characteristics different from those of patients who had tracheotomy performed by other services. Specifically, we found that self-identified Black race, history of neck surgery, ongoing antiplatelet and/or anticoagulation therapy, and morbid obesity were independently associated with greater odds of an otolaryngology service performing tracheotomy among patients with CPF. These factors were not associated with undergoing open tracheotomy. In contrast, patients with more comorbidities were more likely both to undergo tracheotomy with an open approach and to have tracheotomy performed by an otolaryngology service. Our analysis, which assessed patients receiving a tracheotomy at 8 academic institutions that performed both open and percutaneous tracheotomies by multiple specialty services, revealed that otolaryngologists were more likely to perform the procedure among patients with more medical complexity.
History of neck surgery, morbid obesity, and active anticoagulation and/or antiplatelet therapy are clinical characteristics that may complicate tracheotomy.6,7 Morbid obesity is independently associated with increased risk of complications and tracheostomy-related death.6,8 Intraoperative complications associated with both morbid obesity and a history of neck surgery include distortion of neck anatomy that may result in tube misplacement or injury to surrounding structures as well as special positioning, retraction, and approach contributing to prolonged surgical duration.9 Postoperative complications in patients with morbid obesity are multifactorial and include both tracheotomy-specific factors, such as accidental decannulation,10 and associated comorbidities, such as poor wound healing and increased risk of infection, which may require further intervention. Although the present study did not evaluate complications associated with morbid obesity, current literature suggests that morbid obesity is independently associated with more complex tracheotomy.6,11 Furthermore, active anticoagulation and/or antiplatelet therapy are directly associated with an elevated risk for intraoperative bleeding and postoperative hemorrhage. The most common early complication of tracheotomy is bleeding, with an estimated overall incidence of 5.7%.12 In contrast, the current literature suggests that anticoagulation and/or antiplatelet therapy may be associated with increased risk of bleeding, with an estimated incidence of 19% to 27%.13,14
Tracheotomies are increasingly being performed by nonotolaryngology services.15 Recent retrospective studies have demonstrated a 3.3% to 3.8% yearly decline in tracheotomies performed by otolaryngology services at large tertiary centers.15,16 Despite these findings, otolaryngologists with specialized training and airway experience continue to perform tracheotomy even in nonsurgical patients with indications due to CPF. There is some limited literature that supports our hypothesis that patient-specific clinicodemographic factors may influence the decision to consult an otolaryngologist. In a retrospective analysis of 805 tracheotomies performed at a single institution, Alfonso et al17 demonstrated that patients with morbid obesity were more likely to be referred to otolaryngology services than to other services (OR, 3.23; 95% CI, 2.21-4.72). In the present analysis, 2929 tracheotomies were performed by more than 6 specialty services across all institutions, and similar findings were demonstrated. Therefore, although otolaryngologists may be performing fewer tracheotomies,15,16 otolaryngology services are likely consulted to perform tracheotomies for patients with more clinical characteristics that increase surgical complexity compared with other services.
Open tracheotomy technique is commonly indicated in select patients to help mitigate factors that complicate a percutaneous approach, including gross distortion of the neck anatomy (tracheal deviation, morbid obesity, and short neck), history of neck surgery, and elevated bleeding risk. Thus, we chose to assess the same factors that were associated with greater odds of an otolaryngologist performing tracheotomy in models that assessed odds of performing an open surgical tracheotomy. Importantly, we found that a history of neck surgery, morbid obesity, and anticoagulation and/or antiplatelet therapy were independently associated with greater odds of an otolaryngologist performing the tracheotomy but were not associated with use of an open surgical approach.
Our multivariable analyses showed that self-identified Black race was associated with greater odds of an otolaryngologist performing tracheotomy and that self-identified Hispanic patients were less likely to have a tracheotomy performed by an otolaryngologist. We were unable to ascertain the reasons behind these findings. A possibility for these findings is that an unidentified confounding factor was associated with differences in these patient populations. A few studies revealed racial and ethnic differences in adult patients who received a tracheotomy18,19,20 and in outcomes associated with tracheotomy.21,22 A study of 86 246 patients with acute brain injury from the National Inpatient Sample database revealed that compared with White patients, Black patients had 20% increased risk of undergoing tracheotomy, Hispanic and Latino patients had 10% increased risk, and Asian patients had 8% increased risk.20 Furthermore, in a retrospective study of 523 tracheotomies performed at a single institution, multivariable analyses accounting for age, sex, race and ethnicity, insurance status, income, and comorbidities revealed that Hispanic or Latino ethnicity was associated with increased odds of 30-day mortality after tracheotomy (OR, 4.43; 95% CI, 1.26-15.60).22 These data suggest that there may be differences in the allocation, performance, and outcomes of tracheotomy based on race and ethnicity. We acknowledge that race and ethnicity are social constructs that are constantly evolving,23 and the discrepancies identified in our analysis merit further investigation.
