Abstract: PO0121
Deep Learning for Subphenotype Identification in COVID-19-Associated AKI
Session Information
- COVID-19: Health Systems and More
November 04, 2021 | Location: On-Demand, Virtual Only
Abstract Time: 10:00 AM - 12:00 PM
Category: Coronavirus (COVID-19)
- 000 Coronavirus (COVID-19)
Authors
- Jaladanki, Suraj K., Icahn School of Medicine at Mount Sinai, New York, New York, United States
- Dellepiane, Sergio, Icahn School of Medicine at Mount Sinai, New York, New York, United States
- Vaid, Akhil, Icahn School of Medicine at Mount Sinai, New York, New York, United States
- Paranjpe, Ishan, Icahn School of Medicine at Mount Sinai, New York, New York, United States
- Singh, Karandeep, Icahn School of Medicine at Mount Sinai, New York, New York, United States
- Chan, Lili, Icahn School of Medicine at Mount Sinai, New York, New York, United States
- Coca, Steven G., Icahn School of Medicine at Mount Sinai, New York, New York, United States
- Nadkarni, Girish N., Icahn School of Medicine at Mount Sinai, New York, New York, United States
Background
Acute kidney injury (AKI) is common in COVID-19 and associated with increased adverse outcomes. COVID-associated AKI (COVID-AKI) pathophysiology is heterogenous, and deep learning may discover subphenotypes.
Methods
We used data from 5 New York City hospitals from adults admitted between March ‘20-March ‘21 with COVID and AKI, excluding patients with kidney failure. An autoencoder compressed 58 features containing comorbidities, the first laboratory values and vital signs within 48 hours of admission for unsupervised K-means clustering. Outcomes were mortality, dialysis, mechanical ventilation, and ICU admission.
Results
We identified 1634 patients with COVID-AKI and discovered 3 subphenotypes. Subphenotype one had 576 patients (35%); two had 635 patients (39%), and three had 423 patients (26%) (Table 1). Subphenotype three had the lowest median blood pressures, highest median BMI, and highest rates of all outcomes. (Figure 1)
Conclusion
There are distinct subphenotypes in COVID-AKI indicating the heterogeneity of this condition.
Table 1. Demographics of COVID-Associated AKI Subphenotypes.
Overall | Subphenotype 1 | Subphenotype 2 | Subphenotype 3 | P-value | |
N (%) | 1634 | 576 (35.3%) | 635 ( 34.1%) | 423 (25.6%) | |
Age, median (IQR) | 73.0 [63.0,83.0] | 74.0 [63.0,84.0] | 75.0 [65.0,83.0] | 69.0 [59.5,78.0] | <0.001 |
Female, n (%) | 936 (57.3) | 324 (56.2) | 318 (50.1) | 294 (69.5) | <0.001 |
Race, n (%) | |||||
Asian | 63 (3.9) | 11 (1.9) | 31 (4.9) | 21 (5.0) | 0.001 |
African American | 489 (29.9) | 190 (33.0) | 187 (29.4) | 112 (26.5) | |
White | 427 (26.1) | 158 (27.4) | 153 (24.1) | 116 (27.4) | |
Unknown | 655 (40.1) | 217 (37.7) | 264 (41.6) | 174 (41.1) | |
Ethnicity, n (%) | |||||
Hispanic/Latino | 379 (23.2) | 100 (17.4) | 183 (28.8) | 96 (22.7) | <0.001 |
Non-Hispanic/Latino | 995 (60.9) | 351 (60.9) | 390 (61.4) | 254 (60.0) | |
Unknown | 260 (15.9) | 125 (21.7) | 62 (9.8) | 73 (17.3) | |
Clinical Outcomes, n (%) | |||||
ICU Admission | 651 (39.8) | 174 (30.2) | 236 (37.2) | 241 (57.0) | <0.001 |
Dialysis | 147 (9.0) | 49 (8.5) | 57 (9.0) | 41 (9.7) | 0.811 |
Mechanical Ventilation | 220 (13.5) | 74 (12.8) | 72 (11.3) | 74 (17.5) | 0.014 |
Mortality | 886 (54.2) | 298 (51.7) | 335 (52.8) | 253 (59.8) | 0.026 |
Figure 1a: Proportions of ICU admission, dialysis or mechanical ventilation usage, and mortality across subphenotypes. Figure 1b: Top 4 features with the largest log-transformed differences between subphenotypes 1 and 3.
Funding
- NIDDK Support