Low Predictability of Readmissions and Death Using Machine Learning in Cirrhosis.

INTRODUCTION: Readmission and death in cirrhosis are common, expensive, and difficult to predict. Our aim was to evaluate the abilities of multiple artificial intelligence (AI) techniques to predict clinical outcomes based on variables collected at admission, during hospitalization, and at discharge. METHODS: We used the multicenter North American Consortium for the Study of End-Stage Liver Disease (NACSELD) cohort of cirrhotic inpatients who are followed up through 90-days postdischarge for readmission and death. We used statistical methods to select variables that are significant for readmission and death and trained 3 AI models, including logistic regression (LR), kernel support vector machine (SVM), and random forest classifiers (RFC), to predict readmission and death. We used the area under the receiver operating characteristic curve (AUC) from 10-fold crossvalidation for evaluation to compare sexes. Data were compared with model for end-stage liver disease (MELD) at discharge. RESULTS: We included 2,170 patients (57 ± 11 years, MELD 18 ± 7, 61% men, 79% White, and 8% Hispanic). The 30-day and 90-day readmission rates were 28% and 47%, respectively, and 13% died at 90 days. Prediction for 30-day readmission resulted in 0.60 AUC for all patients with RFC, 0.57 AUC with LR for women-only subpopulation, and 0.61 AUC with LR for men-only subpopulation. For 90-day readmission, the highest AUC was achieved with kernel SVM and RFC (AUC = 0.62). We observed higher predictive value when training models with only women (AUC = 0.68 LR) vs men (AUC = 0.62 kernel SVM). Prediction for death resulted in 0.67 AUC for all patients, 0.72 for women-only subpopulation, and 0.69 for men-only subpopulation, all with LR. MELD-Na model AUC was similar to those from the AI models. DISCUSSION: Despite using multiple AI techniques, it is difficult to predict 30- and 90-day readmissions and death in cirrhosis. AI model accuracies were equivalent to models generated using only MELD-Na scores. Additional biomarkers are needed to improve our predictive capability (See also the visual abstract at http://links.lww.com/AJG/B710).

Hepatic Hydrothorax: An Updated Review on a Challenging Disease.

Hepatic hydrothorax is a challenging complication of cirrhosis related to portal hypertension with an incidence of 5-11% and occurs most commonly in patients with decompensated disease. Diagnosis is made through thoracentesis after excluding other causes of transudative effusions. It presents with dyspnea on exertion and it is most commonly right sided. Pathophysiology is mainly related to the direct passage of fluid from the peritoneal cavity through diaphragmatic defects. In this updated literature review, we summarize the diagnosis, clinical presentation, epidemiology and pathophysiology of hepatic hydrothorax, then we discuss a common complication of hepatic hydrothorax, spontaneous bacterial pleuritis, and how to diagnose and treat this condition. Finally, we elaborate all treatment options including chest tube drainage, pleurodesis, surgical intervention, Transjugular Intrahepatic Portosystemic Shunt and the most recent evidence on indwelling pleural catheters, discussing the available data and concluding with management recommendations.

PAI-1 Level Differences in Malignant Plural Effusion, Parapneumonic Pleuritis, and Cardiac Hydrothorax.

BACKGROUND AND OBJECTIVES: Plasminogen activator inhibitor-1 (PAI-1) is a fibrinolytic system enzyme whose role in various fibrinolytic processes is currently unknown. In clinical manifestations of pleural liquids of diverse etiology, various levels of fibrinolytic activity can be observed-parapneumonic processes tend to loculate in fibrin septa, while malignant pleural effusion (MPE) does not. The purpose of this study was to determine possible differences in PAI-1 levels in pleural effusions of varied etiology. MATERIAL AND METHODS: PAI-1 level in pleural effusion and serum was determined in 144 patients with pleural effusions of various etiology (cardiac hydrothorax-42 patients (29.2%), MPE-67 patients (46.5%), parapneumonic pleuritis-27 (18.8%), tuberculous pleuritis-6 patients (4.1%), pancreatogenic pleuritis-1 patient (0.7%) and pulmonary artery thromboembolism with pleuritis-1 patient (0.7%)). RESULTS: The median PAI-1 level (ng/mL) was the highest in the parapneumonic pleuritis group both in the effusion and the serum, with values of 291 (213-499) ng/mL and 204 (151-412) ng/mL, respectively, resulting in a statistically significant difference (p < 0.001) from the cardiac hydrothorax and MPE groups. However, there was no statistically significant difference between PAI-1 levels in the pleural effusion and serum in the cardiac hydrothorax and MPE groups. CONCLUSION: The PAI-1 level in MPE and cardiac hydrothorax was statistically significantly lower than in parapneumonic pleuritis.

