Endomapper dataset of complete calibrated endoscopy procedures

Computer-assisted systems are becoming broadly used in medicine. In endoscopy, most research focuses on the automatic detection of polyps or other pathologies, but localization and navigation of the endoscope are completely performed manually by physicians. To broaden this research and bring spatial Artificial Intelligence to endoscopies, data from complete procedures is needed. This paper introduces the Endomapper dataset, the first collection of complete endoscopy sequences acquired during regular medical practice, making secondary use of medical data. Its main purpose is to facilitate the development and evaluation of Visual Simultaneous Localization and Mapping (VSLAM) methods in real endoscopy data. The dataset contains more than 24 hours of video. It is the first endoscopic dataset that includes endoscope calibration as well as the original calibration videos. Meta-data and annotations associated with the dataset vary from the anatomical landmarks, procedure labeling, segmentations, reconstructions, simulated sequences with ground truth and same patient procedures. The software used in this paper is publicly available.

Predictive Value of Genetic Risk Scores in the Development of Colorectal Adenomas

INTRODUCTION

Unlike colorectal cancer (CRC), few studies have explored the predictive value of genetic risk scores (GRS) in the development of colorectal adenomas (CRA), either alone or in combination with other demographic and clinical factors.

METHODS

In this study, genomic DNA from 613 Spanish Caucasian patients with CRA and 829 polyp-free individuals was genotyped for 88 single-nucleotide polymorphisms (SNPs) associated with CRC risk using the MassArray™ (Sequenom) platform. After applying a multivariate logistic regression model, five SNPs were selected to calculate the GRS. Regression models adjusted by sex, age, family history of CRC, chronic use of NSAIDs, low-dose ASA, and consumption of tobacco were built in order to study the association between GRS and CRA risk. We evaluated the discriminatory capacity using the area under the receiver operating characteristic curve (AUC). The interactions between demographic information and GRS were also analyzed.

RESULTS

Significant associations between high GRS values and risk of CRA for analyzed models were observed. In particular, patients with higher GRS values had 2.3-2.6-fold increase in risk of CRA compared to patients with middle values. Combining sex and age with the GRS significantly increased the discriminatory accuracy of the univariate model with GRS alone. The best model achieved an AUC value of 0.665 (95% CI: 0.63-0.69). The GRS showed a different behavior depending on sex and age.

CONCLUSION

Our findings showed that, besides sex and age, GRS is an important risk factor for development of CRA and may be useful for CRC risk stratification and adaptation of screening programs.

Familial Colorectal Cancer and Genetic Susceptibility: Colorectal Risk Variants in First-Degree Relatives of Patients With Colorectal Cancer

INTRODUCTION

Epidemiological studies estimate that having a first-degree relative (FDR) with colorectal cancer (CRC) increases 2-fold to 3-fold the risk of developing the disease. Because FDRs of CRC patients are more likely to co-inherit CRC risk variants, we aimed to evaluate potential differences in genotype distribution of single nucleotide polymorphisms (SNPs) related to CRC risk between FDRs of patients with nonsyndromic CRC (cases) and individuals with no family history of CRC (controls).

METHODS

We designed a case-control study comprising 750 cases and 750 Spanish Caucasian controls matched by sex, age, and histological findings after colonoscopy. Genomic DNA from all participants was genotyped for 88 SNPs associated with CRC risk using the MassArray (Sequenom) platform.

RESULTS

Ten of the 88 SNPs analyzed revealed significant associations (P < 0.05) with a family history of CRC in our population. The most robust associations were found for the rs17094983G>A SNP in the long noncoding RNA LINC01500 (odds ratio = 0.72; 95% confidence interval: 0.58-0.88, log-additive model), and the rs11255841T>A SNP in the long noncoding RNA LINC00709 (odds ratio = 2.04; 95% confidence interval: 1.19-3.51, dominant model). Of interest, the observed associations were in the same direction than those reported for CRC risk.

DISCUSSION

FDRs of CRC patients show significant differences in genotype distribution of SNPs related to CRC risk as compared to individuals with no family history of CRC. Genotyping of CRC risk variants in FDRs of CRC patients may help to identify subjects at risk that would benefit from stricter surveillance and CRC screening programs.

