Refine
Document Type
- Article (11)
- Preprint (2)
- Doctoral Thesis (1)
Has Fulltext
- yes (14)
Is part of the Bibliography
- no (14)
Keywords
- cirrhosis (2)
- portal hypertension (2)
- ACLF (1)
- Bipolar disorder (1)
- Central Europe (1)
- Cirrhosis (1)
- Diagnostic markers (1)
- Ellenberg indicator values (1)
- Endokrinologie (1)
- Genetics (1)
- Genome-wide association studies (1)
- Liver diseases (1)
- Platelets (1)
- Portal veins (1)
- Prostaglandin (1)
- SARS-CoV-2 (1)
- SPSS (1)
- Thrombosis (1)
- Thromboxane (1)
- Veins (1)
- Viral infection (1)
- acute decompensation (1)
- acute-on-chronic liver failure (1)
- biogeographic legaciese (1)
- bioindication (1)
- computed tomography (1)
- eutrophication (1)
- forest classification (1)
- forest functional similarity (1)
- hepatic encephalopathy (1)
- liver (1)
- nitrogen deposition (1)
- phylogenetic community distance (1)
- portosystemic shunt (1)
- renin-angiotensin system (1)
- soil nutrients (1)
- spontaneous portosystemic shunt (1)
- tropical forests (1)
- vegetation-plot data (1)
The genetic make-up of an individual contributes to the susceptibility and response to viral infection. Although environmental, clinical and social factors have a role in the chance of exposure to SARS-CoV-2 and the severity of COVID-191,2, host genetics may also be important. Identifying host-specific genetic factors may reveal biological mechanisms of therapeutic relevance and clarify causal relationships of modifiable environmental risk factors for SARS-CoV-2 infection and outcomes. We formed a global network of researchers to investigate the role of human genetics in SARS-CoV-2 infection and COVID-19 severity. Here we describe the results of three genome-wide association meta-analyses that consist of up to 49,562 patients with COVID-19 from 46 studies across 19 countries. We report 13 genome-wide significant loci that are associated with SARS-CoV-2 infection or severe manifestations of COVID-19. Several of these loci correspond to previously documented associations to lung or autoimmune and inflammatory diseases3,4,5,6,7. They also represent potentially actionable mechanisms in response to infection. Mendelian randomization analyses support a causal role for smoking and body-mass index for severe COVID-19 although not for type II diabetes. The identification of novel host genetic factors associated with COVID-19 was made possible by the community of human genetics researchers coming together to prioritize the sharing of data, results, resources and analytical frameworks. This working model of international collaboration underscores what is possible for future genetic discoveries in emerging pandemics, or indeed for any complex human disease.
Background & Aims: Spontaneous portosystemic shunts (SPSS) frequently develop in liver cirrhosis. Recent data suggested that the presence of a single large SPSS is associated with complications, especially overt hepatic encephalopathy (oHE). However, the presence of >1 SPSS is common. This study evaluates the impact of total cross-sectional SPSS area (TSA) on outcomes in patients with liver cirrhosis.
Methods: In this retrospective international multicentric study, CT scans of 908 cirrhotic patients with SPSS were evaluated for TSA. Clinical and laboratory data were recorded. Each detected SPSS radius was measured and TSA calculated. One-year survival was the primary endpoint and acute decompensation (oHE, variceal bleeding, ascites) was the secondary endpoint.
Results: A total of 301 patients (169 male) were included in the training cohort. Thirty percent of all patients presented with >1 SPSS. A TSA cut-off of 83 mm2 was used to classify patients with small or large TSA (S-/L-TSA). Patients with L-TSA presented with higher model for end-stage liver disease score (11 vs. 14) and more commonly had a history of oHE (12% vs. 21%, p <0.05). During follow-up, patients with L-TSA experienced more oHE episodes (33% vs. 47%, p <0.05) and had lower 1-year survival than those with S-TSA (84% vs. 69%, p <0.001). Multivariate analysis identified L-TSA (hazard ratio 1.66; 95% CI 1.02–2.70, p <0.05) as an independent predictor of mortality. An independent multicentric validation cohort of 607 patients confirmed that patients with L-TSA had lower 1-year survival (77% vs. 64%, p <0.001) and more oHE development (35% vs. 49%, p <0.001) than those with S-TSA.
