Outcomes of tracheostomy after cardiac surgery through median sternotomy

Introduction

Background

Deep sternal wound infection (DSWI) is a well-known complication of cardiac surgery that contributes significantly to patient mortality and has garnered considerable attention (1).

Rationale and knowledge gap

Although tracheostomy wound complications are frequently reported in approximately 48% patients (2), with minor local infections at the tracheostomy site reportedly progressing to serious complications such as mediastinitis or lung abscess, the potential risks of tracheostomy complications (TCs) remain underestimated. Consequently, detailed analyses of outcomes specific to the tracheostomy site remain limited.

Several factors warrant particular attention to tracheostomy outcomes in patients after cardiac surgery through median sternotomy. First, the proximity of the sternal wound and the tracheostomy site may have a mutual influence. Second, retraction during cardiac surgery can cause blunt injury to the tissues around the sternal notch and neck, overlapping with the tracheostomy site. Third, the use of cardiopulmonary bypass (CPB) during cardiac surgery induces systemic inflammation, potentially affecting wound healing and tracheostomy outcomes. Accordingly, it is essential to investigate tracheostomy outcomes in patients who have undergone cardiac surgery through median sternotomy.

Objective

The aims of this study were to investigate the outcomes of tracheostomy in patients who underwent cardiac surgery through median sternotomy at our institution over a 10-year period, and to identify perioperative factors associated with TCs. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-762/rc).

Methods

Ethics statement

This clinical study was entirely observational and did not require registration. The research was conducted by reviewing medical records, ensuring anonymity of the patients. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The Institutional Review Board of Pusan National University Hospital (IRB No. 2503-020-149) approved the study and waived the need for informed consent because of the retrospective study design.

Patients and data collection

In total, 1,689 patients underwent cardiac surgery through median sternotomy at our center between February 2014 and August 2024, and the electronic medical records of 85 patients who underwent tracheostomy after the surgery were reviewed. After the exclusion of 21 patients who underwent cardiac surgery via thoracotomy or mini-thoracotomy, 64 patients were included in the analysis (Figure 1).

Figure 1 Study flowchart.

Comprehensive data, including patient characteristics, intraoperative details, and early postoperative outcomes, were systematically collected. Patient characteristics included sex, age, body mass index, underlying diseases, smoking history, the European System for Cardiac Operative Risk Evaluation (EuroSCORE) II, history of cardiac surgery or tracheostomy, pro-BNP levels, left ventricular ejection fraction, preoperative use of extracorporeal membrane oxygenation (ECMO) or intra-aortic balloon pump support (IABP), history of preoperative cardiopulmonary resuscitation (CPR), indication for tracheostomy, and time interval between sternotomy and tracheostomy. Intraoperative details included the urgency of the operation, intraoperative transfusion requirements, operative time, use of CPB, CPB duration, requirement for aortic cross-clamping (ACC), ACC duration, use of the left internal mammary artery (LIMA), department performing the tracheostomy, and tracheostomy approach.

Early postoperative outcomes included postoperative ECMO support, IABP support, in-hospital complications, total duration of hospital stay, duration of intensive care unit (ICU) stay, duration of ventilator use, mortality, cause of death, readmission, and tracheostomy wound complications. Tracheostomy wound complications were categorized according to the Clavien-Dindo classification, and major complications were defined as postoperative bleeding, acute kidney injury, mediastinitis, and stroke.

Comparison between patients with and those without TCs

The 64 study participants were divided into TCs (n=11) and no TCs (N-TC; n=53) groups depending on the occurrence of tracheostomy complications. Preoperative patient characteristics, intraoperative data, and early postoperative outcomes were analyzed and compared between the two groups.

