Is there a correlation of sonographic measurements of true vocal cords with gender or body mass indices in normal healthy volunteers?

BACKGROUND: Ultrasound is a readily available, non-invasive technique to visualize airway dimensions at the patient's bedside and possibly predict difficult airways before invasively looking; however, it has rarely been used for emergency investigation of the larynx. There is limited literature on the sonographic measurements of true vocal cords in adults and normal parameters must be established before abnormal parameters can be accurately identified.
OBJECTIVES: The primary objective of the following study is to identify the normal sonographic values of human true vocal cords in an adult population. A secondary objective is to determine if there is a difference in true vocal cord measurements in people with different body mass indices (BMIs). The third objective was to determine if there was a statistical difference in the measurements for both genders.
MATERIALS AND METHODS: True vocal cord measurements were obtained in healthy volunteers by ultrasound fellowship trained emergency medicine physicians using a high frequency linear transducer orientated transversely across the anterior surface of the neck at the level of the thyroid cartilage. The width of the true vocal cord was measured perpendicularly to the length of the cord at its mid-portion. This method was duplicated from a previous study to create a standard of measurement acquisition.
RESULTS: A total of 38 subjects were enrolled. The study demonstrated no correlation between vocal cord measurements and patient's characteristics of height, weight, or BMI's. When accounting for vocal cord measurements by gender, males had larger BMI's and larger vocal cord measurements compared with females subjects with a statistically significant different in right vocal cord measurements for females compared with male subjects.
CONCLUSION: No correlation was seen between vocal cord measurements and person's BMIs. In the study group of normal volunteers, there was a difference in size between the male and female vocal cord size.

INTRODUCTION

The airway is the first priority in the primary assessment of critical patients in the Emergency Department (ED). Studies have shown 1 in 10,000 patients have a difficult airway,[12] and data from the ED has demonstrated that 1% of patients will have a failed airway, requiring surgical intervention to establish the airway.[34] Mallampati scores are often used and correlate well with oropharyngeal visualization and laryngoscopy,[5] but are not amenable to emergency situations. Computed tomography (CT) and magnetic resonance imaging (MRI) may be used to measure laryngeal dimensions and can provide accurate data for both the true and false vocal cords and therefore could address airway or vocal cord concerns.[67] Unfortunately, both these modalities are not ideal for the patient in acute or impending respiratory failure due to airway compromise from such diseases as angioedema or anaphylaxis. Flexible nasolaryngoscopy can assess airway structures by direct visualization. Yet nasolaryngoscopy is not readily available and the physician has to rely on other methods of evaluation.

Ultrasonography is a rapid, readily and widely available non-invasive technique to visualize airway dimensions at the patient's bedside and possibly predict difficult airways before invasively looking when a patient presents with possible impending airway compromise; however, it has rarely been used for emergency investigation of the larynx.[8910] The use of high-frequency ultrasound allows visualization of the airway structures and trachea. Other studies from the emergency and critical care setting have looked at ultrasound as a modality to help confirm endotracheal tube placement[11] or diagnosis epiglottitis.[12] A study done by Adhikari et al. looked at sonographic measurements of anterior neck soft-tissue thickness at the level of hyoid bone and thyrohyoid membrane, but not vocal cord measurements to help distinguish difficult and easy laryngoscopies.[13] There is limited literature on the sonographic measurements of true vocal cords in adults[8] and normal parameters must be established before abnormal parameters can be accurately identified. There has only been one study investigating the parameters for real and false vocal cords according to age group.[8] We conducted a pilot study to identify the normal sonographic values of human true vocal cords in an adult population from the ED and to determine if there is a difference in true vocal cord measurements in people with different body mass indices (BMI's) and within gender groups. If an average size could be established, then further research in the future could help identify potential vocal cord edema and possible airway compromise.

MATERIALS AND METHODS

This study was performed with the approval of the Institutional Review Board of the State University of New York Downstate Medical Center. Written informed consent was obtained from all participants. In this study, laryngeal ultrasounds were performed on adult subjects conveniently selected from our ED faculty, including residents, attending's and nursing staff. Subjects were excluded if they had respiratory or laryngeal complaints, such as dyspnea, cough, sore throat, or hoarseness. Participant's BMIs were calculated based on patient's last recorded height and weight. A normal weight was defined as a BMI of 18.5-24.9, overweight as a BMI of 25-29.9 and obese as a BMI of greater than 30.

The participants were placed supine with their head extended and neck flexed (the “sniffing” position). Sonographic images were obtained using a Sonosite (Bothell, WA, US) HFL38 × 13−6 MHz high frequency linear transducer. The ultrasound transducer was orientated transversely across the anterior surface of the neck at the level of the thyroid cartilage (TC) since as that provided the best window for imaging the vocal cords in our participants. We were able to visualize the true and false vocal cords in all participants by sliding the transducer in a cephalocaudad direction over the TC. The true vocal cords appeared as 2-paired triangular hypoechoic structures [Figure 1]. The false vocal cords lay parallel and cephalad to the true cords and were more hyperechoic in appearance. The true vocal cords, which is made up the vocal cord ligament and vocalis muscle attach to the TC anteriorly and the arytenoid cartilage posteriorly. The width of the vocal cord was measured perpendicular to the length of the cord at its mid-portion. This method was duplicated from a previous study to create a standard of measurement acquisition.[8]

Figure 1

The left image demonstrates an ultrasound image of the true vocal cords using a high frequency array probe in the transverse axial plane. The image on the right is a schematic representation of the ultrasound image on the left. SM: Sternocleidomastoid muscle, TC: Thyroid cartilage, VM: Vocalis muscle, VL: Vocal cord ligament, AC: Arytenoid cartilage

We estimated our sample size with a 95% confidence interval (CI) of 1 mm. This would require a sample size of approximately 40 subjects. To describe our ultrasound findings we used descriptive statistics with 95% CI (statistical package for the social sciences [SPSS] version 15.0, SPSS Chicago Illinois, USA).

