If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The apparent late recognition of esophagitis as a widespread disease has given cause for much discussion (Fig. 1). Surprise has been expressed at the relative lack of reference to it before the mid 19th century and even thereafter it was noted to be rare. Numerous factors reflect the apparent obscurity of the disease until recently. These include a lack of the clear understanding of the anatomy and physiology of the organ, confusion regarding differentiation between its symptomatology and that of the stomach, and the lack of opportunity and technology to adequately study it. Of particular importance in terms of the latter aspect of the problem, is the advent at the turn of the century of rigid endoscopy and the introduction of first bismuth then barium upper gastrointestinal radiological studies. The subsequent evolution of fiber-optic endoscopy in the 1960s provided major access to the esophageal lumen, as did the introduction of manometry and pH probe devices to the elucidation of the function of the lower esophageal sphincter. The delineation of the secretion of acid, the ability to document its relationship to acid peptic disease, and the recognition of the concept of reflux as a related but separate disease process facilitated the ability of physicians to focus on the esophago-gastric junction as an area of considerable medical relevance.
The prolonged anatomic course of the esophagus caused confusion in the attribution of symptoms. Thus, upper area complaints were often considered to be of oral or throat origin, and pathology in the mediastinal course erroneously attributed to the heart. Similarly, at the lower end, the symptoms and their relation to ingested food often led to the conclusion that the problem was of gastric origin. Furthermore, the path of the esophagus deep within the posterior mediastinum rendered its study difficult and speculation rather than evaluation confounded a clear elucidation of its pathology and delayed the evolution of its rational therapy. Indeed the upper third was well into the 20th century, regarded as the province of the otolaryngologist, whereas the lower component considered a part of the stomach. Few physicians laid claim to the middle third because most cardiothoracic surgeons directed their attention to the more dynamic structures of the heart and lungs. Indeed as early as 1946, John Tilden Howard, the secretary of the American Society of Gastrointestinal Endoscopy, had been vocal in pointing out to gastroenterologists the need to examine the esophagus. By the 1950s the advent of therapeutic strategies for esophageal disease rendered it apparent that the esophagus was no longer “terra incognita” or “no man's land” (Fig. 2).
Ye meri or ysopaghus
Although seemingly a straightforward topic the consideration of the esophagus and its disease has proved to be a subject of considerable vexation to physicians. The esophagus has for long languished as an organ not highly regarded by any group and little understood by either laymen or physicians. Some wits even referred to it as the “humble organ” because so little attention had been devoted to it. To some extent this reflected its course through three different anatomical territories—the neck, mediastinum, and abdomen and the reluctance of the medical specialists of each group to claim the organ as their own. Even its name is poorly understood and much conjecture exists regarding the origin of the term. Although the name is considered to have originally derived from an old Homeric term for the gullet, the meaning became transmuted to reflect a combination of swallowing and food thus becoming interpreted as “a tube that conveys food.” Aristotle claimed that the name derived from its narrowness and length and some etymologists have even linked the origin of the word to an osier twig because the old Greek term could also be translated as an “eater of osiers.” Whether this reflected the early usage of such flexible branches in the treatment of esophageal obstructions is not recorded. Although the earliest English usage of the term is recorded in 1398 in a manuscript in the Bodleian collection [
] stated, “the Meri otherwise called Ysophgus is ye way of the mete & this Meri commeth out of the throte and thyrleth the mydryfe unto ye bely or stomacke.”
The etymology of symptomatology
The history of the subject of esophagitis is mired in the confusion regarding the terminology used to describe its symptomatology and by a failure to appreciate the organ as a separate entity from the stomach or the mouth. Thus dyspepsy, or as it more latterly referred to as dyspepsia, was in 1656 referred to by Blount as dyspesie and in 1661 by Lovell as an “imbecility of the stomach, which is a vice of the concocting faculty and is called apepsy, bradyspepsy, or dispepsy and diaphthora.” The recognition that it might be caused by acid and bile seems to reside in the 1829 notation of Southey who in “Epistle” in “Anniversary” noted the sensations evoked “by bile, opinions, and dyspepsy sour”! Lowell who in the 1848 Fable for Critics opined that an individual had been “Brought to death's door of a mental dyspepsy” may have first noted the relationship between the mind, stress, and acid disorders. Similarly the subject of heartburn initially regarded as a heated and embittered state of mind, which is felt but not openly expressed, was described by Shakespeare in Richard III as “A long continued grudge and hearte brennynge betwene the Quenes kinred and the kinges blood.” An awareness of this problem dates back to 1591 when Percivall described “a sharpnes, sowernes of stomack, hartburning.” Thereafter Swan in 1635 expressed the opinion that “Lettice cooleth a hot stomach called heart-burning.” As early as 1607, Topsell noted the potential severity of the disease commenting on “the hearts of them that die of the heart-burning disease.” Of note was in 1747 when Wesley identified that heart burning was “a sharp gnawing pain at the orifice of the Stomach” (Fig. 3).
Nevertheless the esophagus claimed early medical attention given the propensity of foreign bodies to obstruct it and the rapid onset of symptoms once ingress of fluids or liquids into the stomach was impaired. Furthermore there was the disturbing sensation in the chest that caused intense discomfort and was associated with the eating of certain kinds of food. Given the location of the pain and the concept of cardiac dominance in medicine, it was assumed that the origin of the discomfort was the heart and not the esophagus hence cardalgia or cardialgy were terms first used to describe the uneasy burning sensation in the lower part of the chest. The cause was believed to indicate putrefactive fermentation in the stomach. This confusion was further amplified by anatomic nomenclature that referred to the upper portion of the stomach as the cardia. An additional term of cardiodynia was also used in the nosology of Cullen and was a part of a complex classification that sought to characterize a wide variety of “pain” thereby further defining the origin of a particular disease. The time of Harvey and the description of the circulation of the blood and the ebb and flow of bodily fluids through valves and orifices had been acceptably described as examples of flux and reflux. It was therefore well accepted that food might reflux out of the gastric storage vat and the descriptive term employed to describe the consequences of reflux thus alluded not only to the burning sensation, “heartburn” but its sequela, namely cardiospasm. As early as 1597, Gerard in his Herbal (Fig. 4) noted that “a small stonecrop is good for hart-burne,” whereas Buchan in 1790 could state with assurance, “I have frequently known the heart-burn to be cured by chewing green tea.” A more contemporary comment by Beale in 1880 stated, “chalk or magnesia is taken for the relief of heartburn” [
In the earliest of times physicians were aware of the complaint of dyspepsia but were not able to distinguish its exact location. They recognized it as related to the ingestion of food and assumed it to be caused by either “bad” food or disordered digestion. Because the latter was described as “pepsis,” a process that led to what was considered abnormal digestion was referred to as “dyspepsis,” hence the origin of the term dyspepsia. Indeed the usage of the term became an almost all-inclusive description for any upper abdominal symptom. The well-known Roman practice of post cibal vomiting was practiced to relieve gastric bloating and distention. The reflux associated sensations that accompanied the regurgitation of fine Falernian wine and other Roman gustatory delicacies no doubt allowed for a fine early appreciation of the symptoms that were associated with esophagitis.
