Dataset Information

Utilization of Deep Learning for Subphenotype Identification in Sepsis-Associated Acute Kidney Injury.

ABSTRACT:

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.

SUBMITTER: Chaudhary K

PROVIDER: S-EPMC7646246 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:To identify mechanisms associated with sepsis-acute kidney injury based on the expression levels of renal injury biomarkers, neutrophil gelatinase-associated lipocalin, and kidney injury molecule-1 in renal biopsies which may allow the identification of sepsis-acute kidney injury patient subtypes.DesignProspective, clinical laboratory study using "warm" human postmortem sepsis-acute kidney injury kidney biopsies.SettingResearch laboratory at university teaching hospital.SubjectsAdult patients who died of sepsis in the ICU and control patients undergoing tumor nephrectomy.Measurements and main resultsReverse transcription quantitative polymerase chain reaction and immunohistochemical staining were used to quantify messenger RNA and protein expression levels of neutrophil gelatinase-associated lipocalin and kidney injury molecule-1 in the kidney of sepsis-acute kidney injury patients and control subjects. Morphometric analysis was used to quantify renal and glomerular neutrophil gelatinase-associated lipocalin and kidney injury molecule-1 protein levels. Neutrophil gelatinase-associated lipocalin and kidney injury molecule-1 messenger RNA and protein levels were increased in kidneys of sepsis-acute kidney injury patients compared with control kidney tissue. Neutrophil gelatinase-associated lipocalin was localized in the distal tubules, collecting ducts, the adventitia of the renal arterioles, and in the glomerular tufts of renal biopsies from sepsis-acute kidney injury patients. In contrast, kidney injury molecule-1 was localized at the brush border of the proximal tubules. There was no correlation between neutrophil gelatinase-associated lipocalin and kidney injury molecule-1 levels. Furthermore, renal neutrophil gelatinase-associated lipocalin and kidney injury molecule-1 levels were not associated with the extent of renal injury, the severity of critical illness, or serum creatinine levels at either ICU admission or day of expiration. By laser microdissecting glomeruli, followed by reverse transcription quantitative polymerase chain reaction, we identified heterogenous glomerular neutrophil gelatinase-associated lipocalin production in the kidney of sepsis-acute kidney injury patients.ConclusionWe found differences in the expression of neutrophil gelatinase-associated lipocalin and kidney injury molecule-1 in patients with the same syndrome "sepsis-acute kidney injury" meaning there is no single pathway leading to sepsis-acute kidney injury. This underscores the beliefs that there are many/different pathophysiological pathways that can cause sepsis-acute kidney injury. Hence, patients with criteria that meet the definitions of both acute kidney injury and sepsis can be divided into subtypes based on pathophysiological features.