-
Neurogastroenterology and Motility Dec 2022Several magnetic resonance imaging (MRI) protocols have been used to assess gastric emptying (GE) with MRI. This systematic review summarizes the current literature on... (Review)
Review
BACKGROUND
Several magnetic resonance imaging (MRI) protocols have been used to assess gastric emptying (GE) with MRI. This systematic review summarizes the current literature on the topic. The aim was to provide an overview of the available imaging protocols and underline the items that appear most agreed upon and those that deserve further investigation.
METHODS
According to PRISMA guidelines, two independent reviewers conducted a systematic literature search with a pre-specified strategy in different databases. Peer-reviewed articles that utilized MRI techniques to assess GE in healthy volunteers (HVs) were included. The quality and the outcomes of the studies were reported and analyzed.
KEY RESULTS
The literature search yielded 30 studies (531 HVs, weighted mean age 27.4, weighted mean body mass index 23.0 kg/m ), T2-weighted sequences, balanced turbo field echo, and balanced gradient echo were evenly utilized, with volunteers in the supine position (74% of the studies). After overnight fasting, both liquid (56%) and mixed (44%) meals were equally utilized. Segmentation of the volumes was predominantly performed manually (63%) with a reported mean T50 ranging from 7 to 330 min.
CONCLUSIONS & INFERENCES
As observed in this systematic review, MRI is a flexible tool for assessing GE. Different protocols were analyzed, showing an equal capacity to assess the GE. However, many items in these protocols still require further investigation to obtain a common standard and increase this assessment quality.
Topics: Humans; Adult; Gastric Emptying; Meals; Magnetic Resonance Imaging; Fasting; Healthy Volunteers
PubMed: 35340100
DOI: 10.1111/nmo.14371 -
Drug Safety May 2024Glucagon-like peptide 1 receptor agonists (GLP1RAs) are used in the treatment of diabetes and obesity. Their slowing effect of gastric emptying might change oral drug...
BACKGROUND
Glucagon-like peptide 1 receptor agonists (GLP1RAs) are used in the treatment of diabetes and obesity. Their slowing effect of gastric emptying might change oral drug absorption, potentially affecting pharmacokinetics, particularly in the case of medications with a narrow therapeutic index.
PURPOSE
The purpose of this systematic review is to summarize data on drug-drug interactions between GLP1RAs and oral drugs.
DATA SOURCES
The PubMed and EMBASE databases were searched up to November, 1st 2023.
STUDY SELECTION
We selected pharmacokinetic studies of any injectable GLP1RA given with an oral medication, and product prescribing sheets reporting data without access to the original study.
DATA EXTRACTION
Two authors independently extracted the data.
DATA SYNTHESIS
Twenty-two reports and six prescribing sheets were included. Treatment with GLP1RAs resulted in unaffected or reduced C and delayed t of drugs with high solubility and permeability (warfarin, contraceptive pills, acetaminophen), drugs with high solubility and low permeability (angiotensin converting enzyme inhibitors), drugs with low solubility and high permeability (statins) and drugs with low solubility and permeability (digoxin). However, the use of GLP1RAs did not exert clinically significant changes in the AUC or differences in clinically relevant endpoints.
LIMITATIONS
The major limitations of the studies that are included in this systematic review are the enrollment of healthy subjects and insufficient data in conditions that might affect pharmacokinetics (e.g., kidney dysfunction).
CONCLUSIONS
To conclude, reduced C and delayed t of drugs co-administered with GLP1RAs are consistent with the known delayed gastric output by the latter. Nevertheless, the overall drug exposure was not considered clinically significant. Dose adjustments are probably not required for simultaneous use of GLP1RAs with oral medications. Still, results should be carefully generalized to cases of background kidney dysfunction or when using drugs with narrow therapeutic index. The study is registered in PROSPERO: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42022332339 .
Topics: Humans; Angiotensin-Converting Enzyme Inhibitors; Digoxin; Drug Interactions; Glucagon-Like Peptide 1; Warfarin
PubMed: 38273155
DOI: 10.1007/s40264-023-01392-3 -
Langenbeck's Archives of Surgery Aug 2023Reducing clinically relevant post-operative pancreatic fistula (CR-POPF) incidence after pancreatic resections has been a topic of great academic interest. Optimizing... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Reducing clinically relevant post-operative pancreatic fistula (CR-POPF) incidence after pancreatic resections has been a topic of great academic interest. Optimizing post-operative drain management is a potential strategy in reducing this major complication.