Unfortunately little rational therapeutic intervention was available for the treatment of gastric disorders and few bothered or were able to distinguish between the esophagus and the stomach as the exact source of the problem. Indeed, up until the late 19th century, the stomach itself was often not clearly recognized as a source of symptoms that were believed to emanate from a variety of digestive organs, including the liver. Thus there existed much confusion between the diagnosis of biliousness and dyspepsia. From the earliest times chalk, charcoal, and slop diets had been noted to provide symptomatic relief from dyspepsia. In the 17th century, chalk and pearl juleps were used for infant gastric disorders but little comment was made about adult problems except for Sydenham who wrote about gout and dysentery. Jan Heurne, the first proponent of bedside teaching, left a posthumous pamphlet on the diseases of the stomach in 1610. Similarly, Martin Harmes and Ferriol in 1684 and 1668 respectively produced books on diseases of the stomach and intestines. In 1664 Swalbe published a long satire on “the quarrels and opprobria of the stomach” (prosopopoia). The favorite theme of 17th century physicians was the dyspepsia caused by gastric atony implicit in the title “De imbecilitate ventriculi.” This term was continued into the 18th century when it was equated with the “hectic stomach” (Arnold 1743) and later with the term of “embarras gastrique” used by the French physicians. Throughout the 18th century there were various descriptions of gastric and intestinal diseases but no specific and logical remedies were recorded. The doctrine of acidity was widespread and prevalent and the general literature of dyspepsia was large. Thus in 1754, Joseph Black writing on the subject of carbon dioxide commented on “De humore acido a cibo orto” (acid and eating). In 1835 in Paris, Jean Cruveilhier (1791–1873) had written extensively on the pathology of peptic ulcer and the condition was referred to as Cruveilhier's disease. Yet in all these tomes little specific therapy apart from dietary adjustments or homeopathic remedies is apparent (Fig. 5).
The relation of acid (its presence, absence, or exact nature) and its relation to digestion had been considered in detail by several scientists and philosophers. Thus R. Reaumur (1683–1757) had shown that gastric secretion caused digestion in a buzzard, L. Spallanzani [
], whereas J. Hunter (1728–1793) had equivocated. F. Leuret (1747–1851) and J-L Lassaigne (1800–1859) of Paris had claimed it was lactic (1825), Young of Philadelphia that it was phosphoric, W. Prout (1785–1850) (London, 1823), F. Tiedemann (1781–1861), and L. Gmelin (1788–1853) of Heidelberg (1826) that it was hydrochloric acid. The relationship of acid to ulceration was well accepted after the experimental demonstration by Schwartz [
The elucidation of the pathobiology of acid peptic related disease of the gastro-duodenum and esophagus has accelerated dramatically in the last 50 years with the advent of scientific techniques that have facilitated the measurement of acid secretion and thereafter the study of the cell biology of the gastric mucosa. The identification of cellular mechanisms has facilitated the development of therapeutic strategies capable of not only ameliorating the mucosal ulceration but also eradicating or abrogating the putative causes. Whereas acid and pepsin have long been recognized as pathogenic with regard to the mucosa it is only recently that an infective component has been established as of critical relevance to the etiology of gastric and duodenal ulcer disease. No such entity has yet emerged for the esophagus and it is likely that the pathogen or pathogens may well be agents used in the contemporary preservation of food.
Early 20th century management strategies of esophagitis thus focused chiefly on the avoidance of acid containing foods or substances containing capsaicin related compounds, such as peppers and curries. The use of bland diets and milk ingestion was then augmented by the addition of neutralizing compounds and antacids (Fig. 6). Loose clothing, loss of weight, and elevation of the head of the bed also achieved considerable popularity as did a host of proprietary medicines guaranteed to cure reflux, dyspepsia, impotence, and alopecia! Unfortunately, such regimes lead to a significant decline in the quality of life and in addition to side effects (diarrhea, milk/alkali syndrome) associated with the use of excessive alkali and cation containing agents, or severe depression based on failed expectations. Advances in antisepsis, anesthesia, and surgery led to vain attempts to pexy the stomach, resect the acid secreting area, and reduce associated hiatus hernias. The advent of the Nissen wrap and a wide variety of modifications resulted in considerable initial enthusiasm but the high morbidity and failure rate of the procedure mitigated against its widespread acceptance. The subsequent development in the 1970s of histamine–2 receptor antagonists (H2RA) and in the 1980s of proton pump inhibitors dramatically altered the efficacy of treatment to the extent that open surgery virtually disappeared from the therapeutic horizon. The subsequent recognition of a bacterial pathogen of the gastro-duodenal component of acid peptic disease and its antibiotic eradication have more recently further amplified therapeutic gains and resulted in an even greater focus on the residual problem represented by esophagitis. In this respect surgery that had for the most part become obsolete as a method of decreasing acid secretion evolved into a more targeted option and has provided a new lease of life as minimally invasive technology suggested the possibility of a safer and more efficacious therapy. Similarly, pharmacological intervention that had for the most part been directed at the development of acid suppressive agents now assessed the potential of motility agents capable of either facilitating esophageal clearance of refluxate or amplifying lower esophageal sphincter pressure to decrease reflux (Fig. 7). By the beginning of the 21st century the old Iatromechanical school of thought once again surfaced as both endoscopists and laparoscopists sought to bolster the lower esophageal sphincter by considering not only wraps, but also prosthetic devices, collagen injections and radio frequency induced fibrosis. Thus, even at the advent of the new millennium, opinion was divided between the Iatromechanical School and the Iatrochemists who believed in a pharmacological resolution to the problem. The environmentalists who claimed that food additives were the source of the esophagitis and reflux a correlateable epiphenomenon, gained some acceptance but were for the most part regarded as modest contenders in the “cause” of cure. The role of coffee, onion, alcohol, and peppers was for the most part disregarded because treatment of disease by abstinence (eg, the advocacy of celibacy in the treatment of venereal disease) has long been the most unpopular form of therapy. Intervention has been, and will for the foreseeable future always be the most sought after form of therapy!
