Percutaneous tracheostomy

Abstract

Background

Percutaneous tracheostomy (PT) has gained an increasing acceptanceas an alternative to the conventional surgical tracheostomy (ST). Inexperienced hands, and with proper patient selection, it is safe, easy andquick.

Complications

Perioperative complications are comparable with those of ST andthese are mostly minor. An important advantage of PT over ST is that there isno need to move a critically ill patient to the operating room and the rate ofstomal infection is very low. Although data on late complications of PT are notyet sufficient, available reports show a favourable result.

Techniques

Ciaglia's method is the most commonly applied, but no studyhas shown superiority of any of the percutaneous techniques described. Thedecision on which method to use should solely be made depending on the clinicalsituation and the experience of the operator. The learning curve demandscaution, attention to detail and adequate experience on the part of theintensive care physician. Although PT is unfortunately declared'easy', it must be left in the hands of experienced physicians toavoid unnecessary complications, and the risk of overimplementation should bekept in mind.

History

Tracheostomy is one of the oldest surgical procedures. The origin of percutaneous tracheostomy (PT) is not certain, although the Italian surgeon Sanctorius was probably the first to describe the technique in the 16th century. Sheldon et al. [1] used the term percutaneous tracheotomy in 1955 and described the method as an alternative to the surgical route. Toye and Weinstein [2,3] introduced the technique using the Seldinger guidewire and it has since been refined with various modifications [4,5,6,7]. The percutaneous dilatational tracheostomy (PDT) introduced by Ciaglia et al. [4] in 1985, which involves progressive dilatation with blunt-tipped dilators, is the most frequently used and evaluated in the literature [8,9,10,11,12,13,14,15,16,17,18]. In 1989, Schachner et al. [5] introduced a rapid PT technique, Rapitrac, which did not get considerable acceptance because of complications associated with, and reservations towards, the sharp edges of the dilating forceps. In 1990, Griggs et al. [6] reported on a PT technique using a modified Howard-Kelly forceps with a blunt edge and Fantoni et al. [7] reported the translaryngeal tracheostomy technique (TLT).

Indications and timing

Tracheostomy is indicated for prolonged ventilatory support, long-term airway maintenance, and to prevent the complications of long-term translaryngeal intubation. It also eases patient care and the process of weaning from mechanical ventilation. The timing of tracheostomy is still controversial [19,20,21,22,23]. In 1989, a Consensus Conference on Artificial Airways in Patients Receiving Mechanical Ventilation [21] recommended translaryngeal intubation for an anticipated need of up to 10days and a tracheostomy if an artificial airway for more than 21days is anticipated. However, the decision on the time point of tracheostomy should be made on an individual basis and should depend on prognostic evaluations and not on 'calendar watching' [24]. Although early tracheostomy is preferred by some authors [25,26,27,28], there is no adequate comparative study as to the advantages of this approach [23].

Techniques

Currently, the technique by Ciaglia et al. [4] (Cook^® Critical Care, Bjaeverskov, Denmark) is the most widely applied, followed by that of Griggs et al. [6] (Portex^®, Smiths Industries Medical Systems, Hythe, Kent, UK). The TLT method (Mallinckrodt Medical^®, Mirandola, Italy), with the tracheal cannula being inserted through the translaryngeal route, has been reported particularly in Italy and is now under evaluation in several clinics across Europe. The key procedure in all these methods is needle puncture of the trachea and insertion of the Seldinger guidewire. Though the procedure may be carried out under local anesthesia, experience shows that it is safer to perform this under adequate analgosedation and, if necessary, neuromuscular relaxation. The latter is particularly important to suppress the cough reflex which may increase the risk of damage to the posterior tracheal wall with either the puncture needle or dilators. Additionally, infiltration of the proposed site with lidocaine/epinephrine solution may be useful to reduce the risk of bleeding. Hyperextension of the neck for anterior displacement of the trachea is crucial. Therefore, PT is not recommended when manipulation of the cervical spine is contraindicated. There is no study on the implementation of any of these techniques in emergency situations. Furthermore, their use should be weighed carefully in patients with a large goiter, recent neck surgery or inflammatory changes at the proposed site of skin incision.

Ciaglia's technique (percutaneous dilatationaltracheostomy)

Serial dilatation of the trachea is the hallmark of this technique. Originally, Ciaglia et al. [4] described the point of entry to be subcricoidal; however, this was found to be too high, with a risk of subglottic stenosis [29,30,31]. Therefore, the preferred site of entry is now between the first and the second or the second and third tracheal rings [9,12,17,18,31]. Initial skin incision and blunt preparation of the pretracheal tissue may be helpful to identify the tracheal rings, thus avoiding either too high or too low tracheal puncture. After dilatation with the maximal available dilator, a tracheal cannula (inner diameter up to 9mm) can be inserted whilst mounted on a corresponding dilator.

