Figure 11–1. Normal stomach. (Republished from Quick Reference to Upper GI Motility with the permission of Janssen Pharmaceutica.)

The lower esophageal sphincter (LES), like most of the gastrointestinal (GI) tract, consists entirely of smooth muscle. This sphincter is much more rounded in its closure, yet still demonstrates some degree of radial asymmetry, having the higher pressures in the posterolateral direction.³⁷ Innervation of the LES originates from the dorsal motor nucleus of the brain-stem, and the efferent fibers are carried through the vagus nerve and synapse in the myenteric plexus in the region of the LES.

The muscular wall of the esophagus is composed of an inner circular and an outer longitudinal layer, with no serosa overlying the muscle layers. The UES and the upper portion of the tubular esophagus are primarily striated muscle. Recent studies have indicated that smooth muscle occurs in the upper 4 to 5 cm of the human esophagus, although it is quite variable in different individuals and in the different muscle layers. Consistently, greater than the distal half of the human esophagus is entirely smooth muscle.³⁸ Like the LES, the smooth muscle portion of the tubular esophagus is innervated primarily via the vagus nerve from neurons arising in the dorsal motor nucleus connecting to the myenteric plexus.

Physiology

Swallowing, or deglutition, has 3 stages: the oral (voluntary) stage, the pharyngeal (involuntary) stage, and the esophageal stage. These 3 closely coordinated and combined processes are regulated through the swallowing center in the medulla.³⁹

Oral Stage

This preparatory stage includes mechanical disruption of the food and mixing with salivary bicarbonate and enzymes (amylase, lipase). It is an essential process by which the swallowing mechanism is primed. Ingested food is voluntarily moved posteriorly by pistonlike movements of the tongue muscles, forcing the food bolus toward the pharynx and pushing it backward and upward against the palate. Once the food has been delivered to the pharynx, the process becomes involuntary. The oral, preparatory stage obviously requires proper functioning of the striated muscles of the tongue and pharynx and is the stage of swallowing that is likely to be abnormal in patients with neurologic or skeletal muscle disease. Appropriate mentation is also necessary.

Pharyngeal Stage

During this stage of swallowing, the food is passed from the pharynx, through the UES, and into the proximal esophagus. This involuntary process requires the finely tuned coordinated sequences of contraction and relaxation, resulting in transfer of the ingested material, while protecting the airway. The presence of food in the pharynx stimulates sensory receptors, which send impulses to the swallowing center in the brainstem. The central nervous system (CNS) then initiates a series of involuntary responses that include the following:

1. The soft palate is pulled upward and closes the posterior nares.

2. The palatopharyngeal folds are pulled medially, limiting the opening through the pharynx.

3. The vocal folds are closed, and the epiglottis swings backward and down to close the larynx.

4. The larynx is pulled upward and forward by the muscles attached to the hyoid bone, stretching the opening of the esophagus and UES.

5. The UES relaxes. Active relaxation of the usually tonic cricopharyngeus is essential to permit the passive opening of the UES created by the movement of the larynx.

6. Peristaltic contraction of the constrictor muscles of the pharynx produces the force that propels food into the esophagus.

This sequence is a coordinated mechanism that includes impulses carried by 5 cranial nerves. Sensory information to the swallowing center is carried along cranial nerves V, VII, IX, and X. The motor responses from the swallowing center are carried along cranial nerves V, VII, IX, X, and XII and also the ansa cervicalis (C-1 and C-2). This intricate process takes just over 1 second from start to finish and requires coordination of pharyngeal contraction and UES relaxation (Figure 11–3). The UES is only open for approximately 500 milliseconds.

Esophageal Stage

The main function of the esophagus is to transport ingested material from the mouth to the stomach. This active process requires contraction of both the longitudinal and circular muscles of the tubular esophagus and coordinated relaxation of the sphincters. At the onset of swallowing, the longitudinal muscle contracts and shortens the esophagus to provide a structural base for the circular muscle contraction that forms the peristaltic wave. The sequential contraction of esophageal circular smooth muscle from proximal to distal generates the peristaltic clearing wave. The neuromuscular control of this activity will be described below. As opposed to other GI tract smooth muscle, the esophageal smooth muscle has a unique electrical activity pattern, showing only spiked potentials without underlying slow waves. Circular muscle contractions can be characterized into 3 distinct patterns:

1. Primary peristalsis. This is the usual form of a contraction wave of circular muscle that progresses down the esophagus and is initiated by the central mechanisms that follow the voluntary act of swallowing. It follows sequentially the pressure generated in the pharynx and requires approximately 8 to 10 seconds to reach the distal esophagus. The LES relaxes at the onset of the swallow and remains relaxed until it contracts as a continuation of the progressive peristaltic wave. These pressure relationships are shown in Figure 11–4.

2. Secondary peristalsis. This represents a peristaltic contraction of the circular esophageal muscle, which begins without central stimulation. That is to say, it originates in the esophagus as a result of distention and will usually continue until the esophagus is empty. Some food, particularly solid material, requires more than the single primary peristaltic wave for eventual clearance. This is accomplished by the secondary peristaltic waves. Thus, secondary peristalsis is the mechanism for clearing both ingested material and also material that is refluxed from the stomach. Experimentally, secondary peristalsis can be demonstrated by inflating a balloon in the mid-to-upper esophagus.

3. Tertiary contractions. This contraction pattern is identified primarily during barium x-ray studies of the esophagus. It represents a nonperistaltic series of contractile waves that appear as localized segmented indentations in the barium column. It has no known physiologic function.

One of the interesting phenomena seen in the esophagus occurs during the process of rapid sequential swallowing (10 seconds or less between successive voluntary swallows). This process results in inhibition of peristalsis, so-called “deglutitive inhibition.” Peristalsis will be suspended during the continuation of a series of rapid swallows and a large “clearing wave” will occur at the completion of the swallows (Figure 11–5). This phenomenon occurs because of the inhibitory neural discharge that arises from the central swallowing center during swallowing, and also because the esophageal musculature shows a refractoriness, demonstrated to persist for up to 10 seconds.⁴⁰

Importance of the Sphincters

The esophagus is located in the thorax and has negative pressure relative to pressures in the pharynx proximally and the stomach distally. Therefore, the action of the sphincters must maintain constant closure to prevent abnormal movement of air or food into the esophagus. In the absence of a tonically contracted UES, air will flow freely into the esophagus during inspiration. In the presence of a weak LES, gastric contents are not inhibited from refluxing into the distal esophagus, particularly in the recumbent position. Pressure relationships, in and around the esophagus and its sphincters, are shown in Figure 11–6.

Upper Esophageal Sphincter

The UES maintains a constant closure with its strongest forces directed in the anterior-posterior orientation of the sling-shaped attachment of the cricopharyngeus to the cricoid cartilage. Normal pressures in the UES are approximately 100 mm Hg in the anterior-posterior direction and approximately 50 mm Hg laterally.³⁶

Lower Esophageal Sphincter

The tonically contracted LES normally maintains a closing pressure 10 to 45 mm Hg greater than the intragastric pressure below. By convention, LES pressure is measured as a gradient in mm Hg higher than intragastric pressure, which is used as a zero reference (Figures 11–7 and 11–8). At the time of swallowing, the LES relaxes promptly in response to the initial neural discharge from the swallowing center in the brain and stays relaxed until the peristaltic wave reaches the end of the esophagus and produces sphincter closure. During relaxation, the pressure measured within the sphincter falls approximately to the level of gastric pressure; this is by definition “complete” relaxation. Although there has been much controversy over the years, it is now generally accepted that the LES does not have to be located within the diaphragmatic crus to maintain a constant closing pressure. Thus, the presence of a sliding hiatal hernia is not necessarily detrimental to the physiologic function of this sphincter.

The LES maintains 2 important physiologic functions; the first is its role in prevention of gastroesophageal reflux and the second is its ability to relax with swallowing to allow movement of ingested material into the stomach. The mechanism by which the circular smooth muscle of the LES maintains tonic closure has been a subject of considerable investigation over many years. At present, this is felt to be predominantly the result of intrinsic muscle activity, because investigations in animals have demonstrated that resting LES tone persists even after the destruction of all neural input by the neurotoxin tetrodotoxin.⁴¹ In addition, truncal vagotomy does not affect resting LES pressure in humans. Calcium-channel-blocking agents, which exert their effect directly on the circular smooth muscle, will produce decreases in LES pressures in animals and humans.^42,43 There also appears to be some cholinergic tone present in many animal species and in humans, as an injection of atropine or of botulinum toxin (Botox, Allergan) has been shown to produce marked decreases in resting LES pressure.^44,45

The mechanism of relaxation of the LES in response to a swallow has also been a subject of considerable investigation and controversy. The precise neurotransmitter responsible for this response is not definitely known. It is clear that it is not a classic cholinergic or adrenergic agent, because specific pharmacologic blockade of these mechanisms does not inhibit LES relaxation. This is a neural event. It can be reproduced in animals by stimulation of the vagus nerve, and relaxation is inhibited by tetrodotoxin.⁴⁶ Their relationships are summarized in Figure 11–9. Recent studies indicate that the neurotransmitter might be a combination of vasoactive intestinal polypeptide (VIP) and nitric oxide.^46,47

The resting pressure of the LES is dynamic. Pressures measured over long periods of time indicate that LES pressure will vary considerably, even from minute to minute. Much of this is due to the effect of a variety of factors that modulate pressure. These include foods ingested during meals and other events such as cigarette smoking and gastric distention. The normal LES will respond to transient increases in intraabdominal pressure by raising its resting pressure to a greater degree than the pressure increases in the abdomen below. This normal protective mechanism guards against gastroesophageal reflux. In addition, many hormones and other peptide substances produced in the GI tract and in other areas of the body have been shown to affect LES pressure. These are summarized in Table 11–1. Many of these likely represent pharmacologic responses that have been shown to occur after intravenous injection or infusions of these substances in man or animals. Whether they represent truly physiologic actions has not been clarified in most cases. The strongest candidates for physiologic hormonal control of the LES are cholecystokinin, which helps explain the decrease in LES pressure seen after fat ingestion, and progesterone, which explains the decrease in LES pressure that occurs during pregnancy. Finally, various neurotransmitters and pharmacologic agents have been shown to affect LES pressure. These are summarized in Table 11–2. The modulation of LES resting pressure is a complex mechanism that involves the interaction of the LES smooth muscle, neural control, and humoral factors.⁴⁸

Controls of Esophageal Peristalsis

As noted above, esophageal peristalsis is controlled via afferent and efferent neural connections through the swallowing center in the medulla. This central mechanism regulates the involuntary sequence of muscular events that occurs during swallowing (Figures 11–10 and 11–11) and simultaneously inhibits the respiratory center in the medulla so that respiration is stopped during the pharyngeal stage of swallowing.

The direct innervation to the striated muscle of the pharynx and upper esophagus is carried via fibers from the brainstem (nucleus ambiguus) through the vagus nerve. The innervation of the smooth muscle of the distal esophagus and LES arises from the dorsal motor nucleus of the vagus and is carried through cholinergic visceral motor nerves to ganglia in the myenteric plexus (also known as the Auerbach plexus). Noncholinergic, nonadrenergic inhibitory nerves also course within the vagus.

The myenteric plexus in the esophageal portion of the enteric nervous system (the “brain in the gut”) receives efferent impulses from the central nervous system (CNS) and sensory afferents from the esophagus. Thus, impulses travel in 2 directions through this modulating area, which interconnects and regulates signals that result in normal peristalsis in the smooth muscle of the esophagus. One manifestation of the afferent control is the regulation of peristaltic squeezing pressures, to some degree, by the size of the ingested bolus. In addition, dry swallows often fail to provide adequate stimulation for the action of the myenteric plexus as the primary regulatory mechanism of esophageal peristalsis in the smooth muscle portion, as shown by observations that bilateral cervical vagotomy in animals does not abolish peristalsis in this area.

Interesting results have been obtained from in vitro studies of esophageal smooth-muscle preparations.⁴⁹ Using muscle from the opossum esophagus, it has been shown that the longitudinal smooth muscle demonstrates a sustained contraction during electrical field stimulation; this is called the “duration response.” This response is neural and cholinergic, because it can be blocked with both atropine and tetrodotoxin. The circular smooth muscle of the opossum esophagus shows a quite different response. With the onset of electrical stimulation, there is a brief, small contraction at the beginning of the stimulus, known as the “on-response.” This response is quite variable and has no known physiologic role. The on-response is followed by a much larger contraction that occurs after the termination of the stimulus, known as the “off-response.” This response is also neural in origin but is not cholinergic, because it is blocked only by tetrodotoxin and not atropine. Muscle strips taken from different segments of the smooth muscle portion of the esophageal body show progressively longer intervals for the off-response contraction following stimulation while moving distally in the esophagus. This phenomenon has been called the “latency gradient.” These concepts are shown in Figure 11–12.

It has been proposed that these in vitro experiments from the opossum esophagus may help to explain some of the mechanisms of the development of normal peristalsis in the human smooth-muscle esophagus. With the initial swallowing event, an inhibitory neural discharge is sent to the circular smooth muscle of the entire esophagus. The LES relaxes from its resting tonic state. The remainder of the esophageal smooth muscle is already relaxed and shows no measurable change. Rebound contraction occurs following the end of the brief stimulus (the off-response). The latency of the gradient for this off-response, progressing distally down the esophagus, produces the peristaltic contraction wave. Although this concept does not entirely explain all of the phenomena that have been observed in human peristaltic activity, these in vitro observations are consistent with many aspects of normal human physiology. One example is the deglutitive inhibition referred to above. With repetitive swallowing at frequent intervals, the successive inhibitory neural impulses from the swallowing center prevent the contractions of the smooth-muscle portion of the esophagus until the last swallow occurs. The off-response and the latency gradient then allow the single peristaltic clearing wave that usually follows.

Other Considerations

When gastric pressure becomes greater than LES pressure, reflux occurs. It must be remembered, however, that mechanical sphincter dysfunction is not the only cause of reflux symptoms. Gastric pathology (ie, hypersecretion and alkaline gastroesophageal reflux), motility disorders, and other conditions such as impaired gastric emptying (Figures 11–13 and 11–14) must be considered.

Clinical Presentation and Epidemiology of Gastroesophageal Reflux Disease: An Overview

Gastroesophageal reflux disease (GERD) is a spectrum of disease best defined as symptoms and/or signs of esophageal or adjacent organ injury secondary to the reflux of gastric contents into the esophagus or above into the oral cavity or airways. GERD is a common disorder often encountered in clinical practice and presents with a multitude of symptoms. Injury caused by GERD is defined based on symptoms or organ damage, which include esophagitis; inflammation of the larynx, pharynx, and oral cavity; or acute and/or chronic pulmonary injury. This section presents an overview of GERD including typical, atypical, and extraesophageal presentations.

