ATS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Seth Daniel Force
Daniel L. Miller
Rebecca Petersen
Kamal A. Mansour
Joseph Craver
Robert A. Guyton
Joseph I. Miller, Jr
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Force, S. D.
Right arrow Articles by Miller, J. I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Force, S. D.
Right arrow Articles by Miller, J. I., Jr

Ann Thorac Surg 2005;80:618-622
© 2005 The Society of Thoracic Surgeons

Original article: Cardiovascular

Incidence of Deep Sternal Wound Infections After Tracheostomy in Cardiac Surgery Patients

Seth Daniel Force, MD * , Daniel L. Miller, MD, Rebecca Petersen, BA, Kamal A. Mansour, MD, Joseph Craver, MD, Robert A. Guyton, MD, Joseph I. Miller, Jr, MD

Department of Cardiothoracic Surgery, The Emory Clinic, Atlanta, Georgia

Accepted for publication February 9, 2005.

* Address reprint requests to Dr Force, Department of Cardiothoracic Surgery, The Emory Clinic, 1365 Clifton Rd, NE, Bldg A, Ste 2100, Atlanta, GA 30322 (Email: sethforce{at}emoryhealthcare.org).

Presented at the Fifty-first Annual Meeting of the Southern Thoracic Surgical Association, Cancun, Mexico, Nov 2–4, 2004.


   Abstract
Top
 Abstract
Introduction
 Patients and Methods
 Results
 Comment
 Discussion
 References
 
BACKGROUND: Tracheostomy after median sternotomy is associated with an increased risk of deep sternal wound infection (DSWI). However, associated comorbidities in these patients make the exact risk of tracheostomy difficult to discern. Therefore, we performed a retrospective review to determine the risk of DSWI in our cardiac surgery patients who underwent postoperative tracheostomy.

METHODS: Our cardiac surgery database was queried from January 1996 to July 2003. Perioperative morbidities were identified for all patients who underwent tracheostomy after cardiac surgery. Statistical analysis was performed using {chi}2 analysis and Fisher’s exact test.

RESULTS: During the study period, 16,277 cardiac procedures were performed through a median sternotomy. Tracheostomy was performed in 291 patients (1.8%). Deep sternal wound infections occurred in 0.8% of all patients and in 3.4% of patients (9 of 268) who underwent tracheostomy. Mean number of days to tracheostomy was 14.2 in the DSWI group and 15.8 in the non-DSWI group (p = 0.45). In patients with a tracheostomy, preoperative renal failure was the only perioperative comorbidity found to be significantly associated with DSWI (p = 0.03). Overall operative mortality was 3.6%. Mortality was 16.5% for patients with DSWI, 22.7% for patients requiring tracheostomy, and 55% for patients with DSWI and tracheostomy.

CONCLUSIONS: The incidence of DSWI after tracheostomy, in our experience, is not as high as previously reported. Tracheostomy can be performed safely after median sternotomy with a low morbidity and mortality rate. However, if DSWI occurs after tracheostomy, mortality is significant.


   Introduction
Top
 Abstract
 Introduction
Patients and Methods
 Results
 Comment
 Discussion
 References
 
Tracheostomy has long been associated with an increased incidence of deep sternal wound infections (DSWI) and mediastinitis in patients after median sternotomy for cardiac procedures. The exact incidence of sternal infections in this population has ranged, in recent published articles, from 0.4% to 8.6% [1, 3]. However, patients requiring tracheostomy usually have other perioperative comorbidities that may be partly responsible for the increase in wound infections and death after cardiac procedures. The purpose of our study was to identify the incidence of tracheostomy-associated DSWI, identify other risk factors for sternal infections, and determine the overall increase in mortality attributed to these complications.


   Patients and Methods
Top
 Abstract
 Introduction
 Patients and Methods
Results
 Comment
 Discussion
 References
 
After approval by our Institutional Review Board, we queried our cardiac database, for the period of January 1996 to July 2003, for all patients who underwent median sternotomy for cardiac procedures, patients who also had a tracheostomy, and patients who had sternal wound infections. Cases of DSWI that occurred before tracheostomy were placed in the nontracheostomy group when performing statistical analysis. Patients with DSWI were identified from operative notes that detailed sternal washouts, debridements, and muscle flaps after sternotomy. All cardiac procedures were performed by attending surgeons at Emory University Hospital or Crawford Long Hospital. During this time, there were two to four thoracic surgeons who performed tracheostomies. All tracheostomies were performed with an open technique, and although surgical techniques varied, there were no specific precautions taken for this group of patients. Tracheostomies were either performed in the operating room or in the cardiothoracic intensive care unit. Tracheostomy was usually deferred for 10 or more days, after sternotomy, as per institutional policy.

Univariate analysis was performed by Fisher’s exact test and {chi}2 test to identify significant associations between various perioperative morbidities and DSWI. A p value of 0.05 or less was considered significant. Multivariate analysis using logistic regression was then used to determine independent variables associated with DSWI and death. The large sample size allowed for very small effects being recognized as significant at a p value of 0.05; therefore, only variables with a p value of 0.01 or less were considered significant in the multivariate regression. Sex, on/off pump, hypertension, diabetes, chronic obstructive pulmonary disease (COPD), renal failure, current smoker, prior stroke, prior peripheral vascular disease, age, and body mass index (BMI) were evaluated independently to decide which variables to include in the regression analysis; and variables found to be significant at the 0.01 level were selected.


   Results
Top
 Abstract
 Introduction
 Patients and Methods
 Results
Comment
 Discussion
 References
 
Between January 1996 and July 2003, 16,277 cardiac procedures were performed utilizing a median sternotomy. This sample size had an 80% power to show a 3.25% increased risk of DSWI in patients with a tracheostomy with a p value of less than 0.05. Two hundred ninety-one patients (1.8%) required tracheostomy postoperatively. Demographics for the tracheostomy and nontracheostomy groups are shown in Table 1. The patients in the tracheostomy group were more likely to be female and to have a history of congestive heart failure, hypertension, diabetes, COPD, prior stroke, prior myocardial infarction, renal failure, and an ejection fraction less than 50%.


View this table:
[in this window]
[in a new window]
  		Table 1. Demographic Variables for All Tracheostomy and Nontracheostomy Patients
 
Deep sternal wound infection occurred in 124 patients, for an overall rate of 0.8%. Deep sternal wound infections were found to occur in 0.7% of patients not requiring tracheostomy (115 of 15,894) and 3.4% of patients requiring tracheostomy (9 of 268). Characteristics for the tracheostomy patients who developed DSWI are shown in Table 2. The mean number of days to tracheostomy was 14 (SD = 6) in the DSWI group and 16 (SD = 7) in the group without DSWI (p = 0.45). The average time from tracheostomy to diagnosis of DSWI was 25 days. Table 3 shows that, compared with the absence of a tracheostomy, the presence of a tracheostomy was significantly associated with DSWI (p < 0.001, odds ratio 4.8).


View this table:
[in this window]
[in a new window]
  		Table 2. Characteristics of Patients Who Had a Tracheostomy Followed by a Deep Sternal Wound Infection
 

View this table:
[in this window]
[in a new window]
  		Table 3. Risk of Deep Sternal Wound Infection (DSWI) With Tracheostomy
 
Tracheostomy and nontracheostomy patients, with and without DSWI, were compared by univariate analysis. Variables evaluated are shown in Table 4. In patients who underwent tracheostomy, preoperative history of renal failure was the only variable significantly associated with DSWI (p <0.05). In the group of patients who did not require tracheostomy, all the variables evaluated were significantly associated with DSWI (p < 0.05), except family history of coronary artery disease, history of COPD, history of peripheral vascular disease, and presence of cardiogenic shock.


View this table:
[in this window]
[in a new window]
  		Table 4. Univariate Analysis Variables
 
Logistic regression was performed using significant variables found in the univariate analysis. Results of the logistic regression for predictors of DSWI are shown in Table 5. Tracheostomy, age over 68 years, BMI greater than 28, and a history of diabetes were all associated with an increased risk of DSWI. Tracheostomy carried the highest risk with an odds ratio of 5.65. The odds ratio was found to be 4.81 (confidence interval: 2.41 to 9.57, p < 0.0001) when tracheostomy was tested alone as a predictor of DSWI.


View this table:
[in this window]
[in a new window]
  		Table 5. Predictors of Deep Sternal Wound Infection Logistic Regression (n = 97 of 11,051)
 
The mortality rate was 3.6% for the entire group of patients undergoing cardiac surgery, 16.5% for those with DSWI, and 22.7% for those with a tracheostomy. Mortality was 22% for patients with a tracheostomy and no DSWI and 55% for patients with a tracheostomy and DSWI (p = 0.034). Logistic regression was performed to evaluate predictors of mortality (Table 6). In addition to the presence of a tracheostomy, female sex, renal failure, and age over 68 were found to be independent predictors of mortality. The odds ratio was 7.92 (confidence interval: 5.94 to 10.55, p < 0.0001) for tracheostomy when it was tested alone as a predictor of mortality. Table 7 demonstrates the results of testing tracheostomy and DSWI, alone, as predictors of mortality.