Conclusion: This study suggests that TSA >83 mm2 increases the risk for oHE and mortality in patients with cirrhosis. Our results support the clinical use of TSA/SPSS for risk stratification and decision-making in the management of patients with cirrhosis.
Lay summary: The prevalence of spontaneous portosystemic shunts (SPSS) is higher in patients with more advanced chronic liver disease. The presence of more than 1 SPSS is common in advanced chronic liver disease and is associated with the development of hepatic encephalopathy. This study shows that total cross-sectional SPSS area (rather than diameter of the single largest SPSS) predicts survival in patients with advanced chronic liver disease. Our results support the clinical use of total cross-sectional SPSS area for risk stratification and decision-making in the management of SPSS.
In psychiatry, there has been a growing focus on identifying at-risk populations. For schizophrenia, these efforts have led to the development of early recognition and intervention measures. Despite a similar disease burden, the populations at risk of bipolar disorder have not been sufficiently characterized. Within the BipoLife consortium, we used magnetic resonance imaging (MRI) data from a multicenter study to assess structural gray matter alterations in N = 263 help-seeking individuals from seven study sites. We defined the risk using the EPIbipolar assessment tool as no-risk, low-risk, and high-risk and used a region-of-interest approach (ROI) based on the results of two large-scale multicenter studies of bipolar disorder by the ENIGMA working group. We detected significant differences in the thickness of the left pars opercularis (Cohen’s d = 0.47, p = 0.024) between groups. The cortex was significantly thinner in high-risk individuals compared to those in the no-risk group (p = 0.011). We detected no differences in the hippocampal volume. Exploratory analyses revealed no significant differences in other cortical or subcortical regions. The thinner cortex in help-seeking individuals at risk of bipolar disorder is in line with previous findings in patients with the established disorder and corresponds to the region of the highest effect size in the ENIGMA study of cortical alterations. Structural alterations in prefrontal cortex might be a trait marker of bipolar risk. This is the largest structural MRI study of help-seeking individuals at increased risk of bipolar disorder.
Knowledge about the biogeographic affinities of the world’s tropical forests helps to better understand regional differences in forest structure, diversity, composition, and dynamics. Such understanding will enable anticipation of region-specific responses to global environmental change. Modern phylogenies, in combination with broad coverage of species inventory data, now allow for global biogeographic analyses that take species evolutionary distance into account. Here we present a classification of the world’s tropical forests based on their phylogenetic similarity. We identify five principal floristic regions and their floristic relationships: (i) Indo-Pacific, (ii) Subtropical, (iii) African, (iv) American, and (v) Dry forests. Our results do not support the traditional neo- versus paleotropical forest division but instead separate the combined American and African forests from their Indo-Pacific counterparts. We also find indications for the existence of a global dry forest region, with representatives in America, Africa, Madagascar, and India. Additionally, a northern-hemisphere Subtropical forest region was identified with representatives in Asia and America, providing support for a link between Asian and American northern-hemisphere forests.
Objective: Liver stiffness measurement (LSM) is a tool used to screen for significant fibrosis and portal hypertension. The aim of this retrospective multicentre study was to develop an easy tool using LSM for clinical outcomes in advanced chronic liver disease (ACLD) patients.
Design: This international multicentre cohort study included a derivation ACLD patient cohort with valid two-dimensional shear wave elastography (2D-SWE) results. Clinical and laboratory parameters at baseline and during follow-up were recorded. LSM by transient elastography (TE) was also recorded if available. The primary outcome was overall mortality. The secondary outcome was the development of first/further decompensation.
Results: After screening 2148 patients (16 centres), 1827 patients (55 years, 62.4% men) were included in the 2D-SWE cohort, with median liver SWE (L-SWE) 11.8 kPa and a model for end stage liver disease (MELD) score of 8. Combination of MELD score and L-SWE predict independently of mortality (AUC 0.8). L-SWE cut-off at ≥20 kPa combined with MELD ≥10 could stratify the risk of mortality and first/further decompensation in ACLD patients. The 2-year mortality and decompensation rates were 36.9% and 61.8%, respectively, in the 305 (18.3%) high-risk patients (with L-SWE ≥20 kPa and MELD ≥10), while in the 944 (56.6%) low-risk patients, these were 1.1% and 3.5%, respectively. Importantly, this M10LS20 algorithm was validated by TE-based LSM and in an additional cohort of 119 patients with valid point shear SWE-LSM.
