Changing Trends of Color of Different Laryngeal Regions in Laryngopharyngeal Reflux Disease

Abstract

Introduction:

In our previous study, we demonstrated that there might be correlations between laryngopharyngeal reflux disorder and the hue value of laryngoscopic images. And we found that different regions of larynx have different hue values. It was hypothesized that the degree of inflammation varies between different laryngeal regions, due to an acid reflux pattern.

Objective:

The objective of this study was to compare the changing trends of hue values of different laryngeal regions in patients with laryngopharyngeal reflux (LPR) disease.

Methods:

Ninety-seven patients, including 20 pH-positive, 19 pH-negative were tested for LPR through multichannel intraluminal impedance 24-hour pH monitoring, and 58 controls with reflux symptom index less than 13. Laryngoscopic images of all patients were obtained. The hue values of 7 areas of interest, including both sides of the true vocal folds, the false vocal folds, the arytenoids, and the interarytenoid space, were quantified using a hue calculation. The analysis of variance analysis was applied to find if there was significant difference between different groups within each region.

Results:

(1) In the regions of both sides of the true vocal folds and interarytenoid, there was no significant difference between positive group and negative group; (2) in the regions of both sides of false vocal folds, there was no significant difference between negative group and control group; (3) in the regions of both sides of arytenoids, there was neither significant difference between positive group and negative group nor between negative group and control group; (4) in other comparisons, there were significant differences.

Conclusion:

Hue values of separate laryngeal regions are different. In negative group, the hue values of interarytenoid region are similar with positive group, and the sensitivity to the acid are different from different regions. The sensitivity in the true vocal folds may be present.

Keywords

laryngitis laryngopharyngeal reflux color analysis

Introduction

Laryngopharyngeal reflux (LPR) is a common condition in otolaryngologic diseases and it afflicts nearly 50% of all patients with laryngeal or voice disorders.¹ Additionally, almost 15% of patients in outpatient otolaryngology clinics have chronic LPR.^2
-4 Hoarseness, throat clearing, choking sensation, dysphagia, dysphonia, globus, and laryngospasms also are suspected to be associated with LPR.⁵

In the 1990s, Jacob et al⁶ reported that posterior laryngitis could be associated with evidence of gastropharyngeal reflux, suggesting that acid reflux may affect the posterior structure of the larynx. However, the mechanisms on how the acid affects the laryngopharynx are still unknown. In a previous study, we used the hue value of laryngoscopic images as a tool to diagnose LPR and hue value was demonstrated to be a valid quantitative parameter of inflammation.^7,8 Moreover, we found that the hue values of 7 different laryngeal regions were not same. Two false vocal folds and the interarytenoid region tend to have lower hue values, while others do not. This indicates that the degree of inflammation varies between different laryngeal regions. This may be associated with acid reflux pattern.

This study was designed to compare the changing trends of hue values of different laryngeal regions. To perform this, we applied the objective 24-hour multichannel intraluminal impedance (MII-24 h) pH monitoring and the reflux symptom index (RSI) scale to classify LPR patients.

Materials and Methods

Ethical Considerations

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The protocol number is 2017-0515.

Selection/Classification of Patients

In this prospective study, 97 patients who attended the clinic between January 2015 and February 2017 were included. The patients were divided into 3 groups: The first was the pH-positive group in which the patients had the RSI score >13, and the number of proximal acid episodes was greater than or equal to 3, proximal acid exposure time was greater than 1%, or impedance detected proximal acid exposure was less than 4. The second group was pH negative in which the patients’ RSI score >13, but there were fewer than 3 proximal acid episodes, proximal acid exposure timeless was than 1%, and impedance detected proximal acid exposure was less than 4. The third group was the control group who had no complaints about LPR (RSI < 13), but were administered laryngoscopy for routine testing before thyroid surgery. All patients were evaluated subjectively by a RSI and a laryngoscope. With the patients’ consent, the patients with symptoms also underwent 24-hour pH probe monitoring. Laryngopharyngeal reflux was confirmed by total acid exposure time. Laryngopharyngeal reflux is confirmed when the proximal sensor detects pH levels lower than 4 for more than 1% of the time, or the number of acid episodes was at least 3 during a 24-hour monitoring period.^9,10

Laryngeal images were obtained from all patients from one high-resolution laryngoscopy. The grouping based off of the pH monitoring was as follows: 20 patients who were confirmed with LPR (16 men, 4 women; age range: 18-63 years; mean age: 42.2 years), 19 patients who were suspected to have LPR but tested negative (10 men, 9 women; age range: 23-78 years; mean age: 46.3 years), and 58 control patients (25 men, 33 women; age range: 20-68 years; mean age: 45.0 years). All patients in this study were the same descent. There were no significant differences in RSI between age groups and no significant difference in RSI between positive and negative groups (P > .05). However, there were significant differences between the control group and the other 2 groups (P < .05), as shown in Figure 1.