Typical Symptoms

The typical or classic symptoms of GERD are heartburn (pyrosis), defined as substernal burning occurring shortly after meals or on bending over that is relieved with antacids and regurgitation (the spontaneous return of gastric contents into the esophagus or mouth). When present together, heartburn and regurgitation establish the diagnosis with greater than 90% certainty. In clinical practice, heartburn is a daily complaint in 7 to 10% of the population in the United States and at least monthly in about 40 to 50%.^50–52 Over 20 million people in the United States have heartburn at least twice a week and use antacids or other over-the-counter (OTC) antireflux products on a regular basis. Regurgitation is experienced weekly by about 6% of the population according to one study.⁵¹ In the same study, either heartburn or regurgitation was present weekly in 20% of patients surveyed and monthly in 59%. The prevalence of heartburn appears to decrease slightly with increasing age.

Classic heartburn is described typically by patients as a burning sensation under the breastbone with radiation upward toward the throat or mouth. Heartburn generally occurs 1 to 2 hours after meals, following heavy lifting or on bending over. Big meals, spicy foods, citrus products, such as a grapefruit and orange juice, and meals high in fat are more likely to produce heartburn. Colas, coffee, teas, and even beer may have an acidic pH and cause symptoms when ingested. Heartburn may also be caused by medications (Table 11–3). Meals eaten late in the evening, close to bedtime, or taken with alcohol make patients more prone to nighttime symptoms. Patients report often that their symptoms are relieved with an over-the-counter antacid preparation, H₂-receptor antagonists, or even by drinking water.

Although heartburn is often associated with the presence of regurgitation, the spontaneous experience of an acidic or bitter taste in the throat or mouth, these are not synonymous symptoms. Heartburn should not be confused with dyspepsia or a more vague epigastric distress usually localized in the upper abdominal or lower substernal area and associated with nausea, bloating, or fullness after meals. Although dyspepsia (epigastric discomfort) may be a symptom of GERD, it is neither as sensitive nor as specific a symptom as heartburn. The generic terminology of “acid indigestion” used to encompass all symptoms related to GERD is inappropriate; these symptoms must be distinguished for accurate diagnosis and therapy. Waterbrash, the sudden filling of the mouth with a clear, salty fluid, should not be confused with heartburn. This symptom reflects the increase in salivary secretion seen as a reflex response to reflux or regurgitation of gastric acid into an inflamed distal esophagus.

Heartburn is a highly specific symptom of GERD, although GERD is not the only cause. For example, a heartburnlike symptom, suspected to be due to esophageal stasis from outflow obstruction, is described often in patients with achalasia. It is felt that fermentation of undigested food in the esophagus coupled with inflammation may create a heartburnlike sensation in the absence of true GERD. Functional heartburn may also be a component of irritable bowel syndrome. However, if heartburn is the only presenting esophageal symptom, it is likely due to GERD.

Despite the sensitivity and specificity of these two symptoms for the diagnosis of GERD, neither the presence of heartburn and/or regurgitation nor the frequency of these symptoms is predictive of the degree of endoscopic damage to the distal esophagus. Many patients with daily heartburn will have no endoscopic abnormalities. The frequency of heartburn usually does not correlate with the severity of GERD, although nocturnal heartburn suggests the possibility of erosive esophagitis. Only 50 to 60% of patients presenting to a physician with heartburn will have erosive esophagitis seen on a diagnostic endoscopic examination; the remainder will be diagnosed as having nonerosive GERD.⁵³ Severe disease, including Barrett’s esophagus and peptic strictures, may present with infrequent or absent complaints of heartburn.

Most patients with esophagitis will not have progression of their disease beyond the severity of esophagitis seen at the time of initial endoscopy. In a series of 701 patients followed for up to 29 years, only 23% progressed to a more serious grade of esophagitis.⁵⁴ The patient with reflux symptoms and no evidence of esophagitis (nonerosive GERD) has even less likelihood of esophageal disease progression, with less than 15% of patients progressing to a higher grade over 6 months.⁵⁵

Regurgitation is often associated with heartburn and GERD. When the 2 symptoms are present together, the diagnosis of GERD is highly likely. Regurgitation without heartburn should raise suspicion of Barrett’s esophagus (in which acid sensitivity is reduced), achalasia, or other esophageal abnormality. Regurgitation may also be seen as a more prominent symptom in extraesophageal manifestations of GERD, particularly in patients with pulmonary symptoms of reflux, and it may be an important prognostic factor in predicting outcome of therapy.^55–57 Regurgitation is often confused by patients as vomiting. The effortless return of food or fluid in the absence of nausea is an important distinction between these 2 symptoms. Esophageal and extraesophageal symptoms associated with GERD are outlined in Table 11–4.

Extraesophageal (Atypical) Symptoms

A number of so-called atypical or extraesophageal symptoms have been associated with GERD, including unexplained substernal chest pain without evidence of coronary artery disease (noncardiac chest pain), asthma, bronchitis, chronic cough, recurrent pneumonia, hoarseness, chronic posterior laryngitis, globus sensation, otalgia, aphthous ulcers, hiccups, and erosion of dental enamel. In contrast to heartburn and regurgitation, the prevalence of these atypical or extraesophageal symptoms and their frequency in the general population have not been systematically studied until recently. In a large population-based survey of Caucasians in Olmstead County, Minnesota,⁵¹ designed to assess the prevalence of GERD in the general population, unexplained chest pain was seen in 23% of the population yearly and in 4% at least weekly. The frequency of unexplained chest pain surprisingly decreased with age. Forty percent had symptoms for more than 5 years, and 5% reported severe symptoms. Asthma was reported in approximately 9%, bronchitis in approximately 20%, and chronic hoarseness in 15% of patients who had atypical GERD symptoms.

The association of these atypical symptoms with heartburn and regurgitation is controversial. In the Minnesota study,⁵¹ patients with heartburn and regurgitation had one or more atypical symptoms about 80% of the time. Atypical symptoms were more common in patients with frequent GERD symptoms compared to patients with no GERD symptoms. Heartburn or regurgitation was reported in more than 80% of the patients with unexplained chest pain and in 60% with globus sensation. The only exception was asthma. Approximately 60% of patients with asthma, bronchitis, hoarseness, and pneumonia had heartburn or regurgitation. The presence of heartburn is not predictive of otolaryngologic symptoms. However, in a case control study by the Veterans Administration, patients with a discharge diagnosis of erosive esophagitis had twice the prevalence of an associated otolaryngologic symptom as compared with control patients without esophagitis.⁵⁸ Observations in patients presenting with atypical GERD show that frequent heartburn and regurgitation are uncommon complaints; however, the absence of these typical symptoms should not preclude making a diagnosis. Prospective studies using endoscopy and ambulatory pH monitoring find GERD in as many as 75% of patients with chronic hoarseness,³² between 70 and 80% of asthmatics,^59,60 and in 20% of patients with chronic cough.⁶¹

Reflux is a well-recognized cause of atypical chest pain and may be responsible for many (most) of the symptoms in the 75 000 to 150 000 patients who undergo normal coronary angiography in the United States annually.⁶²

Approximately 45% of these patients with unexplained chest pain can be shown to have GERD.⁶³ Esophagitis in this population is less common, being seen in less than 10%.⁶⁴ Endoscopically, esophagitis is seen in 30 to 40% of patients with asthma^65,66 and about 20% of patients with reflux laryngitis. Distinguishing between cardiac and noncardiac chest pain due to GERD is difficult, and they may coexist in the same patient. All of the features of cardiac angina—tight, gripping, vicelike pain radiating to the neck, shoulder, or left arm and associated with exertion—may be seen with GERD also. Long episodes of pain (greater than 1 hour), pain relieved by eating, or pain awakening from sleep are more likely esophageal symptomatology. Antacids or H₂-blockers may relieve chest pain, later proven to be associated with coronary artery disease. It is therefore crucial to rule out cardiac disease before presuming GERD is the cause of chest pain. Omeprazole is an effective treatment for reflux-related, noncardiac chest pain⁶⁷; in patients with infrequent (less than 3 times per week), noncardiac chest pain, high-dose proton pump inhibitor (PPI) therapy, such as omeprazole, may be a sensitive, specific, cost-effective strategy for diagnosing GERD.⁶⁸ However, it must be remembered that not all patients respond to PPI medications. Therefore, persistent chest pain during PPI therapy does not definitively rule out GERD as the etiology of the chest pain as the evaluation of patients with unexplained chest pain remains complex.⁶⁹ For example, the rare syndrome X is a condition involving anginal chest pain with objective signs of ischemia on exercise stress testing or myocardial scintigraphy, but with normal coronary arteries on angiogram. Esophageal hypersensitivity (as opposed to gross functional abnormality) may be an associated finding in these patients; and acid suppression may improve the condition in many patients.⁷⁰ The complex relationship between reflux and chest pain remains incompletely understood. Ambulatory pH monitoring, particularly during continued PPI therapy, remains the gold standard for diagnosis of GERD in this population.

Seventy to 80% of patients with asthma will have associated GERD. Whether there is a cause-and-effect relationship or the coincidental presence of 2 diseases is not clear. A careful history will reveal heartburn or regurgitation in only 50%. Onset of asthma late in life, the absence of a seasonal or allergic component, and onset after a big meal, alcohol consumption, or exercise suggest GERD-related asthma. Empiric treatment with acid reflux suppression followed by pH testing in nonresponders was suggested in one study as the most cost-effective means of determining whether GERD is aggravating a patient’s asthma.⁷¹ This approach seems reasonable, because it has been demonstrated that PPI therapy in asthmatics with gastroesophageal reflux improves peak expiratory flow rate and quality of life.⁷²

Reflux is the third most common cause of chronic cough, after postnasal drip and bronchitis, and in many cases symptoms of postnasal drip may actually be associated with reflux. So the prevalence of reflux as an etiologic factor in chronic cough may be even higher than recognized previously. In the patient with cough, a normal chest x-ray, and no sinonasal postnasal drip, GERD should be considered as the most likely diagnosis.

Hoarseness is the most common otolaryngologic symptom of GERD. Most studies suggest that heartburn is present in only about 50% or less of otolaryngologic patients with extraesophageal manifestations (such as hoarseness) of GERD. However, some authors feel that a careful history may reveal heartburn to be present, at least occasionally, in as many as 75%.⁷³ Other associated symptoms of reflux laryngitis include halitosis, throat clearing, dry cough, coated tongue, globus sensation, tickle in the throat, chronic sore throat, postnasal drip, and others discussed later in this chapter. Difficulty in warming up the voice in the professional singer, voice fatigue, and intermittent laryngitis are associated symptoms. Erosion of the dental enamel may be due to GERD; however, its frequency is not known.

Complications of GERD

GERD may present with severe complications, including peptic stricture, ulceration, iron deficiency anemia, or more importantly Barrett’s esophagus. Barrett’s esophagus is a premalignant condition that involves a change from normal squamous epithelial lining to a metaplastic intestinal type epithelium with typical special staining characteristics. Estimates are that 2 to 10% of patients with GERD will have strictures,⁷⁴ and 10 to 15% will have Barrett’s esophagus.^53,75 Dysphagia, odynophagia (painful swallowing), and upper gastrointestinal (GI) bleeding may occur with these complications of GERD. Slowly progressive dysphagia, particularly for solids, suggests peptic strictures. Liquid and solid dysphagia suggests a GERD-related motility disorder secondary to erosive esophagitis, Barrett’s esophagus, or scleroderma. GERD-related motility disorders are seen with increased frequency in patients with otolaryngologic manifestations of GERD,⁷⁶ even though dysphagia is not usually a presenting symptom. Motility abnormalities pose important complications for the patient considering surgery (see Surgical Therapy for Gastroesophageal Reflux Disease). Odynophagia is uncommon in patients with reflux. Its presence suggests ulceration or inflammation, and it is seen most frequently in infectious or pill-induced esophagitis. Occasionally esophagitis may present with occult, upper gastrointestinal bleeding or iron deficiency anemia.⁷⁴ The frequency of these complications in patients with reflux laryngitis is not known.

Anatomy and Pathophysiology of LPR

Findings highlight the important fact that gastric acid can reflux through the esophagus to the larynx without causing esophageal injury in transit. It has been assumed that this is because distal esophageal mucosa has specialization and defense mechanisms that help it tolerate acid exposure. Esophageal protective mechanisms include peristalsis, which clears acid from the esophagus; a mucosal structure that may be specialized to tolerate intermittent acid contact; the acid-neutralizing capacity of saliva that passes through the esophagus and bicarbonate production in the esophagus, which has been recognized since the 1980s.^77–79 Interestingly, however, if some patients stop reflux treatment after a few months, classic dyspepsia and pyrosis seem to be present commonly when symptoms recur, although this clinically observed phenomenon has not been studied formally. It should be noted that the larynx and pharynx do not have protective mechanisms, such as those found in the esophagus, to protect against mucosal injury. So, exposure to acid and pepsin that might be of no consequence in the distal esophagus may cause substantial symptoms and signs in the larynx and/or pharynx of some patients. Interestingly, preliminary data by Axford et al suggest that laryngeal mucosa has different cellular defenses from those of esophageal mucosa.⁸⁰ They also suggested that there may be specific differences in MUC gene expression and carbonic anhydrase that suggest a pattern of abnormality in patients with LPR.

In addition to prolonged vocal warm-up time, professional singers and actors may complain of vocal practice interference, manifested by frequent throat clearing and excessive phlegm, especially during the first 10 to 20 minutes of vocal exercises or singing. Hyperfunctional technique during speaking and especially singing is also associated with reflux laryngitis. This is probably due to the vocalist’s unconscious tendency to guard against aspiration. Voice professionals can be helped somewhat in overcoming this secondary muscular tension dysphonia (MTD) through voice therapy with speech-language pathologists, singing voice specialists, and acting-voice specialists, but it is difficult to overcome completely until excellent reflux control has been achieved.

In addition to the paucity of typical GERD symptoms in patients with LPR, the tendency for under-diagnosis of LPR has been increased by 3 additional factors. First, the importance of various aspects of the physical examination is underappreciated. Posterior laryngitis and interarytenoid pachydermia are ignored frequently.⁸¹ It is even more common to fail to recognize the causal relationship between reflux and edema with little or no erythema, especially if the edema is diffuse rather than most prominent on the arytenoids. Second, therapeutic medication trials may fail because patients are undermedicated (eg, a PPI only once daily), and assessed before signs of laryngopharyngeal reflux have had time to resolve (which may require a few months or more). Third, routine tests for gastroesophageal reflux disease can be falsely negative. This problem involves not only barium esophagrams, the Bernstein acid-hyperperfusion test, and radionucleotide scanning, but also esophagoscopy and 24-hour pH monitor studies (depending on the norms used). Consequently, laryngologists must maintain a high index of suspicion in the presence of symptoms consistent with LPR, evaluate such patients aggressively, and interpret test results knowledgeably and with awareness of their sensitivities, specificities, limitations, and controversies.

