Determinants of tracheostomy decannulation: an international survey

Tracheostomy is probably the most common surgical procedure performed on critically ill patients [1]. Approximately 10% of mechanically ventilated critically ill patients receive a tracheostomy to facilitate prolonged airway and ventilatory support [2‐5]. The recent development of percutaneous dilational tracheostomy techniques has made tracheostomy a routine procedure commonly performed at the bedside in the intensive care unit (ICU) [6]. Prolonged tracheostomy tube placement may expose patients to an increased risk of late complications, including tracheal stenoses, bleeding, fistulas, infections, and aspiration [7‐11]. Psychological implications are profound with patients experiencing reduced body image perceptions and life satisfaction [12]. Removing a tracheostomy is a fundamental step in rehabilitating a patient recovering from critical illness [13].

The frequency of tracheostomy in the management of patients receiving mechanical ventilation contrasts with the lack of evidence as to when a tracheostomy tube should be removed. It appears that the majority of critically ill tracheostomized patients who survive to ICU discharge can eventually be successfully decannulated [14]. Limited uncontrolled pilot studies [15, 16] and expert guidelines [17] have proposed that decannulation be considered in patients once respiratory mechanics are adequate, mechanical ventilation is no longer needed, upper airway obstruction is resolved, airway secretions are controlled, and swallowing has been evaluated. It is unclear to what extent these publications guide clinician decision-making. To complicate matters, the long-term management of tracheostomized patients often is fragmented between different health care settings and providers because recovery from critical illness takes months [18]. Consequently, the clinicians faced with the decision whether or when to decannulate a tracheostomized patient are often not the same physicians who inserted the tracheostomy and may have limited experience with such patients. Since little is known about how clinicians make decisions to decannulate tracheostomized patients, we conducted a survey to determine the factors clinicians consider to be important in recommending decannulation and to ascertain their opinions regarding the definition of decannnulation failure.

Our study was designed to examine tracheostomy decannulation opinions at major centers around the world. The results demonstrate three major findings. First, clinicians are able to identify patient factors that they believe are important in the decision to decannulate a tracheostomized patient. Second, there is significant variability in clinicians' decannulation opinions. Third, clinicians are able to define decannulation failure and identify what they believe are acceptable rates of failure.

Clinicians are able to identify patient factors that they believe are important in the decision to decannulate a tracheostomized patient. However, there is significant variability in clinicians' opinions. For example, decannulation recommendations varied between physicians and respiratory therapists as well as between clinicians who worked at acute facilities and those who worked at chronic care facilities. Our results are consistent with a growing body of scientific literature suggesting that factors idiosyncratic to health care providers are major determinants of the medical decisions and care that patients receive [24]. Tracheostomy care, therefore, is likely to vary significantly depending on the individual clinician responsible for a patient's care. Our findings highlight the need for clinical studies in tracheostomy care to guide clinical decision-making. Clinicians indicated in our survey that, in determining whether to decannulate a tracheostomized patient, the patient's level of consciousness, ability to tolerate tracheostomy capping, cough effectiveness, secretions, and oxygenation needed to be evaluated. Although the ability to tolerate tracheostomy capping was judged to be an important determinant of tracheostomy decannulation, it did not influence clinicians' recommendations in the clinical scenarios. Previous studies and guidelines have also suggested that maximal expiratory pressure, peak cough flows, arterial blood gases, and upper airway endoscopy may be useful in the decannulation decision-making process, although these factors require special equipment and expertise and are more complicated than the simple bedside criteria employed in our study [15‐17]. We propose that a patient's level of consciousness, cough effectiveness, secretions, and oxygenation be tested in a clinical trial as four simple bedside factors to consider in determining whether to decannulate a tracheostomized patient.

Clinicians understand that tracheostomy decannulation is not without risk. However, there is currently no accepted definition for decannulation failure. Extubation failure is defined by most clinicians and researchers as the need to reinstate mechanical ventilation within 24 to 72 hours of planned extubation [25‐29]. The incidence of extubation failure is reported to be between 2% and 25% of extubation attempts and is associated with increased hospital mortality, prolonged ICU and hospital stays, and more frequent need for long-term acute care [27‐29]. In their study of criteria for extubation and tracheostomy tube removal, Bach and Saporito [16] defined successful decannulation as 'extubation or decannulation and site closure with no consequent respiratory symptoms or blood gas deterioration for at least 2 weeks'. Ceriana and colleagues [15], in evaluating the feasibility of a decisional flowchart for weaning from tracheostomy, defined failure as the 'need to reopen the tracheotomy because of an acute episode or progressive worsening of arterial blood gases not corrected by the application of noninvasive mechanical ventilation'. The two research groups documented reinsertion of artificial airways in 35% of patients at 2 weeks and 3% of patients at 3 months, respectively [15, 16]. Our data suggest that most clinicians would consider reinsertion of an artificial airway within 48 to 96 hours following planned tracheostomy removal to constitute a decannulation failure. Furthermore, clinicians appeared to consider a decannulation failure rate of 2% to 5% to be acceptable. Clinicians' preference for a lower failure rate for decannulation than extubation may be explained by the fact that tracheostomies are generally an intervention of longer duration than endotracheal intubation. Alternatively, tracheostomy may be perceived as a more invasive intervention than intubation with less tolerance for reinstrumentation. In addition, there may be a role for devices such as the Minitrach [13, 15], tracheal button [17], and noninvasive mechanical ventilation [15] to serve as a bridge during the decannulation process to minimize the risk of failure, although we did not address this issue in our survey.

The results of this study need to be interpreted within the context of its limitations. First, the survey instrument used was simple and perhaps imperfect. To ensure that our instrument was economical, we were unable to explore all important aspects of tracheostomy decannulation. Second, we measured what health care workers stated they would do in response to scenarios, although we did not observe how they practice. The survey questions may have generated idealized responses rather than reflect actual practice. Finally, the snowball sampling technique efficiently identified clinical experts but provided for unequal sampling from the different jurisdictions. For example, only a small minority of respondents worked principally in a rehabilitation or chronic care facility. Nevertheless, the respondents were clinicians experienced in the management of patients with tracheostomies and represented more than 100 medical centers.

Our study provides the first survey of contemporary tracheostomy decannulation practices. Our data indicate that clinicians consider a patient's level of consciousness, cough effectiveness, secretions, and oxygenation when determining whether to recommend tracheostomy removal. Clinicians define decannulation failure as the need to reinsert an artificial airway within 48 to 96 hours of planned tracheostomy removal and are willing to accept a 2% to 5% failure rate. The development of evidence-based tracheostomy guidelines will facilitate the safe and effective management of patients with tracheostomies.