Guidelines for the Diagnosis and Management of Gastroesophageal Reflux Disease

Gastroesophageal reflux disease (GERD) is arguably the most common disease encountered by the gastroenterologist. It is equally likely that the primary care providers will find that complaints related to reflux disease constitute a large proportion of their practice. The following guideline will provide an overview of GERD and its presentation, and recommendations for the approach to diagnosis and management of this common and important disease. The document will review the presentations of any risk factors for GERD, the diagnostic modalities and their recommendation for use and recommendations for medical, surgical and endoscopic management including comparative effectiveness of different treatments. Extraesophageal symptoms and complications will be addressed as will the evaluation and management of "refractory" GERD. The document will conclude with the potential risks and side effects of the main treatments for GERD and their implications for patient management. Each section of the document will present the key recommendations related to the section topic and a subsequent summary of the evidence supporting those recommendations. An overall summary of the key recommendations is presented in Table 1. A search of OVID Medline, Pubmed and ISI Web of Science was conducted for the years from 1960–2011 using the following major search terms and subheadings including "heartburn", "acid regurgitation", "GERD", "lifestyle interventions", "proton pump inhibitor (PPI)", "endoscopic surgery," "extraesophageal symptoms," "Nissen fundoplication," and "GERD complications." We used systematic reviews and meta-analyses for each topic when available followed by a review of clinical trials.Table 1: Summary and strength of recommendationsTable 1: (continued)The GRADE system was used to evaluate the strength of the recommendations and the overall level of evidence (1,2). The level of evidence could range from "high" (implying that further research was unlikely to change the authors' confidence in the estimate of the effect) to "moderate" (further research would be likely to have an impact on the confidence in the estimate of effect) or "low" (further research would be expected to have an important impact on the confidence in the estimate of the effect and would be likely to change the estimate). The strength of a recommendation was graded as "strong" when the desirable effects of an intervention clearly outweigh the undesirable effects and as "conditional" when there is uncertainty about the trade-offs. It is important to be aware that GERD is defined by consensus and as such is a disease comprising symptoms, end-organ effects and complications related to the reflux of gastric contents into the esophagus, oral cavity, and/or the lung. Taking into account the multiple consensus definitions previously published (3,4,5), the authors have used the following working definition to define the disease: GERD should be defined as symptoms or complications resulting from the reflux of gastric contents into the esophagus or beyond, into the oral cavity (including larynx) or lung. GERD can be further classified as the presence of symptoms without erosions on endoscopic examination (non-erosive disease or NERD) or GERD symptoms with erosions present (ERD). SYMPTOMS AND EPIDEMIOLOGY Epidemiologic estimates of the prevalence of GERD are based primarily on the typical symptoms of heartburn and regurgitation. A systematic review found the prevalence of GERD to be 10–20% of the Western world with a lower prevalence in Asia (6). Clinically troublesome heartburn is seen in about 6% of the population (7). Regurgitation was reported in 16% in the systematic review noted above. Chest pain may be a symptom of GERD, even the presenting symptom (2,3). Distinguishing cardiac from non-cardiac chest pain is required before considering GERD as a cause of chest pain. Although the symptom of dysphagia can be associated with uncomplicated GERD, its presence warrants investigation for a potential complication including an underlying motility disorder, stricture, ring, or malignancy (8). Chronic cough, asthma, chronic laryngitis, other airway symptoms and so-called extraesophageal symptoms are discussed in a subsequent section. Atypical symptoms including dyspepsia, epigastric pain, nausea, bloating, and belching may be indicative of GERD but overlap with other conditions. A systematic review found that ∼38% of the general population complained of dyspepsia. Dyspepsia was more frequent in GERD patients than those without. These patients were at risk for a new diagnosis of GERD. Epigastric pain, early satiety, belching and bloating were more likely to respond to a PPI therapy compared with nausea. Overall, these symptoms can be considered to be associated with GERD if they respond to a PPI trial (9). A recent systematic review on the burden of GERD on quality of life (QOL) included 19 studies. Patients with disruptive GERD (daily or >weekly symptoms) had an increase