A systematic review on tracheostomy decannulation: a proposal of a quantitative semiquantitative clinical score

In the following chapter we will discuss the main physiological implications for breathing after tracheostomy. In scientific literature, authors usually compare tracheostomy to endotracheal intubation. In our purpose, we intended to compare changes during tracheostomy to spontaneous breathing and the most relevant topics are:

Airflow resistance - Resistance of anatomically normal airways is firstly conditioned by upper airways (up to 80% during nose breathing and 50% during mouth breathing). Therefore, tracheostomy should virtually reduce flow resistance bypassing the respiratory upper airways. Actually, data from scientific literature state the contrary, as the presence of tracheotomy tube, during spontaneous breathing, reduces airway radius leading to increased flow resistance and, of course, increased work of breathing (WOB) [4, 5].

This is according to Poiseuille equation, as the resistance to gas flow through a tube varies inversely with the internal diameter of the tube (in particular, to the 4th power of the radius of the tube when flow is laminar).

Secondly, secretions (stimulated even more by the presence of the tube) still play an important role: adhering to the inner lumen of the tube, they can lead to significant luminal narrowing, according to Wilson et al. [6], 15% of tubes are shrunk three sizes; other experiences from literature ultimately come to the same conclusions [7, 8].

It has been supposed that the inner diameter of a smaller tracheostomy tube may induce increased flow resistance and WOB, if compared to spontaneous breathing. The first data from in vivo studies appeared recently by Valentini et al. trying to define the effect of smaller tracheostomy tube sizes on diaphragm effort. They recorded diaphragm pressure- time product per min (PTPdi/min), tension-time index of the diaphragm (TTdi), and the ratio of respiratory rate to tidal volume (f/VT) for 2 different tracheotomy tube, with an inner diameters 8 mm and 6.5 mm. The use of a smaller diameter resulted in an increase of diaphragmatic effort, decrease of VT and an increase of intrinsic PEEP. In conclusion, the authors assessed that, especially in hard to wean tracheostomized patients, the use of a tracheostomy tube of small size can lead to alterations of some weanibility parameters, otherwise normal with greater size tubes [9].

In addition, a single physiologic study by Criner et al. reported that airways resistance and work of breathing resulted higher with the tube in place, during spontaneous breathing, when compared to breathing after decannulation [10].

Humidification and heating - Normally upper airways play a fundamental function of heating and humidification of inspired air. In tracheostomized patients, the air bypasses the nasopharyngeal cavity and enters directly the tracheobronchial tree; a system of artificial humidification is needed. In the absence of adequate humidification, the epithelium of the trachea is involved in a gradual inflammatory process resulting in squamous metaplasia and, consequentially , impairment of ciliary function and increased risk of respiratory infection [11].

While scientific literature has better investigated tracheotomy effects on respiratory physiology, data about respiratory mechanisms after decannulation are almost totally lacking. Chadda et al. studied respiratory parameters in nine neuromuscular patients after that the upper airways were confirmed to be free of obstructions by fiberoptic bronchoscopy. In this experience, decannulation resulted in an increase of the tidal volume and carbon dioxide partial pressure due to an increase of the dead space and work of breathing [12].

However, the argument is still controversial. In a study of few years ago, Dellweg et al. did not find a significant difference in WOB in favour of tracheostomy when compared to mouth breathing. Decannulation increases or decreases airways resistance depending from case to case and, in particular, from the morphology of the upper airways [13].

To our knowledge, there are only few papers on physiological modifications after decannulation; experts have differing opinion on this topic and so further research and trials are desirable.

Place a tracheostomy enables discharging the patient from ICU to a rehabilitation unit [14]. However, despite decannulation is not risk-free, there is evidence of benefits for tracheostomy tube removal. The tracheostomy tube may cause inflammation and stenosis or excessive cough and may impair swallowing by preventing the physiological trachea’s elevation against the epiglottis in order to prevent aspiration of food or secretions [15].

Furthermore, upper airways are excluded from breathing. At last, in most cases tracheostomized patients are unable to speak; aphonia worsens patient’s quality of life and slows down the recovery process, often leading to anxiety and depression.

