Big data in gastroenterology research

Host-related low-prevalence gastritides: Epidemiological and clinical characterization

INTRODUCTION

A recent consensus meeting (RE.GA.IN) addressed "host-related, low-prevalence gastritis": eosinophilic (EoG), lymphocytic (Hp-pos_LyG and Hp-neg_LyG), collagenous (CollG), and granulomatous gastritis (GrG). Our study evaluates their clinico-epidemiological characteristics.

MATERIALS AND METHODS

We extracted all patients with a diagnosis of EoG, LyG, CollG, and GrG from a clinicopathological database and compared their demographics, clinical and endoscopic characteristics, associated conditions, and clinical awareness to those of all other subjects in the database (controls).

RESULTS

There were 1,781,005 unique patients (median age 57 years; 55.7 % female). Hispanics were overrepresented amongst those with Hp-pos_LyG. Subjects with GrG had a high prevalence of erosions and ulcers. Clinical awareness of these conditions was dismal (<1:10,000 patients). Some clinical manifestations were more common in patients with certain gastritides (e.g., vomiting and diarrhea in CollG; anemia in LyG), but none were sufficiently distinctive to suggest a clinical diagnosis. EoG was associated with EoE; LyG had a strong association with celiac disease; CollG with microscopic colitis; and GrG with Crohn disease.

CONCLUSIONS

The diagnosis of these gastritides (between <1: in 1,000 and 1 in 5000 subjects) rests on histopathology. They remain poorly characterized and clinically neglected. Yet, their associations may herald other conditions: eosinophilic gastrointestinal diseases (EGID), celiac, and Crohn disease. Patients might benefit from increased detection and characterization.

"Beyond MELD" - Emerging strategies and technologies for improving mortality prediction, organ allocation and outcomes in liver transplantation

In this review article, we discuss the model for end-stage liver disease (MELD) score and its dual purpose in general and transplant hepatology. As the landscape of liver disease and transplantation has evolved considerably since the advent of the MELD score, we summarise emerging concepts, methodologies, and technologies that may improve mortality prognostication in the future. Finally, we explore how these novel concepts and technologies may be incorporated into clinical practice.

Cystic fibrosis patients on cystic fibrosis transmembrane conductance regulator modulators have a reduced incidence of cirrhosis

BACKGROUND

Cystic fibrosis transmembrane conductance regulator (CFTR) modulators significantly improve pulmonary function in patients with cystic fibrosis (CF) but the effect on hepatobiliary outcomes remains unknown. We hypothesized that CF patients on CFTR modulators would have a decreased incidence of cirrhosis compared to patients not on CFTR modulators or on ursodiol.

AIM

To investigate the effect of CFTR modulators on the development of cirrhosis in patients with CF.

METHODS

A retrospective analysis was performed using Truven MarketScan from January 2012 through December 2017 including all patients with a diagnosis of CF. Patients were excluded if they underwent a liver transplantation or if they had other etiologies of liver disease including viral hepatitis or alcohol use. Subjects were grouped by use of CFTR modulators, ursodiol, dual therapy, or no therapy. The primary outcome was development of cirrhosis. Kaplan-Meier curves estimated the incidence of cirrhosis and log-rank tests compared incidence curves between treatment groups.

RESULTS

A total of 7201 patients were included, of which 955 (12.6%) used a CFTR modulator, 529 (7.0%) used ursodiol, 105 (1.4%) used combination therapy, and 5612 (74.3%) used neither therapy. The incidence of cirrhosis was 0.1% at 1 year and 0.7% at 4 years in untreated patients, 5.9% and 10.1% in the Ursodiol group, and 1.0% and 1.0% in patients who received both therapies. No patient treated with CFTR modulators alone developed cirrhosis. Patients on CFTR modulators alone had lower cirrhosis incidence than untreated patients (P = 0.05), patients on Ursodiol (P < 0.001), and patients on dual therapy (P = 0.003). The highest incidence of cirrhosis was found among patients treated with Ursodiol alone, compared to untreated patients (P < 0.001) or patients on Ursodiol and CFTR modulators (P = 0.01).

CONCLUSION

CFTR modulators are associated with a reduction in the incidence of cirrhosis compared to other therapies in patients with CF.

Collagenous gastritis: Epidemiology and clinical associations

BACKGROUND

the rare occurrence of collagenous gastritis (CG) makes its epidemiology difficult to investigate. We designed a study to determine the demographic and clinical characteristics as well as the associations of CG with other upper gastrointestinal diseases in a large national clinicopathological database.

