- Email: [email protected]
Cost analysis of intubation-related tracheal injury using a national database
Otolaryngology–Head and Neck Surgery (2010) 143, 31-36
ORIGINAL RESEARCH–GENERAL OTOLARYNGOLOGY
Cost analysis of intubation-related tracheal injury using a national database Nasir I. Bhatti, MD, Atta Mohyuddin, MD, Nancy Reaven, MA, Susan E. Funk, MBA, Kulsoom Laeeq, MD, Vinciya Pandian, MSN, CRNP, Marek Mirski, MD, and David Feller-Kopman, MD, Baltimore, MD; and La Canada, CA Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article. ABSTRACT OBJECTIVE: To perform risk analysis of tracheal injuries caused by endotracheal intubation (ETI) and to estimate the financial impact of these sequelae. STUDY DESIGN: Cost analysis using a national database. SETTINGS: The Agency for Healthcare Research and Quality (AHRQ) 2006 National Inpatient Sample. SUBJECTS AND METHODS: We identified clinical manifestations and treatments of complications associated with endotracheal tubes and codified them into International Classification of Disease-ninth revision diagnosis and procedure codes, intentionally excluding alternative etiologies of tracheal injury. Using the AHRQ 2006 National Inpatient Sample, we then compared patients with tracheal injury coded during the medical or surgical stay for length of stay (LOS) and mean hospital cost with diagnosis-related group (DRG)-matched controls; we also examined readmissions treating tracheal injury. RESULTS: Tracheal injury presents as tracheal stenosis, tracheomalacia, tracheoesophageal fistula, laryngotracheal ulceration, and vocal cord paralysis. A total of 3232 discharge records met criteria for tracheal injury from ETI within the index hospital stay. Average LOS for patients with tracheal injury (6.3 days; 95% confidence interval [CI] 6.0-6.3) exceeded LOS in the uncomplicated sample (5.2 days; CI 5.1-5.3) by 1.1 days. The average hospital cost was $1888 higher with tracheal injury ($10,375 [CI $9762-$10,988] vs $8487 [CI $8266-$8669]). LOS for procedures treating prior tracheal injury averaged 4.7 days and cost an average of $11,025 per discharge. CONCLUSION: Tracheal injury from ETI is associated with a significant increase in healthcare costs that accrue both during the index admission and during subsequent hospitalizations required to treat the injury. © 2010 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved.
E
ndotracheal intubation (ETI) is a common procedure carried out in both emergent and nonemergent clinical scenarios. Although necessary in many circumstances, the placement of an endotracheal tube can be a source of morbidity and mortality. The incidence of airway injury following ETI is estimated to occur with a frequency of one to 11 percent and can occur anywhere from the teeth/lips to the distal trachea.1 Even with short-term intubation, injury to the posterior commissure and subglottis is quite common.2 Whited3 prospectively studied 200 patients and reported an incidence of tracheal stenosis of four percent for intubations lasting five to 10 days and up to 12 percent for intubations lasting 11 to 24 days. The local risk factors associated with tracheal injuries following ETI include anatomical features of the upper airway, trauma, number of intubation attempts, size and mobility of the endotracheal tube (ETT), presence of a balloon cuff, and duration of ETI. Of these factors, duration of ETI was the most important factor for developing posttracheal injuries.4,5 Airway injuries can present acutely either at the time of intubation, while the ETT is in place, or sometime after the intubation. Although some of these injuries are minor and self-limited, others require treatment, which may occur either during the associated hospitalization or sometime thereafter. The cost of treatment for ETI-associated tracheal injury has not previously been examined. The purpose of this study was, first, to identify postintubation tracheal injury and its associated impact on length of stay (LOS) and cost to the index hospital admission and, second, to examine costs related to subsequent readmissions required to treat the injury.
Methods This study was exempt from institutional review board review.
Received September 13, 2009; revised October 22, 2009; accepted November 3, 2009.
0194-5998/$36.00 © 2010 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved. doi:10.1016/j.otohns.2009.11.004
32
Otolaryngology–Head and Neck Surgery, Vol 143, No 1, July 2010
Analysis of Tracheal Injury
Table 1 Tracheal injury codes: ICD-9 ICD code
Description
478.30 478.31 478.32 478.33 478.34 478.79 519.1 519.11 519.19 530.84
Paralysis, vocal cord/larynx NOS Paralysis, vocal cord unilateral/partial Paralysis, vocal cord unilateral/total Paralysis, vocal cord bilateral/partial Paralysis, vocal cord bilateral/total Disease, larynx NEC Disease, trachea/bronchus NEC Acute bronchospasm Trachea & bronchus NEC Fistula, tracheoesophageal
ICD-9, International Classification of Diseases-ninth revision; NOS, not otherwise specified; NEC, not elsewhere classified.
Selection of ICD-9 Diagnosis We performed a search of Medline, Cochrane, International Society for Pharmacoeconomics and Outcome Research, and the Royal Society of Medicine to investigate the incidence of complications, treatment approaches, and medical resource utilization associated with complications related to ETI including tracheal stenosis, tracheomalacia, tracheal perforation, and tracheoesophageal fistula. The literature search was performed iteratively, with the use of basic keywords: [endotracheal tube OR endotracheal intubation] AND [tracheomalacia OR tracheal stenosis OR tracheal perforation OR tracheoesophageal fistula OR tracheal dilation OR tracheal injury]. In preparation for the data analysis, we linked the narrative descriptions of complications and treatments as documented in the literature to the most closely aligned ICD-9 diagnosis (International Classification of Diseases-ninth revision, diagnosis and procedural codes, World Health Organization) (Tables 1 and 2).
Hospital Cost and LOS Data on hospital LOS, charges, and costs were obtained from the 2006 National Inpatient Sample (NIS), Healthcare Cost and Utilization Project (HCUP) managed by the Agency for Healthcare Research and Quality (AHRQ).6 AHRQ calculates LOS using the discharge date minus admission date, thus allowing zero-day hospital inpatient stays. All costs, charges, and LOS results reported in this document were developed by using AHRQ’s discharge weights to ensure that the results are as representative as possible. Hospital cost was estimated from hospital charges by using hospital-specific or group-average all-payer inpatient cost-to-charge ratios calculated by AHRQ. Thus, throughout this report, the “hospital cost” of discharges with or without complications represents the full cost to the hospital of providing the services, irrespective of how much reimbursement (if any) the hospital received.