View this table:
[in this window]
[in a new window]
  		Table 6. Predictors of Mortality, Logistic Regression (n = 637 of 16,277)
 

View this table:
[in this window]
[in a new window]
  		Table 7. Predictors of Mortality, Tracheostomy, and Deep Sternal Wound Infection, Logistic Regression (n = 637 of 16,277)
 

   Comment
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Comment
Discussion
 References
 
Tracheostomy has long been associated with an increased risk of sternal wound infection after median sternotomy for cardiac surgery procedures [2]. Curtis and coworkers [3] reported a 0.7% rate of mediastinitis in nontracheostomy patients and a rate of 8.6% in patients requiring a tracheostomy after median sternotomy. Other authors who evaluated sternal wound infections after cardiac surgery have failed to comment on tracheostomy as a specific risk factor for DSWI [4–6].

Our nontracheostomy patient population had the same rate of DSWI reported by the Curtis group, but only 3.4% of our tracheostomy patients developed sternal wound infection. The exact reason for this lower rate is difficult to discern and may be due to differences in detection and reporting of DSWI, or the larger sample size in our study may have led to a more accurate rate. That is supported by the power of our study, which was able to detect as small as a 3.25% increased incidence of mediastinitis due to tracheostomy.

Tracheostomy was shown to be a predictor of DSWI by univariate and multivariate analysis. Multivariate analysis also revealed diabetes, age, and BMI to be predictors of DSWI, but tracheostomy was found have the greatest odds ratio. Patients with a tracheostomy have almost five times the risk of developing DSWI compared with patients without tracheostomy.

The timing of tracheostomy is thought to play a role in the subsequent development of DSWI by seeding the sternal wound with airway bacteria. Brown and colleagues [7] stated that "the dangers of a tracheostomy so near to a vulnerable wound can be avoided...until the tissue planes are healing" and recommended waiting at least 2 weeks from the time of sternotomy before performing a tracheostomy. Pierce and associates [8] suggested isolating the tracheostomy from the sternotomy wound by placing it through the cricothyroid membrane. One study evaluating early tracheostomy found that 38 of 72 patients (53%) had evidence of superficial wound infection and 18 of the patients had the same bacteria in their wounds and tracheal secretions [1].

The timing of tracheostomy has also been debated as a factor in the development of DSWI. Stamenkovic and colleagues [1] showed that tracheostomy could be performed safely in patients as early as 1 to 2 days after sternotomy but did not say if there was any statistical difference in outcomes between patients who had early versus late tracheostomy. Curtis and associates [3] commented that, in their study, "the time interval between CABG and tracheostomy was not predictive of mediastinitis." However, the time to tracheostomy was relatively late, 19 days in patients who subsequently developed DSWI and 25 days in patients who did not [3]. In our study, the mean number of days to tracheostomy in the group of patients with DSWI was 14, with a mean time to diagnosis of DSWI of 25 days from the time of tracheostomy, which was not statistically different from the time to DSWI in the nontracheostomy group. This finding suggests that timing may not have played a major role in the development of DSWI.

Percutaneous tracheostomy has been advocated to prevent DSWI in patients with sternotomies after cardiac surgery [9, 10]. The apparent benefit is achieved by the tight seal created between the skin and the tracheostomy tube. Byhahn and coworkers [11] suggested that percutaneous tracheostomies could be performed early and with a lower risk of mediastinitis than open tracheostomies. However, there are several problems with these studies. They are all small studies ranging from 50 to 144 patients, and therefore are lacking the statistical power to identify risk factors for DSWI, given its overall low incidence. Additionally, patients with difficult anatomy were not considered for percutaneous tracheostomy in these studies. These would most likely include patients with a high BMI and possibly other risk factors that we found to be predictors of DSWI, thereby giving an advantage to the percutaneous group.

The overall mortality rate in our study was 3.6% but increased to almost 17% with DSWI and to 23% with tracheostomy. Risk of death was increased eight times in patients with a tracheostomy and six times in patients with DSWI. Patients with a tracheostomy and DSWI had an extremely poor outcome, with a 55% mortality. Interestingly, when univariate analysis was performed on all patients with a tracheostomy, renal failure was identified as the only significant covariate among patients with and without DSWI. This result was due to the overall high incidence of comorbidities among all the patients requiring tracheostomy and proves that the need for tracheostomy is an indicator of a high-risk patient.