METHODS

from the IDEA database we extracted all patients with histopathologically documented CG and, in a case-control study, we compared 168 subjects with and 1,286,165 subjects without CG using odds ratios (OR) with their 95% confidence intervals (CI).

RESULTS

the prevalence of CG was 13 per 100,000 EGDs. CG was significantly more common among female than male patients (OR: 1.69, 95% CI: 1.20-2.39) and was characterized by a bi-modal age distribution (first peak in patients aged 10-19, second peak primarily in females aged >60 years). CG patients presented with diarrhea (18%), anemia (12%), weight loss (11%), and vomiting (10%). CG was significantly associated with other lymphocytic disorders of the upper gastrointestinal tract, including celiac sprue (2.12, 1.55-2.88), duodenal intraepithelial lymphocytosis (3.71, 2.30-5.98), and lymphocytic gastritis (23.2, 10.9-49.5). CG persisted in 69% of patients who underwent multiple consecutive endoscopies.

CONCLUSIONS

the epidemiologic features of collagenous gastritis reflect on different etiologies contributing to its occurrence in children and adults.

Acute and Severe Acute Pancreatitis and the Effect of Cannabis in States Before and After Legalization Compared With States Without Legalized Cannabis

OBJECTIVES

Cannabis legalization has increased its use. The incidence of acute pancreatitis (AP) and severe acute pancreatitis (SAP) has also increased. In this study, data on pancreatitis were obtained from 2 states before and after cannabis legalization and compared with 2 states without legalized cannabis.

METHODS

Data were extracted from State Inpatient Databases from the states of Colorado and Washington before recreational cannabis legalization (2011) and after legalization (2015). Arizona and Florida were used as the nonlegalized cannabis states. Multivariable logistic regression models were fit for AP and SAP to determine a trend difference between legalized and nonlegalized cannabis states.

RESULTS

Cannabis use, AP, and SAP increased in all states. The increase in AP and SAP was not significantly different between the states that legalized cannabis use and those that did not. Legalized cannabis states had lower charges for AP and SAP and shorter length of hospitalizations.

CONCLUSIONS

The trend of AP and SAP increased during the study period, but this was not correlated to cannabis use. Cannabis users had lower hospitalization costs and hospital stay. The effects of other confounders such as cannabis dose and delivery methods, alcohol, tobacco, and others need to be studied further as use increases.

A Study on the Effect of Patient Characteristics, Geographical Utilization, and Patient Outcomes for Total Pancreatectomy Alone and Total Pancreatectomy With Islet Autotransplantation in Patients With Pancreatitis in the United States

OBJECTIVES

A selective therapy for pancreatitis is total pancreatectomy and islet autotransplantation. Outcomes and geographical variability of patients who had total pancreatectomy (TP) alone or total pancreatectomy with islet autotransplantation (TPIAT) were assessed.

METHODS

Data were obtained from the Healthcare Cost and Utilization Project National Inpatient Sample database. Weighed univariate and multivariate analyses were performed to determine the effect of measured variables on outcomes.

RESULTS

Between 2002 and 2013, there were 1006 TP and 825 TPIAT in patients with a diagnosis of chronic pancreatitis, and 1705 TP and 830 TPIAT for any diagnosis of pancreatitis. The majority of the TP and TPIAT were performed in larger urban hospitals. Costs were similar for TP and TPIAT for chronic pancreatitis but were lower for TPIAT compared with TP for any type of pancreatitis. The trend for TP and TPIAT was significant in all geographical areas during the study period.

CONCLUSIONS

There is an increasing trend of both TP and TPIAT. Certain groups are more likely to be offered TPIAT compared with TP alone. More data are needed to understand disparities and barriers to TPIAT, and long-term outcomes of TPIAT such as pain control and glucose intolerance need further study.