The first description of the disease by Galen was accurate although somewhat incomplete. Nevertheless, he duly noted that although tumors or paralysis may cause difficulties with swallowing the esophagus when inflamed interfered with swallowing by itself acting as a hindrance caused by the associated excruciating pain. In 1884, Morell Mackenzie defined esophagitis as an “acute idiopathic inflammation of the mucous membranes of the esophagus giving rise to extreme odynophagia and often to aphagia.” It was this Anglicized description that was used to qualify the disease known as “esophagitis” that had first properly been described by John Peter Frank in 1792 (Fig. 8) [
Although similar observations had been made previously by Boehm in 1722 who described an acute pain “which reached down even to the stomach and which was accompanied by hiccup and a constant flow of serum from the mouth,” the latter had not defined the condition as accurately. Honkoop published a thesis on inflammation of the gullet in 1774 and in 1785, Bleuland a physician was struck down with the disease and carefully recorded the details of his illness. The first reports of the condition of “esophagitis” in children originated with Billard who in 1828 published a statement regarding the ailments of newborns. In 1829, Mondière wrote a thesis describing his own severe personal experiences with esophagitis and in addition collected much of the extant material on the subject. His laborious compilation represents more industry than discrimination but the thesis provides a good overview of early 19th century thoughts on the esophagus. Indeed it is from the work of Mondière that much of the literature sources of esophageal diseases were drawn for the next century. These included Velpeau's “The Esophagus–A Dictionary in Four Volumes,” Follin's essay “Considerations of Esophageal Disease” and Copland's Dictionary. In 1835, Graves of Dublin made some observations on the disease but thereafter confined his attentions to the thyroid, achieving eponymous recognition for his contributions to the subject. In 1878, Knott in Dublin published “The Pathology of the Esophagus” and included in it the cases of esophagitis described by Roche, Bourguet, Broussais, and Paletta and some original illustrations of diseases of the organ. The topic of the esophagus was further illuminated by communications from Hamburger, Padova, Laboulbène, and several other distinguished physicians of the time, but little novel information was added to the understanding of the subject.
Although Mackenzie regarded the condition to be of unknown etiology (1884), he carefully defined the chief symptom as excruciating burning or tearing pain—odynophagia—induced by any attempt to swallow or any movement of the laryngeal muscles (Fig. 9). He also noted that such patients developed considerable thirst but were unable to achieve relief by drinking because the pain was so severe. In adults he reported that there was a constant expectoration of frothy saliva and, although patients might not always have a fever, they often became delirious. Mackenzie believed the lesion to represent a diffuse catarrhal inflammation of the mucosa of the upper end of the gullet and felt the diagnosis was based on the history of extreme pain and the absence of pharyngeal inflammation on examination!
Although the description by Mackenzie of esophagitis is different to that now recognized to represent the contemporary understanding of the disease, his is the first lucid attempt to define the subject. His text, “Diseases of the Throat and Nose,” contains a fascinating and detailed analysis and classification of esophagitis but sometimes fails to accurately distinguish between disease of the lower and upper end of the gullet. In the 20th century the term esophagitis has become referable almost solely to an entity involving the lower part of the gullet. Nevertheless, Mackenzie recognized that the inflammatory conditions of the upper part of the gullet were specific and included diphtheria, thrush, tuberculosis, syphilis, actinomycosis, and corrosive damage. Some of the associated non-specific conditions (myalgia, general hyperaesthesia) alluded to by Mackenzie are no longer recognized but as late as 1900, all inflammatory conditions effecting the gullet were still regarded as varieties of “esophagitis.” Of particular interest were the treatments proposed in the management of this vexatious condition. It was considered mandatory that the organ be maintained in a state of absolute rest. Thus feeding was undertaken by nutrient enemata and morphia administered by hypodermic injection to facilitate resolution of inflammation and abolish pain. Poultices were applied along the upper part of the spine or if the pain was particular severe, anodyne embrocations such as oleate of morphia or belladonna lineaments were rubbed into the back. Mondière, following the French practice of the time believed in venesection and cupping and leech applications (12 to 30 at a time) (Fig. 10).
Counter irritation by the application of mustard poultices or moxas was also widely recommended. By 1884, Mackenzie was however able to declare that bleeding or even the local abstraction of blood was of little value and that counter-irritation was ineffective. He favored derivatives and especially recommended the use of extremely hot “pediluvia.” Bleuland used blisters “loco dolenti” between the shoulders with success and Pagenstecher reported the use of hydrochloride of ammonia to great advantage. This agent had long been a favorite of German and Dutch physicians and was used as a remedy for different diseases. Mackenzie was adamant that the passage of bougies was dangerous and should never be attempted because it was likely to cause rupture of the esophagus. Once convalescence commenced the patient could be changed from liquid to a solid diet gradually and if pain returned immediate return to a liquid diet should once again be undertaken.
Despite the attention to the gullet provided by Mackenzie esophagitis per se was not commonly recognized and little known about it. In 1906, Wilder Tileston carefully defined at least twelve different types of ulceration that included carcinoma, corrosive, foreign bodies, acute infectious diseases, decubitus, aneurysms, catarrhal inflammations, those associated with diverticulum, tuberculosis, syphilis, varicose, and ulcers caused by thrush. Tileston was particularly drawn to the disease that he referred to as “peptic ulcer of the esophagus” that he claimed exactly simulated the behavior and appearance of chronic gastric ulcer [
]. He noted that although a rare entity it had initially been described by Albers in 1839 and thereafter sporadically noted by pathologists. No less an authority than C. Rokitansky (1804–1878) had concurred that peptic ulcer of the lower esophagus was a real and definable entity and represented the aftermath of gastric juice in the gullet. Nevertheless when Tileston [
] reviewed the literature as far as 1906 there had only been 44 clear cut examples of the condition published. The ulcers described were usually single and often associated with chronic peptic ulcer in the stomach or the duodenum. They were large penetrating lesions sometimes 6 to 8 cm in length that lay “above the cardiac sphincter” and although often longitudinal might also encircle the entire gullet (Fig. 11).
