Recent changes in findings at oesophagogastroduodenoscopy (OGD): clinical implications and management
Recent changes in findings at oesophagogastroduodenoscopy (OGD): clinical implications and management
Introduction: Fibreoptic endoscopy revolutionised the practice of gastroenterology allowing direct visualisation of the gastrointestinal tract and mucosal biopsy to guide accurate diagnosis and treatment of gastrointestinal (GI) disorders. With so much emphasis over the years on service demands, less attention was paid to how much findings at OGD changed to where the clinical picture at OGD today is in many ways quite different to that of 40-50 years ago. Changed findings at OGD often equate with a change in diagnosis and treatment. The aim of this paper is to describe the clinical implications of recent changes in OGD findings and their management.
Peptic Ulcer Disease (PUD)
The groundbreaking discovery of H. pylori infection 40 years ago pinpointed a critical factor in the causation of peptic ulcer disease (PUD) and gastric neoplasia. Assessment of H. pylori infection status using gastric biopsies is now an integral part of the conduct of OGD and despite initial scepticism, eradication of H. pylori infection has become the treatment of choice for PUD. What was not appreciated at the time of its discovery was how much the prevalence of H. pylori infection was in decline and had been for several decades largely as a result of improvements in socioeconomic conditions in childhood, coupled with improved hygiene and sanitation. There is close correlation between the prevalence of H. pylori infection and the incidence of its related diseases and a fall in disease expression accompanied the falling infection rate. Consequently, the most obvious change in OGD findings over the past four decades is the dramatic decline in H. pylori -positive PUD and its replacement by gastro-oesophageal reflux disease (GORD). International studies show the fall in PUD is a worldwide phenomenon where the prevalence of PUD probably peaked in the early 1980’s and declined thereafter. The fall in PUD is more pronounced for duodenal ulcer compared with gastric ulcer, and most marked in younger age groups.
Idiopathic peptic ulcer disease (IPUD) is diagnosed when ulcers occur where H. pylori, NSAID use, and more unusual causes of PUD have been excluded, including non- H. pylori Helicobacter infection. As H. pylori-positive PUD decreased, IPUD increased. Risk factors for IPUD may be increasing age and multiple comorbidities. In clinical practice, idiopathic ulcers tend to be multiple and recurrent, seem more susceptible to complications, and may be more refractory to treatment.
In the absence of data from randomised controlled trials of treatment of IPUD, current advice is to continue full doses of proton pump inhibitors (PPIs) long-term as maintenance treatment.
Despite the striking decrease in uncomplicated H. pylori-positive PUD, hospitalisation for complicated PUD has not decreased. The problem of peptic ulcer bleeding continues, particularly in elderly females, where about 25% of ulcers present de novo with a complication without antecedent dyspepsia. Systematic review has shown H. pylori infection and NSAIDs are independent and synergistic risk factors for both uncomplicated and bleeding PUD. Clinical trials in high risk patients have shown H. pylori eradication alone may not be enough to eliminate the risk of bleeding and suggest co-prescription of PPIs should continue long-term. Current management of idiopathic bleeding ulcers is poorly defined leaving long-term high-dose PPI therapy as perhaps the best option.
Gastroesophageal Reflux Disease (GORD)
Several studies have documented the increase in the prevalence of GORD at OGD. For instance, a study from Rotherham, UK found a dramatic five-fold increase in the prevalence of GORD at OGD from 1977-2001 where it was described as a new phenomenon. Epidemiological studies, and studies in primary care, have also documented the opposing time trends of PUD and GORD, the significant increase in time in the prevalence of GORD symptoms, increased GORD-related hospitalisation over the past four decades, and the rise in GORD as a worldwide phenomenon.
Upper Gastrointestinal Cancer
The arrival of fibreoptic endoscopy came at a time of significant and opposing changes in the incidence of gastric and oesophageal cancer, similar to what was happening between PUD and GORD. Before 1940, gastric cancer was the leading cause of cancer death in American males but went into sharp decline over the time period of 1930 – 1978. As gastric cancer declined, oesophageal adenocarcinoma (OAC) showed a dramatic increase. Diverging time trends also appeared for gastric cardia cancer. H. pylori -related cardia cancer decreased whereas reflux-related cardia cancer increased and the curves for reflux-related cardia cancer and OAC closely mirrored each other. An important cross-sectional and longitudinal study using cancer registry data from 51 countries confirmed the presence of a worldwide negative association between the incidence and time trends of gastric cancer and OAC, and suggested H. pylori infection might be the key environmental factor exerting opposite effects on the incidence of the two cancers. Studies in H. pylori-positive and H. pylori-negative volunteers found less gastric acidity and decreased density of parietal and chief cells in the H. pylori-positive group which might in turn result in less noxious gastroesophageal refluxate theoretically protecting against OAC.