METHODS
Studies involving pancreatic resections, including both pancreaticoduodenectomy (PD) and distal pancreatic resections (DP), with intra-operative drain placement were screened. Early drain removal was defined as removal before or on the 3rd post-operative day (POD) while late drain removal was defined as after the 3rd POD. The primary outcome was CR-POPF, International Study Group of Pancreatic Surgery (ISGPS) Grade B and above. Secondary outcomes were all complications, severe complications, post-operative haemorrhage, intra-abdominal infections, delayed gastric emptying, reoperation, length of stay, readmission, and mortality.
RESULTS
Nine studies met the inclusion criteria and were included for analysis. The studies had a total of 8574 patients, comprising 1946 in the early removal group and 6628 in the late removal group. Early drain removal was associated with a significantly lower risk of CR-POPF (OR: 0.24, p < 0.01). Significant reduction in risk of post-operative haemorrhage (OR: 0.55, p < 0.01), intra-abdominal infection (OR: 0.35, p < 0.01), re-admission (OR: 0.63, p < 0.01), re-operation (OR: 0.70, p = 0.03), presence of any complications (OR: 0.46, p < 0.01), and reduced length of stay (SMD: -0.75, p < 0.01) in the early removal group was also observed.
CONCLUSION
Early drain removal is associated with significant reductions in incidence of CR-POPF and other post-operative complications. Further prospective randomised trials in this area are recommended to validate these findings.
Topics: Humans; Pancreatectomy; Device Removal; Pancreas; Postoperative Complications; Postoperative Hemorrhage; Intraabdominal Infections
PubMed: 37587225
DOI: 10.1007/s00423-023-03053-6 -
BMC Cancer Apr 2023Pancreaticoduodenectomy (PD) is a complex and traumatic abdominal surgery with a high risk of postoperative complications. Nutritional support, including immunonutrition... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Pancreaticoduodenectomy (PD) is a complex and traumatic abdominal surgery with a high risk of postoperative complications. Nutritional support, including immunonutrition (IMN) with added glutamine, arginine, and ω-3 polyunsaturated fatty acids, can improve patients' prognosis by regulating postoperative inflammatory response. However, the effects of IMN on PD patients' outcomes require further investigation.
METHODS
PMC, EMbase, web of science databases were used to search literatures related to IMN and PD. Data such as length of hospital stay, infectious complications, non-infectious complications, postoperative pancreatic fistula (POPF), delayed gastric emptying (DGE), mortality, systemic inflammatory response syndrome (SIRS) duration, IL-6, and C-reactive protein (CRP) were extracted, and meta-analyses were performed on these data to study their pooled results, heterogeneity, and publication bias.
RESULTS
This meta-analysis involved 10 studies and a total of 572 patients. The results showed that the use of IMN significantly reduced the length of hospital stay for PD patients (MD = -2.31; 95% CI = -4.43, -0.18; P = 0.03) with low heterogeneity. Additionally, the incidence of infectious complications was significantly reduced (MD = 0.42; 95% CI = 0.18, 1.00, P = 0.05), with low heterogeneity after excluding one study. However, there was no significant impact on non-infectious complications, the incidence of POPF and DGE, mortality rates, duration of SIRS, levels of IL-6 and CRP.
CONCLUSION
The use of IMN has been shown to significantly shorten hospital stays and decrease the frequency of infectious complications in PD patients. Early implementation of IMN is recommended for those undergoing PD. However, further research is needed to fully assess the impact of IMN on PD patients through larger and higher-quality studies.
Topics: Humans; Pancreaticoduodenectomy; Immunonutrition Diet; Interleukin-6; Pancreatectomy; Postoperative Complications; Pancreatic Fistula; Length of Stay
PubMed: 37069556
DOI: 10.1186/s12885-023-10820-7 -
HPB : the Official Journal of the... Sep 2021Preoperative chemo- or chemoradiotherapy is recommended for borderline-resectable pancreatic cancer. The aim of this study was to determine the impact of preoperative... (Meta-Analysis)
Meta-Analysis Review
The effect of preoperative chemotherapy and chemoradiotherapy on pancreatic fistula and other surgical complications after pancreatic resection: a systematic review and meta-analysis of comparative studies.