The patients were elderly and many had no symptoms referable to the esophagus until their admission to the hospital. Death usually occurred from perforation into a large vessel, the pericardium, the mediastinum, or the pleural cavity, although some died of pneumonia. Of interest were that few exhibited symptoms of esophageal obstruction. Tileston further reported that the histology of such ulcers was identical to that of chronic gastric ulcer and that the adjacent mucosa was gastric type. He assumed it to be “ectopic” because it lined the lower part of the gut in the mediastinum.
The understanding of ulceration of the esophagus and esophagitis became further obfuscated with the advent of radiology and esophagoscopy at the turn of the 19th century. The use of these diagnostic tools enabled the identification of a variety of esophageal lesions particularly ulcers and inflammation at the lower end of the gullet not previously apparent to clinicians. Thus by the 1920s it was apparent that pathologists, clinicians, and endoscopists, while assuming that they were describing the same entity when they referred to peptic ulcer of the esophagus or esophagitis, were each unclear as to the exact nature of the disease process. In 1929 Chevalier Jackson, a virtuoso esophagoscopist, reported 88 cases in 4000 consecutive endoscopies (Fig. 12) [
], whereas Stewart and Hartfall in the same year claimed that they were able to identify only one example of esophagitis in 10,000 consecutive autopsies.
The subsequent reports of Lyall (1937), Chamberlin (1939), Dick and Hurst (1942) provided further insight into the nature and occurrence of these lower esophageal lesions. The further substantial contributions by Paul Allison and Johnstone of Leeds (1943, 1946, and 1948) left little doubt that the investigators were confident in their ability to recognize and diagnose “peptic ulcer of the esophagus” or “esophagitis.”
The beatitude of Barrett
Norman Barrett believed that the early descriptions of ulcer by Tileston, Stewart, and Lyall differed significantly from those identified by clinicians, such as Allison [
]. He felt that the former were rare and of little clinical significance whereas the latter were common and important entities (Fig. 13). He further proposed that this confusion had arisen because the pathology of the former had been tacked on to the symptomatology of the latter. Indeed it was Barrett's opinion that the condition described in 1948 by Allison actually represented “reflux esophagitis” and that this was the best name for the lesion.
The assessment of the situation at the lower end of the esophagus by Allison was prescient. He was of the opinion that chronic esophageal ulcers could develop into lesions similar to gastric ulcers and noted that the features of esophageal ulcers complicating reflux esophagitis are that they are situated in esophagus and represent digestion of the squamous epithelium. Allison opined that these lesions arose in areas of general acute inflammation and manifested initially as erosions that could either heal or persist depending on the amount of gastric juice that had access to the gullet. Such areas remained as superficial defects for a considerable period of time in the most individuals but eventually when the site had become significantly scarred they burrowed through the muscularis. It was Barrett's proposal that the stricture of the esophagus generated by this chronic inflammation had been mistakenly regarded as the esophagus whereas it was in fact a patch of stomach partially enveloped by peritoneum that had been drawn up by scar tissue into the mediastinum. Barrett therefore concluded that it was this stomach that was the site of the chronic gastric ulcer and that the ulcers that had been described in the lower gullet were in truth gastric and not esophageal and that they were located below the stricture [
To defend his novel and somewhat provocative assertion Barrett asked two questions. First, what is the esophagus? And second, what is meant by “heterotopia” of the gastric mucosa in relation to the esophagus? Barrett regarded the esophagus to be that part of the alimentary canal extending from the pharynx to the stomach lined by squamous epithelium. Regarding heterotopia and ectopia, he believed the terms to be interchangeable and to refer to small islets of gastric epithelium found in the esophagus and surrounded on all sides by normal squamous epithelium. He proposed that the alternative scenario in which gastric mucosa in direct continuity with the rest of the stomach, extended up into mediastinum in one sheet was caused by a congenital and pathological condition that was denoted by the term “short esophagus.” His views on the relationship of islets of gastric epithelium to chronic peptic ulcer of the esophagus were novel. In his opinion such ulcers arose in the esophagus in islets of ectopic gastric mucosa or as a consequence of the secretion of acid into the gullet by such islets. The history of these ectopic islets of gastric mucosa in esophagus is actually intriguing. Although first described by F.A. Schmidt in 1805, the observation was overlooked until Schridde in 1904 reported that microscopic ectopic islets were present in 70% of all gullets and always situated in the post-cricoid region. Taylor of Leeds in 1927 in examining 900 cases had identified in six, at the top of the gullet, areas large enough to be visible with the naked eye. Terracol in 1938 reported similar lesions at the top of the gullet whereas Rector and Connerly (1941) identified in infants fifty-six examples at the upper end and seven lower down. Barrett did not believe that these isolated islets secreted acid or if they did the volume was of such paucity that it could not cause ulcers. Indeed he claimed “nor were there any reports of ulcers associated with such islets.” His interpretation of the papers by Tileston, Stewart, and Lyall was that the massive ulceration described by them was in an area of stomach that extended beyond the crura of the diaphragm and that the ulcers were in actual fact chronic gastric ulcers occurring in a pouch of “mediastinal stomach.” In contradistinction, Lyall and his colleagues had regarded the ulcers to be in the esophagus and had called the columnar mucosa with which they were associated, “heterotopic” (Fig. 14).