Reflux Laryngitis and Other Otolaryngologic Manifestations of Laryngopharyngeal Reflux

Although the majority of otolaryngologists have recently acknowledged the importance of reflux in causing otolaryngologic disease, many authors have recognized the association for more than 2 decades.^{1,3,4,11,12,14,16,32,82–102} Otolaryngologists are becoming increasingly diligent about looking for erythema and edema of the mucosa overlying the arytenoid cartilages, suspecting laryngopharyngeal reflux (LPR) as the underlying problem, and treating it as the primary approach to therapy for various reflux-related conditions. However, additional information has shown that the term “laryngopharyngeal reflux” describes a complex spectrum of abnormalities; and it is important for physicians to understand the latest concepts in basic science and clinical care of LPR. Symptoms and signs related to reflux have been identified in 4 to 10% of all patients seen by otolaryngologists,^18,28,32,103 and it is probable that these estimates are low. Among patients with laryngeal and voice disorders, laryngopharyngeal reflux appears to be strongly associated with, or a significant etiologic cofactor in, about 50% of these patients. Many of the current concepts regarding reflux laryngitis and related controversies have been reviewed recently in the otolaryngologic and gastroenterologic literature.^104–106

Other Clinical Practice Considerations

Treatment considerations in reflux patients are discussed in greater detail later in this chapter Research into appropriate treatment regimens is ongoing, and extensive additional investigation on the consequences of reflux on the larynx and all of the other mucosal surfaces above the cricopharyngeus muscle is needed.

LPR has been found in association with otitis media with effusion not only in the pediatric population, but also in adults.¹⁰⁷ Habesoglu et al studied the effects of LPR on tympanoplasty failure.¹⁰⁸ They reported the results of 147 patients who underwent tympanoplasty who had laryngoscopic evidence of LPR and positive reflux finding scores. The authors suggested that these findings should be considered as a factor in the failure of tympanoplasty, and they recommended that reflux evaluation and treatment should be considered in the treatment of patients with chronic otitis media and ear disease, including eustachian tube dysfunction.^108,109 Other studies have reported that LPR and obesity are risk factors associated with lingual hypertrophy and obstructive sleep apnea.^110,111 A study by Corvo et al reported significant LPR in association with patients known to have Sjögren’s syndrome, confirmed by salivary analysis.¹¹²

Several studies have reported the presence of pepsin and amylase in oral and tracheal secretions.^113,114 Another report¹¹⁵ suggested alterations in mucin gene expression in laryngeal mucosa and/or laryngeal mucosal metaplasia due to MUC3, 4, and 5 MucSAC expression being downregulated in LPR may predispose the larynx to mucosal damage from gastric refluxate. Eckley et al¹¹⁶ studied salivary epidermal growth factor in adult patients with LPR pre- and post-treatment with PPIs. Saliva samples were obtained on 20 patients with LPR before and after a 16-week course of PPI therapy and compared to a control group of 12 healthy patients. Patients with LPR had lower salivary epidermal growth factor concentrations pre- and post-treatment compared with the healthy control group. The authors suggest that this finding indicates a defective mechanism of mucosal protection. Several recent studies have shown that scitintigraphic studies are useful as a screening tool for predicting aspiration in patients with LPR and GERD and in whom fundoplication surgery is being considered.^117,118 Endoscopy, olfactory function, and analysis of color and texture of laryngoscopic images also have been reported as part of the armamentarium in evaluating the presence of and/or predisposition to LPR.^119,120

In treating LPR patients, the clinician is faced with management decisions centered on the entire aerodigestive tract. For example, the presence of pepsin in nasal lavage fluid has been documented in patients undergoing endoscopic sinus surgery.¹²¹ LPR has been confirmed in these patients, suggesting a relationship between chronic rhinosinusitis and LPR. Nasal pepsin may prove to be an avenue of LPR screening.¹²¹ Evidence also links LPR and chronic otitis media,¹²² as discussed above, and surgical outcomes in laryngeal trauma surgery have proven better in patients with preoperative and postoperative PPI treatment.¹²³ Therefore, many different patient groups may benefit from reflux evaluation and treatment.

Pepsin has been found in tracheoesophageal puncture sites, as reported by Bock et al.¹²⁴ They performed tissue biopsy and collected secretions in 17 patients, 12 of whom had a history of GERD/LPR, and pepsin was detected in the majority of their patients. LPR also may play a role in the pathogenesis of dental disease and sleep disorders. Several studies have shown a relationship between LPR and dental erosion.^125,126 In an animal (rat) model study, Higo et al¹²⁷ identified microscopic dental erosion and loss of surface enamel secondary to regurgitation of acid, liquid, and gas, as well as destruction of teeth and supporting structures when they were exposed to gastric and duodenal contents. Ranjitkar et al¹²⁸ reported that casein phosphopeptide-amorphous calcium phosphate can protect teeth from erosion caused by acid and bile products.

Becker et al¹²⁹ studied the role of LPR in patients with complaints of intraoral burning sensations. They placed an oropharyngeal pH monitoring probe at the level of the uvula and found no causal relationship between LPR and intraoral burning. They suggested that PPI therapy is not indicated in this patient population.

Researchers and clinicians have examined a myriad of symptoms in patients with LPR and the relationships between LPR and voice disorders, benign and malignant lesions in the larynx, chronic cough, pulmonary disease, asthma and allergy, and obstructive sleep apnea in both adults and children. Chung et al¹³⁰ examined the relationship between LPR and benign vocal fold lesions. They reported LPR in 65% of their control group, 66% of the vocal nodules group, 75% of the vocal fold polyp group, and 95% of the patients with Reinke’s edema. Saleh¹³¹ suggested links among reflux, postnasal drip, and chronic cough. Randhawa et al¹³² pointed out that it is sometimes difficult to determine the cause of dysphonia, noting that the laryngeal findings on nasopharyngolaryngoscopy may be similar in patients with LPR and in those with allergy. In their small sample population, they diagnosed allergy in 10 patients and LPR in only 3 patients, which they felt raised the question of LPR being overdiagnosed.

Recent studies suggest a relationship between LPR and obstructive sleep apnea. Eskiismir and Kezirian¹³³ suggested that the increased respiratory efforts used by patients with obstructive sleep apnea generate increased intrathoracic pressure, which contributes to increased reflux. Suzuki et al¹³⁴ suggested that relaxation of the LES might be the mechanism of reflux in patients with mild to moderate obstructive sleep apnea. They also found reflux-induced spontaneous arousals. Karkos and colleagues¹³⁵ reported that during sleep, UES pressures decrease significantly. However, they noted a lack of controlled trials and/or meta-analyses that address the correlation between reflux, snoring, and/or apnea. In 2010, Wang et al¹³⁶ examined the concentration of pepsin detected in oropharyngeal secretions in patients with LPR and obstructive sleep apnea. In the LPR population, they found higher levels of pepsin in sputum that correlated with a higher reflux symptom index and higher reflux finding score. In obstructive sleep apnea patients, they reported no relationship between pepsin levels and reflux symptom index.

Eryuksel et al¹³⁷ examined the relationship between LPR and chronic obstructive pulmonary disease. Before and following 2 months of PPI treatment, patients underwent laryngeal examinations and pulmonary function tests and were asked to complete questionnaires for LPR and chronic obstructive pulmonary disease. The authors reported significant improvement in chronic obstructive pulmonary disease symptom index and LPR symptoms and findings on laryngeal examination. De la Hoz and colleagues¹³⁸ studied reflux and pulmonary disease in 9-11 World Trade Center first responders. Their findings suggest that patients with reflux demonstrated reduced forced vital capacity, suggestive of air trapping and lower airway disease.

Physical Examination

Physical examination of patients with throat and voice complaints must be comprehensive. A thorough head and neck examination is always included, with attention to the ears and hearing, nasal patency, signs of allergy, the oral cavity, temporomandibular joints, the larynx, and the neck. In some patients with LPR severe enough to involve the oral cavity, there is also loss of dental enamel. Hence, transparency of the lower portion of the central incisors may be seen occasionally in reflux patients, although it may be more common in patients with bulimia and those who habitually eat lemons. At least a limited general physical examination is included to look for signs of systemic disease that may present as throat or voice complaints. More comprehensive, specialized physical examinations by medical consultants should be sought when indicated.

When the patient has complaints of vocal difficulties, laryngeal examination is mandatory. It should be performed initially using a mirror or flexible fiberoptic laryngoscope; but comprehensive laryngeal examination requires strobovideolaryngoscopy for slow motion evaluation of the vibratory margin of the vocal folds. Formal assessment of the speaking and singing voice also should be performed, when appropriate. Objective voice analysis quantifies voice quality, pulmonary function, valvular efficiency of the vocal folds, and harmonic spectral characteristics. Neuromuscular function can be measured by laryngeal electromyography (EMG). These aspects of the physical examination and tests of voice function are discussed elsewhere in the book and will not be reviewed in this chapter.

Most commonly, laryngoscopy in patients with LPR reveals erythema and edema. Classically, reflux laryngitis involves erythema of the arytenoid cartilages and frequently interarytenoid pachydermia (a knobbled or cobblestone appearance), as well as other signs^{25,96,138–145} (Figures 11–15 and 11–16). However, many additional findings may be observed including edema of the false and true vocal folds; partial effacement or obliteration of the laryngeal ventricle; pseudosulcus (a longitudinal groove extending below the vibratory margin throughout the length of the vocal fold, including the cartilaginous portion); Reinke’s edema; granulomas or ulcers (most commonly in the region of the vocal process); nodules and other masses; an interarytenoid bar; laryngeal stenosis; and other abnormalities. Koufman reported that edema was seen even more common than erythema, having been diagnosed in 89% of 46 patients, compared with 87% who had erythema; 19% with granuloma or granulation tissue; and 2% with ulceration.¹⁴⁶

Belafsky et al¹⁴⁷ developed a reflux finding score (RFS) that rates signs and appears to correlate with the presence of LPR. They advocate use of this instrument in combination with the reflux symptom index (RSI).¹⁴⁸ The reflux finding score depends on observations of subglottic edema, ventricular obliteration, erythema/hyperemia, vocal fold edema, diffuse laryngeal edema, posterior laryngeal hypertrophy, granuloma/granulation tissue, and still, thick endolaryngeal mucus. Although additional research from other centers is needed to confirm the validity and reliability of the RFS (which remains somewhat controversial), the authors found excellent inter- and intraobserver reproducibility (although all observers were practicing at the same medical center); they found the RFS to be an accurate instrument for documenting treatment efficacy in patients with LPR, and it is used widely (including by the authors of this chapter).

In patients with severe LPR, the finding of a hyperactive gag reflex is also common; of interest, they also may have decreased laryngeal sensation. One of us (RTS) has performed functional endoscopic evaluation of sensory threshold testing on patients with LPR and found that responses were diminished prior to treatment and were improved following treatment. These findings are consistent with preliminary observations by Jonathan Aviv, MD (personal communication, 2000).

It should be noted that controversy exists regarding the significance of laryngeal findings. Credible studies of the sensitivity and specificity of laryngoscopy for diagnosis of LPR are needed, although a few initial reports exist in the literature. Carr et al¹⁴⁹ studied 155 children retrospectively. In a chart review of direct laryngoscopy and bronchoscopy findings, they reported a positive predictive value of 100% for the combination of posterior chronic edema with any vocal fold or ventricular abnormality.

McMurray et al¹⁵⁰ evaluated 39 children prospectively with laryngoscopy, bronchoscopy, esophagoscopy, and pH monitoring prior to airway reconstruction. Full-thickness laryngeal mucosal biopsy specimens were obtained from the posterior cricoid area and the interarytenoid area, and esophageal biopsy specimens were obtained. These investigators were unable to demonstrate a correlation among pH probe data, laryngoscopic findings, and histologic findings. Hicks et al¹⁵¹ studied 105 healthy, asymptomatic volunteers. On laryngoscopy, more than 80% had at least one “abnormal” finding, including (in order of frequency from most frequent to least) interarytenoid bar, medial arytenoid granularity, and true vocal fold erythema. This study did not perform 24-hour pH monitoring or any other tests to rule out the presence of “silent” reflux as a cause of the laryngoscopic abnormalities.

Despite many articles exploring signs and symptoms of reflux, including those cited above and other recent literature,^151–166 evidence confirming the diagnostic significance of various complaints and findings is scarce and contradictory. This is due to various problems, including the lack of a standard definition of “normal” in populations being studied. Continued interdisciplinary discourse and multicenter studies should be encouraged to answer important questions regarding the sensitivity and specificity of the many findings associated commonly with laryngopharyngeal reflux, as well as the impact of laryngopharyngeal reflux on quality of life and general health.¹⁶⁷

Pathophysiology

Laryngeal abnormalities may be caused by direct injury or by a secondary mechanism. Direct injury is due to contact of acid and pepsin with laryngeal mucosa, resulting in mucosal damage.^{3,95,96,168–171} Alternatively, irritation of the distal esophagus by acid may cause a reflex mediated by the vagus nerve, resulting in chronic cough and/or throat clearing, which may produce traumatic injury to the laryngeal mucosa.^{83,96,172–176}

Researchers are attempting to delineate pathophysiology. To show LPR as an extra extraesophageal manifestation of GERD, Groome et al¹⁷⁷ hypothesized that GERD patients would have some LPR symptoms if the pathophysiology were truly common. Through a questionnaire administered to 1383 GERD patients, they determined that the prevalence of LPR increases with the severity of GERD.¹⁷⁷ Although based on nonstandard questionnaires, the finding suggests a relationship. A similar study sought to exclude other causes of laryngitis and found LPR in 24% of patients with reflux esophagitis. The presence of LPR was predicted best by age, hoarseness, and hiatal hernia.¹⁷⁸ A third study, strengthened by confirmed diagnoses of GERD and LPR with esophagogastroduodenoscopy (EGD) and 24-hour pH monitoring, respectively, similarly showed that when both GERD and non-GERD patients were treated with PPIs, laryngitis symptoms and signs improved in the GERD group only.¹⁷⁹ These studies seem to point toward a common pathophysiology, suggesting that laryngeal symptoms are indeed caused by acid exposure.

Bile reflux also may cause laryngeal irritation.^180,181 In addition, recent findings raise many new questions about the pathophysiology of LPR. For example, Eckley et al^182,183 report that decreased salivary epidermal growth factor may be associated with LPR and warrants further study; Altman’s discovery of a proton pump in laryngeal serous cells and ducts of submucosal glands is particularly intriguing.¹⁸⁴ It has been established that pepsin in the larynx results in depletion of carbonic anhydrase isoenzyme III (CAI III) and squamous epithelial stress protein (Sep70), 2 laryngeal protective proteins.^185,186 Pepsin is taken up by laryngeal cells and can be reactivated by a drop in pH, as seen in LPR. Pepsin is found in the esophageal mucosa of those with LPR.¹⁸⁷ Interestingly, pepsin irreversibly affects CAI III at pH below 4 only in laryngeal, not esophageal, epithelium.¹⁸⁸ These laryngeal receptors for pepsin may be another future target for intervention. They also might explain the presence of symptoms and signs of LPR with weakly acidic reflux, as pepsin may be active to some degree at any pH between 3 and 6.5,¹⁸⁹ although a longer exposure time may be necessary at pH 5 or above to produce damage.¹⁹⁰ Possible pepsin activity at pH much above 5 remains controversial. Mucin gene (muc) expression also is downregulated in the presence of pepsin.^190,191

Animal studies have shown that acid combined with pepsin (acid-activated enzyme) compromises the integrity of the vocal fold epithelium. In 2010, Habesoglu et al¹⁹² reported that rats exposed to an acidic pH in the presence of pepsin developed edema of the lamina propria, submucous gland hyperplasia, and muscular atrophy. Erickson and Sivansakar¹⁹³ reported that the epithelial barrier resistance of the vocal fold is compromised when exposed to acid and pepsin. They described transepithelial resistance as a marker of epithelial barrier resistance that measures the ability to restrict movement of solute and solvents. The authors pointed out that less than 3 episodes of reflux per week could injure vocal fold mucosa, in contrast to 0 to 50 reflux events daily that are considered normal exposure in the distal esophagus.