Predictive Value of Carcinoembryonic Antigen in Symptomatic Patients without Colorectal Cancer: A Post-Hoc Analysis within the COLONPREDICT Cohort

We aimed to assess the risk of cancer in patients with abdominal symptoms after a complete colonoscopy without colorectal cancer (CRC), according to the carcinoembryonic antigen (CEA) concentration, as well as its diagnostic accuracy. For this purpose, we performed a post-hoc analysis within a cohort of 1431 patients from the COLONPREDICT study, prospectively designed to assess the fecal immunochemical test accuracy in detecting CRC. Over 36.5 ± 8.4 months, cancer was detected in 115 (8%) patients. Patients with CEA values higher than 3 ng/mL revealed an increased risk of cancer (HR 2.0, 95% CI 1.3-3.1), CRC (HR 4.4, 95% CI 1.1-17.7) and non-gastrointestinal cancer (HR 1.7, 95% CI 1.0-2.8). A new malignancy was detected in 51 (3.6%) patients during the first year and three variables were independently associated: anemia (OR 2.8, 95% CI 1.3-5.8), rectal bleeding (OR 0.3, 95% CI 0.1-0.7) and CEA level >3 ng/mL (OR 3.4, 95% CI 1.7-7.1). However, CEA was increased only in 31.8% (95% CI, 16.4-52.7%) and 50% (95% CI, 25.4-74.6%) of patients with and without anemia, respectively, who would be diagnosed with cancer during the first year of follow-up. On the basis of this information, CEA should not be used to assist in the triage of patients presenting with lower bowel symptoms who have recently been ruled out a CRC.

Dietary Fat Patterns and Outcomes in Acute Pancreatitis in Spain

Background/Objective: Evidence from basic and clinical studies suggests that unsaturated fatty acids (UFAs) might be relevant mediators of the development of complications in acute pancreatitis (AP). Objective: The aim of this study was to analyze outcomes in patients with AP from regions in Spain with different patterns of dietary fat intake.

Materials and Methods: A retrospective analysis was performed with data from 1,655 patients with AP from a Spanish prospective cohort study and regional nutritional data from a Spanish cross-sectional study. Nutritional data considered in the study concern the total lipid consumption, detailing total saturated fatty acids, UFAs and monounsaturated fatty acids (MUFAs) consumption derived from regional data and not from the patient prospective cohort. Two multivariable analysis models were used: (1) a model with the Charlson comorbidity index, sex, alcoholic etiology, and recurrent AP; (2) a model that included these variables plus obesity.

Results: In multivariable analysis, patients from regions with high UFA intake had a significantly increased frequency of local complications, persistent organ failure (POF), mortality, and moderate-to-severe disease in the model without obesity and a higher frequency of POF in the model with obesity. Patients from regions with high MUFA intake had significantly more local complications and moderate-to-severe disease; this significance remained for moderate-to-severe disease when obesity was added to the model.

Conclusions: Differences in dietary fat patterns could be associated with different outcomes in AP, and dietary fat patterns may be a pre-morbid factor that determines the severity of AP. UFAs, and particulary MUFAs, may influence the pathogenesis of the severity of AP.

Accuracy of the Narrow-Band Imaging International Colorectal Endoscopic Classification System in Identification of Deep Invasion in Colorectal Polyps

BACKGROUND & AIMS

T1 colorectal polyps with at least 1 risk factor for metastasis to lymph node should be treated surgically and are considered endoscopically unresectable. Optical analysis, based on the Narrow-Band Imaging International Colorectal Endoscopic (NICE) classification system, is used to identify neoplasias with invasion of the submucosa that require endoscopic treatment. We assessed the accuracy of the NICE classification, along with other morphologic characteristics, in identifying invasive polyps that are endoscopically unresectable (have at least 1 risk factor for metastasis to lymph node).

METHODS

We performed a multicenter, prospective study of data collected by 58 endoscopists, from 1634 consecutive patients (examining 2123 lesions) at 17 university and community hospitals in Spain from July 2014 through June 2016. All consecutive lesions >10 mm assessed with narrow-band imaging were included. The primary end point was the accuracy of the NICE classification for identifying lesions with deep invasion, using findings from histology analysis as the reference standard. Conditional inference trees were fitted for the analysis of diagnostic accuracy.

RESULTS

Of the 2123 lesions analyzed, 89 (4.2%) had features of deep invasion and 91 (4.3%) were endoscopically unresectable. The NICE classification system identified lesions with deep invasion with 58.4% sensitivity (95% CI, 47.5-68.8), 96.4% specificity (95% CI, 95.5-97.2), a positive-predictive value of 41.6% (95% CI, 32.9-50.8), and a negative-predictive value of 98.1% (95% CI, 97.5-98.7). A conditional inference tree that included all variables found the NICE classification to most accurately identify lesions with deep invasion (P < .001). However, pedunculated morphology (P < .007), ulceration (P = .026), depressed areas (P < .001), or nodular mixed type (P < .001) affected accuracy of identification. Results were comparable for identifying lesions that were endoscopically unresectable.

CONCLUSIONS

In an analysis of 2123 colon lesions >10 mm, we found the NICE classification and morphologic features identify those with deep lesions with >96% specificity-even in non-expert hands and without magnification. ClinicalTrials.gov number NCT02328066.

[Body mass index and tri-ponderal mass index of 1,453 healthy non-obese, non-undernourished millennial children. The Barcelona longitudinal growth study]

INTRODUCTION

Body mass index-for age (BMI) and tri-ponderal mass index-for-age (TMI) values of healthy non-underweight, non-obese millennial children have not been reported until now. We aimed to obtain these values.