Barrett’s Oesophagus (BO)
Barrett’s oesophagus (BO) is defined as a condition in which metaplastic columnar epithelium that predisposes to cancer development replaces stratified squamous epithelium that normally lines the distal oesophagus. BO is an acquired condition occurring in response to gastroesophageal reflux and as the incidence of GORD has increased, so has the incidence of BO. The rise in BO has also paralleled the obesity epidemic, and abdominal obesity in particular is a strong risk factor for BO. There is also a strong inverse relationship between H. pylori infection and BO, especially Cag A-positive strains, and the apparent protective effect of H. pylori is probably mediated by decreasing GORD in infected patients.
The ability to easily visualise and biopsy the lower oesophagus at OGD raised the possibility that aggressive screening, surveillance and management of BO might make an impact on the rising threat of OAC and detect OAC at an early curable stage. Surveillance for BO is now firmly established as an integral part of the OGD workload. Patients with OAC discovered as a result of BO surveillance present with earlier stage disease, are less likely to have lymph node involvement, and have better short-term life expectancy than patients presenting with symptoms of OAC. Advances in therapeutic endoscopy over the past decade or so have made endoscopic mucosal resection the treatment of choice for high grade dysplasia or intramucosal cancer detected during BO surveillance, obviating the need for oesophagectomy with its attendant risks.
Following endoscopic resection, metachronous neoplasia can then be prevented by radiofrequency ablation of residual Barrett’s mucosa.
The endoscopist has an important role in optimising the value of BO surveillance (Table). High definition white light endoscopy with longer inspection times is associated with increased detection of high grade dysplasia and adenocarcinoma in BO. The optimal biopsy protocol includes sampling of any visible lesions together with systematic four-quadrant biopsies at every 2cm of visible Barrett’s mucosa, the Seattle protocol. Adherence to this protocol is not universal and, ironically, is considerably less in patients where adherence is needed most, those with long segments of BO. Nonadherence to the Seattle protocol has been shown to result in a lower rate of dysplasia detection, whereas adherence can detect dysplasia in 13 times more patients than a nonsystematic biopsy protocol.
Ironically, one of the disappointing aspects of endoscopy surveillance for BO has been its negligible impact on the rising tide of OAC. A number of factors might account for this apparent anomaly.
About 40% of patients with OAC report no history of chronic reflux symptoms and hence are not likely to ever undergo OGD screening or surveillance for BO. The annual cancer conversion rate in BO is only 0.5%, and the majority of patients with BO die of other causes, particularly cardiovascular disease. Also, fewer than 5% of patients with OAC are known to have had BO before they presented with symptoms. It also appears that approximately 50% of OAC has a distinct phenotype without evidence of associated intestinal metaplasia or visible BO segments. This non-BO cancer phenotype appears to be more aggressive and has a worse prognosis than the BO phenotype.
Eosinophilic Gastrointestinal Disorders (EGIDs)
Apart from significant changes in existing diseases, the era of fibreoptic endoscopy has also witnessed the emergence of eosinophilic GI disorders (EGIDs), apparently new clinical disorders. EGIDs are chronic inflammatory conditions characterised by GI symptoms and pathological eosinophilic infiltration of the GI mucosa where secondary causes of tissue eosinophilia have been excluded. Although still rare in routine clinical practice, there appears to be a genuine increase in all types of EGIDs including eosinophilic oesophagitis (EoE), eosinophilic gastritis (EG) and eosinophilic gastroenteritis (EGE). EoE is the most prevalent and well-studied EGID.
EoE
EoE is characterised clinically by symptoms of oesophageal dysfunction and histologically by eosinophil-predominant inflammation with an eosinophil count > 15 eosinophils/ high power field. Since its initial description in 1993, EoE has increased in incidence and prevalence and become a major cause of upper GI morbidity. The natural history of EoE is that of a chronic, perhaps lifelong, inflammatory disorder which if left undiagnosed and untreated leads to oesophageal wall remodelling, and possible stricture formation. There is no evidence that EoE is a premalignant condition.
