Determination of cytokeratin-18 and granzyme B in urine: screening method for the diagnosis of chronic kidney diseases in children

Abstract

Objective. To evaluate the possibilities of determining molecules of apoptosis and endothelial dysfunction in urine as indicators for screening chronic kidney diseases in children.
Materials and methods. 141 patients with chronic kidney diseases were examined. The comparison group included conditionally healthy children (n = 48) without signs of kidney damage and autoimmune process. Quantitative determination of granzyme B, cytokeratin-18, and intercellular adhesion molecule-1 (ICAM-1) was performed by enzyme immunoassay.
Results. The concentration of granzyme B in urine is 0.021 or higher, indicating that the patient has chronic kidney disease. The diagnostic accuracy of this test according to the ROC analysis was 87.0 % (95 % CI (69.0—100.0), sensitivity (Se) of the test was 90.5 %, specificity (Sp) was 93.0 %). The level of cytokeratin-18 in urine of 0.036 and higher also indicates the presence of chronic nephropathy (diagnostic accuracy — 94.9 % (95 % (CI 89.4—100.0), Se — 90.0 %, Sp — 91.3 %). In the case of the ICAM-1 test, the presence of chronic kidney disease can be indicated if the value of this marker in urine is less than 9.8 (diagnostic accuracy according to ROC analysis is 97.4 % (95 % (CI (93.3—100.0); Se — 90.0 %, Sp — 97.6 %). As a result of a comparative assessment of clinical usefulness, the use of a test with the determination of cytokeratin-18 and granzyme B in urine turned out to be the most promising.
Conclusion. When conducting a two-stage screening with sequential tests for the determination of granzyme B and cytokeratin-18 in urine in a population of children containing 0.275 % (1 in 364) of patients with chronic kidney disease, the probability of chronic nephropathy in a patient with a double positive test will be 26.94 %.