Our analysis of patient-level data for tracheal injury was designed to yield data on two sets of complication-related events: first, incremental costs and LOS associated with tracheal injury identified during an index hospital admission for a nontracheal reason; second, the costs of subsequent hospital admissions required for repair of the tracheal injury occurring previously.
Injury Occurring During an Index Hospital Admission for another Reason Our approach was to identify a base population of admitted patients who experienced a tracheal complication and to compare their hospital LOS and costs with a matched population of patients undergoing similar medical or surgical admissions, but experiencing no tracheal injury. Inclusion criteria were age ⱖ 18 years and tracheal injury diagnosis code in any position other than primary diagnosis in the hospital record. We used the tracheal injury diagnostic codes as shown in Table 1. We excluded patients with diagnosis and procedural codes that would act as confounders for tracheal injury not related to ETI. These include codes for neoplasms of the larynx, trachea, and lung; codes for accidents and trauma to the airway; and codes denoting a history of tracheostomy; likewise excluded were patients having a primary diagnosis of a tracheal injury or having an implant procedure code and another procedure that might have caused tracheal injury, such as a bronchoscopy with stenting, thermal ablation, or dilation. Following the inclusion and exclusion criteria shown in Table 3, we identified the diagnosis-related group (DRG) associated with each admission and the number of patients meeting inclusion/exclusion criteria in each DRG. Our analysis of cost and LOS for patients with tracheal injury was based on a net population of 3232 patients in the 73 DRGs in which there was a minimum of 10 patients for which we were able to document a tracheal injury with complications. The minimum of 10 patients was instituted, in part, to comply with patient privacy mandates covering federally maintained databases. Table 2 Tracheal injury repair procedure codes ICD-9 description ICD code
Description
31.5 31.73 31.75 31.79 31.93 31.98 31.99 96.05
Local destruction tracheal lesion Tracheal fistula closure NEC Tracheal reconstruction Tracheal repair NEC Trachea/larynx stent replacement Laryngeal operations NEC Tracheal operations NEC Respiratory tract intubation NEC
ICD-9, International Classification of Diseases-ninth revision; NEC, not elsewhere classified.
Bhatti et al
Cost analysis of intubation-related tracheal injury . . .
33
Table 3 Specific inclusion and exclusion criteria for tracheal injury diagnosis Inclusion criteria ● ●
Age ⱖ18 yrs Tracheal Damage Diagnosis code (Table 1) in any position other than primary diagnosis
Exclusion criteria From the base of adults with a qualifying diagnosis code, we excluded individuals meeting any of the following criteria: ● Diagnosis code for tracheostomy complication (519.0) or history of tracheostomy (V44.0 or V55.0) in any position in the discharge record ● Primary diagnosis in the discharge record that is a tracheal damage diagnosis code (we examine these separately in our analysis of readmissions for tracheal damage repair) ● Tracheostomy procedure code (31.1 or 31.2, 31.21 or 31.29) in any position in the discharge record ● Implant procedure code (the cost of implants adds significant variability to cost comparisons and makes cost estimates unreliable) ● Cancer as a primary diagnosis (CPT codes between 140 and 208.91 in the first diagnosis position in the discharge record) ● Diagnosis code indicating cancer, cancer in situ, or benign neoplasm of the throat, lungs, or other nearby anatomical sites, in any position in the discharge record ● Tracheal injury diagnosis code (874) in any position in the discharge record ● Other procedure that might have caused tracheal damage, such as a bronchoscopy, bronchial dilation, gastroscopy, or esophagoscopy ● Any procedure code we determined was most likely to be done in a readmission rather than in the initial admission involving the ET tube ● Principal event code (ECode1) described an event that we determined was unlikely to be associated with tracheal damage ● DRG 290, 182, or 92; any of which might indicate an alternative clinical explanation for tracheal symptoms ● Total charges data unavailable or invalid in the discharge record, so hospital cost cannot be calculated CPT, Current Procedural Terminology; ET, endotracheal; DRG, diagnosis-related group. Note: Initial analyses showed that ET tube placement was rarely coded in hospital records (procedure code 96.04). Accordingly, we did not require documentation of the ET tube procedure code, but reviewed the data iteratively from multiple perspectives and crafted an extensive list of criteria aimed at excluding discharges in which there might be an alternative explanation for the conditions we identify as tracheal damage.
Matched Sample of Uncomplicated Discharges For each DRG for which we were able to identify at least 10 patients with complications meeting both our inclusion and exclusion criteria, we selected as the matched sample all in-patient discharges for patients aged ⱖ18 years having the same DRG. For the matched population, we applied the same exclusion criteria, and we also excluded patients with any diagnosis code beginning with 99 (referring to “complications of care”). The resulting matched sample consisted of 2,322,642 patients treated under the same DRGs as our “tracheal complications” sample, but absent any of the tracheal diagnoses, procedures, or complications examined in this study.
Tracheal Injury, Readmission To establish the study population of patients requiring readmission for treatment of previously occurring ETI-related tracheal injury, we queried the NIS for hospital admissions in which procedure codes specified that a tracheal repair procedure (Table 2) was documented during an admission
in conjunction with one of the diagnosis codes specifying tracheal injury (Table 1). There were 263 discharge records meeting these basic criteria. After excluding patients who had a tracheostomy procedure code; diagnosis codes specific to complications of tracheostomy, history of tracheostomy, or attention to tracheostomy; a mechanical ventilation code; or no cost data in the hospital record, a final population of 200 patients was available for this analysis.
