Effect of Patient Age on Timing of Inpatient Esophagogastroduodenoscopy and Outcomes for Non-variceal Upper GI Bleeds.

Background Esophagogastroduodenoscopy (EGD) is typically performed within 24 hours of presentation for patients admitted to a hospital for patients presenting with a non-variceal upper gastrointestinal bleed (UGIB). To date, no studies have been performed to identify the impact of patient age on the timing of inpatient EGD and patient outcomes in non-variceal UGIB. Our aim was to assess the differences in the timing of EGD, blood transfusion requirements, development of hemorrhagic shock, development of acute renal failure, mortality, length of stay, and total hospital charges for patients aged 18-59 and those aged 60 and older. Methods Admissions for non-variceal UGIB were identified from the National (Nationwide) Inpatient Sample (NIS) database from 2016 and 2017. Patients who initially presented with hemorrhagic shock were excluded. Patients were divided into two age groups, those aged 18-59 and those aged 60 or older. We classified EGDs as early and delayed. Since the NIS database identifies days as midnight to midnight, we categorized early EGDs as those performed on day 0 and day 1. Delayed EGD were categorized as those performed on days 2 and 3. Multivariate logistic regression was performed on propensity-matched data to compare EGD timing, blood transfusion requirements, development of post-hospitalization hemorrhagic shock, development of acute renal failure, and mortality. The following patient and hospital variables were used in regression models: race, sex, insurance status, income quartile, mortality risk score, illness severity score, admission month, admission day, type of admission, region, bed size, and hospital teaching status. Finally, weighted two-sample T-tests were used to compare the length of stay and total hospitalization cost. Results A total of 12,449 weighted cases of inpatient non-variceal UGIB were included in this study. Patients aged 60 and older were more likely to die during the hospitalization (OR= 1.661, 95%CI: 1.108-2.490, p= 0.014), require blood transfusion (OR= 1.257, 95%CI: 1.131-1.396, p<0.001), and develop acute renal failure (OR= 1.672, 95%CI: 1.447-1.945, p<0.001). Patients aged 60 and older were also less likely to receive an early EGD (OR= 0.850, 95%CI: 0.752-0.961, p= 0.009). Total hospital costs (95%CI: -1397.77 - -4005.68, p<0.001) and length of stay (95%CI: -0.428 - -0.594, p<0.001) were both lower in patients aged 18-59 years. There was no difference in the development of post-hospitalization hemorrhagic shock between the two groups (OR= 0.984, 95%CI: 0.707-1.369, p= 0.923). Conclusions Patients aged 60 and older were less likely to have an early EGD and more likely to have worse outcomes. They had increased rates of inpatient mortality, blood transfusion requirements, development of acute renal failure, increased total hospital costs, and longer lengths of stay. There were no differences in the development of post-hospitalization hemorrhagic shock between the two groups.

Trends in Mortality and Health Care Burden of Cirrhotic Decompensation in Hospitalized Patients: A Nationwide Analysis.

INTRODUCTION: Mortality caused by cirrhosis is now the 14th most common cause of death worldwide and 12th most common in the United States. We studied trends in inpatient mortality and hospitalization charges associated with cirrhotic decompensation from esophageal variceal bleeding, ascites, hepatic encephalopathy, spontaneous bacterial peritonitis, and hepatorenal syndrome from 2007 to 2017. MATERIALS AND METHODS: Using the National Inpatient Sample databases, we first isolated patients 18 years or older with the diagnosis of cirrhosis using International Classification of Diseases, Ninth Revision (ICD-9) or International Classification of Diseases, Tenth Revision (ICD-10) codes. We then identified patients with the admission diagnosis of esophageal variceal bleeding, ascites, hepatic encephalopathy, spontaneous bacterial peritonitis, and hepatorenal syndrome. Time-series regression was used to determine whether a trend occurred over the study period. We also evaluated for patient-related demographic changes over the study period. RESULTS: A total of 259,897 cirrhotic patients with the studied decompensations were captured. During the study period, time-series regression confirmed downtrends in mortality rates and length of stay for all types of decompensations. Conversely, we found increases in hospitalization charges for all types of decompensations. Patient age increased over the study period. Patients were also more likely to be White and pay with. CONCLUSION: From 2007 to 2017, inpatient mortality rates and lengths of stay decreased for cirrhotic decompensations for all causes of decompensation. Total charges, conversely, increased for all causes.

Analysis of pancreatic pseudocyst drainage procedural outcomes: a population based study.