BACKGROUND
Preoperative chemo- or chemoradiotherapy is recommended for borderline-resectable pancreatic cancer. The aim of this study was to determine the impact of preoperative therapy on surgical complications in patients with resected pancreatic cancer.
METHODS
This systematic review and meta-analysis included studies reporting on the rate of surgical complications after preoperative chemo- or chemoradiotherapy versus immediate surgery in pancreatic cancer patients. The primary endpoint was the rate of grade B/C POPF. Pooled odds ratios were calculated using random-effects models.
RESULTS
Forty-one comparative studies including 25,389 patients were included. Vascular resections were more often performed after preoperative therapy (29.4% vs. 15.7%, p < 0.001). Preoperative therapy was associated with a lower rate of grade B/C POPF as compared to immediate surgery (pooled OR 0.47, 95%CI 0.38-0.58). This reduction was mostly obtained by preoperative chemoradiotherapy (OR 0.46, 95%CI 0.29-0.73), but not by preoperative chemotherapy alone (OR 0.83, 95%CI 0.59-1.16). No difference was demonstrated for major morbidity, mortality, postpancreatectomy haemorrhage, delayed gastric emptying and overall morbidity.
CONCLUSION
Preoperative chemo- and chemoradiotherapy in patients with pancreatic cancer appears to be safe with respect to POPF and other surgical complications as compared to immediate surgery. The reduced rate of POPF appears to be attributable to preoperative chemoradiation.
Topics: Chemoradiotherapy; Humans; Neoadjuvant Therapy; Pancreatectomy; Pancreatic Fistula; Pancreatic Neoplasms; Postoperative Complications
PubMed: 34099372
DOI: 10.1016/j.hpb.2021.04.027 -
The Cochrane Database of Systematic... Feb 2023Upper endoscopy is the definitive treatment for upper gastrointestinal haemorrhage (UGIH). However, up to 13% of people who undergo upper endoscopy will have incomplete... (Review)
Review
BACKGROUND
Upper endoscopy is the definitive treatment for upper gastrointestinal haemorrhage (UGIH). However, up to 13% of people who undergo upper endoscopy will have incomplete visualisation of the gastric mucosa at presentation. Erythromycin acts as a motilin receptor agonist in the upper gastrointestinal (GI) tract and increases gastric emptying, which may lead to better quality of visualisation and improved treatment effectiveness. However, there is uncertainty about the benefits and harms of erythromycin in UGIH.
OBJECTIVES
To evaluate the benefits and harms of erythromycin before endoscopy in adults with acute upper gastrointestinal haemorrhage, compared with any other treatment or no treatment/placebo.
SEARCH METHODS
We used standard, extensive Cochrane search methods. The latest search date was 15 October 2021.
SELECTION CRITERIA
We included randomised controlled trials (RCTs) that investigated erythromycin before endoscopy compared to any other treatment or no treatment/placebo before endoscopy in adults with acute UGIH.