Statistical analysis

Baseline characteristics of the two groups were analyzed, and intraoperative data and postoperative outcomes were compared using the independent t-test or Wilcoxon rank-sum test for continuous variables and the chi-square test or Fisher’s exact test for categorical variables. Perioperative factors related to tracheostomy complications were assessed, and logistic regression analysis was used to identify associated factors. When complete separation occurred in a univariate model, Firth’s penalized logistic regression was applied to obtain bias-reduced estimation. Due to a small number of complication events, the multivariable model was restricted to two clinically relevant predictors, and Firth’s penalized likelihood approach was again employed. A P value <0.05 was considered statistically significant. All statistical analyses were performed using R version 4.2.2 (R Core Team, 2020).

Results

Baseline characteristics

The median [range] age of the 64 enrolled patients was 68.5 [60.7; 76.0] years, and 50% (n=32) patients were male. The most common underlying condition was hypertension (68.6%), followed by diabetes mellitus (32.8%) and chronic obstructive pulmonary disease (32.8%). The median EuroSCORE II was 5.0% [2.1%; 11.2%]. A history of previous cardiac surgery was observed for six patients (9.4%), and two patients (3.1%) had a history of tracheostomy before cardiac surgery. Severe left ventricular dysfunction was present in 11 (17.2%) patients, preoperative ECMO was required for seven (10.9%) patients, and eight (12.5%) patients underwent CPR prior to surgery.

Postoperative respiratory failure requiring prolonged mechanical ventilation was the primary indication for tracheostomy in 37 (57.8%) patients. The median duration between cardiac surgery and tracheostomy was 12.5 [8.0; 15.0] days (Table 1). The mean follow-up time was 103.9±131.4 days, and total follow-up time was 6,648 days.

Intraoperative data

Aortic replacement was the most common procedure, performed in 26 (40.6%) patients, followed by valve procedures in 17 (26.6%) patients and coronary artery bypass grafting (CABG) in 13 (20.3%) patients. CPB was used for 60 (93.6%) patients, and ACC was required for 58 (90.6%) patients. On the day of cardiac surgery, a median of 7.5 [5.0; 10.0] units of packed red blood cells (pRBCs) was transfused; this included intraoperative transfusions. Tracheostomy was predominantly performed in the thoracic and cardiovascular surgery department (n=56; 87.5%), and the majority were surgical tracheostomies (n=59; 92.2%). Percutaneous dilatational tracheostomy was performed for a small proportion of patients (n=5; 7.8%; Table 2).

Outcomes of tracheostomy

During the postoperative course, 21 (32.8%) patients required mechanical circulatory support, including ECMO and IABP. The incidence of in-hospital major adverse cardiac and cerebrovascular events (MACCE) was 78.1% (n=50), and major complications occurred in 47 (73.4%) patients. The most common complication was pneumonia, affecting 54 (84.4%) patients. Sternal wound dehiscence and infection were observed in seven (10.9%) patients, while tracheostomy wound complications were more frequent, occurring in 11 (17.2%) patients. The in-hospital mortality rate was 40.6% (n=26), with multiorgan failure being the most common cause of death (n=9; 34.6%). The hospital readmission rate was 23.4% (n=15; Table 3).

Comparisons between the TC and N-TC groups

A history of redo tracheostomy was significantly more prevalent in the TC group than in the N-TC group (P=0.03). The time interval between cardiac surgery and tracheostomy was shorter in the TC group, with a median of 10.0 [8.5; 14.5] days compared with a median of 13.0 [8.0; 17.0] days in the N-TC group. However, this difference was not statistically significant (Table 1).

The median units of pRBCs transfused intraoperatively and on the day of cardiac surgery were significantly more in the TC group than in the N-TC group [12.0 (range, 6.0–16.0) vs. 7.0 (range, 5.0–9.0), P=0.042; Table 2]. Although the total hospital stay was similar between the two groups, the ICU stay was significantly shorter in the N-TC group by 412 h (P=0.03; Table 3). Among patients in the TC group (n=11), the underlying cardiac procedures were distributed as follows: aortic surgery (n=4), CABG (n=3), including one off-pump case, valve surgery (n=2), and other operations: pulmonary thromboembolectomy and right-atrial repair after cardiac trauma (n=2).

