Discussion
The main finding of the current revision is that in the scientific literature there are no studies with high levels of evidence, such as randomized clinical trials, on decannulation protocols in pediatrics. Nor analytic observational studies on cohorts of patients, where decannulation protocols are evaluated in a prospective way. Most studies were retrospective studies and case series, with limited numbers of patients.
The study of Wirtz et al(14)  evaluated a decannulation protocol whose key point was to restrict the used resources. This approach makes it unique amongst the rest of the studies. The patients who were included had stable lung function with no breathing obstruction and with no less than 2 months without ventilatory support. The specific protocol consisted in laryngoscopy and bronchoscopy, prior to decannulation, with the aim of testing the respiratory tract. The airways are evaluated through a transnasal approached flexible bronchoscopy and the tracheostomy cannula is taken away in the operating room if the airways are considered to be eligible for decannulation.  Unless any complications occur, the patient is discharged the next day. No occlusion or cannula reduction, nor polysomnography are performed. According to this study, the spontaneous ventilation during endoscopy is an important factor, since it allows the testing for obstruction or dynamic collapse. The authors believe this to be a higher ranked test, over daily routine capping, for decannulation processes suggested in other protocols.
The implementation of the reduction of the tracheostomy tube diameter and its occlusion are common in most protocols. However, its use is not universal nor standardized.
One of the arguments supporting this procedure is that it may not lead only to a successful decannulation, but also to a better adjustment to the child to the changing physiology of their breathing tract. The claims against the use of this protocol on a daily basis is that the reduction in the size of the lumen may result in cannula obstruction, due to its small diameter.  During the plugging, the airway cross-sectional area is reduced to such a level that those patients who would not tolerate it, would tolerate decannulation.
Regarding polysomnography, in 1996, Tunkel et al,(17)published a study on the utility of polysomnography (PSG) during the assessment when planning decannulation. PSG provides objective data on the dynamic factors influencing the upper airways permeability when the pharyngeal muscle tone is lowered to its top. The authors reached the conclusion that the sleep studies close to the normal ranges of PSG are correlated to a successful decannulation.
The existing literature is composed of retrospective revisions and case series, and there are many discrepancies when it comes to define what a “favorable” PSG is, for determining the tracheostomy tube extraction.(18-19)
Currently, the PSG role has expanded amongst pediatric patients with obstructive apneas and central apneas, neuromuscular disorders as well as respiratory support dependent patients. Gurbani et al,(18) in the Cincinnati Children’s Hospital Medical Center (CCHMC) studied the breathing and sleeping parameters associated with a successful decannulation, and the agreement between microlaryngoscopy, bronchoscopy(MLB) and polysomnography(PSG).
Microlaryngoscopy is the gold standard for the anatomy evaluation of the airways prior to decannulation. According to this study, certain PSG parameters which include the obstructive index (OI), apnea/hypopnea index (AHI) and hypoventilation, with a favorable MLB, seem to be good predictors for successful decannulation in patients with complex airways. They concluded that the combined use of MLB and PSG rises the predictability of successful decannulation in these children, above any other factor on its own. So, PSG must be considered a crucial part during the evaluation process for decannulation, especially amongst children with important breathing airways problems. Robinson et al,(19) also concluded that PSG may be a complementary trial for decannulation preparation. In the study of Lee  et al,(10) the authors studied the utility of polysomnography with occluded tracheostomy tube, as a complement for endoscopy, for predicting the decannulation outcome.
In a retrospective review of the patients who had been tested with a PSG before the test, 30 children were included. 26 out of those 30 had a successful decannulation and 4 of them failed to finish the procedure. The predictive factors of the decannulation outcome were the index of apnea/hypopnea and desaturation events. They concluded that it is a useful and objective tool to be used as a complement of endoscopy and the evaluation of the functional airways.
Moreover, the anxiety generated by the decannulation process is something that should be also taken into account. The perspective of decannulation may cause significant anxiety on the patient and his family.
Previously, in other studies such as Black et al,(22)and Waddell et al,(23) panic was mentioned as a cause of failed decannulation. The pediatric decannulation protocol suggested in Great Ormond Street Hospital in United Kingdom (UK) includes the prior evaluation and mandatory psychological preparation.
Finally, given the evidence so far it is not possible to set as the ultimate truth any of the measures or variables which were part of these protocols, although it would be possible to give certain recommendations.
Mitchell(16), who was previously mentioned in this article, claimed that no consensus was reached regarding many topics, such as the tube change frequency in a mature tracheostomy, or the size of the cannula for children during nocturnal ventilation.
Considering the limitations of the existing studies it is possible to mention as key points the study design, mostly retrospective, case series and with a limited number of recruited patients.
When this review was finished, a multicentric study was published by Schweiger and col (41)evaluating the factors associated to a successful decannulation amongst the pediatric population. The study was placed in 4 high complexity hospital centers in Brazil, in the otorhinolaryngology department. Medical registers of under-18 tracheotomized patients were analyzed in a retrospective approach. The most common cause was post intubation laryngitis.
The outcomes showed that age and post intubation laryngitis were associated with a higher probability of successful decannulation.  On the other hand, chronic neurological and breathing comorbidities were linked to lower decannulation chances.
The outcomes of the decannulation predictors are also important since they may help in the planning of the decannulation process. And, more importantly, they may be useful for advising health workers, the patient’s family or tutor to have realistic expectations about the outcome.