This study proves that DSWI, alone or after tracheostomy, occurs infrequently after major cardiac surgery procedures. Other authors have suggested that perioperative comorbidities such as obesity, COPD, and postoperative inotropic support have led to the highest risk of DSWI [4, 5]. In our study, however, tracheostomy more than any other independent variable was associated with an increased incidence of DSWI; and when the two occur together, a very high risk of death can be expected. Fortunately, DSWI and tracheostomy together occur with a low incidence in patients who already have other significant comorbidities increasing the risk of complications after cardiac surgery.


   Discussion
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Comment
 Discussion
References
 
DR ANDREW C. FIORE (St. Louis, MO): We always try and avoid tracheostomy, but if you are forced to do this procedure after a sternotomy, from your analysis, when is the safest time to do it?

DR FORCE: The answer is no, and obviously that has been a hot topic that everybody has talked about, timing; 14 days has been discussed. We actually had an institutional policy to not perform tracheostomies prior to the tenth postoperative day. So if you looked at our two groups of patients, those who developed a wound infection and those who didn’t, the time to tracheostomy was about 2 weeks, and I think the standard deviation was around 7, but there was no difference between the two groups.

The problem with trying to define the best time to perform a tracheostomy is that sternal wound infections occur rarely. So if you had to carve out the patients who had it, and we broke it down, there aren’t enough in any group, 7 days or 8 days, whatever, to determine a statistically significant difference. In addition, you would have to do a prospective study to really have a good study to say whether prior to 7 days, after 7, or more is the correct time. So in answer, no, we don’t know the answer to that.

DR LYNN H. HARRISON (New Orleans, LA): That was a very nice report of a quite extensive series that showed findings consistent with those reported by a number of other groups. Unfortunately, the associated risk factors are elements over which cardiac surgeons have minimal control; if we could choose not to operate on morbidly obese patients, diabetics or patients in renal failure, I am sure we would all do so. More useful would be an examination I think of those surgeon-controlled factors that are associated with deep sternal wound infection and aside from the issue of tracheostomy. Have you looked at that? Have you looked at the incidence of infection in terms of, for example, exuberant versus conservative users of electrocautery or the exact timing of the administration of preoperative antibiotics and so forth, the things that might in fact enable us to pick our way through this minefield without getting our legs blown off?

DR FORCE: We looked at the perioperative morbidities in the patients. It is hard to define exuberant users of electrocautery, and it was hard for us to go back and determine the exact timing of the antibiotics. We did look at things like on-pump versus off-pump, cross-clamp time, pump time, variables that were more easy to define in such a large patient population, and the ones that were significant in a multivariate analysis were the ones that I showed.

DR STEPHEN D. CASSIVI (Rochester, MN): Seth, thank you for a very well presented and very thoughtful paper. As a cardiothoracic surgeon like you who is interested in general thoracic surgery, I would like to hear your comments on the after topic. Your paper concentrated on tracheostomy after cardiac surgery with median sternotomy. I would like to know in this day and age of the Phoenix of lung volume reduction surgery rising up again, what your comments are with regard to that particular population. We know that LVRS can be done by thoracoscopy, but you and I both trained with Dr Joel Cooper at Washington University in St. Louis who taught us how to do it by median sternotomy. I think this group has a different cornucopia of comorbidities than the group that you presented. I would just like to hear your comments regarding this problem in that group.

DR FORCE: The most obvious risk factor would certainly be that the majority of those patients are on steroids. I think the one thing as a thoracic surgeon that this study proved to me is that although we are probably never going to know the answer, it probably doesn’t matter how, when, or why you do the tracheostomy. These patients are sick, and that was the point of the talk. You saw the demographics. These patients are sick going into surgery and they are sick coming out of surgery. They have a significant increase in mortality, tracheostomy, and deep sternal wound infection, and probably you are better off performing a tracheostomy in them early.

The big question is, does it make any difference? These patients aren’t just sick from their sternal wound infections. They have renal failure, strokes and pneumonias so is it really going to make a difference whether you trach them at 7 days or 10 days in terms of wound infections or their overall outcome? The answer with such a low incidence is probably no, we probably will never be able to show that performing tracheostomies at 6 versus 7 versus 8 versus 10 days makes a difference. I tend to perform tracheostomies in my patients early just because I think it is more comfortable for the patients, and we can wean the patients from the ventilator more aggressively. I did not specifically look at the population of lung volume reduction, but it would be an interesting thing to look at.

