Talley, Nicholas J.

Background The Nepean Dyspepsia Index (NDI) has been in widespread use since its publication in 1999 and the addition of a short form in 2001. The NDI was one of the first disease-specific quality-of-life instruments created for functional dyspepsia (FD). However, its psychometric properties have never been validated in an independent sample. Aim This study aimed to evaluate the validity and reliability of the NDI in an a-priori driven approach in an independent population. Patients and methods In 289 individuals who fulfilled the Rome criteria for FD enrolled in a randomized placebo-controlled trial (FD treatment trial), we examined construct validity, convergent validity, and internal consistency. Results Construct validity was supported in its 25-item unweighted and weighted forms as well as the 10-item short form. All items in the 25-item form yielded considerable (>0.5) standardized loadings on their respective latent variables and all reached statistical significance ( P <0.0001), supporting their relationships with the hypothesized domains. Convergent validity was strongly supported, with every domain being correlated with multiple external instruments; the majority of correlations were in the range 0.3–0.5 (in absolute values). The items comprising each domain showed good internal consistency, with the lowest value of Chronbach α at 0.80. Scores based on the short form (10-item) version of the NDI correlated strongly with the full 25-item form (tension ρ =0.88, interference ρ =0.94, eat/drink ρ =0.95, knowledge ρ =0.84 and work/study ρ =0.97; all P <0.0001). Conclusion The NDI is a valid instrument that can be used to measure the disease-specific impact of FD on quality of life.

AbstractBackgroundIt is unknown why functional gastrointestinal disorders (FGIDs) overlap and limited information exists on risk factors for those with overlap. Our aim was to estimate the prevalence of combinations of FGIDs including reflux (FGIDs‐gastroesophageal reflux [GER]), and evaluate potential risk factors for people with multiple disorders in a representative US community.MethodsA population‐based study was conducted by mailing a valid GI symptom questionnaire to an age‐ and gender‐stratified random sample of residents of Olmsted County, MN. Rome III definitions were used to identify people with FGIDs, and GER was defined by weekly or more frequent heartburn or acid regurgitation. The prevalence of people meeting multiple symptom complexes was estimated. Moreover, potential risk factors for people with multiple disorders were evaluated.Key ResultsA total of 3548 people provided data for each of the necessary symptom questions (mean age: 61±16 years, 54% female). Among these 3548 subjects, 2009 (57%) had no FGIDs‐GER, 906 (26%) had a pure FGID‐GER, 372 (10%) had 2 FGIDs‐GER, and 261 (7%) had 3 or more FGIDs‐GER. Somatization as assessed by a higher Somatic Symptom Checklist score (OR=3.3, 95% CI [2.7,4.1]) was associated with an increased odds for those with 3 or more FGIDs‐GER compared to subjects with a pure FGID‐GER adjusting for age and gender.Conclusions and InferencesSymptom complex overlap is common rather than rare in the community. GER is an integral symptom complex associated with both upper and lower FGIDs. Somatization is a strong risk factor for multiple FGIDs.

AbstractBackgroundEarly life events have been found to be associated with irritable bowel syndrome (IBS) suggesting a role in development of functional disorders. The study aim was to identify potential perinatal risk factors for adult IBS.MethodsUtilizing a population‐based nested case–control design, cases who met modified Rome III criteria for IBS and age‐ and‐gender matched controls were identified using responses from prior mailed surveys to a random sample of Olmsted County residents. Medical records of eligible respondents were reviewed for perinatal events of interest. The association of early life events with subsequent case status was assessed using conditional logistic regression.Key ResultsOf 3 417 respondents, 513 were born in Olmsted County and 108 met criteria for IBS. Due to missing records, 89 pairs were included in the final analyses. Logistic regression revealed only birth weight as a predictor of IBS. Lower birth weight increased the odds for IBS (OR = 1.54 [95% CI = (1.12, 2.08), p = 0.008]). Median birth weight was 3.35 kg (range: 1.96–5.24) and 3.57 kg (range: 2.18–4.59) for cases and controls, respectively. Maternal age, delivery method, and antibiotic exposure were not associated with IBS status but this study was only powered to detect large odds ratios.Conclusions and InferencesLower birth weight was observed as a risk factor for IBS. It is not clear if in utero developmental delays directly lead to IBS or if low birth weight is a prospective marker for subsequent early life problems leading to IBS.

Objective IBS shows genetic predisposition, but adequately powered gene-hunting efforts have been scarce so far. We sought to identify true IBS genetic risk factors by means of genome-wide association (GWA) and independent replication studies. Design We conducted a GWA study (GWAS) of IBS in a general population sample of 11 326 Swedish twins. IBS cases (N=534) and asymptomatic controls (N=4932) were identified based on questionnaire data. Suggestive association signals were followed-up in 3511 individuals from six case-control cohorts. We sought genotype-gene expression correlations through single nucleotide polymorphism (SNP)-expression quantitative trait loci interactions testing, and performed in silico prediction of gene function. We compared candidate gene expression by real-time qPCR in rectal mucosal biopsies of patients with IBS and controls. Results One locus at 7p22.1, which includes the genes KDELR2 (KDEL endoplasmic reticulum protein retention receptor 2) and GRID2IP (glutamate receptor, ionotropic, delta 2 (Grid2) interacting protein), showed consistent IBS risk effects in the index GWAS and all replication cohorts and reached p=9.31×10 −6 in a meta-analysis of all datasets. Several SNPs in this region are associated with cis effects on KDELR2 expression, and a trend for increased mucosal KDLER2 mRNA expression was observed in IBS cases compared with controls. Conclusions Our results demonstrate that general population-based studies combined with analyses of patient cohorts provide good opportunities for gene discovery in IBS. The 7p22.1 and other risk signals detected in this study constitute a good starting platform for hypothesis testing in future functional investigations.

AbstractBackgroundThe etiology of irritable bowel syndrome (IBS) is not been fully elucidated, but childhood trauma may disturb the brain–gut axis and therefore be important. Thus, we conducted a family based case–control study of IBS cases and their relatives with the aims to (i) determine the frequency of childhood trauma among IBS cases and controls as well as their relatives, and (ii) assess childhood trauma among IBS cases with affected relatives (familial IBS).MethodsOutpatients with IBS, matched controls, and their first‐degree relatives completed a self‐report version of Bremner' Early Trauma Inventory. Percent of cases and controls with a family history were compared and odds ratios were computed using chi‐squared test; recurrence risks to relatives were computed using logistic regression and generalized estimating equations.Key ResultsData were collected from 409 cases, 415 controls, 825 case relatives, and 921 control relatives. IBS cases had a median age of 50 and 83% were women. Of IBS cases, 74% had experienced any general trauma compared to 59% among controls, yielding an odds ratio of 1.56 (95% CI: 1.13–2.15, p < 0.008). There were no statistical differences between IBS relatives and control relatives with regards to lifetime trauma.Conclusions & InferencesIBS is associated with childhood trauma, and these traumas often occur prior to onset of IBS symptoms. This provides further insight into how traumatic childhood events are associated with development of adult IBS.

A community sample of 610 adults were grouped into those who reported no sexual abuse experience, others who reported noncoercive sexual contact with an individual that was at most 4 years older, and those who reported more severe sexual abuse. The first two groups did not differ from each other on current social support, trauma-specific symptomatology, and somatic complaints whereas the more severe sexual abuse group reported more problems on these dimensions. Further analyses suggested that of the two variables, coercion was more closely associated with problematic outcome than was the existence of an age difference.