TCs and associated perioperative factors

TCs were classified according to the Clavien-Dindo classification. The most frequent complication was massive bleeding requiring surgical control (Grade IV), which occurred in five (45.6%) patients. Other complications included wound redness managed with daily dressing and wound care (Grade I) in three (27.3%) patients, tracheostomy wound infection requiring antibiotic therapy (Grade II) in two (18.2%) patients, and a wound requiring irrigation and revision surgery in the operating room (Grade III) in one (9.1%) patient. No grade V complications (tracheostomy-related deaths) were observed (Table 4).

The relationship between TCs and perioperative factors was analyzed using Firth’s penalized logistic regression. In multivariable analysis, a significant association was found between TCs and the use of LIMA (odds ratio: 8.64, 95% confidence interval: 1.43–52.19, P=0.02). The units of pRBCs transfused intraoperatively and on the day of cardiac surgery (odds ratio: 1.27, 95% confidence interval: 1.09–1.49, P=0.001) were also significantly associated with TCs (Table 5).

Discussion

In the present study, 3.8% patients who underwent cardiac surgery through median sternotomy required tracheostomy, with a TC rate of 17.2%. The in-hospital mortality rate for patients with TCs was 40.6%, while the incidence of MACCE was 78.1%, indicating a high complication rate. In comparison with the present findings, a recently published study focusing on patients who underwent tracheostomy after cardiac surgery through sternotomy reported a tracheostomy rate of 1.9% and an overall mortality rate of 21.3%, both lower than those observed in our study (1). These differences may be attributed to the higher baseline surgical risk in our patient population. Patients requiring tracheostomy are expected to represent a more critically ill subgroup with a more complex postoperative course. Consequently, a higher proportion of patients requiring tracheostomy in a cohort is likely to show worse postoperative outcomes. Moreover, our cohort likely included patients with a significantly higher cardiac surgical risk. The median EuroSCORE II in our study population was 5.0% [2.1%; 11.2%], whereas in the study by Tsai et al., the estimated mortality risks in the DSWI and non-DSWI groups were 0.185% and 0.235%, respectively (1).

Post-sternotomy mediastinitis is associated with a high mortality rate ranging from 14% to 47% (4,5). Tracheostomy is recognized as an independent risk factor for DSWI in post-sternotomy patients, with tracheostomy wound infection being the most common complication that potentially leads to mediastinitis (1). However, the risk associated with tracheostomy-related complications is often underestimated, resulting in the lack of comprehensive analysis of tracheostomy outcomes. Given the reciprocal effects of sternotomy wounds and tracheostomy, understanding tracheostomy outcomes and their influencing factors is essential to optimize patient management and minimize complications.

The timing of tracheostomy with respect to healing of the median sternotomy wound has been a subject of ongoing debate. Previous studies recommended avoiding early tracheostomy because of its significant association with sternal wound dehiscence (6,7). However, no large-scale study has specifically analyzed the risks associated with tracheostomy wound complications and their outcomes (8). In this study, the interval between cardiac surgery and tracheostomy was not significantly associated with TCs. However, several caveats should be considered when interpreting this result. At our center, most cardiac surgeons maintain a minimum interval of 1 week between sternotomy and tracheostomy in patients requiring prolonged mechanical ventilation. This practice is based on the assumption that a shorter interval increases the risk of procedural complications and mediastinitis. Anatomically, tracheostomy is performed at the third tracheal ring, which is in close proximity (approximately 2–3 cm) to the upper margin of the sternotomy wound. The risk is particularly elevated in patients with a short neck, those undergoing aortic arch replacement, and those undergoing off-pump CABG, where the incision is extended by an additional 2 cm from the sternal notch. Furthermore, during cardiac surgery, the soft tissue above the sternal notch is forcefully retracted caudally using an Army-Navy retractor to optimize surgical exposure. However, this maneuver can result in blunt trauma to the area adjacent to the tracheostomy site. From a wound healing perspective, the inflammatory phase begins on postoperative day 1 and persists until approximately day 4. It is followed by the proliferative phase, which overlaps and continues for 3–10 days. The proliferative phase is critical for re-epithelialization, which serves as a barrier against infection (9,10). Therefore, tracheostomy performed during the inflammatory phase may have a significant impact on wound healing by prolonging inflammation, delaying the transition to the proliferative phase, and increasing the risk of infection. However, in our study, tracheostomy was performed at a median of 13.0 and 10.0 days after cardiac surgery in the N-TC and TC groups, respectively; thus, it was performed after inflammatory phase; this might influence the result which showed the lack of a significant difference in complication rates between the two groups. Although the present study detected no statistically significant association between tracheostomy timing and postoperative complications, further research—including an analysis of the effects of earlier tracheostomy—is warranted to clarify this relationship.