in time off work and decrease in work productivity. Low scores on sleep scales were seen compared with patients with less frequent symptoms. A decrease in physical functioning was also seen (10). Nocturnal GERD has a greater impact on QOL compared with daytime symptoms. Both nocturnal symptoms and sleep disturbances are critical to elucidate when evaluating the GERD patient (11). The balance of evidence suggests that symptom frequency does not change as we age, however, the intensity of symptoms may decrease after the age of 50 (12). Aging increases the prevalence of erosive esophagitis, Los Angeles (LA) grades C and D (13). Barrett's esophagus increases in prevalence after age 50, especially in Caucasian males (14). There are little data addressing the features of GERD in women distinct from men. Patients with erosive esophagitis are more likely to be men, and women are more likely to have NERD. Barrett's esophagus is more frequent in men compared with women (15). The gender ratio for esophageal adenocarcinoma is estimated to be 8:1 male to female (14). There is a definite relationship between GERD and obesity. Several meta-analysis suggest an association between body mass index (BMI), waist circumference, weight gain and the presence of symptoms and complications of GERD including ERD and Barrett's esophagus (16,17). The ProGERD study, likely the largest of its kind (>5,000 patients) used logistic regression analysis to identify several independent risk factors for ERD. The odds for higher degrees of ERD increased as BMI rose (18). It is of greatest concern that there has been a well-documented association between BMI and carcinoma of the esophagus and gastric cardia (19). Establishing the diagnosis of GERDFigure: No Caption available.The diagnosis of GERD is made using some combination of symptom presentation, objective testing with endoscopy, ambulatory reflux monitoring, and response to antisecretory therapy. (Table 2) The symptoms of heartburn and regurgitation are the most reliable for making a presumptive diagnosis based on history alone; however, these are not as sensitive as most believe. A systematic review of seven studies found the sensitivity of heartburn and regurgitation for the presence of erosive esophagitis to be 30–76% and the specificity from 62–96% (20). Empiric PPI therapy (a PPI trial) is a reasonable approach to confirm GERD when it is suspected in patients with typical symptoms. A response to therapy would ideally confirm the diagnosis; however, a well done meta-analysis suggested some limitations of this approach with a sensitivity of 78% and specificity of 54% (21). Therefore, empiric therapy (or a so called PPI trial) has some limitations.Table 2: Diagnostic testing for GERD and utility of testsNon-cardiac chest pain has often been associated with the presence of GERD, and can be the presenting symptom. A meta-analysis found a high probability that non-cardiac chest pain responds to aggressive acid suppression (22). This study supported earlier work suggesting the efficacy and cost effectiveness of a PPI trial (PPI twice daily in variable doses) in patients with chest pain in whom a cardiac cause had been excluded. However, a more recent systematic review suggested that the response of non-cardiac chest pain to a PPI trial was significantly higher than placebo in patients with objective evidence of GERD (ERD on endoscopy and/or abnormal pH monitoring) (23). The response to PPIs compared with placebo was almost non-existent in the absence of objective documentation of GERD. As such, a diagnostic evaluation with endoscopy and pH monitoring should be considered before a PPI trial (24). The presence of heartburn in conjunction with chest pain was not predictive of PPI response of the chest pain component. Dysphagia has historically been an alarm symptom or warning sign and an indication for early endoscopy to rule out a GERD complication. Respiratory symptoms have been associated with GERD, based on retrospective case–control studies. In addition, dental erosions, erosion of dental enamel, sinusitis, chronic laryngitis and voice disturbance have similarly been associated with GERD. These are discussed later in the article. Overall, heartburn and regurgitation remain reliable symptoms of GERD as does non-cardiac chest pain. Other symptoms, while associated with GERD, are not as reliable. The causal relationship between GERD and the so-called atypical and extraesophageal manifestations remains difficult with only a history. Barium radiographs have been historically considered part of the potential diagnostic armamentarium in the patient with esophageal symptoms, including GERD. Although well-performed barium esophagrams with double contrast can detect signs of esophagitis, the overall sensitivity of this test is extremely low (25). The finding of barium reflux above the thoracic inlet with or without provocative maneuvers including the water siphon test does increase the sensitivity of the barium test; however, not sufficiently to be recommended as a diagnostic