Conclusion: The M10LS20 algorithm allows risk stratification of patients with ACLD. Patients with L-SWE ≥20 kPa and MELD ≥10 should be followed closely and receive intensified care, while patients with low risk may be managed at longer intervals.
Rationale: Early reperfusion in patients experiencing acute ischemic stroke is critical, especially for patients with large vessel occlusion who have poor prognosis without revascularization. S olitaire™ stent retriever devices have been shown to immediately restore vascular perfusion safely, rapidly, and effectively in acute ischemic stroke patients with large vessel occlusions.
Aim: The aim of the study was to demonstrate that, among patients with large vessel, anterior circulation occlusion who have received intravenous tissue plasminogen activator, treatment with S olitaire revascularization devices reduces degree of disability 3 months post stroke.
Design: The study is a global multicenter, two‐arm, prospective, randomized, open, blinded end‐point trial comparing functional outcomes in acute ischemic stroke patients who are treated with either intravenous tissue plasminogen activator alone or intravenous tissue plasminogen activator in combination with the Solitaire device. Up to 833 patients will be enrolled.
Procedures:Patients who have received intravenous tissue plasminogen activator are randomized to either continue with intravenous tissue plasminogen activator alone or additionally proceed to neurothrombectomy using the S olitaire device within six‐hours of symptom onset.
Study Outcomes: The primary end‐point is 90‐day global disability, assessed with the modified R ankin S cale (mRS). Secondary outcomes include mortality at 90 days, functional independence (mRS ≤ 2) at 90 days, change in N ational I nstitutes of H ealth S troke S cale at 27 h, reperfusion at 27 h, and thrombolysis in cerebral infarction 2b/3 flow at the end of the procedure.
Analysis:Statistical analysis will be conducted using simultaneous success criteria on the overall distribution of modified R ankin S cale (R ankin shift) and proportions of subjects achieving functional independence (mRS 0–2).
Introduction: Recent animal studies have shown that the alternate renin-angiotensin system (RAS) consisting of angiotensin-converting enzyme 2 (ACE2), angiotensin-(1–7) (Ang-(1–7)) and the Mas receptor is upregulated in cirrhosis and contributes to splanchnic vasodilatation and portal hypertension. To determine the potential relevance of these findings to human liver disease, we evaluated its expression and relationship to the patients’ clinical status in subjects with cirrhosis. Methods: Blood sampling from peripheral and central vascular beds was performed intra-operatively for cirrhotic patients at the time of liver transplantation (LT) or trans-jugular intra-hepatic portosystemic shunt (TIPS) procedures to measure angiotensin II (Ang II) and Ang-(1–7) peptide levels and ACE and ACE2 enzyme activity. Relevant clinical and hemodynamic data were recorded pre-operatively for all subjects and peripheral blood sampling was repeated 3 months or later post-operatively. Results: Ang-(1–-7) and ACE2 activity were up-regulated more than twofold in cirrhotic subjects both at the time of LT and TIPS and levels returned to comparable levels as control subjects post-transplantation. Ang-(1–7) levels correlated positively with the degree of liver disease severity, as measured by the model for an end-stage liver disease (MELD) and also with clinical parameters of pathological vasodilatation including cardiac output (CO). There were strong correlations found between the ACE2:ACE and the Ang-(1–7):Ang II ratio highlighting the inter-dependence of the alternate and classical arms of the RAS and thus their potential impact on vascular tone. Conclusions: In human cirrhosis, the alternate RAS is markedly upregulated and the activation of this system is associated strongly with features of the hyperdynamic circulation in advanced human cirrhosis.