The routine use of bronchoscopy during PT, apart from TLT, is not yet settled. There are reports of lower rates of acute complications under endoscopic guidance [8,13,32]. However, there is no adequate controlled study showing that endoscopic-guided tracheostomy is superior to the 'blind' one. Furthermore, the significance of operator experience, anatomical consideration and individualization in decision making is not discussed in these studies. Additionally, resultant hypercarbia should be considered when choosing endoscopic-guided PT for the critically ill and/or patients with head injuries [33]. However, endoscopic guidance plays a decisive role in the training of physicians, during PT on patients with a difficult anatomy, and to remove aspirated blood.

Another controversial issue is whether bronchoscopy can better define the exact location of tracheal puncture. A cadaver study by Dexter [34] showed that correct 'blind' puncture in the intended intercartilaginous space was achieved in only 45% of cases. Another post-mortem study [35] reported accurate placement of the tracheal cannulas in 76% of cases. Until now, studies using bronchoscopic guidance during PDT have concentrated on the confirmation of the initial airway puncture. Therefore, a controlled study is necessary to settle these issues. In any case, a bronchoscope must be readily available in case of an emergency.

The average time required to perform the dilatational tracheostomy is 10-15min [12,14,17,18,36]. Although Ciaglia's technique has already been carried out successfully on children [37], there are still reservations on its use in this age group due to the marked elasticity of the tracheal tissue.

Griggs' technique

The distinctive feature of this technique is the use of a pair of modified Howard-Kelly forceps for blunt dilatation of the pretracheal and intercartilaginous tissue after insertion of the guidewire into the trachea and skin incision. The average time required for a tracheostomy is about 5min, but it can also be accomplished in about 1min [38,39] (unpublished personal observation). Applying this method on patients with a short and/or thick neck may be difficult, if not dangerous, particularly while attempting to perform intercartilaginous dilatation. Although none of these percutaneous techniques have been evaluated for emergency use, this method could possibly be applied in such a situation following proper patient selection.

Translaryngeal tracheostomy (Fantoni's technique)

For translaryngeal tracheostomy, in contrast to the other techniques, the initial puncture of the trachea is carried out with the needle directed cranially and the tracheal cannula inserted with a pull-through technique along the orotracheal route. The cannula is then rotated downward using a plastic obturator. The main advantage of TLT is that there is hardly any skin incision required, and therefore practically no bleeding is observed. Furthermore, there is minimum pressure on the trachea and pretracheal tissue. It has also been successfully carried out on infants and children [7]. It may be particularly useful in patients with bleeding diathesis and goiter. The procedure can be carried out under endoscopic guidance only, and rotating the tracheal cannula downward may pose a problem, thus demanding more experience. There is also an apnea phase of about 60-90s during the procedure [7,40]; this technique should therefore be contraindicated in patients with severe respiratory insufficiency requiring extreme forms of mechanical ventilation (high positive end-expiratory pressure, high inspiratory oxygen concentration). Additionally, since the tracheal cannula is pulled through the oropharynx, the significance of contamination of the cannula with oropharyngeal bacterial flora in the development of (aspiration) pneumonia or other airway infections should be investigated.

Complications

The advantages of PT are that it is a simple, fast, and minimally invasive bedside procedure leading to less stress to the patient compared with surgical tracheostomy (ST). Although there are differences between authors as to what is considered worth reporting, the rate of perioperative complications for Ciaglia's technique is between 4.1% and 12%, the majority of these being minor with the rate decreasing with experience [8,9,13,14,15,16,17,18]. Bleeding is the most common perioperative complication (Table 1). A rare and life threatening complication of tracheostomy is a tracheo-innominate artery fistula which has also been reported after PDT [10,41]. This may occur with a tracheostomy below the third or fourth tracheal ring. Another important complication is damage to the posterior tracheal wall due to the puncture needle or dilators, which is usually minimal but may have serious consequences in a few cases. As in any other invasive procedure, the rate of complications depends not only on the inherent problems of the technique but also on the experience of the operating physician [14,17], as well as on a proper patient selection. Our prospective observation on 234 PDTs demonstrates the learning curve which should be taken into consideration when discussing complication rates (Fig. 1; unpublished data).

Mortality due to PT is rare and this is reported to be due to bleeding [10,15,18], bronchospasm [14], cardiac arrhythmia [8], and premature decannulation [42]. Stomal infection is rare (0-3.3%) and mostly minor, since the stoma fits snugly around the cannula and there is hardly any tissue devitalization [8,9,11,13,14,15,16,17,36,38,43].

Figures on late complications after decannulation, including tracheal stenosis, hoarseness and tracheomalacia, are difficult to analyze since the criteria applied by the authors differ and the diagnostic intensity varies. Ciaglia and Graniero [9] reported only one case of mild voice change among 52 decannulated patients, whereas Hill et al. [14] observed symptomatic tracheal stenosis in 3.7%. Marx et al. [16] reported two cases of tracheal stenosis that required tracheoplasty among their 254 patients. In a detailed analysis using tomography of the trachea on 54 decannulated patients, van Heurn et al. [30] reported a tracheal stenosis of 10-25% in 11 patients, between 25-50% in two patients, and more than 50% in one patient. In 41 patients examined at least 6 months after decannulation, Law et al. [44] found a tracheal stenosis of 10% in four asymptomatic patients by means of laryngotracheoscopy and spirometry. Walz et al. [18] also reported a tracheal stenosis of at least 10% in about 40% of their follow-up patients.