In 2009, Johnston et al¹⁹⁴ reported that pepsin invades the laryngeal epithelial cells by receptor-mediated endocytosis and that pepsin activity is maximal at a pH of 2. Bulmer et al¹⁹⁵ in 2010 showed that the effects of acid and pepsin exposure on porcine laryngeal mucosa were similar to the effects observed in the human larynx in patients diagnosed with LPR. Samuels and Johnston¹⁹⁶ reported that the presence of pepsin in the airway indicates reflux. However, they reported that there are only 2 methods used to identify pepsin in the airways: immunologic and enzymatic, both of which have advantages and disadvantages. Richter¹⁹⁷ stated that gastric acid combined with pepsin and bile salts is known to be causally related to the development of chronic esophagitis and Barrett’s esophagus. In contrast, they reported that weak or non-acid refluxate has not been shown to cause damage to the esophageal or extraesophageal structures, including the larynx and the lungs.

Histopathologic inflammation and its association with the pathogenesis of esophageal and extraesophageal disease has been the subject of ongoing research. Some studies that have focused on histology have produced interesting findings. One found significantly increased CD8⁺ lymphocytes in the epithelium of patients with LPR, with proportionally more in the luminal epithelial layer. Additionally, nonclassical major histocompatibility complex (MHC) molecule expression was found to be involved in the response to refluxate, suggesting that relying upon classical markers may lead to erroneous conclusions about the response of laryngeal epithelium to reflux.¹⁹⁸ Wada et al¹⁹⁹ evaluated histopathologic inflammation of the upper esophagus in comparison with the lower esophagus. When compared with their control group, they found inflammation of the upper esophagus to be significantly greater in patients with abnormal laryngopharyngeal symptoms and significantly greater lower esophageal inflammatory histopathology in patients with classic reflux symptoms (ie, GERD).

Park et al²⁰⁰ used transmission electron microscopy to examine cellular damage of the esophageal epithelium by gastric refluxate. They measured the intracellular space and quantified its dilatation. They compared 2 groups: patients who had LPR without GERD and patients who had LPR with GERD. They reported that the intracellular space of the esophageal epithelium was significantly more dilated in the LPR-with-GERD group. Amin and colleagues²⁰¹ reported that dilatation of intracellular spaces of esophageal epithelium is considered a specific marker in GERD. They studied a group of patients with LPR and sore throat in whom they found dilatation of intracellular spaces in oropharyngeal biopsy specimens. They also found dilatation of the intracellular spaces in the laryngeal mucosa in animal models exposed to pepsin.

There are important pathophysiologic differences between LPR patients and GERD patients. For example, combined upright and recumbent or nighttime reflux is typical for GI patients with GERD. Upright reflux and regurgitation also are the least common pattern in this population. However, patients with LPR are more likely to experience upright reflux commonly throughout the day,^16,28,32,202 often even in the absence of supine reflux.

We have observed some patients with LPR who experience reflux exclusively (and constantly) when they sing. Motility abnormalities have been demonstrated with higher frequency in patients with LPR, resulting in delayed acid clearance in one study.²⁰³ In contrast, Postma et al²⁰⁴ demonstrated that patients with GERD have significantly longer esophageal acid clearance times than those measured in patients with LPR. However, it is not unusual for LPR patients to have abnormal upper esophageal sphincter (UES) function. In 1978, Gerhardt et al²⁰⁵ showed that experimental instillation of acid in the distal esophagus in patients with esophagitis and in normal controls produces increase in UES tone. This phenomenon does not occur normally in many patients with LPR,¹⁰³ although an increase in resting UES pressure has been demonstrated in patients with reflux laryngitis.²⁰⁵ In 2010, based on their research, Szczesniak and colleagues²⁰⁶ suggested upregulation of the esophago-UES relaxation as a possible mechanism in the pathophysiology of reflux laryngitis. They found that the UES relaxation reflex induced by rapid air insufflation of the esophagus is upregulated in patients with posterior laryngitis compared with healthy controls. They also found that this group of patients had a higher pharyngo-UES contractile reflex threshold, which is considered a mechanism of airway protection.

Vardouniotis et al²⁰⁷ reported that a hypotensive lower esophageal sphincter (LES) is a significant pathophysiologic component of LPR, noting that complex molecular mechanisms involved in the function of the LES and genetic factors are involved in tissue protection from reflux. They reported that compared with the esophageal lining, laryngeal and pharyngeal mucosa is more susceptible to tissue damage from refluxate because the larynx is not protected by peristalsis or buffered by salivary bicarbonate. They reported that transforming growth factor-beta 1, which inhibits inflammatory response, has shown gene overexpression in postcricoid fibroblasts and that fibroblast growth factor 2 has shown decreased expression. Cheng et al²⁰⁸ suggested that the microphage activation caused by gastric acid exposure should be considered also in the pathogenesis of GERD and aspiration-induced lung disease.

Chong and Jalihal²⁰⁹ reported a greater percentage of LPR symptoms in patients with heterotrophic gastric mucosal patch (HGMP) of the distal esophagus. HGMP is ectopic gastric mucosa typically observed distal to the UES and thought to be congenital. Due to the fact that HGMP can produce acid, the authors suggested that this may be the etiology of LPR in some patients. In fact, in their study, patients with HGMP experienced LPR symptoms to a greater degree (73.1%) compared with their non-HGMP patients (25.9%).

Salminen et al and Basseri et al^210,211 have reported that the endoscopic prevalence of HGMP is low, ranging from 0.1 to 10% in reported patients 16 to 75 years old. Typical symptoms in patients with HGMP include dysphagia, globus pharyngeus, cough, hoarseness, and shortness of breath. They reported that the HGMP is a source of acid production, and this has been supported by pH monitor studies showing gastric acid in this area. Parietal cells with oxyntic mucosa are the most common histologic type reported.

Laryngopharyngeal reflux can affect anyone, but it appears to be particularly common and symptomatic in professional voice users, especially singers. This is true for several reasons. First, the technique of singing involves “support” by the forceful compression of the abdominal muscles designed to push the abdominal contents superiorly and pull the sternum down. This action compresses the air in the thorax, thereby generating force for the stream of expired air, but it also compresses the stomach and works against the LES. Singing is an athletic endeavor, and the mechanism responsible for reflux in singers is similar to that associated with reflux following other athletic activities, lifting, and other conditions that alter abdominal pressure, such as pregnancy (which is also influenced by hormonal factors). It has been established clearly by Clark et al²¹² that reflux is induced by exercise even in asymptomatic, young volunteers (mean age, 28 years). They demonstrated that running induced reflux more often than did exercise with less bodily agitation, such as bicycling, but both forms of aerobic exercise caused reflux, as did weight lifting in some patients. Postprandial exercise-induced reflux has a similar pattern, but with a greater amount of refluxate. It may be that the effect of exercise on reflux is even more pronounced in patients with GERD or LPR than it is in research subjects with no history of reflux symptoms, although this question has not been studied.

Second, many singers do not eat before performing because a full stomach interferes with abdominal support and promotes reflux. Performances usually take place at night. Consequently, the singer returns home hungry and eats a large meal before going to bed.

Third, performance careers are particularly stressful. Psychological stress has been associated with esophageal motility disorders (which may be associated with reflux) and with other gastroenterologic conditions, such as irritable bowel syndrome.²¹³ Psychological stress alone acts to increase the amplitude of esophageal contractions.²¹⁴ Stress also may affect the production of gastric acid. If psychological stress increases LPR, it may create a vicious cycle. Pharyngeal stimulation may cause transient LES relaxation directly, or it may lower the threshold for triggering gastric distention.²¹⁵

Fourth, many singers pay little attention to good nutrition, frequently consuming caffeine, fatty foods (including fast foods), spicy foods, citrus products (especially lemons), and tomatoes (including pizza and spaghetti sauce). In addition, because of the great demands that singers place on their voices, even slight alterations caused by peptic mucositis of the larynx produce symptoms that may impair performance. Thus, singers are more likely to seek medical care because of reflux symptoms than are individuals with fewer vocal demands. However, careful inquiry and physical examination reveal similar problems among many nonsinger patients. Most of the voice problems associated with reflux laryngitis appear to be due to direct mucosal damage from proximal reflux. The effects of distal reflux alone on laryngeal function have not been studied.

Voice abnormalities and vocal fold pathology may be due to reflux of gastric acid onto the vocal folds. Severe coughing may cause vocal fold hemorrhage or mucosal tears, sometimes leading to permanent dysphonia by causing scar that obliterates the layers of the lamina propria and adheres the epithelium to deeper layers. Aspiration caused by reflux also makes reactive airway disease difficult to control. Even mild pulmonary obstruction impairs voice support. Consequently, afflicted patients subconsciously strain to compensate with muscles in the neck and throat, which are designed for delicate control, not for power source functions.^216,217 This behavior is typically responsible for the development of vocal fold nodules and other lesions related to voice abuse. Although it appears likely that some extraesophageal symptoms of reflux are due to stimulation of the vagus nerve rather than (or in addition) to topical irritation, the role of vagal reflexes in reflux laryngitis remains to be clarified.

Posterior Laryngitis and Related Conditions

In addition to erythema and edema, more serious vocal fold pathology may be caused by reflux laryngitis. In 1968, Cherry and Margulies⁹⁶ recognized that reflux laryngitis might be a causative factor in contact ulcers and granulomas of the posterior portion of the vocal folds, conditions that are discussed in detail below. They also observed that treatment of peptic esophagitis resulted in resolution of vocal process granulomas. Delahunty and Cherry⁹⁷ followed up on this observation by applying gastric juice to the vocal processes of 2 dogs and applying saliva to the vocal processes of a third dog who was used as a control. The control dog’s vocal folds remained normal; the other dogs developed granulomas at the sites of repeated acid application. The experiment by Delahunty and Cherry is particularly interesting. The posterior portion of the left vocal fold of 2 dogs was exposed to gastric acid for a total of only 20 minutes per day, 5 days out of every 7, for a total of 29 days of exposure in a 39-day period. A total of 20 minutes out of 24 hours may not seem like an extensive exposure period; however, erythema and edema were apparent in both dogs by the fourth day of the first week. At the beginning of the second week, the larynges appeared normal after the 2-day rest period. However, visible reaction was provoked within 2 days after application was resumed, and the vocal folds never regained normal appearance. Marked inflammation, thickening, and irregularities were apparent in both dogs by the fourth week, and epithelial slough at the site of acid contact occurred on day 29 in one dog and day 32 in the other. Granulation tissue appeared shortly thereafter. A similar procedure was performed on a control animal by applying saliva instead of gastric juice to the vocal fold, and the vocal fold remained normal. This research suggests that even relatively short periods of acid exposure may cause substantial abnormalities in laryngeal mucosa. Since then, numerous authors have recognized the importance of reflux laryngitis as a causative factor in laryngeal ulcers and granulomas, including intubation granuloma.^{12,13,77,83,96,218–221} In addition to its etiological involvement in intubation granuloma, reflux laryngitis has long been recognized as a contributing factor to posterior glottic stenosis, especially following intubation.²²² Olson has suggested that it may also be a causative factor in cricoarytenoid joint arthritis through chronic inflammation and ulceration, beginning on the mucosa and involving the synovial cricoarytenoid joint.⁷ We have encountered this problem, as well. In addition to posterior glottic and supraglottic stenosis, subglottic stenosis has been reported as a complication of reflux.^104,223

Laryngeal Granulomas

Laryngeal granulomas are a particularly vexing problem for patients and their physicians. Granulomas, like contact ulcers of the larynx, usually occur on the posterior aspect of the vocal folds, often on or above the cartilaginous portion. They may be unilateral, although it is also common to see a sizable granuloma on one side and a contact ulcer on the other. Patients with ulcers or granulomas may complain of pain (laryngeal or referred otalgia), a globus sensation, hoarseness, painful phonation, and occasionally hemoptysis. Surprisingly, even large granulomas are often asymptomatic. These benign lesions usually contain fibroblasts, collagenous fibers, proliferated capillaries, leukocysts, and sometimes ulceration. Although the term “granuloma” is universally accepted, these laryngeal lesions actually are not granulomas histopathologically, but rather chronic inflammatory lesions. However, granulomas and ulcers may mimic more serious lesions such as carcinoma, tuberculosis, and granular cell tumor. Consequently, the clinical diagnosis of laryngeal granuloma must always be made with caution and must be considered tentative until the patient has been followed over time and a good response to treatment has been observed.

Understanding the etiology of laryngeal ulcers and granulomas is essential to clinical evaluation and treatment. Traditionally, ulcers and granulomas in the region of the vocal processes have been associated with trauma, especially intubation injury. However, they are also seen in young, apparently healthy professional voice users with no history of intubation or obvious laryngeal injury. In fact, the vast majority of granulomas and ulcerations (probably even those from intubation) are caused or aggravated by laryngopharyngeal reflux disease. In some patients, muscular tension dysphonia producing forceful vocal process contact may be contributory or causal.

Evaluation of patients with laryngeal ulcers or granulomas begins with a comprehensive history and physical examination. In addition to elucidating specific voice complaints and their importance to the individual patient’s life and profession, the history is designed to reveal otolaryngologic and systemic abnormalities that may have caused dysphonia. Special attention is paid to symptoms of voice abuse and of laryngopharyngeal reflux, as listed above. It must be remembered that reflux laryngitis is commonly not accompanied by pyrosis or dyspepsia in these patients. The history also seeks specifically symptoms consistent with asthma, including voice fatigue following extensive voice use. Exercise-induced asthma can be provoked by the exercise of voice use, and even mild reactive airway disease undermines the power source of the voice and may lead to compensatory muscular tension dysphonia and consequent laryngeal granuloma or ulcer. Inquiry also investigates systematically all body systems for evidence of other diseases that present with laryngeal pathology. It is important to include a psychological assessment. Excessive stress may lead to increased acid production, abnormal esophageal function, and symptomatic reflux and to muscular tension dysphonia. In such cases, it is important to identify and treat the underlying stressor, as well as the symptomatic expressions of the stress.

Mirror examination usually reveals the presence of a granuloma or ulcer, but more sophisticated evaluation is invaluable. In the presence of suspected laryngeal granuloma or ulcer, the author (RTS) routinely performs strobovideolaryngoscopy using both flexible and rigid endoscopes. Flexible endoscopic examination reveals patterns of phonation and is extremely helpful in identifying muscular tension dysphonia and determining phonatory behaviors associated with forceful adduction. Recent observations (Steven Zeitels, MD, personal communication, 1997, and the author’s [RTS] experience) suggest that some granuloma patients have a vocal fold closure pattern with initial forceful vocal process contact. This implies an adduction strategy with lateral cricoarytenoid dominance, and this observation is important in the treatment of granulomas that are refractory to therapy (medical and/or surgical) or granulomas that recur. Rigid laryngeal stroboscopic examination provides magnified, detailed information of the lesions under slow-motion light, allowing analysis of their composition (solid granulomas versus fluid-filled cysts) and their effects on phonation. This examination also permits assessment of other areas of the vocal folds to rule out separate lesions (eg, vocal fold scar) that may be the real cause of the patient’s voice complaint.

