This Month in Gastroenterology

Abstract

Complications of ColonoscopyColonoscopy is a key examination for the investigation of several abdominal symptoms. Colonoscopy is also recommended for the screening for and prevention of colorectal cancer, through detection of polyps and polypectomy. Several case series have reported on the incidence of complications of colonoscopy, like bleeding or perforation, but no data are available on the incidence and risk factors of colonoscopy from a population-based perspective.In this issue of Gastroenterology, Rabeneck et al report on a population-based cohort of subjects undergoing outpatient colonoscopy in 4 Canadian provinces over a 12-month period. All individuals 50–75 years old undergoing outpatient colonoscopy during this period were identified using the physicians claims database. Because the study sought subjects undergoing screening colonoscopy, those with a previous diagnosis of colorectal cancer, inflammatory bowel disease, or colonic resection were excluded. Therapeutic colonoscopies for endoscopic hemostasis, stent insertion, or reduction of a sigmoid volvulus were also excluded. Complications were identified by analysis of diagnostic codes for hospitalizations within the first 30 days after the colonoscopy, and by direct analysis of a representative selection of hospital charts. The impact of risk factors related to the patient, endoscopist, and procedure type and setting on the prevalence of complications was analyzed.A total of 97,091 persons (54.2% women) between the age of 50 and 75 underwent outpatient colonoscopy in the 4 provinces during the evaluation period. A polypectomy was performed during 24.3% of the procedures. The pooled rates of colonoscopy-related bleeding and perforation per 1,000 procedures were 1.64 ± 0.36 and 0.85 ± 0.13, respectively (mean ± standard error). The estimated colonoscopy-related death rate per 1,000 colonoscopies was 0.074 ± 0.033.The combined outcome of bleeding or perforation was 6 times higher with a polypectomy. The risk for perforation was higher in older patients, those with increased comorbidity, and in those undergoing a polypectomy. The risk for bleeding was associated with male gender, greater age, and polypectomy. In addition, having the colonoscopy performed by a low-volume procedure endoscopist was associated with increased odds of bleeding or perforation. The risk of complications was significantly increased below a threshold of 300 colonoscopies per year (Figure 1). The setting (hospital or private office or clinic) and endoscopist's specialty (gastroenterologist, internist, surgeon, or other) did not significantly influence the risk of complications.This is the first study to report rates of bleeding and perforation and risk of death after colonoscopy based on data from usual clinical practice. The paper provides important information for patients undergoing colonoscopies, and for the endoscopist performing them, especially when polypectomy is indicated in male or elderly patients. The study was not able to evaluate other potentially relevant risk factors such as the number of polyps removed, use of aspirin or nonsteroidal drugs, or quality of bowel preparation.See page 1899.Drug-Induced Liver InjuryThe clinical presentation of drug-induced liver injury ranges from asymptomatic liver test abnormalities to symptomatic acute liver disease, prolonged jaundice and disability, or overt acute or subacute liver failure. The pathogenesis and risk factors of the condition are poorly understood, and it has been proposed that immunoallergic reactions or abnormalities in the metabolism of the agent underlie severe drug-induced liver injury in susceptible patients. More recently, it has become clear that, beside prescribed drugs, nonprescription drugs, including herbal remedies and over-the-counter dietary supplements may also cause “drug-induced” liver injury.Because of the relatively rarity of the condition and the variable nature of putative causal agents, a multicenter approach seems necessary to enhance current understanding of drug-induced liver injury. In 2003, 5 academic centers in the United States joined in a Drug-Induced Liver Injury Network, to prospectively identify and follow up patients with drug-induced liver injury. This observational, multicenter study enrolled patients with a strong clinical suspicion of a liver injury event that was caused by a medication or a dietary supplement. Patients with underlying immune-mediated liver disease, liver or bone marrow transplantation, or suspected acetaminophen liver toxicity were excluded. In this issue of Gastroenterology, Chalasani et al report on the first 300 patients enrolled into the study.All patients underwent a detailed medical and clinical history, including intake of agents and herbs, laboratory testing at inclusion, and at follow-up at 6, 12, and up to 24 months. Based on the dominant type of laboratory test abnormalities, patients were subdivided into hepatocellular, cholestatic, or mixed type drug-induced liver injury. In addition, the severity of the liver injury and the strength of the causal association between the implicated agent and the liver injury event was graded on 5-level scales.A total of 300 patients (60% women) were enrolled from 2004 to 2007, the majority (93%) adults; only 6% had preexisting liver disease. The median duration between first exposure to the implicated agent and recognition of drug-induced liver injury was 42 days. The