Printer-friendly version

Esophageal reflux and the larynx

There is a seemingly pervasive belief that all or nearly all voice disorders are caused by reflux. That is, some ENT doctors and some laryngologists believe that all voice disorders are caused by reflux. As well, there seems to be a tendency on the part of many ENTs and some laryngologists to attribute hoarseness to reflux when nothing else obvious stands out.

I believe that the cause of this current diagnostic epidemic (hoarseness is caused by esophageal reflux) is due to my colleague's reliance on diagnosing voice disorders solely by looking at the voice box. However, problems with the voice box are nearly always functional in their manifestation and not always easily visualized. In a common scenario, a patient’s complaint of hoarseness is evaluated solely by a visual examination of the larynx. The physician pronounces the voice box “red” or states that there is swelling on the back of the voice box between the vocal cords. A prescription is written for an anti-reflux medication and a dietary and physical regimen such as avoiding caffeine and chocolate, etc. and raising the head of the bed. There are even handouts provided that state a red larynx may be caused by reflux.

Acid reflux or laryngopharyngeal reflux (LPR) are terms used to describe the belief that stomach contents regurgitate frequently back up the esophagus and enter the back of the throat. There is the belief that the lining of the voice box or larynx cannot protect itself from the chemicals or acid present in the stomach and that causes some sort of reaction or burn on the voicebox. Nearly all of us has thrown up at some point and tasted the contents of our stomach. Some of us have heartburn which is frequently attributed to the stomach contents regurgitating up into the esophagus.

So there are at least two issues here. Do people regurgitate acid or stomach contents up into their throat frequently? If they do, does it cause problems with the voice?

Color Red

Before we answer these questions, there is the problem that the color red has no impact whatsoever on the function of the voice. The skin (mucosa) lining the vocal tract is always red or pink – look inside your mouth. In fact the pronouncement of “redness” suggests that the physician has an inadequate view of the vocal cords. Typical fiberoptic endoscopes blur the image of the throat lining with their pixelated view, and a mirror exam of the larynx is so distant that the best that can be seen is often the general sense of the color of the mucosa: red. However, with newer cameras, the fine detail of the mucosa lining the voice box can be made out. The mucosa is actually translucent and beneath the mucosa can be seen tiny blood vessels. They can be thin, straight, tortuous, dilated, pedunculated, mal-oriented, etc. The vessels themselves are always red from their content (blood) and the more vessels there are, the redder the vocal cords.


The second problem lies in describing the mucosa in the area between the vocal cords which is very loosely attached to the underlying structures (cartilage and muscle). The mucosa is loosely attached so that one set of cartilage may move over the other without restriction. This loose mucosa can swell and change color. It can be smooth and whitish or it can be redundant and floppy, sometimes resembling elephant skin vaguely, and called pachydermia. However, this tissue is not involved in making sound, so it's characteristics and the description of it are generally a diversion from the problem at hand – hoarseness.


A laryngologist has an algorithm for understanding the larynx or voicebox. The predominant algorithm in existence today and the one still being widely taught is to use the eye to identify visual abnormalities on the voice box and then to attribute the complaint (hoarseness) to the visual anomaly. This algorithm is fraught with problems, not the least of which is that a visual non-regularity of the voice box is often no more than randomly associated with a functional change in the voice. I even see patients with big bumps on their voice box that do not need to be removed because the bumps are not causing the patient’s complaints and are benign. The actual cause of the hoarseness is missed and the hoarseness persists even after "treatment" because attention was focused on the visual problem and not on the problem generating the abnormal "voice."

It is true that this visual algorithm has worked in many cases. Clearly a big lump on the edge of the vocal cords is highly likely to be impairing the vibration of the vocal cords and thus the voice. So the algorithm works often enough, but there is considerable room for improvement.

I believe there is a much more suitable algorithm for understanding the voice, that will one day replace our present visual assessment of the voice. If you think of the vocal cords as strings that vibrate (analogous to many stringed instruments from piano to violin to guitar) and when they vibrate without any interference, they produce what on paper would appear be a sine wave, a smooth curvilinear line oscillating about an axis. However, the vocal cords act more like a wind instrument than a string instrument - in that they leak air when they are vibrating. The air passing between the vocal cords, air that is not involved in vibrating them, flows turbulently and generates white noise. That is, a whisper that involves no vocal cord motion at all, is almost completely turbulent air flow and thus white noise.