Results Index Admission Tracheal Injury We were able to match hospital LOS and cost data for 73 medical and surgical DRGs. The weighted average results of the comparison between average cost and LOS between patients experiencing complications of tracheal injury and those not experiencing tracheal injury are listed in Table 4. Having tracheal injury as a complication of ETI adds an average of just over one day to the average LOS (6.3 days; 95% CI 6.0-6.6 vs 5.2 days; 95% CI 5.1-5.3) and almost
34
Otolaryngology–Head and Neck Surgery, Vol 143, No 1, July 2010
Table 4 Tracheal damage: Impact on hospital length of stay and costs for the initial admission for diagnosis-related groups with >10 eligible patients with complications
ALOS and cost Weighted average
Uncomplicated index cases
Index cases with tracheal injury
Differences (complicateduncomplicated)
ALOS
ALOS
ALOS Avg cost
Avg cost
Avg cost
5.2 $8487 6.3 $10,375 (95% CI 5.1-5.3 d) (95% CI $8266-$8669) (95% CI 6.1-6.3 d) (95% CI $9762-($10,988)
1.1
$1888
ALOS, average length of stay; Avg, average; CI, confidence interval. Source: Reference 6. (Hospital cost is estimated from total charges by using hospital-specific or group average all-payer inpatient cost-to-charge ratios provided by the publisher.)
and colleagues12 performed tracheostomy for severe PITS that prevented extubation in 1.8 percent (142/7600 patients). In our analysis, we searched patient records for the most commonly cited tracheal complications and procedures, compared hospital costs and LOS for patients with and without documented tracheal complications, and examined the costs of readmission for procedures to treat the tracheal injury. Our analysis suggests that the initial tracheal injury is associated with an increased hospital LOS of 1.1 days (6.3 days vs 5.2 days) and an added cost of $1888 ($10,375 vs $8487), whereas readmissions to treat tracheal injury have an LOS averaging 4.7 days and average costs of $11,025. We excluded patients with tracheostomies from this analysis as it would be difficult to assign causation to the endotracheal tube versus the tracheostomy tube in a patient with tracheal injury such as tracheal stenosis or malacia. Accordingly, our criteria are likely to have excluded some of the most costly patients requiring repair of tracheal injury caused by ETI. A large number of studies, going back many decades, directly associate the development of tracheal stenosis to high cuff pressures and link the decrease in tracheal injury since the 1970s to the development of lower-pressure cuffs. Unfortunately, low-pressure cuffs can be overinflated, and
$2000 in hospital costs ($10,375; CI $9762-$10,988 vs $8487; 95% CI $8266-$8669).
Readmissions to Repair Tracheal Injury For our patient population of 200 who were readmitted for treatment of a previously occurring ETI-related tracheal injury, hospital costs averaged approximately $11,025 per admission and hospital LOS averaged 4.7 days (Table 5). Typical procedures performed during the readmissions included bronchoscopic treatment such as laser, electrocautery, balloon dilation, and stenting, as well as surgical resection and reconstruction (Table 2).
Discussion Data examining the incidence and costs associated with post-intubation tracheal injury are sparse. Airway injury can start within a few hours after intubation.7,8 In a prospective study by Kastanos’ group,9 the incidence of post-intubation tracheal stenosis (PITS) was 10 percent. Schneider et al10 cites incidence rates of tracheal perforations from 0.005 percent for single lumen endotracheal tubes to 0.19 percent for double lumen tubes. Grillo et al11 reported an incidence of tracheal injuries in 20 (3.8%) patients in a group of 521 patients intubated for surgical procedures. Likewise, Deeb
Table 5 Readmissions to treat previous tracheal injuries Diagnosis category of index cases
Weighted average age (yrs)
% Discharges
Weighted average LOS (d)
Weighted average charges (U.S.$)
Weighted average cost (U.S.$)
Tracheal stenosis, tracheomalacia Tracheoesophageal fistula All other conditions* All readmissions
50 61 56 52
81.4 11.7 6.9 100
4.4 8.1 2.4 4.7
30,496 54,633 16,139 32,335
10,016 20,905 6161 11,025
LOS, length of stay. *Category includes vocal cord paralysis, bronchospasm, and laryngotracheal ulceration.
Bhatti et al
Cost analysis of intubation-related tracheal injury . . .
cuff-related tracheal injury remains an ongoing problem.13-15 Other risk factors for PITS include female sex (75%), obesity (66%), a history of diabetes mellitus (35.4%), hypertension (51.6%), cardiovascular disease (45.1%), and being a current smoker (38.7%).16 It does not appear that age is associated with the development of postintubation tracheal injury. In our study, the average age is nearly the same among patients with and those without tracheal injury. Our study has some limitations. First, this is a review of numbers extrapolated from national databases. In light of the low incidence of tracheal injury associated with ETI, performing a large, prospective study would be quite difficult. The ideal study would prospectively follow all patients undergoing elective and emergent intubation as well as document several variables including patient height and weight, endotracheal tube size, Mallampati classification, type and number of airway interventions, and so forth. The investigators would also need to prospectively follow these patients with airway CT scans or bronchoscopy or both, whether they were having symptoms or not. Second, because we were not able to verify the use of ETT in every admission, it is possible that some patients with tracheal injury due to non-ETT factors, such as complications of thyroidectomy, were included in our study population, despite our stringent exclusion criteria. Third, many complications associated with ETI are not apparent until the patient has been extubated and has left the hospital. Additionally, patients often do not develop symptoms of tracheal stenosis, such as exertional dyspnea, until the tracheal lumen is approximately 8 mm in diameter, and this may take months to years to develop.17 It is therefore probable that our study underestimates the cost associated with readmission because we are examining readmissions during a 12-month cross-sectional timeframe rather than in a longitudinal study. Finally, the management of patients with tracheal injury is best performed in a multidisciplinary approach.18 Many tracheal stenosis cases require multiple minimally invasive procedures, and some patients are ultimately referred for surgical resections.19-21 Because our study examined only the costs of inpatient readmissions, the costs of these multiple outpatient interventions are not accounted for. Similarly, the indirect costs associated with time off from work and other such economic issues are also not addressed in our study. Despite the limitations of our study, it is thus far the largest investigation of post-intubation tracheal injury. To our knowledge, this is the only study examining hospital cost and LOS following tracheal injures due to ETI. It is important to note that in the current regulatory environment, hospitals do not receive payment for treatment of certain complications, called “never events,” such as catheter-related blood stream infections. In the future, it is possible that hospitals will not be reimbursed for the additional costs associated with post-intubation tracheal injury. Although some reimbursement is generally expected for
35
separate hospital admissions, Medicare and other payers are increasingly reluctant to pay extra for care needed to treat medical errors and adverse results of hospital care.