Autoimmune Endokrinopathien entstehen durch die Aktivierung pathogener TLymphozyten, welche in de Lage sind, bestimmte Ziel-Zellen mittels Zytokinen oder anderer Mechanismen zu zerstören. Typ I Diabetes mellitus ist ein durch Insulinmangel bedingter Defekt der Glukoseverwertung, der durch eine autoimmune Zerstörung der pankreatischen Beta-Zellen verursacht wird. Als häufigste chronische Erkrankung Jugendlicher und junger Erwachsener in den Industrienationen stellt sie durch ihre stetig steigende Inzidenz, ihr immer früheres Erkrankungsalter und nicht zuletzt die steigende allgemeine Lebenserwartung eine bedeutende medizinische und ökonomische Herausforderung dar. Die zunehmende Aufklärung der mannigfaltigen Wechselbeziehungen des Immunsystems und der starke Einfluss des Vitamin D auf die Entwicklung und Funktion des Immunsystems führten zu der Überlegung, dass auch der Vitamin D Rezeptor (VDR) in Verbindung mit seinem biologisch aktivsten Liganden 1,25 (OH)2D3 einen Einfluss auf die Entwicklung einer Autoimmunendokrinopathie haben könnte. In Tierexperimenten konnte gezeigt werden, dass 1,25 (OH)2D3 und auch seine Analoga die Produktion des Interleukins IL-12 hemmen, welches eine wichtige Rolle in der Entwicklung der Th1-Helferzellen ausübt, gleichzeitig wird die Produktion des Interleukins IL-10 stimuliert, welches die Entwicklung von Th1-Helferzellen blockiert. Dieser Zelltyp ist bei der Entwicklung eines Typ 1 Diabetes mellitus ursächlich beteiligt. 1,25 (OH)2D3 kann auch die Entwicklung der Th2 Helferzellen inhibieren, welche in der Pathogenese einer Hashimoto Thyreoiditis involviert sind. In betroffenen Geweben, in denen pathogen-wirksame Th1-Zellen die Bildung zytotoxischer T-Zellen und aktivierter Makrophagen bewirken, kann 1,25 (OH)2D3 und seine Analoga die Anzahl der Th1-Zellen herunterregeln. Des weiteren regt Vitamin D die Bildung von CD4+CD25+-regulatorischen T-Zellen und Th2-Zellen an, die ihrerseits wiederum hemmend auf Th1-Zellen wirken. Bisher haben bereits diverse Studien bestätigt, das funktionelle Auswirkungen verschiedener Einzelnukleotid-Polymorphismen in kodierenden und nicht kodierenden Regionen des VDR- Gens in kausalem Zusammenhang mit dem Auftreten eines Typ 1 Diabetes Mellitus stehen. In dieser Arbeit zeigten sich erniedrigte 1,25 (OH)2D3 Spiegel in deutschen Patienten mit Typ 1 Diabetes Mellitus. Eine vergleichbare Studie wurde in der finnischen Bevölkerung durchgeführt und bestätigt unsere Resultate. Dies spiegelt die zentrale Rolle des Vitamin D bei der Entwicklung eines Typ 1 Diabetes Mellitus wieder. Um den kausalen Zusammenhang zwischen molekularen und funktionellen Beziehungen des Vitamin D Rezeptors näher zu definieren, liegt die Frage nahe, inwiefern eine Korrelation der schon bekannten VDR- Polymorphismen und neu etablierten mit Vitamin D Spiegeln in Patienten und Kontrollen besteht, weiterhin, ob auf der Ebene der Genexpression schon ein signifikanter Unterschied bei Patienten mit Typ 1 Diabetes Mellitus zu beschreiben ist. Zusätzlich wurden sieben neue VDR- Polymorphismen in Familien, mit jeweils einem an Typ 1 Diabetes Mellitus erkrankten Kind, untersucht. Bestimmte Transmissionsmuster wurden bereits für die VDR- Polymorphismen ApaI, BsmI, FokI und TaqI bei Typ 1 Diabetes Mellitus beschrieben und eine signifikante Korrelation hergestellt. In dieser Doktorarbeit zeichnet sich eine grenzwertig signifikante Korrelation zwischen dem VDR- Polymorphismus 2238135 und Typ 1 Diabetes Mellitus ab. Das Allel „C“ wurde häufiger auf die erkrankten Kinder transmittiert. Dieses Ergebnis ließ sich ebenfalls in der Gegenüberstellung von Patienten zu Kontrollen nachweisen, in der Allelfrequenzanalyse stellte sich das Allel „C“ ebenfalls als dominant heraus. Für die anderen sechs untersuchten VDR- Polymorphismen konnte keine signifikante Korrelation zu Typ 1 Diabetes Mellitus in der deutschen Population nachgewiesen werden. Jedoch konnten bei vier der sieben VDR- Polymorphismen signifikante Haplotypen beschrieben werden, die den Zusammenhang zu Typ 1 Diabetes Mellitus erneut