DATA COLLECTION AND ANALYSIS
We used standard Cochrane methods. Our primary outcomes were 1. UGIH-related mortality and 2. serious adverse events. Our secondary outcomes were 1. all-cause mortality, 2. visualisation of gastric mucosa, 3. non-serious adverse events, 4. rebleeding, 5. blood transfusion, and 5. rescue invasive intervention. We used GRADE criteria to assess the certainty of the evidence for each outcome. MAIN RESULTS: We included 11 RCTs with 878 participants. The mean age ranged from 53.13 years to 64.5 years, and most participants were men (72.3%). One RCT included only non-variceal haemorrhage, one included only variceal haemorrhage, and eight included both aetiologies. We defined short-term outcomes as those occurring within one week of initial endoscopy. Erythromycin versus placebo Three RCTs (255 participants) compared erythromycin with placebo. There were no UGIH-related deaths. The evidence is very uncertain about the short-term effects of erythromycin compared with placebo on serious adverse events (risk difference (RD) -0.01, 95% confidence interval (CI) -0.04 to 0.02; 3 studies, 255 participants; very low certainty), all-cause mortality (RD 0.00, 95% CI -0.03 to 0.03; 3 studies, 255 participants; very low certainty), non-serious adverse events (RD 0.01, 95% CI -0.03 to 0.05; 3 studies, 255 participants; very low certainty), and rebleeding (risk ratio (RR) 0.63, 95% CI 0.13 to 2.90; 2 studies, 195 participants; very low certainty). Erythromycin may improve gastric mucosa visualisation (mean difference (MD) 3.63 points on 16-point ordinal scale, 95% CI 2.20 to 5.05; higher MD means better visualisation; 2 studies, 195 participants; low certainty). Erythromycin may also result in a slight reduction in blood transfusion (MD -0.44 standard units of blood, 95% CI -0.86 to -0.01; 3 studies, 255 participants; low certainty). Erythromycin plus nasogastric tube lavage versus no intervention/placebo plus nasogastric tube lavage Six RCTs (408 participants) compared erythromycin plus nasogastric tube lavage with no intervention/placebo plus nasogastric tube lavage. There were no UGIH-related deaths and no serious adverse events. The evidence is very uncertain about the short-term effects of erythromycin plus nasogastric tube lavage compared with no intervention/placebo plus nasogastric tube lavage on all-cause mortality (RD -0.02, 95% CI -0.08 to 0.03; 3 studies, 238 participants; very low certainty), visualisation of the gastric mucosa (standardised mean difference (SMD) 0.48 points on 10-point ordinal scale, 95% CI 0.10 to 0.85; higher SMD means better visualisation; 3 studies, 170 participants; very low certainty), non-serious adverse events (RD 0.00, 95% CI -0.05 to 0.05; 6 studies, 408 participants; very low certainty), rebleeding (RR 1.13, 95% CI 0.63 to 2.02; 1 study, 169 participants; very low certainty), and blood transfusion (MD -1.85 standard units of blood, 95% CI -4.34 to 0.64; 3 studies, 180 participants; very low certainty). Erythromycin versus nasogastric tube lavage Four RCTs (287 participants) compared erythromycin with nasogastric tube lavage. There were no UGIH-related deaths and no serious adverse events. The evidence is very uncertain about the short-term effects of erythromycin compared with nasogastric tube lavage on all-cause mortality (RD 0.02, 95% CI -0.05 to 0.08; 3 studies, 213 participants; very low certainty), visualisation of the gastric mucosa (RR 1.19, 95% CI 0.79 to 1.79; 2 studies, 198 participants; very low certainty), non-serious adverse events (RD -0.10, 95% CI -0.34 to 0.13; 3 studies, 213 participants; very low certainty), rebleeding (RR 0.77, 95% CI 0.40 to 1.49; 1 study, 169 participants; very low certainty), and blood transfusion (median 2 standard units of blood, interquartile range 0 to 4 in both groups; 1 study, 169 participants; very low certainty). Erythromycin plus nasogastric tube lavage versus metoclopramide plus nasogastric tube lavage One RCT (30 participants) compared erythromycin plus nasogastric tube lavage with metoclopramide plus nasogastric tube lavage. The evidence is very uncertain about the effects of erythromycin plus nasogastric tube lavage on all the reported outcomes (serious adverse events, visualisation of gastric mucosa, non-serious adverse events, and blood transfusion).
AUTHORS' CONCLUSIONS
We are unsure if erythromycin before endoscopy in people with UGIH has any clinical benefits or harms. However, erythromycin compared with placebo may improve gastric mucosa visualisation and result in a slight reduction in blood transfusion.
Topics: Female; Humans; Male; Middle Aged; Endoscopy; Erythromycin; Gastrointestinal Hemorrhage; Metoclopramide; Treatment Outcome
PubMed: 36723439
DOI: 10.1002/14651858.CD013176.pub2 -
Gut and Liver Sep 2020Ghrelin agonists are emerging prokinetic agents for treating gastroparesis. Although recent clinical trials have demonstrated their efficacy in patients with diabetic... (Meta-Analysis)
Meta-Analysis
BACKGROUND/AIMS
Ghrelin agonists are emerging prokinetic agents for treating gastroparesis. Although recent clinical trials have demonstrated their efficacy in patients with diabetic gastroparesis (DG), the impact of such agents on symptoms and gastric dysmotility remains unclear. We performed a systematic review and meta-analysis to evaluate the efficacy and safety of ghrelin agonists in patients with DG.
