SciELO journals
Browse
1/1
4 files

Urine microscopy as a biomarker of Acute Kidney Injury following cardiac surgery with cardiopulmonary bypass

dataset
posted on 2019-10-23, 03:17 authored by João Carlos Goldani, José Antônio Poloni, Fabiano Klaus, Roger Kist, Larissa Sgaria Pacheco, Elizete Keitel

Abstract Introduction: Acute kidney injury (AKI) occurs in about 22% of the patients undergoing cardiac surgery and 2.3% requires renal replacement therapy (RRT). The current diagnostic criteria for AKI by increased serum creatinine levels have limitations and new biomarkers are being tested. Urine sediment may be considered a biomarker and it can help to differentiate pre-renal (functional) from renal (intrinsic) AKI. Aims: To investigate the microscopic urinalysis in the AKI diagnosis in patients undergoing cardiac surgery with cardiopulmonary bypass. Methods: One hundred and fourteen patients, mean age 62.3 years, 67.5 % male, with creatinine 0.91 mg/dL (SD 0.22) had a urine sample examined in the first 24 h after the surgery. We looked for renal tubular epithelial cells (RTEC) and granular casts (GC) and associated the results with AKI development as defined by KDIGO criteria. Results: Twenty three patients (20.17 %) developed AKI according to the serum creatinine criterion and 76 (66.67 %) by the urine output criterion. Four patients required RRT. Mortality was 3.51 %. The use of urine creatinine criterion to predict AKI showed a sensitivity of 34.78 % and specificity of 86.81 %, positive likelihood ratio of 2.64 and negative likelihood ratio of 0.75, AUC-ROC of 0.584 (95%CI: 0.445-0.723). For the urine output criterion sensitivity was 23.68 % and specificity 92.11 %, AUC-ROC was 0.573 (95%CI: 0.465-0.680). Conclusion: RTEC and GC in urine sample detected by microscopy is a highly specific biomarker for early AKI diagnosis after cardiac surgery.

History

Usage metrics

    Jornal Brasileiro de Patologia e Medicina Laboratorial

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC