Introduction
In recent years, percutaneous dilatational tracheostomy (PDT), performed according
to the technique by Ciaglia, has gained widespread acceptance for airway access in
patients requiring prolonged mechanical ventilation. Reasons for the preference for
PDT over surgical tracheostomy (ST) include its simplicity of performance, its rapid
placement, the lower rate of postoperative complications and better cosmetic results.
The traditional method of performing tracheostomies in critically ill patients requires
transfer from the intensive care unit to the operation theatre where a surgical team
performs a surgical tracheostomy. This involves a full dissection of the pretracheal
tissue and the insertion of the tracheostomy tube into the trachea under direct vision.
So far ST and PDT for prolonged mechanical ventilation have been compared in terms
of benefits and drawbacks, time of duration, indication and complication rate of the
respective procedures.
We studied a large number of patients managed with PDT or ST in order to compare the
factors related to the choice of intervention, duration of intervention, as well as
complication rate.
Methods
Patient population
The study population included all consecutive medical, surgical and neurological ICU
patients in the medical intensive care unit requiring elective tracheostomy. Medical
records were retrieved retrospectively between May 2003 and December 2008. The PDT
and ST were evenly distributed over the study period. After the indication for tracheostomy,
the choice of operation method was determined by patient factors. Typical factors
such as high BMI, short neck in obese patients, increased thyroid volume, previous
damage of upper airways, previous head and neck operations, affect the choice for
the ST method. The PDT was preferred in patients without the before mentioned factors.
Data recording
The following variables were recorded: ICU category (medical, surgical, neurological),
age, gender, comorbidities, reason for admission, reason for tracheostomy, duration
of intubation before tracheostomy, duration of the procedure, intraprocedural complications,
postprocedural complications, techniques used to perform the tracheostomy (patient
factors), physician (surgeon or anaesthetist), location (operating theatre or ICU),
indication (respiratory failure), acute physiological score (APS) and concurrent antibiotic
therapy.
Procedures
General approach
Cuffed tracheostomy tubes were used in both groups. Tube sizes varied from 8 to 10.
Cuff pressure was the minimal occluding pressure and was checked regularly. Cervical
anatomical features were evaluated for the presence of a palpable cricoid cartilage
at least 3 cm above the sternal angle on appropriate extension.
Both groups were examined bronchoscopically prior to the tracheostomy procedure. The
endotracheal tube was withdrawn into the subglottic larynx and the trachea was inspected
in order to evaluate the presence of any tracheal lesion and to aspirate bronchial
secretions. At the end of the procedure, another bronchoscopy was performed: the bronchoscope
was passed through the tracheostomy tube and then through the mouth in order to check
the stoma and the tube in order to evaluate the presence of endotracheal hemorrhage
and to aspirate bronchial secretions.
Tracheostomy wound dressings were changed once a day or as often as necessary to maintain
a dry stoma.
Surgical techniques
A team of thoracic surgeons performed all of the surgical tracheostomies in this study.
The surgical team consisted of the head of the department of thoracic surgery, one
surgical consultant and residents. Procedures performed with open surgical technique
included additionally an anesthetist for general anaesthesia and endotracheal tube
removal.
A modification of the surgical technique described by Grillo [1 ] was used.
A 2 cm horizontal incision was made. The superficial fascia and strap muscles were
retracted laterally to expose the cricoid cartilage and thyroid gland. The thyroid
was retracted either superiorly or the isthmus ligated depending on individual anatomy.
Hemostasis was archieved prior to entering in trachea. A horizontal incision was made
between the second and third tracheal rings and an inferiorly based tracheal flap
(reversed H-shaped incision) was sutured to the skin (Björk flap). The stumps of the
two divided rings were retracted with two clamps and a tracheostomy tube of appropriate
size was inserted into the trachea. The strap muscles were reapproached with one or
two 2 – 0 Vicryl sutures and the skin was loosely sutured with one stitch at both
sides of the tracheostomy tube. The endotracheal tube was removed, and the tracheostomy
tube inserted under direct visualization. After cuff inflation, confirmation of placement
was registered by auscultation of lung fields and bronchoscopy.
