Produção Nacional
Revisado por pares
Evaluation of the association between voice formants and difficult facemask ventilation
2019; Lippincott Williams & Wilkins; Volume: 36; Issue: 12 Linguagem: Inglês
10.1097/eja.0000000000001108
ISSN1365-2346
AutoresClístenes Crístian de Carvalho, Danielle Melo da Silva, Antonio Deusany de Carvalho, Fernando Jorge Firmino Nóbrega, Flávia Augusta de Orange,
Tópico(s)Voice and Speech Disorders
ResumoEditor, Facemask ventilation remains a crucial component of airway management. It is also of significant importance when it is used to 'bail out' difficult airways. Nevertheless, problematical difficult facemask ventilation (DFMV) is still not addressed properly since the literature is lacking accurate predictors.1 In this regard, studies have shown that upper airway anatomy is associated with specific voice measurements such as formant frequencies – more intense frequencies within the spectrum of voice frequencies (Fig. 1).2,3 Each formant relates to a specific region of upper airway as follows: formant 1 (F1) is related to the opening of mouth and vertical displacement of tongue; formant 2 (F2) is associated with horizontal displacement of tongue; formant 3 (F3) has a connection with anterior oral and pharyngeal cavities; and formant 4 (F4) is probably related to length of the laryngeal tube.4,5 We therefore aimed to evaluate whether such a relationship might be translated into an association between voice formants and DFMV, during general anaesthesia for elective surgical procedures.Fig. 1: Graph of the voice frequencies as a function of the intensity. The blue curve demonstrates a real-patient spectrum of voice frequencies and the pink, a representation of the formants. Note that the formants are the more intense frequencies within the spectrum of voice frequencies. F1 means first formant; F2, second formant; F3, third formant and F4, fourth formant. Note that each formant represents a peak of intensity.A prospective study was conducted in patients scheduled for elective surgical procedures under general anaesthesia who agreed to participate and signed the informed consent form or the informed assent form in case of patients under 18 years of age. Ethical approval for this study (Ethical Committee protocol No. 1253520) was provided by the Ethical Committee of Centro de Ciências da Saúde of Universidade Federal de Pernambuco, Recife, Pernambuco, Brazil on 1 October 2015. Patients unable to pronounce vowels, those not ventilated with facemask, and those under 15 years of age were excluded. We collected data on age, sex, height, weight, BMI, ASA physical status, modified Mallampati, characteristics of formants and difficulty of facemask ventilation by the Han and Kheterpal6 classification system. From each selected patient at a transitional waiting hall and before transport to the operating room, five phonemes were recorded (/a/, /ε/, /i/, /ɔ/ and /u/, named a, e, i, o and u, respectively)7 using the microphone at the base of the Apple iPhone 5, model A1457, US version from Apple New York, NY. The first five formants (f1, f2, f3, f4 and f5) from each phoneme were extracted for analysis, making 25 formants per patient; the third formant of phoneme /i/ is described as if3, for example. General anaesthesia induction was performed in a standard protocol by a 3-year-experienced anaesthesiologist who collected data on the Han and Kheterpal classification. Statistical analyses were performed on 431 patients' data. Table 1 summarises the demographic and airway descriptive data. The bivariate statistical analysis demonstrated an association between the Han and Kheterpal classification and obesity (P = 0.001), ASA physical status (P = 0.000), weight (P = 0.000), BMI (P = 0.001), if5 (P = 0.032) and of5 (P = 0.033). A prediction model was set by a logistic regression analysis including only the two significant formants in the bivariate analysis. This model featured an area under the receiver operating characteristic curve of 74%; a sensitivity of 80%; a specificity of 82.2%; a positive predictive value of 20%; and a negative predictive value of 98.6%.Table 1: Demographic data and Mallampati classificationIn summary, the results of this preliminary study provide evidence for an association between voice and DFMV. Thus, voice formants may be considered as an alternative valuable information, in addition to already known factors for DFMV, to predict DFMV in patients scheduled for general anaesthesia. The voice analysis can be made automatically and can work as a simple and objective method to evaluate several sound parameters at once. However, large and multicentre studies should be conducted to confirm the initial findings, and to further refine how voice formants could be integrated in a predictive score for difficult mask ventilation. Acknowledgements relating to this article Assistance with the letter: the authors are grateful to the 'Hospital das Clínicas da Universidade Federal de Pernambuco' (UFPE), Department of Anaesthesiology, Recife, Pernambuco, Brazil, the institution in which this study was conducted. Financial support and sponsorship: none. Conflict of interest: none.
Referência(s)