Our study lacked tracheotomy-specific patient outcomes and was limited to mortality within 30 days of the tracheotomy. There was no association of 30-day mortality with open tracheotomy technique. We did, however, find that tracheotomies performed by cardiothoracic surgery and pulmonology services were associated with greater odds of death within 30 days of tracheotomy. We suspect that this finding is likely a reflection of inherent patient factors not well defined by our database.
Limitations
This study has several important limitations to consider, including its retrospective design and, as with all large database studies, erroneous data collection and coding errors. Furthermore, consultation by otolaryngology services is influenced by institutional factors, including experience in tracheotomy technique, resources, and availability of specialists, which may have introduced selection bias. To address this, we limited our study to institutions that performed both open and percutaneous tracheotomies and that included more than 2 specialty services performing tracheotomies. Future studies could assess early and late complications associated with tracheotomies performed by otolaryngologists who operate on more patients with history of neck surgery, morbid obesity, and elevated bleeding risks than specialists in other services, as demonstrated in the present study.
Conclusions
This cohort study found that history of neck surgery, morbid obesity, and anticoagulation and/or antiplatelet therapy were independently associated with greater odds of an otolaryngology service performing tracheotomy compared with other specialist services. Although tracheotomies are increasingly performed by nonotolaryngology services,15,16 the findings suggest that otolaryngologists are consulted to perform tracheotomy in a subset of patients with complex clinical characteristics (in particular, challenging anatomic scenarios) different from those of patients who have tracheotomy performed by other services.
eTable 1. Distribution of Tracheotomies Performed Among Services
eTable 2. Baseline Clinicodemographics—Open vs Percutaneous Tracheotomy
eTable 3. Logistic Regression—Factors Associated With Otolaryngology vs Other Services Performing Tracheotomy
eTable 4. Logistic Regression—Factors Associated With Open vs Percutaneous Tracheotomy Technique
eTable 5. Factors Associated With Mortality Within 30 Days of Tracheotomy
Data Sharing Statement
References
- 1.Yu M. Tracheostomy patients on the ward: multiple benefits from a multidisciplinary team? Crit Care. 2010;14(1):109. [DOI] [PMC free article] [PubMed]
- 2.Cheung NH, Napolitano LM. Tracheostomy: epidemiology, indications, timing, technique, and outcomes. Respir Care. 2014;59(6):895-915. doi: 10.4187/respcare.02971 [DOI] [PubMed] [Google Scholar]
- 3.Schwann NM. Percutaneous dilational tracheostomy: anesthetic considerations for a growing trend. Anesth Analg. 1997;84(4):907-911. doi: 10.1213/00000539-199704000-00037 [DOI] [PubMed] [Google Scholar]
- 4.Susanto I. Comparing percutaneous tracheostomy with open surgical tracheostomy: both will coexist until robust evidence becomes available. BMJ. 2002;324(7328):3-4. doi: 10.1136/bmj.324.7328.3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Panuganti BA, Pang J, Francis DO, et al. Clinicodemographic predictors of tracheotomy tube size and decannulation: a multiinstitutional retrospective cohort study on tracheotomy. Ann Surg. 2022. [DOI] [PubMed] [Google Scholar]
- 6.El Solh AA, Jaafar W. A comparative study of the complications of surgical tracheostomy in morbidly obese critically ill patients. Crit Care. 2007;11(1):R3. doi: 10.1186/cc5147 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Rana S, Pendem S, Pogodzinski MS, Hubmayr RD, Gajic O. Tracheostomy in critically ill patients. Mayo Clin Proc. 2005;80(12):1632-1638. doi: 10.4065/80.12.1632 [DOI] [PubMed] [Google Scholar]