H. pylori infection appears to have a protective effect against EoE.
The endoscopist has an important role in both diagnosis and management of EoE. First, EoE can only be diagnosed by endoscopy and oesophageal biopsies. EoE has some characteristic endoscopy findings including mucosal oedema, rings, exudates, furrows, and strictures and an endoscopic reference score (EREFS) can be calculated based on these findings. The EREFS score helps standardise recognition and reporting of the main endoscopic features, has excellent interobserver and intraobservor agreement, and is also responsive to treatment. EoE has a patchy distribution and increasing the number of biopsies increases diagnostic yield. Recent European guidelines suggest taking at least six oesophageal biopsies from at least two locations focusing on areas with mucosal abnormality. Second, symptoms or noninvasive biomarkers cannot be relied on to monitor EoE. Hence, whether treatment is with PPIs, swallowed topical steroids, dietary measures, or novel biological agents such as dupilumab, repeat OGD and biopsies are necessary to accurately monitor disease activity and document remission. Third, oesophageal stricture is a key complication of EoE and endoscopic dilatation is an important part of the treatment algorithm to help relieve dysphagia and prevent food impaction.
In contrast to benign peptic strictures, mucosal tears causing postprocedural retrosternal pain lasting several days are common after dilatation in EoE and patients should be advised about this possibility before the procedure. Careful stepwise dilatation should be performed by experienced endoscopists adhering to the “rule of 3” and published guidelines. Non-oesophageal EGIDs in the upper GI tract include EG and EGE and endoscopy findings in EG and EGE are normal in about two-thirds of cases but can include mucosal erythema, ulceration, nodularity, and friability. Hence the importance of taking multiple gastric and duodenal biopsies regardless of endoscopic appearances. In terms of prognosis, decreased tissue eosinophilia is associated with improvement in symptoms, and in endoscopic and histological findings. Hence, repeat OGD and biopsies to assess response to treatment remains an integral part of the care plan for EGID patients.
Gastric Polyps
A gastric polyp is an abnormal growth of tissue originating in the epithelium or submucosa and protruding into the stomach lumen. Over the last few decades, there have been dramatic changes in the spectrum of gastric polyps, both in terms of polyp type, frequency, and anatomical site. In the early years of OGD, hyperplastic polyps (HPs) and adenomas were the dominant polyp types and the emphasis in clinical practice was on assessment of malignant potential. If H. pylori is found in the surrounding gastric mucosa, HPs often regress after successful eradication therapy. The late 1990s saw a progressive decline in the prevalence of HPs coupled with a steep increase in the prevalence of sporadic fundic gland polyps (FGPs). That trend has continued to the present day. Long-term use of PPIs and the absence of H. pylori have emerged as important factors underlying what appears to be an emerging epidemic of sporadic FGPs. An important study from the Netherlands found 1. long-term PPI use was associated with an increased risk of FGPs while short-term (<1 yr) was not, 2. the PPI effect was a class effect, 3. FGPs could occur without PPI use, 4. none of the patients with FGPs had H. pylori or signs of gastritis, and 5. the risk of dysplasia in PPI-associated FGPs was negligible. The epidemic of sporadic FGPs appears to be a worldwide phenomenon and mirrors the exponential rise in the use of PPIs, both appropriate and inappropriate, evident over the past 30 years.
H. pylori
In the early years of H. pylori eradication, H. pylori-positive PUD was common, eradication and ulcer cure rates were high, and patient reported outcomes were good. With the fall off in PUD, H. pylori infection is now encountered more often in patients with either functional dyspepsia and normal endoscopy findings, or in the setting of GORD, situations where it is often unclear if eradication of infection might be of clinical value. Patients with functional dyspepsia who obtain sustained relief of symptoms after eradication therapy are now classed as having H. pylori-associated dyspepsia, and it is acknowledged that H. pylori eradication in GORD does not fundamentally change the chronic relapsing nature of the condition. Over the past 30 years, H. pylori itself has become more difficult to eradicate. Treatment regimens that were successful in the 1990s have been undermined by the steep rise of antimicrobial resistance and the efficacy of standard triple therapy in particular has declined as resistance to clarithromycin has increased.