Data for the Griggs' technique are few. The rate of perioperative complications is about 4% [38,39]. Late complication, particularly tracheal stenosis, was observed by Griggs et al. [38] in one out of 153 cases.

For TLT, Fantoni and Ripamonti [7] reported bleeding in 2.8%, although this was attributed to ample skin incision in the initial experimentation phase. Another prospective study on a small group of patients also showed only minimal complications [40]. No late complication was observed by Fantoni and Ripamonti [7] in nine autopsies and 20 adults after decannulation. However, the duration of cannulation was not mentioned. An adequate comparative study is necessary to investigate whether the rate of late complications is indeed lower than that for the other percutaneous techniques.

Pathological studies on the trachea after PT are scarce. In an autopsy study of 12 cases with PDT, van Heurn et al. [45] reported a fracture of one or more tracheal rings in 11 cases, two of whom had a fracture of the cricoid. Destruction and necrosis of one or more tracheal rings was also observed in those cases cannulated for more than 10 days. Transverse rupture of the anterior tracheal wall with or without fracture of neighboring rings is considered as the typical lesion following PDT by Walz and Schmidt [35]. As these authors have already pointed out, certain complications, particularly too high tracheostomy and ring fracture, can be avoided by attention to detail during the procedure. Exact palpation of the tracheal rings is crucial before starting the percutaneous technique, and this can be improved by blunt dissection of the pretracheal tissue when using the Cook and Portex kits and, in case of difficult anatomy, by applying endoscopic guidance. Furthermore, too much pressure on the trachea during cannulation must be avoided.

Figure 1

The learning curve: perioperative complications duringpercutaneous dilatational tracheostomy.

Table 1 Perioperative complications (%) during percutaneous dilatational tracheostomy

Percutaneous tracheostomy versus surgical tracheostomy

Comparing PT with historical data of complications for ST is erroneous and may give a biased picture. Furthermore, due to different definitions of complications used by authors, these figures should be interpreted cautiously. Nevertheless, comparative studies have shown that PT has certain advantages [36,38]. Firstly, it can be performed immediately once the decision is made and few personnel are needed. In contrast, ST requires more organization and, if it is to be done in the operating room, time scheduling. ST involves the transport of mostly critically ill patients out of the intensive care unit to the operating room, which is often a complex co-ordinated effort and may endanger the patient. The time required for PT is about one-quarter that for the surgical route [36,38], which implies less stress to the patient and better use of available resources.

The rate of perioperative complications for ST does not generally differ from that for PT. A prospective study by Stock et al. [20] revealed a rate of 6.0%. Two large retrospective studies reported rates between 5.4% and 6.3% for acute complications [46,47]. In a prospective comparison of Griggs' technique with standard ST, Griggs et al. [38] reported rates of 3.9% and 8.1%, respectively, for perioperative complications.

However, the rate of stomal infection for ST is significantly higher (6.8-22.2%) [36,38,48], which has been associated with the larger wound surface and tissue devitalization. Late complications of ST, particularly tracheal stenosis, are reported to be low, ranging between 0-1.1% [46,48,49,50].

Although cost analysis between PT and ST is not easy because of varying reimbursement systems and hospital structures, available studies show that PT is considerably cheaper than the surgical route [8,13,14,36,41,51]. It is common sense that if fewer personnel and no operating room time are required, and the patient need not be moved, then the overall cost of PT has to be lower than that of ST.

Conclusion

Percutaneous tracheostomy has already replaced the surgical route in several intensive care units and it is indeed the procedure of choice in the majority of cases. This is attributable to the fact that, in experienced hands, it is safe, easy and quick, and there is no need to move the patient to the operating room. Perioperative complications are at least comparable with those of surgical tracheostomy and most of them are minor. With proper patient selection, operator experience and attention to detail, complication rates can be reduced that may have an influence on late complications.

An important advantage of PT over the surgical route is the very low rate of stomal infection. Several reports have also shown that PT is cheaper than ST, which is of course important at a time when resources are limited. Despite all the virtues of the percutaneous technique, the role of ST in cases with contraindications for PT, difficult anatomies and failed PTs remains unchallenged. The decision on which method to use should solely be made depending on the clinical situation and the experience of the operator. The fact that a technique is declared 'easy' should not lead to an attitude that every physician may get a chance to try it. PT must be left in the hands of physicians with enough experience, although at the moment there are no criteria to define this quality.

The discussion on the routine use of bronchoscopy during PT is not yet settled. Although this is a requirement during TLT, there is no adequate controlled study on the superiority of routine endoscopic guidance during dilatational tracheostomy. However, it is indispensable for training purposes and during PT on patients with difficult anatomy. Moreover, a bronchoscope must be at hand during PT in case an emergency situation arises.

No study has shown superiority of any of the three methods reported, although TLT is still under evaluation and not widely in use. These techniques must be judged by their safety, ease of performance and long-term effects, not merely by the rapidity with which they can be performed. Finally, in our enthusiasm to embrace new techniques, we must not get lured into their overimplementation.