Utilization of Deep Learning for Subphenotype Identification in Sepsis-Associated Acute Kidney Injury
Copyright © 2020 by the American Society of Nephrology..
BACKGROUND AND OBJECTIVES: Sepsis-associated AKI is a heterogeneous clinical entity. We aimed to agnostically identify sepsis-associated AKI subphenotypes using deep learning on routinely collected data in electronic health records.
DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: We used the Medical Information Mart for Intensive Care III database, which consists of electronic health record data from intensive care units in a tertiary care hospital in the United States. We included patients ≥18 years with sepsis who developed AKI within 48 hours of intensive care unit admission. We then used deep learning to utilize all available vital signs, laboratory measurements, and comorbidities to identify subphenotypes. Outcomes were mortality 28 days after AKI and dialysis requirement.
RESULTS: We identified 4001 patients with sepsis-associated AKI. We utilized 2546 combined features for K-means clustering, identifying three subphenotypes. Subphenotype 1 had 1443 patients, and subphenotype 2 had 1898 patients, whereas subphenotype 3 had 660 patients. Subphenotype 1 had the lowest proportion of liver disease and lowest Simplified Acute Physiology Score II scores compared with subphenotypes 2 and 3. The proportions of patients with CKD were similar between subphenotypes 1 and 3 (15%) but highest in subphenotype 2 (21%). Subphenotype 1 had lower median bilirubin levels, aspartate aminotransferase, and alanine aminotransferase compared with subphenotypes 2 and 3. Patients in subphenotype 1 also had lower median lactate, lactate dehydrogenase, and white blood cell count than patients in subphenotypes 2 and 3. Subphenotype 1 also had lower creatinine and BUN than subphenotypes 2 and 3. Dialysis requirement was lowest in subphenotype 1 (4% versus 7% [subphenotype 2] versus 26% [subphenotype 3]). The mortality 28 days after AKI was lowest in subphenotype 1 (23% versus 35% [subphenotype 2] versus 49% [subphenotype 3]). After adjustment, the adjusted odds ratio for mortality for subphenotype 3, with subphenotype 1 as a reference, was 1.9 (95% confidence interval, 1.5 to 2.4).
CONCLUSIONS: Utilizing routinely collected laboratory variables, vital signs, and comorbidities, we were able to identify three distinct subphenotypes of sepsis-associated AKI with differing outcomes.
| Errataetall: |
CommentIn: Clin J Am Soc Nephrol. 2020 Oct 8;:. - PMID 33034570 |
|---|---|
| Medienart: |
E-Artikel |
| Erscheinungsjahr: |
2020 |
|---|---|
| Erschienen: |
2020 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:15 |
|---|---|
| Enthalten in: |
Clinical journal of the American Society of Nephrology : CJASN - 15(2020), 11 vom: 06. Nov., Seite 1557-1565 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Chaudhary, Kumardeep [VerfasserIn] |
|---|
| Links: |
|---|
| Anmerkungen: |
Date Completed 22.11.2021 Date Revised 21.04.2023 published: Print-Electronic CommentIn: Clin J Am Soc Nephrol. 2020 Oct 8;:. - PMID 33034570 Citation Status MEDLINE |
|---|
| doi: |
10.2215/CJN.09330819 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM31603195X |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM31603195X | ||
| 003 | DE-627 | ||
| 005 | 20231225160210.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231225s2020 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.2215/CJN.09330819 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1053.xml |
| 035 | |a (DE-627)NLM31603195X | ||
| 035 | |a (NLM)33033164 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Chaudhary, Kumardeep |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Utilization of Deep Learning for Subphenotype Identification in Sepsis-Associated Acute Kidney Injury |
| 264 | 1 | |c 2020 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 22.11.2021 | ||
| 500 | |a Date Revised 21.04.2023 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a CommentIn: Clin J Am Soc Nephrol. 2020 Oct 8;:. - PMID 33034570 | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a Copyright © 2020 by the American Society of Nephrology. | ||
| 520 | |a BACKGROUND AND OBJECTIVES: Sepsis-associated AKI is a heterogeneous clinical entity. We aimed to agnostically identify sepsis-associated AKI subphenotypes using deep learning on routinely collected data in electronic health records | ||