implicated agent was a single prescription medication in 73%, a single or multiple dietary supplements in 9%, and a combination of agents in 18% of the patients. The most frequently implicated single prescription medications were antimicrobials (46%), central nervous system drugs (15%), immunomodulators (6%) and analgesics (5%). The likelihood of drug-induced liver injury as the reason for the liver injury considered definite in 32%, highly likely in 41%, probable in 14%, and possible in 10%. The dietary supplement classes most commonly associated with drug-induced liver injury included agents used for muscle building or weight loss.The drug-induced liver injury was hepatocellular in 57%, cholestatic in 23%, and mixed in 20%. Hepatocellular type drug-induced liver injury was more prevalent in younger and female subjects. The degree of severity of the liver disease was considered in 27%, moderate in 19%, moderate-hospitalized in 33%, severe in 15%, and resulting in death or liver transplantation in 6%. In multivariate analysis, the presence of diabetes (odds ratio, 2.69; 95% confidence interval, 1.14–6.45) and alcohol consumption (odds ratio, 0.3; 95% confidence interval, 0.15–0.76) and were independent predictors of severe drug-induced liver injury.Liver test abnormalities peaked on average between 1 (alkaline phosphatase) and 7 (bilirubin) days after recognition of potential drug-induced liver injury. The median duration for a 50% decrease in bilirubin from peak value was 13 days. At 6-month follow-up, 13.6% of enrolled patients were considered to have chronic drug-induced liver injury, 8% had died, and 2.1% had undergone liver transplantation. Chronic or severe evolution of drug-induced liver injury was not significantly influenced by the implicated agent, patient age, or clinical pattern (Figure 2). The mortality was significantly higher in patients with hepatocellular drug-induced liver injury with peak serum total bilirubin level of >2.5 mg/dL.Figure 2Outcomes of drug-induced liver injury according to implicated agent, patient age, and pattern. No significant differences were observed in relation to these characteristics.View Large Image Figure ViewerDownload Hi-res image Download (PPT)The present study is the first to provide a prospective assessment and follow-up of patients with suspected drug-induced liver injury. The study shows that a multitude of agents may induce liver injury, and helps to identify additional agents that can be associated with drug-induced liver injury. The study identified coexistent diabetes mellitus as an independent risk factor for more severe liver injury; alcohol consumption was a negative predictive factor. Drug-induced liver injury with jaundice from hepatocellular liver injury carries a high mortality rate. Further studies are needed to improve understanding of the underlying pathogenesis and to develop early detection strategies.See page 1924.A DNA Non-Repair Signature Distinguishing MYH-Associated PolypsThe human mutated Y homolog (MYH) gene encodes a DNA glycosylase that is operative in base excision repair of DNA to repair oxidative damage, principally 8-oxoG:A nucleotide pairs. Germline mutations within the MYH gene cause an autosomal-recessive condition that can be initially confused with the autosomal-dominant familial adenomatous polyposis (FAP), now termed MYH-associated polyposis. The type of polyps in MYH-associated polyposis has been generally thought limited to adenomas. Additionally, the absence of MYH function might provide a unique signature in polyps that can help to distinguish this condition.In the study by Boparai et al, 17 patients with biallelic MYH mutation were analyzed for types of polyps present, and subsequently polyp DNA was assessed for the type of mutation in the APC mutation cluster region (codons 1250–1550) and K-ras codon 12. Eight of 17 (47%) unrelated patients with MYH-associated polyposis had ≥1 hyperplastic polyp (HP) and/or sessile serrated adenoma (SSA) in addition to tubular adenomas, with 3 patients having >10 HPs and/or SSAs. Tubular adenomas from MYH-associated polyposis patients showed exclusively G:C to T:A transversions, with none in control tubular adenomas (Table 1). Additionally, nearly all K-ras mutations detected in adenomas, HPs, and SSAs from MYH-associated polyposis patients were G:C to T:A transversions (48/51 polyps [94%]) compared with a more predictable distribution for controls (2/7 polyps [29%]; Table 1).Table 1Comparison of Distribution of G:C to T:A Transversions in the APC-Mutation Cluster Region and K-ras Codon 12 Mutations in MYH-Associated Polyps Versus ControlsSomatic mutation spectrumPatients with MYH-associated polyposis (n = 8)Control groupP valuePolypsAD (n = 22)HP (n = 63)SSA (n = 10)AD (n = 17)HP (n = 24)SSA (n = 17)APC mutation900700G:C → T:A9 (100%)00000<0.0001⁎Adenomas from MYH-associated polyposis patients versus control adenomas.K-ras mutation5456452<0.0001⁎⁎Hyperplastic polyps (HPs) and sessile serrated adenomas (SSAs) from MYH-associated polyposis patients versus control HPs/SSAs.G:C → T:A5 (100%)42 (93%)6 (100%)3 (75%)1 (20%)1 (50%)<0.0001⁎⁎Hyperplastic polyps (HPs) and sessile serrated adenomas (SSAs) from MYH-associated polyposis patients versus control HPs/SSAs. Adenomas from MYH-associated polyposis patients versus control adenomas. Hyperplastic polyps (HPs) and sessile serrated adenomas (SSAs) from MYH-associated polyposis patients versus control HPs/SSAs. Open table in a new tab The