Evidence-based practice guidelines for the diagnosis of dysphagia in children

Abstract

We present an evidence-based guideline for the diagnosis of dysphagia in children. The articles included were retrieved by electronic search in Medline, Lilacs, and Cochrane databases. The search was restricted to those published between January 1996 and December 2007, in English or Spanish, and only to those including children (0-18 years). Manual search of papers cited by primary articles and non-systematic reviews was also done. The search strategy identified 605 articles and 58 were chosen for further analysis. Evaluation of the papers was carried out by all the authors and recommendations were done by consensus. The guideline was developed by using the AGREE instrument (Appraisal of Guidelines Research and Evaluation) and the GRADE model (Grading Strength of Recommendations and Quality of Evidence in Clinical Guidelines) from the American College of Chest Physicians.

The diagnostic approach to children with dysphagia begins with clinical suspicion, and thorough clinical history and physical examination should precede any diagnostic test. We found that in children with clinical features that suggest dysphagia, videofluoroscopy is the recommended imaging technique to confirm this condition (Recommendation 1C). This test is considered to be the gold standard to demonstrate aspiration and/or penetration of contrast material into the airway (Recommendation 1A). Ultrasound is considered as the best diagnostic imaging technique in children with dysphagia associated with tongue abnormalities (Recommendation 1C). On the other hand, electromyography might be used as a screening test in children aged 5 years or more with disorders of the facial muscles (Recommendation 2B). Finally, magnetic resonance imaging seems to provide the best diagnostic yield in those children with cervical lesions responsible for disrupting deglutition dynamics (Recommendation 1C).

In conclusion, available articles on diagnostic tests for dysphagia have serious methodological deficiencies and do not allow recommendations with better levels of evidence.