|
|
||||||||
Ann Thorac Surg 2005;79:1879-1885
© 2005 The Society of Thoracic Surgeons
Department of Cardiothoracic Surgery, New York Presbyterian HospitalCornell University, New York, New York
Accepted for publication October 22, 2004.
* Address reprint requests to Dr Lee, Dept of Cardiothoracic Surgery, New York Presbyterian HospitalCornell University, 525 East 68th St, M-4, New York, NY 10021 (E-mail: lyl2003{at}med.cornell.edu).
| Abstract |
|---|
|
|
|---|
METHODS: During 3 years, 86 tracheostomies were performed in more than 4,000 patients who underwent cardiac surgery, 59 open and 27 PDT. A retrospective analysis was performed comparing clinical and financial outcomes of the two groups.
RESULTS: There were no significant differences in demographics, medical histories, operations, or complications between open and PDT except the open group experienced more postoperative arrhythmias (70% [41 of 59] versus 44% [12 of 27], p < 0.05). Total savings associated with 1 year of PDT was $84,000, for a projected discounted savings of $283,000 during the study period. A sensitivity analysis of critical economic variables (number of tracheostomies per year, cost of operating room per minute, cost of intensive care unit bed per day) was included to evaluate the impact on cost savings. The net present value analysis, which discounts future savings by an appropriate interest rate, yielded a range of projected savings of PDT more than 5 years of $73,000 to $541,000 with a best estimate of $304,000 using figures established from our 3-year experience with PDT. Sensitivity analysis of the net present value for each critical variable was $227,000 per day of reduced intensive care unit length of stay, $180,000 per cost of operating room avoidance, $100,000 per intensive care unit bed cost per day, and $11,000 per additional tracheostomy per year.
CONCLUSIONS: There were no significant clinical differences between open and PDT in cardiac surgery patients during the 3-year study period; however, PDT offered significant cost savings.
| Introduction |
|---|
|
|
|---|
Historically, patients requiring elective surgical airways have undergone tracheostomy using an open surgical technique originally described by Jackson [5]. The open approach has been the standard of care since its inception and provides the standard against which other techniques should be measured. More recently, percutaneous dilatational tracheostomy (PDT) has gained popularity as a viable alternative to open tracheostomies because of its simplicity, ability to be performed at the bedside without transport of a potentially critically ill patient to the operating room (OR), and decreased cost [69]. In light of this development, the two methods have been compared to assess their clinical differences in terms of morbidity and mortality in a number of studies for a range of patient populations [1012]. Various authors have reported differing accounts of the efficacy and morbidity of PDT. Lin and colleagues [13] raised concerns regarding the possibility of posterolateral wall lacerations, whereas Norwood and associates [14] found 31% of patients had greater than 10% tracheal stenosis in long-term follow-up of critically ill patients who had undergone PDT. Others, such as Van Heurn and coworkers [15], observed better outcomes, especially in the hands of more experienced surgeons.
Although the literature generally supports the use of PDT in most patient populations, the appropriateness of its general use is not entirely accepted [16]. We reviewed our experience with both open tracheostomy (open) and PDT in a group of cardiac surgical patients during a 3-year period in a university hospital. The use of PDT was proposed as a more efficient means to secure a tracheostomy in critically ill cardiac surgical patients and represents a programmatic change in the Department of Cardiothoracic Surgerys approach to airway management. We compare these two groups on the basis of morbidity, mortality, and financial costs and savings associated with each technique.
| Material and Methods |
|---|
|
|
|---|
Open tracheostomies were performed in the OR under general anesthesia, which required an attending cardiothoracic surgeon or fellow, an anesthesiologist, nurses, and transportation of the patient from the intensive care unit (ICU). This technique followed standard surgical practices as described by Jackson [5]. The patients neck was slightly extended and prepared in a sterile fashion. A 3-cm transverse incision was made over the second tracheal ring. This incision was carried down to the level of the trachea, with splitting of the muscles in the midline and careful division of the thyroid when necessary. Sharp division of the second or third tracheal ring was completed, and a cuffed tracheostomy was inserted as the endotracheal tube was removed. The tracheostomy was then secured with Prolene sutures (Ethicon, Somerville, NJ) and umbilical tape placed around the patients neck.