In Europe, gastroenterologists in routine clinical practice often prescribe eradication therapy for only seven to ten days (69%), use low dose PPIs (48%), prescribe standard triple therapy where it is likely to be ineffective (46%), and seem slow to adapt to published recommendations. Selecting H. pylori therapy based on prior antibiotic exposure is good practice, yet in recent US surveys, only 38% of healthcare practitioners asked patients about prior antibiotic exposure, and only 58% checked for confirmation of cure after treatment.
Obtaining gastric biopsies specifically for H. pylori culture and sensitivity is rare in routine OGD practice, related to factors such as logistical challenges, lack of laboratory back-up, and cost. H. pylori is also a fastidious bacterium where culture is time-consuming and not always successful. Molecular antibiotic resistance testing using Next Generation Sequencing (NGS) methods offers a simpler, faster alternative to traditional culturebased testing and the presence of mutations detected by molecular tests correlates well with culturebased susceptibility tests. NGS methods simultaneously identify DNA mutations in the H. pylori genome responsible for resistance to all the antibiotics commonly used to treat the infection and could conceivably replace culture methods and at least encourage more susceptibility-based H. pylori treatment in clinical practice.
Coeliac Disease (CD)
Coeliac disease (CD) is an immune-mediated enteropathy triggered in genetically susceptible individuals by ingestion of gluten. The last 50 years or more have seen a true increase in the incidence and prevalence of CD. At the same time, the clinical picture of CD has changed.
Presentation with classical features such as diarrhoea and weight loss has decreased, and CD now increasingly presents with nonclassical features such as iron deficiency anaemia or mimicking irritable bowel syndrome.
Presentation to specialties other than gastroenterology and screendetected CD has increased. Age at presentation has increased while severity of histopathological damage has decreased. In a study that assessed the relationship between CD and H. pylori status, CD was diagnosed in 4.4.% of H. pylori-positive patients compared with 8.8% in H. pylorinegative patients and this strong inverse relationship remained after adjustment for age, gender, and socioeconomic status.
Irrespective of time trends in disease frequency and phenotype, small bowel biopsy using endoscopic forceps biopsy of the descending duodenum is the most reliable method of diagnosing villous atrophy. More recently, a single biopsy technique has been shown to improve the quality of duodenal biopsy specimens and ideally endoscopists should take only one biopsy specimen per pass of the forceps. Villous atrophy can be confined to the duodenal bulb and collection of biopsy specimens from D1 can increase the sensitivity of CD detection by 10%. Published guidelines therefore suggest 4 biopsy specimens should be taken from the descending duodenum and 1 or 2 biopsies from the duodenal bulb. A study from the US showed adherence to the proposed standard of submitting 4 or more specimens occurred in only 35% of OGDs, and it is not unusual for coeliac patients to have had a prior OGD where duodenal biopsies were not taken, and the opportunity missed to diagnose the condition. Duodenal biopsy was also more likely if the duodenal mucosa looked abnormal despite the fact that endoscopic markers of CD are most associated with severe pathological changes and lack diagnostic sensitivity. OGD also has an important role in monitoring CD. Similar to Inflammatory Bowel Disease, the goal of treatment in CD that ensures the best patient outcome is complete mucosal healing. Coeliac serology has excellent sensitivity and specificity in the diagnosis of CD but falls down when it comes to monitoring patients on gluten free diet. Likewise, symptoms in coeliac patients are poorly predictive of mucosal status on gluten-free diet and villous atrophy can occur in the absence of symptoms. Hence, follow up OGD and duodenal biopsy remains the optimal test to assess mucosal recovery on gluten-free diet.
Discussion
Substantial changes have occurred in OGD findings over the past 40 years or so and the speed of change suggests environmental rather than genetic factors are responsible. The changes are summarised in the Figure and suggest a transition has occurred from features suggesting a H. pylori-positive “phenotype” (DU and GU, distal gastric cancer, hyperplastic or adenomatous gastric polyps) to a H. pylorinegative “phenotype” (GORD, oesophageal and cardia cancer, IPUD, FGPs, EGIDs, increased CD). By reducing gastric acidity, H. pylori may have been protective against reflux disorders whereas the decline in H. pylori with resultant increased gastric acidity has contributed to the rise in reflux related disorders.