The final conclusion that Barrett arrived at in his provocative but seminal comments on esophagitis was that the word had now become a blunderbuss term and was being used to cover many different pathological lesions. He believed that its usage should always be qualified by the descriptive adjective, “reflux esophagitis” and that this condition was common and could produce ulceration of the esophagus and stricture formation. He further believed that this particular lesion was separate to the condition regarded by pathologists as “peptic ulcer” of the esophagus that he felt to be an example of a congenital short esophagus. In the latter there were neither evidence of general inflammation, nor stricture formation but a part of the stomach extended up into the mediastinum and even into the neck and it was in this type of stomach that a typical chronic gastric ulcer could form [
Although there are many debates in regard to who first noted reflux esophagitis and what its precise etiology might be, the contributions of Winkelstein require consideration. At the 85th Annual Session of the American Medical Association held in Cleveland on 13 June 1934, Winkelstein of Mt. Sinai Hospital, New York, presented to the Section of Gastroenterology and Proctology a paper entitled “Peptic Esophagitis: A new Clinical Entity” [
]. He noted that although the causes of esophagitis were generally regarded as either irritative (alcohol, tobacco), specific (syphilis, tuberculosis) or secondary to complications of cardiospasm, diverticuli or neoplasm he believed that there existed a different and separate entity. “Recently I have observed some patients with a type of esophagitis that does not seem to fit into this classification. The features of these cases are so distinctive as to impress one with the probability that they form a separate clinical entity” (Fig. 15).
In the discussion of this pivotal presentation the doyen of American esophageal disease, the venerable Chevalier Jackson of Philadelphia rose to comment stating, “I am in hearty accord….Chronic esophagitis is a common disease due to a variety of causes not all of which are known or, at least proved.…The cause of this kind of esophagitis is similar to the causes of gastric ulcer….I think that the word peptic is justified by the fact that all of these ulcers in the esophagus, stomach and duodenum are within the area of the gastric juice, and for parallel reasons this form of esophagitis, which is also within the area flooded by the gastric juice. I think it is quite properly called peptic.”
In 1935, in the 16 March issue of the Journal of the American Medical Association, Winkelstein further noted, “one cannot avoid the suspicion that the disease in these five cases is possibly a peptic esophagitis ie, an esophagitis resulting from the irritant action on the mucosa of free hydrochloric acid and pepsin.” Indeed it was the culmination of Winkelstein's proposal and Allison's later concept of a chronic reflux of gastric contents [
] that finally assured the place of the term reflux esophagitis in the literature. The subsequent arguments as to whether anatomical and mechanical factors, such as a hiatus hernia were responsible engendered much discussion. In 1968, E.D. Palmer in a powerful paper cast considerable doubt on the relationship between hiatus hernia and esophagitis. A short time previously he similarly disparaged the proposed relationship between the spiral bacteria noted in gastric biopsies and washings and peptic ulcer disease! In a twenty-two year prospective study, he reported that many patients with hernia had neither reflux symptoms nor esophagitis and that many other patients had esophagitis in the absence of a hiatus hernia. At this stage the role of the lower esophageal sphincter became an area of critical relevance in establishing the primary mechanism of reflux disease. The further expansion of the gastroesophageal horizon of disease was proposed in 1962 when J.H. Kennedy implicated reflux symptoms in the production of pulmonary symptomatology and in 1968, Cherry and Margulies suggested that laryngeal abnormalities might be secondary to gastroesophageal reflux [
] as far back as 1580, historically, the hiatus hernia had hardly been described before the radiograph era, and reflected the classic autopsy technique of that time, in which the esophagus was cut just above the diaphragm and taken out with the heart and the lungs, losing all its connections with its abdominal segment and with the stomach. In 1889 [
], Postempski reported the successful repair of a diaphragmatic hernia by a transthoracic approach and within a year, six further cases had been reported in his clinic. The advent of radiological contrast studies of the gastro-intestinal tract led to a better appreciation of the prevalence of hiatus hernias such that R. Carman at the Mayo clinic in 1924 could document 18 cases within a one year period (Fig. 16).
Bernstein in reviewing the subject of hiatus hernia [
] in 1947 noted some of the causes for the low numbers of diagnoses of the condition. “At autopsy, with all muscles relaxed and the intra-abdominal pressure diminished, this condition may easily be overlooked, and only scattered reports of a few cases were therefore known in the literature of the pre-radiological era. The conventional technique of x-ray examination of the stomach with the patient in upright posture usually also fails to visualize these hernias. Examination in recumbent or even Trendelenburg position with application of manual pressure toward the upper abdomen is necessary to produce and demonstrate the condition under the fluoroscope. It is usual for these hernias to disappear as soon as the patient is brought back into upright posture or the increased abdominal pressure is released.”
The rarity of the condition may best be considered by the work of H. Eppinger who in 1911 summarized the literature on diaphragmatic hernia and reported that of 635 cases of herniation through various portions of the diaphragm only 11 involved the esophageal hiatus [
]. In 1923 Richards could review 23 observations. By 1925, 30 cases had been seen at the Mayo Clinic. It was therefore only when such a specially adapted technique of examination was introduced by Akerlund that the relative frequent occurrence of this condition was generally recognized [
]. Akerlund reported 30 cases in 1926 and proposed a classification of hiatus hernia that for many years was regarded as definitive. From then on the number soon became legion. Knothe observed 300 cases at the Charité Hospital in Berlin, representing eight per cent of all patients subjected to gastro-intestinal roentgen examination. In 1930 Ritvo, a Boston radiologist, published a series of 60 cases, all in adults and in commenting on the differential diagnosis, cited cardio-esophageal relaxation as a distinct pathological process [
]. Although Ritvo reported epigastric pain, heartburn, nausea, vomiting, and regurgitation as clinical correlates of “esophageal orifice hernia,” he failed to postulate gastroesophageal reflux as the cause of these symptoms. Harrington subsequently reported 680 cases of diaphragmatic hernia seen at the Mayo Clinic. From 1930–1946, 984 articles on diaphragmatic hernia were quoted in the Quarterly Cumulative Index Medicus. Bernstein himself noted that in the two years before his publication he had identified 38 cases of diaphragmatic hernia in 994 gastric examinations (3.8 per cent). From all these publications, containing reports of numerous cases each, and from his experience he stated, “hernia of the esophageal hiatus is a very common condition indeed.” He concluded that on the average the incidence in relation to the number of patients examined varies from 2% to 5%.
] and others in this century concentrated on anatomic problems and for practical purposes, surgical intervention was concerned principally with returning the stomach to the abdomen. In 1935 Winklestein [
] in 1947, commented on the clinical relevance of this consideration. It became recognized that hiatal hernia was a common condition and Harrington reported that over half of the patients undergoing laparotomy for other conditions could be shown to have an asymptomatic hiatal hernia. It was thus accepted that only at the time reflux esophagitis was present did hiatal hernias become symptomatic. In 1949, Kramer and Ingelfinger [
] and others made it possible to accurately determine the necessity for surgical correction and also enabled an objective evaluation of the completeness of the surgical repair.
