World Journal of Cardiovascular Surgery, 2013, 3, 114-118 Published Online July 2013 (
Early versus Late Tracheostomy for Ventilator
Dependence after Cardiovascular Surgery:
Long-Term Results
Joseph M. Ladowski*, Hannah E. Downey, Bennet J. Ladowski, Joseph S. Ladowski
Lutheran Hospital of Indiana, Fort Wayne, USA
Email: *,,
Received June 11, 2013; revised July 11, 2013; accepted July 18, 2013
Copyright © 2013 Joseph M. Ladowski et al. This is an open access article distributed under the Creative Commons Attribution Li-
cense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Although concerns exist about the possibility of infecting a sternotomy wound by performing a tracheostomy early after
cardiac surgery, it has been shown that tracheostomy performed before postoperative day 10 can improve outcomes for
cardiovascular surgery patients who require long-term ventilation. We retrospectively reviewed all charts for patients
who underwent cardiovascular surgery and required tracheostomy at a single-institution site from January 2005 to July
2012. Patients were divided into two groups based upon whether their tracheostomy was performed less than 10 days
after their initial operation (early tracheostomy) or 10 days after their initial operation (late tracheostomy). Preopera-
tive demographics and postoperative complications were studied. There were no new mediastinitis episodes following
tracheostomy. Early tracheostomy was associated with higher 90-day (74.4% vs 56.8%) and 180-day (47.4% vs 28.7%,
p < 0.047) survival when compared to late tracheostomy. Early tracheostomy after cardiovascular surgery does not
seem to be associated with an increased rate of mediastinitis but is associated with improved survival.
Keywords: Surgery Cardiovascular; Injury Lung; Intensive Care; Mediastinitis
1. Introduction
Long-term ventilator dependency following cardiovascu-
lar surgery occurs in up to 20% of all patients [1]. For
patient comfort and to facilitate ease of mechanical ven-
tilation, tracheostomy is commonly performed upon ven-
tilator-dependent patients. The timing of tracheostomy is
open to debate. Some investigators have indicated that
patients who undergo earlier tracheostomy (less than ten
days postoperatively) have decreased length of stay,
morbidity and mortality [2]. Others have found that ear-
lier tracheostomy in noncardiovascular surgery patients
is associated with shorter times on mechanical ventilation
and lower costs [3]. This study is a retrospective analysis
of 137 consecutive patients who underwent cardiovascu-
lar surgery and required tracheostomy for long-term ven-
tilator dependence. We analyzed these patients’ records
to determine whether early or late tracheostomy would
affect development of sternotomy wound infections
and whether a strategy of early sternotomy would allow
for improved survival.
2. Materials and Methods
We retrospectively reviewed the charts of 137 consecu-
tive patients who underwent cardiovascular surgery and
subsequent tracheostomy from January 2005 to July 2012
at Lutheran Hospital of Indiana. The types of initial op-
erations are summarized in Table 1.
The time interval from initial operation to tracheo-
stomy was decided clinically based upon the surgeon’s
Table 1. Types of operations.
Operation Number
Coronary Artery Bypass 88
Abdominal Aortic Aneurysm (open) 13
Thoracic Aortic Aneurysm (open) 9
Single Valve Replacement or Repair 15
Double Valve Replacement or Repair 11
Combined Revascularization and Valve Replacement 1
Total 137
*Corresponding author.
opyright © 2013 SciRes. WJCS
impression, aided by pulmonary consultation, that the pa-
tient would require prolonged ventilation. Tracheostomy
was performed surgically with standard transverse neck
incision and cephalocaudad tracheal incision followed by
insertion of a tracheostomy appliance under direct vision.
During this period, none of our tracheostomy procedures
were done in a percutaneous fashion.
Statistical analysis of data was performed by subject-
ing survival data to Kaplan-Meier analysis. Significance
was determined if p was <0.05 by Mantel-Cox log rank
Sixteen variables were studied for their univariate
(Fisher’s exact test for frequencies and analysis of vari-
ance for comparison of means) influence upon mortality.
