Mortality associated with gastrointestinal bleeding events: Comparing short-term clinical outcomes of patients hospitalized for upper GI bleeding and acute myocardial infarction in a US managed care setting

Effect of dialysis modalities on risk of hospitalization for gastrointestinal bleeding

Dialysis patients are at risk of both thromboembolic and bleeding events, while thromboembolism prevention and treatment may confer a risk of major bleeding. Gastrointestinal (GI) bleeding is a great concern which can result in high subsequent mortality rates. Our object was to clarify whether hemodialysis (HD) and peritoneal dialysis (PD) confer different incidence of GI bleeding, and further assist individualized decision-making on dialysis modalities. We conducted a population-based retrospective cohort study which included all incident dialysis patients above 18 years old derived from the National Health Insurance database from 1998 to 2013 in Taiwan. 6296 matched pairs of HD and PD patients were identified. A propensity score matching method was used to minimize the selection bias. The adjusted hazard ratio for GI bleeding was 1.13 times higher in the HD group than in the PD group, and data from the unmatched cohort and the stratified analysis led to similar results. Among subgroup analysis, we found that the use of anticoagulants will induce a much higher incidence of GI bleeding in HD patients as compared to in PD patients. We concluded that PD is associated with a lower GI bleeding risk than HD, and is especially preferred when anticoagulation is needed.

[Mortality Risk Factors for Inpatients with Ischemic Heart Disease Complicated with Gastrointestinal Bleeding]

OBJECTIVE

To investigate the risk factors of in-hospital mortality in patients with combined ischemic heart disease (IHD) and gastrointestinal bleeding (GIB).

METHODS

Patients who were hospitalized and received treatment for IHD combined with GIB at West China Hospital, Sichuan University between Jan. 2015 and Jan. 2018 were included in the study. Information concerning their baseline data, comorbidities, history of anticoagulant and antiplatelet medication, laboratory data on admission, and in-hospital treatments was collected. In-hospital death of all causes was taken as the primary endpoint event of the study, and multivariate logistic regression analysis was conducted to identify the independent risk factors of mortality during their hospital stay for this specific type of patients. Then, receiver operating characteristic ( ROC) curve was drawn and the area under curve ( AUC) was calculated accordingly.

RESULTS

A total of 395 patients met the enrollment criteria and were included in the study. Among them, 342 patients were discharged after their condition improved, and 53 patients died during hospitalization. Analysis of the cause of death revealed that cardiogenic death was the leading cause of death (54.7%), which was followed by infection-caused death (24.5%). Logistic regression analysis revealed that patients with ST-segment elevation myocardial infarction (STEMI) had a 2.527-fold risk of mortality compared with patients with non-acute coronary syndrome (odds ratio [ OR]=2.527, 95% confidence interval [ CI]: 1.152-8.277, P=0.043), and patients with comorbidity of chronic renal disease (CKD) had a 2.89-fold risk of mortality ( OR=2.89, 95% CI:1.187-7.037, P=0.019). It was also shown the higher level of WBC count ( OR=1.123, 95% CI: 1.057-1.193, P<0.001) and lower hemoglobin ( OR=1.014, 95% CI: 1.003-1.025, P=0.013) on admission were related to in-hospital mortality. On the other hand, endoscopy ( OR=0.305, 95% CI: 0.103-0.881, P=0.029) was identified as a protective factor in hospital treatment that decreased the risk of in-hospital mortality. ROC curve was drawn by combining the aforementioned variables to predict in-hospital mortality, which had an AUC of 0.79.

CONCLUSION

The actual type of IHD being STEMI, the patient's condition being complicated with chronic kidney disease, and having high white blood cells and low hemoglobin levels upon admission were considered independent risk factors for in-hospital death outcome of IHD patients complicated with GIB, while undergoing endoscopy during hospitalization was considered as a protective factor.

Detection of Bleeding Events in Electronic Health Record Notes Using Convolutional Neural Network Models Enhanced With Recurrent Neural Network Autoencoders: Deep Learning Approach

BACKGROUND

Bleeding events are common and critical and may cause significant morbidity and mortality. High incidences of bleeding events are associated with cardiovascular disease in patients on anticoagulant therapy. Prompt and accurate detection of bleeding events is essential to prevent serious consequences. As bleeding events are often described in clinical notes, automatic detection of bleeding events from electronic health record (EHR) notes may improve drug-safety surveillance and pharmacovigilance.