study indicates that the spectrum of polyps in MYH-associated polyposis patients is not exclusively adenomas, but quite commonly includes HPs and SSAs in its phenotype. Furthermore, these polyps demonstrate G:C to T:A transversions consistent with loss of MYH function that occurs markedly less frequently in polyps from patients without MYH-associated polyposis, and may be able to help distinguish this condition from FAP, hyperplastic polyposis, and sporadic cases.See page 2014.Glucagon Receptor in the Liver: More Than Glucose Homeostasis and a Matter of Hepatocyte SurvivalGlucagon is a pancreatic-derived hormone that stimulates hepatic gluconeogenesis and glycogenolysis and is counter-regulatory to the hormone insulin. Glucagon exerts its cellular function through the glucagon receptor (Gcgr). Mice lacking this receptor exhibit modest fasting hypoglycemia, reduced adiposity, and enhanced insulin sensitivity. Glucagon-like peptides GLP-1 and GLP-2 regulate glucose homeostasis and nutrient absorption, respectively, but in addition regulate pancreatic β-cell islet and enterocyte apoptosis through their distinct G-protein–coupled receptors, respectively. The role of the Gcgr on hepatocyte survival has not been previously examined.In the study by Sinclair et al, Gcgr−/− mice were utilized and compared with wild-type Gcgr mice. Isolated wild-type mouse hepatocytes (and BHK fibroblasts transfected with the rat Gcgr) treated with the Jo2 antibody to activate the Fas apoptotic pathway showed characteristic signs of apoptosis at the cell level, including cleaved caspase-3, all of which were attenuated after treatment with glucagon. Although glucagon caused activation of PKA and PI3K pathways and reduced MAPK phosphorylation, inhibition of these pathways did not interfere with glucagon's ability to attenuate Jo2-induced apoptosis. Glucagon also stimulates cAMP accumulation and Epac, a family of cAMP-regulated nucleotide exchange factors that act independently of PKA, and stimulation of Epac mimicked the effects of glucagons on apoptosis. Gcgr−/− mice had more rapid morbidity and increased mortality than Gcgr+/+ mice in response to proapoptotic Jo2 treatment, and demonstrated worse liver histology, higher levels of serum transaminases, and more extensive cleaved caspase-3 immunopositivity. Viral transfection of Gcgr into isolated Gcgr−/− mouse hepatocytes further demonstrated the antiapoptotic actions of glucagon, showing a direct role for Gcgr. In vivo viral transfection of Gcgr into Gcgr−/− mice demonstrated a reduction in plasma glucagons levels and an increase in plasma glucose levels compared with control mice, but also a significant reduction in hepatic injury after Jo2 treatment and a marked reduction in cleaved caspase-3 in their livers (Figure 3). Additionally, exogenous glucagons administration to wild-type mice reduced liver injury and increased hepatocyte survival.Figure 3Adenovirus rescue of the glucagon receptor in Gcgr−/− mice attenuates Jo2-induced apoptosis. (A) Fed-state levels of plasma glucagon in Gcgr−/− mice 5 days after intravenous administration of 1 × 109 pfu of either Ad-LacZ (control), or Ad-rGcgr (recombinant glucagons receptor); glucagon levels in nontransduced Gcgr−/− mice are also shown. Data are mean values ± standard error of the mean (n = 6–8 mice per group). ***P < .001, Ad-LacZ versus Ad-Gcgr–transduced Gcgr−/− mice or Gcgr+/+ mice. (B, C). Five days after administration of adenovirus as indicated in A, Gcgr−/− mice were injected with 10 μg Jo2 (n = 6–8 mice per group). After 4 hours, liver samples were taken for analysis. (B) Individual hepatic apoptosis scores were assessed. Horizontal line, median. (C) Immunohistochemical and quantitative detection of cleaved caspase-3 in liver sections from either Ad-LacZ– or Ad-rGcgr–infected Gcgr−/− mice after Jo2 administration (original magnification, 50×).View Large Image Figure ViewerDownload Hi-res image Download (PPT)This study indicates that glucagon directly reduces hepatocyte apoptosis and loss of Gcgr signaling increases hepatocyte susceptibility to liver injury. Thus, the glucose homeostatic roles of glucagon are extended to encompass hepatocyte survival.See page 2096. Complications of ColonoscopyColonoscopy is a key examination for the investigation of several abdominal symptoms. Colonoscopy is also recommended for the screening for and prevention of colorectal cancer, through detection of polyps and polypectomy. Several case series have reported on the incidence of complications of colonoscopy, like bleeding or perforation, but no data are available on the incidence and risk factors of colonoscopy from a population-based perspective.In this issue of Gastroenterology, Rabeneck et al report on a population-based cohort of subjects undergoing outpatient colonoscopy in 4 Canadian provinces over a 12-month period. All individuals 50–75 years old undergoing outpatient colonoscopy during this period were identified using the physicians claims database. Because the study sought subjects undergoing screening colonoscopy, those with a previous diagnosis of colorectal cancer, inflammatory bowel disease, or colonic resection were excluded. Therapeutic colonoscopies for endoscopic hemostasis, stent insertion, or reduction of a sigmoid volvulus were also excluded. Complications were identified by analysis of diagnostic codes for hospitalizations within the first 30 days after the colonoscopy, and by direct analysis of a representative selection