INTRODUCTIONLaryngomalacia is the most common cause of stridor in children, being the culprit for 65 to 75% of the stridor cases1. It affects more men then women, at a 2:1 rate. This condition is established by the laryngeal failure in keeping its lumen open during inhaling. Inspirational stridor is the main disease manifestation2 which is usually self limited and resolves in the first 12 to 18 months of life. Nonetheless, in 10% of the cases it is necessary to operate, because the respiratory obstruction can cause severe dyspnea, dysphagia, growth delay and obstructive sleep apnea1.
Its etiology is still unknown. Numerous theories try to explain its etiopathogenesis. The one most commonly accepted today states that laryngomalacia results from a neuromuscular immaturity, causing supraglottic hypotonia and bringing laxity to these structures during inhaling2. Another theory is associated with a reduction in support from the cartilaginous framework. Another possible etiology is laryngeal muscles hypotonia.
Diagnosis is based on clinical and epidemiological history of newborns and infants with inspiratory stridor, which worsens with restlessness or crying, and improves during sleep. Usually, direct laryngoscopy is the gold standard exam to diagnose stridor3-5. Since laryngomalacia is the most common cause of stridor and, usually the most benign cause, a dynamic laryngeal assessment by means of a nasolaryngoscopy through flexible endoscope has shown good results, although it does not replace direct laryngoscopy with the rigid endoscope in indicated cases.6 However, endoscopic image assessment has not been properly systematized until current days, with many confounding factors associated with examiner's experience and anatomical variation inherent to human beings. Therefore, endoscopic image evaluations have not been systematized enough, with confounding factors associated with the observer's experience and to the anatomical variations inherent to human beings. Thus, we still need to establish, in a controlled fashion, which anatomical alterations prove to be more constantly present when evaluated by nasal-fibroscopy, setting up the parameters that must be evaluated in all cases with suspicion of laryngomalacia. It seems that this has not been done until current times.
The goal of the present investigation is to establish videolaryngoscopy diagnostic accuracy regarding laryngomalacia, when performed by observers with variable clinical experience in laryngology.
MATERIALS AND METHODSThis is an agreement study, used to assess the reliability of a diagnostic test. The project was evaluated by the Ethics in Research Committee of the Federal University of Minas Gerais, under protocol # 571/07.
Eighteen patients with prior diagnosis of laryngomalacia - based on clinical, epidemiological and anatomical data, were evaluated. The videolaryngoscopic exams of these patients were compared in relation to the agreement index from four different examiners. The videolaryngoscopic assessment protocol used was based on the most frequently observed anatomical alterations found in this disorder2. The parameters considered were:
1) anterior arytenoid drawing during inhaling;
2) tubular epiglottis (omega-shape) which collapses during inhaling;
3) short aryepiglottic folds;
4) posterior epiglottic drawing;
5) good or bad view of the vocal folds;
6) edema on the posterior laryngeal structures. The exams were carried out with a flexible nasal-laryngoscope and recorded on a video-tape.
The examiners had between t (examiners 1, 2 and 3) and 15 years (examiner 4) of clinical experience in children laryngology. They all had prior knowledge that the patients had been diagnosed with laryngomalacia and also knew what the anatomical criteria to be considered were.
The analyses were carried out in such a way that one observer did not know what the others had found. Later on, all the observers analyzed the exams together and reached a consensus, which was then considered reference. Therefore, in order to establish a gold standard, we considered the consensus of the exams and not the total number of patients. We were interested in checking the agreement between the alterations noted by the examiners in relation to the consensus (gold standard). In comparing, each examiner was assessed as to the number of correct findings in relation to the gold standard.
RESULTSThe analysis encompassing the eighteen laryngoscopies carried out by each examiner and the consensus on the assessment are reported on Table 1. The average agreement of all the parameters considered in relation to the consensus of the four examiners was of 88.2%. The data associated with such analysis is specified on Table 2.
As to the individual analysis, examiner 4, with 15 years of clinical experience, was the one who most disagreed in terms of the "Posterior Laryngeal Edema" parameter, when compared to the other examiners with 2 years of clinical experience (Table 2).