bestätigen. Dies zeigt, wie die alleinige Analyse eines VDR- Polymorphismus scheinbar keine Assoziation zum Typ 1 Diabetes Mellitus zu haben scheint, aber durch die Bildung von Allelkombinationen sich signifikante Korrelationen für die gleichen Polymorphismen ausbilden können. Die gemeinsame Analyse verleiht den untersuchten VDR-Polymorphismen erst die Signifikanz. Das Gleichgewicht war jedoch deutlich zugunsten der signifikant seltener transmittierten und damit protektiven Haplotypen verschoben. Nur „GCTG“ wurde häufiger transmittiert (62,7% vs. 37,3%, p= 0,0087), während die folgenden 3 Haplotypen „TCTG“, „TCGG“ und „GCGG“ seltener transmittiert wurden und sich als signifikant herausstellten. „TCGG“ stellte sich als grenzwertig signifikanter Haplotyp heraus, während „GCGG“ als signifikante Allelkombination anzusehen ist. Am stärksten jedoch zeigte sich der diabetogen- protekive Effekt beim Screening des „TCTG“ Haplotypen, der fast gar nicht an die erkrankten Nachkommen transmittiert wurde (99,1% vs. 0,9%, p= 0,0006). In der vorliegenden Doktorarbeit bestätigt sich einerseits die molekulare Korrelation zwischen bestimmten Polymorphismen des VDR- Gens und dem Auftreten eines Typ 1 Diabetes Mellitus. Andererseits konnten vier neue Haplotypen bestimmt werden, die im Rahmen des durchgeführten Screenings in der deutschen Bevölkerung das Risiko für die Entstehung eines Diabetes als schützende oder prädisponierende Faktoren mitbestimmen. 1,25(OH)2D3, der stärkste immunregulatorische Vitamin D Metabolit, stellt für die medikamentöse Prävention und Therapie der autoimmunen Erkrankungen auch weiterhin ein interessantes Ziel dar, unter Umständen auch über die Beeinflussung seiner Stoffwechselwege, deren molekulare Korrelation hier weiterführend geklärt wurde.
Within the last 30 years the role of nitrogen in Central European forests has changed fundamentally from limiting resource to environmental problem. As the retrospective tracking of nutrient availability by soil chemical and biogeochemical measurements faces serious problems, bioindication based on understorey species composition is indispensable for monitoring broad-scale eutrophication. Based on a broad survey of more than 100,000 forest vegetation plots accessible in electronic data-bases from Germany and adjacent countries, we calculated unweighted average Ellenberg nutrient values (mN) as a proxy of plant-available macronutrients. Based on the quantiles of the frequency distribution of mN in a regionally stratified sample, we define five trophic classes, which can be used to compare dimensionless mN values. We studied spatial patterns of average nutrient values within 17 regions and compared the periods from 1899 to 1975 and 1976 to 2006. After 1975 eutrophic (mN > 5.67) and hypertrophic (mN > 6.28) conditions were common everywhere except in the Alps and Saxony-Anhalt, but very oligotrophic conditions (mN < 3.44) were still widespread in regions with nutrient-poor bedrock. Before 1975 mN of plots had been lower than after 1975 in all but the southeastern regions. Between the pre- and post-1975 data the proportion of hypertrophic plots increased from 5.7 to 11.8%, and that of very oligo-trophic plots decreased from 14.6 to 8.3%. To remove bias resulting from uneven distribution, the dataset was stratified by five tree layer dominance types, period and region and resampled. In pre-1975 plots medians of mN increased in the order Pinus sylvestris, Quercus spp., Picea abies, Fagus sylvatica and Alnus spp, whereas the increase of mN was highest in forest types with historically low nutrient values. Therefore, the widespread change in mN must be attributed to the pronounced vegetation changes in Quercus and Pinus stands, indicating the importance of land-use change, i.e. recovery of nutrient cycles after hundreds of years of exploitation through coppicing, grazing and litter use. The analysis confirms eutrophication as a megatrend of modern vegetation change and demonstrates the high research potential of linking vegetation plot databases across large regions.