METHODS
A search of common electronic databases (MEDLINE, Embase, and Cochrane Central Register of Controlled Trials) was preformed, using keyword combinations that referenced ghrelin and DG and retrieving all eligible randomized controlled trials (RCTs) of ghrelin agonists versus placebo in patients with DG. The primary outcome measure was the change in patient-reported overall gastroparesis symptom scores. Secondary outcomes included the change in gastric emptying time, specific symptoms related to gastroparesis, and adverse events. A random-effects model was applied to all study outcomes. Heterogeneity among studies was determined by the chi-square test and I statistics.
RESULTS
We selected six RCTs of patients with DG (n=557) for meta-analysis. Ghrelin agonist administration (vs placebo) significantly improved overall gastroparesis symptoms (standardized mean difference, -0.34; 95% confidence interval, -0.56 to -0.13) and significantly improved symptoms related to gastroparesis, including nausea, vomiting, early satiety, and abdominal pain. Adverse events recorded for ghrelin agonists and placebo did not differ significantly. There was no significant heterogeneity among eligible studies.
CONCLUSIONS
Compared with placebo, ghrelin agonists are effective and well-tolerated for the treatment of DG.
Topics: Abdominal Pain; Diabetes Mellitus; Gastric Emptying; Gastroparesis; Ghrelin; Humans; Vomiting
PubMed: 31816671
DOI: 10.5009/gnl19103 -
Diabetes, Obesity & Metabolism Nov 2021To present an overview of exendin(9-39)NH usage as a scientific tool in humans and provide recommendations for dosage and infusion regimes. (Review)
Review
AIM
To present an overview of exendin(9-39)NH usage as a scientific tool in humans and provide recommendations for dosage and infusion regimes.
METHODS
We systematically searched the literature on exendin(9-39)NH and included for review 44 clinical studies reporting use of exendin(9-39)NH in humans.
RESULTS
Exendin(9-39)NH binds to the orthosteric binding site of the glucagon-like peptide-1 (GLP-1) receptor with high affinity. The plasma elimination half-life of exendin(9-39)NH after intravenous administration is ~30 minutes, requiring ~2.5 hours of constant infusion before steady-state plasma concentrations can be expected. Studies utilizing infusions with exendin(9-39)NH in humans have applied varying regimens (priming with a bolus or constant infusion) and dosages (continuous infusion rate range 30-900 pmol/kg/min) with subsequent differences in effects. Administration of exendin(9-39)NH in healthy individuals, patients with diabetes, obese patients, and patients who have undergone bariatric surgery significantly increases fasting and postprandial levels of glucose and glucagon, but has inconsistent effects on circulating concentrations of insulin and C-peptide, gastric emptying, appetite sensations, and food intake. Importantly, exendin(9-39)NH induces secretion of all L cell products (ie, in addition to GLP-1, also peptide YY, glucagon-like peptide-2, oxyntomodulin, and glicentin) complicating use of exendin(9-39)NH as a tool to study the isolated effect of GLP-1.
CONCLUSIONS
Exendin(9-39)NH is selective for the GLP-1 receptor, with numerous and complex whole-body effects. To obtain GLP-1 receptor blockade in humans, we recommend an initial high-dose infusion, followed by a continuous infusion rate aiming at a ratio of exendin(9-39)NH to GLP-1 of 2000:1. Highlights Exendin(9-39)NH is a competitive antagonist of the human GLP-1 receptor. Exendin(9-39)NH has been used as a tool to delineate human GLP-1 physiology since 1998. Exendin(9-39)NH induces secretion of GLP-1 and other L cell products. Reported effects of exendin(9-39)NH on insulin levels and food intake are inconsistent. Here, we provide recommendations for the use of exendin(9-39)NH in clinical studies.
Topics: C-Peptide; Glucagon; Glucagon-Like Peptide 1; Glucagon-Like Peptide-1 Receptor; Humans; Insulin; Peptide Fragments; Receptors, Glucagon
PubMed: 34351033
DOI: 10.1111/dom.14507 -
Langenbeck's Archives of Surgery Aug 2020The effect of immunonutrition (IM) on postoperative outcomes has been investigated in gastrointestinal cancer surgery; however, strong evidence regarding IM in partial... (Meta-Analysis)
Meta-Analysis
BACKGROUND
The effect of immunonutrition (IM) on postoperative outcomes has been investigated in gastrointestinal cancer surgery; however, strong evidence regarding IM in partial pancreatoduodenectomy (PD) is lacking. This study evaluated the effect of IM on short-term outcomes in patients undergoing PD.