BACKGROUND: A pancreatic pseudocyst is a collection of fluid surrounded by a well-defined wall that contains no solid material. Studies on outcomes of pancreatic pseudocyst drainage have largely been limited to small cohorts. This study aims to take a population based approach to evaluate differences in inpatient outcomes among laparoscopic, percutaneous, and endoscopic drainage for pancreatic pseudocysts. METHODS: The National Inpatient Sample database was used to identify inpatient stays for pancreatic pseudocysts in which a single drainage approach was conducted. Baseline characteristic differences were compared with Rao-Scott chi squared and Mann-Whitney U tests. Propensity score matching controlling for clinical and demographic covariates followed by multivariable regression was used to pairwise compare drainage outcomes. Primary outcomes were length of stay, total charge, mortality, and disposition. Secondary outcomes were procedure related complication rates. RESULTS: Among a total of 35,640 weighted pancreatic pseudocyst cases, 3235 underwent drainage via a single procedure. Percutaneous was the most frequent drainage method performed (44.5%) and was more likely to be performed at nonteaching hospitals than laparoscopic (17% vs 9%, p = 0.04). Percutaneous drainage was associated with longer LOS (aIRR 1.42, 95% CI 1.07-1.86, p = 0.01) versus endoscopic and lower rates of routine disposition (aOR 0.45, 95% CI 0.23-0.89, p = 0.02) relative to endoscopic and laparoscopic (aOR 0.41, 95% CI 0.27-0.61, p < 0.01) drainage. There were no differences in primary outcomes in laparoscopic versus endoscopic drainage. Percutaneous drainage was associated with higher rates of septic shock than laparoscopic drainage (aOR 2.59, 95% CI 1.15-5.82, p = 0.02). CONCLUSIONS: Endoscopic and laparoscopic pancreatic pseudocyst drainage are associated with the least short term procedure related complications and more favorable in-hospital outcomes compared to percutaneous approaches. However, percutaneous drainage was the most commonly performed method in the 2017 NIS database.

Do Socio-Demographics Play a Role in the Prevalence of Red Flags and Pursuant Colonoscopies in Patients With Irritable Bowel Syndrome?

Background Irritable bowel syndrome (IBS) is a "brain-gut disorder" that lacks laboratory, radiologic, or physical exam findings. Colonoscopies are not routinely performed unless "red flag" symptoms, such as bleeding or abnormal weight loss, are present. Socio-demographics have been implicated as sources of potential disparities in appropriate care. Aims We hypothesize that the incidence of red flag symptoms and pursuant colonoscopies differ by socio-demographic status in patients with IBS. Methods Patients diagnosed with IBS were extracted from the National Inpatient Sample 2001-2013 using the International Classification of Diseases, Ninth Revision (ICD-9) codes. Gastrointestinal bleed, blood in stool, weight loss, and anemia were pooled into red flag symptoms. Colonoscopies during the admission were identified using ICD-9 procedural codes. Chi-square analysis and binomial logistic regression were used to evaluate potential disparities with α<0.01. Results Patients with Medicaid or Medicare or those without insurance had higher odds of presenting with red flag symptoms compared to those with private insurance. Medicaid patients and uninsured patients had higher odds of undergoing colonoscopies. All patients that were not Caucasian had higher odds of presenting with red flags and subsequently undergoing colonoscopies. Older patients had higher odds of presenting with concerning red flag symptoms but lower odds of undergoing colonoscopies. Conclusions The incidence of red flag symptoms and performance of colonoscopies differed by socio-demographics in patients with IBS. Patients with non-private or those without insurance were more likely to have red flags and undergo a colonoscopy. Age and race also increased rates of red flag symptoms while having a mixed effect on pursuant colonoscopies. This may represent discrepancies in healthcare utilization in a vulnerable population.

Effect of hospital teaching status on endoscopic retrograde cholangiopancreatography mortality and complications in the USA.