Application of Big Data analysis in gastrointestinal research

Big Data, which are characterized by certain unique traits like volume, velocity and value, have revolutionized the research of multiple fields including medicine. Big Data in health care are defined as large datasets that are collected routinely or automatically, and stored electronically. With the rapidly expanding volume of health data collection, it is envisioned that the Big Data approach can improve not only individual health, but also the performance of health care systems. The application of Big Data analysis in the field of gastroenterology and hepatology research has also opened new research approaches. While it retains most of the advantages and avoids some of the disadvantages of traditional observational studies (case-control and prospective cohort studies), it allows for phenomapping of disease heterogeneity, enhancement of drug safety, as well as development of precision medicine, prediction models and personalized treatment. Unlike randomized controlled trials, it reflects the real-world situation and studies patients who are often under-represented in randomized controlled trials. However, residual and/or unmeasured confounding remains a major concern, which requires meticulous study design and various statistical adjustment methods. Other potential drawbacks include data validity, missing data, incomplete data capture due to the unavailability of diagnosis codes for certain clinical situations, and individual privacy. With continuous technological advances, some of the current limitations with Big Data may be further minimized. This review will illustrate the use of Big Data research on gastrointestinal and liver diseases using recently published examples.

Big data in IBD: a look into the future

Big data methodologies, made possible with the increasing generation and availability of digital data and enhanced analytical capabilities, have produced new insights to improve outcomes in many disciplines. Application of big data in the health-care sector is in its early stages, although the potential for leveraging underutilized data to gain a better understanding of disease and improve quality of care is enormous. Owing to the intrinsic characteristics of inflammatory bowel disease (IBD) and the management dilemmas that it imposes, the implementation of big data research strategies not only can complement current research efforts but also could represent the only way to disentangle the complexity of the disease. In this Review, we explore important potential applications of big data in IBD research, including predictive models of disease course and response to therapy, characterization of disease heterogeneity, drug safety and development, precision medicine and cost-effectiveness of care. We also discuss the strengths and limitations of potential data sources that big data analytics could draw from in the field of IBD, including electronic health records, clinical trial data, e-health applications and genomic, transcriptomic, proteomic, metabolomic and microbiomic data.

The Use of International Classification of Diseases Codes to Identify Patients with Pancreatitis: A Systematic Review and Meta-analysis of Diagnostic Accuracy Studies

Background

Hospital discharge codes are increasingly used in gastroenterology research, but their accuracy in the setting of acute pancreatitis (AP) and chronic pancreatitis (CP), one of the most frequent digestive diseases, has never been assessed systematically. The aim was to conduct a systematic literature review and determine accuracy of diagnostic codes for AP and CP, as well as the effect of covariates.

Methods

Three databases (Pubmed, EMBASE and Scopus) were searched by two independent reviewers for relevant studies that used International Classification of Disease (ICD) codes. Summary estimates of sensitivity, specificity and positive predictive value were obtained from bivariate random-effects regression models. Sensitivity and subgroup analyses according to recurrence of AP and age of the study population were performed.

Results

A total of 24 cohorts encompassing 18,106 patients were included. The pooled estimates of sensitivity and specificity of ICD codes for AP were 0.85 and 0.96, respectively. The pooled estimates of sensitivity and specificity of ICD codes for CP were 0.75 and 0.94, respectively. The positive predictive value of ICD codes was 0.71 for either AP or CP. It increased to 0.78 when applied to incident episode of AP only. The positive predictive value decreased to 0.68 when the ICD codes were applied to paediatric patients.

Conclusion

Nearly three out of ten patients are misidentified as having either AP or CP with the indiscriminate use of ICD codes. Limiting the use of ICD codes to adult patients with incident episode of AP may improve identification of patients with pancreatitis in administrative databases.

Liver cirrhosis and cancer: comparison of mortality

BACKGROUND

Liver cirrhosis is known to have low survival rate, and its assessment in relation with other fatal diseases will help us design appropriate health interventions. This study compares the mortality of liver cirrhosis with that of five major cancers (lung, colorectal, stomach, liver, and breast cancers).

METHODS AND RESULTS

We used the National Health Insurance Service-National Sample Cohort (NHIS-NSC) which provides data for 1,025,340 representative samples of the 46,605,433 people in Korea from 2002 to 2010. During the 8 years, 800 out of 2609 liver cirrhosis patients died and 1316 out of 4852 patients with the five major cancers died. When we estimated the mortality between liver cirrhosis and five major cancers, the relative mortality for liver cirrhosis was greater [hazard ratio 1.47 (95% CI 1.28-1.67) after age, gender, area of residence, type of insurance, insurance premium level (proxy for income level), and comorbidities were adjusted for]. When a sensitivity analysis was performed by excluding patients with both liver cirrhosis and one of the five cancers, the relative mortality was still greater for liver cirrhosis [hazard ratio 1.27 (95% CI 1.10-1.47)]. Furthermore, when we limited liver cirrhosis patients to those with decompensated liver cirrhosis, the relative mortality of decompensated liver cirrhosis was even greater than that of the five cancers [hazard ratio 1.82 (95% CI 1.51-2.20)].

