A machine learning approach for identification of gastrointestinal predictors for the risk of COVID-19 related hospitalization.

Background and aim: COVID-19 can be presented with various gastrointestinal symptoms. Shortly after the pandemic outbreak, several machine learning algorithms were implemented to assess new diagnostic and therapeutic methods for this disease. The aim of this study is to assess gastrointestinal and liver-related predictive factors for SARS-CoV-2 associated risk of hospitalization. Methods: Data collection was based on a questionnaire from the COVID-19 outpatient test center and from the emergency department at the University Hospital in combination with the data from internal hospital information system and from a mobile application used for telemedicine follow-up of patients. For statistical analysis SARS-CoV-2 negative patients were considered as controls in three different SARS-CoV-2 positive patient groups (divided based on severity of the disease). The data were visualized and analyzed in R version 4.0.5. The Chi-squared or Fisher test was applied to test the null hypothesis of independence between the factors followed, where appropriate, by the multiple comparisons with the Benjamini Hochberg adjustment. The null hypothesis of the equality of the population medians of a continuous variable was tested by the Kruskal Wallis test, followed by the Dunn multiple comparisons test. In order to assess predictive power of the gastrointestinal parameters and other measured variables for predicting an outcome of the patient group the Random Forest machine learning algorithm was trained on the data. The predictive ability was quantified by the ROC curve, constructed from the Out-of-Bag data. Matthews correlation coefficient was used as a one-number summary of the quality of binary classification. The importance of the predictors was measured using the Variable Importance. A 2D representation of the data was obtained by means of Principal Component Analysis for mixed type of data. Findings with the p-value below 0.05 were considered statistically significant. Results: A total of 710 patients were enrolled in the study. The presence of diarrhea and nausea was significantly higher in the emergency department group than in the COVID-19 outpatient test center. Among liver enzymes only aspartate transaminase (AST) has been significantly elevated in the hospitalized group compared to patients discharged home. Based on the Random Forest algorithm, AST has been identified as the most important predictor followed by age or diabetes mellitus. Diarrhea and bloating have also predictive importance, although much lower than AST. Conclusion: SARS-CoV-2 positivity is connected with isolated AST elevation and the level is linked with the severity of the disease. Furthermore, using the machine learning Random Forest algorithm, we have identified the elevated AST as the most important predictor for COVID-19 related hospitalizations.

Gastrointestinal tract involvement in mixed connective tissue disease (Sharp syndrome).

Mixed connective tissue diseases (MCTD) is a very rare autoimmune disease connecting clinical signs of systemic lupus, systemic sclerosis, polymyositis and rheumatoid arthritis. Clinical manifestations are very diverse. In some patients, the digestive tract is affected in varying degrees. The esophagus is affected most often, and patients are complaining of dysphagia. Morphologically, this disorder is similar to the injure in systemic scleroderma. In this case, we describe a unique case of a severe damage of digestive tract manifested by esophageal motility disorders, cachectization, ascites, and repeated ileus conditions.

S-POEM in treatment of achalasia and esophageal epiphrenic diverticula - single center experience.

Background: Standard treatment for esophageal epiphrenic diverticula associated with achalasia includes surgical diverticulectomy, myotomy and anterior fundoplication. However, several case reports published recently suggest that endoscopic approach using per oral endoscopic myotomy is a safe and effective alternative.Methods: This is a retrospective review of a single center case series of patients with achalasia and epiphrenic diverticula. During the treatment, the POEM guided on the opposite site of the diverticular neck without diverticulotomy was performed. Symptomatic outcome was evaluated 3 months after procedure and afterwards with the median follow-up time of 24 months. High resolution manometry was performed 3 months after the procedure.Results: Seven patients with esophageal epiphrenic diverticula were included. POEM was successfully performed in all patients, with no complications in the periprocedural period. We observed a significant reduction of Eckardt score and the relaxation pressure of the lower esophageal sphincter (31.8 vs. 8.8 mmHg, p < .0001).Conclusions: POEM is a promising approach in the management of achalasia and esophageal epiphrenic diverticula. We demonstrated its safety, efficiency and ability to provide symptom reduction and decrease of the LES relaxation pressure even without diverticulotomy. Multicentric studies on larger cohorts of patients and with longer follow-up time are required to confirm these results.

The Effect of Proton Pump Inhibitor Withdrawal on Dabigatran Etexilate Plasma Levels in Patients With Atrial Fibrillation: A Washout Study.