Figure 1.

Reflux symptom index scores between pH-positive, pH-negative, and control groups. RSI indicates reflux symptom index.

Image Collection

Images were obtained during a laryngoscope through only one flexible high-resolution laryngoscope (model EV-NE; XION; Berlin, Germany). There was no standard nasal sidedness to the flexible laryngoscopy; the side selected was based on accessibility or minimal level of nasal congestion. However, the selection of a particular side should not have a significant effect on the overall classification. The images were taken near the midline of the larynx, regardless of which side was chosen. The minor sidedness distance should not significantly change the hue values. The white balance function was used to exclude other potential factors that may interfere with the true image color and to standardize the images to what was present. After the images were saved as JPEG files, the hue values were analyzed. Images with a noticeable blur or glare were omitted from the study.

Data Collection

Images were deidentified and evaluated with a customized color analysis script developed on Matlab 2013b.The color analysis tool calculates the hue value of several manually drawn regions of a laryngoscopic image using the algorithm explained in Witt et al.⁷ For each image, specific regions of the larynx were outlined for each area of interest. This includes the left and right true vocal folds, left and right false vocal folds, left and right arytenoids, and the interarytenoid space. Using a digital sketchpad, the boundaries were manually drawn, as shown on Figure 2. In our case, processing images using regional calculations was preferable as compared to a global calculation because hue variations may be localized to small areas of the larynx with little variation present in other locations. Dividing the image into manually drawn regions would allow parts of the image to be grouped such that the area would have similar characteristics in physical appearance. This was done to reduce variations in our results between each individual region. The deidentification of the images eliminated bias as the evaluator was blinded to the group to which each image belonged.

Figure 2.

Laryngoscopic image in the color analysis tool interface with laryngeal regions of interest outlined. (A) Right true vocal fold, (B) left true vocal fold, (C) right false vocal fold, (D) left false vocal fold, (E) right arytenoid, (F) left arytenoid, (G) interarytenoid.

As demonstrated in Figure 3, the red color is the reference point of the traditional color circle, which holds a domain from 0 to 360 degrees. For each region, the hue parameter was standardized around the red color to a scale ranging from −180 to 180 degrees. The pixels surrounded by the drawn boundaries had their individual hue angles measured using the same algorithm described in Witt et al⁷, and an average hue of each region was recorded.

Figure 3.

A 2-D hue value wheel, indicating perceived color based on red as 0° (image from Witt et al).

Data and Statistical Analysis

To compare the average hue values of the 7 regions between the 3 groups, we applied a repeated measures analysis. We then obtained a curve describing the trend made by these 7 measurements. We also compared the hue values of regions for each side of the larynx (both including interarytenoid region) to obtain the trend of both sides. Then, we compared the trends among the 3 groups to see if there were any significant differences.

Results

From the 7 regions within the larynx, we obtained the hue values of each region. In order to eliminate the systematic error, we calculated the predicted values and their standard error (shown in Table 1). Figures 4 through 6 show the plots of the mean values of hue of the 7 regions, the right side and the left side of larynx, respectively.

Table 1.

Predicted Average Hue Values of Each Region.^a

Region	pH Positive	pH Negative	Control Group
Right arytenoid	4.26 ± 0.85	5.88 ± 0.87	7.70 ± 0.50
Left arytenoid	4.68 ± 0.82	6.95 ± 0.84	8.84 ± 0.48
Interarytenoid	5.59 ± 0.76	6.99 ± 0.78	9.26 ± 0.45
Right false vocal fold	6.53 ± 0.67	10.03 ± 0.69	11.05 ± 0.39
Left false vocal fold	6.95 ± 0.75	10.54 ± 0.77	11.42 ± 0.44
Right true vocal fold	11.31 ± 0.93	13.43 ± 0.95	17.41 ± 0.54
Left true vocal fold	10.89 ± 0.95	13.67 ± 0.98	17.08 ± 0.56

^aAverage ± standard error (SE).

Figure 4.

Comparison of average hue values in the 7 regions of the larynx. Region 1, right arytenoid; 2, left arytenoid; 3, interarytenoid; 4, right false vocal fold; 5, left false vocal fold; 6, right vocal fold; 7, left vocal fold.

Figure 5.

Comparison of the average hue values in the right side of larynx of different groups. Region 1, right arytenoid; 2, interarytenoid; 3, right false vocal fold; 4, right vocal fold.