| 520 | |a DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: We used the Medical Information Mart for Intensive Care III database, which consists of electronic health record data from intensive care units in a tertiary care hospital in the United States. We included patients ≥18 years with sepsis who developed AKI within 48 hours of intensive care unit admission. We then used deep learning to utilize all available vital signs, laboratory measurements, and comorbidities to identify subphenotypes. Outcomes were mortality 28 days after AKI and dialysis requirement | ||
| 520 | |a RESULTS: We identified 4001 patients with sepsis-associated AKI. We utilized 2546 combined features for K-means clustering, identifying three subphenotypes. Subphenotype 1 had 1443 patients, and subphenotype 2 had 1898 patients, whereas subphenotype 3 had 660 patients. Subphenotype 1 had the lowest proportion of liver disease and lowest Simplified Acute Physiology Score II scores compared with subphenotypes 2 and 3. The proportions of patients with CKD were similar between subphenotypes 1 and 3 (15%) but highest in subphenotype 2 (21%). Subphenotype 1 had lower median bilirubin levels, aspartate aminotransferase, and alanine aminotransferase compared with subphenotypes 2 and 3. Patients in subphenotype 1 also had lower median lactate, lactate dehydrogenase, and white blood cell count than patients in subphenotypes 2 and 3. Subphenotype 1 also had lower creatinine and BUN than subphenotypes 2 and 3. Dialysis requirement was lowest in subphenotype 1 (4% versus 7% [subphenotype 2] versus 26% [subphenotype 3]). The mortality 28 days after AKI was lowest in subphenotype 1 (23% versus 35% [subphenotype 2] versus 49% [subphenotype 3]). After adjustment, the adjusted odds ratio for mortality for subphenotype 3, with subphenotype 1 as a reference, was 1.9 (95% confidence interval, 1.5 to 2.4) | ||
| 520 | |a CONCLUSIONS: Utilizing routinely collected laboratory variables, vital signs, and comorbidities, we were able to identify three distinct subphenotypes of sepsis-associated AKI with differing outcomes | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a AKI | |
| 650 | 4 | |a acute kidney injury | |
| 650 | 4 | |a acute renal failure | |
| 650 | 4 | |a deep learning | |
| 650 | 4 | |a dialysis | |
| 650 | 4 | |a mortality | |
| 650 | 4 | |a subtypes | |
| 650 | 7 | |a Lactic Acid |2 NLM | |
| 650 | 7 | |a 33X04XA5AT |2 NLM | |
| 650 | 7 | |a Creatinine |2 NLM | |
| 650 | 7 | |a AYI8EX34EU |2 NLM | |
| 650 | 7 | |a L-Lactate Dehydrogenase |2 NLM | |
| 650 | 7 | |a EC 1.1.1.27 |2 NLM | |
| 650 | 7 | |a Alanine Transaminase |2 NLM | |
| 650 | 7 | |a EC 2.6.1.2 |2 NLM | |
| 650 | 7 | |a Glutamyl Aminopeptidase |2 NLM | |
| 650 | 7 | |a EC 3.4.11.7 |2 NLM | |
| 650 | 7 | |a Bilirubin |2 NLM | |
| 650 | 7 | |a RFM9X3LJ49 |2 NLM | |
| 700 | 1 | |a Vaid, Akhil |e verfasserin |4 aut | |
| 700 | 1 | |a Duffy, Áine |e verfasserin |4 aut | |
| 700 | 1 | |a Paranjpe, Ishan |e verfasserin |4 aut | |
| 700 | 1 | |a Jaladanki, Suraj |e verfasserin |4 aut | |
| 700 | 1 | |a Paranjpe, Manish |e verfasserin |4 aut | |
| 700 | 1 | |a Johnson, Kipp |e verfasserin |4 aut | |
| 700 | 1 | |a Gokhale, Avantee |e verfasserin |4 aut | |
| 700 | 1 | |a Pattharanitima, Pattharawin |e verfasserin |4 aut | |
| 700 | 1 | |a Chauhan, Kinsuk |e verfasserin |4 aut | |
| 700 | 1 | |a O'Hagan, Ross |e verfasserin |4 aut | |
| 700 | 1 | |a Van Vleck, Tielman |e verfasserin |4 aut | |
| 700 | 1 | |a Coca, Steven G |e verfasserin |4 aut | |
| 700 | 1 | |a Cooper, Richard |e verfasserin |4 aut | |
| 700 | 1 | |a Glicksberg, Benjamin |e verfasserin |4 aut | |
| 700 | 1 | |a Bottinger, Erwin P |e verfasserin |4 aut | |
| 700 | 1 | |a Chan, Lili |e verfasserin |4 aut | |
| 700 | 1 | |a Nadkarni, Girish N |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Clinical journal of the American Society of Nephrology : CJASN |d 2006 |g 15(2020), 11 vom: 06. Nov., Seite 1557-1565 |w (DE-627)NLM172123720 |x 1555-905X |7 nnns |
| 773 | 1 | 8 | |g volume:15 |g year:2020 |g number:11 |g day:06 |g month:11 |g pages:1557-1565 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.2215/CJN.09330819 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 15 |j 2020 |e 11 |b 06 |c 11 |h 1557-1565 | ||