An Unusual Presentation in Urinothorax.

Urinothorax is defined as the presence of urine in the pleural cavity. Leakage from the urinary tract can cause urinoma with retroperitoneal urine collection, and secondarily, urinothorax. We report the case of a 35-year-old female who presented with dyspnoea and right-sided chest pain. Chest radiograph revealed a right-sided pleural effusion. The patient had undergone left-sided ovarian cystectomy three months ago, had sustained a left-sided ureteric injury that required ureteric stent placement. Urinothorax was suspected as a consequence of ureteric injury; pleural fluid to serum creatinine ratio was found to be greater than one, confirming the diagnosis.

Hepatic hydrothorax.

Hepatic hydrothorax is defined as a pleural effusion in patients with liver cirrhosis in the absence of cardiopulmonary disease. The estimated prevalence among patients with liver cirrhosis is approximately 5-6%. The pathophysiology involves the passage of ascitic fluid from the peritoneal cavity to the pleural space through diaphragmatic defects. The diagnosis is made from clinical presentation and confirmed by diagnostic thoracentesis with pleural fluid analysis. The initial medical management is sodium restriction and diuretics, but liver transplantation provides the only definitive therapy. For patients who are not transplant candidates and those who await organ availability, other therapeutic modalities that are to be considered include transjugular intrahepatic portosystemic shunt placement, videoassisted thoracoscopic surgery repair, pleurodesis, and vasoconstrictors (eg, octreotide and terlipressin). The primary therapeutic goals are to reduce ascitic fluid production and improve symptoms to bridge the time for liver transplantation.

Angiogenic and antiangiogenic factors before and after resolution of maternal mirror syndrome.

Mirror syndrome is a rare condition that involves fetal hydrops, placentomegaly and severe maternal edema. The pathogenesis of this syndrome mimics endothelial dysfunction observed in pre-eclampsia. We report a case of maternal mirror syndrome caused by bilateral fetal hydrothorax that resolved after intrauterine pleuroamniotic shunt placement. At the time of the clinical manifestation there was an antiangiogenic state similar to that seen in pre-eclampsia, which resolved after fetal treatment. Our findings suggest that mirror syndrome is a manifestation of a broad spectrum of pathological conditions that induces an antiangiogenic state.

Hepatic hydrothorax in a child and its management.

Hepatic hydrothorax is a rare complication of portal hypertension. The optimal treatment for this condition is liver transplantation. Liver transplantation is significantly more manageable in children who weigh more than 8 kg. Here, an implantable pleural access device was used in a 5-month-old infant for painless iterative punctures to relieve respiratory symptoms, while waiting for liver transplantation and the patient's growth. The patient underwent successful transplantation 3 months later with a more optimal weight.

Pleural effusion in liver disease.

Hepatic hydrothorax is the paradigmatic pleural effusion in liver cirrhosis. It is defined as a pleural effusion in a patient with portal hypertension and no cardiopulmonary disease. The estimated prevalence of this complication in patients with liver cirrhosis is 5 to 6%. Its pathophysiology involves movement of ascitic fluid from the peritoneal cavity into the pleural space through diaphragmatic defects. Thoracentesis and pleural fluid analysis are necessary for diagnosis. Initial management consists of sodium restriction, diuretics, and therapeutic thoracentesis. A transjugular intrahepatic portosystemic shunt may provide a bridge prior to liver transplantation. Spontaneous bacterial empyema is the infection of a preexisting hydrothorax. The more frequent bacteria involved are ENTEROBACTERIACEAE and gram-positive cocci. Antibiotic therapy is the cornerstone of therapy. This article reviews etiology, clinical manifestations, and therapy of these two complications of liver cirrhosis and portal hypertension.

Respiratory mechanics and fluid dynamics after lung resection surgery.

