|Home | About | Journals | Submit | Contact Us | Français|
Dyspepsia is a common term used for a heterogeneous group of abdominal symptoms. Functional dyspepsia (FD) is the focus of this review. The 2006 Rome III criteria defined FD and its subgroups, postprandial distress syndrome (PDS) and epigastric pain syndrome (EPS). FD is a very common condition with a high prevalence throughout the world, adversely affecting the quality of life of patients. The pathophysiology of FD has been under investigation during the past two decades. Multiple mechanisms such as abnormal gastric emptying, visceral hypersensitivity, impaired gastric accommodation, and central nervous system factors are likely involved. Several tests are available for the assessment of various physiologic functions possibly involved in the pathogenesis of FD, and some of these could be used in clinical practice, helping to understand the abnormalities underlining patients’ complaints. Currently, the possibilities of pharmacological therapy for FD are still limited, however, experience of using prokinetics, tricyclic antidepressants, selective serotonin-reuptake inhibitors (SSRIs), proton-pump inhibitors (PPIs), and several alternative techniques has been accumulated. The different combinations of alterations in physiologic gastrointestinal and central nervous system functions result in the very heterogeneous nature of FD so combined approaches to these patients could be beneficial in challenging cases.
Dyspepsia is a common term used to characterize abdominal pain centered in the epigastrium, sometimes combined with other gastrointestinal complaints. Historically the word ‘dyspepsia’ was used for a heterogeneous group of abdominal symptoms. Functional (previously nonulcer) dyspepsia (FD) is the focus of this review, and usually indicates abdominal discomfort or pain with no obvious organic cause that could be identified by endoscopy.
Stomach pain and discomfort have been reported since ancient times. The term ‘dyspepsia’ originates from the Greek ‘δυς-’ (dys-) and ‘πέψη’ (pepse), popularly known as indigestion. It was first recorded in the mid 18th century and since then it has been widely used [Baron et al. 2006]. In the 18th century dyspepsia was thought to be one of the ‘nervous disorders’ along with hypochondria and hysteria [Hare, 1991].
In addition to the term ‘functional dyspepsia’, several other descriptions of dyspepsia are in use, each of which reflects various amounts of investigation into upper gastrointestinal symptoms of the patient. Uninvestigated dyspepsia refers to patients with either new or possibly recurrent dyspeptic symptoms in whom no investigations have previously been undertaken. After those investigations dyspeptic complaints may be called investigated dyspepsia and should be differentiated into organic dyspepsia and FD. Organic dyspepsia means that there is a clear anatomic or pathophysiologic reason for the dyspeptic complaints, such as an ulcer disease or mass. In contrast, when a diagnosis of FD has been made, it means that a number of investigations were performed including upper gastrointestinal endoscopy, and were found to be normal [Jones, 2002] (Figure 1).
In 1994, the Rome criteria were developed in an attempt to meet the clinical need to describe systematically functional gastrointestinal disorders. The recommended definition for FD was a symptom or set of symptoms that are considered by most physicians to originate from the gastroduodenal region. Specific symptoms could include epigastric pain, epigastric burning, postprandial fullness, early satiation, bloating in the upper abdomen, nausea, and vomiting. The Rome criteria were subsequently modified in 2000 and 2006. The Rome I and II criteria did not account for meal-related symptoms and this was the fundamental change in Rome III criteria [Talley et al. 2008a, 1999]. According to the most recent 2006 Rome III criteria FD must include one or more of following symptoms: bothersome postprandial fullness, early satiation, epigastric pain, epigastric burning with no evidence of structural disease, including the use of upper endoscopy, which is likely to explain the symptoms. Criteria should be fulfilled for at least 3 months with symptom onset at least 6 months previously (Table 1).
FD includes multiple types of patients with heterogeneous complaints and possibly different underlining pathophysiology. There is accumulating evidence that distinct subgroups of uninvestigated dyspepsia exist in the general population, suggesting that separate evaluation and treatment strategies might be needed [Aro et al. 2009; Choung et al. 2007]. However, it is difficult to identify such groups because of significant overlap. Rome III introduced epigastric pain syndrome (EPS) and postprandial distress syndrome (PDS) subgroups to distinguish between different symptom clusters in FD. EPS is intermittent pain or a burning sensation localized in the epigastrium of at least moderate severity, at least once per week, not relieved by defecation or passage of flatus and not fulfilling criteria for gallbladder or sphincter of Oddi disorders. PDS is postprandial fullness after ordinary sized meals and early satiation preventing finishing a regular meal at least several times per week (Tables 2–4).
Older terms that represent FD are ‘nonulcer’, ‘idiopathic’ or ‘essential dyspepsia’. The term nonulcer dyspepsia is still popular but no longer recommended because it implies that the patient has symptoms similar to peptic ulcer disease, without having an actual ulcer on endoscopy examination. However, the spectrum of symptoms is broad in FD and is not defined by ulcer-like pain only.
FD is a very common condition with a high prevalence throughout the world. It affected up to 29% of the population in a study of employees in the USA [Shaib and El-Serag, 2004]. In a large population-based study from Europe 20.6% of subjects had experienced epigastric symptoms during the preceding 12-month period [Piessevaux et al. 2009]. In the Korean population the prevalence of uninvestigated dyspepsia was 11.7% [Jeong et al. 2008]. As in other functional disorders, dyspeptic symptoms are more frequent in women 24.4% versus 16.6% [Piessevaux et al. 2009; Welen et al. 2008].