Evaluation also includes at least a formal assessment by a speech-language pathologist (SLP) skilled in voice evaluation and care. In the author’s (RTS) center, we also include objective voice analysis and a vocal stress assessment with a singing voice specialist (even with nonsingers). In addition to a laryngologist, SLP, and singing voice specialist, other members of the voice team are often used, depending on the patient’s problems. Additional team members include an acting-voice specialist, psychologist, psychiatrist, otolaryngologic nurse clinician, and consulting pulmonologist, neurologist, gastroenterologist, and others. The information provided by these evaluations helps establish the degree to which voice abuse/misuse is present, and it guides the design of an individualized therapy plan.

Reflux must be suspected in virtually all cases of granuloma. It can be evaluated by 24-hour pH monitor, barium swallow with water siphonage (routine barium swallows are not satisfactory for diagnosing reflux, and the accuracy of barium swallow even with water siphonage is debatable), other tests, and/or a therapeutic trial of medical management. If there is historical evidence of prolonged reflux symptoms, endoscopic evaluation to rule out Barrett’s esophagus is often advisable. It may be appropriate to biopsy the presumed granuloma at the same time. If a therapeutic trial of medications without confirmatory tests is elected, marked improvement in symptoms and signs should occur following daily use of a PPI (before breakfast and dinner) within 2 to 3 months. Treatment for laryngopharyngeal reflux should be aggressive.

The efficacy of oral corticosteroids for treatment of laryngeal granulomas and ulcers has not been proven, but they are used commonly on the basis of anecdotal evidence, especially for small or medium-sized granulomas and ulcers that appear acutely inflamed. For these conditions, low doses of steroids for longer periods are usually given, such as triamcinolone 4 mg twice a day for 3 weeks. Steroid inhalers are not recommended. They may lead to laryngitis or laryngeal Candida infections, and prolonged use may cause vocal fold atrophy.

At the end of 2 months of therapy including antireflux measures, voice therapy, and possibly steroids, substantial improvement in the appearance of the larynx should be seen. Ulcers should be healed, and granulomas should be substantially smaller. Patients should be examined after 1 month to be certain that the lesions are not enlarging. If they appear worse, biopsy should be performed promptly. However, it should be noted that complete healing may take 8 months or more.^224,225 Repeated strobovideolaryngoscopic examinations allow comparison of lesion size over time. If improvements are noted, aggressive therapy and close follow-up can be continued until the mass lesion disappears or stabilizes. If the mass does not disappear, or if response to the first 2 months of aggressive therapy produces no substantial improvement, biopsy should be performed to rule out carcinoma and other diseases. If the surgeon is reasonably certain that the lesion is a granuloma, injection of an aqueous steroid preparation (such as dexamethasone) into the base of the lesion at the time of surgery may be helpful. As long as a good specimen is obtained, the laser may be used for resection of suspected granulomas because the lesions are usually not on the vibratory margin, and they often are friable. However, the author (RTS) usually uses traditional instruments to avoid the third-degree burn caused by the laser (even with a microspot) in the treatment of this chronic, irritative condition.

It is essential that causative factors, especially reflux and voice abuse, be treated preoperatively and controlled strictly following laryngeal surgery. The patient is kept on therapeutic doses of a PPI prior to surgery and for at least 6 weeks following surgery. Surgeons should not hesitate to use omeprazole (Prilosec) 20 mg as frequently as 4 times a day or the equivalent dose of another PPI under these circumstances and, if necessary, to add an H₂ blocker (ranitidine, 300 mg) at bedtime. Following surgery, absolute voice rest (writing pad) is prescribed until the surgical area has remucosalized. This is usually approximately 1 week and virtually never longer than 10 to 14 days. There are no indications for more prolonged absolute voice rest, although relative voice rest (limited voice use) is recommended routinely. Voice therapy is reinstituted on the day when phonation is resumed, and frequent short therapy sessions and close monitoring are maintained throughout the healing period.

As previously stated, granulomas recur in some patients. In all such cases, aggressive reevaluation of reflux with 24-hour pH monitor studies is warranted because granulomas are seen commonly in patients with reflux laryngitis. Often endoscopy and biopsy of esophageal and postcricoid mucosa is appropriate. Twenty-four-hour pH monitor studies should be conducted not only off all medications, but also when the patient is taking a PPI or H₂ blocker. A few patients are resistant to PPIs and will have normal acid secretions despite even 80 mg of omeprazole daily or the equivalent dose of another PPI; and some patients may respond to PPIs initially and then develop resistance. In such patients, H₂ blockers may be effective. When medical management of reflux is insufficient, laparoscopic fundoplication can be considered for patients with recurring granuloma. Voice use must also be optimized and monitored with the help of the SLP, always, and other members of the voice team when indicated. The laryngologist and voice therapists must be sure that good vocal technique is carried over outside the medical office into the patient’s daily life.

Occasionally, even after excellent reflux control (including fundoplication), surgical removal including steroid injection into the base of the granulomas, and voice therapy, patients may develop multiply recurrent granulomas. Medical causes other than reflux and muscular tension dysphonia must be ruled out, particularly granulomatous diseases including sarcoidosis and tuberculosis, and neoplasm such as granular cell tumors. Pathology slides from previous surgical procedures should be reviewed. When it has been established that the recurrent lesions are typical laryngeal granulomas occurring in the absence of laryngopharyngeal reflux, the cause is almost always phonatory trauma. When voice therapy has been insufficient to permit adequate healing, some of these uncommonly difficult patient problems can be solved by temporary paresis of selected vocal fold adductor muscles (particularly the lateral cricoarytenoid) using botulinum toxin (Botox, Allergan, Irvine, California) injection. Although this treatment approach has been effective, it is not utilized ordinarily as initial therapy and is appropriate only for selected cases.

Delayed Wound Healing

In addition to its possible carcinogenic potential, the chronic irritation of reflux laryngitis may be responsible for failure of wound healing. Reflux appears to delay the resolution not only of vocal process ulcers and granulomas, but also of healing following vocal fold surgery. For this reason, otolaryngologists are becoming increasingly aggressive about diagnosing and treating reflux before subjecting patients to vocal fold surgery, even for conditions unrelated to the reflux.

Stenosis

As noted above, laryngeal stenosis has been associated with reflux.^{32,84,104,105} Koufman has reported laryngopharyngeal reflux in 92% of his patients with laryngeal stenosis, all of whom were documented by 24-hour pH monitoring studies.³² This is consistent with an earlier report by Little, Koufman, et al in which the authors were able to produce nonhealing ulcerations and subglottic stenosis experimentally in canines by applying gastric acid and pepsin to injured subglottic mucosa.³ Long-term control of laryngopharyngeal reflux is essential to success in treating laryngeal stenosis.

Globus Pharyngeus

The sensation of a lump in the throat or globus pharyngeus is associated commonly with laryngopharyngeal reflux. The literature on the association of globus with reflux does not provide definitive guidance.^{15,16,89,90,225–229} However, reflux has been found in 23 to 90% of patients with globus.^{32,90,225–228,230,231}

Smit et al studied 27 patients with globus pharyngeus alone, 20 patients with hoarseness alone (more than 3 months duration), and 25 patients with both globus and hoarseness.²³¹ Using dual-probe pH monitoring, pathologic reflux was diagnosed if patients had a pH below 4 for more than 0.1% of the total time, more than 0.2% of time in the upright position, and more than 0% of the time in the supine position in the proximal probe, or if they had more than 3 reflux episodes with pH below 4. The proximal probe was placed visually at the UES, and the distal probe was 15 cm from the proximal probe. Only 30% of patients with globus but without hoarseness had pathologic reflux. Similar findings were reported by Wilson et al (23%),²²⁷ Curran et al (38%),²²⁸ and Hill et al (30.8%).²³⁰ Smit et al found that only 35% of patients with hoarseness alone had pathologic reflux. However, 72% (18 of 25) of patients with globus and hoarseness had pathologic reflux. Sixty-five percent of patients with pathologic GERD had abnormal findings during esophagoscopy, including 2 patients with Barrett’s mucosa. The diagnosis of laryngopharyngeal reflux should be considered in patients with globus pharyngeus, and diagnostic evaluation and therapeutic trial with PPIs are warranted.^231–233

Laryngospasm

Laryngospasm is forceful, involuntary adduction of the vocal folds. It is associated with airway obstruction that is often severe enough to cause the patient to panic. Typically laryngospasm occurs suddenly and without warning. It may be precipitated by laughing or exercise, or it may occur with no apparent precipitating event. Nighttime attacks that awaken the patient are common. Reflux is a well-recognized cause of laryngospasm. The mechanism may be related to chemoreceptors on the epiglottis that respond to a pH of 2.5 or below by eliciting laryngospasm.²³³ Loughlin et al also demonstrated that this reflex is dependent on a functioning superior laryngeal nerve.²³⁴ In our experience, laryngopharyngeal reflux is the cause of paroxysmal laryngospasm in nearly all patients with this condition, and most respond to aggressive antireflux therapy.

Muscle Tension Dysphonia

The relationship between laryngopharyngeal reflux and muscle tension dysphonia remains uncertain, but there is reason to consider an association possible. Koufman and coworkers found a 70% incidence of laryngopharyngeal reflux in patients with structural vocal fold lesions associated commonly with muscle tension dysphonia, including nodules, Reinke’s edema, hematoma, ulcers, and granuloma.³² Chronic reflux laryngitis causes irritation that leads not only to an inflammatory response, but also to laryngeal hyperirritability. Laryngospasm is the extreme manifestation of this condition. However, hyperfunctional posturing of the laryngeal muscles in response to chronic irritation, or as a defense against unpredictably timed episodes of laryngeal aspiration of acid, conceivably could lead to hyperfunctional patterns of voice use. Alternatively, in some patients, laryngopharyngeal reflux and muscle tension dysphonia may occur coincidentally, but injury to the vocal fold mucosa by acid and pepsin may make the vocal folds more prone to injury and to the development of structural lesions associated with phonotrauma. Traditionally, otolaryngologists and speech-language pathologists have viewed muscle tension dysphonia as a primary condition in the majority of cases. In the author’s (RTS) opinion, a high percentage of patients with muscle tension dysphonia have an underlying disorder such as reflux laryngitis or superior laryngeal nerve paresis that may have been responsible for the patient’s hyperfunctional voice disorder. In all patients with voice abnormalities, including muscle tension dysphonia, it is essential to seek out and treat the primary etiological condition.

Paroxysmal vocal fold movement disorder is a laryngeal dystonia characterized by intermittent glottic obstruction by adduction of the vocal folds on inspiration. It is also called paradoxical vocal fold adduction, respiratory dysphonia, and other names, including paradoxical vocal fold movement disorder (PVFMD). Yelken et al²³⁵ point out that PVFMD “mimics asthma,” and patients are often diagnosed incorrectly initially. In their patients, they reported no relationship between asthma attacks and severity and PVFMD. However, they found LPR and allergy to be prevalent in their patients with PVFMD.

Cough associated with PVFMD was studied by Murry and colleagues.²³⁶ They suggested that chronic cough associated with PVFMD might be due to laryngeal sensory deficits secondary to chronic acid exposure in the laryngopharynx, which triggers the cough reflex. They suggested also that the cough reflex could be an adaptive mechanism to clear particulate matter from the laryngopharynx.

Reinke’s Edema

Prolonged acid/pepsin irritation of the laryngeal mucosa can result in significant alterations in laryngeal tissues including carcinoma, as discussed below. Reinke’s edema appears to be one such tissue alteration. Koufman demonstrated abnormal 24-hour pH monitoring results in a majority of their patients with Reinke’s edema,³² and this is consistent with our experience. In the author’s (RTS) opinion, in many cases, it is unclear whether reflux is the primary cause of Reinke’s edema or is a cofactor with other laryngeal mucosal irritants such as smoking, hyperfunctional voice use, or hypothyroidism. However, we evaluate all patients with Reinke’s edema for reflux and treat LPR aggressively. Many patients seeking optimal restoration of voice quality require surgical treatment despite good reflux control, voice therapy, smoking cessation, and correction of any thyroid abnormalities. Good reflux control should be maintained long term, but it is especially critical in the immediate postoperative period, as discussed previously in the section on delayed wound healing.

Carcinoma

The association of gastroesophageal reflux disease with Barrett’s esophagus and esophageal carcinoma has been well established. It is now thought possible that LPR is associated with laryngeal malignancy, as well.^237–241 Delahunty biopsied the posterior laryngeal mucosa in a patient with reflux laryngitis and reported epithelial hyperplasia with parakeratosis and papillary down-growth.⁷⁸ In the 1980s, Olson and others reported on patients (including young, nonsmokers, nondrinkers) with posterior laryngeal carcinoma in whom he believed reflux to be a cofactor.¹ This issue was addressed also by Morrison.²⁴² He reported 6 cases of vocal fold carcinoma in patients who had severe reflux but had never smoked. In 1997, Olson reaffirmed that the relationship between reflux and cancer is not conclusive.²⁴³

The mechanisms by which reflux may cause laryngeal cancer remains speculative. Both smoking and alcohol consumption promote reflux by lowering lower esophageal sphincter pressure, impairing esophageal motility and mucosal integrity, increasing gastric acid secretion, and delaying gastric emptying. So, a high incidence of reflux in laryngeal cancer patients who smoke and drink is not surprising. However, the association does not explain how LPR may act as a cofactor in these patients or as a primary factory in patients who do not smoke and drink. Richtsmeier et al²⁴⁴ have suggested that a deficiency in T-cell mediated immunity is causally related to immunodeficiency in cancer patients. There is a subgroup of suppressor T cells with histamine-type 2 receptors. Cimetadine, a histamine-type 2 receptor antagonist, inhibits the expression of suppressor T cells and enhances immune responses. Richtsmeier and Eisele found that skin test anergy in laryngeal cancer can be reversed by cimetadine.²⁴⁵ This led Richtsmeier et al to recommend the use of an H₂ blocker not only to treat reflux in laryngeal cancer patients, but also to address their underlying immune dysfunction,²⁴⁶ although this thinking is not accepted widely.

Although some questions remain regarding the relationship between LPR and laryngeal carcinoma, the studies cited above, as well as more recent evidence,²⁴⁶ suggest that the 2 conditions are probably associated. At present, patients with laryngeal cancer, or those at risk to develop laryngeal cancer, should be screened for reflux; antireflux therapy should be instituted when it is present. Cancer surveillance is reasonable even in patients without known risk factors other than chronic LPR. The long-term efficacy of such treatment with regard to prevention of malignancy remains unknown, but we have seen resolution of laryngeal structural abnormalities, including suspicious leukoplakia, in patients with LPR alone, and even in patients who continue smoking and consuming alcohol. Koufman has had similar experiences.²⁴⁷

In 1988, Ward and Hanson recognized reflux as a potential cofactor for the development of laryngeal cancer, particularly in nonsmokers.²⁴⁸ In 1991, Koufman documented LPR in 84% of 31 consecutive patients with laryngeal carcinoma, only 58% of whom were active smokers.³² Frieje et al reviewed retrospectively 23 patients with T1 and T2 carcinomas of the larynx, and they concluded that GERD plays a role in the etiology of carcinoma of the larynx particularly in patients who lack typical risk factors (14 of their patients had quit smoking more than 15 years prior to developing laryngeal carcinoma) and may act as a co-carcinogen in smokers and drinkers.²⁴⁹ In 1997, Koufman and Burke felt that the causal relationship between LPR and laryngeal malignancy remained unproven, but noted that most patients who develop laryngeal malignancy have LPR in addition to being smokers.²⁴⁷ Until more definitive data are available, we believe that long-term antireflux therapy in these patients should be considered.