Pathophysiological role of prostanoids in coagulation of the portal venous system in liver cirrhosis
(2019)
Background: Prostanoids are important regulators of platelet aggregation and thrombotic arterial diseases. Their involvement in the development of portal vein thrombosis, frequent in decompensated liver cirrhosis, is still not investigated.
Methods: Therefore, we used pro-thrombotic venous milieu generation by bare metal stent transjugular intrahepatic portosystemic shunt insertion, to study the role of prostanoids in decompensated liver cirrhosis. Here, 89 patients receiving transjugular intrahepatic portosystemic shunt insertion were included in the study, and baseline levels of thromboxane B2, prostaglandin D2 and prostaglandin E2 were measured in the portal and the hepatic vein.
Results: While the hepatic vein contained higher levels of thromboxane B2 than the portal vein, levels of prostaglandin E2 and D2 were higher in the portal vein (all P<0.0001). Baseline concentrations of thromboxane B2 in the portal vein were independently associated with an increase of portal hepatic venous pressure gradient during short term follow-up, as an indirect sign of thrombogenic potential (multivariable P = 0.004). Moreover, severity of liver disease was inversely correlated with portal as well as hepatic vein levels of prostaglandin D2 and E2 (all P<0.0001).
Conclusions: Elevated portal venous thromboxane B2 concentrations are possibly associated with the extent of thrombogenic potential in patients with decompensated liver cirrhosis.
Trial registration: ClinicalTrials.gov identifier: NCT03584204.
Acute deterioration of liver cirrhosis (e.g., infections, acute‐on‐chronic liver failure [ACLF]) requires an increase in cardiac contractility. The insufficiency to respond to these situations could be deleterious. Left ventricular global longitudinal strain (LV‐GLS) has been shown to reflect left cardiac contractility in cirrhosis better than other parameters and might bear prognostic value. Therefore, this retrospective study investigated the role of LV‐GLS in the outcome after transjugular intrahepatic portosystemic shunt (TIPS) and the development of ACLF. We included 114 patients (48 female patients) from the Noninvasive Evaluation Program for TIPS and Their Follow‐Up Network (NEPTUN) cohort. This number provided sufficient quality and structured follow‐up with the possibility of calculating major scores (Child, Model for End‐Stage Liver Disease [MELD], Chronic Liver Failure Consortium acute decompensation [CLIF‐C AD] scores) and recording of the events (development of decompensation episode and ACLF). We analyzed the association of LV‐GLS with overall mortality and development of ACLF in patients with TIPS. LV‐GLS was independently associated with overall mortality (hazard ratio [HR], 1.123; 95% confidence interval [CI],1.010‐1.250) together with aspartate aminotransferase (HR, 1.009; 95% CI, 1.004‐1.014) and CLIF‐C AD score (HR, 1.080; 95% CI, 1.018‐1.137). Area under the receiver operating characteristic curve (AUROC) analysis for LV‐GLS for overall survival showed higher area under the curve (AUC) than MELD and CLIF‐C AD scores (AUC, 0.688 versus 0.646 and 0.573, respectively). The best AUROC‐determined LV‐GLS cutoff was −16.6% to identify patients with a significantly worse outcome after TIPS at 3 months, 6 months, and overall. LV‐GLS was independently associated with development of ACLF (HR, 1.613; 95% CI, 1.025‐2.540) together with a MELD score above 15 (HR, 2.222; 95% CI, 1.400‐3.528). Conclusion: LV‐GLS is useful for identifying patients at risk of developing ACLF and a worse outcome after TIPS. Although validation is required, this tool might help to stratify risk in patients receiving TIPS.
Background: Biological psychiatry aims to understand mental disorders in terms of altered neurobiological pathways. However, for one of the most prevalent and disabling mental disorders, Major Depressive Disorder (MDD), patients only marginally differ from healthy individuals on the group-level. Whether Precision Psychiatry can solve this discrepancy and provide specific, reliable biomarkers remains unclear as current Machine Learning (ML) studies suffer from shortcomings pertaining to methods and data, which lead to substantial over-as well as underestimation of true model accuracy.