Although the TC and N-TC groups showed significant differences with regard to a history of previous tracheostomy and the number of transfused pRBC units on the postoperative day, multivariable logistic regression analysis identified the use of LIMA as a potential contributing factor for TCs. A previous study in patients undergoing CABG have shown that bilateral internal mammary artery use may reduce sternal perfusion, thereby increasing the risk of complications such as DSWI (11). Infected sternal wounds close to the tracheostoma can also result in complications at the tracheostomy site. In addition, LIMA does not directly supply the lower cervical region; however, it may contribute to its perfusion through collateral circulation. The incidence of variations in the superior branches of the internal mammary artery, including anastomoses with the inferior thyroid artery, have been reported in up to 10% patients (12). In cases where LIMA is harvested, this could potentially reduce perfusion to the tracheostomy site, negatively affecting wound healing and clinical outcomes. As our study did not include direct measurements of regional perfusion, the proposed association between reduced perfusion and impaired tracheostomy site healing remains speculative. Future studies incorporating objective perfusion assessment are needed to validate this hypothesis.

Prolonged ICU stay is associated with increased mortality, and 30–50% ICU survivors experience physical, cognitive, or psychological sequelae, leading to reduced quality of life (13-15). In our study, the ICU stay was significantly longer in the TC group than in the N-TC group, likely because of the severity of complications. Notably, 55% (n=6) of the 11 patients in the TC group showed complications with a Clavien-Dindo grade of III or higher. Patients with milder complications (Grade I or II) are typically managed medically, allowing for timely transfer from ICU to the general ward. In contrast, patients with Clavien-Dindo Grade III or higher complications often require additional surgical intervention, transfusion, or prolonged ICU monitoring following the intervention. Our findings emphasize that TCs, particularly Clavien-Dindo Grade III or higher complications, should not be considered minor postoperative events, but rather as critical factors influencing long-term patient outcomes.

This study had several limitations. First, it was a single-center, retrospective study with a relatively small sample size, which may introduce potential selection bias. In addition, the cohort was heterogeneous in terms of surgical procedure and technique, yet the limited sample size and low event count precluded subgroup analysis to adjust for heterogeneity. Second, we analyzed only short-term outcomes, limiting our ability to assess the long-term effects of TCs. Third, the study spanned 11 years, during which surgical techniques and ICU management protocols likely evolved, potentially influencing the results. Lastly, despite limiting the multivariable model to two clinically relevant variables and applying Firth’s penalized logistic regression to mitigate small sample bias, the coefficient estimation remains imprecise and the confidence intervals wide; consequently, these results should be viewed as hypothesis-generating rather than definitive. Despite these limitations, to the best of our knowledge, this study represents the largest-scale analysis focusing specifically on tracheostomy outcomes in post-sternotomy patients. Future studies with larger cohorts and controlled variables, including surgical techniques, postoperative care protocols, and patient comorbidities, will provide further insights into the factors that contribute to TCs after cardiac surgery through median sternotomy.