These variables are summarized in Table 2. All patient
data was kept anonymous for reporting purposes.
3. Results
Outcomes for the 137 patients are summarized in Figure
1. Total mortality for the group was 42% at one year.
We began by analyzing the effect of earlier (less than
ten days of initial operation) versus later (greater than or
equal to ten days after initial operation) timing of tra-
cheostomy. Thirty-three of the patients underwent tra-
cheostomy within 10 days of their initial operation while
104 patients underwent tracheostomy greater than 10
days following their initial operation. Those patients who
Table 2. Variables studied for their possible association
with mortality.
1. Gender
2. Age
3. Redo surgery
4. Active smoker
5. History of alcohol abuse
6. Preoperative renal failure
7. Preoperative body mass index
8. Preoperative treatment for hypertension
9. Preoperative treatment for dyslipidemia
10.Preoperative treatment for diabetes mellitus
11.Postoperative sepsis
12.Postoperative renal failure requiring dialysis
13.Postoperative encephalopathy
14.Postoperative stroke
15.Postoperative myocardial infarction
16.Postoperative treatment for hyperglycemia
Died in the ICU
(n = 12)
Admitted to
the TCU
(n = 121)
Home Alive
(n = 4)
Died in the TCU
(n = 12)
Admitted to
an ECU
(n = 26)
Discharged to setting
other than
their own home
alive (n = 4)
Died in the ECU
(n = 19)
Alive in the ECU
(n = 48)
Died at Home
(n = 3)
Alive at Home
(n = 23)
All Patients
(n = 137)
(n = 26)
TCU: Transitional Care Unit
ECU: Extended Care Unit
Figure 1. Outcome for 137 patients who underwent tracheostomy after cardiovascular surgery.
Copyright © 2013 SciRes. WJCS
underwent tracheostomy within ten days of their initial Importantly, the early and late tracheostomy groups
uire tracheostomy for failure to wean
, other investigations have found that pa-
Table 3. Comparison of types of initial operations in early and late tracheostomy groups.
Category p Value
procedure had 90- and 180-day survivals of 74.4% and
74.4% respectively while patients who underwent tra-
cheostomy greater than ten days following their initial
procedure had 90- and 180-day survivals of 56.8% and
28.7% respectively (p < 0.047).
We next attempted to analyze whether or not the 33
patients who had early tracheostomy were similar in de-
mographic makeup to the 104 patients who had late tra-
cheostomy. The two groups were not significantly dif-
ferent in terms of gender, history of alcohol abuse, height,
weight, body mass index (BMI), or preoperative presence
of diabetes, hypertension, dyslipidemia, or renal failure.
The two groups did differ regarding smoker status as
50% of the late tracheostomy group had never smoked
compared to 39.4% of the early tracheostomy group (p =
0.034). The late tracheostomy group was also older than
the early tracheostomy group with 70.2% of the former
being greater than 69 years while only 54.6% of the latter
were older than 69 (p = 0.012). Additionally we analyzed
the type of initial surgical procedures that the patients
had undergone. The early and late tracheostomy groups
did not differ in type of initial operation (Table 3).
We then attempted to analyze whether the early versus
late tracheostomy groups differed in incidences of post-
operative complications. Table 4 summarizes these data.
The late tracheostomy group had a lower incidence of
postoperative stroke and a lower incidence of significant
hyperglycemia. The incidence of dialysis-dependent re-
nal failure, cardiac arrest, sepsis, myocardial infarction,
and encephalopathy was not different between the two
ho had undergone sternotomy for initial operation did
not have a significant difference in incidence of devel-
oping sternal wound complications following their tra-
cheostomy (none in each group).