All ultrasounds were performed by two ultrasound fellowship trained emergency physicians (MS, LB). Measurements were obtained in millimeters by using distance calipers.[8] To compare agreement between both sonographers, intra class correlation (ICC) of the sonographic measurements was calculated on all subjects who were enrolled. The sonographic measurements were the best measurement taken of the vocal cords. Examinations were performed by the first operator (MS) and then by the second operator (LB) independently on the same day. The operators were blinded to the others examination results.

RESULTS

In this preliminary study at our institution, we found an average vocal cord size of 7.5 mm with a standard deviation of 3.2 mm. In total, 38 subjects were enrolled and the results are broken down as follows: 55% were female and the average BMI was 25.9 ± 5.1 [Table 1]. Left vocal cord measurements (7.83 ± 1.99) were slightly but not statistically significantly larger than right cord measurements (7.31 ± 2.19) for all subjects [Table 2].

Table 1

Subject characteristics

Table 2

Subject characteristics by gender

When accounting for vocal cord measurements by gender, males had larger BMI's and larger vocal cord measurements compared to females subjects with a statistically significant different in right vocal cord measurements for females (6.66 [5.9-7.18]) compared with male (8.24 [6.86-9.61]) subjects (P = 0.007).

The intra-class correlation was 0.633 (95% CI, 0.235-0.824), which shows substantial agreement between both observers.

The study demonstrated a sonographic average vocal cord measurement for males and females separately. No correlation between vocal cord measurements and patient's characteristics of gender, height, weight, or BMI's [Tables 3 and 4] were noted.

Table 3

Correlations between airway measurements to patient characteristics

Table 4

Comparison of airway measurements by basal metabolic index categories

DISCUSSION

This is the first study to evaluate and measure the true vocal cords of normal healthy subjects with varying degrees of BMI's in an ED setting.

Our study showed the left vocal cord had slight larger, but not statistically different measurements compared with the right, but when accounting for gender there was a statistical difference in measurements of the right vocal cord between male and female subjects. No correlation was seen between vocal cord measurements and a person's BMI.

The true vocal cords were visualized in all participants. The parameter of vocal cord measurement our pilot study focused on was the width rather than both width and length. Visualization of the true vocal cords was 100% in this pilot study and this can be explained by the fact that our volunteers were younger, which also correlates with Hu et al.[8] findings of having better visualization rates with younger participants. Furthermore by sampling a younger population we were less likely to encounter older patients who may have developed calcification of the TC, which may have impeded vocal cord visualization. Vocal cord measurements were greater in males than those in female adults and these findings are similar to previous reports.[81415] Our measurements are larger than previously reported vocal cord measurements.[814] This may be explained by ethnic differences, a younger adult population as well as operator technique and different measuring points taken. The left true vocal cord measurement was larger but not statistically significant compared with the right true vocal cord. This difference can most likely be explained by the operator's technique and their defined measuring end-points. The reliability ICC was good for our measurements. Intra-class correlation was used since there is not an established gold standard for measuring vocal cord size that could be used for comparison. Our results show that sonography is reliable and reproducible method for measuring the true vocal cords.

The practical implications of such a study is the potential to correlate true vocal cord measurements in patients presenting with angioedema who potentially need airway access and the treating physician doesn’t have access to direct naso-laryngoscopy. In a future, more powered study, based on our pilot data, we hope to establish standard measurements of true vocal cords as visualized through ultrasound. If a standard measurement can be established, these measurements can be applied to patients with presumed edema of the true vocal cords and possible impending airway failure and arrest. The availability and portability of ultrasound enables many physicians the ability to quickly assess the airway.

Further work is also necessary to characterize the sonographic anatomy in patients with clinical abnormalities of the vocal cords; here, a comparison with more established imaging techniques such as CT, MRI, or direct laryngoscopy would be useful.

Limitations

There are a few inherent limitations to this pilot study. The first being the small sample size studied. The sample was chosen for statistical significance; with the difference between male and female now identified, the two groups should be studied separately in the future to establish an average vocal cord measurement and not together. A larger sample size may show a statistical difference in vocal cord measurements for males and females, or correlation with a person's BMI. Measurements were taken during quiet respiration, but measurements taken between both operators may have been taken during different phases of respiration and measurements may be slightly different if taken during inspiration than expiration. There is also potential variability between the two operators (MS, LB) measurements, but this was mitigated by the substantial agreement between operators (ICC = 0.63). Further studies should include the ability of novice sonographers to evaluate the airway with ultrasound.

Furthermore, the measurements were taken by ultrasound-trained faculty, though if shown to be of value, this skill set can be taught to the standard practitioner. Finally, another important limitation is that there is no comparison to a gold standard that accurately measures the vocal cords.

CONCLUSION

There is an evidence that ultrasound is a reliable and reproducible method for measuring the true vocal cords. No correlation was seen between vocal cord measurements and person's BMI. The data supports, in normal healthy volunteers, an average vocal cord measurement may be established in a future study with a larger sample size separately evaluating male and female subjects.