Conclusion The financial impact of post-intubation tracheal injury is significant, adding to the index admission cost and LOS as well as to future costs for treatment and repair of the injury. Prospective studies are required to obtain further data about indirect costs, as well as the costs associated with outpatient management of these patients.
Author Information From the Departments of Otolaryngology–Head and Neck Surgery (Drs. Bhatti, Mohyuddin, Laeeq, and Pandian), Anesthesiology and Critical Care Medicine (Dr. Mirski), and Interventional Pulmonology (Dr. Feller-Kopman), Johns Hopkins Hospital, Baltimore; and Strategic Health Resources (Drs. Reaven and Funk), La Canada. Corresponding author: Nasir I. Bhatti, MD, 601 North Caroline St, Ste 6241, Baltimore, MD 21287-0910. E-mail address: [email protected]. This article was presented at the 2009 AAO–HNSF Annual Meeting & OTO EXPO, San Diego, CA, October 4-7, 2009.
Author Contributions Nasir I. Bhatti, substantial contributions to concept and design, drafting of the article or revising it critically, final approval of version to be published; Atta Mohyuddin, drafting of the introduction and methods sections, literature review for ICD-9 diagnosis, final approval of version to be published; Nancy Reaven, assistance with formulating ICD-9 diagnosis and data analysis, final approval of version to be published; Susan E. Funk, contributions to drafting and revision of article, final approval of version to be published; Kulsoom Laeeq, statistical analysis and interpretation of data; final approval of version to be published; Vinciya Pandian, proof reading, editing, and contribution to revision of manuscript, final approval of version to be published; Marek Mirski, assistance with formulating a plan for methodology and results, final approval of version to be published; David Feller-Kopmann, original idea, substantial contributions to study concept and design and interpretations of data, drafting and critical revision of article, final approval of version to be published.
Disclosures Competing interests: Nancy Reaven, Covidien Healthcare: provided grant for research (no support for article); Susan E. Funk, Covidien Healthcare: provided grant for research (none for article). Sponsorships: Study was partially funded by Covidien Healthcare under an agreement that the study be conducted independently to reduce funding bias. Accordingly, the study was designed, performed, analyzed, interpreted, and written by the investigators without the involvement, review, or prepublication approval of Covidien.
References 1. Koshkareva Y, Gaughan JP, Soliman AM. Risk factors for adult laryngotracheal stenosis: a review of 74 cases. Ann Otol Rhinol Laryngol 2007;116:206 –10. 2. Hawkins DB. Glottic and subglottic stenosis from endotracheal intubation. Laryngoscope 1977;87:339 – 41.
36
Otolaryngology–Head and Neck Surgery, Vol 143, No 1, July 2010
3. Whited RE. A prospective study of laryngotracheal sequelae in longterm intubation. Laryngoscope 1984;94:367–77. 4. Gaynor EB, Greenberg SB. Untoward sequelae of prolonged intubation. Laryngoscope 1985;95:1461–7. 5. Stauffer JL, Oslen DE, Petty TL. Complications and consequences of endotracheal intubation and tracheostomy: a prospective study of 150 critically ill adult patients. Am J Med 1981;70:65–76. 6. HCUP Nationwide Inpatient Sample (NIS). Health Cost and Utilization Project (HCUP). 2006. Rockville (MD): Agency for Healthcare Research and Quality. Available at: http://www.hcup-us.ahrq.gov/ nisoverview.jsp. Accessed July 5, 2008. 7. Donnelly WH. Histopathology of endotracheal intubation: an autopsy study of 99 cases. Arch Pathol 1969;88:511–20. 8. Yang KL. Tracheal stenosis after a brief intubation. Anesth Analg 1995;80:625–7. 9. Kastanos N, Estopa MR, Perez M, et al. Laryngotracheal injury due to endotracheal intubation: incidence, evolution and predisposing factors. A prospective long term study. Crit Care Med 1983;11:362–7. 10. Schneider T, Storz K, Dienmann H, et al. Management of iatrogenic tracheobronchial injuries: a retrospective analysis of 29 cases. Ann Thorac Surg 2007;83:1960 – 4. 11. Grillo HC, Donahue HD, Mathisen JD, et al. Post intubation tracheal stenosis: treatment and results. J Thorac Cardiovasc Surg 1995;109: 486 –93. 12. Deeb ZE, Williams JB, Campbell TE. Early diagnosis and treatment of laryngeal injuries from prolonged intubation in adults. Otolaryngol Head and Neck Surg 1990;120:25–9.
13. Grillo HC, Cooper JD, Geffin B, et al. A low pressure cuff for tracheostomy tubes to minimize tracheal injury: a comparative clinical trial. J Thorac Cardiovasc Surg 1971;62:898 –907. 14. Cooper JD, Grillo HC. Experimental production and prevention of injury due to cuffed tracheal tubes. Surg Gynecol Obstet 1969;129: 1235– 41. 15. Seegobin RD, Van Hasselt GL. Endotracheal cuff pressure and tracheal mucosal blood flow: endoscopic study of effects of four large volume cuffs. Br Med J 1984;288:965– 8. 16. Zias N, Chroneou A, Tabba MK, et al. Post tracheostomy and post intubation tracheal stenosis: report of 31 cases and review of the literature. BMC Pulm Med 2008;8:18. 17. Hollingsworth HM. Wheezing and stridor. Clin Chest Med 1987;8: 231– 40. 18. Brichet A, Verkindre C, Dupont J, et al. Multidisciplinary approach to management of postintubation tracheal stenosis. Eur Resp J 1999;13: 888 –93. 19. Cavaliere S, Bezzi M, Toninelli C, et al. Management of post-intubation tracheal stenosis using the endoscopic approach. Monaldi Arch Chest Dis 2007;67:73– 80. 20. Macchiarini P, Verhoye JP, Chapelier A, et al. Partial cricoidectomy with primary thyrotracheal anastomosis for postintubation subglottic stenosis. J Thorac Cardiovasc Surg 2001;121:68 –76. 21. Sittel C, Blum S, Streckfuss A, et al. Cricotracheal resection in nontracheotomized adults: a prospective case series. Ann Otol Rhinol Laryngol 2008;117:288 –94.