Research has shown that GERD is a factor in the development of esophageal cancer.²⁴⁹ Studies also have examined the potential carcinogenic role of LPR. Conflicting data exist, complicated recently by published stories that show no carcinogenic effects but that were limited by short periods of acid injury.^250,251 Although some studies have shown LPR as a risk factor in animal models, others have not, and the true relationship between LPR and laryngeal malignancy remains uncertain. However, there are enough data indicating a possible link to suggest that at present, known reflux patients should be screened for laryngeal cancer and be made aware of this risk.²⁵²

Given the risk of esophageal adenocarcinoma in reflux patients, the value of EGD screening has been assessed. It has been argued that LPR symptoms are better indicators of esophageal adenocarcinoma than are gastroesophageal symptoms.²⁵³ Studies have shown esophagitis in 12 to 18% of LPR patients, and of 3 to 7% Barrett’s metaplasia patients.^254,255 Additionally, in patients on long-term PPI’s, Helicobacter pylori is known to accelerate the loss of specialized gastric glands, causing atrophic gastritis and gastric cancer.²⁵⁶ H pylori, breath, or fecal testing or biopsy may screen for this infection. Diagnosis of hiatal hernia would also potentially affect treatment and is facilitated by EGD. Some have suggested routine EGD for patients complaining of heartburn; others suggest it for all LPR patients.²⁵⁴

Until the true relationship is defined, screening with either transnasal esophagoscopy or EGD (which may be performed at the time of routine colonoscopy in appropriate patients) is advisable. Many gastroenterologists are now advising visualization of the larynx during EGD. This has led to the diagnosis of LPR in up to 4.6% of patients undergoing EGD for GERD.²⁵⁷

Current literature reflects ongoing research on the causal effects of gastric refluxate (pepsin) and the carcinogenic effects on the larynx, pharynx, and upper airway epithelium.^258–262

Laryngopharyngeal (LPR) has been considered a risk factor in the development of squamous cell carcinoma of the larynx, but it remains unproven.

Sudden Infant Death Syndrome and Other Pediatric Considerations

Laryngopharyngeal reflux is important in the pediatric population, although it has been studied much less extensively than has reflux in adults. Unlike adults, infants and young children are unable to complain of symptoms associated with LPR. Nevertheless, LPR has been associated with various problems in infants and children including halitosis, dysphonia, laryngospasm, laryngomalacia, asthma, pneumonia, sleep apnea, and sudden infant death syndrome (SIDS).^{6,24,263–285} The diagnosis can be established by laryngoscopy and bronchoscopy, and 24-hour pH monitor studies. Children can be treated with H₂ blockers and/or PPIs, and fundoplication is appropriate in selected cases, particularly in patients with life-threatening complications of reflux.

Evidence suggests that SIDS may be causally related to acid reflux into the larynx. Hence, SIDS must join laryngeal and esophageal cancer at the top of the list of serious otolaryngologic consequences of reflux laryngitis. Wetmore investigated the effects of acid on the larynges of maturing rabbits by applying solutions of acid or saline at 15-day intervals up to 60 days of age.²⁶⁴ Because the larynx is not only a site of resistance in the airway, but also contains the afferent limb for reflexes that regulate respiration, he discovered that acid exposure resulted in significant obstructive, central, and mixed apnea. Gasping respirations and frequent swallowing were observed as associated symptoms. Central apnea occurred in all age groups but had a peak incidence at 45 days. Acid-induced obstructive apnea in rabbits is similar to obstructive apnea previously recognized in human infants with gastroesophageal disease. However, the demonstration of acid-induced central apnea produced by acid stimulation of the larynx is more ominous. Central apnea has been demonstrated in other animal models as a result of different forms of laryngeal stimulation. Central apnea resulting in fatal asphyxia has also been described in several animal models. Wetmore’s study²⁶⁴ suggests that gastroesophageal reflux alone is capable of triggering fatal central apnea. This is particularly compelling when one recognizes that the peak incidence of central apnea occurring at 45 days in the rabbit corresponds well within the peak incidence of SIDS in humans, which occurs between 2 and 4 months of age.

Otolaryngologists should be aware of the prevalence of LPR/GERD in children and infants. Studies have shown that reflux is associated commonly with regurgitation and vomiting, disturbed sleep patterns, colic, gastrointestinal pain, croup, and hoarseness in the pediatric population.^265–288

Monitoring of pH is valuable to infants with laryngitis, including pharyngeal monitoring, which may diagnose undertreated LPR or LPR missed on esophageal monitoring.²⁸⁹ LPR is a known common cause of hoarseness in children and should be in the differential diagnosis of dysphonia.²⁹⁰ It may be misdiagnosed as recurrent croup when reflux triggers intermittent airway obstruction.²⁹¹ The diagnosis of LPR is still often missed in children with hoarseness or frequent respiratory disorders.

Barreto et al²⁹² examined the laryngeal and phonatory effects associated with untreated growth hormone deficiencies. Values for roughness, breathiness, and strain were higher, and LPR signs were also more common in this population. More research is needed. Otolaryngologists should be familiar with this association.

Ongoing research continues to examine the relationship between LPR and otitis media in children. Several studies have confirmed the presence of pepsin and pepsinogen in children with otitis media with effusion.^293,294 One study reported positive H pylori results in 6 of 31 children (19%) with middle ear effusion.²⁹⁵ Miura et al²⁹⁶ suggested that reflux disease in children with chronic otitis media with effusion appears to be factorial, but the “cause-effect” relationship is unclear. They suggested that, based on current research, “anti-reflux therapy for otitis media cannot be endorsed.” Other studies have identified increased presence of reflux in children with upper respiratory infections associated with cough, runny nose, otitis media, and chronic rhinosinusitis.^297–299 Another study reported the presence of pepsin in the tears of children with LPR.³⁰⁰ Andrews and Orobello³⁰¹ compared biopsies of the posterior cricoid region and nasopharyngeal pH results in diagnosing LPR in children. They reported their retrospective review of 63 patients ages 6 months to 17 years, and found 80% of patients tested positive for reflux by both methodologies. Katra et al²⁹⁵ investigated the relationship between H pylori in hyperplasia of the adenoids and reflux episodes in children detected by impedance and pH monitoring. Their study population was small, 30 children with a mean age of 5.34 years. The children underwent adenoidectomy and pH/impedance monitoring with a proximal impedance sensor 1 cm above the UES. Their results confirmed their hypothesis that reflux episodes that reached the UES could have a significant role in H pylori reaching lymphoid tissue in the nasopharynx and the development of adenoid hyperplasia in children.

Diagnostic Tests for Reflux

The approach to diagnosis of laryngopharyngeal reflux (LPR) in a general or otolaryngologic practice includes careful physical examination and diagnostic testing. This section discusses the use of each of these modalities in the management of GERD in general, with specific reference to the otolaryngologic patient (Table 11–5).

Therapeutic Trial

When a patient presents with typical heartburn and regurgitation, diagnostic studies may not be needed. Relief of symptoms after a therapeutic trial with H₂-antagonists, prokinetic agents, or PPIs for 8 to 12 weeks can confirm that the symptoms are secondary to GERD. Because heartburn generally is absent in the otolaryngologic patient, the endpoint of the therapeutic trial is dependent on other presenting symptoms, and diagnostic tests are often necessary to confirm the diagnosis. Historical clues that otolaryngologic symptoms may be due to GERD, specifically LPR, include morning hoarseness, halitosis, excess phlegm, dry mouth, throat clearing, and others.

If a therapeutic trial is used in a patient with suspected GERD and otolaryngologic symptoms, higher doses of antireflux therapy, usually with a PPI, for longer periods of time are needed. However, neither the cost-effectiveness nor clinical efficacy of any medical regimen in patients with LPR has been tested. We currently use a PPI twice a day initially for a minimum of 8 to 12 weeks as a therapeutic trial for laryngeal symptoms suspected to be due to reflux.

It should be emphasized that patients with reflux laryngitis frequently require more intensive therapy with higher doses of H₂ blockers or earlier use of PPIs than patients with dyspepsia in the absence of laryngeal symptoms and signs. In addition to monitoring symptoms and signs of reflux laryngitis, response to treatment is best judged by combined intraesophageal and intragastric pH monitoring of patients while they are receiving treatment. Such studies are worthwhile even when patients are taking PPIs, because some patients are omeprazole-resistant,^168,169 and resistance to other protein pump inhibitors may occur, as well. Our recent observations suggest that omeprazole resistance also can develop in patients who respond well initially to the medication. Moreover, it must be recognized that a normal pH 24-hour monitor study does not indicate the absence of reflux. Rather, it demonstrates the absence of acid reflux. Regurgitation of pH-neutral liquid may still be present and may produce symptoms, especially in singers and actors. Study of this phenomenon and its optimal management is needed badly. At the present time, although there are no data to support the superiority of surgery over medical therapy for LPR patients, it appears that selected patients may benefit from surgery over medicine, especially considering the efficacy and decreased morbidity associated with laparoscopic fundoplication and the potential costs and risks associated with the use of H₂ blockers or PPIs for periods of many years. If endoscopic suturing and Stretta techniques prove efficacious, one or both of these techniques may be useful, but at present, they have not been studied in LPR or compared to surgery or medical therapy.

Barium Radiographs

Barium studies are relatively inexpensive and widely available for use in the diagnosis of esophageal disease. When evaluating the esophagus, a double-contrast barium swallow is needed for optimal assessment. An upper GI (gastrointestinal) series usually results in insufficient evaluation of esophageal function, concentrates excessively on the stomach and duodenum, and does not give enough attention to potential mucosal or motility abnormalities in the esophagus. A hiatal hernia is the most common abnormality seen on barium swallow. However, up to 60% of the adult population will have a hiatal hernia,³⁰² making this a nonspecific finding and not diagnostic of GERD. Free reflux is seen in up to 30% of “normal” patients and may be absent in up to 60% of patients with GERD established by pH monitoring,³⁰³ making the barium study an insensitive and nonspecific study for GERD. It has been suggested that reflux of barium to or above the carina or to the thoracic inlet is indicative of the potential for aspiration and is useful as an aid in the diagnosis of GERD-associated laryngitis. There are no prospective or controlled studies to substantiate this clinical impression. This finding is reported usually with the patient in the supine position, making this observation of relatively little use. The so-called “high” reflux on a barium study has not been well correlated with proximal acid exposure on ambulatory pH monitoring. Barium swallow with water siphonage has been used to aid in the diagnosis of reflux in otolaryngology patients. Patients may show abnormalities on barium swallow with water siphonage, which may be interpreted as confirming a diagnosis of pathologic reflux, although interpretations should be made with caution as the true positive predictive value has not been confirmed. However, barium swallow with water siphonage has more value than recognized by many radiologists. The literature on this subject was reviewed in 1994 by Ott.³⁰² Because early reports revealed a wide discrepancy in reflux detection rates, barium esophagrams were considered insensitive, and provocative tests (ie, water siphonage) were believed to increase the sensitivity at the expense of specificity. Thompson et al found that a reflux detection rate increased to 70% when using the water siphonage test, as compared with 26% for spontaneous reflux.³⁰³ However, this gain in sensitivity may be counterbalanced by the low specificity of this test.

In professional singers and actors especially, barium swallow with water siphonage seems to provide a good clinical approximation of daily reflux episodes. To optimize mucosal function, it is essential for singers and actors to remain well hydrated. Consequently they drink large amounts of water, routinely carry water bottles with them, and drink substantial quantities shortly before they sing. This routine behavior is similar to the water siphon portion of the barium swallow, which raises the question of whether positive water siphonage tests may provide useful information, at least in professional voice users, even when a 24-hour pH monitor study is normal. Specific mucosal abnormalities on double-contrast barium studies, such as thickening of esophageal mucosal folds, erosions, or esophageal ulcers, are seen in a minority of patients with GERD, making this study relatively insensitive for this diagnosis. The diagnosis of Barrett’s esophagus is conclusive also by a barium swallow.

The optimal use of the barium study is to evaluate patients with suspected complications of GERD, such as motility abnormalities or peptic stricture that are commonly seen in patients with solid and/or liquid dysphagia. A barium swallow can identify rings, webs, or other obstructive lesions including carcinoma that are seen in patients with dysphagia, but these are unusual complications of GERD. A solid bolus such as a marshmallow or a barium cookie can be given to help localize the site of obstruction in a patient with solid dysphagia.

Although the barium swallow allows demonstration that reflux is occurring and can demonstrate mucosal injury, it is of inconsistent value in establishing a diagnosis of GERD. Its best use is in evaluation of the patient with dysphagia, and it should be performed in conjunction with endoscopy in these patients. Nevertheless, in some patients, barium esophagram provides important additional information that may be missed without radiologic imaging or esophagoscopy.^28,303,304 In a series of 128 patients, for example, barium studies showed esophagitis in 18%, a lower esophageal ring in 14%, and peptic stricture in 3% of patients.^305,306 Consequently, if endoscopy is not planned, patients with LPR should be considered for further evaluation by barium swallow.

Radionucleotide Studies

Scintigraphic studies have been suggested as valuable in diagnosis of GERD.³⁰⁷ A radioisotope (Technetium 99m-sulfur colloid) marker is mixed with a measured quantity of liquid (usually H₂O), and graded abdominal compression is used to unmask reflux. Originally proposed as a sensitive test, its reliability has been questioned, and it is no longer considered a useful investigation.³⁰⁸

Endoscopy

Endoscopy is used to document mucosal disease and establish a diagnosis of erosive esophagitis or Barrett’s metaplasia. When patients with frequent heartburn and regurgitation are studied prospectively, erosive esophagitis is seen in 45 to 60% of patients.³⁰⁹ The others will have nonerosive disease (mucosal edema, hyperemia, or a normal-appearing esophagus). Erosive esophagitis suggests a serious form of GERD in which patients require continuous medical therapy with a PPI or antireflux surgery for effective symptom relief and healing. Barrett’s esophagus is seen in 10 to 15% of reflux patients undergoing endoscopy.⁹⁴ Unfortunately, there is no classic presentation of Barrett’s esophagus, but it is most common in white males over 50 years of age.⁷⁵

Erosive esophagitis is uncommon in patients with extraesophageal symptoms. Although 50% of patients with unexplained chest pain and normal coronary arteries have GERD, the prevalence of erosive esophagitis is 10% or less.⁶⁴ GERD-associated asthma and evidence of esophagitis on endoscopy have been reported in 30 to 40% of adult patients.^59,60 In patients with reflux laryngitis, erosive esophagitis is seen in only 20 to 30%, making this study of low diagnostic yield in GERD.¹⁶

There are no absolute indications for endoscopy in the patient with suspected GERD. In general, endoscopy is performed in patients who do not respond to a therapeutic trial of medical therapy, patients with symptoms for greater than 5 years to rule out Barrett’s metaplasia, and patients with the “alarm” symptoms of dysphagia—odynophagia, weight loss, anemia, or gastrointestinal bleeding.³¹⁰

Endoscopic findings may help predict the prognosis and outcome of medical therapy. Patients with erosive esophagitis will almost always require longterm PPI therapy for healing and symptom relief. Recurrence of erosive esophagitis is seen in up to 80% of patients within 3 to 6 months following the discontinuation of medications³¹¹; these patients usually require continuous pharmacologic therapy for effective long-term control. Because patients with nonerosive esophagitis seldom progress to more severe forms of esophagitis, they can be managed with a range of pharmacologic treatments. Endoscopy is useful for long-term treatment planning in difficult-to-manage cases.