Thoracic surgery that requires resection of a portion of lung or of a whole lung profoundly alters the mechanical and fluid dynamic setting of the lung-chest wall coupling, as well as the water balance in the pleural space and in the remaining lung. The most frequent postoperative complications are of a respiratory nature, and their incidence increases the more the preoperative respiratory condition seems compromised. There is an obvious need to identify risk factors concerning mainly the respiratory function, without neglecting the importance of other comorbidities, such as coronary disease. At present, however, a satisfactory predictor of postoperative cardiopulmonary complications is lacking; postoperative morbidity and mortality have remained unchanged in the last 10 years. The aim of this review is to provide a pathophysiologic interpretation of the main respiratory complications of a respiratory nature by relying on new concepts relating to lung fluid dynamics and mechanics. New parameters are proposed to improve evaluation of respiratory function from pre- to the early postoperative period when most of the complications occur.

A left-sided cerebrospinal fluid hydrothorax and a right ventricular-peritoneal shunt: a unique clinical case study.

We describe the case of a 84-year-old woman with a right ventricular-peritoneal shunt and a left-sided pleural effusion, the analysis of which was positive for cerebrospinal fluid. We consider the potential causative mechanisms. Our patient was managed conservatively due to her frailty, the effusion being asymptomatic and her preference not to pursue further invasive diagnostic testing. This case report is unique due to the contralateral nature of the effusion to the shunt, which has not been described before in the literature.

Indwelling Pleural Catheters in Hepatic Hydrothorax: A Single-Center Series of Outcomes and Complications.

BACKGROUND: Treatment of hepatic hydrothorax (HH) generally involves sodium restriction, diuretics, and serial thoracentesis. In more advanced cases, transjugular intrahepatic portosystemic shunt and liver transplantation may be required. Previously, indwelling tube drainage has been avoided due to concerns regarding high complication rates and overall poor outcomes. Recently, indwelling pleural catheters (IPCs) have been proposed as a novel treatment option for HH. METHODS: This study was a retrospective review of patients who had undergone IPC placement for HH over a 10-year period at a large liver transplant referral center. We tracked outcomes, including complication rates and liver transplantation, as well as biomarkers of nutritional status. RESULTS: Sixty-two patients underwent IPC placement between 2007 and 2017, with 33 IPCs (53%) placed as a bridge to liver transplantation. Complications were recorded in 22 patients (36%); empyema was the most common, diagnosed in 10 patients (16.1%). Ten patients evaluated for liver transplantation underwent successful transplantation following IPC placement. There were statistically significant decreases in both BMI and serum albumin levels following IPC placement. CONCLUSIONS: IPCs represent a potential treatment for refractory HH and should be used with caution in patients eligible for liver transplantation. Ideally, IPC use for these patients would be evaluated by a multidisciplinary team. IPC use may lead to small decreases in BMI and serum albumin levels in patients over time.

Hepatic hydrothorax: current concepts of pathophysiology and treatment options.

Pleural effusions develop in 6-10% of patients with end-stage liver disease. Although, commonly seen in conjunction with ascites, isolated hepatic hydrothorax can occur in a small number of patients with cirrhosis. Refractory hepatic hydrothorax particularly poses a challenging therapeutic dilemma as treatment options are limited at best in these patients. Current patho-physiologic understanding of this disorder, as a cause, points towards the presence of diaphragmatic defects responsible for the shift of fluid from the peritoneal to the pleural cavity. When sodium restriction and diuretic treatment fail, liver transplantation remains the most definitive therapy in these refractory cases. However, transjugular intrahepatic porto-systemic shunt (TIPS), or video-assisted thoracoscopic (VATS) repair of the diaphragmatic defects (with or without pleurodesis) are effective strategies in those who are not transplant candidates or those awaiting organ availability. Hepatic hydrothorax, especially when refractory to medical treatment, poses a challenging management dilemma. An early recognition and familiarity with available treatment modalities is crucial to effectively manage this exigent complication of cirrhosis.

Hepatic hydrothorax occurring rapidly after manual abdominal compression.

Hepatic hydrothorax is a relatively infrequent but potentially serious complication of liver cirrhosis that often causes respiratory dysfunction. Several hypotheses for the development of hepatic hydrothorax have been suggested to explain a transdiaphragmatic shift of ascitic fluid through small defects between the peritoneal cavity and the pleural space. However, the rapid development of hydrothorax within several hours is seldom encountered. In addition, the causal factors for rapid passage of ascitic fluid into the pleural cavity are unknown. This report describes a patient with liver cirrhosis who suffered rapid development of a hydrothorax after manual compression of the abdomen.

Frequency of Hepatic Hydrothorax and its Association with Child Pugh Class in Liver Cirrhosis Patients.