Nonsurgical approach
The technique according to Ciagla [2 ] was used in all patients. For procedures performed percutaneously, the Ciaglia Percutaneous
Tracheostomy introducer set was used in all cases of PDT. PDTs were done in the ICU,
bedside, under general anesthesia. All PDTs were performed under fiberoptic control
to monitor the tracheal puncture and the insertion of the dilators and the tracheostomy
tube. The patient was positioned with the neck extended and ventilated with 100 %
fraction of inspired oxygen. The endotracheal tube was withdrawn under direct vision
until the tip was at the level of the subglottic larynx. The cricoids and the first
tracheal rings were identified by palpation. A 1 – 1,5 cm skin incision was made and
the catheter-introducer needle was inserted into tracheal lumen between the first
and second ring under fiberoptic control. The J-tip wire and the guiding catheter
assembly were then used to introduce the percutaneous dilators until the required-size
tracheostomy tube could be inserted.
Outcomes
All procedures were timed from skin incision to insertion of the tracheostomy tube.
Perioperative (initiation of the procedure and 24-hours postoperatively) complications
were routinely recorded. Possible perioperative complications included hemorrhage
requiring exploration/ligation, tracheostomy tube displacement, pneumothorax, premature
extubation of the endotracheal tube with consequently respiratory failure, deterioration
in cardiopulmonary status requiring intervention, inability to complete PDT or conversion
to open surgical tracheostomy, death secondary to a complication of the tracheostomy.
All patients were examined by the investigators daily until intensive care discharge
to detect any complications. Postoperative complications were defined as those occurring
from postoperative day 1 to day 21 or decannulation (tracheostomy site infection).
Statistics
Comparison between groups was performed by means of the Chi-squared test for categorical
variables and by means of the t-test for continuous variables (or by nonparametric
Mann-Whitney-U test in case of not normally distributed data). All analyses were performed
using SPSS software (SPSS 19.0; SPSS Inc., Chicago, IL, USA). All tests of significance
were two-tailed, and α was set at 0.05.
Results
Patient Characteristics
The study population consisted in a total of n = 378 patients, including 112 women
(29.6 %) and 266 men (70.4 %); the average age was 61.0 ± 13.9 years. Further demographical
and preoperative clinical characteristics of the population, classified according
to the tracheostomy method, are shown in [
Table 1
].
Table 1
Demographic and preoperative, clinical characteristics.
Parameters
PDT (n = 209)
ST (n = 169)
p
Age [years]
59.9 ± 14.1
62.3 ± 13.5
0.101
Gender [n/%]
female
male
53/25.4
156/74,6
59/34.9
110/65.1
0.043
BMI [kg/m2 ]
23.5 ± 4.2
26.2 ± 5.5
< 0.001
Ventilated days, preoperative
10.1 ± 6.1
20.3 ± 51.2
0.431
Indications [n/%]
weaning
long term ventilation
158/75.6
51/24.4
105/62.1
64/37.9
0.005
Primary diagnosis [n/%]
cardiac
pulmonary
CNS
diabetes
neoplasia
polytrauma
12/5.7
120/57.4
64/30,6
1/0.5
0/0.0
12/5.7
18/10.7
76/45.0
64/37.9
0/0.0
2/1.2
9/5.3
0.065
Days in ICU
35.3 ± 23.5
25.8 ± 23.1
< 0.001
ICU [n/%]
none
interdisciplinary
thoracic surgery
general surgery
mixed
neurology
4/1.9
55/26.3
50/23.9
36/17.2
4/1.9
60/28.7
29/17.2
37/21.9
28/16.6
23/13.6
8/4.7
44/26.0
< 0.001
Previous operation(s)
no
yes
194/92.8
15/7.2
100/59.2
69/40.8
< 0.001
Data from a total number of 378 tracheostomies were evaluated; of these, 209 were
percutaneous dilatational tracheostomies (PDT) according to the Ciaglia [2 ] method and 169 open surgery tracheostomies (ST) according to the modified Grillo
[1 ] method.
In every case the indication for tracheostomy was the presence of respiratory failure
necessitating prolonged weaning or permanent mechanical ventilation.
Preoperative comparison of groups
Age. There was no significant difference in age (59.9 ± 14.1 years (range: 24.3 – 88.5)
vs . 62.3 ± 13.5 years (range: 17.8 – 86.7; p = 0.101).
Gender. Both groups consisted in the majority of male patients (75 % in the PDT and 65 %
in the ST group). The greater proportion of male patients in the PDT group compared
to the ST group was significant (p = 0.043).