A number of clinically important early and late complications have been evaluated, including granulation tissue, tracheal stenosis, tracheomalacia, tracheo-esophageal fistula, ventilator-associated pneumonia and aspiration. According to literature, chronic tracheostomy in severe Chronic Obstructive Pulmonary Disease (COPD) patients is associated with a higher frequency of exacerbations requiring antibiotic treatment [16]. The clinical relevance of these complications may lead to death and most of patients weaned from mechanical ventilation through tracheostomy should undergo early decannulation.

As it is hard to schematize the approach for decannulation, we decided to perform the following revision of literature, with the aim of supporting or retracting current statements. Which are then current issues in managing tracheostomy? How does evidence-based medicine handle the process of weaning from tracheostomy tube? Nowadays clinical habit derives from physiopathological knowledge, from personal experience and from professional practice: there is limited evidence in literature regarding specifically decannulation processes. Little is known about how clinicians decide to decannulate patients. Which are the criteria for choice?

A literature search has been performed using the Cochrane Central Register of Controlled Trials (CENTRAL); MEDLINE via Ovid SP; EMBASE via Ovid SP; EBSCO. Research included, but was not limited to, 3 keywords (tracheostomy, decannulation, weaning). Abstracts were independently reviewed: for those studies fitting the inclusion criteria on the basis of the title and abstract, full-text was achieved. Reference lists were also examined for any additional relevant studies not identified through the former search. We included studies published from January 1, 1995 until March 31, 2014. Such studies examined a population more than 18 years old, who received tracheostomy for any clinical reason, excluded neuromuscular diseases, and hospitalized in any ward (ICU, rehabilitation wards, etc.); any sort of review and expert opinion has been excluded by our survey. Including English language restriction was applied. A flow chart diagram of the search strategy and study selection is provided in Figure 1.

The parameters considered in literature and also reported in Table 1 are worth of deeper analysis.

First of all, we identified two different types of parameters: Quantitative parameters: objective, described by means of numerical values and provided with cut-off (such as, ability to tolerate tube capping > 24h); Semi-quantitative parameters: objective, but however not be easily described by mean of a numerical value (as, for instance, swallowing function). Our purpose is to suggest a clinical score including all parameters – both quantitative and semiquantitative –considered by available scientific literature in order to evaluate the feasibility of tracheostomy tube removal.

In our idea, objective quantitative parameters shall be taken into greater account in decisional process. Other parameters should instead be evaluated, when it is possible, according to a binary system (e.g. dysphagia yes/no).

Our intent is to give a high score (e.g. 20 pts) to patients fitting quantitative objective parameters. When such requirements are missing, the score of each single parameter will be 0 pts. This choice is to underline the fundamental importance we assign to those parameters, being the most frequent measurable ones taken in consideration by current scientific literature; in lack of those parameters, scientific evidence seems to predict a negative outcome for decannulation.

In the second place, our clinical score must comprehend as well semi-quantitative objective clinical parameters (e.g. dysphagia and secretions) and subjective parameters (e.g. clinician experience). In our systematic review the utility of bronchoscopy before and during decannulation appears as an important tool that should be considered in clinical practice.

Most of these parameters respond to the need for a straightforward and binary evaluation (yes/no = 5pts/0pts); a smaller numerical value has been assigned to them in order to lower their weight on the overall score versus objective parameters.

With reference to the two different types of parameters taken into account, we suggest to name the proposed score “QsQ score”, that is “Quantitative semi Quantitative score”. In Table 2 such approach is summarized.

We suggest an hypothetical score, that requires discussion and a prospective validation study. For a practical use we will name objective quantitative parameters ‘major criteria’, and semi-quantitative or subjective parameters ‘minor criteria’. If all main criteria are satisfied, regardless of minor criteria, decannulation with high probability of positive outcome can be assumed.

If only one of the two major criteria is satisfied, a careful evaluation of minor criteria should be required, assuming a good probability of positive outcome when the majority of minor criteria is satisfied. The same probability category reported above could be applied if, in lack of major criteria, all of minor criteria are satisfied. Finally, if none of the major criteria and less than three minor criteria are satisfied, a low probability of positive outcome can be assumed.