The evolution of the surgical strategies
During the first half of the 20th century, surgeons were somewhat mechanistically inclined and were thus especially excited by the concept that an organ that protruded through a muscle, possessed a sac, and was called a hernia should indeed be treated like a hernia. Thus, if the patient was deemed capable of sustaining a major operation that consisted in an anatomical repositioning of the stomach then such was the resolution of the problem. If the operation was deemed risky, the alternative of ablating the phrenic nerve or snipping the hiatal rim to release the supposedly “throttled” stomach was considered acceptable. As usual there existed some discussion as to how large the hernia should be to qualify for fixing and, of course, whether the repair should be done through the chest or abdomen. Surgeons who had failed to comprehend the significance of gastroesophageal reflux and were uncertain of the precise relationship of hiatus hernia to esophagitis responded by proposing a wide variety of operations destined to restore anatomical “purity” to the region. Thus the subject of hiatal hernia was relegated to chapters dealing with inguinal, scrotal, and other celomic “ruptures.”
By 1950 there was no doubt that hiatus hernia had to be corrected surgically by placing the abdominal segment of the esophagus, the cardia, and the fundus into the abdomen below the diaphragm. In the 1940s, Pettersson [
] had tried to correct the hiatus hernia by paralyzing the left hemi-diaphragm, which he, accomplished by crushing the phrenic nerve. This idea had been proposed earlier by Harrington and Manseck. However, the results were unpredictable and the method was therefore abandoned.
The demise of this doctrine of anatomical preoccupation was ushered in by publication of Allison's classic article in 1948 [
] in which he proposed that the symptoms of an ordinary sliding hiatal hernia derived from acid damage inflicted on the lower esophagus secondary to reflux (Fig. 17). To remedy this situation Allison described a procedure whereby he united the two halves of the right crus and reattached the splayed-out phreno-esophageal membrane to the undersurface of the diaphragm. This operation was accepted as the logical treatment of hiatus hernia and it was believed that the anatomic repair would also correct the pathophysiology. Unfortunately this technique failed to prevent gastroesophageal reflux, and often even aggravated the situation. Indeed in 1973, Allison himself reviewed 421 of his own cases, operated between 1950 and 1970, and reported a recurrence of the hernia or gastroesophageal reflux in 49% of sliding hiatus hernia repairs. This experience was shared by numerous surgeons of the time including Sweet, Husfeldt, Pettersson, Wanke, Rehbein, and others. As a result in 1955 Boerema proposed the anterior gastropexy—fixation of the lesser curvature of the stomach to the anterior abdominal wall—as the sole necessary procedure in the surgical repair of hiatus hernia.
Unfortunately it now seems apparent that the latter component of the technique probably distracted and opened the dysfunctional sphincter even more. Because reposition of the hernia alone proved valueless in preventing gastroesophageal reflux, new techniques were developed to re-establish or accentuate the angle of His, thus forming an antireflux valve. Duhamel was among the first to use such a technique and in January 1953 published his first eight cases of reconstruction of the angle of His with six excellent and two good results. Thereafter Bettex proposed a broader back-to-back suturing of the fundus to the lower esophagus, and achieved better long-term results.
Although Allison's operation represented a novel idea and he was among the first to embrace the notion of the relationship to hernia and reflux the procedure failed to endure and enthusiasm for the “reflux thesis” might have waned considerably had it not been for the more or less simultaneous and independent achievements in the Kantonsspital, Basel, and the Frenchay Hospital in Bristol by Rudolf Nissen and Ronald Belsey, respectively.
Nissen and the art of fundoplication
The first rational surgical therapy for the condition now ubiquitously referred to under the “generic” acronym of GERD can be attributed to Rudolf Nissen and the technique of fundoplication (Fig. 18). In his original account of the procedure [
] Nissen, who was at that time chairman of the department of surgery at the Bürgerspital, Switzerland, treated two patients with chronic reflux disease. He termed the operation as a “gastroplication.” In the original description of the open fundoplication, two broad folds were wrapped around the esophagus, and fixed anterior to the esophagus utilizing four to five sutures. In the course of this operation major exposure of the esophago-gastric junction was required and an incision of the lesser omentum undertaken, leading to the transection of numerous branches of the vagal nerve, particularly the right trunk. This inadvertent denervation was responsible post operatively for considerable gastric motility related morbid symptomatology. As a result the original fundoplication was modified by Nissen and his colleagues and by others, to preserve the vagal branches. The “anterior wall technique” (Rossetti–Nissen total fundoplication) [
] uses the anterior wall of the gastric fundus to create the total wrap, thus obviating the extensive mobilization and division of the short gastric vessels. A further modification of the original fundoplication procedure, the “short floppy cuff” introduced in 1977 by Donahue and Bombeck [
]. This technical modification proposed a complete mobilization of the fundus and gastroesophageal junction, a short wrap length (2.0 to 2.5 cm) and preservation of the vagal nerves. In addition to the initial total fundoplication procedures used [
], a partial fundoplication may be performed whereby partial fundal wraps are placed either anterior or posterior to the gastroesophageal junction. A variety of similar procedures have been described and include a posterior crurally fixed partial fundoplication (Toupet), or an anterior 180° fundoplication (Dor). Watson's operation comprises a full mobilization of the lower esophagus and gastroesophageal junction, crural repair, fixation of the esophagus to the crura, and anterior 180° Dor-type fundoplication (the Belsey Mark IV [
] consists of an anterior 270° partial fundoplication, fixed to the undersurface of the diaphragm, and performed through the left chest—more on this later). In general it was believed (by surgeons) that such procedures had a good outcome and were associated with fewer side effects, such as dysphagia and gas bloat [
]. In particular it was apparent that the partial fundoplications were most useful in patients with evidence of significant esophageal hypomotility. Overall the variety of procedures, their high morbidity and the absence of rigorous evaluation of the outcome has, however, generated a sense of suboptimal enthusiasm for their usage.