BMI. The mean body mass index in the PDT group was 23.5 ± 4.2 kg/m2 , thus in the normal weight range (range: 15 – 36 kg/m2 ), whereas the ST group at 26.2 ± 5.5 kg/m2 (range: 17 – 42 kg/m2 ) corresponded to a classification as pre-obese (p < 0.001).
Indication. Weaning was the indication in 75.6 % of the PDT and 62.1 % of the ST patients (p = 0.005).
Primary Diagnosis. More than 80 % of the patients in both groups displayed pulmonary primary disease
or CNS disease. The most frequent condition in the remaining patients was cardiac
disease. Overall, differences between the groups were not significant (p = 0.065), however, pulmonary disease was more frequent in the PDT group.
Preoperative duration of ventilation. The preoperative duration of ventilation was not significantly different (10.1 ± 6.1
versus 20.3 ± 51.2 days, p = 0.431), however, distribution of ventilation duration
in the ST group was extremely wide as expressed by a high standard variation.
ICU stay. Duration of ICU stay in PDT patients was 35.3 ± 23.5 days (range: 0 – 156 days)
and significantly longer than that of ST patients (25.8 ± 23.1 days; range: 0 – 151
days). The mean difference was 9.5 days (p < 0.001).
Intensive Care Unit/Specialist Field. The breakdown according to specialist fields of the intensive care unit indicated
a significant difference between the groups (p < 0.001).
Previous operations. 40.8 % of the ST group of patients had one or more previous surgeries, compared to
only 7.2 % in the PDT group (p < 0.001).
Comparison of methods PDT vs. ST: perioperative and postoperative clinical parameters
The results of the comparison of peri- and postoperative parameters in both groups
are summarized in [
Table 2
].
Table 2
Method comparison PDT vs. ST: peri- and postoperative clinical parameters.
Parameter
PDT (n = 209)
ST (n = 169)
p
Duration of procedure [min]
18.2 ± 10.0
38.2 ± 14.2
< 0.001
Diameter of tracheal cannula [mm]
8.4 ± 0.5
9.1 ± 0.7
< 0.001
Days of ventilation, postoperative
14.7 ± 20.2
12.7 ± 18.7
0.152
Tracheostomy occlusion [n/%]
yes
no
unknown
1/0.5
207/99.0
1/0.5
15/8.9
146/86.4
8/4.7
< 0.001
Pathogens [n/%]
no
yes
86/41.1
123/58.9
94/55.6
75/44.4
0.005
Antimicrobial treatment [n/%]
no
yes
106/50.7
103/49.3
122/72.2
47/27.8
< 0.001
Complications [n/%]
no
yes
191/91.4
18/8.6
155/91.7
14/8.3
0.909
Surgeons [n/%][1 ]
head physician/senior physician
specialist in advanced training
junior doctor in advanced training
82/39.2
94/45.0
33/15.8
118/69.8
50/29.6
1/0.6
< 0.001
Number of revisions [n/%]
0
1
2
202/96.7
7/3.3
0/0.0
165/97.6
3/1.8
1/0.6
0.346
Fatalities [n/%]
no
yes
149/71.3
60/28.7
142/84.0
27/16.0
0.003
1 „none“ refers to patients who were transferred immediately postoperatively to the
neurological early rehabilitation ward.
Duration of the procedure. On average the PDT lasted 8.2 min (range: 2 – 60 min), the ST 38.2 min (range: 12 – 88
min). The mean group difference was 20 min (p < 0.001).
Diameter of the Tracheal cannula. In the case of PDT, significantly smaller tracheal cannula were used then in the
case of ST (8.4 ± 0.5 mm vs. 9.1 ± 0.7 mm; p < 0.001).
Airway infection. Investigation of tracheobronchial aspirates revealed higher rates of pathogens in
the pre-interventional bronchoscopy (59 % of the PDT patients and 44 % of the ST patients
p = 0.005).
Furthermore, the differentiation according to pathogen groups showed a significant
group difference (p = 0.019). An overview of the pathogen distribution is shown in [
Fig. 1
].
Antimicrobial treatment. It was administered in 49.3 % of the PDT patients compared to 27.8 % of the ST patients
(p < 0.001).
Fig. 1 Microbial pathogens in tracheobronchial samples. Investigation of tracheobronchial
aspirates revealed higher rates of pathogens in the pre-interventional bronchoscopy,
59 % of PDT patients and 44 % of the ST patients (p = 0,005).The differentiation according
to pathogen groups showed a significant group difference (p = 0.019); G +, gram positive
bacteria; G–, gram negative bacteria; aveb, anaerobic organism; nonfarm, non fermenter;
fungi,fungal infection, ST, surgical tracheostomy; PDT, percutaneous dilatational
tracheostomy.