The obesity epidemic that engulfed the developed world and now extending to developing countries is another likely causative factor that increased as H. pylori decreased. Obesity, particularly central obesity, is closely linked with increasing rates of reflux oesophagitis, BO, OAC, and cardia cancer. Interestingly, possible links have also been suggested between the absence of H. pylori per se and obesity where the lack of H. pylori, especially during childhood, might enhance the risk of obesity later in life, possibly related to increased levels of the appetite-stimulant hormone ghrelin in the H. pylori-negative stomach.
The changed findings at OGD have clinical implications and pose some potential dilemmas for everyday clinical practice. PPIs are among the commonest prescribed medications worldwide and prescription of PPIs has increased about 350% in the past 30 years. The rise in PPI prescribing is sustained by the parallel epidemic of obesity and GORD and the fact that GORD itself is a chronic relapsing disorder where adequate relief of symptoms is often PPI-dependant. However, long-term PPI treatment drives the growth and proliferation of FGPs, now the commonest gastric polyp encountered at OGD. Although currently looked on as non-neoplastic and not warranting ongoing surveillance, we have to admit the long-term prognosis for FGPs is unknown. H. pylori infection is now less common, often diagnosed in settings where the clinical impact of eradication is uncertain, and increasing antibiotic resistance has made successful eradication more difficult to achieve. Current management of H. pylori has to address these new realities. Ideally, eradication therapy should be based on the results of antimicrobial sensitivity testing, not empirical, and new molecular testing might ultimately make that approach more attainable during a busy endoscopy list.
Apart from the discovery of H. pylori, one of the unexpected developments of the era of fibreoptic endoscopy has been the emergence of EGIDs, particularly EoE, as new disorders. The inflammatory process in EGIDs has many of the hallmarks of atopic or allergic disorders which have increased worldwide over the last few decades. CD is a lifelong autoimmune enteropathy which has also increased in incidence. EoE and CD both show an inverse relationship with H. pylori infection as if H. pylori in some way protects against these conditions. The debate goes on as to whether the apparent permissive effect on autoimmune and allergic diseases results from the absence of H. pylori per se or is part of the impact of the broader Hygiene Hypothesis. Hygiene Hypothesis suggests early life exposure to a diverse immune stimulating microbiome is fundamental to establishing a competent tolerant immune system whereas depletion of these stimuli in childhood fosters immune deviation and predisposes to the development of autoimmune or allergic diseases later in life in genetically susceptible individuals.
Recent changes in OGD findings also have implications for the conduct of OGD where accurate diagnosis now seems to depend less on appearances and more on obtaining comprehensive and adequate mucosal biopsies. In the oesophagus, non-adherence to validated biopsy protocols significantly reduces the chance of detecting dysplasia in BO and diagnosing EoE. In the stomach, optimal diagnosis of H. pylori infection requires mucosal biopsies from the gastric antrum and body and OGD performed at least two weeks and four weeks off PPI and antibiotic therapy, respectively. In the duodenum, CD is best diagnosed using good quality single biopsy specimens from the first and second parts of the duodenum. Similar biopsy considerations apply to monitoring response to treatment, particularly in EoE and CD, where symptoms and blood tests give inadequate guidance and documenting mucosal healing is increasingly important for long-term prognosis.
Conclusions
Findings at OGD have undergone a series of important changes over the past 40-50 years driven by environmental factors such as the falling prevalence of H. pylori infection, the obesity epidemic, and the huge rise in prescribing of PPI’s and antibiotics. There appears to be a transition from a H. pyloripositive “phenotype” 40 years ago to a H. pylori-negative “phenotype” today. This transition has important implications for the diagnosis of GI disorders, prescribing, and the conduct of OGD.
Sincere thanks to Dr. Gerard Giblin, Mater University Hospital Dublin, for technical help with the Figure and to Mrs. Debbie Burke for expertly typing the manuscript.
REFERENCES
1. O’Connor Humphrey J. Review: Forty years of Helicobacter pylori infection and changes in findings at esophagogastroduodenoscopy. Helicobacter 2023; 00: e13026. https://doi.org/10.1111/hel.13026
2. Malfertheiner P, et al. Management of Helicobacter pylori infection: the Maastricht 6th/Florence consensus report. Gut.2022 Aug 8: gutjnl- 2022- 327745
3. Wuesten BLAM, et al. Diagnosis and management of Barrett esophagus: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy 2023; 55:1124-1146
Read the full magazine: March HPN
Read our Latest News