OBJECTIVE: To determine the frequency of hepatic hydrothorax and its association with Child Pugh Class in patients with liver cirrhosis. STUDY DESIGN: Descriptive, analytical study. PLACE AND DURATION OF STUDY: Jinnah Postgraduate Medical Centre, Karachi, Medical Unit-III, (Ward- 7), from June 2012 to May 2013. METHODOLOGY: All patients with established diagnosis of decompensated chronic liver disease were included. Detailed history, thorough physical examination, routine laboratory investigations, chest X-ray and abdominal ultrasound were carried out in all patients to find out the presence of pleural effusion and ascites, respectively. Fifty milliliters of pleural fluid was aspirated in all patients with pleural effusion using the transthoracic approach, taking ultrasound guidance, wherever required. Fluid was sent for microscopic, biochemical, and microbial analysis. SBEM defined if pleural fluid with polymorphonuclear (PMN) cell count > 500 cells/mm3 or positive culture with PMN cell count > 250 cells/mm3 with exclusion of a parapneumonic effusion. RESULTS: Two hundred and six patients met the inclusion criteria, with mean age of 41.25 ±13.59 years. Among them, 149 (72.3%) were males and 57 (27.7%) females. Twenty-three (11.2%) had hydrothorax; right sided involvement was in 18 (78.3%) subjects, 3 (13%) had left sided while bilateral pleural effusion was found in 2 (8.7%) cases. SBEM was found in 07 (30.43%) cases. Mean serum albumin 3.125 ±0.71 gram/dl. There was association between serum albumin levels and hydrothorax. Asignificant association of hydrothorax with Child Pugh scoring system (p=0.018), but not with serum albumin (p=0.15). CONCLUSION: The frequency of hepatic hydrothorax has a significant association with hepatic function as assessed by Child Pugh scoring system, but not with serum albumin.

The morphology of diaphragmatic defects in hepatic hydrothorax: thoracoscopic finding.

BACKGROUND: Until now, the pathophysiology of hepatic hydrothorax has been moot. We discuss (on the basis of gross videothoracoscopy findings in 11 cases and the literature) the pathogenesis and clinical presentation of this complex condition. METHODS: We prospectively studied 11 patients (age, 31-73 years; 6 men and 5 women) with refractory hepatic hydrothorax (Child-Pugh class B-C) who underwent thoracoscopic repair of diaphragmatic defects. The diaphragmatic defects were examined intraoperatively. RESULTS: The diaphragmatic defects stemming from hepatic hydrothorax were classified into 4 morphologic types: type I, no obvious defect (1 patient); type II, blebs lying on the diaphragm (4 patients); type III, broken defects (fenestrations) in the diaphragm (8 patients); and type IV, multiple gaps in the diaphragm (1 patient). The type of diaphragmatic defect did not correlate with the volume occupied by the pleural effusion in the preoperative chest radiograms. CONCLUSIONS: The finding of this study allowed hepatic hydrothorax pathophysiology to be directly visualized, and further studies concerning the treatment of hepatic hydrothorax might be based on these mechanisms.

Effect of CCL4-induced cirrhosis on the pathophysiologic course of acute myocardial infarction in nonarteriosclerotic vs arteriosclerotic male rats.

Arteriosclerotic and nonarteriosclerotic rats were treated with carbon tetrachloride (CCL4) to induce cirrhosis of the liver. Massive myocardial infarction was then induced in intact and CCL4-treated animals. During acute necrosis (Days 1 thru 3), animals were killed at 4, 8, 12 and 24 h on Days 1 and 2, and during myocardial repair on Days 4, 5 and 8. During the induction of cirrhosis, animals developed polydypsia, polyuria, and hyperglycemia; during myocardial infarction, the arteriosclerotic + cirrhotic animals developed severe and persistent congestive heart failure, i.e., hydrothorax. Adrenal and thymus gland weights and corticosterone levels indicated that cirrhosis per se increased pituitary--adrenal activity, particularly in arteriosclerotic animals. Enzyme levels of SGOT and SGPT demonstrated severe hepatic damage due to cirrhosis and acute myocardial infarction. Blood triglycerides and cholesterol responded abnormally in cirrhotic animals during acute myocardial ischemia due to their entrapment within hepatic cells. The cirrhotic animals manifested poor myocardial repair with persistent foci of necrosis, calcification, and a high incidence of large, occlusive, atrial thrombi. It is suggested that cirrhosis interferes with lipid metabolism and adrenal steroid conjugation leading to abnormal levels of mineralocorticoids which favor congestive heart failure, poor myocardial repair, and atrial thrombosis.