of hospital charts. The impact of risk factors related to the patient, endoscopist, and procedure type and setting on the prevalence of complications was analyzed.A total of 97,091 persons (54.2% women) between the age of 50 and 75 underwent outpatient colonoscopy in the 4 provinces during the evaluation period. A polypectomy was performed during 24.3% of the procedures. The pooled rates of colonoscopy-related bleeding and perforation per 1,000 procedures were 1.64 ± 0.36 and 0.85 ± 0.13, respectively (mean ± standard error). The estimated colonoscopy-related death rate per 1,000 colonoscopies was 0.074 ± 0.033.The combined outcome of bleeding or perforation was 6 times higher with a polypectomy. The risk for perforation was higher in older patients, those with increased comorbidity, and in those undergoing a polypectomy. The risk for bleeding was associated with male gender, greater age, and polypectomy. In addition, having the colonoscopy performed by a low-volume procedure endoscopist was associated with increased odds of bleeding or perforation. The risk of complications was significantly increased below a threshold of 300 colonoscopies per year (Figure 1). The setting (hospital or private office or clinic) and endoscopist's specialty (gastroenterologist, internist, surgeon, or other) did not significantly influence the risk of complications.This is the first study to report rates of bleeding and perforation and risk of death after colonoscopy based on data from usual clinical practice. The paper provides important information for patients undergoing colonoscopies, and for the endoscopist performing them, especially when polypectomy is indicated in male or elderly patients. The study was not able to evaluate other potentially relevant risk factors such as the number of polyps removed, use of aspirin or nonsteroidal drugs, or quality of bowel preparation.See page 1899. Colonoscopy is a key examination for the investigation of several abdominal symptoms. Colonoscopy is also recommended for the screening for and prevention of colorectal cancer, through detection of polyps and polypectomy. Several case series have reported on the incidence of complications of colonoscopy, like bleeding or perforation, but no data are available on the incidence and risk factors of colonoscopy from a population-based perspective. In this issue of Gastroenterology, Rabeneck et al report on a population-based cohort of subjects undergoing outpatient colonoscopy in 4 Canadian provinces over a 12-month period. All individuals 50–75 years old undergoing outpatient colonoscopy during this period were identified using the physicians claims database. Because the study sought subjects undergoing screening colonoscopy, those with a previous diagnosis of colorectal cancer, inflammatory bowel disease, or colonic resection were excluded. Therapeutic colonoscopies for endoscopic hemostasis, stent insertion, or reduction of a sigmoid volvulus were also excluded. Complications were identified by analysis of diagnostic codes for hospitalizations within the first 30 days after the colonoscopy, and by direct analysis of a representative selection of hospital charts. The impact of risk factors related to the patient, endoscopist, and procedure type and setting on the prevalence of complications was analyzed. A total of 97,091 persons (54.2% women) between the age of 50 and 75 underwent outpatient colonoscopy in the 4 provinces during the evaluation period. A polypectomy was performed during 24.3% of the procedures. The pooled rates of colonoscopy-related bleeding and perforation per 1,000 procedures were 1.64 ± 0.36 and 0.85 ± 0.13, respectively (mean ± standard error). The estimated colonoscopy-related death rate per 1,000 colonoscopies was 0.074 ± 0.033. The combined outcome of bleeding or perforation was 6 times higher with a polypectomy. The risk for perforation was higher in older patients, those with increased comorbidity, and in those undergoing a polypectomy. The risk for bleeding was associated with male gender, greater age, and polypectomy. In addition, having the colonoscopy performed by a low-volume procedure endoscopist was associated with increased odds of bleeding or perforation. The risk of complications was significantly increased below a threshold of 300 colonoscopies per year (Figure 1). The setting (hospital or private office or clinic) and endoscopist's specialty (gastroenterologist, internist, surgeon, or other) did not significantly influence the risk of complications. This is the first study to report rates of bleeding and perforation and risk of death after colonoscopy based on data from usual clinical practice. The paper provides important information for patients undergoing colonoscopies, and for the endoscopist performing them, especially when polypectomy is indicated in male or elderly patients. The study was not able to evaluate other potentially relevant risk factors such as the number of polyps removed, use of aspirin or nonsteroidal drugs, or quality of bowel preparation. See page 1899. Drug-Induced Liver InjuryThe clinical presentation of drug-induced liver injury ranges from asymptomatic liver test abnormalities to symptomatic acute liver disease, prolonged jaundice and disability, or overt acute or subacute liver failure. The pathogenesis and risk factors of the condition are poorly understood, and it has been proposed that immunoallergic reactions or abnormalities in the metabolism of the agent underlie