Belsey of Bristol
Whereas it may appear that the discovery by Nissen of the beneficial effect of fundoplication [
] was a serendipitous observation it nevertheless remains a fact that in undertaking surgery that folded the stomach over an esophagogastric anastomosis to protect it from leaking, he inadvertently also protected the esophagus from the predictable acid reflux associated with resection of the lower esophageal sphincter. On the contrary the surgical strategy adopted by Belsey reflected the culmination of many years of observations in the Frenchay endoscopy unit and subsequent careful correlation of these findings with the symptomatology of the patient. Using a rigid 50-cm esophagoscope and examining the sedated but awake patient in the sitting position, Belsey concluded that competency at the cardia depended on its lying well below the diaphragm. Thus at the time the gastroesophageal junction was dislocated from its usual relationship with the right crus, the cardia gaped and gastric contents could be detected as rising into the esophagus with deep inspiration.
Belsey described this situation as “a patulous cardia” and proposed that the operative goal should be to fix the gastroesophageal junction 2 or 3 cm below the diaphragm (Fig. 19). A series of different surgical strategies were thereafter undertaken in an attempt to maximize the efficacy of this procedure. The Mark I operation was essentially the Allison approach, whereas the Mark II and III procedures represented various degrees of fundoplication. These culminated in the crescentic overlay of stomach in the Mark IV procedure that was achieved by trial and error as a means of obtaining tissue more substantial than the naked esophagus into which the anchoring sutures could be passed. Of interest is Belsey's criteria of a good operation included the ability to teach the procedure to others and achieve durable relief from reflux while preserving the other esophageal functions (ie, agreeable swallowing, venting of gas, protection of the airway, and maintenance of the capacity to vomit). The Mark III operation, with three rows of plicating sutures, failed both tests because it was difficult to teach and because the valve was more competent than was acceptable. During the early trial period (1949–1955) with the Mark I, II, and III operations, unsatisfactory results were identified in about a third of the patients; there were also seven post-operative deaths. In contrast, 85% of the 632 patients operated on between 1955 and 1962 had a good or excellent result. Belsey delayed formal publication of the Mark IV repair procedure for patients with the isolated condition of “patulous cardia” for six years before considering it to be of sufficient durability to publish the results [
]. Indeed the most salient lesson of this classic report was the remarkable restraint exhibited by Belsey in deferring publication until more than two decades had passed and more than 1,000 patients had been treated. Pearson, in a recent presidential address to the American Association of Thoracic Surgery [
], quoted Belsey as saying “The battlefields of surgery are strewn with the remains of promising new operations which perished in the follow-up clinic.” The lessons of the publication remain as valid now as they were almost fifty years ago when gastroesophageal reflux was firmly established as a correctable affliction if the appropriate patient was selected and an experienced surgeon identified. Of particular interest is the recognition that these results were obtained based on astute observations made decades before the availability of cineradiography, manometry, pH studies, and flexible endoscopy.
The introduction of laparoscopic antireflux surgery in 1991 by Geagea [
] in Belgium had a profound impact on the management strategy of patients with GERD. In much the same fashion as had followed the introduction of laparoscopic cholecystectomy, the novel procedure was widely embraced by surgeons anxious to use technical skills to resolve the vexatious problem of GERD. As a result of the perceived advantages of the procedure, surgeons began to undertake laparoscopic fundoplication often without adequate training and unaware of the appropriate indications. Although the advantages of minimally invasive surgery were evident, initial enthusiasm resulted in failure to initiate prospective randomized trials to critically evaluate the relative merits of the open procedure compared with long-term PPI usage. Nevertheless, early results seemed to confirm all the advantages of the minimally invasive approach (once the learning curve had been negotiated) with early outcomes not significantly different to those reported for the open procedure.
In general, the laparoscopic approach follows the same principles as the open Nissen's fundoplication (Fig. 20). A particular advantage is the excellent visibility of the gastro-esophageal junction at operation and the atraumatic nature of the procedure as compared with a major upper abdominal incision. A short (1 to 2 cm) wrap is advocated to avoid postoperative dysphagia. Other modifications that have been proposed include either division of the short gastric vessels to establish a tension-free fundoplication using a mobile posterior portion of the fundus or, alternatively, nondivision with the use of a more anterior portion of the fundus.
The pediatric scenario
Gastroesophageal reflux and peptic esophagitis were referred to in the medical literature as early as the 18th century by J.P. Frank; first case of esophagitis in a child was published in Paris in 1828 by Billard [
]. By 1931, the British otolarygologists Findlay and Kelly had described nine children with partial intrathoracic stomach, shortening of the esophagus, and stenosis just above the cardia in a paper entitled “Congenital Shortening of the Esophagus and the Thoracic Stomach Resulting Therefrom.” But the role of gastroesophageal reflux was overlooked, and cause and effect inverted, as may be judged by review of the title. Subsequently, between 1941 and 1943, Allison [
] in Paris, tried to synthesize all that was then known and for the first time considered hiatus hernia, esophageal relaxation or chalasia, gastroesophageal reflux, peptic esophagitis, short esophagus, and peptic esophageal stricture as different forms and grades of the same affliction. As early as 1962, in the United States Ravitch expressed a similar opinion in regard to the plurality of the disease process [
At about the same time a general feeling emerged that the entity of hiatus hernia represented a gradation of a single disease process and that minor forms could develop into major forms or in infants even disappear by the age of one to two years as reported by Carré [
]. Those physicians who were involved in pediatric care were of the opinion that all forms of hiatus hernia could lead to peptic esophagitis and even considered that this was more frequent with the minor forms rather than the major hernias. Without treatment, 20% of patients with hiatus hernia culminated in peptic stenosis, which was in the pediatric population a major incentive to propose surgical intervention. Thus by the early 1950s it was well accepted that the malformation of the lower esophagus and cardia, which is called hiatus hernia, led to gastroesophageal reflux and in 20% of cases, to esophageal stenosis. Thus pediatric surgeons were satisfied with the description of the pathologic entity and with its role in the pathogenesis of gastroesophageal reflux, of peptic esophagitis, and of peptic stenosis of the esophagus. Furthermore it seemed that surgical correction of this malformation had a sound scientific rationale (Fig. 21).