Percutaneous dilatational tracheostomy was performed at the bedside in the ICU by either a cardiothoracic attending surgeon or fellow and the patients nurse, with 1% lidocaine for local anesthesia, morphine sulfate for pain control, and midazolam or propofol for sedation. Percutaneous dilatational tracheostomy was performed using a standard modified Seldinger technique with concurrent fiberoptic bronchoscopy to ensure correct placement of the tracheostomy in a manner similar to that described by Ciaglia and associates [6]. The endotracheal tube was pulled back under bronchoscopic guidance to a level just below the vocal cords, allowing the entire procedure to be performed under direct vision. A 2-cm skin incision was made over the second tracheal ring. The air column was then accessed between the second and third tracheal rings in the midline with a finder needle. After guidewire placement and initial dilatation, a single Blue Rhino dilator was used to dilate the trachea to 36F, followed by placement of a tracheostomy. All tracheostomies in this study were performed electively; none were emergent. Each of 6 attending surgeons assumed responsibility for performing tracheostomy on his patients, and there were no differences in the rates of incidence among the surgeons. The primary indications for performing a tracheostomy were respiratory failure, pulmonary hygiene, and airway protection.
The financial implications regarding the change in clinical practice from open to PDT was assessed along with clinical outcome. A discounted cash-flow analysis was performed to project the potential cost savings of PDT for a 5-year period. This method discounts future savings to adjust for the time value of money [17], referred to as a net present value (NPV). An analysis using a range of values for several critical variables determined the sensitivity of the NPV to each variable, namely the cost of an ICU bed, average number of tracheostomies per year, and cost of OR time.
SPSS Version 10.0 (SPSS, Inc, Chicago, IL) was used for all statistical calculations. Microsoft Excel 2000 (Microsoft Corp, Redmond, WA) was used for the financial analysis.
| Results |
|---|
|
|
|---|
Postoperative Outcome
As would be expected for patients requiring tracheostomy, patients in this study experienced a greater number of complications and higher mortality than patients not requiring prolonged ventilatory support, which may be a reflection of the severity of their preoperative illness. Greater than 90% of patients in either group were in New York Heart Association class 3 or 4, whereas less than 10% of the patients in either group were in New York Heart Association class 1 or 2. Table 1 lists the postoperative outcomes for the open and PDT groups. The ICULOS was 53.7 ± 47.5 days for open and 43.0 ± 33.4 days for PDT. Similarly, the trend in total length of stay was longer in the open group compared with the PDT group, 60.9 ± 54.9 and 45.1 ± 33.1 days, respectively, although they were not significantly different. The median ICULOS for the open and PDT groups was 36 and 35 days, respectively. The duration of endotracheal intubation was 14.5 ± 9.4 days with a range of 0 to 49 days before tracheostomy. There was a single case of a tracheostomy being performed on day 0, which was associated with a lost airway in the operating room. The indications for open and PDT include primary respiratory failure (56% [33 of 59] versus 56% [15 of 27]), pneumonia with subsequent respiratory failure (24% [14 of 59] versus 30% [8 of 27]), adult respiratory distress syndrome (15% [9 of 59]) versus 7% [2 of 27]), and cerebrovascular accident (5% [3 of 59] versus 7% [2 of 27]). The marginal difference between the two groups in regard to adult respiratory distress syndrome possibly reflects an ICU policy change in more recent years to delay tracheostomy in patients with adult respiratory distress syndrome until their respiratory status improves.
|
Financial Analysis
The NPV of cost savings associated with the institution of PDT was estimated to be $304,000, which represents the savings for a 5-year period. Table 2 shows the financial model and the baseline assumptions. The assumptions in the model follow directly from our experience with open tracheostomy and PDT. The average number of tracheostomies per year is approximately 29. Liberal estimates on the amount and cost of medications used for the PDT were used. The estimate of OR time reflects the time associated with setting up the room, procedural time, and room clean up. The decreased ICULOS was estimated by using the difference between the median LOS for the open and PDT groups. To assess the time value of money, a discount rate of 15% was used in the financial analysis.
|
|
|
| Comment |
|---|
|
|
|---|
Other major limitations are the duration and size of the study. Although no instances of tracheal stenosis were encountered in follow-up, other investigators report it as a possible long-term complication of PDT with a range of incidence from 0% to 31% [14, 1821]. Some of the methodological problems with these studies lie in the rates of follow-up, which may lead to selection bias and overestimation of the true incidence of tracheal stenosis. Still, this problem warrants consideration in PDT. We found no evidence of clinically relevant tracheal stenosis in our study, albeit the patients were only observed through their postoperative hospital admission. Furthermore, clinical signs and symptoms are not sensitive for the detection of tracheal stenosis [2, 22] and generally require computed tomographic scanning or bronchoscopy for accurate assessments. Van Heurn and colleagues [15], in their study of late complications of PDT, concluded that there was a correlation between the level of experience of the person performing the tracheostomy and risk for developing late tracheal stenosis, with an odds ratio between the most experienced surgeon and the other surgeons of 30.87 (2.87 to 332.0), p = 0.03. Unfortunately, because of the retrospective nature of this study, the long-term data on tracheal stenosis are unable to be assessed.
