The evidence interval and the Bayesian evidence value: On a unified theory for Bayesian hypothesis testing and interval estimation

Comparative efficacy of early TIPS, Non-early TIPS, and Standard treatment in patients with cirrhosis and acute variceal bleeding: a network meta-analysis

BACKGROUND

Cirrhosis is a chronic disease characterized by chronic liver inflammation and diffuse fibrosis. A combination of vasoactive drugs, preventive antibiotics, and endoscopy is the recommended standard treatment for patients with acute variceal bleeding; however, this has been challenged. We compared the effects of early transjugular intrahepatic portosystemic shunt (TIPS), non-early TIPS, and standard treatment in patients with cirrhosis and acute variceal bleeding.

MATERIALS AND METHODS

The present network meta-analysis was conducted in accordance with the criteria outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses and Assessing the methodological quality of systematic reviews guidelines. The review has been registered with the International Prospective Register of Systematic Reviews. The PubMed, Embase, Cochrane Library, ClinicalTrials.gov, and World Health Organization-approved trial registry databases were searched for randomized controlled trials (RCTs) evaluating early TIPS, non-early TIPS, and standard treatment in patients with cirrhosis and acute variceal bleeding.

RESULTS

Twenty-four RCTs (1894 patients) were included in the review. Compared with standard treatment, early TIPS [odds ratio (OR), 0.53; 95% credible interval (Cr), 0.30-0.94; surface under the cumulative ranking curve (SUCRA), 98.3] had a lower risk of all-cause mortality (moderate-to-high-quality evidence), and early TIPS (OR, 0.19; 95% CrI, 0.11-0.28; SUCRA, 98.2) and non-early TIPS (OR, 0.30; 95% CrI, 0.23-0.42; SUCRA, 1.8) were associated with a lower risk of rebleeding (moderate-to-high-quality evidence). Early TIPS was not associated with a reduced risk of hepatic encephalopathy, and non-early TIPS (OR, 2.78; 95% CrI, 1.89-4.23, SUCRA, 0) was associated with an increased incidence of hepatic encephalopathy (moderate-to-high-quality evidence). There was no difference in the incidence of new or worsening ascites (moderate-to-high-quality evidence) among the three interventions.

CONCLUSION

Based on the moderate-to-high quality evidence presented in this study, early TIPS placement was associated with reduced all-cause mortality [with a median follow-up of 1.9 years (25th-75th percentile range 1.9-2.3 years)] and rebleeding compared to standard treatment and non-early TIPS. Although early TIPS and standard treatment had a comparable incidence of hepatic encephalopathy, early TIPS showed superiority over non-early TIPS in this aspect. Recent studies have also shown promising results in controlling TIPS-related hepatic encephalopathy. However, it is important to consider individual patient characteristics and weigh the potential benefits against the risks associated with early TIPS. Therefore, we recommend that clinicians carefully evaluate the patient's condition, considering factors such as severity of variceal bleeding, underlying liver disease, and overall clinical status, before making a treatment decision. Further well-designed RCTs comparing early TIPS with non-early TIPS are needed to validate these findings and provide more definitive guidance.

Changes in postictal cerebral perfusion are related to the duration of electroconvulsive therapy-induced seizures

OBJECTIVE

Postictal symptoms may result from cerebral hypoperfusion, which is possibly a consequence of seizure-induced vasoconstriction. Longer seizures have previously been shown to cause more severe postictal hypoperfusion in rats and epilepsy patients. We studied cerebral perfusion after generalized seizures elicited by electroconvulsive therapy (ECT) and its relation to seizure duration.

METHODS

Patients with a major depressive episode who underwent ECT were included. During treatment, 21-channel continuous electroencephalogram (EEG) was recorded. Arterial spin labeling magnetic resonance imaging scans were acquired before the ECT course (baseline) and approximately 1 h after an ECT-induced seizure (postictal) to quantify global and regional gray matter cerebral blood flow (CBF). Seizure duration was assessed from the period of epileptiform discharges on the EEG. Healthy controls were scanned twice to assess test-retest variability. We performed hypothesis-driven Bayesian analyses to study the relation between global and regional perfusion changes and seizure duration.

RESULTS

Twenty-four patients and 27 healthy controls were included. Changes in postictal global and regional CBF were correlated with seizure duration. In patients with longer seizure durations, global decrease in CBF reached values up to 28 mL/100 g/min. Regional reductions in CBF were most prominent in the inferior frontal gyrus, cingulate gyrus, and insula (up to 35 mL/100 g/min). In patients with shorter seizures, global and regional perfusion increased (up to 20 mL/100 g/min). These perfusion changes were larger than changes observed in healthy controls, with a maximum median global CBF increase of 12 mL/100 g/min and a maximum median global CBF decrease of 20 mL/100 g/min.

SIGNIFICANCE

Seizure duration is a key factor determining postictal perfusion changes. In future studies, seizure duration needs to be considered as a confounding factor due to its opposite effect on postictal perfusion.

Reducing the false discovery rate of preclinical animal research with Bayesian statistical decision criteria

The success of preclinical research hinges on exploratory and confirmatory animal studies. Traditional null hypothesis significance testing is a common approach to eliminate the chaff from a collection of drugs, so that only the most promising treatments are funneled through to clinical research phases. Balancing the number of false discoveries and false omissions is an important aspect to consider during this process. In this paper, we compare several preclinical research pipelines, either based on null hypothesis significance testing or based on Bayesian statistical decision criteria. We build on a recently published large-scale meta-analysis of reported effect sizes in preclinical animal research and elicit a non-informative prior distribution under which both approaches are compared. After correcting for publication bias and shrinkage of effect sizes in replication studies, simulations show that (i) a shift towards statistical approaches which explicitly incorporate the minimum clinically important difference reduces the false discovery rate of frequentist approaches and (ii) a shift towards Bayesian statistical decision criteria can improve the reliability of preclinical animal research by reducing the number of false-positive findings. It is shown that these benefits hold while keeping the number of experimental units low which are required for a confirmatory follow-up study. Results show that Bayesian statistical decision criteria can help in improving the reliability of preclinical animal research and should be considered more frequently in practice.