DR ULRICH FRANKE (Jena, Germany): Congratulations for your excellent presentation. I have one comment regarding the surgical technique. We analyzed our results using the minitracheostomy compared with the dilation tracheostomy and we didn’t find any difference. The minitracheostomy might be a little less traumatic.

DR FORCE: That is obviously a big point, and I think you are talking about looking at the percutaneous tracheostomies, and most of those studies have come out of Europe. The benefit with that is thought to be that there is a tight seal between the skin and the tracheostomy tube, and so the soiling of the wound is much less, and if you look at some of the studies, the rates of sternal wound infection are much lower than we presented. In fact, that 0.4% that I showed, for instance, at the beginning came from a percutaneous tracheostomy trial.

I perform percutaneous tracheostomies, too, but excluding patients with a high BMI, the big, heavy patient with no neck who probably also has diabetes, selects for a healthier group of patients. Probably the risk factors are not necessarily due to soiling at that point but probably the other high risk factors for the patients going into the tracheostomy. We will perform percutaneous tracheostomies on those patients, but I can tell you, given the usual BMI of the patient undergoing cardiac surgery, it is very difficult to perform this safely.


   References
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Comment
 Discussion
 References
 

  1. Stamenkovic SA, Morgan IS, Pontefract DR, Campanella C. Is early tracheostomy safe in cardiac patients with median sternotomy incisions? Ann Thorac Surg 2000;69:1152-1154.[Abstract/Free Full Text]
  2. Sanfelippo PM, Danielson GK. Complications associated with median sternotomy J Thorac Cardiovasc Surg 1972;63:419-423.[Medline]
  3. Curtis JJ, Nicole CC, McKenney CA, et al. Tracheostomya risk factor for mediastinitis after cardiac operation. Ann Thorac Surg 2001;72:731-734.[Abstract/Free Full Text]
  4. Bitkover CY, Garlund B. Mediastinitis after cardiovascular operationa case-control study of risk factors. Ann Thorac Surg 1998;65:36-40.[Abstract/Free Full Text]
  5. The Parisian Mediastinitis Study Group Risk factors for deep sternal wound infection after sternotomya prospective, multicenter study. J Thorac Cardiovasc Surg 1996;111:1200-1207.[Abstract/Free Full Text]
  6. Ashley ESD, Carroll DN, Engemann JJ, et al. Risk factors for postoperative mediastinitis due to methicillin-resistant Staphylococcus aureus Clin Infect Dis 2004;38:1555-1560.[Abstract/Free Full Text]
  7. Brown AH, Braimbridge MV, Panagopoulos P, Sabar EF. The complications of median sternotomy J Thorac Cardiovasc Surg 1969;58:189-197.[Medline]
  8. Pierce WS, Tyers GFO, Waldhausen JA. Effective isolation of a tracheostomy from a median sternotomy wound J Thorac Cardiovasc Surg 1973;66:841-842.[Medline]
  9. Hubner N, Rees W, Böckelmann M, Christmann U, Warnecke H. Percutaneous dilatational tracheostomy done early after cardiac surgeryoutcome and incidence of mediastinitis. Thorac Cadiovasc Surg 1998;46:89-92.
  10. Westphal K, Lohrengel A, Wilke HJ, Rinne T, Wichelhaus TA, Lischke V. Percutaneous dilatational tracheostomy in a cardiac surgical ICU Z Hertz Thorax Gefaßchir 1998;12:177-181.
  11. Byhahn C, Rinne T, Halbig S, et al. Early percutaneous tracheostomy after median sternotomy J Thorac Cardiovasc Surg 2000;120:329-334.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Eur J Cardiothorac SurgHome page
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 - 1081.
[Abstract] [Full Text] [PDF]

Home page
 Ann. Thorac. Surg.Home page
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]

Home page
 Ann. Thorac. Surg.Home page
L. D. Berrizbeitia
Invited commentary
Ann. Thorac. Surg., December 1, 2007; 84(6): 1991 - 1992.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Seth Daniel Force
Daniel L. Miller
Rebecca Petersen
Kamal A. Mansour
Joseph Craver
Robert A. Guyton
Joseph I. Miller, Jr
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Force, S. D.
Right arrow Articles by Miller, J. I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Force, S. D.
Right arrow Articles by Miller, J. I., Jr

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