test without dysphagia (26). The endoscope has long been the primary tool used to evaluate the esophageal mucosa in patients with symptoms suspected due to GERD. Findings of GERD include erosive esophagitis, strictures, and a columnar lined esophagus ultimately confirmed to be Barrett's esophagus. As such, endoscopy has excellent specificity for the diagnosis of GERD especially when erosive esophagitis is seen and the LA classification is used (27). However, the vast majority of patients with heartburn and regurgitation will not have erosions (or Barrett's) limiting upper endoscopy as an initial diagnostic test in patients with suspected GERD (28). Endoscopy allows for biopsy of rings and strictures and screening for Barrett's. Although epidemiologic risk factors for Barrett's esophagus have been well-defined (age over 50, symptoms for>5–10 years, obesity, male sex) the sensitivity and specificity of these symptoms for abnormal endoscopy makes the utility of screening for Barrett's a controversial topic. Recent data indicate that it may be reasonable to perform endoscopy for screening in certain high-risk groups in particular overweight white males over the age of 50 with chronic GERD symptoms (12). The finding of any Barrett's esophagus segment has been associated with pathologic GERD and generally obviates the need for pH testing (29). In a 2009 study, 90% of short-segment BE patients were found to have abnormal pH-impedance testing (30). The addition of esophageal biopsies as an adjunct to an endoscopic examination has been re-emphasized because of the increased prevalence of eosinophilic esophagitis (EoE). Many clinicians routinely biopsy the esophagus in patients with reflux-type symptoms to look for EoE in the setting of an endoscopy that does not reveal erosive changes. Unfortunately, differentiating GERD from EoE using only biopsy is difficult and risks making a diagnosis and instituting treatment without supportive data. Low eosinophil counts in the distal esophagus while suggestive of GERD are not specific. In addition, a high eosinophil count may be seen with GERD and respond to PPIs (PPI responsive eosinophilia) (31). The sensitivity of the other histologic findings; basal cell hyperplasia, elongation of the rete pegs, papillary elongation, and even neutrophils, are of limited clinical usefulness (32,33). There are no studies examining the efficacy of PPIs based on microscopic findings alone. The use of routine biopsy of the esophagus to diagnose GERD cannot be recommended in a patient with heartburn and a normal endoscopy based on current literature. In addition, the practice of obtaining mucosal biopsies from a normal appearing esophagogastric junction has not been demonstrated to be useful in GERD patients (34). Esophageal manometry is of limited value in the primary diagnosis of GERD. Neither a decreased lower esophageal sphincter pressure, nor the presence of a motility abnormality is specific enough to make a diagnosis of GERD. Manometry should be used to aid in placement of transnasal pH-impedance probes and is recommended before consideration of antireflux surgery primarily to rule out achalasia or severe hypomotility (scleroderma-like esophagus), conditions that would be contraindications to Nissen fundoplication, but not to tailor the operation. Ambulatory reflux monitoring (pH or impedance-pH) is the only test that allows for determining the presence of abnormal esophageal acid exposure, reflux frequency, and symptom association with reflux episodes. Performed with either a telemetry capsule (usually 48 h) or transnasal catheter (24 h), pH monitoring has excellent sensitivity (77–100%) and specificity (85–100%) in patients with erosive esophagitis; however, the sensitivity is lower in those with endoscopy-negative reflux symptoms (<71%) when a diagnostic test is more likely to be needed (24). A consensus statement (35) suggested that impedance added to pH monitoring increased the sensitivity of reflux monitoring to close to 90%. Telemetry capsule pH monitoring offers increased patient tolerability and the option to extend the monitoring period to 48 or perhaps to 96 h. The additional monitoring period allows for combining and on and off therapy study in selected situations and offers additional opportunity to correlate symptoms with acid reflux. Catheter-based monitoring allows for the addition of impedance and detection of weakly acidic or non-acid reflux. Optimal use of these two options is certainly debated as is whether to test on or off therapy. As a true diagnostic test (is abnormal acid exposure present) and for evaluation before considering surgery in a patient with NERD an off therapy test is recommended. The use of on and off therapy monitoring in refractory GERD is discussed subsequently. When symptom correlation is required, the decision is more difficult. The two symptom association measures most often used are symptom index (SI) and symptom association probability (SAP). Both have methodological shortcomings that have been reviewed elsewhere (36) and