Urine and serum concentrations of Cytokeratin 19 in preeclampsia

Objective: To evaluate the usefulness of Cytokeratin 19 as biomarker for the diagnosis of preeclampsia.

Study design: Cytokeratin 19 protein fragment CYFRA 21-1 was measured by means of electrochemiluminescence immunoassays in urine and serum samples of 32 women with preeclampsia and 32 samples of normotensive healthy singleton pregnancies at random, matched for gestational age, as controls.

Results: The median serum concentrations of CYFRA 21-1 in controls and women with preeclampsia were 2.4 (range 1.3-6.6)ng/mL and 4.4 (range 2.1-16.2)ng/mL, respectively (p<0.001). The median urine concentrations of CYFRA 21-1 in controls and women with preeclampsia were 13.7 (range 0.7-441.4)ng/mL and 11.8 (range 1.5-338.6)ng/mL, respectively (p=0.629). Calculation of a ROC curve to study the use of serum CYFRA 21-1 concentration as a predictor of preeclampsia revealed cut-off points with the highest sum of specificity and sensitivity of 3.2ng/mL, leading to specificity of 75% and sensitivity of 84%. A similar curve calculated for CYFRA 21-1 in urine showed an area under the curve of 0.536 meaning no predictive power. The correlation between urinary excretion of protein in 24h and serum concentrations of CYFRA 21-1 in the case group was r=0.26, which is not significant (p=0.258). The correlation between proteinuria and urine values of CYFRA 21-1 was r=0.10, which also is not significant (p=0.666).

Conclusion: Serum levels of Cytokeratin 19 fragment are increased in women with preeclampsia. However this does not result in a significant difference in CYFRA 21-1 levels in maternal urine. Thus Cytokeratin 19 fragment may prove to be a valuable biomarker for preeclampsia in serum but not urine. We propose further longitudinal studies to investigate the role of Cytokeratin 19 in maternal serum of women with preeclampsia.