The Research On Symptom Pattern In Functional Dyspepsia A Factor And Cluster Analysis

Part I Symptom pattern in functional dyspepsia and its evolutionObjective:To study the symptom pattern in functional dyspepsia and its evolution in the general population.Methods:From June2008to November2009functional dyspepsia patients who came from five clinical centers in Wuhan, Beijing, Shanghai, Guangzhou and Xi’an city and fulfilled the Rome Ⅲ criteria were recruit to complete clinical questionnaire and12-month follow-up survey. The baseline and follow-up data of patients were collected and collated to establish the Epidate3.0software database for statistical analysis with R language3.0.1.Results:1. The severity and frequency of different dyspepsia symptoms are significant different in functional dyspepsia patients. The severity and frequency score of postprandial fullness, epigastric pain, epigastric burning, early satiety and belching are higher than other dyspepsia symptoms.2. There are significant positive correlations between the severity and frequency of dyspepsia symptoms and significant correlations between the severities of dyspepsia symptoms and concomitant gastrointestinal symptoms in functional dyspepsia patients.3. The cumulative score of severity of dyspepsia symptoms in functional dyspepsia patients overlapped with irritable bowel syndrome and/or gastro-esophageal reflux disease are higher than patients without irritable bowel syndrome and/or gastro-esophageal reflux disease.4. The severity of dyspepsia symptoms and concomitant gastrointestinal symptoms in PDS&EPS is often higher than in PDS alone and in EPS alone.5. PDS&EPS is more prone to overlap with irritable bowel syndrome and/or gastro-esophageal reflux disease than PDS alone and EPS alone.6. After12months follow-up684(72.53%) cases have symptom resolution and no longer fulfill the Rome III criteria for functional dyspepsia.7. The subgroups of functional dyspepsia are instability and often transition over time.8. Multivariate logistic regression analysis revealed that the drug therapies are not the reason for functional dyspepsia subgroups transition (P>0.05).9. Compared with the baseline after12month follow-up the cumulative score of severity of dyspepsia symptoms in PDS alone (P<0.05) significant reduced but not significant reduced in EPS alone (P>0.05) and PDS&EPS (P>0.05).Conclusion:1. Functional dyspepsia is a heterogeneous condition with intricate heterogeneity in the dyspepsia symptom complex and pathogenesis and often overlapped with other functional gastrointestinal disorders. The symptom patterns in functional dyspepsia lack stability and vary over time with spontaneous resolution and subgroup transition.2. Compared with PDS alone and EPS alone there are more complex pathological mechanism and more severe dyspepsia symptoms in PDS&EPS. PDS&EPS is more prone to overlap with other functional gastrointestinal disorders than PDS alone and EPS alone. Part II the factor and cluster analysis of the symptom pattern in functional dyspepsiaObjective:To clarify the determinant symptom factors and factor structure of symptom pattern in functional dyspepsia with factor analysis and determine the cluster structure characteristic of functional dyspepsia patients by cluster analysis based on the results of the factor analysis for exploring the significance of the factor and cluster analysis in clinical research of functional dyspepsia.Methods:To collect and collate the baseline and follow-up data of functional dyspepsia patients who fulfilled the Rome Ⅲ criteria and establish Epidate3.0software database for statistical analysis with R language3.0.1. Principal component factor analysis was applied to reduce data in statistical analysis and determine the few unable to directly measured potential variables (symptom factors) underlying the symptom pattern in functional dyspepsia. Subsequently, nonhierarchical k-means cluster analysis based on the factor scores derived earlier was performed to determine the cluster structure characteristic of functional dyspepsia patients. And then the differences in gastrointestinal symptom severity in each cluster are analyzed with Kruskal-wallis H test and wilcoxon rank sum test. The difference in constituent ratio of Rome Ⅲ criteria subgroups (PDS alone/EPS alone/PDS&EPS) in each cluster are analyzed with partitions of χ2method. And the difference in prevalence of irritable bowel syndrome and gastro-esophageal reflux disease in each cluster are analyzed with Fisher exact probability method and partitions of x2method respectively.Results:The factor analysis revealed the presence of several independent and separated symptom factors including postprandial fullness/early satiety factor, epigastric pain/abdominal discomfort factor, reflux factor, belching factor, nausea/vomiting factor, diarrhea factor, abdominal distension/constipation factor and irritable gut factor and representing the specific symptom groups with different pathological mechanisms, respectively. There are undifferentiated clusters with undetermined symptoms characteristic in both the cluster structures of the cardinal dyspepsia symptoms and upper gastrointestinal symptoms of functional dyspepsia patients. But the clusters based on the whole gastrointestinal symptoms all have determined symptoms characteristic such as postprandial fullness/early satiety, epigastric pain/abdominal discomfort, belching, reflux, nausea/vomiting/abdominal bloating/abdominal distension, irritable gut/diarrhea, respectively. The prevalence of irritable bowel syndrome is the highest in the cluster predominated with irritable gut/diarrhea (30.94%) and followed by the cluster predominated with nausea/vomiting/abdominal bloating/abdominal distension (12.96%). The prevalence of gastro-esophageal reflux disease is the highest in the cluster predominated with reflux (57.14%) and followed by the cluster predominated with nausea/vomiting/abdominal bloating/abdominal distension (48.15%) and the cluster predominated with irritable gut/diarrhea (31.65%).Conclusion:1. The more gastrointestinal symptoms are included in the factor and cluster analyses, the more critical symptom characteristics can be comprehensively interpreted in functional dyspepsia.2. Compared with the arbitrarily predefined classification of functional dyspepsia the result derived from the factor and cluster analysis can more objectively and accurately reflect the symptoms characteristic of patients in different subsets (clusters).3. The patient in the clusters predominated with the symptom characteristics such as irritable gut/diarrhea and nausea/vomiting/abdominal bloating/abdominal distension and reflux have more complex symptom pattern and intricate pathogenesis.