severe drug-induced liver injury in susceptible patients. More recently, it has become clear that, beside prescribed drugs, nonprescription drugs, including herbal remedies and over-the-counter dietary supplements may also cause “drug-induced” liver injury.Because of the relatively rarity of the condition and the variable nature of putative causal agents, a multicenter approach seems necessary to enhance current understanding of drug-induced liver injury. In 2003, 5 academic centers in the United States joined in a Drug-Induced Liver Injury Network, to prospectively identify and follow up patients with drug-induced liver injury. This observational, multicenter study enrolled patients with a strong clinical suspicion of a liver injury event that was caused by a medication or a dietary supplement. Patients with underlying immune-mediated liver disease, liver or bone marrow transplantation, or suspected acetaminophen liver toxicity were excluded. In this issue of Gastroenterology, Chalasani et al report on the first 300 patients enrolled into the study.All patients underwent a detailed medical and clinical history, including intake of agents and herbs, laboratory testing at inclusion, and at follow-up at 6, 12, and up to 24 months. Based on the dominant type of laboratory test abnormalities, patients were subdivided into hepatocellular, cholestatic, or mixed type drug-induced liver injury. In addition, the severity of the liver injury and the strength of the causal association between the implicated agent and the liver injury event was graded on 5-level scales.A total of 300 patients (60% women) were enrolled from 2004 to 2007, the majority (93%) adults; only 6% had preexisting liver disease. The median duration between first exposure to the implicated agent and recognition of drug-induced liver injury was 42 days. The implicated agent was a single prescription medication in 73%, a single or multiple dietary supplements in 9%, and a combination of agents in 18% of the patients. The most frequently implicated single prescription medications were antimicrobials (46%), central nervous system drugs (15%), immunomodulators (6%) and analgesics (5%). The likelihood of drug-induced liver injury as the reason for the liver injury considered definite in 32%, highly likely in 41%, probable in 14%, and possible in 10%. The dietary supplement classes most commonly associated with drug-induced liver injury included agents used for muscle building or weight loss.The drug-induced liver injury was hepatocellular in 57%, cholestatic in 23%, and mixed in 20%. Hepatocellular type drug-induced liver injury was more prevalent in younger and female subjects. The degree of severity of the liver disease was considered in 27%, moderate in 19%, moderate-hospitalized in 33%, severe in 15%, and resulting in death or liver transplantation in 6%. In multivariate analysis, the presence of diabetes (odds ratio, 2.69; 95% confidence interval, 1.14–6.45) and alcohol consumption (odds ratio, 0.3; 95% confidence interval, 0.15–0.76) and were independent predictors of severe drug-induced liver injury.Liver test abnormalities peaked on average between 1 (alkaline phosphatase) and 7 (bilirubin) days after recognition of potential drug-induced liver injury. The median duration for a 50% decrease in bilirubin from peak value was 13 days. At 6-month follow-up, 13.6% of enrolled patients were considered to have chronic drug-induced liver injury, 8% had died, and 2.1% had undergone liver transplantation. Chronic or severe evolution of drug-induced liver injury was not significantly influenced by the implicated agent, patient age, or clinical pattern (Figure 2). The mortality was significantly higher in patients with hepatocellular drug-induced liver injury with peak serum total bilirubin level of >2.5 mg/dL.The present study is the first to provide a prospective assessment and follow-up of patients with suspected drug-induced liver injury. The study shows that a multitude of agents may induce liver injury, and helps to identify additional agents that can be associated with drug-induced liver injury. The study identified coexistent diabetes mellitus as an independent risk factor for more severe liver injury; alcohol consumption was a negative predictive factor. Drug-induced liver injury with jaundice from hepatocellular liver injury carries a high mortality rate. Further studies are needed to improve understanding of the underlying pathogenesis and to develop early detection strategies.See page 1924. The clinical presentation of drug-induced liver injury ranges from asymptomatic liver test abnormalities to symptomatic acute liver disease, prolonged jaundice and disability, or overt acute or subacute liver failure. The pathogenesis and risk factors of the condition are poorly understood, and it has been proposed that immunoallergic reactions or abnormalities in the metabolism of the agent underlie severe drug-induced liver injury in susceptible patients. More recently, it has become clear that, beside prescribed drugs, nonprescription drugs, including herbal remedies and over-the-counter dietary supplements may also cause “drug-induced” liver injury. Because of the relatively rarity of the condition and the variable nature of putative causal agents, a multicenter approach seems necessary to enhance current understanding of drug-induced liver injury. In 2003, 5 academic centers in the United States joined in a Drug-Induced Liver Injury Network, to prospectively