prospective data to validate the ability of these symptom association measures to predict response to treatment is scarce. Both the SI and SAP have been validated when pH monitoring is performed off therapy in a patient with heartburn. A positive test on therapy, coupled with a symptom relationship, theoretically suggests GERD as a cause for symptoms but outcome studies are lacking for any symptom other than heartburn. For patient management, a strongly positive SI or SAP may suggest the need for a therapeutic intervention and a negative result supports the notion that the patient's symptoms are unlikely to be due to reflux. However, these indices should not be used in isolation and other reflux monitoring parameters as well the patient's presentation have to be taken into account. The relationship between H. pylori infection and GERD is controversial. As such, a full discussion is beyond the scope of this article. One issue most often discussed is whether treatment of H. pylori should be altered because of an exacerbation of GERD and if patients on long-term PPIs require screening and subsequent eradication of the bug to prevent the possibility of increasing risk of gastric cancer. A meta-analysis of 12 studies found no increase in GERD (erosive esophagitis) in patients with dyspeptic symptoms who were eradicated compared with those not. This same study found, in subgroup analysis, patients with peptic ulcer disease might experience the new onset of GERD symptoms after H. pylori eradication (37). Concern for the use of long-term PPI therapy in patients with H. pylori infection has been raised because of the potential for development of atrophic gastritis in infected patients on long-term PPI (38). This study prompted a Food and Drug Administration (FDA) review panel that concluded that the evidence was not sufficient to recommend testing of all patients on long-term PPI. The flaws in this study and lack of observational data on negative outcomes lead us to recommend against screening of GERD patients for H. pylori despite the European recommendation in favor of screening (39). GERD is frequent during pregnancy, manifests as heartburn, and may begin in any trimester. One study found onset of 52% in the first trimester, 40% in the second trimester, and 8% in the third trimester (40). Among 607 pregnant women attending an antenatal clinic, 22% experienced heartburn in the first trimester, 39% in the second, and 72% in the third, whereas only 14% of these women reported mild heartburn before their pregnancy (41). Severity also increased throughout pregnancy. Significant predictors of heartburn are increasing gestational age, heartburn before pregnancy, and parity. Maternal age is inversely correlated with heartburn. Race, pre-pregnancy BMI, and weight gain in pregnancy do not correlate with the onset of heartburn. Despite its frequent occurrence during pregnancy, heartburn usually resolves after delivery (42). Pregnancy and amount of weight gain during pregnancy were risk factors for frequent GERD symptoms 1 year post delivery (43). No other GERD symptom has been studied in pregnancy. The diagnosis of GERD during pregnancy should be based on symptoms and treatment symptom-based. Additional diagnostic testing is generally not required for the majority of patients with suspected GERD. In the occasional pregnant patient who does require testing, upper endoscopy is the test of choice, but should be reserved for patients whose symptoms are refractory to medical therapy or who have suspected complications. If possible however, endoscopy should be delayed until after the first trimester. It is uncommon to require ambulatory pH monitoring during pregnancy. Management of GERDFigure: No Caption available.SUMMARY OF THE EVIDENCE Lifestyle interventions are part of therapy for GERD. (Table 3) Counseling is often provided regarding weight loss, head of bed elevation, tobacco and alcohol cessation, avoidance of late-night meals, and cessation of foods that can potentially aggravate reflux symptoms including caffeine, coffee, chocolate, spicy foods, highly acidic foods such as oranges and tomatoes, and foods with high fat content.Table 3: Efficacy of lifestyle interventions for GERDA systematic review (44) evaluated the effect of dietary and other lifestyle modifications on lower esophageal sphincter pressure, esophageal pH, and GERD symptoms. Consumption of tobacco (12 trials), chocolate (2 trials), and carbonated beverages (2 trials) and right lateral decubitus position (3 trials) were shown to lower pressure of the lower esophageal sphincter (LES), whereas consumption of alcohol (16 trials), coffee and caffeine (14 trials), spicy foods (2 trials), citrus (3 trials), and fatty foods trials) had no There was an increase in esophageal acid exposure with tobacco and alcohol consumption in addition to of chocolate and fatty However, tobacco and alcohol cessation trials) were not shown to esophageal pH, or GERD symptoms. In addition, there have been no studies conducted to that have shown clinical in GERD symptoms