Given the rarity of erosive esophagitis, we do not use endoscopy routinely as the initial study in patients with suspected GERD-related otolaryngologic disease, chest pain, asthma, or cough, preferring ambulatory dual-probe pH monitoring or a therapeutic trial of antireflux medications as the initial diagnostic test.

Esophageal Biopsy

Biopsy and cytology are of limited value in evaluation of the patient with GERD unless Barrett’s esophagus or malignancy is suspected, and the author (POK) biopsies only the esophagus in these patients. The light microscopic signs of GERD—elongation of rete pegs and hyperplasia of the basal cell layer do not distinguish between acute and chronic disease and do not help predict response to therapy. The microscopic signs of active esophagitis, polymorphonuclear leukocytes and eosinophils, are seen in a minority of adult patients, so they are insensitive diagnostic findings. Biopsy may be more useful in the pediatric population where the frequency of these findings is higher.

If Barrett’s metaplasia is suspected, a systematic biopsy protocol should be followed to confirm the diagnosis and rule out dysplasia or carcinoma. Endoscopic surveillance with biopsies to rule out dysplasia every 1 to 2 years is the current standard of practice for management of patients with Barrett’s metaplasia.⁷⁴

Prolonged Ambulatory pH Monitoring

Prolonged (16–24 hour) pH monitoring is the most important study to quantify esophageal reflux and determine whether symptoms are related to GERD. The study is performed by placing an antimony catheter (2 mm diameter) transnasally into the distal esophagus with an electrode placed 5 cm above the lower esophageal sphincter, which is identified by esophageal manometry (Figure 11–17). Precise positioning is important for accuracy in the interpretation of results. The probe is connected to a small microcomputer that is worn on a belt or clipped to the waist so that the patient can be monitored in an ambulatory setting. Activity can be tailored to provoke reflux in the setting in which symptoms are typically produced. For example, a patient with chronic hoarseness who sings professionally will be reminded to sing during the study. We have found some patients who reflux constantly during singing, and rarely at other times (see Figure 11–17).

Multiple electrodes can be placed on a single catheter to monitor intragastric and intraesophageal pH, distal esophageal, and proximal esophageal acid exposure, or all 3 simultaneously. Abnormal acid exposure in the proximal esophagus, just below the upper esophageal sphincter, predicts the potential for aspiration in patients with otolaryngologic symptoms. An intragastric electrode allows monitoring of the gastric acid response to antireflux therapy. Several investigators have placed probes above the upper esophageal sphincter in the hypopharynx^312,313 to document reflux above the UES, thus being more certain of aspiration as the cause of symptoms. Unfortunately, this placement creates difficulty in standardizing the distance between the proximal and distal probes and causes difficulty in placement of the distal esophageal probe 5 cm above the LES, the standard used in developing normal values. Probes in the hypopharynx can be uncomfortable; normal values are not available; and pharyngeal probe data are occasionally subject to interpretation error, including incorrect diagnosis of reflux due to probe drying and acidic food or liquid ingestion, both artifacts resulting in a drop of pH to less than 4 that is not a true reflux episode. Some investigators who believe that pharyngeal probe placement is important and that data are valid and reliable, for example, Koufman and colleagues,³¹² place the proximal probe immediately above the cricopharyngeus, posterior to the larynx. This position prevents drying of the probe and reportedly produces valid data. They consider any pharyngeal acid exposure (even one episode) abnormal. Although placement of the proximal probe (just above the cricopharyngeus or just below it) remains controversial, the need for dual-probe pH monitoring is clear. Without a proximal probe, the sensitivity of single-probe esophageal pH monitoring has been shown to be only 62% for LPR190; that is, 38% of patients had pharyngeal reflux with distal esophageal parameters that were considered within normal limits. At present normal values for distal reflux at 5 cm above the lower esophageal sphincter and for proximal reflux 20 cm above the sphincter are available, making this a more useful protocol.^95,313 Normal values vary slightly between laboratories and should be included with reports from the laboratory performing the procedure.

The microcomputer (data logger) has a symptom button that allows recording of up to 6 symptoms during a single study. The patient is asked to push the symptom button as well as to record symptoms on a diary card. This allows correlation of reflux events with symptoms in order to determine a symptom index,³¹⁴ which is especially valuable in patients with asthma, cough, and chest pain and allows correlation between symptoms and reflux in patients who have continued symptoms on medical therapy. Symptom correlation in the otolaryngologic patient may be more difficult on a single study, particularly when symptoms are continuous and not produced by a single reflux episode. This scenario is more likely in patients with laryngitis or chronic sore throat. Other symptoms such as throat clearing, cough, or symptoms provoked by singing may be correlated with single reflux episodes.

Prolonged pH monitoring is used in patients with heartburn to establish a diagnosis when symptoms have not responded to a trial of antireflux therapy and endoscopy is negative. In this case, a single-channel pH probe can be placed with the distal probe 5 cm above the lower esophageal sphincter. Symptoms are correlated with reflux, and reflux frequency is assessed. Patients with known GERD who have heartburn and regurgitation not responding to medical therapy can be monitored while still on therapy with a dual-channel intragastric and distal esophageal probe to assess the adequacy of gastric acid suppression, to assess esophageal reflux frequency, and to correlate symptoms with reflux events. Patients with continued esophageal acid exposure and/or symptoms may require additional therapy.

Patients with otolaryngologic symptoms, or other upper airway symptoms suggestive of GERD, are ideal candidates for prolonged ambulatory pH monitoring. We prefer performing monitoring early in the clinical course to establish a diagnosis and symptom correlation when possible. Dual-channel pH monitoring with one electrode 5 cm above the LES and a second probe 20 cm above in the proximal esophagus just below the UES (Figure 11–18) is our procedure of choice. Abnormal distal esophageal acid exposure can be documented, establishing a diagnosis of GERD. Abnormal proximal reflux can be demonstrated, suggesting the potential for aspiration and lending stronger probability that the otolaryngologic symptom is due to GERD. If symptom correlation can be demonstrated, this will establish the diagnosis. The presence of proximal reflux appears to predict the response to medical therapy in patients with pulmonary disease.^315–317 This is less clear in the otolaryngologic patient. A small percentage of patients will have normal percent time of distal esophageal acid exposure, but demonstrable increased frequency of proximal reflux or reflux into the hypopharynx. This is seen in up to 30% of patients with otolaryngologic symptoms. In one study of 10 patients with reflux laryngitis, 3 of 10 (30%) demonstrated hypopharyngeal reflux with normal distal acid exposure.³¹⁸ In a larger, retrospective series of patients with pulmonary disease, 12% of patients reviewed had only abnormal proximal reflux³¹⁷—a group that would have been diagnosed incorrectly as normal had only a single-channel study been performed. These patients should be considered abnormal and treated aggressively. Studies in our laboratory have shown that dual-electrode pH recording can document abnormal distal and proximal esophageal reflux induced by singing. A singing challenge (shown in Figure 11–17) can also unmask significant reflux in patients with otherwise normal 24-hour pH monitor studies.

Studies in adults with a variety of otolaryngologic symptoms demonstrate abnormal amounts of acid in reflux in up to 75% of patients.¹⁶ Abnormal acid exposure has been documented in upright and supine positions, although upright reflux seems more common. Ambulatory pH monitoring is most useful in assessing response to antireflux therapy, particularly in the patient who has failed to respond to a therapeutic trial of a PPI twice daily for 8 to 12 weeks. Intragastric, distal esophageal, and proximal esophageal pH can be monitored while therapy is continued. Adequacy of intragastric acid suppression can be assessed, as can the presence of esophageal acid exposure and correlation between reflux events and symptoms. This study is particularly valuable in patients who do not respond (or are resistant) to PPIs. If acid reflux is still present, treatment can be increased or modified. If adequate acid suppression is achieved and symptoms persist, alternative diagnosis can be sought. However, the definition of “adequate acid suppression” is more complex than it might appear. For example, Shi et al³¹⁷ reviewed 771 consecutive patients who had undergone 24-hour pH monitoring studies. Statistically significant association was found between symptoms during reflux episodes in 96 patients (12.5%) with esophageal acid exposures that were within the laboratory’s normal values. In these patients, the duration of reflux episodes was shorter and the pH of reflux episodes was higher than in patients who were diagnosed with GERD. The authors hypothesized that the underlying pathological feature in these patients was hypersensitivity to acid. However, there are other possible explanations. For example, it is possible that normative values have been established at levels higher than desirable.

An area of some controversy is the evaluation of the patient with continued symptoms but no esophageal acid exposure on 24-hour pH monitoring. This is when a question of alkaline or pH-neutral reflux may be raised. Current pH monitoring technology makes it possible to detect bilirubin pigment (bile) by use of a bilitec probe in addition to standard probes used for 24-hour pH monitoring. However, current data using this technique suggest that esophageal bile reflux rarely, if ever, occurs in the absence of acid reflux, making the bilitec probe useful only in select patients (see Approach to the Patient with GERD, below). The diagnosis of alkaline reflux should not be made based solely on the rise in pH above 7; careful analysis of pH rises above 7 coupled with symptom correlation may suggest alkaline reflux; but it is rarely, if ever, diagnosed conclusively by pH testing. A new approach to detecting pH-neutral reflux utilizes measurements of electrical impedance at numerous points along an esophageal probe. This device detects the presence of liquid in the esophagus regardless of pH, and sequential measurements along the chain of sensors indicate whether the liquid is traveling from proximal to distal or is refluxing from distal to proximal. This technology is now available commercially and has proven useful. However, the decision to proceed with surgical management for the suspicion of alkaline or pH-neutral reflux is a clinical one and must be made after careful consultation among the health care team, including the patient. Current pH technology cannot confirm this diagnosis definitively.

Although prolonged pH monitoring is used extensively in the evaluation of patients with suspected LPR, controversy exists as to its overall sensitivity and specificity. This is particularly important when addressing the issue of hypopharyngeal reflux. A recent study by Shaker and colleagues³¹⁸ found a similar number and duration of hypopharyngeal reflux events in control subjects and patients with suspected LPR. Another study³¹⁹ revealed no major difference between hypopharyngeal acid exposure in 36 patients with LPR signs and symptoms and healthy controls. Using a newer methodology for triple-probe monitoring, Maldonado et al³²⁰ found a 10% prevalence of abnormal pH in normal controls. These data remind us of the importance of standardization of measurement of hypopharyngeal reflux to optimally understand the causal relationship between this phenomenon and LPR.

Sato et al³²¹ described their experience using a tetra-probe 24-hour pH monitoring system. The proximal probe is placed in the hypopharynx; a second probe is placed in the mid-esophagus; the third probe is placed a few centimeters above the LES (but not specifically 5 cm); and the distal sensor in placed in the stomach. This system is advantageous in that it measures pH simultaneously at all 4 sites, providing information regarding the relationship of these values. However, in our opinion, there are significant disadvantages associated with failure to have a sensor 5 cm above the LES, a standard site that permits comparison with other literature. In a separate report,³²² the authors described their results using the tetra-probe 24-hour pH monitoring system to evaluate patterns of LPR and GERD and to determine the validity of using pH 5 as an indicator of LPR. The data suggest that pH levels less than 4 and less than 5 are indicative of significant LPR events. An interesting study was reported in 2010 by Zelenik et al.³²³ They performed a prospective dual-probe pH monitoring study in 46 patients with complaints of globus pharyngeus for more than 3 months. They collected data from this group using a pH less than 4 and from a second data group using pH both less than 4 and less than 5. Extraesophageal reflux was found in 23.9% of their patients using analysis of pH less than 4. When data were calculated using pH both less than 4 and less than 5, they found extraesophageal reflux in an additional 4 (8.7%) patients whose reflux had not been detected when pH less than 4 alone was used.

In addition, abnormal findings on pH monitoring do not necessarily predict patient response to therapy. Similarly, a recent placebo-controlled study³²⁴ of 145 patients treated with either esomeprazole or placebo found that symptomatic or laryngeal improvement was independent of pretherapy pH-monitoring results. The apparent dichotomy in the clinical usefulness of hypopharyngeal pH monitoring is likely due to several key factors:

1. Lack of consensus on the duration and amount of reflux that constitute abnormal pharyngeal acid exposure.

2. High operator dependence and variability of probe positioning (eg, use of manometry vs direct visualization)

3. Variable sensitivity of pH monitoring in detecting reflux, which may vary from day to day.³²⁵

All these variables have been reported to result in a diagnostic yield from 14 to 83%.³²⁶ The yield may improve with better data evaluation parameters. Reichel and Issing³²⁷ suggest the use of reflux area indices of 4 and 5, which would include time at pH 5. The data regarding pepsin activity at pH 5 have generated somewhat less controversy than the data on activity at higher pH. The reflux area index is calculated from the number and duration of proximal reflux events and the degree to which the pH drops below a given value.³²⁷ Incorporating several data points and including a reflux area index of 5 may increase the sensitivity of pH monitoring.

As such, 24-hour pH monitoring cannot be used to conclusively establish or rule out reflux as the cause of suspected LPR. Until the value of other tests such as wireless esophageal pH monitoring and/or impedance monitoring (see later section) has been established, an empiric trial with an effective dose of PPI remains the most important determinant of the relationship between reflux and LPR symptoms.

Ongoing research continues to examine the value of pH impedance monitoring in the diagnosis and treatment of patients with LPR.^328–333 In 2015, Gooi et al³³⁴ reported the findings of the American Bronchoesophageal Association (ABBA) evaluation of the changes in the management and evaluation of patients with LPR over a 10-year period between 2002 and 2012. 426 ABBA members were e-mailed questionnaires to be completed online, of whom 63 (14.8%) responded. Their responses showed that dual pH probe testing remained highly regarded for sensitivity and specificity in the evaluation of LPR but was being used less often in 2012 compared to 2002 (63.8% vs 78.3%, respectively). Empirical medical management of LPR was more common in 2012 (82.6% compared to 56.3% in 2002).