METHODS
A systematic literature review of randomized controlled trials was conducted to identify the studies investigating the IM effect on outcomes in PD. Random-effects meta-analyses were conducted to calculate the pooled risk ratio (RR). Studies were evaluated using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach.
RESULTS
Five studies were included in the meta-analysis. IM was associated with a lower incidence of overall complications (RR 0.74; 95% confidence interval (CI) 0.58, 0.94; P = 0.01; I = 0%) and infectious complications (RR 0.60; 95% CI 0.42, 0.84; P = 0.003; I = 0%). However, no significant association was noted in the incidence of major complications (RR 0.68; 95% CI 0.41, 1.12; P = 0.13), mortality (RR 0.79; 95% CI 0.16, 3.99; P = 0.78), postoperative pancreatic fistula (RR 0.92, 95% CI 0.59, 1.46; P = 0.74), and delayed gastric emptying (RR 1.09; 95% CI 0.55, 2.15; P = 0.81).
CONCLUSIONS
IM administration in PD can prevent the incidence of overall and infectious complications postoperatively (GRADE recommendation: moderate). However, IM has no impact on major complications, mortality, and PD-specific complications (GRADE recommendation: low).
Topics: Enhanced Recovery After Surgery; Humans; Immunomodulation; Nutritional Support; Pancreaticoduodenectomy; Postoperative Complications
PubMed: 32613544
DOI: 10.1007/s00423-020-01916-w -
Clinical Gastroenterology and... Jan 2020Gastric emptying (GE) is involved in the regulation of appetite. We compared times of GE after different bariatric endoscopic and surgical interventions and associations... (Meta-Analysis)
Meta-Analysis
BACKGROUND & AIMS
Gastric emptying (GE) is involved in the regulation of appetite. We compared times of GE after different bariatric endoscopic and surgical interventions and associations with weight loss.
METHODS
We performed a comprehensive search of publication databases, through September 14, 2018, for randomized and nonrandomized studies reporting outcomes of weight-loss surgeries. Two independent reviewers selected and appraised studies. The outcome of interest was GE T (min), measured before and after the procedure. A random-effects model was used to pool the mean change in T (min) after the intervention. We performed a meta-regression analysis to find associations between GE and weight loss. Heterogeneity was calculated using the I statistic. Methodologic quality was assessed.
RESULTS
From 762 citations, the following studies were included in our analysis: 9 sleeve gastrectomies, 5 intragastric balloons, and 5 antral botulinum toxins. After sleeve gastrectomy, the pooled mean reduction in GE T at 3 months was 29.2 minutes (95% CI, 40.9-17.5 min; I = 91%). Fluid-filled balloons increased GE T by 116 minutes (95% CI, 29.4-203.4 min; I = 58.6%). Air-filled balloons did not produce a statistically significant difference in GE T. Antral botulinum injections increased GE T by 9.6 minutes (95% CI, 2.8-16.4 min; I = 13.3%). Placebo interventions reduced GE T by 6.3 minutes (95% CI, 10-2.6 min). Changes in GE were associated with weight loss after sleeve gastrectomy and intragastric balloons, but not botulinum toxin injections.
CONCLUSIONS
In a systematic review and meta-analysis, we found that sleeve gastrectomy reduced GE T whereas fluid-filled balloons significantly increased GE T. Air-filled balloons do not significantly change the time of GE, which could account for their low efficacy. Antral botulinum toxin injections produced small temporary increases in GE time, which were not associated with weight loss. Changes in GE time after surgical and endoscopic bariatric interventions correlated with weight loss and might be used to select interventions, based on patients' physiology.
Topics: Bariatric Surgery; Endoscopy, Gastrointestinal; Gastric Emptying; Humans; Obesity, Morbid; Randomized Controlled Trials as Topic; Treatment Outcome; Weight Loss
PubMed: 30954712
DOI: 10.1016/j.cgh.2019.03.047