Tracheotomy Outcomes in Super Obese Patients

Importance  Surgeons need to understand the expected outcomes for super obese patients undergoing tracheotomy to appropriately counsel patients and families about likely risks and benefits.

Objective  To determine the outcomes, complications, and mortality after tracheotomy in super obese patients (those with a body mass index [BMI] greater than 50).

Design, Setting, and Participants  A retrospective review was conducted of billing records from a tertiary care academic medical center from November 1, 2010, through June 30, 2013, to identify patients undergoing tracheotomy. Medical records were reviewed to identify patients with a BMI (calculated as weight in kilograms divided by height in meters squared) greater than 50 and a control group with a BMI of 30 to 50. Patient characteristics, including BMI, age, race/ethnicity, primary diagnosis for hospitalization, medical comorbidities, and surgical technique, were measured.

Main Outcomes and Measures  The primary outcome measure was dependence on tracheostomy at discharge. Secondary outcomes included rates of ventilator dependence, mortality, postoperative complications, and discharge disposition.

Results  The super obese population included 31 patients and was predominantly African American (20 patients [65%]) and female (21 patients [68%]). Mean BMI of super obese patients was 64.0 (range, 50.2-95.5). The obese patient population was mainly African American (25 patients [74%]) and female (17 patients [50%]). Twenty-five of 31 super obese patients (81%) were discharged with a tracheotomy tube in place, compared with 16 of 34 obese patients (52%). Seven patients (23%) in the super obese group were ventilator dependent at discharge, compared with 4 patients (13%) in the obese group. Only 2 of the super obese patients (3%) were decannulated before discharge, compared with 15 (44%) in the obese group. In-hospital mortality was similar for the 2 groups (super obese, 4 patients [13%] and obese, 3 patients [9%]). The overall complication rate was 19% in the super obese group (6 patients) compared with 6% in the obese group (2 patients). Super obese patients were less likely to be discharged to a health care facility (17 patients [55%]) compared with patients in the obese group (22 patients [65%]).

Conclusions and Relevance  Tracheotomy in super obese patients is a safe and effective strategy for airway management. Critically ill, super obese patients have a high likelihood of remaining dependent on a tracheotomy or ventilator at the time of discharge.

Obesity rates continue to rise in the United States, with Mississippi at the forefront of this epidemic. The prevalence of individuals with a body mass index (BMI) (calculated as weight in kilograms divided by height in meters squared) greater than 50, commonly referred to as super obesity in the literature, has grown 3 times as fast as the prevalence of those with a BMI greater than 30.¹ Increasing cervical and submental adipose tissue decreases visualization and surgical access to the airway.² Standard-size tracheotomy tubes have been shown to fit poorly in the airways of obese and super obese patients owing to the increased distance from the skin to the tracheal wall.³ Logistical factors, such as transferring patients to a standard operating table, can also prove to be very difficult. As airway experts, otolaryngologists are typically called on to handle patients with the most difficult neck anatomy, including that due to severe obesity.

When considering ambulatory surgical procedures, super obese patients have more frequent postoperative complications and poorer outcomes compared with those with a lower BMI.^3,4 Super obese patients have significantly increased respiratory resistance, increased peak inspiratory pressure, and a decrease in dynamic respiratory compliance during bariatric surgery.⁵ Systolic and diastolic blood pressure and mean arterial pressure were also found to be reduced intraoperatively in super obese patients.⁵ These factors result in less respiratory reserve and at least a theoretical increased risk with airway surgery. The literature is divided on whether obese and morbidly obese patients show a greater risk of complications or mortality when undergoing tracheotomy.^2,6-12

Outcomes after tracheotomy in patients with a BMI greater than 50 have not been reported in the English-language literature, to our knowledge. It is important to understand the expected outcomes for these patients so that surgeons can appropriately counsel patients and families about likely risks, benefits, and the necessary care that will be required once the patient is discharged from the hospital. Analyzing complications also gives surgeons the opportunity to identify potential sources of complications so they can be avoided, thus improving the safety of this procedure. Understanding complications gives the patient, family, and surgeon more information with which to make a decision regarding care.

Our goal was to determine the primary outcome of tracheostomy status at the time of discharge in super obese (BMI >50) vs obese (BMI of 30-50) patients. Furthermore, we investigated outcomes related to mortality, ventilator dependence, disposition, and complication rate. We hypothesize that the super obese patient population undergoing tracheotomies at our institution as inpatients are more likely to have complications related to the procedure that will affect the course of their hospital stay when compared with an obese patient population undergoing the same procedure.