Figure 6.

Comparison of the average hue values in the left side of larynx of different groups. Region 1, left arytenoid; 2, interarytenoid; 3, left false vocal fold; 4, left vocal fold.

Then, we applied the analysis of variance analysis to find if there was significant difference between different groups within each region (shown in Table 2). From these, we could see that:

In the regions of both sides of the true vocal folds and interarytenoid, there was no significant difference between positive group and negative group (P = .34 and .13 for right and left sides, P = .60 for interarytenoid);

In the regions of both sides of false vocal folds, there was no significant difference between negative group and control group (P = .61 and .97 for right and left sides);

In the regions of both sides of arytenoids, there was neither significant difference between positive group and negative group (P = .56 and .16 for right and left side), nor between negative group and control group (P = .21 and .16 for right and left sides);

In other comparisons, there were significant differences (P < .05).

Table 2.

Results of the ANOVA Analysis Between the Positive, Negative, and Control Groups in the 7 Laryngeal Regions.

Region		Positive	Negative	Control
Right true vocal fold	Positive	–	0.34	0.00^a
	Negative	0.34	–	0.00^a
	Control	0.00^a	0.00^a	–
Left true vocal fold	Positive	–	0.13	0.00^a
	Negative	0.13	–	0.01^a
	Control	0.00^a	0.01^a	–
Right false vocal fold	Positive	–	0.00^a	0.00^a
	Negative	0.00^a	–	0.61
	Control	0.00^a	0.61	–
Left false vocal fold	Positive	–	0.00^a	0.00^a
	Negative	0.00^a	–	0.97
	Control	0.00^a	0.97	–
Right arytenoid	Positive	–	0.56	0.00
	Negative	0.56	–	0.21
	Control	0.00^a	0.21	–
Left arytenoid	Positive	–	0.16	0.00^a
	Negative	0.16	–	0.16
	Control	0.00^a	0.16	–
Interarytenoid	Positive	–	0.60	0.00^a
	Negative	0.60	–	0.04^a
	Control	0.00^a	0.04^a	–

Abbreviation: ANOVA, analysis of variance.

^aP < 0.05.

Discussion

Jacob et al reported in 1991 that posterior laryngitis without ulceration could be associated with gastropharyngeal reflux as documented by pH monitoring.⁶ In a prospective study, Kamel et al reported the elimination of symptoms and signs of chronic posterior laryngitis during treatment with omeprazole.¹¹ Hanson et al reported that the color value of the posterior laryngeal structures and vocal folds in the patients who were treated for reflux laryngitis were significantly changed.¹² However, the mechanism of how acid reflux affects different regions in larynx has not been elucidated clearly.

In our previous study, we demonstrated that there might be correlations between laryngopharyngeal reflux disorder (LPRD) and the hue value of laryngoscopic images, where hue values of patients with LPRD tend to be lower. Furthermore, we found that different regions of larynx have different hue values. The regions of both arytenoids and the interarytenoid regions are closer to the inlet of esophagus. The hue value of these 2 regions are almost the highest of the 7 regions, except for the true vocal fold and false vocal fold regions.^7,12,13 It was hypothesized that this phenomenon was correlated with the presence of air in the reflux was predominantly during upright position and the pattern of reflux out of the upper esophageal sphincter (UES) was spilling, spurting, and spraying.^14,15

Based on Figures 4 through 6 and Table 2, there were significant differences between hue values in every region when comparing the positive group and the control group. When comparing the negative and the control group, there was no significant difference in both sides of arytenoid regions and false vocal folds. Additionally, there was no significant difference when we compared positive with negative groups in both sides of vocal folds, arytenoids, and interarytenoid region.

Comparing to the control group, the negative group has no significant different hue value in both sides of vocal folds, arytenoids, and interarytenoid region with the positive group. Indicating the same level of inflammation of these regions between the 2 groups. Suggesting that these regions may be affected by acid or pepsin already in the negative group. As previously stated, the negative group was comprised of patients with some symptoms of LPRD and RSI > 13, but were negative in MII-24 h pH monitoring. So, we hypothesized that this group might be presenting a developing stage of LPRD. We should give them the advisement on changing lifestyle, particularly on diet and physical exercise, also we should follow them up to learn if the pH monitoring result turn to be positive in the future.

Moreover, our results show that there were significant differences between the negative and control group in the true vocal fold regions, but no significant difference in both false vocal folds regions. This phenomenon was also reported by Hanson et al, who thought that the vocal fold regions were much more sensitive to the acid.¹²

While reflux finding score (RFS) covers only a few characteristics of a laryngoscope image and is judged subjectively, image analysis based on computer recognition and machine-learning could process the images with more objective and comprehensive parameters. Furthermore, this method gives us a possibility to grade the LPR disease. Although there may be biases of such as different angles, distance, or luminance, they affect the RFS either. Future work should focus on collecting more patients and tracing changes in laryngoscopic images, including studies within the negative group.