identify and follow up patients with drug-induced liver injury. This observational, multicenter study enrolled patients with a strong clinical suspicion of a liver injury event that was caused by a medication or a dietary supplement. Patients with underlying immune-mediated liver disease, liver or bone marrow transplantation, or suspected acetaminophen liver toxicity were excluded. In this issue of Gastroenterology, Chalasani et al report on the first 300 patients enrolled into the study. All patients underwent a detailed medical and clinical history, including intake of agents and herbs, laboratory testing at inclusion, and at follow-up at 6, 12, and up to 24 months. Based on the dominant type of laboratory test abnormalities, patients were subdivided into hepatocellular, cholestatic, or mixed type drug-induced liver injury. In addition, the severity of the liver injury and the strength of the causal association between the implicated agent and the liver injury event was graded on 5-level scales. A total of 300 patients (60% women) were enrolled from 2004 to 2007, the majority (93%) adults; only 6% had preexisting liver disease. The median duration between first exposure to the implicated agent and recognition of drug-induced liver injury was 42 days. The implicated agent was a single prescription medication in 73%, a single or multiple dietary supplements in 9%, and a combination of agents in 18% of the patients. The most frequently implicated single prescription medications were antimicrobials (46%), central nervous system drugs (15%), immunomodulators (6%) and analgesics (5%). The likelihood of drug-induced liver injury as the reason for the liver injury considered definite in 32%, highly likely in 41%, probable in 14%, and possible in 10%. The dietary supplement classes most commonly associated with drug-induced liver injury included agents used for muscle building or weight loss. The drug-induced liver injury was hepatocellular in 57%, cholestatic in 23%, and mixed in 20%. Hepatocellular type drug-induced liver injury was more prevalent in younger and female subjects. The degree of severity of the liver disease was considered in 27%, moderate in 19%, moderate-hospitalized in 33%, severe in 15%, and resulting in death or liver transplantation in 6%. In multivariate analysis, the presence of diabetes (odds ratio, 2.69; 95% confidence interval, 1.14–6.45) and alcohol consumption (odds ratio, 0.3; 95% confidence interval, 0.15–0.76) and were independent predictors of severe drug-induced liver injury. Liver test abnormalities peaked on average between 1 (alkaline phosphatase) and 7 (bilirubin) days after recognition of potential drug-induced liver injury. The median duration for a 50% decrease in bilirubin from peak value was 13 days. At 6-month follow-up, 13.6% of enrolled patients were considered to have chronic drug-induced liver injury, 8% had died, and 2.1% had undergone liver transplantation. Chronic or severe evolution of drug-induced liver injury was not significantly influenced by the implicated agent, patient age, or clinical pattern (Figure 2). The mortality was significantly higher in patients with hepatocellular drug-induced liver injury with peak serum total bilirubin level of >2.5 mg/dL. The present study is the first to provide a prospective assessment and follow-up of patients with suspected drug-induced liver injury. The study shows that a multitude of agents may induce liver injury, and helps to identify additional agents that can be associated with drug-induced liver injury. The study identified coexistent diabetes mellitus as an independent risk factor for more severe liver injury; alcohol consumption was a negative predictive factor. Drug-induced liver injury with jaundice from hepatocellular liver injury carries a high mortality rate. Further studies are needed to improve understanding of the underlying pathogenesis and to develop early detection strategies. See page 1924. A DNA Non-Repair Signature Distinguishing MYH-Associated PolypsThe human mutated Y homolog (MYH) gene encodes a DNA glycosylase that is operative in base excision repair of DNA to repair oxidative damage, principally 8-oxoG:A nucleotide pairs. Germline mutations within the MYH gene cause an autosomal-recessive condition that can be initially confused with the autosomal-dominant familial adenomatous polyposis (FAP), now termed MYH-associated polyposis. The type of polyps in MYH-associated polyposis has been generally thought limited to adenomas. Additionally, the absence of MYH function might provide a unique signature in polyps that can help to distinguish this condition.In the study by Boparai et al, 17 patients with biallelic MYH mutation were analyzed for types of polyps present, and subsequently polyp DNA was assessed for the type of mutation in the APC mutation cluster region (codons 1250–1550) and K-ras codon 12. Eight of 17 (47%) unrelated patients with MYH-associated polyposis had ≥1 hyperplastic polyp (HP) and/or sessile serrated adenoma (SSA) in addition to tubular adenomas, with 3 patients having >10 HPs and/or SSAs. Tubular adenomas from MYH-associated polyposis patients showed exclusively G:C to T:A transversions, with none in control tubular adenomas (Table 1). Additionally, nearly all K-ras mutations detected in adenomas, HPs, and SSAs from MYH-associated polyposis patients were G:C to T:A transversions (48/51 polyps [94%]) compared with a more predictable distribution for controls (2/7 polyps [29%]; Table 1).Table 1Comparison of Distribution of G:C to T:A Transversions in the APC-Mutation Cluster Region and K-ras Codon 12 Mutations in MYH-Associated Polyps Versus ControlsSomatic mutation spectrumPatients with MYH-associated polyposis (n = 8)Control groupP valuePolypsAD (n = 22)HP (n = 63)SSA (n = 10)AD (n = 17)HP (n = 24)SSA (n = 17)APC mutation900700G:C → T:A9 (100%)00000<0.0001⁎Adenomas from MYH-associated polyposis patients versus control adenomas.K-ras mutation5456452<0.0001⁎⁎Hyperplastic polyps (HPs) and sessile serrated adenomas (SSAs) from MYH-associated polyposis patients versus control HPs/SSAs.G:C → T:A5 (100%)42 (93%)6 (100%)3 (75%)1 (20%)1 (50%)<0.0001⁎⁎Hyperplastic polyps (HPs) and sessile serrated adenomas (SSAs) from MYH-associated polyposis patients versus control HPs/SSAs. Adenomas from MYH-associated polyposis patients versus control adenomas. Hyperplastic polyps (HPs) and sessile serrated adenomas (SSAs) from MYH-associated polyposis patients versus control HPs/SSAs. Open table in a new tab The study indicates that the spectrum of polyps in MYH-associated polyposis patients is not exclusively adenomas, but quite commonly includes HPs and SSAs in its phenotype. Furthermore, these polyps demonstrate G:C to T:A transversions consistent with loss of MYH function that occurs markedly less frequently in polyps from patients without MYH-associated polyposis, and may be able to help distinguish this condition from FAP, hyperplastic polyposis, and sporadic cases.See page 2014. The human mutated Y homolog (MYH) gene encodes a DNA glycosylase that is operative in base excision repair of DNA to repair oxidative damage, principally 8-oxoG:A nucleotide pairs. Germline mutations within the MYH gene cause an autosomal-recessive condition that can be initially confused with the autosomal-dominant familial adenomatous polyposis (FAP), now termed MYH-associated polyposis. The type of polyps in MYH-associated polyposis has been generally thought limited to adenomas. Additionally, the absence of MYH function might provide a unique signature in polyps that can help to distinguish this condition. In the study by Boparai et al, 17 patients with biallelic MYH mutation were analyzed for types of polyps present, and subsequently polyp DNA was assessed for the type of mutation in the APC mutation cluster region (codons 1250–1550) and K-ras codon 12. Eight of 17 (47%) unrelated patients with MYH-associated polyposis had ≥1 hyperplastic polyp (HP) and/or sessile serrated adenoma (SSA) in addition to tubular adenomas, with 3 patients having >10 HPs and/or SSAs. Tubular adenomas from MYH-associated polyposis patients showed exclusively G:C to T:A transversions, with none in control tubular adenomas (Table 1). Additionally, nearly all K-ras mutations detected in adenomas, HPs, and SSAs from MYH-associated polyposis patients were G:C to T:A transversions (48/51 polyps [94%]) compared with a more predictable distribution for controls (2/7 polyps [29%]; Table 1). The study indicates that the spectrum of polyps in MYH-associated polyposis patients is not exclusively adenomas, but quite commonly includes HPs and SSAs in its phenotype. Furthermore, these polyps demonstrate G:C to T:A transversions consistent with loss of MYH function that occurs markedly less frequently in polyps from patients without MYH-associated polyposis, and may be able to help distinguish this condition from FAP, hyperplastic polyposis, and sporadic cases. See page 2014. Glucagon Receptor in the Liver: More Than Glucose Homeostasis and a Matter of Hepatocyte SurvivalGlucagon is a pancreatic-derived hormone that stimulates hepatic gluconeogenesis and glycogenolysis and is counter-regulatory to the hormone insulin. Glucagon exerts its cellular function through the glucagon receptor (Gcgr). Mice lacking this receptor exhibit modest fasting hypoglycemia, reduced adiposity, and enhanced insulin sensitivity. Glucagon-like peptides GLP-1 and GLP-2 regulate glucose homeostasis and nutrient absorption, respectively, but in addition regulate pancreatic β-cell islet and enterocyte apoptosis through their distinct G-protein–coupled receptors, respectively. The role of the Gcgr on hepatocyte survival has not been previously examined.In the study by Sinclair et al, Gcgr−/− mice were utilized and compared with wild-type Gcgr mice. Isolated wild-type mouse hepatocytes (and BHK fibroblasts transfected with the rat Gcgr) treated with the Jo2 antibody to activate the Fas apoptotic pathway showed characteristic signs of apoptosis at the cell level, including cleaved caspase-3, all of which were attenuated after treatment with glucagon. Although glucagon caused activation of PKA and PI3K pathways and reduced MAPK phosphorylation, inhibition of these pathways did not interfere with glucagon's ability to attenuate Jo2-induced apoptosis. Glucagon also stimulates cAMP accumulation and Epac, a family of cAMP-regulated nucleotide exchange factors that act independently of PKA, and stimulation of Epac mimicked the effects of glucagons on apoptosis. Gcgr−/− mice had more rapid morbidity and increased mortality than Gcgr+/+ mice in response to proapoptotic Jo2 treatment, and demonstrated worse liver histology, higher levels of serum transaminases, and more extensive cleaved caspase-3 immunopositivity. Viral transfection of Gcgr into isolated Gcgr−/− mouse hepatocytes further demonstrated the antiapoptotic actions of glucagon, showing a direct role for Gcgr. In vivo viral