4. Discussion
Patients who req
from a ventilator after major cardiovascular surgery are
known to have a markedly increased likelihood of mor-
tality with hospital mortality as high as 49% [4]. It has
been suggested that earlier tracheostomy might lead to
better survival [2] for these patients.
Early tracheostomy after sternotomy has been found to
be associated with increased risk of mediastinitis by one
investigator [5]. This finding was not confirmed by sub-
sequent studies [6-8]. We found no mediastinitis that
developed after tracheostomy in the early or late group
and were reassured that early tracheostomy does not
necessarily increased the incidence of mediastinitis after
As expected
nts who require tracheostomy after major cardiovas-
cular surgery suffer increased mortality if they have sig-
nificant preoperative renal failure or stroke morbidity
present [9]. Similarly, patients who experience worsen-
ing organ function after cardiovascular surgery are at
increased risk for death [10]. Although our late tracheo-
stomy group was older, they had fewer smokers and ex-
perienced a lower incidence of postoperative stroke and a
lower incidence of postoperative hyperglycemia. Despite
Level/Statistic Early Late Total
N 33) 104) (24.1% (75.9%137
Emergency Surgery < 0.999
CABG + 1
Valve Only 1
Type of Surgery
24 (72.7%) 77 (74.0%) 101
Yes 9 (27.3%) 27 (26.0%) 36
8 (24.2%) 28 (26.9%) 36
3 (9.1%) 27 (26.0%) 30
CABG + 2 8 (24.2%) 14 (13.5%) 22
AAA 5 (15.2%) 8 (7.7%) 13
TAA 4 (12.1%) 5 (4.8%) 9
3 (9.1%) 2 (11.5%)15
Valve × 2 2 (6.1%) 9 (8.7%) 11
Other 0 (0.0%) 1 (1.0%) 1
No 3 (69.7%) 0 (67.3%)93
Yes 10 (30.3%) 34 (32.7%) 44
Copyright © 2013 SciRes. WJCS
Table st-operatitions inus late tracheostomy groups.
Category p Value
4. Pove complica earl y ve r s
Level/Stat Early Late Total
N 33) 104%) (24.1% (75.9137
No 17 (51.5%) 47 (45.2%) 64
Renal Failure 0.554
Hyperglycemia 0.031
Cardiac Arrest 0.359
1 6
Myocardial Infarction 0.425
21 127
Stroke 0.013
Encephalopathy 0.617
Yes 16 (48.5%) 57 (54.8%) 73
No 27 (81.8%) 83 (79.8%) 110
Yes 6 (18.2%) 21 (20.2%) 27
No 18 (54.5%) 78 (75.0%) 96
Yes 15 (45.5%) 26 (25.0%) 41
No 30 (90.9%) 100 (96.2%) 130
Yes 3 (9.1%) 4 (3.8%) 7
No 9 (57.6%) 0 (57.7%) 79
Yes 14 (42.4%) 44 (42.3%) 58
No 32 (97.0%) 103 (99.0%) 135
Yes 1 (3.0%) 1 (1.0%) 2
No 7 (81.8%) 00 (96.2%)
Yes 6 (18.2%) 4 (3.8%) 10
No 28 (84.8%) 83 (7.8%) 111
Yes 5 (15.2%) 21 (20.2%) 26
a higher prevalence of these latter three risk factors, the
ire tracheostomy following cardiovas-
sis, data acquisition, manu-
[1] S. C. Murthy, A. C. Arroliga, P. A. Walts, et al., “Venti-
latory Dependar Surgery,” Jour-
early tracheostomy group was found to have improved
survival compared to the late tracheostomy group.
Limitations of this study include the fact that this was
a retrospective analysis with no attempt to randomize
patients to early or later tracheostomy. None of the tra-
cheostomies were performed with percutaneous tech-
niques, which actually serves to provide standardization
of technique throughout this study. Although the decision
to perform early or later tracheostomy was made based
upon subjective criteria of whether or not the patient
could be weaned from mechanical ventilation, there does
not seem to be a deliberate effort to perform later tra-
cheostomy on the sicker patients, as evidenced by the
lower incidence of postoperative complications (stroke
and hyperglycemia) in the late tracheostomy patients.