ORIGINAL RESEARCH–GENERAL OTOLARYNGOLOGY
Cost analysis of intubation-related tracheal injury using a national database Nasir I. Bhatti, MD, Atta Mohyuddin, MD, Nancy Reaven, MA, Susan E. Funk, MBA, Kulsoom Laeeq, MD, Vinciya Pandian, MSN, CRNP, Marek Mirski, MD, and David Feller-Kopman, MD, Baltimore, MD; and La Canada, CA Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article. ABSTRACT OBJECTIVE: To perform risk analysis of tracheal injuries caused by endotracheal intubation (ETI) and to estimate the financial impact of these sequelae. STUDY DESIGN: Cost analysis using a national database. SETTINGS: The Agency for Healthcare Research and Quality (AHRQ) 2006 National Inpatient Sample. SUBJECTS AND METHODS: We identified clinical manifestations and treatments of complications associated with endotracheal tubes and codified them into International Classification of Disease-ninth revision diagnosis and procedure codes, intentionally excluding alternative etiologies of tracheal injury. Using the AHRQ 2006 National Inpatient Sample, we then compared patients with tracheal injury coded during the medical or surgical stay for length of stay (LOS) and mean hospital cost with diagnosis-related group (DRG)-matched controls; we also examined readmissions treating tracheal injury. RESULTS: Tracheal injury presents as tracheal stenosis, tracheomalacia, tracheoesophageal fistula, laryngotracheal ulceration, and vocal cord paralysis. A total of 3232 discharge records met criteria for tracheal injury from ETI within the index hospital stay. Average LOS for patients with tracheal injury (6.3 days; 95% confidence interval [CI] 6.0-6.3) exceeded LOS in the uncomplicated sample (5.2 days; CI 5.1-5.3) by 1.1 days. The average hospital cost was $1888 higher with tracheal injury ($10,375 [CI $9762-$10,988] vs $8487 [CI $8266-$8669]). LOS for procedures treating prior tracheal injury averaged 4.7 days and cost an average of $11,025 per discharge. CONCLUSION: Tracheal injury from ETI is associated with a significant increase in healthcare costs that accrue both during the index admission and during subsequent hospitalizations required to treat the injury. © 2010 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved.
E
ndotracheal intubation (ETI) is a common procedure carried out in both emergent and nonemergent clinical scenarios. Although necessary in many circumstances, the placement of an endotracheal tube can be a source of morbidity and mortality. The incidence of airway injury following ETI is estimated to occur with a frequency of one to 11 percent and can occur anywhere from the teeth/lips to the distal trachea.1 Even with short-term intubation, injury to the posterior commissure and subglottis is quite common.2 Whited3 prospectively studied 200 patients and reported an incidence of tracheal stenosis of four percent for intubations lasting five to 10 days and up to 12 percent for intubations lasting 11 to 24 days. The local risk factors associated with tracheal injuries following ETI include anatomical features of the upper airway, trauma, number of intubation attempts, size and mobility of the endotracheal tube (ETT), presence of a balloon cuff, and duration of ETI. Of these factors, duration of ETI was the most important factor for developing posttracheal injuries.4,5 Airway injuries can present acutely either at the time of intubation, while the ETT is in place, or sometime after the intubation. Although some of these injuries are minor and self-limited, others require treatment, which may occur either during the associated hospitalization or sometime thereafter. The cost of treatment for ETI-associated tracheal injury has not previously been examined. The purpose of this study was, first, to identify postintubation tracheal injury and its associated impact on length of stay (LOS) and cost to the index hospital admission and, second, to examine costs related to subsequent readmissions required to treat the injury.
Methods This study was exempt from institutional review board review.
Received September 13, 2009; revised October 22, 2009; accepted November 3, 2009.
0194-5998/$36.00 © 2010 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved. doi:10.1016/j.otohns.2009.11.004
32
Otolaryngology–Head and Neck Surgery, Vol 143, No 1, July 2010
Analysis of Tracheal Injury
Table 1 Tracheal injury codes: ICD-9 ICD code
Description
478.30 478.31 478.32 478.33 478.34 478.79 519.1 519.11 519.19 530.84
Paralysis, vocal cord/larynx NOS Paralysis, vocal cord unilateral/partial Paralysis, vocal cord unilateral/total Paralysis, vocal cord bilateral/partial Paralysis, vocal cord bilateral/total Disease, larynx NEC Disease, trachea/bronchus NEC Acute bronchospasm Trachea & bronchus NEC Fistula, tracheoesophageal
ICD-9, International Classification of Diseases-ninth revision; NOS, not otherwise specified; NEC, not elsewhere classified.
Selection of ICD-9 Diagnosis We performed a search of Medline, Cochrane, International Society for Pharmacoeconomics and Outcome Research, and the Royal Society of Medicine to investigate the incidence of complications, treatment approaches, and medical resource utilization associated with complications related to ETI including tracheal stenosis, tracheomalacia, tracheal perforation, and tracheoesophageal fistula. The literature search was performed iteratively, with the use of basic keywords: [endotracheal tube OR endotracheal intubation] AND [tracheomalacia OR tracheal stenosis OR tracheal perforation OR tracheoesophageal fistula OR tracheal dilation OR tracheal injury]. In preparation for the data analysis, we linked the narrative descriptions of complications and treatments as documented in the literature to the most closely aligned ICD-9 diagnosis (International Classification of Diseases-ninth revision, diagnosis and procedural codes, World Health Organization) (Tables 1 and 2).
Hospital Cost and LOS Data on hospital LOS, charges, and costs were obtained from the 2006 National Inpatient Sample (NIS), Healthcare Cost and Utilization Project (HCUP) managed by the Agency for Healthcare Research and Quality (AHRQ).6 AHRQ calculates LOS using the discharge date minus admission date, thus allowing zero-day hospital inpatient stays. All costs, charges, and LOS results reported in this document were developed by using AHRQ’s discharge weights to ensure that the results are as representative as possible. Hospital cost was estimated from hospital charges by using hospital-specific or group-average all-payer inpatient cost-to-charge ratios calculated by AHRQ. Thus, throughout this report, the “hospital cost” of discharges with or without complications represents the full cost to the hospital of providing the services, irrespective of how much reimbursement (if any) the hospital received.