SUBJECTS AND METHODS

Longitudinal growth study (1995-2017) of 1,453 healthy non-underweight, non-obese millennial children, from birth (n = 477) or from 4 years of age (n = 976) to 18 years in girls and 19 years in boys (25,851 anthropometric measurements).

RESULTS

In each sex, mean BMI-for-age values increased from birth to one year, declined until 5and increased from then onwards. Mean TMI-for-age values decreased abruptly during the first 6years of age and slowly thereafter, in both sexes. Although, at some ages, mean BMI-for age values differed statistically between sexes, differences were scant and of poor clinical significance. The same occurred for TMI-for-age values. BMI-for-age cut-off values to define underweight status (-2 SD) were similar to those proposed by Cole and the WHO for both sexes. However, BMI-for-age cut-off values to define obesity (+2 SD) were lower in both sexes (1.0-5.3) than those proposed by Cole and similar to those proposed by the WHO until 12 in girls and 14 in boys and lower (1.0-4.8) from these ages onwards.

CONCLUSIONS

BMI-for-age and TMI-for-age values of healthy non-underweight, non-obese millennial children are provided. No clinically relevant differences were observed between sexes. These values may be used to measure underweight status and obesity in present pediatric populations and to evaluate the relationship between BMI-for-age and TMI-for-age in a clinical setting.

Incomplete type of intestinal metaplasia has the highest risk to progress to gastric cancer: results of the Spanish follow-up multicenter study

BACKGROUND AND AIM

In high or moderate risk populations, periodic surveillance of patients at risk of progression from gastric precursor lesions (PL) to gastric cancer (GC) is the most effective strategy for reducing the burden of GC. Incomplete type of intestinal metaplasia (IIM) may be considered as the best candidate, but it is still controversial and more research is needed. To further assess the progression of subtypes of IM as predictors of GC occurrence.

METHODS

A follow-up study was carried-out including 649 patients, diagnosed with PL between 1995-2004 in 9 participating hospitals from Spain, and who repeated the biopsy during 2011-2013. Medical information and habits were collected through a questionnaire. Based on morphology, IM was sub-classified as complete (small intestinal type, CIM) and incomplete (colonic type, IIM). Analyses were done using Cox (HR) models.

RESULTS

At baseline, 24% of patients had atrophic gastritis, 38% CIM, 34% IIM, and 4% dysplasia. Mean follow-up was 12 years. 24 patients (3.7%) developed a gastric adenocarcinoma during follow-up. The incidence rate of GC was 2.76 and 5.76 per 1,000 person-years for those with CIM and IIM, respectively. The HR of progression to CG was 2.75 (95% CI 1.06-6.26) for those with IIM compared with those with CIM at baseline, after adjusting for sex, age, smoking, family history of GC and use of NSAIDs.

CONCLUSIONS

IIM is the PL with highest risk to progress to GC. Sub-typing of IM is a valid procedure for the identification of high risk patients that require more intensive surveillance.

Endoscopic surveillance in patients with multiple (10-100) colorectal polyps

BACKGROUND AND STUDY AIMS

Endoscopic surveillance in patients with multiple colorectal polyps aims to reduce colorectal cancer (CRC) incidence and mortality, as well as the need for colorectal surgery. The aim of this study was to determine the risk of developing CRC or the need for surgery during endoscopic surveillance in a cohort of patients with multiple (10 - 100) colorectal polyps.

PATIENTS AND METHODS

This was a multicentrer, longitudinal, observational study in 15 CRC high risk clinics in Spain, carried out between January 2009 and December 2010. Patients who were included in the EPIPOLIP trial and had at least 1 year of follow-up were included in the study. The primary outcome of interest was the incidence of CRC at least 1 year following the initial colonoscopy. The secondary outcome was the need for colorectal surgery.

RESULTS

A total of 265 patients were followed for a median of 3.8 years. Patients underwent a median of 5 colonoscopies, and 17 patients (6.4 %) were diagnosed with CRC. A total of 32 patients (12.1 %) underwent surgery, including 15 (5.7 %) for prophylaxis without a diagnosis of CRC. The corresponding incidence density rates for CRC and colorectal surgery were 1.4 (95 % confidence interval [CI] 0.7 to 2.1) and 2.7 (95 %CI 1.7 to 3.6) per 100 patient-years, respectively. Only the presence of symptoms at first colonoscopy was independently associated with CRC diagnosis (hazard ratio [HR] 7.7, 95 %CI 1.1 to 59.3) and colorectal surgery (HR 4.6, 95 %CI 1.02 to 20.6).

CONCLUSIONS

Patients with more than 10 neoplastic polyps required frequent colonoscopies within a short follow-up period. More than 10 % of patients required colorectal surgery within 4 years, more than half for incident CRC.