- 8.Darrat I, Yaremchuk K. Early mortality rate of morbidly obese patients after tracheotomy. Laryngoscope. 2008;118(12):2125-2128. doi: 10.1097/MLG.0b013e3181847a78 [DOI] [PubMed] [Google Scholar]
- 9.Fattahi T, Chafin C, Bunnell A. Tracheostomy in the morbidly obese: difficulties and challenges. J Oral Maxillofac Surg. 2017;75(7):1372-1375. doi: 10.1016/j.joms.2016.12.030 [DOI] [PubMed] [Google Scholar]
- 10.Marshall RV, Haas PJ, Schweinfurth JM, Replogle WH. Tracheotomy outcomes in super obese patients. JAMA Otolaryngol Head Neck Surg. 2016;142(8):772-776. doi: 10.1001/jamaoto.2016.1089 [DOI] [PubMed] [Google Scholar]
- 11.Byhahn C, Lischke V, Meininger D, Halbig S, Westphal K. Peri-operative complications during percutaneous tracheostomy in obese patients. Anaesthesia. 2005;60(1):12-15. doi: 10.1111/j.1365-2044.2004.03707.x [DOI] [PubMed] [Google Scholar]
- 12.Delaney A, Bagshaw SM, Nalos M. Percutaneous dilatational tracheostomy versus surgical tracheostomy in critically ill patients: a systematic review and meta-analysis. Crit Care. 2006;10(2):R55. doi: 10.1186/cc4887 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Pasin L, Frati E, Cabrini L, et al. Percutaneous tracheostomy in patients on anticoagulants. Ann Card Anaesth. 2015;18(3):329-334. doi: 10.4103/0971-9784.159802 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Huang YH, Tseng CH, Chan MC, Lee BJ, Lin CH, Chang GC. Antiplatelet agents and anticoagulants increased the bleeding risk of bedside percutaneous dilational tracheostomy in critically ill patients. J Formos Med Assoc. 2020;119(7):1193-1200. [DOI] [PubMed]
- 15.Patel HH, Siltumens A, Bess L, Camacho F, Goldenberg D. The decline of tracheotomy among otolaryngologists: a 14-year review. Otolaryngol Head Neck Surg. 2015;152(3):465-469. doi: 10.1177/0194599814563513 [DOI] [PubMed] [Google Scholar]
- 16.Bowen AJ, Nowacki AS, Benninger MS, Lamarre ED, Bryson PC. Is tracheotomy on the decline in otolaryngology? a single institutional analysis. Am J Otolaryngol. 2018;39(2):97-100. [DOI] [PubMed]
- 17.Alfonso KP, Kaufman MR, Dressler EV, Liu M, Aouad RK. Otolaryngology consultation tracheostomies and complex patient population. Am J Otolaryngol. 2017;38(5):551-555. [DOI] [PubMed]
- 18.Plocienniczak M, Sambhu KM, Noordzij JP, Tracy L. Impact of socioeconomic demographics and race on laryngotracheal stenosis etiology and outcomes. Laryngoscope. 2023;133(4):908-913. [DOI] [PubMed]
- 19.Shaw JJ, Santry HP. Who gets early tracheostomy? evidence of unequal treatment at 185 academic medical centers. Chest. 2015;148(5):1242-1250. doi: 10.1378/chest.15-0576 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Jones RC, Creutzfeldt CJ, Cox CE, et al. Racial and ethnic differences in health care utilization following severe acute brain injury in the United States. J Intensive Care Med. 2021;36(11):1258-1263. doi: 10.1177/0885066620945911 [DOI] [PubMed] [Google Scholar]
- 21.Park C, Bahethi R, Yang A, Gray M, Wong K, Courey M. Effect of patient demographics and tracheostomy timing and technique on patient survival. Laryngoscope. 2021;131(7):1468-1473. doi: 10.1002/lary.29000 [DOI] [PubMed] [Google Scholar]
- 22.Bahethi R, Park C, Yang A, et al. Influence of insurance status and demographic factors on outcomes following tracheostomy. Laryngoscope. 2021;131(7):1463-1467. doi: 10.1002/lary.28967 [DOI] [PubMed] [Google Scholar]
- 23.Lindeborg M, Din T, Araya-Quezada C, et al. Race and ethnicity in otolaryngology academic publications. Otolaryngol Head Neck Surg. 2022;166(6):1196-1203. doi: 10.1177/01945998221084201 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Distribution of Tracheotomies Performed Among Services
eTable 2. Baseline Clinicodemographics—Open vs Percutaneous Tracheotomy
eTable 3. Logistic Regression—Factors Associated With Otolaryngology vs Other Services Performing Tracheotomy
eTable 4. Logistic Regression—Factors Associated With Open vs Percutaneous Tracheotomy Technique
eTable 5. Factors Associated With Mortality Within 30 Days of Tracheotomy
Data Sharing Statement