Complications. Both groups showed almost the same complication rates of 8.6 % (PDT) and 8.3 % (ST)
(p = 0.909). However, the type of complications was different. Subglottic tracheal stenosis
was observed in 6 ST patients (3.6 %) but none in PDT patients (p = 0.006). On the other hand, 11 patients (5.3 %) of the PDT group displayed injury
to the tracheal posterior wall but none of the ST patients (p = 0.002) ([
Fig. 2
]).
Fig. 2 Intra- and postoperative complications. Complications were observed in 31 patients
from 378 (8.2 %). PDT n = 18 (8,6 %); ST: n = 14 (8,3 %). Both groups showed almost
the same complication rates (p = 0,909). The types of complications were different
in the groups. ST, surgical tracheostomy; PDT, percutaneous dilatational tracheostomy.
Operating surgeons. Over two thirds of the open tracheostomies were performed by the chief surgeon and
residents. Most of the percutaneous dilatational tracheostomies (45.0 %) were performed
by specialist doctors in advanced training ([
Fig. 3
]). The group difference was significant (p < 0.001).
Fig. 3 The operating team. In this study the 209 percutaneous dilatational tracheostomies
were performed by anesthesiologists and the 169 conventional surgical tracheostomies
by thoracic surgeons. Over 2/3 of the surgical tracheostomies were performed by the
chief surgeon and residents. Most of the percutaneous dilatational tracheostomies
(45,0 %) were performed by specialist doctors in advanced training. The group difference
was significant (p < 0,001).
Revisions. In 7 patients (3.3 %) of the PDT group and 3 patients (1.8 %) of the ST group revision
was indicated, in one further patient of the ST group even 2 revisions. The group
difference was not significant (p = 0.346).
Fatalities. 28.7 % and 16.0 %, respectively, of the patients died in the PDT and the ST group. The
higher fatality rate in the PDT group was significant (p = 0.003).
Discussion
Main Results
The main results of our study are the following: 1) patients with a high BMI and associated
difficulties predominantly underwent ST. The ST was more frequently performed by experienced
operators; 2) PDT is a faster and more straightforward procedure than ST; 3) a few
complications were observed and related to procedure complications in PDT and long-term
complications in ST.
Comparability
The two groups were similar in terms of age and underlying disease.
All ST- and PDT-procedures were performed by the same team of thoracic surgeons and
intensivists, respectively, each invariably using the same technique. Therefore, the
results obtained are not affected by complications related to the operative learning
curve. Nevertheless, in our study PDT was carried out on many occasions by doctors
undergoing specialist training and junior doctors; this was quite the opposite of
ST which in two thirds of cases was primarily been carried out by the head of department
and (senior) consultant. The ST procedure obviously was regarded as an increased-risk
procedure and hence was carried out by experienced surgeons.
Techniques
Our technique of ST involved gaining access to the airway through a reversed H-shaped
incision between the second and third tracheal ring so as to reduce the potential
risk related to tracheostomies performed at higher levels (laryngo-tracheal stenosis)
or at lower levels (trachea-innominate artery fistula). We preferred a reversed H-shaped
incision since it seems to ensure optimal healing of the stoma by preserving vascularization
of the two tracheal flaps which close up like the shutters of a window after decannulation.
Our technique of PDT involved gaining access to the airway between the first and second
tracheal ring to reduce any risk of damaging the cricoid cartilage and to prevent
bleeding from lesions to the thyroid isthmus. All PDTs were performed under endoscopic
guidance to monitor the tracheal puncture and the insertion of the dilators and the
tracheostomy tubes.
Factors related to the choice of intervention and operator
We observed that patients with a higher BMI, short neck in obese patients, increased
thyroid volume, previous damage of upper airways, previous head and neck operations,
predominantly underwent ST whereas PDT was preferred in patients without the before
mentioned factors.
Interestingly, ST procedures were performed more frequently by experienced operators.
This is comparable with the study of Melloni [3 ]. In this study he evaluated the surgical team, the technique and experience of each
operator. However, the complication rate in this study is higher than in ours. The
validity of the study is questionable because of the small number of patients. Other
literature does not address the learning curve.
Time of the procedure
The average time taken for the procedure was 18.2 min for PDT and 38.2 min for ST.
Comparable results concerning the significant time difference for the two groups were
arrived at in the study by Jackson [4 ]. A lot of studies evaluated procedure time and show likewise results.
Detection of tracheobronchial pathogens during procedure
Overall, there was a high rate of positive tracheobronchial samples prior to the procedure.
It was highest in the PDT group. This may be primarily explained by the higher rates
of patients with underlying pulmonary disease and on antimicrobial treatment. As regards
pathogen classes, gram-positive pathogens were predominantly found. All pathogens
were regarded as colonization and in fact, no infectious complications related to
any of the procedures could be observed.
Complication types and rates
The complication rates in our study were the same for both procedures (8.6 % PDT and
8.3 % ST). However, we observed a small number of intraoperative complications in
the PDT group and a postoperative long-term complication in the ST group. Six ST patients
developed a subglottic tracheal stenosis over time which did not occur with any of
the PDT patients. On the other hand, 11 patients in the PDT group experienced an injury
to the posterior wall of the trachea during the procedure. The risk for such complication
is related to the lack of infrastructure in the intensive care unit as compared with
the operating theatre. One should note the advantages of the operating theatre resulting
from the size of the working area, lighting and optimum patient positioning thanks
to the operating table functions which are not present with bedside PDT. The structural
deficiencies of the intensive care unit include limited possibilities for positioning
the patient owing to the soft mattress, and a working area which has limited space
and is poorly lit.
We didn’t observe any fracture of tracheal ring in both methods. We think, this is
an advantage of the bronchoscopic guided procedure. The potential risk of damaging
the cricoids during PDT must always be carefully considered, especially in elderly
patients with heavily calcified larynxes.
The fatality rates of our study albeit significantly different in both groups were
not related to the interventions, but caused by the underlying disease.
Several randomized clinical trials comparing PDT with ST have been performed, and
usually show, that both operative and short-term post-operative complications are
comparable.
Graham et al. in 1996 [5 ] retrospectively compared 31 patients who underwent surgical tracheostomy to 29 patients
who underwent dilatative tracheostomy in the ICU. Their study did not demonstrate
any significant outcome difference between the two groups. Holdgaard et al. [6 ] performed a prospective study evaluating short term complications and found statistically
significant decreases in procedure time from 15 to 11.5 minutes, decreases in minor
operative and postoperative bleeding and decrease in minor and major infections with
the PDT method.
Crofts et al. [7 ] performed a prospective randomized study comparing surgical tracheostomy with bedside
PDT found no statistically significant difference in morbidity between the two groups
and no morbidities in either group. They concluded that both methods could be performed
safely.
Practical advantages and disadvantages for PDT
One advantage of PDT is, that it can be carried out at any time in the intensive care
unit without having to draw on planned operating theatre capacities. However, the
above-mentioned disadvantages of carrying out the procedure in the intensive care
unit remain to be considered. Care of long-term ventilated patients is simpler with
ST in respect of tube-changing and caring for the tracheostoma. Changing the tube
on a narrow dilated tracheostoma often leads over the course of time to bleeding as
a result of the development of necrosis, lesions and increased granulation tissue
with the consequent risk of tracheal stenosis. With patients who had previously undergone
surgery in the neck area, the anatomical conditions are distorted due to scar tissue,
which means that a ST is indicated in order to avoid unexpected complications.
Strengths and limitations
The strength of our study is the large number of patients treated over time by a very
homogeneous operating team. Despite of being a retrospective study, all descriptive
and operating data were completely recorded on a routine basis. Nevertheless, this
was not a randomized study, and specific advantages of one method over the other could
not be systematically assessed. Furthermore, we did not specifically evaluate the
timing of the procedure (early versus late tracheostomy).
Conclusions
Both techniques applied in a large population proved to be safe. The advantages of
PDT included a more rapid procedure and a lack of long-term complications. On the
other hand, ST had comparable results. The choice of intervention was the patient
factor (obesity, increased thyroid volume, previous damage of upper airways). In our
opinion these factors will always dictate the choice of intervention. The results
show, that both techniques are adequate for tracheostomy and exist side by side without
replacing one another.
ST obviously requires higher skills and is more frequently performed by experienced
operators.
Abbreviations
Min:
minimum
Max:
maximum
MW:
mean value
ST:
surgical tracheostomy
PDT:
percutaneous dilatational tracheostomy
SD:
standard deviation