In the early 1960s the American literature contained few papers on hiatus hernia in children and in both the first edition (1962) and the second edition (1969) of Pediatric Surgery, the chapter on hiatus hernia was written by an Englishman, David Waterston. On the contrary the European pediatric literature commented in detail on what was considered to be a well-recognized disorder of children. Nevertheless most North American pediatricians and surgeons claimed the condition to be rare and a European phenomenon. They opined that few such children were seen in America, stating: “we never see such cases in America, probably because we feed our infants differently and allow them to lay on their bellies” [
]. Oddly, Randolph worked in Washington, DC, only a few miles away from the great medical centers in Baltimore and Philadelphia, where hiatus hernia had been previously claimed to be nonexistent! By the 1970s papers on gastro-esophageal reflux and hiatus hernia began to appear all over the United States and in 1979, Randolph in the third edition of Benson's Pediatric Surgery, and in 1980 Holder and Ashcraft [
]. Thus fundoplication, according to Nissen's and Belsey's Mark IV operation, became widely accepted as the best methods of treatment in the pediatric age group. This state of mind persisted until the late 1980s when the recognition that adequate acid suppression initially by the histamine receptor antagonist class of drugs was as if not more effective than surgery. The subsequent introduction of the proton pump inhibitor (PPI) class of drugs, particularly in a suspension that could be easily ingested by young children allowed therapy to move towards a pharmacological strategy in most patients.
Current management status
The introduction in the early 1990s of laparoscopic antireflux procedures prompted a wave of enthusiasm for reconsideration of the surgical option for the management of GERD. Indeed the advances that have been occurring in minimally invasive surgery have resulted in a renaissance of the fundoplication procedure initially described by Nissen in 1956 [
]. Initial enthusiastic reports have reactivated debate regarding the relative merits of medical versus surgical treatment. There is little doubt that laparoscopic fundoplication has been proven to be a safe procedure in experienced hands especially if rigorous and appropriate indications for surgery are used. Although prospective randomized studies of its efficacy are few and the follow up period available for evaluation is short the procedure has certainly become worthy of consideration in selected patients. There is however currently no general agreement on the indications for either open or laparoscopic antireflux surgery. This is a critical issue because a major variable in determining the outcome of surgery is the criterion used for patient selection. In certain countries it is possible that it may be a cost-effective alternative particularly when long-term medical therapy is contemplated.
The goals of GERD management are to prevent reflux and its symptomatic and pathological sequelae. In essence this requires the introduction of a therapeutic strategy that targets either the defective mechanisms, which facilitate reflux, or decreases the acid content and volume of the reflux itself (Fig. 22). Options available include either modulation of motility or sphincteric pressure or neutralization or inhibition of acid secretion. Unfortunately, while drugs can be delivered by way of existing anatomic routes, the surgical option includes all the sequelae of general anesthesia and invasion of the peritoneal cavity. Although the introduction of laparoscopic surgery engendered initial enthusiasm caused by its non-invasive nature early results are no better than noted with open surgery. Surgery in an albeit unpredictable, but definable fashion increases lower esophageal sphincter pressure, thus ameliorating reflux. Laparoscopic fundoplication seems to have some advantage as regards decreased hospital stay and a more rapid return to work supporting a notion that is somewhat cost-effective. Nevertheless, the wide variety of modifications of the surgical operations available to increase the pressure zone of the incompetent sphincter, suggest that none is perfect. Thus the recent advent of laparoscopic surgery should be regarded as not having essentially altered the principles of the procedure but simply facilitated access to the gastro-esophageal junction. Indeed individual surgeons performing what they believe to be the same operation often disagree on the precise details, and it has thus been difficult to compare surgical results between centers and between individuals. Generally, the more procedures the surgeon undertakes, the more likely is the patient to be cured of reflux because the learning curve for this operation is somewhat lengthy. Even in the best of hands the likelihood of complications varies between 10% to 15% and the mortality rate is 0.2% to 1%.
Clearly, an experienced surgeon and selection of the correct patient are the most important determinants of surgical success. Given the potential problems of surgery, a careful preoperative assessment of the patient is mandatory as is a joint discussion with a gastroenterologist to determine that an appropriate management strategy is in place. It is necessary to make a positive diagnosis of GERD before surgery, either by endoscopy or 24-hour pH measurement. Similarly delineation of peristaltic function and LES pressure by a manometry is important. The identification of individuals with poor motor function who will not benefit from surgery is critical to avoid iatrogenically induced accentuation of the GERD. A trial of a proton pump inhibitor may be a useful predictor of likely surgical success—if the patient's symptoms are markedly improved by this drug, then surgery will probably help. Apart from the myriad of complications associated with major surgery (bleeding, perforation of the esophagus, splenic injury, and sepsis), dysphagia and a disabling inability to belch may occur. After more than a decade of experience with PPIs no evidence exists to support any substantial adverse effects, thus most physicians have come to accept their overall primacy as the therapeutic option in the treatment of esophagitis. In the absence of any pharmacological prokinetic agent capable of specifically targeting the lower esophageal sphincter therapy, the treatment of esophagitis remains essentially symptomatic in the absence of a definable etiology.
The current vogue in the historical evolution of the management of the problem of reflux is represented by augmentation procedures for the lower esophageal sphincter. Rather than employ a transperitoneal approach, these are directed at the sphincter by the trans-esophageal route and include stitching, collagen injection and radio-frequency–induced fibrosis. It is however probable that these techniques will suffer all the drawbacks of any mechanical intervention but somewhat decrease the morbidity of open, albeit minimally invasive surgery. Similarly, a specific pharmacotherapeutic probe targeting the lower esophageal sphincter, while long fantasized, remains to be identified.
In all likelihood it will transpire that reflux disease represents an electroconductive disorder of the gastric pacemaker that results in an ectopic gastric electric rhythm and intermittent valvular dysfunction (pylorus and lower esophageal sphincter) coupled with abnormal peristaltic activity. The effects of acid and pepsin or bile are simply adjunctive to an as yet undefined gastric dysrhythmia and probably accentuated by as yet undetermined food preservative agents now a common part of the 21st century fascination with instant culinary products. Indeed as the delineation of abnormal cardiac electrophysiology led to the delineation of a previously obscure clinical problem (cardiac arrhythmia) it may be postulated that the delineation of the electrophysiology of the gastric cardia may similarly illuminate the definitive therapy of a disease that at present may be regarded as nothing less than God's Exquisite Revenge on Doctors! (GERD).