Kidney Week

Abstract: PUB005

Predicting In-Hospital Mortality Among COVID-19 Pneumonia Patients With AKI

Session Information

• Publication Only

Category: Coronavirus (COVID-19)

• 000 Coronavirus (COVID-19)

Authors

• Wilhelm, David J., University of Louisville, Louisville, Kentucky, United States

David J. Wilhelm,

• Ali, T'shura, University of Louisville, Louisville, Kentucky, United States

T'shura Ali,

• Brier, Michael E., University of Louisville, Louisville, Kentucky, United States

Michael E. Brier,

• Caster, Dawn J., University of Louisville, Louisville, Kentucky, United States

Dawn J. Caster,

• Arnold, Forest W., University of Louisville, Louisville, Kentucky, United States

Forest W. Arnold,

• Huang, Jiapeng, University of Louisville, Louisville, Kentucky, United States

Jiapeng Huang,

Background

COVID-19 has been identified as a disease causing respiratory failure but is now known to affect the kidneys among other organs. Several studies among COVID-19 patients have shown a significant association between acute kidney injury (AKI) and mortality. There is limited data examining if the effect of AKI on mortality is different across variants. The main objective of this study is to examine the association between AKI and in-hospital mortality among COVID-19 pneumonia (PNA) patients during the original strain and the delta variant.

Methods

Data was obtained from a retrospective analysis of patients hospitalized with COVID-19 PNA from March 2020 until March 2021. The database had two cohorts: the original strain and the delta variant. The presence of AKI was confirmed by an examination of medical records for 612 patients using the AKIN criteria (creatinine ≥0.3 mg/dL above baseline). Chronic kidney disease was defined by estimated Glomerular Filtration Rate (eGFR) calculated using the CKD-EPI 2021 equation. Logistic regression was used to estimate relative risk (RR) for mortality using factors in Table 1.

Results

AKI was present in 414 patients (67.6%). Of the 612 patients reviewed, there were 443 survivors and 169 non-survivors at discharge. Among the non-survivors, there were a higher proportion of AKI (84%) and males (64%) and a lower proportion of African Americans (30%). The non-survivors were younger (67 years) and had a lower eGFR (37 mL/min). Logistic regression results are shown in Table 1.

Conclusion

Analyses showed that among patients hospitalized with COVID-19 PNA, the RR of in-hospital mortality was 3.28 times higher for those with AKI compared to those without AKI. We found no significant difference for in-hospital mortality between the two cohorts when adjusted for presence of AKI. Other findings showed that males may have a greater risk of mortality as compared to females and those of African American race may have a potential survival advantage.

Table 1