Interestingly, in one study the prevalence of dyspeptic symptoms was found to increase significantly with age (7.7% at age 15–17 years; 17.6% at 18–24 years, 18.3% at 25–34 years, 19.7% at 35–44 years, 22.8% at 45–54 years, 23.7% at 55–64 years, and 24.4% above 65 years; q=0.084; p<0.0005). Other demographic variables, such as social level or degree of urbanization were not found to influence the prevalence of dyspeptic symptoms [Piessevaux et al. 2009].
Patients with dyspepsia have reduced health-related quality of life because their symptoms, particularly abdominal pain and indigestion, cause emotional distress, problems with food and drink, and impaired vitality [Halling et al. 2008], and it may be more impaired in patients with FD than in patients with other conditions who present for upper endoscopy [Welen et al. 2008; Talley et al. 1995]. The impact of disease on quality of life is comparable in patients with FD and chronic ‘organic’ diseases such as asthma and inflammatory bowel disease [Sundberg et al. 2009; Lix et al. 2008].
FD produces heavy economic burdens by demanding extensive medical care and diagnostic procedures and also by causing an inability to work or participate in educational activities. Among people with dyspeptic symptoms, 61.4% reported having consulted a physician. Of these, 74.1% reported several visits to a physician.
A large proportion of dyspeptic subjects underwent diagnostic testing: 36.5% reported undergoing upper gastrointestinal endoscopy and 12.4% had to stay away from work or educational activities as a result of their dyspeptic symptoms during the year [Piessevaux et al. 2009].
Cost data on FD remain limited, but a Swedish study estimated that the total costs of dyspepsia, peptic ulcer, and gastroesophageal reflux disease in 1997 was US$424million, or US$63 per adult in Sweden [Agreus and Borgquist, 2002]. The AGA burden of illness study suggested that the total direct costs of irritable bowel syndrome (IBS) (in 2000) were US$1.7 billion, while the indirect costs were US$20.2 billion [Sandler et al. 2002], and the impact of FD might be comparable.
There is a large overlap between gastroesophageal reflux disease (GERD), IBS, and FD, and often they coexist in the same patient. For practical proposes it is important to understand what complaints and symptoms are dominant.
In the past, heartburn itself was sufficient to define dyspepsia. Only recently in Rome III criteria was heartburn removed from the symptoms that refer to FD [Talley et al. 2008a]. Patients with the main complaint of typical heartburn most probably have GERD as a primary diagnosis and should be distinguished from patients with FD. In the study of Piessevaux et al.  33.8% of people with FD also had GERD-like symptoms. The overlap is greatest between nonerosive and functional heartburn and FD [Keohane and Quigley, 2007]. The clinical spectrum that extends from FD to nonerosive reflux disease (NERD) may be present in up to 70% of GERD disease in the community. It is very difficult to estimate the true prevalence of FD-NERD overlap due to a lack of uniformity in the definitions [Quigley and Keohane, 2008]. The people who have both FD and GERD may likely benefit more from proton pump inhibitor (PPI) treatment [Talley and Vakil, 2005].
The overlap with IBS is also common and even more difficult to differentiate since abdominal pain is the main symptom in both. Constipation delays gastric emptying and is commonly accompanied by upper gastrointestinal symptoms; on the other hand, bowel symptoms are common in FD. In one study 46% of FD patients had FD and IBS, and those patients were more likely to be female [Corsetti et al. 2004]. In a longitudinal study with over 12 years follow up, 40% of subjects with FD or IBS had switched symptomatology [Halder et al. 2007]. In a recent meta-analysis by Ford et al. , the prevalence of IBS in subjects with dyspepsia was 37% (95% confidence interval (CI), 30–45%) compared with 7% (95% CI, 5–10%) in subjects without dyspepsia. The pooled odds ratio (OR) for IBS in subjects with dyspepsia was 8 (95% CI, 5.74–11.16). The degree of overlap between the two conditions varied from 15 to 42%, depending on diagnostic criteria used for each [Ford et al. 2009]. This association might indicate common pathophysiologic mechanisms.
Distinguishing between gastroparesis and FD is often almost impossible because gastroparesis is one of the pathophysiologic mechanisms causing dyspepsia and often has the same symptoms. It is not clear whether those conditions have causal relationships or coexisting phenomena, and the treatment probably should target both. Unfortunately, some medications for FD may worsen gastroparesis and vice versa.
Obesity was associated with increased risk of gastrointestinal symptoms in a community-based, case–control study in Olmsted County, MN, USA, while increasing body mass index was associated with increased upper gastrointestinal symptoms, bloating, and diarrhea [Delgado-Aros et al. 2004b]. Other studies showed that weight increase (>10lb (0.45kg)) over time was not associated with the reporting of new symptoms of painful dyspepsia, or with their disappearance, but there was an association with dyspepsia and dysmotility features [Cremonini et al. 2006].
The relationship between weight loss and dyspepsia has been studied and dyspepsia symptoms were found to be associated with weight loss. Nausea, vomiting, and meal-related complaints such as postprandial fullness were most strongly associated with weight loss. Clusters of patients, based on symptoms, were strongly differentiated in terms of weight loss. Patients with vomiting, and early satiety/postprandial fullness reported 25–30% weight loss prevalence over the previous 3 months compared with around 10% prevalence in clusters characterized by low dyspepsia symptom burden [Jones et al. 2008].
An intriguing issue is the prevalence of psychiatric comorbidity in functional gastrointestinal disorders (FGID) and in the subgroup of FD. Aro et al.  in a recently reported Swedish population-based study found that anxiety but not depression is linked to uninvestigated dyspepsia and FD. Interestingly, in this study only PDS and not EPS was associated with anxiety [Aro et al. 2009]. Other studies, conducted in different population cohorts showed an association with psychological distress, frequently seeking health care [Koloski et al. 2003], a family history of dyspepsia and ulcer disease, and the use of tranquilizers [Bernersen et al. 1996].