OBJECTIVE

We aimed to develop a natural language processing (NLP) system to automatically classify whether an EHR note sentence contains a bleeding event.

METHODS

We expert annotated 878 EHR notes (76,577 sentences and 562,630 word-tokens) to identify bleeding events at the sentence level. This annotated corpus was used to train and validate our NLP systems. We developed an innovative hybrid convolutional neural network (CNN) and long short-term memory (LSTM) autoencoder (HCLA) model that integrates a CNN architecture with a bidirectional LSTM (BiLSTM) autoencoder model to leverage large unlabeled EHR data.

RESULTS

HCLA achieved the best area under the receiver operating characteristic curve (0.957) and F1 score (0.938) to identify whether a sentence contains a bleeding event, thereby surpassing the strong baseline support vector machines and other CNN and autoencoder models.

CONCLUSIONS

By incorporating a supervised CNN model and a pretrained unsupervised BiLSTM autoencoder, the HCLA achieved high performance in detecting bleeding events.

Acute Myocardial Infarction with Simultaneous Gastric Perforation

Acute myocardial infarction and perforated peptic ulcer disease with associated peritonitis are both medical emergencies requiring urgent intervention. This patient presented with both emergencies simultaneously. Current literature is devoid of guidance as to which should be addressed initially. A multidisciplinary discussion was conducted leading to a unanimous decision for initiating percutaneous coronary intervention (PCI). After successful PCI, the patient was immediately taken to the operating room for laparoscopic repair of the perforated viscous. Subsequent to the operative repair, the patient became hemodynamically unstable and a repeat electrocardiogram demonstrated complete right coronary occlusion. Shock ensued and the patient died in the intensive care unit despite this plan of care. It is our opinion that this case reveals the need for expert panels to devise decision algorithms for concomitant presentations of life-threatening diseases.

Risk factors of in-hospital mortality among patients with upper gastrointestinal bleeding and acute myocardial infarction

Background/Aims

Patients with simultaneous upper gastrointestinal bleeding (UGIB) and acute myocardial infarction (AMI) have higher mortality than patients with either GIB or AMI. We aimed to assess the incidence and risk factors of in-hospital mortality in patients with UGIB and AMI.

Patients and Methods

A total of 243 patients with UGIB and AMI were enrolled during 2012-2017. Clinical and laboratory data were collected and analyzed for clinical characteristics and potential risk factors of in-hospital mortality.

Results

Among the 243 patients, 60 in-hospital deaths were observed (in-hospital mortality rate of 24.7%). Patients who died were older than the survivors (78.7 ± 6.6 vs. 72.6 ± 10.5 years, P < 0.001). Compared with survivors, patients who died showed increased peak white blood cell (WBC) count (9.74 ± 4.72 vs. 7.60 ± 2.91 × 109/L, P= 0.002), serum creatinine levels [134 (106, 190) vs. 97 (79, 125) mmol/L, P= 0.014], peak blood urine nitrogen levels (16.31 ± 8.48 mmol/L vs. 9.86 ± 6.33 mmol/L, P < 0.001), and peak brain natriuretic peptide (BNP) amounts [13,250 (6071, 30,000) vs. 3598 (728, 12,842) pg/mL, P < 0.001]. Meanwhile, patients who died also displayed lower minimum hemoglobin levels (78.3 ± 21.1 vs. 86.3 ± 22.3 g/L, P= 0.018) and minimum platelet counts (184.3 ± 79.1 vs. 214.6 ± 80.1 × 109/L, P= 0.013). In multivariable logistic analysis, age [OR (95% CI) =1.118 (1.053-1.186), P < 0.001], peak WBC count [OR (95% CI) =1.252 (1.113-1.407), P < 0.001], minimum platelet count [OR (95% CI) = 0.994 (0.989-1.000), P= 0.032], and peak BNP levels [OR (95% CI) =3.880 (1.761-8.550), P= 0.001] were independent predictors of in-hospital mortality.

