Tracheostomy decannulation methods and procedures for assessing readiness for decannulation in adults: a systematic...

Despite the undisputable benefits of tracheostomy, it has been reported to have links with impaired communication, reduced quality of life and a risk of health complications such as bleeding, tracheal stenosis and in some cases resulting in mortality. There is a paucity of literature on tracheostomy decannulation methods and procedures, leaving the decision to expert opinion and institutional guidelines. This study aimed to map evidence on methods and procedures of tracheostomy decannulation in adults and assessment of readiness for decannulation, to reveal knowledge gaps and inform further research. We conducted a systematic search of peer reviewed and grey literature on PubMed/MEDLINE, Google Scholar, Union Catalogue of Theses and Dissertations via SABINET Online, World Cat Dissertations and Theses via OCLC, WHO library and governmental websites from 1985 to present. Following title screening, abstract and full article screening was performed by two independent reviewers guided by the eligibility criteria. Data from included studies were extracted, collated, summarized and synthesized into the following themes: assessment, removal, monitoring and definition of failure of decannulation. Quality of the included studies was assessed using the Mixed Methods Appraisal Tool version 2011. Twenty-five out of 51 screened articles were eligible for data extraction. There was wide variation in the assessment methods employed across and within similar patient groups. The common themes that emerged in the assessment for readiness for decannulation are informed consent, clinical stability, airway patency, physiological decannulation, swallowing assessment, level of consciousness, effectiveness of cough and clearance of secretions. In conclusion, the current body of evidence is inadequate and requires further research, particularly validation of different parameters used. A protocol approach to decannulation may be inappropriate but rather an algorithmic approach using validated parameters.