Conclusions

The present study demonstrated that short-term mortality and morbidity rates were high for patients who underwent tracheostomy after cardiac surgery through median sternotomy. The timing of tracheostomy after sternotomy was not significantly associated with TCs. However, TCs were associated with LIMA use, and patients who experienced TCs had a prolonged ICU stay. Therefore, careful management and close surveillance of the tracheostoma are critical.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-762/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-762/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-762/prf

Funding: None.

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-762/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Pusan National University Hospital (IRB No. 2503-020-149) and individual informed consent was waived because of the retrospective study design.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Tsai YC, Phan K, Stroebel A, et al. Association between post-sternotomy tracheostomy and deep sternal wound infection: a retrospective analysis. J Thorac Dis 2016;8:3294-300. [Crossref] [PubMed]
2. Goldenberg D, Ari EG, Golz A, et al. Tracheotomy complications: a retrospective study of 1130 cases. Otolaryngol Head Neck Surg 2000;123:495-500. [Crossref] [PubMed]
3. Dindo D, Demartines N, Clavien PA, et al. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients. Ann Surg 2004;240:205-13. [Crossref] [PubMed]
4. Loop FD, Lytle BW, Cosgrove DM, et al. J. Maxwell Chamberlain memorial paper. Sternal wound complications after isolated coronary artery bypass grafting: early and late mortality, morbidity, and cost of care. Ann Thorac Surg 1990;49:179-87. [Crossref] [PubMed]
5. Molina JE. Primary closure for infected dehiscence of the sternum. Ann Thorac Surg 1993;55:459-63. [Crossref] [PubMed]
6. Brown AH, Braimbridge MV, Panagopoulos P, et al. The complications of median sternotomy. J Thorac Cardiovasc Surg 1969;58:189-97.
7. Hübner N, Rees W, Seufert K, et al. Percutaneous dilatational tracheostomy done early after cardiac surgery--outcome and incidence of mediastinitis. Thorac Cardiovasc Surg 1998;46:89-92. [Crossref] [PubMed]
8. Bösel J, Niesen WD, Salih F, et al. Effect of Early vs Standard Approach to Tracheostomy on Functional Outcome at 6 Months Among Patients With Severe Stroke Receiving Mechanical Ventilation: The SETPOINT2 Randomized Clinical Trial. JAMA 2022;327:1899-909. [Crossref] [PubMed]
9. Fitridge RThompson MSchultz GS, et al. Principles of Wound Healing. 2011.
10. Landén NX, Li D, Ståhle M. Transition from inflammation to proliferation: a critical step during wound healing. Cell Mol Life Sci 2016;73:3861-85. [Crossref] [PubMed]
11. Parissis H, Parissis M. Up-to-Date, Skeletonized or Pedicle Bilateral Internal Mammary Artery; Does It Matter? Ann Thorac Cardiovasc Surg 2023;29:53-69. [Crossref] [PubMed]
12. Bergman RA, Afifi AK, Miyauchi R. Thyroidea Ima (of Neubauer) Artery. Available online: https://www.anatomyatlases.org/AnatomicVariants/Cardiovascular/Text/Arteries/ThyroideaIma.shtml
13. Moitra VK, Guerra C, Linde-Zwirble WT, et al. Relationship Between ICU Length of Stay and Long-Term Mortality for Elderly ICU Survivors. Crit Care Med 2016;44:655-62. [Crossref] [PubMed]
14. Hope AA, Gong MN, Guerra C, et al. Frailty Before Critical Illness and Mortality for Elderly Medicare Beneficiaries. J Am Geriatr Soc 2015;63:1121-8. [Crossref] [PubMed]
15. Muscedere J, Waters B, Varambally A, et al. The impact of frailty on intensive care unit outcomes: a systematic review and meta-analysis. Intensive Care Med 2017;43:1105-22. [Crossref] [PubMed]