Our analysis of patient-level data for tracheal injury was designed to yield data on two sets of complication-related events: first, incremental costs and LOS associated with tracheal injury identified during an index hospital admission for a nontracheal reason; second, the costs of subsequent hospital admissions required for repair of the tracheal injury occurring previously.
Injury Occurring During an Index Hospital Admission for another Reason Our approach was to identify a base population of admitted patients who experienced a tracheal complication and to compare their hospital LOS and costs with a matched population of patients undergoing similar medical or surgical admissions, but experiencing no tracheal injury. Inclusion criteria were age ⱖ 18 years and tracheal injury diagnosis code in any position other than primary diagnosis in the hospital record. We used the tracheal injury diagnostic codes as shown in Table 1. We excluded patients with diagnosis and procedural codes that would act as confounders for tracheal injury not related to ETI. These include codes for neoplasms of the larynx, trachea, and lung; codes for accidents and trauma to the airway; and codes denoting a history of tracheostomy; likewise excluded were patients having a primary diagnosis of a tracheal injury or having an implant procedure code and another procedure that might have caused tracheal injury, such as a bronchoscopy with stenting, thermal ablation, or dilation. Following the inclusion and exclusion criteria shown in Table 3, we identified the diagnosis-related group (DRG) associated with each admission and the number of patients meeting inclusion/exclusion criteria in each DRG. Our analysis of cost and LOS for patients with tracheal injury was based on a net population of 3232 patients in the 73 DRGs in which there was a minimum of 10 patients for which we were able to document a tracheal injury with complications. The minimum of 10 patients was instituted, in part, to comply with patient privacy mandates covering federally maintained databases. Table 2 Tracheal injury repair procedure codes ICD-9 description ICD code
Description
31.5 31.73 31.75 31.79 31.93 31.98 31.99 96.05
Local destruction tracheal lesion Tracheal fistula closure NEC Tracheal reconstruction Tracheal repair NEC Trachea/larynx stent replacement Laryngeal operations NEC Tracheal operations NEC Respiratory tract intubation NEC
ICD-9, International Classification of Diseases-ninth revision; NEC, not elsewhere classified.
Bhatti et al
Cost analysis of intubation-related tracheal injury . . .
33
Table 3 Specific inclusion and exclusion criteria for tracheal injury diagnosis Inclusion criteria ● ●
Age ⱖ18 yrs Tracheal Damage Diagnosis code (Table 1) in any position other than primary diagnosis
Exclusion criteria From the base of adults with a qualifying diagnosis code, we excluded individuals meeting any of the following criteria: ● Diagnosis code for tracheostomy complication (519.0) or history of tracheostomy (V44.0 or V55.0) in any position in the discharge record ● Primary diagnosis in the discharge record that is a tracheal damage diagnosis code (we examine these separately in our analysis of readmissions for tracheal damage repair) ● Tracheostomy procedure code (31.1 or 31.2, 31.21 or 31.29) in any position in the discharge record ● Implant procedure code (the cost of implants adds significant variability to cost comparisons and makes cost estimates unreliable) ● Cancer as a primary diagnosis (CPT codes between 140 and 208.91 in the first diagnosis position in the discharge record) ● Diagnosis code indicating cancer, cancer in situ, or benign neoplasm of the throat, lungs, or other nearby anatomical sites, in any position in the discharge record ● Tracheal injury diagnosis code (874) in any position in the discharge record ● Other procedure that might have caused tracheal damage, such as a bronchoscopy, bronchial dilation, gastroscopy, or esophagoscopy ● Any procedure code we determined was most likely to be done in a readmission rather than in the initial admission involving the ET tube ● Principal event code (ECode1) described an event that we determined was unlikely to be associated with tracheal damage ● DRG 290, 182, or 92; any of which might indicate an alternative clinical explanation for tracheal symptoms ● Total charges data unavailable or invalid in the discharge record, so hospital cost cannot be calculated CPT, Current Procedural Terminology; ET, endotracheal; DRG, diagnosis-related group. Note: Initial analyses showed that ET tube placement was rarely coded in hospital records (procedure code 96.04). Accordingly, we did not require documentation of the ET tube procedure code, but reviewed the data iteratively from multiple perspectives and crafted an extensive list of criteria aimed at excluding discharges in which there might be an alternative explanation for the conditions we identify as tracheal damage.
Matched Sample of Uncomplicated Discharges For each DRG for which we were able to identify at least 10 patients with complications meeting both our inclusion and exclusion criteria, we selected as the matched sample all in-patient discharges for patients aged ⱖ18 years having the same DRG. For the matched population, we applied the same exclusion criteria, and we also excluded patients with any diagnosis code beginning with 99 (referring to “complications of care”). The resulting matched sample consisted of 2,322,642 patients treated under the same DRGs as our “tracheal complications” sample, but absent any of the tracheal diagnoses, procedures, or complications examined in this study.
Tracheal Injury, Readmission To establish the study population of patients requiring readmission for treatment of previously occurring ETI-related tracheal injury, we queried the NIS for hospital admissions in which procedure codes specified that a tracheal repair procedure (Table 2) was documented during an admission
in conjunction with one of the diagnosis codes specifying tracheal injury (Table 1). There were 263 discharge records meeting these basic criteria. After excluding patients who had a tracheostomy procedure code; diagnosis codes specific to complications of tracheostomy, history of tracheostomy, or attention to tracheostomy; a mechanical ventilation code; or no cost data in the hospital record, a final population of 200 patients was available for this analysis.