Telemetry Capsule pH Monitoring

The development of telemetry capsule-based (tubeless) pH monitoring has added to laryngologists’ armamentarium and ability to monitor patients with suspected reflux disease. The system is safe, well tolerated, and reliable, allowing 48-hour assessment of esophageal acid exposure. It is unobtrusive, comfortable, and allows longer duration of acid monitoring. Its use in LPR has not been studied, and the technology is limited by current inability to place the capsule proximally due to discomfort and bulk. Further refinements in methodology, recording protocols, and diagnostic accuracy have made this an extremely useful test in typical GERD patients. One of us (POK) uses the telemetry capsule to perform off-therapy monitoring in patients prior to antireflux surgery to be certain that abnormal acid exposure is present prior to an operation. This new technology will likely become valuable in assessing patients with LPR.³³⁵

Role of Combined Multichannel Intraluminal Impedance pH

Combined multichannel intraluminal impedance (MII) and pH monitoring is a promising technique that identifies gas, liquid, and mixed gastroesophageal reflux and allows differentiation of them into acid (pH <4), weak acid (pH 4.1–7) and nonacid (pH > 7) types. The current catheter allows simultaneous monitoring of intragastric and distal esophageal pH with the ability to assess the height of refluxate from 3 to 17 cm above the LES. Early data suggest that a number of patients with so-called extraesophageal disease may have symptoms associated with weak acid or nonacid reflux, although the clear relationship of this type of reflux to LPR remains to be studied. Wu³³⁶ reported that combining MII and dual-channel pH monitoring increased the diagnostic yield by ~10% compared with combined MII and single-channel pH monitoring systems. Lee and colleagues³³⁷ reported a 2-fold increase in diagnosing reflux in patients with LPR symptoms, using combined dual-channel impedance/pH-metry. Loots et al³³⁸ reported that the addition of MII to standard pH monitoring increased the yield of symptom association in children and infants with confirmed reflux. Combined MII/pH monitoring is a useful tool in evaluating all patients with persistent symptoms of acid-suppressive therapy, particularly those with LPR.³³⁹

Pharyngeal Monitoring

Although the gold standard for diagnosis always has been pH testing, many physicians rely on the reflux finding score (RFS) and reflux symptom index (RSI) as means to diagnose laryngopharyngeal reflux (LPR) and indications to begin treatment with PPI or H₂-blockers. In addition, many believe that a reduction in RSI/RSF scores reflects improvement or resolution of LPR. However, numerous studies have raised questions about the utility or validity of these indicators, and over the safety and effectiveness of empiric prescription of antireflux medications. In a 2011 prospective study of 82 participants (72 patients and 10 controls), Musser et al³⁴⁰ studied the agreement between RSI, RFS, and pH probe findings at pH levels of 4 and 5. Regardless of which pH criterion was used, RSI and RFS failed to identify the LPR patients from controls, and failed to correlate with severity of disease. In addition, in 2005, Park et al³⁴¹ evaluated 57 patients with globus sensation with 24-hour pH probe, RSI and RFS scores. In this study, RFS and RSI showed low specificity, and there was no significant difference between test and control groups. These findings suggest that both RSI and RFS should not be used as the sole diagnostic test for initiating PPI therapy.

Routine, long-term use of PPI in patients with elevated RSI/RFS, without proven LPR, may be ill-advised because of potential complications of therapy discussed elsewhere in this chapter. In addition, while PPIs are widely prescribed, resolution of LPR symptoms has been equivocal and often incomplete. Patients presenting with symptoms consistent with LPR are difficult to diagnose definitively, with most technologies providing no solid evidence on which a diagnosis or treatment can be based.³⁴²

Borrowing from the standards of gastroesophageal reflux disorder (GERD), the suggested definition of LPR has been a single episode of pH <4 or 1% total time below a pH of 4 on dual lumen probe.^343–345 Although standard 24-hour monitor technique has good diagnostic utility for the presence of LPR, it often causes significant patient discomfort, both in placement and throughout the monitoring period.^346,347 Other studies have reported lack of sleep and dysphagia with a dual lumen probe.³⁴⁸ Hence, patients who alter their eating and sleeping habits may have tests that are not representative of their disease state. A new probe has been developed for monitoring the oropharyngeal pH and for the diagnosis of at least severe LPR. Multiple studies have demonstrated the ability of this device to detect reflux events compared to a traditional pH probe.

In 2009, Golub et al³⁴⁸ studied 15 patients with symptoms of LPR by simultaneously placing a dual lumen pH probe and the Restech (Houston, Texas) oropharyngeal probe. This study revealed that the correlation between the 2 probes for reflux events was 0.95 (P < .001). In a separate study, Wiener et al³⁴⁶ compared the Restech probe to 3 traditional pH probes (2 esophageal, 1 pharyngeal), in 15 patients with symptoms of LPR. In this study, all events detected by the Restech probe were preceded by events in the distal esophagus. This study revealed the Restech probe to be a sensitive device that can detect pH events that originate in the distal esophagus and migrate toward the oropharynx.

In addition to the high correlation with traditional pH, the Restech probe offers some significant advantages. First, sensors in conventional pH probe require a liquid environment to function properly. Traditional probes placed in the oropharynx have a tendency to dry out and lead to “false-positive” readings (pseudo-reflux). The Restech probe does not require a liquid environment to provide real-time pH readings. Second, the sensors for traditional probes are located on the side of the catheter, which leads to the possibility of masking by the mucosal wall and “false-negative” readings. The sensor on the Restech probe is located on the teardrop-shaped tip, which decreases the possibility of mucosal masking. Furthermore, the placement of a traditional probe requires manometric (radiographic in infants) confirmation of placement. The placement of the Restech probe can be confirmed visually with simple oropharyngeal examination by localizing the red LED light in the catheter tip.

Currently, multiple studies have addressed the utility of the Restech probe in a clinical setting. A recent study by Friedman et al³⁴⁹ revealed that in a group of 163 patients with suspected LPR, there was no correlation between RSI and the oropharyngeal environment. In addition, the calculated positive and negative predictive values for RSI and severe reflux (positive Ryan Score) were 44.3 and 58.7%, respectively. The authors recommend that diagnosis and treatment decisions regarding LPR should be on the basis of symptoms (RSI), signs (RSF or oropharyngeal findings), and a confirmatory test (pH testing). In a recent prospective study³⁵⁰ of 18 patients with symptoms of LPR, 100% of patients with positive Ryan Scores (by Restech) responded to PPI therapy, while less than half of those with a negative study responded. While the specific pH profile of Ryan-negative PPI responders needs to be elucidated, these results suggest that the Restech probe may be useful in determining which patients will respond to PPI therapy. In a separate 2011 study by Friedman et al,³⁵¹ 143 patients were offered either empiric PPI treatment (70 patients) or Restech-based treatment (73 patients) for symptoms of LPR. This study revealed that patients tested with the Restech probe had significantly greater compliance with medication (68.5 vs 50%, P = .019) and lifestyle modification (82.2 vs 25.7%, P < .01) as well as significantly greater reduction in RSI (36.6 vs 24%, P = .023). Collectively, these studies suggest Restech has utility in the diagnosis of LPR and selection of patients who might benefit from treatment, as well as in improving patient compliance.

Based on current literature and the side effect profile of antireflux medication, questions must be considered about whether to treat all symptomatic patients empirically without first determining their oropharyngeal pH profile. More research is needed (and is in progress). If the Restech probe is really as reliable and valid as traditional dual lumen probes and offers some advantages over dual lumen probes placed below the UES or in the oropharynx, it may become a standard diagnostic tool. While often associated with GERD, LPR is a different disease state that presents its own set of problems in terms of diagnosis and treatment protocols. This becomes especially true with patients in the “gray zone” presenting with mild to moderate reflux. Currently, multiple groups are using the Restech probe to help better determine disease state/severity. This determination should lead to improved ability to predict those who will respond well to standard antireflux therapy versus those who need more aggressive management. Equally important is the elimination of empiric treatment for those who do not have LPR and faster work-up to identify the true etiology of their symptoms and signs.

Esophageal Manometry

Manometry will establish abnormal LES pressure or esophageal motility and is necessary preoperatively to evaluate contraction amplitude in the esophageal body. A single measurement of LES pressure is rarely low in patients with GERD. In the author’s (POK) experience, only 4% of patients with GERD have a low LES pressure.³⁵² Esophageal motility abnormalities are found more frequently. The most common finding appears to be ineffective esophageal motility (IEM) (amplitude of contraction in the distal esophagus less than 30 mm Hg occurring with 30% or more of water swallows). In our experience, this is the most common abnormality in patients with GERD, seen in approximately 35% of patients with esophagitis.³⁵² IEM appears even more common in GERD-related laryngitis, asthma, and cough.⁵⁷ Esophageal manometry is performed prior to antireflux surgery to establish the presence or absence of ineffective esophageal motility. The surgeon will usually perform a Nissen fundoplication (360° wrap) in patients with normal peristalsis and a Toupet procedure (240° wrap) in patients with significant IEM. Patients with IEM and respiratory symptoms do not appear to respond as well to antireflux surgery if respiratory complaints are the presenting symptom.

Esophageal manometry is important also for proper placement of probes for pH monitoring. Although the proximal probe can be positioned accurately by direct vision using a flexible fiberoptic laryngoscope, mirror, or telescope, without manometry there is no way to position the distal probe precisely. However, when compared to manometry, even proximal probe placement was accurate in only 70% of cases in the study by Johnson et al.³⁵³ Distal probe placement was accurate in just 40% of cases using estimated interprobe distance. Using fixed interprobe distances of 15 cm and 20 cm, distal probe placement was accurate in only 3% (for 15 cm) or 40% (for 20 cm) of cases. These errors are critical because the normative values for distal esophageal acid exposure were established with the distal probe positioned 5 cm above the LES.^354–357 Even slight modifications in the distance between the distal probe and the lower esophageal sphincter cause substantial changes in acid exposure data.^358–361

Hiatal Hernia

A hiatal hernia is not predictive of reflux as a cause of the patients’ symptoms.³⁶² Up to 60% of patients over the age of 60 will demonstrate a hiatal hernia identified on barium swallow examination. One study suggested that only 9% of patients with a radiographically demonstrated hernia have typical reflux symptoms.⁷⁷

Hernias do change the relationship of the LES and crural diaphragm. The LES is displaced above the diaphragm. The low pressure in the hernia can act as a reservoir for acid, allowing earlier reflux during LES relaxations, and may delay esophageal clearance.³⁶³ Patients with large hernias who also have low LES pressure may be more prone to reflux³⁶⁴ if changes occur in intraabdominal pressure.

Evaluation for Helicobactor pylori

Helicobactor pylori (H pylori) appears to play a role in the development of chronic type B gastritis, gastroduodenal ulcer disease, and gastric carcinoma.^365,366 Its significance in gastroesophageal reflux disease remains uncertain, and it is unclear whether it is necessary to treat H pylori in reflux patients, or even advisable to do so. H pylori can be detected through serologic determination of immunoglobin (IgG) antibodies to the organism. This blood test is performed using an enzyme-linked immunosorbent assay (ELISA)/2-step indirect sandwich assay on directly coated microtiter plates.³⁶⁷ This test is believed to have a sensitivity of 94% and specificity of 85%.³⁶⁷

Researchers continue to examine the relationship between H pylori infection, LPR, and laryngeal diseases.^368,369 Suipsinskiene et al³⁷⁰ performed a prospective case control study to examine the presence of H pylori through biopsy of the larynx in patients with laryngeal cancer and patients with benign laryngeal disease, including LPR and vocal fold polyps. They examined the biopsy results from 67 adult patients with benign and malignant laryngeal disease and compared them with a control group of 11 patients. They reported H pylori infection in greater than one-third of the patients studied, with the majority found in patients with laryngeal cancer (46.2%) and in those with chronic laryngitis (45.5%). The authors reported that these findings differed significantly from the control group (9.1%) (p < 0.05), adding that they found no significant relationship between LPR-related symptoms and H pylori detected in the larynx. They stated that H pylori can colonize in patients with benign laryngeal disease and laryngeal cancer, but that further research is needed. Another study by Islam et al³⁷¹ also examined the presence of H pylori in patients with benign and malignant laryngeal disease. They performed a prospective study of 50 patients who underwent microlaryngoscopy over a 2-year period. Their patient population was diagnosed as having LPR based on a reflux symptom index (RSI) greater than 12, and a reflux finding score (RFS) greater than 6. The patients were diagnosed with H pylori based on a positive urea breath test, HP citotoxin-associated gene A(CAGA)-IgG and HP-IGG test results. Intraoperatively, 2 surgical specimens were obtained, one from the interarytenoid area, and one from the primary vocal fold lesion. The authors reported that H pylori was not found in any interarytenoid specimen, and they found no histologic evidence of H pylori in the vocal fold pathology specimens. They also reported “there was no difference between RSS-positive and RSS-negative patients in terms of HP-IGG and UBT,” adding that, “the presence of HP in the gastric mucosa does not have an effect on the RSS and RSI.” Their study conclusions disagree with other reports in the literature and with the author’s (RTS) experience.

Approach to the Patient With GERD-Related Otolaryngologic Abnormalities

In patients with GERD-related otolaryngologic complaints, a thorough history, physical examination, and laryngoscopy should be performed (Figure 11–19). If dysphagia is present, a functional endoscopic evaluation of swallowing (FEES) or videoendoscopic swallowing evaluation should be considered, and a barium swallow should be obtained to rule out stricture or motility abnormalities. The clinician’s dilemma revolves around the choice of early diagnostic testing with prolonged pH monitoring or institution of a therapeutic trial of medication. The “best” approach is not clear. Although diagnostic testing with ambulatory pH monitoring would be ideal, there are several limitations: (1) pH monitoring is not always available; (2) the sensitivity and specificity are clearly not 100%; (3) patients do not reflux with the same frequency every day; and (4) variability in both distal and proximal esophageal acid exposure time in patients with extraesophageal GERD is common, increasing the possibility of a false-negative pH study if physiologic acid exposure is seen on a single study.

If the history and laryngoscopic examination raise a high clinical suspicion of GERD or LPR, if prolonged monitoring is not available, if frequent heartburn and regurgitation are present, or if there is endoscopic documentation of GERD or LPR, a therapeutic trial of antireflux therapy is a reasonable initial choice. An early study with empiric omeprazole, 40 mg at bedtime, in patients with suspected reflux laryngitis found a 67% response in patients with laryngeal symptoms suggestive of GERD.³⁷² Another study found 70% success with empiric omeprazole 20 mg BID for a similar time period.³⁷³ We use a trial of a PPI twice daily in combination with dietary and behavior modification initially for 8 to 12 weeks. If the patient does not respond, pH monitoring should be performed while PPI therapy is continued. A dual-channel probe with intragastric and distal esophageal electrodes should be placed to ascertain adequate gastric acid suppression and to assess the presence of esophageal acid exposure. If distal esophageal acid exposure is seen more than 1.2% of the time, this is definitely abnormal and additional medical therapy is indicated.³⁷⁴ “Normal” esophageal acid exposure, particularly when any proximal esophageal acid exposure is documented, may not always be a negative study. A positive symptom index may be seen even in patients with “normal” distal acid exposure, and this, too, is abnormal and warrants additional therapy. The absence of any esophageal acid exposure and presence of adequate gastric acid suppression (pH >4, 50% of total time) suggests adequate medical therapy in most patients, and an alternative diagnosis should be considered. If GERD-associated otolaryngologic disease is documented, endoscopy is indicated in many cases to rule out Barrett’s esophagus prior to initiating long-term medical therapy or surgery.

Stay updated, free articles. Join our Telegram channel

Join