CONCLUSIONS

The mortality of liver cirrhosis is greater than that of the five major cancers. This implies the need to prioritize appropriate health interventions for liver cirrhosis.

Personalized treatment of patients with very early hepatocellular carcinoma

Hepatocellular carcinoma (HCC), in its very early stage, is heterogeneous both in terms of liver function (i.e., presence or absence of portal hypertension, model for end-stage liver disease score, Child-Pugh score 5 or 6, bilirubin level) and tumor characteristics (i.e., location, alpha-fetoprotein values, pathological features such as microvascular invasion, tumor grade and satellitosis). Existing evidence in comparing different curative options for patients with very early HCC is poor due to small sample sizes and lack of solid subgroup analyses. Large observational studies are available, with the potential to identify effective interventions in different subgroup of patients and to discover which treatments work "in a real world setting". These studies suggest some important treatment selection strategies in very early HCC patients. According to extent of liver resection, and liver function, percutaneous ablation or liver resection are the recommended first line therapies in these patients. Laparoscopic surgery (resection or ablation) is the preferable strategy when the tumor is in the surface of the liver or close to extra-hepatic organs. Due to scarce donor resources and competition with patients at high transplant benefit (HCC patients unsuitable for non-transplant radical therapies and non-HCC patients with decompensated cirrhosis), transplantation is recommended only as second line therapy in patients with very early stage HCC in case of tumor recurrence or liver failure after ablation or liver resection.

The Chronic Pancreatitis International Classification of Diseases, Ninth Revision, Clinical Modification Code 577.1 Is Inaccurate Compared With Criterion-Standard Clinical Diagnostic Scoring Systems

OBJECTIVES

Epidemiological studies utilizing administrative databases use the International Classification of Diseases, Ninth Revision, Clinical Modification code (577.1) to identify patients with chronic pancreatitis (CP). We hypothesized that coding of CP in these databases is inaccurate.

METHODS

We retrospectively reviewed the records of 1343 consecutive patients having an International Classification of Diseases, Ninth Revision, Clinical Modification code 577.1 between October 1, 2005, and November 1, 2008. We labeled patients as definite CP or non-CP, defined as fulfilling any of the 3 diagnostic criteria for definite CP: Mayo, Ammann's, and the Japanese Pancreas Society criteria.

RESULTS

Six hundred fifty-eight subjects (49%) had definite CP. Definite CP among Mayo, Ammann's, and Japanese Pancreas Society criteria was similar (49.0, 42.1, and 43.8, respectively); 84.3% of the definite CP fulfilled all 3 criteria, 6.7% fulfilled 2, and 9.0% filled 1 criterion. Etiologies of definite CP were definite/suspected nonalcohol (57.8%) or alcohol (33.5%) and not reported (8.7%). In non-CP (n = 685), 93.7% had available imaging (5.7% had endoscopic ultrasonography ≥5 or Cambridge II-III scores), and 63.5% had symptoms suggestive of CP but did not fulfill other features for definite CP. The CP versus non-CP groups had similar mean ages but significantly more men and patients ever smoking and ever drinking alcohol.

CONCLUSIONS

Fifty-one percent of subjects coded as CP do not fulfill the diagnostic criteria for definite CP. Relying solely on the International Classification of Diseases, Clinical Modification code for CP in administrative databases may lead to erroneous epidemiological conclusions.

Foodomics as part of the host-microbiota-exposome interplay

The functional complexity of human gut microbiota and its relationship with host physiology and environmental modulating factors, offers the opportunity to investigate (i) the host and microbiota role in organism-environment relationship; (ii) the individual functional diversity and response to environmental stimuli (exposome); (iii) the host genome and microbiota metagenomes' modifications by diet-mediated epigenomic controls (nutriepigenomics); and (iv) the genotype-phenotype "trajectories" under physiological and disease constraints. Systems biology-based approaches aim at integrating biological data at cellular, tissue and organ organization levels, using computational modeling to interpret diseases' physiopathological mechanisms (i.e., onset and progression). Proteomics improves the existing gene models by profiling molecular phenotypes at protein abundance level, by analyzing post-translational modifications and protein-protein interactions and providing specific pathway information, hence contributing to functional molecular networks. Transcriptomics and metabolomics may determine host ad microbiota changes induced by food ingredients at molecular level, complementing functional genomics and proteomics data. Since foodomics is an -omic wide methodology may feed back all integrative data to foster the omics-based systems medicine field. Hence, coupled to ecological genomics of gut microbial communities, foodomics may highlight health benefits from nutrients, dissecting diet-induced gut microbiota eubiosis mechanisms and significantly contributing to understand and prevent complex disease phenotypes.