In our study, only 27% of the patients had a diagnosis of chronic obstructive pulmonary disease documented in their charts, which was the basis for our coding of past medical history. This stands in contrast to the 59% of patients with a documented history of smoking. We suspect that the actual number of patients with chronic obstructive pulmonary disease or emphysema was probably significantly higher; however, we are limited by the retrospective nature of this study to substantiate this claim. Obviously, in a prospective study of patients requiring postoperative tracheostomy, it would be worthwhile to assess the pulmonary function of patients with an extensive smoking history or with clinical signs and symptoms of pulmonary disease.
Inasmuch as the study population consists entirely of cardiothoracic patients, our study may not be able to be generalized to other population groups in medical or surgical ICUs. Despite limiting the scope of this study to the cardiothoracic ICU, the investigators have experience with both open and PDT in other ICU settings and other institutions, and have found similar outcomes. No cases of body habitus or aberrant anatomy, such as goiter or morbid obesity, prohibited the use of PDT in our population.
Complications
In regard to complications, our study is consistent with other reports on the use of PDT within this narrowly focused cohort [10, 12, 2325]. The concern for peritracheostomy infection, mediastinitis, and sternal wound infections appears unsubstantiated. In a thorough study, Byhahn and coworkers [23] found no evidence of cross-contamination of sternal wound infections and PDT. Likewise, Westphal and associates [10] and Stamenkovic and colleagues [26] report no chest wound infections associated with PDT. These authors speculate that the small skin incision and minimal disruptions of tissue planes required for PDT reduce the risk of peritracheostomy infection. Based on our results, we agree with their conclusion about the risk of peristoma infection associated with PDT.
Some investigators, such as Lin and colleagues [13], have reported severe lacerations of the trachea with PDT, which required major surgical interventions. Massick and associates [16] also reported a greater than expected complication rate and inability to complete the PDT, which necessitated open tracheostomy. Although these are known risks of PDT, they can be minimized with good technique. Like Westphal and coworkers [10], we found that routine use of fiberoptic bronchoscopy enables the surgeon to observe accurate placement of the Seldinger needle and guidewire before dilatation, which may prevent lateral placement of the dilator, the cause of tracheal laceration as well as trauma to the posterior wall of the trachea. Although concurrent use does not completely eliminate the risk of such accidents, it significantly reduces it, and is considered standard of care in our department. In our experience in the cardiothoracic ICU as well as other ICUs within our hospital, there have been no tracheal lacerations, and all attempted PDTs were completed without the need to convert to an open approach.
Our study demonstrated no significant differences in the complication rates between open and PDT except for an increased incidence of postoperative arrhythmias in the open group, 70% (41 of 59) versus 44% (12 of 27), p < 0.05. It is difficult to explain this difference, although it is possible that avoiding the transportation of patients on multiple intravenous medications to and from the OR and the more extensive open procedure may predispose these patients to a slightly higher incidence of arrhythmias. However, there were no changes in the patients heart rhythm during the performance of the tracheostomies regardless of approach. It is possible that the difference in incidence of arrhythmias represents an
error, and no true difference exists between the groups. Although not measured as an endpoint in this study, Hubner and colleagues [25] reported a better cosmetic result with PDT. We agree with their conclusions, based on the smaller incision size and minimal disruption of tissue planes in PDT, but cannot offer supportive data because of the lack of long-term follow-up. The high incidence of complications in this study and similar studies reflects the level of critical illness associated with patients who require tracheostomy and is not a reflection of the procedures themselves. The salient point remains that there is little or no difference in the postoperative complication rates between open and PDT.