or complications associated with cessation of coffee, caffeine, chocolate, spicy foods, carbonated fatty foods, or A recent systematic review concluded that there was lack of evidence that consumption of carbonated beverages or GERD gain even in with a normal BMI has been associated with new onset of GERD symptoms studies have demonstrated in GERD symptoms with weight gastric but not has been demonstrated to be in of GERD symptoms A large case–control study based on the demonstrated a 40% in frequent GERD symptoms for women who their BMI by or more compared with of the position has been associated with of esophageal pH and GERD symptoms. have demonstrated in GERD symptoms and esophageal pH with head of bed using or options for patients lifestyle interventions include or PPI therapy. A meta-analysis published in demonstrated that the placebo response in GERD clinical and was lower in patients with erosive esophagitis and PPI compared with with PPI therapy has been associated with and decreased compared with and placebo for patients with erosive esophagitis A meta-analysis demonstrated for all grades of erosive esophagitis using PPI therapy compared with or placebo The overall proportion of or treatment was with PPIs or placebo PPIs a significantly and placebo PPIs provided more heartburn PPIs are associated with a greater of symptom in patients with ERD compared to patients with NERD the symptom For patients with reflux a systematic review demonstrated for PPI therapy compared with and for heartburn the of with over the risk for heartburn primary efficacy in for PPI was confidence for and for In a PPIs were more than and There are seven available PPIs including that can be and are available only by and to in efficacy for symptom between PPIs A meta-analysis published in examining efficacy of PPI therapy for of erosive esophagitis included studies patients) for and there was a increase in the probability of of erosive esophagitis with an risk of and needed to of The by LA of erosive esophagitis were 50, and an 8% increase in the probability of GERD symptom at The clinical of this is of the PPIs with the of and should be before to an has been demonstrated to more nocturnal gastric pH in the first of sleep compared with other PPIs when each is at this effect to any clinical outcomes including symptom further is a delayed PPI in of compared only with demonstrated in esophageal pH in and the of to the any time of the of to in of erosive esophagitis was demonstrated in with in study As it would be expected that of patients with ERD would on PPI therapy and of patients with NERD. of GERD symptoms after a of PPI therapy has been found in of patients and does not in patients PPI or twice The evaluation and management of patients with response are discussed in the refractory GERD section. factors for lack of symptom have included patients with of presence of extraesophageal symptoms, and lack of PPIs are most in pH when taken before a and are generally less when taken at The to this rule to be for the of to have efficacy in pH of and can pH when at is common in practice Although PPI is common in clinical there is limited data to this practice. from trial demonstrated that in GERD patients refractory to patients to therapy daily was as as increasing to twice daily There is no data to PPIs more than in or PPI therapy should be for GERD patients who to have symptoms after PPI is and in patients with complications including erosive esophagitis and Barrett's esophagus. In patients found to have of the patients will off of PPIs over time For patients found to have LA esophagitis, will by In patients found to have any of retrospective studies have suggested a decreased risk for in patients PPI the other studies have demonstrated that patients with NERD and GERD can be with or PPI therapy. In a trial published in of NERD patients to of on were in at compared with of patients on In a systematic review of PPI studies were included in NERD with NERD and mild esophagitis, and studies with The for patients in the were to for patients on PPI therapy and to placebo in patients with but not for patients with ERD. therapy to is option for NERD patients options for GERD patients with response to PPI therapy are The addition of has been recommended for patients with symptoms refractory to PPI. This approach after multiple pH studies demonstrated pH One study suggested potential of pH after a of therapy In of this study and a lack of prospective clinical trial use of a might be most if on as needed in patients with symptoms and patients with objective evidence on pH monitoring of esophageal acid reflux despite PPI therapy with in addition to PPI therapy is option often considered for these has been shown to increase esophageal and gastric data additional of to PPI therapy has not been therapy of with has not been shown to be more compared with or therapy The of has been limited by system side effects including and in of patients in the absence of there is no for in GERD. For the of patients who may from a option is a