Novel machine learning-based point-score model as a non-invasive decision-making tool for infected ascites in patients with hydropic decompensated liver cirrhosis: A retrospective multicentre study.
Abstract Purpose This study aimed to assess the distinctive features of patients with infected ascites and liver cirrhosis and develop a scoring system allowing to accurately identify patients who do not require abdominocentesis to rule out infected ascites. Methods A total of 700 episodes of patients with decompensated liver cirrhosis undergoing abdominocentesis between 2006 and 2020 were included. 532 spontaneous bacterial peritonitis episodes and 37 secondary peritonitis episodes were compared to a control group of 125 patients with 131 episodes of non-infected ascites. Overall, 34 clinical, drug, and laboratory features were evaluated using machine learning to identify key differentiation criteria and integrate them into a point-score model. Results The most important distinction criteria between infected and non-infected ascites were inflammatory markers C-reactive protein and leukocyte count, the occurrence of organ failure, fever, and comorbidities. In total, 11 discriminatory features were selected using a Lasso regression model to establish a point-score model. Considering a pre-test probability for infected ascites of 10%, 15%, and 25%, the negative and positive predictive values of the point-score model for infected ascites were 98.1%, 97.0%, 94.6% and 14.9%, 21.8%, and 34.5%, respectively. Besides the main model, a simplified model was generated, containing only features that are fast to collect, and revealed similar predictive values. Conclusions Our point-score model appears to be a promising non-invasive approach to rule out abdominocentesis in clinical routine with high negative predictive values in patients with hydropic decompensated liver cirrhosis. Diagnosis of infected ascites, on the other hand, requires abdominocentesis..
| Medienart: |
|
|---|
Preprint| Erscheinungsjahr: |
2022 |
|---|---|
| Erschienen: |
2022 |
| Enthalten in: |
ResearchSquare.com - (2022) vom: 10. Aug. Zur Gesamtaufnahme - year:2022 |
|---|
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Würstle, Silvia [VerfasserIn] |
|---|
| Links: |
Volltext [kostenfrei] |
|---|
| Themen: |
|---|
| doi: |
10.21203/rs.3.rs-1930434/v1 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
XRA036832138 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | XRA036832138 | ||
| 003 | DE-627 | ||
| 005 | 20230429190255.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 220811s2022 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.21203/rs.3.rs-1930434/v1 |2 doi | |
| 035 | |a (DE-627)XRA036832138 | ||
| 035 | |a (ResearchSquare)10.21203/rs.3.rs-1930434/v1 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Würstle, Silvia |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Novel machine learning-based point-score model as a non-invasive decision-making tool for infected ascites in patients with hydropic decompensated liver cirrhosis: A retrospective multicentre study. |
| 264 | 1 | |c 2022 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 520 | |a Abstract Purpose This study aimed to assess the distinctive features of patients with infected ascites and liver cirrhosis and develop a scoring system allowing to accurately identify patients who do not require abdominocentesis to rule out infected ascites. Methods A total of 700 episodes of patients with decompensated liver cirrhosis undergoing abdominocentesis between 2006 and 2020 were included. 532 spontaneous bacterial peritonitis episodes and 37 secondary peritonitis episodes were compared to a control group of 125 patients with 131 episodes of non-infected ascites. Overall, 34 clinical, drug, and laboratory features were evaluated using machine learning to identify key differentiation criteria and integrate them into a point-score model. Results The most important distinction criteria between infected and non-infected ascites were inflammatory markers C-reactive protein and leukocyte count, the occurrence of organ failure, fever, and comorbidities. In total, 11 discriminatory features were selected using a Lasso regression model to establish a point-score model. Considering a pre-test probability for infected ascites of 10%, 15%, and 25%, the negative and positive predictive values of the point-score model for infected ascites were 98.1%, 97.0%, 94.6% and 14.9%, 21.8%, and 34.5%, respectively. Besides the main model, a simplified model was generated, containing only features that are fast to collect, and revealed similar predictive values. Conclusions Our point-score model appears to be a promising non-invasive approach to rule out abdominocentesis in clinical routine with high negative predictive values in patients with hydropic decompensated liver cirrhosis. Diagnosis of infected ascites, on the other hand, requires abdominocentesis. | ||
| 650 | 4 | |a Biology |7 (dpeaa)DE-84 | |
| 650 | 4 | |a 570 |7 (dpeaa)DE-84 | |
| 700 | 1 | |a Hapfelmeier, Alexander |e verfasserin |4 aut | |
| 700 | 1 | |a Karapetyan, Siranush |e verfasserin |4 aut | |
| 700 | 1 | |a Studen, Fabian |e verfasserin |4 aut | |
| 700 | 1 | |a Isaakidou, Andriana |e verfasserin |4 aut | |
| 700 | 1 | |a Schneider, Tillman |e verfasserin |4 aut | |
| 700 | 1 | |a Schmid, Roland M. |e verfasserin |4 aut | |
| 700 | 1 | |a von Delius, Stefan |e verfasserin |4 aut | |
| 700 | 1 | |a Gundling, Felix |e verfasserin |4 aut | |
| 700 | 1 | |a Triebelhorn, Julian |e verfasserin |4 aut | |
| 700 | 1 | |a Burgkart, Rainer |e verfasserin |4 aut | |
| 700 | 1 | |a Obermeier, Andreas |e verfasserin |4 aut | |
| 700 | 1 | |a Mayr, Ulrich |e verfasserin |4 aut | |
| 700 | 1 | |a Heller, Stephan |e verfasserin |4 aut | |
| 700 | 1 | |a Rasch, Sebastian |e verfasserin |4 aut | |
| 700 | 1 | |a Lahmer, Tobias |e verfasserin |4 aut | |
| 700 | 1 | |a Geisler, Fabian |e verfasserin |4 aut | |
| 700 | 1 | |a Chan, Benjamin |e verfasserin |4 aut | |
| 700 | 1 | |a Turner, Paul E. |e verfasserin |4 aut | |
| 700 | 1 | |a Rothe, Kathrin |e verfasserin |4 aut | |
| 700 | 1 | |a Spinner, Christoph D. |e verfasserin |4 aut | |
| 700 | 1 | |a Schneider, Jochen |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t ResearchSquare.com |g (2022) vom: 10. Aug. |
| 773 | 1 | 8 | |g year:2022 |g day:10 |g month:08 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.21203/rs.3.rs-1930434/v1 |z kostenfrei |3 Volltext |
| 912 | |a GBV_XRA | ||
| 912 | |a SSG-OLC-PHA | ||
| 951 | |a AR | ||
| 952 | |j 2022 |b 10 |c 08 | ||