If we can imagine an opera singer performing an aria that is clear and smooth sounding to our ears, then nearly all of the air passing through the vocal cords is producing a very regular sine wave and there is almost no turbulent flow. Most phonations are in between these extremes of the aria and the whisper. Some amount of white noise is often well tolerated and considered part of a normal voice. For example a folk singer often leaks a fair amount of air while singing and we consider that breathy style a pleasant type of vocalization depending on our musical tastes. So air leak can be a component of voicing. We can think of the pure oscillation as the signal – the desired part, the component of sound that we can say is a certain pitch on the piano, say middle C. We can think of the air leak as noise and describe a voice as having a certain signal to noise ratio (SNR). At some point as we increase the noise relative to the signal we start to call that sound hoarseness. The irregular part (noise) of the vibration may start to overwhelm the regular part of the vibration or signal.

Our brains are designed to filter out the signal from the noise. That is why we can listen to a radio station that is not entirely clear and still enjoy it to a degree. At some point, as we get farther away from a given radio station, depending on our level of interest in what is on that radio station, we give up on listening to it. That is analogous to the point at which a person with a “hoarse” voice seeks medical care.

This brings me to a third point. I have seen over 5,000 voice patients and I have attributed the cause of the hoarseness to esophageal reflux induced laryngitis in only two cases. There may be problems with my assumption. First almost every patient that I see for vocal troubles is already being treated for reflux with a medication called a proton pump inhibitor {eg: Omeprazole (brand names: Losec®, Prilosec®) Lansoprazole (brand names: Prevacid®, Zoton®, Inhibitol®)Esomeprazole (brand names: Nexium®) Pantoprazole (brand names: Protonix®, Somac®, Pantoloc®) Rabeprazole(brand names: Rabecid®, Aciphex®, Pariet®)} so perhaps everyone who would get better on the medication has already done so and they do not come to see me and I only see the failures (I do not really believe this).

On the other hand, I have never encountered an audible and visible vocal cord lesion that would resolve on the above medication and then recur when the medication stopped and resolve again when the medication was resumed – much as Koch’s postulates are utilized to determine whether a specific agent is truly the cause of a disease. Additionally, I nearly always find some other reasonable explanation for hoarseness or the change in the voice and upon treatment, the hoarseness resolves, convincing me that reflux is not the etiology of almost all voice disorders. Additionally, I have never seen a physician proponent of the reflux explanation of hoarseness ever stand up at a meeting, show me a video of a "red" larynx before treatment with a hoarse quality and then show me an after video of a "non-red" larynx without hoarseness.

So, if you have been diagnosed with reflux as the cause of your vocal disorder and especially if you are not improving, it is not likely that you need more of the medication or that you need an anti-reflux surgery to fix your hoarse voice. It is quite likely that you need a more precise diagnosis. 

21 June 2010: See also article on over-medication by doctors with anti-reflux medication - The Reflex To Treat Reflux

12 February 2012: A singer and I ran a little experiment since the singer claimed that reflux medicine have helped his voice a great deal. I saw him three times, and each time I did not know whether he was taking any anti-reflux medication or not. After the three examinations, he revealed to me that he was taking ranitidine (Zantac) for two weeks prior to one of the examinations. The notable finding was that - when he was not taking the medication, his secretions were very thick and sticky. When he was taking the medication ranitidine, an Histamine-2 blocker, his secretions were very thin. I don't know that acid reflux played any role, but it seems that ranitidine, a histamine blocker seemed to thin the secretions.

This might explain why a great many laryngeal disorders seem partially improved, at least for some pure time, with treatment by medications such as ranitidine (Zantac) or perhaps cimetidine (Tagamet). Whenever there is an impairment of vibration on the vocal cords, secretions will tend to collect at the point of impairment. For example, if there is a nodule or elevation on the vocal cord, secretions will collect at the nodule because the does not vibrate as well at the nodule. Similarly, bowed vocal cords will tend to collect secretions at either end. Consequently, the thinning of secretions would lead to less stickiness at the ends of the vocal cords.

At this point, it seems to me that at least one class of anti-reflux medications, histamine blockers, thin secretions and to that extent can improve the voice by this mechanism.