Other interesting observations have been made about psychiatric comorbidities. Van Oudenhove et al.  found symptom severity and weight loss in FD are determined by psychosocial factors (e.g. depression, abuse history) and somatization, and to a lesser extent by gastric sensorimotor function. Geeraerts et al.  noted that a history of abuse was associated with alterations in gastric sensorimotor function in FD. FD patients reporting childhood sexual abuse had significantly lower discomfort thresholds during gastric distension. Sexual abuse particularly, rather than physical abuse, may influence gastric sensitivity and motor function [Geeraerts et al. 2009].
The pathophysiology of FD has been widely investigated in the past two decades, however, no definite single reason was identified (Figure 2). Multiple mechanisms are likely involved. Specific combinations of physiologic, genetic, environmental, and psychological factors in an individual patient lead to a variety of symptoms called dyspepsia. There is growing evidence that brain–gut interactions at different levels are also involved.
Dysmotility is a main focus of research in the pathophysiology of dyspepsia. Numerous studies have documented abnormal gastroduodenal motility in FD patients [Tack et al. 1998]. The abnormalities range from delayed to accelerated gastric emptying, abnormal antral and fundic contractions, and accommodation issues in the fundus and antrum. The symptoms of FD may be explained by disturbed motility both during and after a meal.
Accommodation is the ability of the stomach to distend appropriately to the size and timing of a meal, allowing an increase in gastric volume without an increase in pressure. Accommodation problems may be expressed as pain during meal ingestion or early satiation. In the study of Bredenoord et al. , 47% of FD patients were found to have impaired accommodation by single photon emission computed tomography (SPECT). In a study with a fundic barostat, impaired accommodation was shown in 40% of patients with FD, and it was associated with the symptom of early satiety [Tack et al. 1998].
SPECT has also been used for assessment of gastric volumes and accommodation. This application is based on the unique capability of both parietal (oxyntic) and nonparietal (mucous) cells to take up and excrete 99mTc-pertechnetate from the circulating blood pool. The use of SPECT to quantify the gastric accommodation response was first proposed and further validated by the Mayo group [De Schepper et al. 2004; Bouras et al. 2002]. Decreased meal size and postmeal symptoms in dyspeptic symptoms were shown to be associated with low fasting gastric volumes and faster gastric emptying [Delgado-Aros et al. 2004a].
Delayed gastric emptying was documented in FD patients in a number of studies [Haag et al. 2004]. One study found delayed emptying of solids and liquids in 23 and 35% of patients, respectively. Delayed gastric emptying of solids was associated with postprandial fullness and vomiting. Delayed emptying of liquids was also associated with postprandial fullness and severe early satiety [Sarnelli et al. 2003a]. Interestingly, another study showed abnormal accelerated gastric outflow in patients measured using a c13 breath test. Around 27% of patients with PDS had accelerated gastric emptying in the early postcibal period associated with PDS-related dyspepsia [Zai and Kusano, 2009].
More than two-thirds of patients with FD show abnormalities with electrogastrography (EGG) and antral/duodenal manometry [Sha et al. 2009]. A subgroup of FD patients with postprandial symptoms show abnormal intragastric distribution of food, independent of the gastric emptying rate [Troncon et al. 1994]. However, this condition cannot be described as dysmotility only; indeed, promotility agents seem to be only partially helpful in FD [Hiyama et al. 2007].
Visceral hypersensitivity plays an important role in FD, as in other FGID. Fundic barostat studies have been used to detect gastric hypersensitivity [Troncon et al. 1994; Mearin et al. 1991]. Recently, Kindt et al. [2009a] demonstrated the existence of a relationship between symptoms, gastric emptying measured by scintigraphy, and gastric sensitivity measured by barostat inflation. Patients with hypersensitivity to distension had significantly higher scores for fullness/satiety. Patients with delayed emptying had significantly higher scores for heartburn/regurgitation, nausea/vomiting, fullness/satiety, bloating, and lower abdominal pain [Kindt et al. 2009a].
On the other hand, another group found that no clear relationship could be demonstrated between postprandial symptoms and proximal stomach function in a study assessing Nepean Dyspepsia Index postprandial symptoms evoked by a drinking test or barostat inflation [Boeckxstaens et al. 2002].
The rationale for the wide use of acid-suppression medications in functional dyspepsia was that acid could be involved in pain-producing mechanisms. Ironically, normal gastric acid secretion was documented in individuals with FD [Collen and Loebenberg, 1989]. One of the mechanisms probably involved in producing the pain might be hypersensitivity to normal acid secretion [Simren et al. 2003], as gastric acid secretion is normal in individuals with FD [Collen and Loebenberg, 1989].
The role of Helicobacter pylori in the pathogenesis of FD has been extensively explored. H. pylori may cause inflammation and dysmotility, probably initiates visceral hypersensitivity, and alters acid secretion, Large population studies have shown that H. pylori is more frequently detected in dyspeptic patients than in controls [O’Morain, 2006], other smaller studies demonstrated a lack of association between H. pylori positive or negative status and FD symptoms [Sarnelli et al. 2003b; Rhee et al. 1999].
The positive relationship between FD and gastrointestinal infections has been reported. Salmonella gastroenteritis was found to be a significant risk factor not only for IBS but also for dyspepsia; at 1 year of follow up after acute S. gastroenteritis, one in seven subjects developed dyspepsia [Mearin et al. 2005]. Another study found increased prevalence of FD after Giardia lamblia infection [Hanevik et al. 2009].