Results Index Admission Tracheal Injury We were able to match hospital LOS and cost data for 73 medical and surgical DRGs. The weighted average results of the comparison between average cost and LOS between patients experiencing complications of tracheal injury and those not experiencing tracheal injury are listed in Table 4. Having tracheal injury as a complication of ETI adds an average of just over one day to the average LOS (6.3 days; 95% CI 6.0-6.6 vs 5.2 days; 95% CI 5.1-5.3) and almost
34
Otolaryngology–Head and Neck Surgery, Vol 143, No 1, July 2010
Table 4 Tracheal damage: Impact on hospital length of stay and costs for the initial admission for diagnosis-related groups with >10 eligible patients with complications
ALOS and cost Weighted average
Uncomplicated index cases
Index cases with tracheal injury
Differences (complicateduncomplicated)
ALOS
ALOS
ALOS Avg cost
Avg cost
Avg cost
5.2 $8487 6.3 $10,375 (95% CI 5.1-5.3 d) (95% CI $8266-$8669) (95% CI 6.1-6.3 d) (95% CI $9762-($10,988)
1.1
$1888
ALOS, average length of stay; Avg, average; CI, confidence interval. Source: Reference 6. (Hospital cost is estimated from total charges by using hospital-specific or group average all-payer inpatient cost-to-charge ratios provided by the publisher.)
and colleagues12 performed tracheostomy for severe PITS that prevented extubation in 1.8 percent (142/7600 patients). In our analysis, we searched patient records for the most commonly cited tracheal complications and procedures, compared hospital costs and LOS for patients with and without documented tracheal complications, and examined the costs of readmission for procedures to treat the tracheal injury. Our analysis suggests that the initial tracheal injury is associated with an increased hospital LOS of 1.1 days (6.3 days vs 5.2 days) and an added cost of $1888 ($10,375 vs $8487), whereas readmissions to treat tracheal injury have an LOS averaging 4.7 days and average costs of $11,025. We excluded patients with tracheostomies from this analysis as it would be difficult to assign causation to the endotracheal tube versus the tracheostomy tube in a patient with tracheal injury such as tracheal stenosis or malacia. Accordingly, our criteria are likely to have excluded some of the most costly patients requiring repair of tracheal injury caused by ETI. A large number of studies, going back many decades, directly associate the development of tracheal stenosis to high cuff pressures and link the decrease in tracheal injury since the 1970s to the development of lower-pressure cuffs. Unfortunately, low-pressure cuffs can be overinflated, and
$2000 in hospital costs ($10,375; CI $9762-$10,988 vs $8487; 95% CI $8266-$8669).
Readmissions to Repair Tracheal Injury For our patient population of 200 who were readmitted for treatment of a previously occurring ETI-related tracheal injury, hospital costs averaged approximately $11,025 per admission and hospital LOS averaged 4.7 days (Table 5). Typical procedures performed during the readmissions included bronchoscopic treatment such as laser, electrocautery, balloon dilation, and stenting, as well as surgical resection and reconstruction (Table 2).
Discussion Data examining the incidence and costs associated with post-intubation tracheal injury are sparse. Airway injury can start within a few hours after intubation.7,8 In a prospective study by Kastanos’ group,9 the incidence of post-intubation tracheal stenosis (PITS) was 10 percent. Schneider et al10 cites incidence rates of tracheal perforations from 0.005 percent for single lumen endotracheal tubes to 0.19 percent for double lumen tubes. Grillo et al11 reported an incidence of tracheal injuries in 20 (3.8%) patients in a group of 521 patients intubated for surgical procedures. Likewise, Deeb
Table 5 Readmissions to treat previous tracheal injuries Diagnosis category of index cases
Weighted average age (yrs)
% Discharges
Weighted average LOS (d)
Weighted average charges (U.S.$)
Weighted average cost (U.S.$)
Tracheal stenosis, tracheomalacia Tracheoesophageal fistula All other conditions* All readmissions
50 61 56 52
81.4 11.7 6.9 100
4.4 8.1 2.4 4.7
30,496 54,633 16,139 32,335
10,016 20,905 6161 11,025
LOS, length of stay. *Category includes vocal cord paralysis, bronchospasm, and laryngotracheal ulceration.
Bhatti et al
Cost analysis of intubation-related tracheal injury . . .
cuff-related tracheal injury remains an ongoing problem.13-15 Other risk factors for PITS include female sex (75%), obesity (66%), a history of diabetes mellitus (35.4%), hypertension (51.6%), cardiovascular disease (45.1%), and being a current smoker (38.7%).16 It does not appear that age is associated with the development of postintubation tracheal injury. In our study, the average age is nearly the same among patients with and those without tracheal injury. Our study has some limitations. First, this is a review of numbers extrapolated from national databases. In light of the low incidence of tracheal injury associated with ETI, performing a large, prospective study would be quite difficult. The ideal study would prospectively follow all patients undergoing elective and emergent intubation as well as document several variables including patient height and weight, endotracheal tube size, Mallampati classification, type and number of airway interventions, and so forth. The investigators would also need to prospectively follow these patients with airway CT scans or bronchoscopy or both, whether they were having symptoms or not. Second, because we were not able to verify the use of ETT in every admission, it is possible that some patients with tracheal injury due to non-ETT factors, such as complications of thyroidectomy, were included in our study population, despite our stringent exclusion criteria. Third, many complications associated with ETI are not apparent until the patient has been extubated and has left the hospital. Additionally, patients often do not develop symptoms of tracheal stenosis, such as exertional dyspnea, until the tracheal lumen is approximately 8 mm in diameter, and this may take months to years to develop.17 It is therefore probable that our study underestimates the cost associated with readmission because we are examining readmissions during a 12-month cross-sectional timeframe rather than in a longitudinal study. Finally, the management of patients with tracheal injury is best performed in a multidisciplinary approach.18 Many tracheal stenosis cases require multiple minimally invasive procedures, and some patients are ultimately referred for surgical resections.19-21 Because our study examined only the costs of inpatient readmissions, the costs of these multiple outpatient interventions are not accounted for. Similarly, the indirect costs associated with time off from work and other such economic issues are also not addressed in our study. Despite the limitations of our study, it is thus far the largest investigation of post-intubation tracheal injury. To our knowledge, this is the only study examining hospital cost and LOS following tracheal injures due to ETI. It is important to note that in the current regulatory environment, hospitals do not receive payment for treatment of certain complications, called “never events,” such as catheter-related blood stream infections. In the future, it is possible that hospitals will not be reimbursed for the additional costs associated with post-intubation tracheal injury. Although some reimbursement is generally expected for
35
separate hospital admissions, Medicare and other payers are increasingly reluctant to pay extra for care needed to treat medical errors and adverse results of hospital care.