Box Section Ref ID

Inpatient hospital billing records from November 1, 2010, through June 30, 2013, at a tertiary care medical center in Jackson, Mississippi, were searched for the Current Procedural Terminology codes pertaining to adult tracheotomy (codes 31600 and 31601) to identify patients who had undergone tracheotomy. The University of Mississippi Institutional Review Board approved this study and waived the requirement for patient consent. Medical records were initially analyzed for data on height, weight, and BMI. Patients with a BMI greater than 50 were further analyzed, along with a comparable control group consisting of patients with a BMI of 30 to 50. Inpatient billing records revealed 811 patients who underwent tracheotomy. After review of these patients’ medical records, 31 patients with a BMI greater than 50 and 34 patients with a BMI of 30 to 50 were identified.

Detailed information was collected about each patient including weight, height, BMI, year of surgery, race/ethnicity, age, sex, health insurance status at the time of hospitalization, primary diagnosis for hospitalization, medical history, hospital length of stay, length of time receiving ventilator assistance, number of failed extubations, fraction of inspired oxygen before tracheotomy, service performing the tracheotomy, method of tracheotomy, details of the surgical technique, tracheotomy size, amount of blood loss, and presence of postoperative complications. Race/ethnicity was determined by the information obtained in the patient’s medical record and was used as a descriptive statistic. Medical comorbidities, including type 2 diabetes, hypertension, congestive heart failure, chronic heart disease, pulmonary hypertension, lung disease, chronic renal insufficiency, and obstructive sleep apnea, were recorded based on documentation of these conditions in the medical record during the patient’s hospitalization.

The use of different surgical techniques, including cervical defatting and Bjork flap, were recorded. Patients’ medical records were searched for documentation of postoperative complications, including pneumothorax, bleeding, accidental decannulation, death within 30 days of the procedure, and tracheal vascular fistulas.

We tested the association between BMI categories and nominal and interval study outcomes and comorbidities with the χ² and t test, respectively. We calculated the relative risk of significant predictors of outcomes. To study the independent effect of BMI on tracheotomy status at discharge, we entered each of the 8 comorbidities individually in 8 logistic regression models with BMI status as the other predictor variable and tracheotomy status at discharge as the outcome variable. For measures of effect size, we used odds ratios (ORs) and 95% CIs for logistic regressions, risk differences (RDs) and 95% CIs for difference in proportions, and Cohen d for interval scaled differences, based on recommendations by Ellis.¹³P = .05 was considered statistically significant. The study was powered using a β of 0.20 to detect a difference in the primary study outcome variable of tracheotomy status at discharge of 35% (ie, 50% vs 85%).

The super obese patient population was predominantly African American (20 [65%]) and female (21 [68%]) (Table 1). Mean BMI of super obese patients was 64.0 (range, 50.2-95.5). The obese patient population was mainly African American (25 [74%]) and female (17 [50%]). Comparison of age, race, and sex within the studied population and control group failed to reveal a statistically significant difference between the 2 groups.

For the primary study outcome, there was a significant association between BMI status and tracheotomy status at discharge. Patients in the super obese group were 80% more likely to have a tracheotomy at discharge (relative risk, 1.80; 95% CI, 1.26-2.57). For secondary outcome measures, we failed to find a significant association between BMI and death before discharge (RD, 0.0085; 95% CI, −0.14 to 0.17), ventilator dependence at discharge (RD, 0.121; 95% CI, −0.08 to 0.32), and disposition (death vs health care facility, RD, 0.05; 95% CI, −0.14 to 0.28; and death vs home, RD, 0.01; 95% C, −0.28 to 0.46).

As detailed in Table 2, Quiz Ref IDpatients in the super obese group had nonsignificant increases in rates of diabetes (RD, 0.23; 95% CI, −0.03 to 0.45; P = .06) and congestive heart failure (RD, 0.17; 95% CI, −0.04 to 0.38; P = .08). There was a significant association between BMI and chronic heart disease (RD, 0.23; 95% CI, 0.02-0.43; P = .02), pulmonary hypertension (RD, 0.19; 95% CI, 0.02-0.38; P = .007), lung disease (RD, 0.39; 95% CI, 0.16-0.58; P < .001), chronic renal insufficiency (RD, 0.25; 95% CI, 0.07-0.45; P = .002), and obstructive sleep apnea (RD, 0.55; 95% CI, 0.34-0.70; P < .001). The association between BMI and hypertension was nonsignificant (RD, 0.16; 95% CI, −0.07 to 0.37; P = .13).

We failed to show any statistical significance for the hospital length of stay (Cohen d = 0.042), length of time receiving ventilator assistance (Cohen d = 0.11; P = .65), number of failed extubations (Cohen d = –0.222; P = .43), and the fraction of inspired oxygen before tracheotomy (Cohen d = –0.178; P = .52) (Table 3). The association between BMI and disposition (death or discharge to health care facility or home) failed to reach statistical significance (death vs health care facility, RD, 0.06; 95% CI, −0.17 to 0.31; death vs home, RD, 0.056; 95% CI, −0.32 to 0.41; P = .82).