Methods: Addressing these issues, we quantify classification accuracy on a single-subject level in N=1,801 patients with MDD and healthy controls employing an extensive multivariate approach across a comprehensive range of neuroimaging modalities in a well-curated cohort, including structural and functional Magnetic Resonance Imaging, Diffusion Tensor Imaging as well as a polygenic risk score for depression.
Findings Training and testing a total of 2.4 million ML models, we find accuracies for diagnostic classification between 48.1% and 62.0%. Multimodal data integration of all neuroimaging modalities does not improve model performance. Similarly, training ML models on individuals stratified based on age, sex, or remission status does not lead to better classification. Even under simulated conditions of perfect reliability, performance does not substantially improve. Importantly, model error analysis identifies symptom severity as one potential target for MDD subgroup identification.
Interpretation: Although multivariate neuroimaging markers increase predictive power compared to univariate analyses, single-subject classification – even under conditions of extensive, best-practice Machine Learning optimization in a large, harmonized sample of patients diagnosed using state-of-the-art clinical assessments – does not reach clinically relevant performance. Based on this evidence, we sketch a course of action for Precision Psychiatry and future MDD biomarker research.
Investigators in the cognitive neurosciences have turned to Big Data to address persistent replication and reliability issues by increasing sample sizes, statistical power, and representativeness of data. While there is tremendous potential to advance science through open data sharing, these efforts unveil a host of new questions about how to integrate data arising from distinct sources and instruments. We focus on the most frequently assessed area of cognition - memory testing - and demonstrate a process for reliable data harmonization across three common measures. We aggregated raw data from 53 studies from around the world which measured at least one of three distinct verbal learning tasks, totaling N = 10,505 healthy and brain-injured individuals. A mega analysis was conducted using empirical bayes harmonization to isolate and remove site effects, followed by linear models which adjusted for common covariates. After corrections, a continuous item response theory (IRT) model estimated each individual subject’s latent verbal learning ability while accounting for item difficulties. Harmonization significantly reduced inter-site variance by 37% while preserving covariate effects. The effects of age, sex, and education on scores were found to be highly consistent across memory tests. IRT methods for equating scores across AVLTs agreed with held-out data of dually-administered tests, and these tools are made available for free online. This work demonstrates that large-scale data sharing and harmonization initiatives can offer opportunities to address reproducibility and integration challenges across the behavioral sciences.
Prognostic assessment of patients with liver cirrhosis allocated for implantation of a transjugular intrahepatic portosystemic shunt (TIPS) is a challenging task in clinical practice. The aim of our study was to assess the prognostic value of the CLIF‐C AD (Acute Decompensation) score in patients with TIPS implantation. Transplant‐free survival (TFS) and 3‐month mortality were reviewed in 880 patients who received de novo TIPS implantation for the treatment of cirrhotic portal hypertension. The prognostic value of the CLIF‐C AD score was compared with the Model for End‐Stage Liver Disease (MELD) score, Child‐Pugh score, and albumin‐bilirubin (ALBI) score using Harrell’s C concordance index. The median TFS after TIPS implantation was 40.0 (34.6‐45.4) months. The CLIF‐C AD score (c = 0.635 [0.609‐0.661]) was superior in the prediction of TFS in comparison to MELD score (c = 0.597 [0.570‐0.623], P = 0.006), Child‐Pugh score (c = 0.579 [0.552‐0.606], P < 0.001), and ALBI score (c = 0.573 [0.545‐0.600], P < 0.001). However, the CLIF‐C AD score did not perform significantly better than the MELD‐Na score (c = 0.626 [0.599‐0.653], P = 0.442). There were no profound differences in the scores’ ranking with respect to indication for TIPS implantation, stent type, or underlying liver disease. Subgroup analyses revealed that a CLIF‐C AD score >45 was a predictor of 3‐month mortality in the supposed low‐risk group of patients with a MELD score ≤12 (14.7% vs. 5.1%, P < 0.001). Conclusion: The CLIF‐C AD score is suitable for prognostic assessment of patients with cirrhotic portal hypertension receiving TIPS implantation. In the prediction of TFS, the CLIF‐C AD score is superior to MELD score, Child‐Pugh score, and ALBI score but not the MELD‐Na score.