BACKGROUND: Several studies demonstrated that proton pump inhibitors (PPIs) co-administrated with dabigatran in patients with atrial fibrillation (AF) decreased dabigatran trough and peak plasma levels. However, it is still unknown whether this interaction is reversible or not, and whether the withdrawal of PPI would lead to normalization of dabigatran plasma levels. AIM OF STUDY: The aim of this study was to determine the effect of PPI withdrawal on dabigatran plasma levels in patients with AF. METHODS: This pilot prospective study enrolled 23 AF patients on long-term dabigatran and PPI therapy (omeprazole 20 mg twice daily or pantoprazole 40 mg once daily). Dabigatran trough and peak levels (ng/mL) were tested on PPI and after a 2-week period of PPI withdrawal with Hemoclot Thrombin Inhibitor Assay. RESULTS: The analysis of dabigatran plasma levels demonstrated significant elevation in trough dabigatran levels after 2 weeks of PPI withdrawal (97.2 ± 79.7 vs. 163.8 ± 105.5 ng/mL; P < 0.05). Moreover, significantly higher peak dabigatran levels were observed after 2 weeks of PPI withdrawal (142.4 ± 102.8 vs. 255 ± 129.5 ng/mL; P ≤ 0.001). CONCLUSIONS: This study showed that a 2-week period of PPI withdrawal lead to a significant increase in dabigatran trough and peak plasma levels in patients with AF.

Comprehensive analysis of acidic pharyngeal reflux before and after proton pump inhibitor treatment in patients with suspected laryngopharyngeal reflux.

OBJECTIVES: The usefulness of pharyngeal pH monitoring in patients with symptoms attributed to laryngopharyngeal reflux (LPR) has been questioned. One problem is the uncertainty whether the pharyngeal pH monitoring captures the aspects of LPR which are responsible for symptoms. We aimed to gain more insight into this problem by performing a comprehensive analysis of acidic pharyngeal reflux before and after the treatment with proton pump inhibitors (PPIs) in patients with suspected LPR. METHODS: We used simultaneous pharyngeal and distal esophageal 24-hour pH/impedance monitoring to establish the gastroesophageal origin of pharyngeal reflux, and an unbiased approach to analysis by evaluating a whole range of pharyngeal reflux acidity (pH < 6, pH < 5.5, pH < 5.0, pH < 4.5 and pH < 4.0). RESULTS: PPI treatment substantially (by ~50%) improved the symptoms attributed to LPR. In contrast, PPI did not reduce the number of pharyngeal reflux episodes or duration of pharyngeal acid exposure at any pH level. This was also true in a subgroup of patients considered to be good responders to PPI (symptoms improvement by ~75%). Furthermore, good responders did not have more acidic pharyngeal reflux than the patients who were less responsive to PPI. CONCLUSIONS: PPI treatment did not reduce acidic pharyngeal reflux despite substantially improving the symptoms attributed to LPR. This may be because pharyngeal pH monitoring does not quantitatively capture the aspects of LPR responsible for symptoms or because acid causes the symptoms also by mechanisms other than LPR. Our results argue against the utility of pharyngeal pH monitoring in patients with suspected LPR.

The infusion of menthol into the esophagus evokes cold sensations in healthy subjects but induces heartburn in patients with gastroesophageal reflux disease (GERD).

Recent studies in animal models have reported that some afferent fibers innervating the esophagus express the cold receptor TRPM8. In the somatosensory system the stimulation of TRPM8 leads to cold sensations and in certain circumstances alleviates pain. It is therefore hypothesized in this paper that the esophageal infusion of the TRPM8 activator menthol evokes cold sensations from the esophagus and alleviates heartburn in humans. The esophageal infusion of menthol (3 mM, 20 min) evoked cold sensations in 11 of 12 healthy subjects. In striking contrast, the esophageal infusion of menthol evoked heartburn in 10 of 10 patients with gastroesophageal reflux disease (GERD). In healthy subjects the cold sensation evoked by menthol was perceived only as a minor discomfort as evaluated by the visual analog scale (VAS score 1.9 ± 0.3 on the scale 1-10). However, in patients with GERD the menthol-induced heartburn was perceived as painful (VAS score 5.6 ± 0.6, P < 0.01 compared to healthy subjects). It is concluded that the sensations evoked by esophageal infusion of menthol change from relatively nonpainful cold sensations in healthy subjects to painful heartburn sensations in patients with GERD. These qualitative and quantitative changes indicate substantial alterations in afferent signaling mediating sensations from the esophagus in patients with GERD.

Acidic Pharyngeal Reflux Does Not Correlate with Symptoms and Laryngeal Injury Attributed to Laryngopharyngeal Reflux.