DISCUSSIONNasal-laryngoscopy is a very important exam in otorhinolaryngology, and it is increasingly more indicated in the evaluation of upper airway affections. Since it is an examiner-dependent exam, it is subject to perception variations mainly associated with the examiner's clinical experience.
As far as the laryngomalacia diagnosis is concerned, the agreement average for the parameters assessed was of 88.2% among examiners. The most reliable parameter was "short aryepiglottic folds", of which agreement rate was of 95.8%. The less reliable was "not visualizing the vocal folds", in which the agreement was of 81.4%. Not being able to see the vocal folds in the videolaryngoscopic evaluation has been associated with the prognosis, because some authors use such information, together with the clinical picture, in order to indicate surgery7.
In the present investigation, the agreement rate, considering all the parameters evaluated, was high, proving that the exam is reliable for the diagnosis of laryngomalacia, even when performed by experts with little experience in laryngology.
Two parameters must be stressed:
1) vocal folds visualization or not, because examiners with less experience in laryngology had difficulty in identifying it, and this is one of the parameters considered for surgery indication7;
2) edema on the posterior laryngeal structures8, was more easily identified by the examiner with greater clinical experience.
CONCLUSIONNasal-laryngoscopy is a good exam for the diagnosis of laryngomalacia, bearing a sensitivity of 88.2% - regardless of examiner's clinical experience. The most reliable parameter was the evaluation of short aryepiglottic folds. For cases of possible surgical indication and in order to assess the presence of edema on the posterior laryngeal structures, clinical experience in children laryngology is paramount.
REFERENCES1. Olney DR, Greinwald JH, Smith RJ, Bauman N. Laryngomalacia and its Treatment. Laryngoscope 1999;109(11):1770-5.
2. Duprat AD, Campos CAH, Costa HOO. Tratado de Otorrinolaringologia (vol 4) - 1a ed. São Paulo: Roca; 2003; 440-50.
3. Holinger LD. Diagnostic endoscopy of the pediatric airway. Laryngoscope 1989;99:346-48.
4. Berkowitz RG. Neonatal upper airway assessment by awake flexible laryngoscopy. Ann Otol Rhinol Laryngol 1998;107(1):75-80.
5. Moumoulidis I, Gray RF, Wilson T. Outpatient fiber-optic laryngoscopy for stridor in children and infants. Eur Arch Otorhinolaryngol 2005:262(3):204-7.
6. Waters KA, Woo P, Mortelliti AJ. Assessment of the infant airway with videorecorded flexible laryngoscopy and the objective analyses of vocal fold abduction. Otolaryngol Head Neck Surg 1996;114(4):554-61.
7. Zalzal GH, Anon JB, Cotton RT. Epigottoplasty for the treatment of laryngomalacia. Ann Otol Rhinol Laryngol 1987;96:72-6.
8. Ungkanot K, Friedman E, Sulek M. Retrospective analyses of airway endoscopy in patients less than 1-month old. Laryngoscope 1998;108(11):1724-28.
1 PhD in Otolaryngology, Adjunct Professor - Department of Otolaryngology - Medical School - UFMG.
2 PhD. Adjunct Professor - Department of Otolaryngology - Medical School - UFMG.
3 Resident physician in Otolaryngology - University of Minas Gerais Medical School.
4 Resident physician in Otolaryngology - University of Minas Gerais Medical School.
5 Resident physician in Otolaryngology - University of Minas Gerais Medical School.
6 Resident physician in Otolaryngology - University of Minas Gerais Medical School.
Faculdade de Medicina da UFMG e Hospital das Clínicas da UFMG
Send Correspondence to: Tânia Mara Assis Lima - Rua do Ouro 229/502 Bairro Serra Belo Horizonte MG 30220-000.
Paper submitted to the ABORL-CCF SGP (Management Publications System) on November 08th, 2006 and accepted for publication on January 2nd, 2006. cod. 3501.