BACKGROUND: Our aim was to assess the differences in outcomes of cholecystitis, pancreatitis, gastrointestinal (GI) bleed, GI perforation, and mortality in teaching versus nonteaching hospitals nationwide among therapeutic and diagnostic ERCPs. We hypothesized that complication rates would be higher in teaching hospitals given greater patient complexity. METHODS: Inpatient diagnostic and therapeutic ERCPs were identified from the National Inpatient Sample (NIS) from 2008 to 2012. The presence of ACGME-approved residency programs is required to qualify as a teaching hospital. Nonteaching urban and rural hospitals were grouped together. We identified hospital stays complicated by pancreatitis, cholecystitis, GI hemorrhage, perforation, and mortality. Logistic regression propensity-matched analysis was performed in SPSS to compare differences in complication rates between teaching and nonteaching hospitals. RESULTS: A total of 1,466,356 weighted cases of inpatient ERCPs were included in this study: of those, 367 and188 were diagnostic, 1,099,168 were therapeutic, 766,230 were at teaching hospitals, and 700,126 were at nonteaching hospitals. Mortality rates were higher in teaching hospitals when compared to nonteaching hospitals for diagnostic (OR 1.266, p < 0.001) and therapeutic ERCPs (OR 1.157, p = 0.001). There was no significant difference in rates of post-ERCP cholecystitis, pancreatitis, or perforation between the two groups. Among diagnostic ERCPs, GI hemorrhage was higher in teaching compared to nonteaching hospitals (OR 1.181, p = 0.003). Likewise, length of stay was increased in teaching hospitals (7.9 vs 6.9 days, p < 0.001, for diagnostic and 6.5 vs 5.8 days, p < 0.001, for therapeutic ERCPs). CONCLUSIONS: In conclusion, teaching hospitals were noted to have a higher mortality rate associated with inpatient ERCPs as well as higher rates of GI hemorrhage in diagnostic ERCPs, which may be due to a higher comorbidity index in those patients admitted to teaching hospitals.

Hospital teaching status on the outcomes of patients with esophageal variceal bleeding in the United States.

BACKGROUND: Acute variceal bleeding is a major complication of portal hypertension and is a leading cause of death in patients with cirrhosis. There is limited data on the outcomes of patients with esophageal variceal bleeding in teaching versus nonteaching hospitals. Because esophageal variceal bleeding requires complex management, it may be hypothesized that teaching hospitals have lower mortality. AIM: To assess the differences in mortality, hospital length of stay (LOS) and cost of admission for patients admitted for variceal bleed in teaching versus nonteaching hospitals across the US. METHODS: The National Inpatient Sample is the largest all-payer inpatient database consisting of approximately 20% of all inpatient admissions to nonfederal hospitals in the United States. We collected data from the years 2008 to 2014. Cases of variceal bleeding were identified using the International Classification of Diseases, Ninth Edition, Clinical Modification codes. Differences in mortality, LOS and cost were evaluated for patients with esophageal variceal bleed between teaching and nonteaching hospitals and adjusted for patient characteristics and comorbidities. RESULTS: Between 2008 and 2014, there were 58362 cases of esophageal variceal bleeding identified. Compared with teaching hospitals, mortality was lower in non-teaching hospitals (8.0% vs 5.3%, P < 0.001). Median LOS was shorter in nonteaching hospitals as compared to teaching hospitals (4 d vs 5 d, P < 0.001). A higher proportion of non-white patients were managed in teaching hospitals. As far as procedures in nonteaching vs teaching hospitals, portosystemic shunt insertion (3.1% vs 6.9%, P < 0.001) and balloon tamponade (0.6% vs 1.2%) were done more often in teaching hospitals while blood transfusions (64.2% vs 59.9%, P = 0.001) were given more in nonteaching hospitals. Using binary logistic regression models and adjusting for baseline patient demographics and comorbid conditions the mortality, LOS and cost in teaching hospitals remained higher. CONCLUSION: In patients admitted for esophageal variceal bleeding, mortality, length of stay and cost were higher in teaching hospitals versus nonteaching hospitals when controlling for other confounding factors.

Trends in Post-Therapeutic Endoscopic Retrograde Cholangiopancreatography Gastrointestinal Hemorrhage, Perforation and Mortality from 2000 to 2012: A Nationwide Study.

BACKGROUND/AIMS: Recent trends in complications following inpatient therapeutic Endoscopic Retrograde Cholangiopancreatography (ERCP) remain poorly defined. We studied trends of gastrointestinal (GI) hemorrhage, perforation, and mortality following inpatient therapeutic ERCPs from 2000 to 2012 with the hypothesis that ERCPs would have down trending complication rates. METHODS: First, we isolated therapeutic ERCPs in patients 18 years or older using the International Classification of Diseases, Ninth Edition in the 2000 to 2012 National Inpatient Sample databases. Procedures complicated by hemorrhage, perforation, and mortality were identified. Multivariate logistic regressions were used to calculate trends in complication rates and secondary variables, including hospital and patient demographics. Time series regressions were then built for each complication to assess for trends from 2000 to 2012. RESULTS: The mortality rate decreased from 1.77 to 1.24%, a trend that was confirmed by time series regression. Perforation rates increased from 0.07 to 0.10% for therapeutic ERCPs. However, time series regression did not show a significant trend. GI hemorrhage rates increased from 1.36 to 1.57% and this uptrend was confirmed by our time series regression. CONCLUSION: Therapeutic ERCPs have become safer, as demonstrated by a down trending mortality rate. Over the same time, GI hemorrhage rates trended upwards, while no change was noted in perforation rates.