Conclusions

Patients with UGIB and AMI had a high in-hospital mortality, which was independently associated with age, peak WBC count, minimum platelet count, and peak BNP levels.

Spontaneously reported haemorrhagic adverse events associated with rivaroxaban and dabigatran in Australia

OBJECTIVES

The objective of our study was to describe spontaneously reported haemorrhagic adverse events associated with rivaroxaban and dabigatran in Australia.

METHODS

Data were sourced from the Australian Therapeutic Goods Administration (TGA) Database of Adverse Event Notifications between June 2009 and May 2014. Records of haemorrhagic adverse events in which rivaroxaban or dabigatran was considered as a potential cause were analysed.

RESULTS

There were 240 haemorrhagic adverse events associated with rivaroxaban and 504 associated with dabigatran. Age was specified for 164 (68%) haemorrhages associated with rivaroxaban, of which 101 occurred in people aged ⩾75 years. Age was specified for 437 (87%) haemorrhages associated with dabigatran, of which 300 occurred in people aged ⩾75 years. Time from treatment initiation to haemorrhage was specified for 122 (51%) haemorrhages associated with rivaroxaban, with 69 (57%) haemorrhages occurring within 30 days of rivaroxaban initiation. Time from treatment initiation to haemorrhage was specified for 253 (50%) haemorrhages associated with dabigatran, with 123 (49%) haemorrhages occurring within 30 days of dabigatran initiation. Gastrointestinal (GI) haemorrhages were the most frequent type of haemorrhages associated with both rivaroxaban (n = 105, 44%) and dabigatran (n = 302, 60%). Data were available on the severity of haemorrhage for 101 (42%) haemorrhages associated with rivaroxaban, with haemorrhage leading to death in 17 people. The severity of haemorrhage was specified for 384 (76%) haemorrhages associated with dabigatran, with haemorrhage leading to death in 61 people.

CONCLUSIONS

Our study highlights the need for research on the haemorrhagic complications of anticoagulation in clinical care. A considerable proportion of reported haemorrhagic events occurred within 30 days of rivaroxaban and dabigatran initiation. This highlights the importance of considering bleeding risk at the time of treatment initiation.

Hemophagocytic Lymphohistiocytosis and Gastrointestinal Bleeding: What a Surgeon Should Know

This paper presents to the surgical community an unusual and often ignored cause of gastrointestinal bleeding. Hemophagocytic syndrome or hemophagocytic lymphohistiocytosis (HLH) is a rare medical entity characterized by phagocytosis of red blood cells, leucocytes, platelets, and their precursors in the bone marrow by activated macrophages. When intestinal bleeding is present, the management is very challenging with extremely high mortality rates. Early diagnosis and treatment seem to be the most important factors for a successful outcome. We present two cases and review another 18 from the literature.

Management of variceal and nonvariceal upper gastrointestinal bleeding in patients with cirrhosis

Acute upper gastrointestinal haemorrhage remains the most common medical emergency managed by gastroenterologists. Causes of upper gastrointestinal bleeding (UGIB) in patients with liver cirrhosis can be grouped into two categories: the first includes lesions that arise by virtue of portal hypertension, namely gastroesophageal varices and portal hypertensive gastropathy; and the second includes lesions seen in the general population (peptic ulcer, erosive gastritis, reflux esophagitis, Mallory-Weiss syndrome, tumors, etc.). Emergency upper gastrointestinal endoscopy is the standard procedure recommended for both diagnosis and treatment of UGIB. The endoscopic treatment of choice for esophageal variceal bleeding is band ligation of varices. Bleeding from gastric varices is treated by injection with cyanoacrylate. Treatment with vasoactive drugs as well as antibiotic treatment is started before or at the same time as endoscopy. Bleeding from portal hypertensive gastropathy is less frequent, usually chronic and treatment options include β-blocker therapy, injection therapy and interventional radiology. The standard of care of UGIB in patients with cirrhosis includes careful resuscitation, preferably in an intensive care setting, medical and endoscopic therapy, early consideration for placement of transjugular intrahepatic portosystemic shunt and, sometimes, surgical therapy or hepatic transplant.