The 8 logistic regression models testing the association between BMI and tracheotomy status at discharge while controlling for the individual comorbidities resulted in BMI being a significant predictor of tracheotomy status at discharge in 7 models. Controlling for diabetes, super obese patients had a 12-fold increase in the odds of being discharged with a tracheotomy, compared with obese patients (OR, 12.1; 95% CI 2.3-61.8); controlling for hypertension, the OR was 10.8 (95%, CI 2.1-54.6), controlling for congestive heart failure, the OR was 11.9 (95% CI, 2.3-62.5), controlling for heart disease, the OR was 10.1 (95% CI, 1.9-52.7), controlling for pulmonary hypertension, the OR was 9.4 (95% CI, 1.8-47.1), controlling for lung disease, the OR was 13.1 (95% CI, 2.1-79.3), and controlling for chronic renal insufficiency, the OR was 8.9 (95% CI, 1.7-44.9). For the model that included obstructive sleep apnea, the predictive value of BMI was nonsignificant (OR, 4.5; 95% CI, 0.82 to 24.4).

Quiz Ref IDPatients’ primary diagnoses for hospitalization were separated into neurologic, infectious, trauma, or other to allow comparison with other literature.⁹ For the super obese group, neurologic diagnoses were the primary diagnoses for hospitalization in 4 patients (13%), infection in 7 patients (23%), trauma in 2 patients (7%), and other for the remaining 18 patients (Table 1). For the obese group, neurologic diagnoses were the primary diagnoses for hospitalization in 18 patients (53%), followed by trauma in 6 patients (18%), and other in 10 patients (29%).

Otolaryngology was the service most likely to perform the tracheotomy in super obese patients (20 [65%]). General surgery was the next most common service in super obese patients (9 [29%]), with intensivists performing 2 tracheotomies (7%) in super obese patients. Intensivists performed 16 tracheotomies (47%) in the control group, followed by otolaryngology (10 [30%]) and general surgery (8 [24%]).

Nine of the tracheotomies in the super obese group (29%) were performed via a bedside percutaneous dilation tracheotomy, compared with 32 tracheotomies (94%) in the obese group. Ten super obese patients (32%) underwent cervical defatting and 9 super obese patients (29%) had a tracheal margin sewn to the skin or Bjork flap as part of the procedure.

The most common style of tracheotomy tube placed in super obese patients was an extended length tracheotomy tube (16 [52%]), with the remainder receiving either cuffed or uncuffed traditional length tracheotomy tubes. Two patients in the obese group (6%) required an extended length tracheotomy tube.

Quiz Ref IDThere was a very low rate of reported complications, with only 2 super obese patients losing more than 20 mL of blood during the procedure. Postoperative tracheitis was the most common complication, occurring in 3 super obese patients (10%). Accidental decannulation occurred in 1 super obese patient (3%) and minor postoperative hemorrhage requiring nonsurgical intervention occurred in 2 super obese patients. The overall complication rate for super obese patients was 19% (6 patients). The obese group maintained similar low rates of reported complications, with 4 patients losing more than 20 mL of blood during the procedure. Postoperative tracheitis and accidental decannulation did not occur in any obese patients. Minor postoperative hemorrhage requiring surgical intervention occurred in 2 obese patients.

The super obese population had a high rate of medical comorbidities, including 19 patients (61%) with diabetes, 26 (84%) with hypertension, 9 (29%) with congestive heart failure, 6 (19%) with pulmonary hypertension, 14 (45%) with lung disease, 8 (26%) with chronic renal insufficiency, and 18 (58%) with obstructive sleep apnea (Table 2). There was a statistically significantly difference in most of these comorbidities when compared with the obese group. Patients in the obese group had a lower rate of diabetes (13 [38%]) and very few patients had pulmonary hypertension, congestive heart failure, chronic renal insufficiency, or obstructive sleep apnea.

Quiz Ref IDThe most common outcome was tracheotomy dependence, in 25 super obese patients (81%). Ventilator dependence at discharge was also relatively common, seen in 7 super obese patients (23%). Four super obese patients (13%) died while hospitalized and 17 (55%) were transferred to a health care facility. Only 2 super obese patients (7%) were decannulated before discharge. Patients in the obese group were less likely to be tracheotomy dependent on discharge, with 16 (47%) being dependent and only 4 (12%) being ventilator dependent at discharge. Most patients in the obese group (22 [65%]) were discharged to a health care facility, 3 (9%) died while hospitalized, and 15 (44%) were decannulated before discharge. We failed to show any statistically significant difference between the super obese and obese groups when considering the discharge location.