transfection of Gcgr into Gcgr−/− mice demonstrated a reduction in plasma glucagons levels and an increase in plasma glucose levels compared with control mice, but also a significant reduction in hepatic injury after Jo2 treatment and a marked reduction in cleaved caspase-3 in their livers (Figure 3). Additionally, exogenous glucagons administration to wild-type mice reduced liver injury and increased hepatocyte survival.This study indicates that glucagon directly reduces hepatocyte apoptosis and loss of Gcgr signaling increases hepatocyte susceptibility to liver injury. Thus, the glucose homeostatic roles of glucagon are extended to encompass hepatocyte survival.See page 2096. Glucagon is a pancreatic-derived hormone that stimulates hepatic gluconeogenesis and glycogenolysis and is counter-regulatory to the hormone insulin. Glucagon exerts its cellular function through the glucagon receptor (Gcgr). Mice lacking this receptor exhibit modest fasting hypoglycemia, reduced adiposity, and enhanced insulin sensitivity. Glucagon-like peptides GLP-1 and GLP-2 regulate glucose homeostasis and nutrient absorption, respectively, but in addition regulate pancreatic β-cell islet and enterocyte apoptosis through their distinct G-protein–coupled receptors, respectively. The role of the Gcgr on hepatocyte survival has not been previously examined. In the study by Sinclair et al, Gcgr−/− mice were utilized and compared with wild-type Gcgr mice. Isolated wild-type mouse hepatocytes (and BHK fibroblasts transfected with the rat Gcgr) treated with the Jo2 antibody to activate the Fas apoptotic pathway showed characteristic signs of apoptosis at the cell level, including cleaved caspase-3, all of which were attenuated after treatment with glucagon. Although glucagon caused activation of PKA and PI3K pathways and reduced MAPK phosphorylation, inhibition of these pathways did not interfere with glucagon's ability to attenuate Jo2-induced apoptosis. Glucagon also stimulates cAMP accumulation and Epac, a family of cAMP-regulated nucleotide exchange factors that act independently of PKA, and stimulation of Epac mimicked the effects of glucagons on apoptosis. Gcgr−/− mice had more rapid morbidity and increased mortality than Gcgr+/+ mice in response to proapoptotic Jo2 treatment, and demonstrated worse liver histology, higher levels of serum transaminases, and more extensive cleaved caspase-3 immunopositivity. Viral transfection of Gcgr into isolated Gcgr−/− mouse hepatocytes further demonstrated the antiapoptotic actions of glucagon, showing a direct role for Gcgr. In vivo viral transfection of Gcgr into Gcgr−/− mice demonstrated a reduction in plasma glucagons levels and an increase in plasma glucose levels compared with control mice, but also a significant reduction in hepatic injury after Jo2 treatment and a marked reduction in cleaved caspase-3 in their livers (Figure 3). Additionally, exogenous glucagons administration to wild-type mice reduced liver injury and increased hepatocyte survival. This study indicates that glucagon directly reduces hepatocyte apoptosis and loss of Gcgr signaling increases hepatocyte susceptibility to liver injury. Thus, the glucose homeostatic roles of glucagon are extended to encompass hepatocyte survival. See page 2096. Bleeding and Perforation After Outpatient Colonoscopy and Their Risk Factors in Usual Clinical PracticeGastroenterologyVol. 135Issue 6PreviewThe most widely quoted complication rates for colonoscopy are from case series performed by expert endoscopists. Our objectives were to evaluate the rates of bleeding, perforation, and death associated with outpatient colonoscopy and their risk factors in a population-based study. Full-Text PDF Causes, Clinical Features, and Outcomes From a Prospective Study of Drug-Induced Liver Injury in the United StatesGastroenterologyVol. 135Issue 6PreviewIdiosyncratic drug-induced liver injury (DILI) is among the most common causes of acute liver failure in the United States, accounting for approximately 13% of cases. A prospective study was begun in 2003 to recruit patients with suspected DILI and create a repository of biological samples for analysis. This report summarizes the causes, clinical features, and outcomes from the first 300 patients enrolled. Full-Text PDF Hyperplastic Polyps and Sessile Serrated Adenomas as a Phenotypic Expression of MYH-Associated PolyposisGastroenterologyVol. 135Issue 6PreviewMYH-associated polyposis (MAP) is a disorder caused by a bi-allelic germline MYH mutation, characterized by multiple colorectal adenomas. These adenomas typically harbor G:C→T:A transversions in the APC and K-ras genes caused by MYH deficiency. Occasional hyperplastic polyps (HPs) have been described in MAP patients but a causal relationship has never been investigated. We examined the presence of HPs and sessile serrated adenomas (SSAs) in 17 MAP patients and studied the occurrence of G:C→T:A transversions in the APC and K-ras gene in these polyps. Full-Text PDF Glucagon Receptor Signaling Is Essential for Control of Murine Hepatocyte SurvivalGastroenterologyVol. 135Issue 6PreviewGlucagon action in the liver is essential for control of glucose homeostasis and the counterregulatory response to hypoglycemia. Because receptors for the related peptides glucagon-like peptide-1 and glucagon-like peptide-2 regulate β-cell and enterocyte apoptosis, respectively, we examined whether glucagon receptor (Gcgr) signaling modulates hepatocyte survival. Full-Text PDF