BACKGROUND: Laryngopharyngeal reflux (LPR) is suspected when the symptoms are attributed to the penetration of acidic gastroesophageal reflux (GER) into the larynx. However, the relationships between the intensity of LPR and symptoms and laryngeal injury have not been elucidated. Several factors confound the study of LPR, namely pH is monitored in the pharynx (pharyngeal reflux) but the pharyngeal acidity (pH) required to induce laryngeal injury is unknown, the GER origin of pharyngeal acid is not always established, and a recent treatment with proton pump inhibitors (PPI) confounds the analysis. AIMS: We aimed to limit these confounding factors to analyze the relationship between LPR and symptoms and laryngeal injury. METHODS: We used dual pharyngeal and distal esophageal 24-h pH/impedance monitoring to establish GER origin of pharyngeal reflux, we used an unbiased approach to analysis by evaluating a whole range of acidity (pH < 6, pH < 5.5, pH < 5.0, pH < 4.5 and pH < 4.0) in patients with suspected LPR without PPI for > 30 days. RESULTS: Pharyngeal reflux was (median[IQR]) 14[8-20.5] and 4[1.5-6.5] pharyngeal reflux episodes with pH < 6.0 and pH < 5.5, respectively. Pharyngeal reflux with pH < 5.0 was rare. Comprehensive analysis did not reveal any correlation between symptoms (reflux symptom index) or laryngeal injury (reflux finding score) and the number of pharyngeal reflux episodes or duration of pharyngeal acid exposure at any pH level. CONCLUSION: Unbiased comprehensive approach did not reveal any relationship between acidic pharyngeal reflux and the symptoms or laryngeal injury attributed to LPR. Limited clinical usefulness of pharyngeal monitoring reported by others is unlikely due to confounding factors.

Cough reflex sensitivity does not correlate with the esophageal sensitivity to acid in patients with gastroesophageal reflux disease.

The sensitization of cough reflex observed in patients with gastroesophageal reflux disease (GERD) is attributed to activation of vagal C-fibers innervating the esophagus by acid, while the heartburn in GERD is mediated by esophageal acid sensitive C-fibers derived from (dorsal root ganglia) DRG. Here we explored the relationship between cough reflex sensitivity (CRS) and esophageal sensitivity to acid. We evaluated CRS to capsaicin inhalation and esophageal sensitivity to acid (intensity of heartburn evoked by esophageal infusions of acid pH = 3, 2 and 1) in patients with GERD and chronic heartburn before and 3 months after proton pump inhibitor (PPI) treatment. There was no correlation between CRS and esophageal sensitivity to acid at any pH tested. PPI treatment substantially reduced esophageal sensitivity to acid but did not affect CRS. We conclude that a simple direct relationship between CRS and esophageal sensitivity to acid is unlikely. The results indicate that spinal and vagal afferent pathways from the esophagus are probably influenced separately in subjects with GERD.

Idiopathic Pan-Colonic and Small-Intestine Varices.

Idiopathic colonic varices represent a rare source of gastrointestinal haemorrhage with a presumed incidence around 0.0007%. Herein, we present a case of idiopathic colonic and small-intestine varices. According to our knowledge, this case report is the first description of both pan-colonic and small-intestine idiopathic varices of this extent. A young male patient without any previous notable medical history was admitted to the hospital because of massive enterorrhagia with haemodynamic instability. Colonoscopy revealed massive pan-colonic varices. After stabilization, numerous diagnostic procedures were performed in order to investigate the aetiology of pan-colonic varices without any explanation of the patient's condition. In addition, capsule endoscopy revealed varices through the whole length of the small intestine. The final diagnosis was idiopathic varices of the colon and small intestine. Because of the rapid clinical stabilization, the single incident of haemorrhage and the extension of the disease, a conservative approach was chosen (venotonics and ß-blockers). During the 12-month follow-up period, the patient reported no gastrointestinal haemorrhage.

Peroral endoscopic myotomy in achalasia and large epiphrenic diverticulum.

Symptomatic epiphrenic diverticula are mostly treated surgically with laparoscopic diverticulectomy, myotomy and anterior fundoplication. However, in case the patient does not agree with surgical therapy or is contraindicated, there are limited ways of alternative treatment. We present a case report of a 72-year-old female patient with severe dysphagia, regurgitation, paroxysmal cough, weight loss and malnutrition who was diagnosed with achalasia and large epiphrenic diverticulum. She did not give consent to the proposed surgical treatment. Alternatively, peroral endoscopic myotomy (POEM) was carried out. In one-year follow up, we observed complete symptom resolution, significant weight gain, improvement of nutritional status and no complications. We suggest that POEM could serve as an effective and safe alternative treatment for patients with achalasia and esophageal epiphrenic diverticula.