Our series of patients experienced a very high rate of tracheotomy dependence and very low rates of decannulation. The 81% rate of tracheotomy dependence among super obese patients is higher than in other studies examining the rates of tracheotomy dependence in obese patients. A recent study by Byrd et al⁹ reported a tracheotomy dependence rate of 49.0% in 50 of 102 obese patients undergoing tracheotomy. Our control group of obese patients had a 47% rate of tracheotomy dependence, which is more consistent with the results obtained in the recent study by Byrd et al. Similarly, the decannulation rate of 7% among super obese patients in our study was also significantly lower than in the study by Byrd et al.⁹ We cannot exclude differences in clinical practice compared with other institutions, although this possibility seems unlikely owing to general adherence to the clinical consensus statement at our institution.

This elevated rate of tracheotomy dependence is not surprising given the altered respiratory physiology in super obese patients.⁵ The high level of tracheotomy dependence is likely explained by obesity hypoventilation syndrome, which is defined as the presence of obesity (BMI ≥30), chronic hypercapnia (Paco₂ ≥45 mm Hg), and sleep-disturbed breathing.¹⁴ In the management of patients with obesity hypoventilation syndrome, the use of a tracheotomy is usually reserved for patients with significant upper airway resistance.¹⁵

The rate of in-hospital mortality is similar to, although slightly lower than, that in other studies examining the mortality rate in tracheotomy (2%-35%).^2,8-10,16 In our series, the mortality rate was 13% among super obese patients. The reason for this lower mortality rate may be owing to the relatively young super obese patient population, with a mean age of 50 years.

There was an overall complication rate for super obese patients of 19% (6 patients), which is comparable with complication rates seen in other series of obese and morbidly obese patients undergoing tracheotomy.^2,7,8 Of the 4 super obese patients who died while hospitalized after the tracheotomy, 3 were African American. There was no correlation found between any single medical comorbidity or combination of medical comorbidities and death. None of the 4 deaths could be attributed to complications of the tracheotomy.

Quiz Ref IDThe literature is in agreement that percutaneous dilated tracheotomy is safe and effective when performed on critically ill obese patients.^6,11,17,18 Guinot et al¹¹ identified no difference in complication rates when comparing tracheotomy in obese and nonobese patients. Our study helped to confirm that these previously reported data on obese patients parallel those of super obese patients.

One somewhat unexpected finding in this case series is the high rate of discharge to other health care facilities. The low rate of discharge home reflects the enormous complexity of caring for an obese patient who is dependent on a tracheostomy. Owing to lack of follow-up, it is not clear for how long many of these patients remained in these health care facilities.

Limitations of this study include the small sample size, which is difficult to overcome in a single-institution experience on a relatively rare, although increasingly common, disease process. Multi-institutional data would improve the sample size. The control group helps to correlate the indication that super obese patients are more likely to be tracheotomy dependent on discharge. The lack of long-term follow-up is another weakness. Long-term follow-up would provide more information for better counseling of patients on long-term rates of tracheostomy and ventilator dependence.

This study serves as a starting point for further investigation into the long-term outcomes of super obese patients. Appropriate preoperative counseling is vital for super obese patients with high rates of tracheostomy dependence, ventilator dependence, and discharge to other health care facilities. Further research is necessary to determine long-term outcomes of super obese patients undergoing tracheotomy.

Outcomes have not previously been reported, to our knowledge, for super obese patients as a specific population after tracheotomy. Overall, this series demonstrates that tracheotomy is a safe procedure for airway management of super obese individuals, with the caveat that these patients will need extra care after discharge. Appropriate counseling needs to be conducted with the patients and their families about the low rates of decannulation and high rates of discharge to other medical institutions such as skilled nursing facilities. Factors most likely contributing to these findings include diabetes, chronic heart disease, pulmonary hypertension, lung disease, obstructive sleep apnea, and chronic renal insufficiency seen in super obese patients.

Accepted for Publication: April 7, 2016.

Corresponding Author: Ryan V. Marshall, MS, School of Medicine, University of Mississippi Medical Center, 2500 N State St, Jackson, MS 39216 (rvmarsha@gmail.com).

Published Online: May 26, 2016. doi:10.1001/jamaoto.2016.1089.

Author Contributions: Mr Marshall and Dr Haas had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Marshall, Haas, Schweinfurth.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Marshall, Haas.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Marshall, Haas, Replogle.

Obtained funding: Schweinfurth.

Administrative, technical, or material support: Marshall, Haas, Schweinfurth.

Study supervision: Haas, Schweinfurth.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.

Additional Contributions: Laura House, MD, University of Mississippi Medical Center, assisted with data collection. She was not compensated for her contribution.