BIOLOGICAL SIGNIFICANCE

Besides transcriptomics and proteomics there is a growing interest in applying metabolic profiling to food science for the development of functional foods. Indeed, one of the biggest challenges of modern nutrition is to propose a healthy diet to populations worldwide, intrinsically respecting the high inter-individual variability, driven by complex host/nutrients/microbiota/environment interactions. Therefore, metabolic profiling can assist at various levels for the development of functional foods, starting from screening for food composition to identification of new biomarkers to trace food intake. This current approach can support diet intervention strategies, epidemiological studies, and controlling of metabolic disorders worldwide spreading, hence ensuring healthy aging. With high-throughput molecular technologies driving foodomics, studying bidirectional interactions of host-microbial co-metabolism, innate immune development, dysfunctional nutrient absorption and processing, complex signaling pathways involved in nutritional metabolism, is now likely. In all cases, as microbiome pipeline efforts continue, it is possible that enhanced standardized protocols can be developed, which may lead to new testable biological and clinical hypotheses. This Review provides a comprehensive update on the current state-of-the-art of the integrated -omics route in food, microbiota and host co-metabolism studies, which may revolutionize the design of new dietary intervention strategies.

Big Data Application in Biomedical Research and Health Care: A Literature Review

Big data technologies are increasingly used for biomedical and health-care informatics research. Large amounts of biological and clinical data have been generated and collected at an unprecedented speed and scale. For example, the new generation of sequencing technologies enables the processing of billions of DNA sequence data per day, and the application of electronic health records (EHRs) is documenting large amounts of patient data. The cost of acquiring and analyzing biomedical data is expected to decrease dramatically with the help of technology upgrades, such as the emergence of new sequencing machines, the development of novel hardware and software for parallel computing, and the extensive expansion of EHRs. Big data applications present new opportunities to discover new knowledge and create novel methods to improve the quality of health care. The application of big data in health care is a fast-growing field, with many new discoveries and methodologies published in the last five years. In this paper, we review and discuss big data application in four major biomedical subdisciplines: (1) bioinformatics, (2) clinical informatics, (3) imaging informatics, and (4) public health informatics. Specifically, in bioinformatics, high-throughput experiments facilitate the research of new genome-wide association studies of diseases, and with clinical informatics, the clinical field benefits from the vast amount of collected patient data for making intelligent decisions. Imaging informatics is now more rapidly integrated with cloud platforms to share medical image data and workflows, and public health informatics leverages big data techniques for predicting and monitoring infectious disease outbreaks, such as Ebola. In this paper, we review the recent progress and breakthroughs of big data applications in these health-care domains and summarize the challenges, gaps, and opportunities to improve and advance big data applications in health care.

Software for enhanced video capsule endoscopy: challenges for essential progress

Video capsule endoscopy (VCE) has revolutionized the diagnostic work-up in the field of small bowel diseases. Furthermore, VCE has the potential to become the leading screening technique for the entire gastrointestinal tract. Computational methods that can be implemented in software can enhance the diagnostic yield of VCE both in terms of efficiency and diagnostic accuracy. Since the appearance of the first capsule endoscope in clinical practice in 2001, information technology (IT) research groups have proposed a variety of such methods, including algorithms for detecting haemorrhage and lesions, reducing the reviewing time, localizing the capsule or lesion, assessing intestinal motility, enhancing the video quality and managing the data. Even though research is prolific (as measured by publication activity), the progress made during the past 5 years can only be considered as marginal with respect to clinically significant outcomes. One thing is clear-parallel pathways of medical and IT scientists exist, each publishing in their own area, but where do these research pathways meet? Could the proposed IT plans have any clinical effect and do clinicians really understand the limitations of VCE software? In this Review, we present an in-depth critical analysis that aims to inspire and align the agendas of the two scientific groups.