Patients with FD often report their symptoms in relation to certain foods, however data about dietary compounds and their relationship with symptoms is controversial [Carvalho et al. 2009; Pilichiewicz et al. 2009]. Most probably the intolerance for foods suggests a hypersensitivity issue, and not a uniform effect of certain dietary components on FD symptoms.
Brain imaging studies shed some light into the processing and perception of different stimuli from the gastrointestinal tract. Vandenberghe and colleagues conducted a study in FD patients with symptoms of hypersensitivity, and positron emission tomography (PET) imaging was performed during proximal gastric distention. In patients with hypersensitivity, activation of components of the lateral pain system and bilateral frontal inferior gyri, putatively involved in the regulation of hunger and satiety, occurred at significantly lower distention pressures than in healthy volunteers [Vandenberghe et al. 2007].
The higher prevalence of psychiatric comorbidity among patients with FGID is well known. Investigation of psychological factors characterizing the FD personality identifies some traits such as association with familial aggregation, sleep dysfunction, somatization [Gathaiya et al. 2009], and anxiety [Halder et al. 2007]. These findings suggest common triggers which may initiate pathologic cycle processes involving central nervous and gastrointestinal systems.
There is increased recognition of the involvement of the immune system in FGID [Kindt et al. 2009c], based on the observation of common onset timing after acute gastrointestinal infections. Some observations may indicate impaired ability of the immune system to terminate the inflammatory response after acute insult [Kindt et al. 2009b]. Study among patients with persisting abdominal symptoms 6 months after Giardia eradication found 24.3% patients had FD according to Rome II criteria [Hanevik et al. 2009]. Other factors include neuropeptide and hormonal influences on the functioning of the gastrointestinal tract. Total and preprandial ghrelin levels were significantly lower in FD [Lee et al. 2009; Takamori et al. 2007]. Other neuropeptides such as motilin [De Smet et al. 2009], cholecystokinin (CCK), peptide YY [Pilichiewicz et al. 2008], and leptin [Lankarani et al. 2004] have been shown to be altered in FD. Potential genetic links of FD are also emerging. The single-nucleotide polymorphism in the 825T allele of the secondary messenger GNB3 gene [Camilleri and Zinsmeister, 2009; Van Lelyveld et al. 2008], has been reported, perhaps promoting an abnormal perception of visceral pain in FD [Jankowski and Talley, 2009].
The evaluation and management of patients with uninvestigated dyspepsia and investigated dyspepsia should be distinct. Patients with uninvestigated dyspepsia are evaluated and managed usually by primary care physicians. Options include upper endoscopy, testing and treatment for H. pylori, and empirical antisecretory therapy. Alarm symptoms, when present, make upper endoscopy warranted, although the yield of ulcer disease or cancer is low [Delaney et al. 2005]. Early investigation by endoscopy may benefit some patients with dyspepsia but is not cost effective as part of an initial management strategy for uninvestigated dyspepsia [Delaney et al. 2005].
The test and treat H. pylori approach can be cost effective and reduces the number of endoscopies. This approach is the most beneficial in populations with a high prevalence of H. pylori [Talley and Vakil, 2005; Moayyedi et al. 2003a]. In the USA where there is a low prevalence of H. pylori, empirical acid suppression with a PPI is probably the most cost-effective option in younger patients (age 30 years). In older patients, H. pylori test and treat becomes more cost effective [Barton et al. 2008]. In persons with predominant heartburn, GERD should be assumed as a main diagnosis and PPI treatment is indicated [Delaney et al. 2003].
The diagnosis of FD can be made based on Rome III criteria. Despite progress in understanding the physiology of FD using different techniques, there is no single diagnostic test that can confirm the diagnosis; however, several tests are available for assessment of various physiologic functions possibly involved in the pathogenesis of FD (Table 5). After upper endoscopy, there are several other tests for determining the pathophysiology of FD, as listed in Table 5, with gastric emptying studies being the most available. Most of these tests are experimental and not clinically available.
Upper endoscopy is required by Rome III for the diagnosis of FD. Interestingly, negative endoscopy was not found to have ‘reassurance’ value in patients with FD [Van Kerkhoven et al. 2006]. A normal upper endoscopy also helps in dividing dyspepsia into organic and functional. This is often a clinical challenge, since an attribution of endoscopic findings to patients’ symptoms is not always straightforward. The decision as to whether endoscopic findings, especially minor abnormalities, should be accepted as an explanation of the patient’s complaints or the patient has FD is often an empirical decision based on subjective personal experience. On the other hand, we know that significant endoscopic findings such as ulcers, H. pylori infection, or masses may be completely asymptomatic and discovered accidentally [Boehme et al. 2007; Gerson et al. 2002]. One study, exploring the factors that make a difference between symptomatic and symptomatic status, compared FD patients with gastric reddish streaks and asymptomatic counterparts. FD patients with the gastric reddish streaks exhibited increased somatization, more stressful life events, less belief in religion, and less tea consumption compared with their asymptomatic counterparts [Chen et al. 2009].
From all the physiologic stomach functions, assessment of emptying of gastric contents is the most available. Abnormal gastric emptying is a common finding in FD. To measure gastric emptying, scintigraphy, paracetamol absorption test, c13-octanoic acid or Spirulina breath tests, ultrasonography, and magnetic resonance imaging (MRI) have been reported. Of those, scintigraphy is the most frequently used. Delayed gastric emptying has been documented by gastric scintigraphy in a large proportion (up to 45%) of dyspeptic patients [Haag et al. 2004], especially those with PDS. The paracetamol absorption test can measure the gastric emptying of liquids. The c13 breath test can measure the gastric emptying of solids or liquids and can achieve an accuracy comparable with gastric scintigraphy [Ghoos et al. 1993].
Gastric accommodation can be evaluated by SPECT [Bredenoord et al. 2003; Kim et al. 2001], ultrasound, MRI, and barostat (see above). Imaging studies are noninvasive and allow visualization of the gastric wall by different technologies. Ultrasound is a user-dependent tool, which requires significant expertise, therefore, the standardization of results have been difficult.
For the assessment of sensory function the barostat, tensostat, and satiety test were developed. The advantage of the barostat is the simultaneous study of accommodation and sensory thresholds; however, oral-gastric intubation with an intragastric balloon is not easily administrated. Clinically, this test may provide an indication of individual sensitivity to distension stimuli and gastric accommodation [Kindt et al. 2009a; Tack et al. 2001]. The tensostat, a computerized pump that can be programmed to apply a fixed tension level on the gastric wall, is currently only a research tool. The authors that developed this tool suggest using it to study the perception that selectively related to tension [Distrutti et al. 1999].
The water/nutrient (satiety) drinking test is easy and noninvasive, and is proposed as an alternative to sensory studies performed with a gastric balloon. Subjects ingest mineral water or a nutrient liquid at a defined rate (e.g. 15ml/min), the maximal tolerated volume, and the symptoms 30min after ingestion are recorded. In one study, FD patients ingested significantly less water (p<0.001) and had impaired filling of the distal part of the stomach (p=0.001) after the drink test [Van Den Elzen et al. 2007]. Drinking capacity is mainly determined by the antral volume, with reduced antral filling in FD [Boeckxstaens et al. 2001; Van Den Elzen et al. 2007]. In the study of Tack et al. [2003b] satiety scores were higher and maximum satiety occurred at fewer calories in subjects with dyspepsia.
Myoelectrical activity of the stomach is recorded by antroduodenal manometry [Thumshirn et al. 1997] or EGG [Pfaffenbach et al. 1997]. Antroduodenal manometry is performed with a pressure sensor on a catheter introduced orally or nasally into the upper gastrointestinal tract. The catheter is fluoroscopically placed across the antroduodenal region. Antroduodenal manometry was abnormal in 74% of dyspeptics in a survey of 31 FD patients [Troncon et al. 1994; Mearin et al. 1991].
EGG is a recorder with three superficial skin electrodes; two of the electrodes lie over the antrum, and the third skin electrode is placed in the isolateral triangle. Patients with FD revealed rhythm abnormalities as pre- and postprandial tachygastrias [Pfaffenbach et al. 1997]. ECG was abnormal in 71% of the patients in the same study of 31 FD patients [Troncon et al. 1994; Mearin et al. 1991]. EGG and antroduodenal manometry may complement each other in demonstrating gastric motor dysfunction in FD patients [Sha et al. 2009].
Even if the findings of these studies itself are not diagnostic they might help to understand better the physiologic background of the patient’s symptoms and to a certain degree direct the therapeutic choices. In clinical practice, after the diagnosis of FD is established, it is useful to perform gastric emptying measurements in these patients. Documentation of gastric emptying time has practical implications, for example, prolonged gastric emptying time may support the diagnosis of PDS, or significant gastroparesis will be a guide to avoiding medications that delay gastric emptying.
Validated questionnaires were developed for FD. One of them, the Nepean Dyspepsia Index was specifically generated for trials in FD. The quality of life score was shown to provide good discrimination between individuals meeting and not meeting the Rome criteria for FD [Jones and Talley, 2009].
Patient assessment of upper gastrointestinal symptoms (PAGI-SYM) is the severity index questionnaire that was recently developed and validated for the evaluation of therapeutic responsiveness in FD. PAGI-SYM scores are mainly correlated with gastric emptying rate and gastric hypersensitivity [Kindt et al. 2009a; Rentz et al. 2004].
The severity of dyspepsia assessment (SODA) questionnaire measures dyspepsia-related health to serve as the primary outcome measure for randomized clinical trials, and was shown to be responsive to clinically meaningful change [Rabeneck et al. 2001].
The questionnaires do not help to differentiate between different types of dyspepsia patients by symptoms or pathophysiology, and mostly have been used for defining patients for clinical studies.
In general, pharmacologic agents may be classified in several ways: by site of action, central or peripherally acting; by primary physiological effect, prokinetics versus fundic relaxants; by therapeutic targets, such as acid secretion, H. pylori infection, gastric emptying, etc.; and, finally, older agents versus newer developments (Table 6).
Currently, the possibilities of pharmacological therapy for FD are still limited (Figure 3). Several agents have emerged and quickly disappeared because of side effects or a disappointing lack of effectiveness over the past decade.
In the most recent meta-analysis by Moayyedi et al.  from 10 randomized controlled trial (RCTs) evaluating 3347 participants compared PPI with placebo. The average response to PPI therapy was 34% (with a 25% response to placebo). Overall there was a statistically significant benefit of PPI over placebo (relative risk reduction (RRR)=13%; 95% CI, 4–20%), and number needed to treat was 10. There was no statistically significant difference between full versus half-dose PPI, omeprazole 10mg versus 20mg daily, or lansoprozole 15mg versus 30mg daily (RRR=2%, 95% CI, 4–8%) [Moayyedi et al. 2006]. One study reported that PPI therapy showed greater efficacy in H. pylori-positive, nonulcer dyspepsia participants whilst seven studies found no association between H. pylori status and PPI efficacy. Five trials found that PPI therapy was more effective in nonulcer dyspepsia participants with reflux-like symptoms while three trials did not find any association with the symptom subgroup [Moayyedi et al. 2006].
Twelve RCTs compared H2 receptor antagonists (H2RAs) ranitidine 150mg daily or twice daily and famotidine 20mg twice daily, with placebo evaluating a total of 2183 participants. Overall an average of 54% of participants had an improvement in dyspepsia with H2RA therapy compared with 40% in the placebo arm. Overall there was a statistically significant benefit of H2RAs over placebo in terms of improvement of dyspepsia (RRR=23%; 95% CI, 8–35%). The number needed to be treated to improve one case of dyspepsia that would not have been improved by placebo was seven (95% CI, 5–21%) [Moayyedi et al. 2006].
There is evidence of a small benefit in eradicating H. pylori in nonulcer dyspepsia. The updated Cochrane review identified 12 trials and evaluated a total of 2903 patients [Moayyedi et al. 2003a]. All trials used antisecretory dual or triple therapy, and most defined a dyspepsia cure as no symptoms or mild symptoms not interfering with daily activities. The mean placebo response rate at 3–12 months was 29% (range 7–51%), and the mean H. pylori eradication therapy response rate was 37% (range 21–62%).
Overall there was a small but statistically significant benefit of H. pylori eradication therapy at 12 months (relative risk (RR) of remaining dyspeptic with H. pylori eradication therapy 0.91; 95% CI, 0.86–0.95). Fifteen (95% CI, 10–28) H. pylori-infected nonulcer dyspepsia patients needed to be treated with eradication therapy to cure one extra case of nonulcer dyspepsia [Moayyedi et al. 2003a].
A recent meta-analysis [Moayyedi et al. 2003a] suggests a small but consistent benefit from H. pylori eradication with 8% pooled risk reduction compared with placebo at 12 months of follow up. However, the high number (17) needed to be treated may suggest that only in a small proportion of dyspeptic patients is H. pylori the dominant contributor to dyspeptic symptoms.
Bismuth salts were compared with placebo in nine RCTs. The majority of the trials evaluating bismuth assessed the role of H. pylori eradication in nonulcer dyspepsia. There was a trend towards bismuth salts being more effective than placebo although this was of marginal statistical significance (RRR=40%; 95% CI, 65–3%), with no difference between H. pylori infected individuals and uninfected [Moayyedi et al. 2006].
Antacids were not significantly better than placebo (RR=1.02; 95% CI, 0.76–1.36) as reported in the one RCT evaluating this intervention in 109 participants [Moayyedi et al. 2006].
Sucralfate and misoprostol did not show significant improvement in dyspepsia symptoms with treatment [Rentz et al. 2004].
Peppermint oil and caraway oil have been tried in the treatment of FD. The activity of peppermint oil is thought to be primarily spasmolytic due to its calcium-antagonistic properties. Caraway oil increases muscular tone. They are used in combination for a potential synergistic effect [May et al. 2000]. Small studies demonstrated a favorable risk–benefit ratio of peppermint oil and caraway oil for the treatment of FD [Holtmann et al. 2003b]. In one study, the intensity of pain was reduced by 40% versus baseline in the group treated with peppermint oil/caraway oil (coated capsule bid, each containing 90mg peppermint oil and 50mg caraway oil) and by 22% in the placebo group [May et al. 2000].
Two trials with the antimuscarinic agent pirenzepine reported improvements in global dyspepsia symptoms. There was a significant reduction in dyspepsia in treated participants (RRR=50%; 95% CI, 19–69%) [Boehme et al. 2007], although the studies were small, for example, in one study 60 patients were randomized evenly between pirenzepine and placebo [Hradsky and Wikander, 1982].
Meta-analysis of prokinetics (14 trials, 1053 patients) showed an RRR of 48% (95% CI, 23–63%) with four being the number needed to be treated (95% CI, 3–7) [Moayyedi et al. 2003b]; however, most of the studies of prokinetics agents were small and only some of them were randomized. This suggests that the significant effect of prokinetics in FD in this meta-analysis may be due to publication bias.
Since gastric emptying and accommodation are two distinct processes in gastric motility that might be impaired in FD, the pharmacologic agents could be divided into two main groups: affecting gastric emptying and modifying fundic relaxation.
One of the oldest and widely used drugs is metoclopramide. It is a central and peripheral D2 antagonist with gastrokinetic and antiemetic activity [Orihata and Sarna, 1994]. The usual dose is 10mg three times daily. Its multiple central nervous system (CNS), due to the crossing of the blood–brain barrier, and other side effects are well known, and at least 30% of patients cannot tolerate this treatment for a long period. Metoclopramide is not the drug of choice in FD because data about its efficacy in FD is limited [Fumagalli and Hammer, 1994; Archimandritis et al. 1992], and it has a low safety profile.
Domperidone is a dopamine antagonist with peripheral action only, without central side effects. It increases lower esophageal sphincter pressure and antroduodenal motility. However, longer-term administration of domperidone did not accelerate gastric emptying compared with the baseline. The Cochrane meta-analysis suggested the beneficial effects of domperidone 10–20mg three times daily versus placebo on a global rate of dyspeptic symptoms [Veldhuyzen Van Zanten et al. 2001].
Cisapride has prokinetic effects including stimulation of esophageal peristalsis and antroduodenal motility, and acceleration of gastric emptying [Wiseman and Faulds, 1994]. Two meta-analyses of cisapride 5–10mg three times daily treatment compared with placebo showed significant improvement of dyspeptic symptoms [Pilichiewicz et al. 2008; Moayyedi et al. 2006]. Cisapride was withdrawn from the US market in 2000 after it was linked to the occurrence of fatal cardiac arrhythmias [Wysowski and Bacsanyi, 1996].
One RCT compared cisapride with metoclopramide in 60 nonulcer dyspepsia patients [Fumagalli and Hammer, 1994]. Another compared cisapride with domperidone in 84 patients [Halter et al. 1997], and the third study compared metoclopramide with domperidone in 138 patients [Van Outryve et al. 1993]. There was no statistically significant difference between these interventions, although there was a trend for cisapride to be more effective (cisapride versus domperidone: RRR, 48%; 95% CI, 72–84%; cisapride versus metoclopramide: RRR, 43%; 95% CI, 75–81%). There was no difference between metoclopramide and domperidone (RRR, 1%; 95% CI, 30–45%) [Moayyedi et al. 2003b].
Tegaserod is a 5-HT4 partial agonist and also a 5-HT2b antagonist [De Maeyer et al. 2008]. Tegaserod accelerates gastric emptying and enhances gastric accommodation [Tack et al. 2003c]. It was approved by the US Food and Drug Administration for constipation predominant IBS [Tack et al. 2005]. In FD studies with Tegaserod 6mg twice daily improvement was shown in symptoms and gastric emptying [Vakil et al. 2008]. However, the drug was used for FD only in clinical trials. In 2007, tegaserod was withdrawn from market because of the increased incidence of cardiac ischemic events [Schiller and Johnson, 2008].
Tandospirone citrate is a new 5-HT1A agonist. It was shown to improve symptoms in FD patients in a RCT of 144 patients with improvement in abdominal pain (p=0.02) and discomfort (p=0.002) [Miwa et al. 2009].
Itopride is a D2 antagonist and acetylcholinesterase inhibitor, which is marketed in some Asian countries. Phase IIb of a placebo-controlled trial of 554 patients found significant efficacy of itopride in FD. Around 41% of the patients receiving placebo were symptom-free or had marked improvement compared with 57%, 59%, and 64% of those receiving itopride at doses of 50, 100, or 200 mg three times daily, respectively (p<0.05) [Holtmann et al. 2006]. However, while two phase III similar placebo-controlled clinical trials were conducted (international and North America), no significant improvement was observed [Talley et al. 2008b].
Improvement of gastric accommodation seems to be an attractive physiologic target. Several pharmacologic agents were shown to influence gastric accommodation by relaxing gastric fundus. Some of them also alternated the perception of a stimulus at the gastric wall. These agents were mostly tried in clinical trials and are not widely available. Selective serotonin-reuptake inhibitors (SSRIs) can relax gastric fundus and influence gastric sensorimotor function, as shown in the study of Tack and colleagues with paroxetine 20mgqd [Tack et al. 2003a].
Sumatriptan is a 5HT1B/D receptor agonist. At a dose of 6mg subcutaneously, sumitripan relaxes the gastric fundus and increases perception thresholds in healthy volunteers [Tack et al. 2000], and in patients with gastric hypersensitivity it decreases sensitivity to gastric distention meal-induced symptoms [Tack et al. 2004]. The limiting factor to its potential use might be the frequent side effects.
Tegaserod has some effect on gastric accommodation as mentioned above. Nitric oxide [Matsumoto et al. 2009], as well as sildenafil [Sarnelli et al. 2004] are reported to influence gastric relaxation and potentially improve accommodation.
Visceral hypersensitivity may be an underlining cause of symptoms in dyspeptic patients, therefore, using drugs that can modify stimulus perception threshold is reasonable.
Antidepressants may potentially modify several components in FD such as treating underling psychiatric condition (e.g. depression and anxiety), influencing central processing of pain stimuli, and increasing the perception threshold peripherally.
Meta-analysis of antidepressant and antianxiety agents of 13 studies found that in 11 studies participants showed symptom improvement. The main limits of this meta-analysis is were the variable definitions of FD (GERD and IBS included), and inclusion of open studies [Hojo et al. 2005].
Tricyclic agents (amitriptiline, desipramine, dianserin, and clomipramine) have been successfully used in a variety of functional gastrointestinal disorders as shown in the meta-analysis of Jackson et al. . The meta-analysis included 11 studies on patients with different FGID, but only three of them were performed on FD patients. In healthy volunteers, amitriptiline slows gastric emptying of solids, but it does not significantly affect gastric emptying or satiation, and reduces nausea after a high calorie liquid load [Bouras et al. 2008]. The other benefits of tricyclics in FD may come from alterations of stimulus perception by nerve endings, positive changes in mood, and sleep. The side effects, such as weakness, drowsiness, and other anticholinergic phenomena, are frequently dose dependent and should be monitored carefully. SSRIs selectively inhibit the reuptake of serotonin in both the brain and gut.
Sertraline had no effect on gastric sensitivity or compliance, or somatic pain tolerance in healthy humans in one study [Ladabaum and Glidden, 2002]. Pretreatment with paroxetine did not alter the thresholds for perception and discomfort during barostat distention, but significantly enhanced the amplitude of meal-induced fundus relaxation on healthy volunteers. These data suggest that the release of 5-hydroxytryptamine, probably at the level of the enteric nervous system, is involved in the control of the accommodation reflex in humans, and SSRI may be beneficial to patients with impaired postprandial fundus relaxation [Tack et al. 2003b].
The serotonin and norepinephrine reuptake inhibitor venlafaxine was not more effective than placebo in FD patients in a randomized, double-blind, placebo-controlled trial of 160 patients published recently [Van Kerkhoven et al. 2008].
Newer agents that have a potential to alter stimuli perception in the gastrointestinal tract are under intensive investigation and they include opioid receptor antagonists, N-methyl D-aspartate receptor antagonists, neurokinin antagonists, and capsaicin. CCK antagonists may have a role in reducing postprandial symptoms.
Acupuncture has received considerable attention recently. A double-blind and cross-over study in patients with FD evaluated the therapeutic value of transcutaneous electroacupuncture and showed marked improvement in dyspepsia symptoms with possible mechanisms that may be associated with an increase in high frequency heart rate variability and the modulation of neuropeptide Y [Liu et al. 2008]. Several interesting studies are trying to gain an insight of the mechanism of action of this technique in FD patients. PET computed tomography at baseline and after a course of acupuncture suggested that the anterior cingulate cortex, prefrontal cortices, and the caudate tail are involved in processing gastric perceptions in FD patients and the deactivation of the primary somatosensory area and the cerebella contributes to acupuncture stimulation [Zeng et al. 2009]. Another study found electroacupuncture to accelerate solid gastric emptying and improve dyspeptic symptoms in patients with FD [Xu et al. 2006].
In the RCT of Calvert et al. , hypnotherapy was found highly effective for short- and long-term treatment of FD in improving the symptoms, quality of life, and reducing the need for medications, providing major economic advantages.
Hamilton et al.  in a RCT of psychotherapy in patients with chronic FD showed that psychodynamic-interpersonal psychotherapy may have both short- and long-term effects in patients with dyspepsia in improving the symptoms, and there was no difference in outcome between patients with normal and abnormal gastric emptying. However, a recent Cochrane review evaluating the impact of psychological therapies in FD concludes that there is insufficient evidence to confirm the efficacy of psychological intervention despite good results in individual studies [Soo et al. 2005]. Breathing exercises with vagal biofeedback may benefit patients with FD with postprandial symptoms [Hjelland et al. 2007].
Multicomponent herbal preparations from various plants containing complex extracts with a large number of different active substances [Rentz et al. 2004] are popular among the patients. In one placebo-controlled, double-blind study it was found that artichoke leaf extract in the treatment of patients with FD was effective [Holtmann et al. 2003a]. Herbal extract STW 5 (Iberogast) showed efficacy in FD in a placebo-controlled study [Von Arnim et al. 2007]. It stimulates gastric relaxation and antral motility and in this way may contribute to the reported therapeutic efficacy of Iberogast [Pilichiewicz et al. 2007].
Chinese herbal medicine appears to be effective in the treatment of FD and no serious side effects were identified according to available literature. A 2009 meta-analysis of 14 randomized controlled studies revealed that modified xiao yao san compared with prokinetic drugs reduced symptoms (OR 3.26, 95% CI 2.24–4.47). There was evidence that modified xiao yao san and prokinetic drugs compared with prokinetic drugs alone reduced symptoms (OR 4.32, 95% CI 2.64–7.08). However, the evidence remains weak due to publication bias and methodological flaws, which may amplify the therapeutic benefit of modified xiao yao san [Qin et al. 2009].
Japanese herbal medicine, rikkunshi-to, has been shown to have multiple effects on gastric motor function, enhances gastric relaxation, decreases gastric blood flow, and influences gastric myoelectric activity [Suzuki et al. 2009]. In a randomized, placebo-controlled trial, gastric emptying and gastrointestinal symptoms were evaluated in 42 patients with FD. After 7 days of treatment, gastric emptying was significantly accelerated and gastrointestinal symptoms were significantly reduced in patients treated with rikkunshi-to, indicating that it has a prokinetic action on gastric emptying and may be useful in treating FD [Tatsuta and Iishi, 1993].
A major limitation of herbal medicines is the lack of standardization in both the quantity and quality of ingredients. For this reason, the physiological and clinical effects of herbal medicine may not be consistent. In addition, few studies have been designed to test the efficacy of herbal medicine in a definitive manner. However, the combination of extracts with various gastrointestinal active ingredients appears to be reasonable for a heterogeneous condition such as FD [Rosch et al. 2006].
Percutaneous implantation of gastric electrodes, a novel technique applied in animals and in patients with FD, has been used for testing gastric-electrical stimulation and the study of gastric electrophysiology. In one patient with FD the weekly frequency of the referral symptoms decreased [Elfvin et al. 2007].
The development of pharmacological treatment for FD is challenging. The patient population is very large, as is the need for effective management. Several obstacles persist. Most drugs target only one factor, while multiple pathophysiological mechanisms can be involved and the patient group is very heterogeneous. Exclusion of patients with coexistent GERD is not always possible when conducting trials. Also, overlap with other psychiatric and functional gastrointestinal disorders is common. Endpoints of dyspepsia studies are difficult to define and measurements of efficacy by current patient reported outcomes may still not be sensitive enough to detect treatment effects despite validation. Complementary therapies such as herbal medicine and hypnotherapy require further validation.
Studying the pathophysiology can help us to understand better the processes underlining the patients’ complaints although direct correlation to symptom severity is challenging. However, there may be a role in the subcategorization of patients physiologically, which may allow better response to a therapeutic approach. The different combinations of alterations in physiologic gastrointestinal and CNS functions result in the very heterogeneous nature of FD so combined approaches to these patients could be beneficial in challenging cases.
Braden Kuo is a consultant for SmartPill, clinical trial for Evoke. Rita Brun has no conflicts of interest.