5. Conclusion
Patients who requ
cular surgery have significant risk of death with one year
mortality of 42%. A strategy of early tracheostomy,
within ten days of the initial cardiovascular procedure,
does not lead to an increase in risk of mediastinitis and
seems to be associated with better chance for survival.
6. Acknowledgements
J. M. Ladowski—data analy
H. E. Downey—data acquisition and analysis and ma-
nuscript preparation.
B. J. Ladowski—data acquisition and manuscript pre-
J. S. Ladowski—original concept, data analysis, manu-
script review, guarantor of paper.
ence after Cardiovascul
nal Thorac Cardiovasc Surgery, Vol. 134, No. 2, 2007,
pp. 484-490. doi:10.1016/j.jtcvs.2007.03.006
[2] J. Devarajan, A. Vydyanathan, M. Xu, et al., “Early Tra-
cheostomy Is Associated with Improved Outcome in Pa-
tients Who Require Prolonged Mechanical Ventilation
after Cardiac Surgery,” Journal of the American College
of Surgeons, Vol. 214, No. 6, 2012, pp. 1008-1016.
[3] A. O. Brook, C. S. Sherman, J. Moley, et al., “Early Ver-
sus Late Tracheostomy in Patients Who Require Pro-
longed Mechanical Ventilation,” American Journal of
Critical Care, Vol. 9, No. 5, 2001, pp. 352-359.
[4] H. Ballota, T. Kandil, L. Generali, et al., “Tracheostomy
after Cardiac Operation: In-Hospital and Long-Term Sur-
vival,” The Annals of Thoracic Surgery, Vol. 92, No. 2,
2011, pp. 528-533. doi:10.1016/j.athoracsur.2011.02.002
[5] J. J. Curtis, N. C. Clark, C. A. McKenney, et al., “Tra-
Copyright © 2013 SciRes. WJCS
cheostomy: A Risk Factor for Mediastinitis after Cardiac
Operations,” The Annals of Thoracic Surgery, Vol. 72,
No. 3, 2001, pp. 731-734.
[6] S. A. Stamenkovic, I. S. Morgan, D. R. Pontefract and C.
Campanella, “Is Early Trache
tients with Median Sternotomy Incisio
ostomy Safe in Cardi
ns?” The Annals of
ac Pa-
Thoracic Surgery, Vol. 69, No. 4, 2000, pp. 1152-1154.
[7] P. B. Rahmanian, D. H. Adams, J. G. Castillo, et al.,
“Tracheostomy Is Not a Risk Factor for Deep Sternal
Wound Infections after Cardiac Surgery,” The Annals of
Thoracic Surgery, Vol. 84, No. 6, 2007, pp. 1984-1991.
[8] S. D. Force, D. L. Miller, R. Petersen, et al., “Incidence
of Deep Sternal Wound Infections after Tracheostomy in
Cardiac Surgery Patients,” The Annals of Thoracic Sur-
gery, Vol. 80, No. 2, 2005, pp. 618-621.
[9] M. Engoran, N. F. Buderer and A. Zacharia
Survival and Health Status after Pro
s, “Long Term
longed Mechanical
Ventilation after Cardiac Surgery,” Critical Care Medi-
cine, Vol. 28, No. 8, 2000, pp. 2742-2749.
[10] P. A. Walts, S. C. Murthy, A. C. Arroliga,
cheostomy after Cardiovascular Surgery: A
et al., “Tra-
n Assessment
of Long-Term Outcome,” The Journal of Thoracic and
Cardiovascular Surgery, Vol. 131, No. 4, 2006, pp. 830-
837. doi:10.1016/j.jtcvs.2005.09.038
Copyright © 2013 SciRes. WJCS