Although the differences in the ICULOS (53.7 ± 47.5 days for open and 43.0 ± 33.4 days for PDT) and the total length of stay (60.9 ± 54.9 days and 45.1 ± 33.1 day) were not statistically significant, this may simply reflect a lack of power in our study. The variation in ICULOS and total length of stay is substantial in both groups and would require much larger sample sizes to discriminate between the two groups. For this reason, we decided to use the difference in median ICULOS, which favored PDT by 1 day, when developing our financial model. This approach and result make intuitive sense because the time from the decision to perform a tracheostomy to its completion is minimized with PDT. This is attributable to the facility with which the investigators could execute the decision to perform tracheostomy, ie, there was no delay in acquiring OR time and the PDT was performed at the bedside, usually in less than 10 minutes. Ultimately, this translates into a quicker discharge from the ICU for patients who require tracheostomy before being considered for rehabilitation placement.
Both the 30-day survival and survival to discharge were similar between the two groups. Again, this suggests the technical approach of tracheostomy does not affect clinical outcome for patients requiring tracheostomy. The main determinants of survival remain more a function of the patients preoperative condition and postoperative physiologic status rather than method of tracheostomy.
Financial Analysis
In the current economic environment, any new approach to existing surgical or medical problems will be subject to some form of economic review. There are many ways to assess the financial impact of new therapies or the delivery of standard therapies. We chose to use a discounted cash-flow analysis with sensitivity testing of various key variables to elucidate their impact on the final cost savings. The model could have used more sophisticated modeling techniques, such as stochastic assessments of the critical variables. However, the increased elegance of such a model would not have altered the qualitative assessment or conclusions about PDT. The baseline analysis suggested a substantial cost savings despite a large range for the NPV as assessed for a 5-year period. Our analysis broke down each critical variable into a change in the NPV per unit of cost variable (
NPV/U of variable). Pareto analysis of the
NPV/U of variable informs us where the bulk of the savings rests. This type of analysis can test the validity and intuitive sense of the model and its conclusions. Not surprisingly, the main cost drivers were the ICULOS and the obviation of OR time, which does not account for anesthesia and OR nursing staff. We disagree with Massick and colleagues [16] that an open bedside tracheostomy offers a superior approach financially or operatively to PDT because they do not account for the OR instrument tray required to perform the procedure. We routinely perform bronchoscopy after all tracheostomies, regardless of approach. Thus, the use of bronchoscopy is a cost-neutral item in our institution and offers no cost savings as it does in the analysis of Massick and associates [16].
Conclusions
The growing body of literature on PDT suggests it offers a safe alternative to open tracheostomy provided strict attention to detail and adherence to protocol is followed. We conclude that there are no significant clinical differences between open and PDT in cardiothoracic patients; however, the percutaneous technique offers significant cost savings when assessed on a discounted cash flow basis for a 5-year period.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. W. Gromann, O. Birkelbach, and R. Hetzer Balloon Dilatational Tracheostomy: Initial Experience with the Ciaglia Blue Dolphin Method Anesth. Analg., June 1, 2009; 108(6): 1862 - 1866. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. P. Williams, M. J. Rosen, J. Jin, M. F. McGee, S. J. Schomisch, and J. Ponsky Objective Analysis of the Accuracy and Efficacy of a Novel Fascial Closure Device Surgical Innovation, December 1, 2008; 15(4): 307 - 311. [Abstract] [PDF] |
||||
![]() |
S. Nundy, A. Mukherjee, J. B. Sexton, P. J. Pronovost, A. Knight, L. C. Rowen, M. Duncan, D. Syin, and M. A. Makary Impact of Preoperative Briefings on Operating Room Delays: A Preliminary Report Arch Surg, November 1, 2008; 143(11): 1068 - 1072. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. L. Ngaage, A. R. Cale, S. Griffin, L. Guvendik, and M. E. Cowen Is post-sternotomy percutaneous dilatational tracheostomy a predictor for sternal wound infections? Eur. J. Cardiothorac. Surg., June 1, 2008; 33(6): 1076 - 1079. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. B. Rahmanian, D. H. Adams, J. G. Castillo, J. Chikwe, and F. Filsoufi Tracheostomy is Not a Risk Factor for Deep Sternal Wound Infection After Cardiac Surgery Ann. Thorac. Surg., December 1, 2007; 84(6): 1984 - 1991. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. De Leyn, L. Bedert, M. Delcroix, P. Depuydt, G. Lauwers, Y. Sokolov, A. Van Meerhaeghe, and P. Van Schil Tracheotomy: clinical review and guidelines Eur. J. Cardiothorac. Surg., September 1, 2007; 32(3): 412 - 421. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |