Tracheostomy is one of the most common techniques used in the Intensive Care Unit (ICU). Although the procedure is assumed to be thousands of years old, modern tracheostomy was introduced with the description of surgical tracheostomy (ST) by Jackson in 1909.1 In that period, inflammatory or tumor processes causing airway obstruction were the main indication of the technique. With advances in the fields of surgery, critical care and positive pressure mechanical ventilation, tracheostomy received a new indication: prolonged mechanical ventilation. For years this was a feared technique, often associated with serious and event fatal complications; physicians therefore attempted to avoid the procedure, and patients remained intubated for weeks and even months.2 In 1985, Bishop et al.3 published an experimental study on the consequences of prolonged intubation, and that same year Ciaglia et al.4 described percutaneous dilatation tracheostomy (PDT), which proved simple, rapid and safe. Furthermore, the technique could be performed at the patient bedside, leading to a new way of viewing surgical access to the airway in the critically ill patient.

Five years after the description of PDT by Ciaglia et al.,4 Griggs et al.5 introduced tracheostomy using guide wire dilating forceps (GWDF) (Portex Limited, Hythe, Kent, United Kingdom). With this technique the tracheostoma is opened using blunt-tipped forceps (designed with a cannulated internal margin through which a metal guide wire can be inserted) previously advanced over the metal guide in the tracheal lumen. In 1998 the classical technique of Ciaglia was modified using a single dilator6 (Ciaglia Blue Rhino [CBR], Cook Critical Care, Bloomington, IN, USA). With respect to the original technique, this modification offered the advantage that the stoma is produced by a single dilatation, thereby avoiding the need for successive dilatations. Fantoni and Ripamonti7 in turn developed the translaryngeal technique (Mallinckrodt, Mirandola, Italy) – a complex system in which the stoma is made by exteriorizing a dilator from within the trachea, using a specially designed tracheostomy cannula and a rigid bronchoscope. In 2002, Frova and Quintel8 described the PercuTwist technique (Rüsch GmbH, Kernen, Germany), in which the tracheostoma is made using a rotation system. Another modification of the classical technique of Ciaglia was introduced in 2008 (Ciaglia Blue Dolphin [CBD], Cook Critical Care, Bloomington, IN, USA), involving dilatation performed by insufflating a balloon.9

The techniques most widely used in Spain in the early years of the last decade were the Griggs technique, the Ciaglia PDT, and the Ciaglia Blue Rhino.10 At that time the balloon dilatation technique had not been developed, and the PercuTwist procedure was being introduced. An additional 5 European surveys have evaluated the use of percutaneous tracheostomy (PT) in the ICU, and their results are shown in Fig. 1.11–15

Figure 1.

Use of percutaneous tracheostomy (PT) in European Intensive Care Units.

(0.2MB).

The indications of tracheostomy can be summarized as follows: (1) need for airway permeation; (2) prevention of laryngotracheal damage caused by prolonged intubation; and (3) maintenance of adequate tracheobronchial hygiene, facilitating the aspiration of secretions in patients with spontaneous secretion clearing problems.

The classical contraindications of PT have been obesity, a short neck, gout or neck deformities, coagulation disorders (INR>1.5) or thrombopenia (platelet count<50,000/mm3), children under 16 years of age, the need to establish an emergency surgical airway, probable or confirmed cervical spine damage, previous neck surgery or tracheostomy, infection in the surgical zone, and high oxygen and PEEP requirements.16–18

As experience with PT has increased, so has clinician confidence with the technique, while the restrictions in high risk patients have decreased. Nevertheless, it must be taken into account that young children, cervical injury and infection of the surgical zone remain absolute contraindications. Some risk groups and the pediatric population deserve special mention.

A number of reasons make it difficult to obtain a rigorous estimate of the perioperative complications of PT: (1) the baseline conditions of the patient may play a predisposing role; (2) the complications are directly conditioned to the degree of experience, which has been evaluated in only a few studies to date; (3) some of the more serious complications of PT can be avoided by adopting adequate safety measures, though these have not been widely used in the comparative studies; and (4) the definition of each event varies considerably among the different studies. In this regard, standardization of these aspects would allow us to obtain more objective and homogeneous information on the acute complications of the different types of PT. In this regard we propose the definitions of the complications of PT shown in Table 1, modified from Durbin.28

Surgical tracheostomy has been the gold standard for comparison with PT. Table 2 shows the results of the randomized studies29–42 that have compared ST and PT (predominantly PDT). Only two studies found the incidence of intra- or postoperative complications to be significantly greater with PDT than with ST.

Few studies have focused on the late complications (some of which are inherent to tracheostomy, regardless of the technique used) of PT, and the existing publications moreover involve small samples due to the high in-hospital and middle-term mortality among critical patients susceptible to PT. This makes it difficult to obtain representative samples for study (Table 3).43–54 The most feared complications are fistula of the innominate artery, tracheomalacia, tracheoesophageal fistula and tracheal stenosis. Other less serious complications are tracheocutaneous fistula, hoarseness and/or voice changes, or esthetic alterations. In relation to the more serious complications, the following should be noted: (1) tracheal stenosis causes no symptoms until 50–75% of the lumen is obliterated, and most cases therefore can be subclinical (Fig. 2); (2) the mortality rate associated with fistula of the innominate artery is close to 100%, and the preventive measures include not performing tracheostomy too low, in order to avoid resting of the cannula on the artery, and avoiding excessive movements on the part of the patient; (3) a tracheoesophageal fistula may be produced when performing PT, secondary to posterior tracheal wall damage, or on a delayed basis as a consequence of excessive balloon insufflation or posterior wall damage caused by continuous trauma produced by the tracheostomy cannula tip; and (4) tracheomalacia is a result of damage to the tracheal cartilage, causing the trachea to become distensible, and in acute ventilated patients the tracheomalacia zone must be passed by introducing a longer tracheostomy tube while waiting definitive treatment.

Figure 2.

Finding (in the context of protocolized follow-up) in an asymptomatic patient 6 months after percutaneous tracheostomy using the Ciaglia Blue Dolphin technique. (A) Spiral CAT reconstruction showing anteroposterior displacement of the tracheal cartilage, and (B) causing 15% stenosis of the tracheal lumen.

(0.12MB).

Each type or modality of PT has its own characteristics, and this has led to the conduction of comparative studies in search of the safest option. However, these studies have been heterogeneous and with small samples. Table 4 summarizes the prospective and randomized studies published to date.55–64

Fibrobronchoscopy is the most widely used safety method, though its application as a routine coadjuvant technique is heterogeneous. In effect, while in Spain fibrobronchoscopy is reportedly used in 16% of the ICUs,10 its application in other countries in our setting varies between 36%13 and 98%.15 The routine use of fibrobronchoscopy has been the subject of controversy, since the endoscopic guide produces a rise in airway pressure, hypoventilation and an increase in intracranial pressure – a situation that contraindicates its utilization in patients with acute neurological damage. Moreover, fibrobronchoscopy increases the cost of the procedure – causing some authors to consider its use only in cases with anatomical difficulties or in the context of learning curves.57 Nevertheless, the technique is simple and is recommended by some guides66 in the absence of contraindications, with a view to preventing serious complications such as false passage or damage to the posterior tracheal wall.

Capnography is useful for confirming the intratracheal location of the needle at the time of puncture. Mallick et al.67 have compared capnography with fibrobronchoscopy in 55 patients subjected to PT using the CBR technique. Twenty-nine patients were randomized to endoscopic control and 26 to capnographic control. The authors found capnography to be comparable to bronchoscopy in confirming the location of the needle in the tracheal lumen, and concluded that it may be an alternative to endoscopic guidance when the latter cannot be used or is contraindicated. The disadvantage of capnography is the lack of direct visualization of the precise position of the needle and metal guide within the trachea.

Cervical ultrasound can be useful for locating aberrant vascular structures that may cause bleeding. Kollig et al.68 performed PDT in 72 patients, initially using cervical ultrasound, followed by bronchoscopy. Based on the ultrasound findings, the puncture site had to be modified in 24% of the cases. There were no cases of damage to the posterior tracheal wall or bleeding. In this respect, ultrasound is a noninvasive procedure and may prove useful in selected patients.

The optimum timing of tracheostomy in the ventilated patient remains the subject of debate. The difficulty lies in defining prolonged mechanical ventilation and predicting which patients are going to need it.

The comparative studies of early versus delayed tracheostomy have been unable to draw firm conclusions. There has been heterogeneity in defining the term prolonged mechanical ventilation, and difficulty in defining which patients are candidates for such ventilation, as well as heterogeneity in the type of patients included, randomization defects, and consequently heterogeneity of the results obtained. Since 1984, a total of 8 randomized prospective studies have been published69–76 (Table 5). Two of them71,72 found no advantages with early tracheostomy versus the delayed technique. One of these studies71 reported important difficulties in performing randomization, while the other72 was carried out in critical burn patients. The results of the three most recent studies are contradictory. While Rumbak et al.74 found early tracheostomy (within the first 48h of mechanical ventilation) to involve a shorter duration of mechanical ventilation, a shorter stay in the ICU, and lesser pneumonia and mortality, Blot et al.75 observed no differences in terms of mortality, pneumonia, days without mechanical ventilation or stay in the ICU between the patients subjected to tracheostomy in the first four days (early) versus those in which the technique was carried out after more than 7 days (delayed). In the latest and also the largest randomized prospective trial, designed to assess the incidence of ventilation associated pneumonia among patients subjected to early (6–8 days) or delayed tracheostomy (13–15 days), Terragni et al.76 observed no differences between the two groups.

Three meta-analyses have addressed this issue. The study by Maziak et al.77 included two retrospective studies and three randomized trials with a total of 396 trauma patients with neutropenia and acute respiratory failure. The results do not suggest that the timing of tracheostomy influences the duration of mechanical ventilation or laryngeal damage. The limitations result from the heterogeneity of the included studies, heterogeneity in timing, a lack of information on how randomization was carried out, defective randomizations, and the inclusion of studies predating the PT era. In their meta-analysis, Griffiths et al.78 included 5 randomized studies comprising 406 clinical, surgical, trauma and burn patients. The timing of early tracheostomy was 0–7 days after the start of ventilation or admission to the ICU. The authors concluded that early tracheostomy results in a lesser duration of mechanical ventilation and a shorter stay in the ICU, but they recorded no differences in terms of pneumonia or mortality. Dunham and Ransom79 in turn included trauma patients in a meta-analysis comprising four retrospective studies and 5 randomized trials. The results showed no influence of early tracheostomy upon the development of pneumonia or patient mortality. Here again, the heterogeneity of the included studies did not allow the drawing of firm conclusions.

The literature suggests that early tracheostomy affords no advantages in terms of ventilation associated pneumonia or mortality, though it seems reasonable to use the technique in patients who are likely to require prolonged mechanical ventilation, in view of its benefits compared with intubation in terms of well-being, stay in the ICU and time on mechanical ventilation.

The term prolonged mechanical ventilation is characterized by important variability, and its definition is conditioned by the setting in which it is used. In our proposal of a prolonged ventilation probability model,80 we established a period of 7 days, on the grounds that this is the time limit for the possible appearance of laryngeal damage secondary to endotracheal intubation.3 Other studies have aimed to identify groups of patients with a high probability of requiring prolonged mechanical ventilation in both selected populations81–83 and in the general population.84 All of them have limitations, and no validated model is available. Therefore, considering the lack of tools that could help in the decision-making process, the timing of PT should be established on an individualized basis.

Tracheostomy is one of the most common techniques in the ICU, and PT is a widespread practice in our setting. The popularity of the technique is based on its safety, easy execution, and the fact that it can be performed at the patient bedside. Fibrobronchoscopy has gained acceptance and is now recommended as a safety technique. Although it is difficult to define the incidence of perioperative complications (which should be analyzed for each individual PT modality), it can be regarded as no greater than that associated with ST. Nevertheless, the incidence of late complications has been little studied to date, and surveys involving larger patient samples are therefore needed to clarify this aspect. The contraindications of PT have become relativized as confidence with the technique has increased, though it must be taken into account that children, neck injuries and infection of the surgical zone remain absolute contraindications. The optimum timing of tracheostomy in the ventilated patient remains the subject of debate. The results of randomized studies contrasting early versus delayed tracheostomy are heterogeneous. Consequently, it can only be affirmed that the timing of PT should be established on an individualized basis.

The authors state that they have no conflicts of interest.