There are some limitations of this study. There was not a consensus on diagnosing LPRD through MII-24 h pH monitoring. The normative values for all parameters acquired during MII-24 h pH monitoring have not been established yet.⁵ Furthermore, there was still the possibility of a misdiagnosis to the false positive and false negative results.¹⁶ Finally, MII-24 h pH monitoring was regarded as a poor predictor of the severity of patients’ symptoms and signs, which should have helped us to determine the developing stage of the disease.¹⁷

Conclusion

This study further demonstrates that hue values of separate laryngeal regions are different. In negative group, the hue values of interarytenoid region are similar with positive group, and the sensitivity to the acid are different from different regions. The sensitivity in the true vocal folds may be present.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Beijing Health Promotion Association, A71490-05. The authors have no other funding, financial relationships, or conflicts of interest to disclose.

ORCID iD

Yan Yan

References

Koufman

. The otolaryngologic manifestations of gastroesophageal reflux disease (GERD): a clinical investigation of 225 patients using ambulatory 24-hour pH monitoring and an experimental investigation of the role of acid and pepsin in the development of laryngeal injury. Laryngoscope. 1991;101(4 pt 2 suppl 53):1–78.

Vaezi

Hicks

Abelson

Richter

. Laryngeal signs and symptoms and gastroesophageal reflux disease (GERD): a critical assessment of cause and effect association. Clin Gastroenterol Hepatol. 2003;1(5):333–344.

DeVault

. Overview of therapy for the extraesophageal manifestations of gastroesophageal reflux disease. Am J Gastroenterol. 2000;95(8 suppl 1):S39–S44.

Postma

Johnson

Koufman

. Treatment of laryngopharyngeal reflux. Ear Nose Throat J. 2002;81(9 suppl 2):24–26.

Asaoka

Nagahara

Matsumoto

Hojo

Watanabe

. Current perspectives on reflux laryngitis. Clin J Gastroenterol. 2014;7(6):471–475.

Jacob

Kahrilas

Herzon

. Proximal esophageal pH-metry in patients with “reflux laryngitis.” Gastroenterology. 1991;100(2):305–310.

Witt

Chen

Mielens

, et al. Detection of chronic laryngitis due to laryngopharyngeal reflux using color and texture analysis of laryngoscopic images. J Voice. 2014;28(1):98–105.

Al-Ramahi

Liu

Yan

Jiang

. Validation of the laryngopharyngeal reflux color and texture recognition compared to pH-probe monitoring. Laryngoscope. 2017;127(3):665–670.

Dobhan

Castell

. Normal and abnormal proximal esophageal acid exposure: results of ambulatory dual-probe pH monitoring. Am J Gastroenterol. 1993;88(1):25–29.

10.

Muderris

Gokcan

Yorulmaz

. The clinical value of pharyngeal ph monitoring using a double-probe, triple-sensor catheter in patients with laryngopharyngeal reflux. Arch Otolaryngol Head Neck Surg. 2009;135(2):163–167.

11.

Kamel

Hanson

Kahrilas

. Omeprazole for the treatment of posterior laryngitis. Am J Med. 1994;96(4):321–326.

12.

Hanson

Jiang

Chi

. Quantitative color analysis of laryngeal erythema in chronic posterior laryngitis. J Voice. 1998;12(1):78–83.

13.

Wan

Yan

, et al. LPR: how different diagnostic tools shape the outcomes of treatment. J Voice. 2014;28(3):362–368.

14.

Babaei

Bhargava

Mittal

. Upper esophageal sphincter during transient lower esophageal sphincter relaxation: effects of reflux content and posture. Am J Physiol Gastrointest Liver Physiol. 2010;298(5):G601–G607.

15.

Wang

Jimin

Songtao

, et al. Gastroesophago-larygotracheal reflux and its syndrome. Clinical Misdiagnosis & Mistherapy. 2010;23(5):401–405.

16.

Wang

. Application of 24-hour multichannel intraluminal impedance in combination with pH monitoring for the diagnosis of laryngopharyngeal reflux: a pilot study. Zhonghua Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2015;50(7):564–568.

17.

Noordzij

Khidr

Desper

Meek

Reibel

Levine

. Correlation of Ph probe-measured laryngopharyngeal reflux with symptoms and signs of reflux laryngitis. Laryngoscope. 2002;112(12):2192–2195.