Conclusion The financial impact of post-intubation tracheal injury is significant, adding to the index admission cost and LOS as well as to future costs for treatment and repair of the injury. Prospective studies are required to obtain further data about indirect costs, as well as the costs associated with outpatient management of these patients.
Author Information From the Departments of Otolaryngology–Head and Neck Surgery (Drs. Bhatti, Mohyuddin, Laeeq, and Pandian), Anesthesiology and Critical Care Medicine (Dr. Mirski), and Interventional Pulmonology (Dr. Feller-Kopman), Johns Hopkins Hospital, Baltimore; and Strategic Health Resources (Drs. Reaven and Funk), La Canada. Corresponding author: Nasir I. Bhatti, MD, 601 North Caroline St, Ste 6241, Baltimore, MD 21287-0910. E-mail address: [email protected]. This article was presented at the 2009 AAO–HNSF Annual Meeting & OTO EXPO, San Diego, CA, October 4-7, 2009.
Author Contributions Nasir I. Bhatti, substantial contributions to concept and design, drafting of the article or revising it critically, final approval of version to be published; Atta Mohyuddin, drafting of the introduction and methods sections, literature review for ICD-9 diagnosis, final approval of version to be published; Nancy Reaven, assistance with formulating ICD-9 diagnosis and data analysis, final approval of version to be published; Susan E. Funk, contributions to drafting and revision of article, final approval of version to be published; Kulsoom Laeeq, statistical analysis and interpretation of data; final approval of version to be published; Vinciya Pandian, proof reading, editing, and contribution to revision of manuscript, final approval of version to be published; Marek Mirski, assistance with formulating a plan for methodology and results, final approval of version to be published; David Feller-Kopmann, original idea, substantial contributions to study concept and design and interpretations of data, drafting and critical revision of article, final approval of version to be published.
Disclosures Competing interests: Nancy Reaven, Covidien Healthcare: provided grant for research (no support for article); Susan E. Funk, Covidien Healthcare: provided grant for research (none for article). Sponsorships: Study was partially funded by Covidien Healthcare under an agreement that the study be conducted independently to reduce funding bias. Accordingly, the study was designed, performed, analyzed, interpreted, and written by the investigators without the involvement, review, or prepublication approval of Covidien.
References 1. Koshkareva Y, Gaughan JP, Soliman AM. Risk factors for adult laryngotracheal stenosis: a review of 74 cases. Ann Otol Rhinol Laryngol 2007;116:206 –10. 2. Hawkins DB. Glottic and subglottic stenosis from endotracheal intubation. Laryngoscope 1977;87:339 – 41.
36
Otolaryngology–Head and Neck Surgery, Vol 143, No 1, July 2010
3. Whited RE. A prospective study of laryngotracheal sequelae in longterm intubation. Laryngoscope 1984;94:367–77. 4. Gaynor EB, Greenberg SB. Untoward sequelae of prolonged intubation. Laryngoscope 1985;95:1461–7. 5. Stauffer JL, Oslen DE, Petty TL. Complications and consequences of endotracheal intubation and tracheostomy: a prospective study of 150 critically ill adult patients. Am J Med 1981;70:65–76. 6. HCUP Nationwide Inpatient Sample (NIS). Health Cost and Utilization Project (HCUP). 2006. Rockville (MD): Agency for Healthcare Research and Quality. Available at: http://www.hcup-us.ahrq.gov/ nisoverview.jsp. Accessed July 5, 2008. 7. Donnelly WH. Histopathology of endotracheal intubation: an autopsy study of 99 cases. Arch Pathol 1969;88:511–20. 8. Yang KL. Tracheal stenosis after a brief intubation. Anesth Analg 1995;80:625–7. 9. Kastanos N, Estopa MR, Perez M, et al. Laryngotracheal injury due to endotracheal intubation: incidence, evolution and predisposing factors. A prospective long term study. Crit Care Med 1983;11:362–7. 10. Schneider T, Storz K, Dienmann H, et al. Management of iatrogenic tracheobronchial injuries: a retrospective analysis of 29 cases. Ann Thorac Surg 2007;83:1960 – 4. 11. Grillo HC, Donahue HD, Mathisen JD, et al. Post intubation tracheal stenosis: treatment and results. J Thorac Cardiovasc Surg 1995;109: 486 –93. 12. Deeb ZE, Williams JB, Campbell TE. Early diagnosis and treatment of laryngeal injuries from prolonged intubation in adults. Otolaryngol Head and Neck Surg 1990;120:25–9.
13. Grillo HC, Cooper JD, Geffin B, et al. A low pressure cuff for tracheostomy tubes to minimize tracheal injury: a comparative clinical trial. J Thorac Cardiovasc Surg 1971;62:898 –907. 14. Cooper JD, Grillo HC. Experimental production and prevention of injury due to cuffed tracheal tubes. Surg Gynecol Obstet 1969;129: 1235– 41. 15. Seegobin RD, Van Hasselt GL. Endotracheal cuff pressure and tracheal mucosal blood flow: endoscopic study of effects of four large volume cuffs. Br Med J 1984;288:965– 8. 16. Zias N, Chroneou A, Tabba MK, et al. Post tracheostomy and post intubation tracheal stenosis: report of 31 cases and review of the literature. BMC Pulm Med 2008;8:18. 17. Hollingsworth HM. Wheezing and stridor. Clin Chest Med 1987;8: 231– 40. 18. Brichet A, Verkindre C, Dupont J, et al. Multidisciplinary approach to management of postintubation tracheal stenosis. Eur Resp J 1999;13: 888 –93. 19. Cavaliere S, Bezzi M, Toninelli C, et al. Management of post-intubation tracheal stenosis using the endoscopic approach. Monaldi Arch Chest Dis 2007;67:73– 80. 20. Macchiarini P, Verhoye JP, Chapelier A, et al. Partial cricoidectomy with primary thyrotracheal anastomosis for postintubation subglottic stenosis. J Thorac Cardiovasc Surg 2001;121:68 –76. 21. Sittel C, Blum S, Streckfuss A, et al. Cricotracheal resection in nontracheotomized adults: a prospective case series. Ann Otol Rhinol Laryngol 2008;117:288 –94.